Movement Disorders: Handbook of Clinical Neurophysiology, Vol 1

Handbook of Clinical Neurophysiology Series Editors

Jasper R. Daube Managed Care Department, Mayo Clinic, 200 First Street SW; Rochester, MN 55905, USA and

Francois Mauguiere Functional Neurology and Epilepsy Department, Htipital Neurologique Pierre Wertheimer, 59 Boulevard Pinel, F-69394 Lyon Cedex 03, France

Volume 1 Movement Disorders Volume Editor

Mark Hallett Human Motor Control Section, NIH, Building 10, Room 5N226, 10 Center Drive, MSC 1428, Bethesda, MD 20892-1428, USA

2003

ELSEVIER Amsterdam • Boston • London • New York • Oxford • Paris • San Diego • San Francisco • Singapore • Sydney • Tokyo

ELSEVIER B.V. Sara Burgerhartstraat 25 P.O. Box 211, WOO AE Amsterdam, The Netherlands

© 2003 Elsevier B.V. All rights reserved. This work is protected under copyright by Elsevier, and the following terms and conditions apply to its use: Photocopying Single photocopies of single chapters may be made for personal use as allowed by national copyright laws. Permission of the Publisher and payment of a fee is required for all other photocopying, including multiple or systematic copying, copying for advertising or promotional purposes, resale, and all forms of document delivery. Special rates are available for educational institutions that wish to make photocopies for non-profit educational classroom use. Permissions may be sought directly from Elsevier's Health Science Rights Department, Elsevier Inc, 625 Walnut Street, Philadelphia, PA 19106, USA, phone: (+ I) 215 238 7869, fax: (+ I) 215 2382239, E-mail: [email protected]. You may also complete your request on-line via the Elsevier homepage (http://www.elsevier.com). by selecting 'Customer Support' and then 'Obtaining Permissions'.

In the USA, users may clear permissions and make payments through the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA; phone: (+ I) (978) 7508400, fax: (+ 1) (978) 7504744, and in the UK through the Copyright Licensing Agency Rapid Clearance Service (CLARCS), 90 Tottenham Court Road, London WIP OLP,UK; phone: (+44) 207 631 5555; fax: (+44) 207 6315500. Other countries may have a local reprographic rights agency for payments. Derivative Works Tables of contents may be reproduced for internal circulation, but permission of Elsevier is required for external resale or distribution of such material. Permission of the Publisher is required for all other derivative works, including compilations and translations.

Electronic Storage or Usage Permission of the Publisher is required to store or use electronically any material contained in this work, including any chapter or part of a chapter. Except as outlined above, no part of this work may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without prior written permission of the Publisher. Address permissions requests to: Elsevier Health Science Rights Department, at the phone, fax and e-mail addresses noted above. Notice No responsibility is assumed by the Publisher for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions or ideas contained in the material herein. Because of rapid advances in the medical sciences, in particular, independent verification of diagnoses and drug dosages should be made.

First edition 2003

ISBN: 0444-50725-6 ISSN: 1567-4231 (series)

8 The paper used in this publication meets the requirements of ANSlINlSO Z39.48-1992 (Permanence of Paper). Printed in The Netherlands.

Foreword

Clinical neurophysiology encompasses the application of a wide variety of electrophysiologic methods to the analysis of normal function and the diagnosis and treatment of diseases involving the central nervous system, peripheral nervous system and muscles. A small number of these methods are applied to a single category disease, but most are useful in multiple clinical settings. Previous editions of the Handbook of Clinical Neurophysiology have focused on categories of testing and the many ways they can be applied. The steady increase in growth of subspecialties in neurology and the study of disorders of the nervous system have led to a need for a compilation of the application of the whole range of physiologic methods used for the major categories of neurologic disease. Each volume will be designed to serve as the ultimate reference source for academic clinical neurophysiologists, and as a reference that will provide subspecialists in an area, they will need to fully understand, assess and treat disorders in their patients. As such these volumes will also serve as a major teaching text for trainees in that subspecialty. Subsequent volumes will include all of the clinical disorders served by clinical neurophysiology: the epilepsies, autonomic dysfunction, peripheral nerve disease, muscle disease, motor system disorders, somatosensory system disorders, behavioral disorders, visual and auditory system disorders, and monitoring neural function. Each will focus on the advances in one of these major areas of clinical neurophysiology. Each volume will include critical discussion of new knowledge in basic neurophysiology, approaches to characterization of disease type, localization, severity and prognosis with detailed discussion of advances in techniques to accomplish these. It is recognized that some techniques will be discussed in more than one volume, but with different focuses in each of them. Each volume will include an overview of the field, followed by a section that includes a detailed description of each of the CNP techniques used in the category of disorders, and a third section discussing specific diseases. The latter will include how to evaluate each and comparison of relative contribution of each of the methods of evaluation. A final section will discuss ongoing research studies and anticipated future advances. Selection of movement disorders as the first volume is particularly appropriate in view of the many advances in the application of clinical neurophysiology in these disorders. We are privileged to have one of the world's leaders in the clinical neurophysiology of movement disorders as the volume editor. He has done a superb job of assembling the world leaders in the description of the methods and in their application to particular categories of disease. Jasper R. Daube Francois Mauguiere

Series Editors

List of Contributors

M.Aramideh

Department of Neurology/Clinical Neurophysiology, Medical Center Alkmaar, P.O. Box 501, 1800 AM Alkmaar, The Netherlands.

C.Ardouin

INSERM 318, Joseph Fourier University, Centre Hospitalier Universitaire, Pavillon B, BP 217,38043 Grenoble Cedex 9, France.

J.U.J. Allum

Department of ORL, University Hospital, Basel, Switzerland.

P.G. Bain

Imperial College School of Medicine, Charing Cross Hospital Campus, Pulham Palace Road, London W6 8RF, UK.

A.L. Benabid

INSERM 318, Joseph Fourier University, Centre Hospitalier Universitaire, Pavillon B, BP 217,38043 Grenoble Cedex 9, France.

A. Benazzouz

INSERM 318, Joseph Fourier University, Centre Hospitalier Universitaire, Pavillon B, BP 217,38043 Grenoble Cedex 9, France.

R. Benecke

Department of Neurology, University of Rostock, Gehlsheimer Strasse 20, D-18147 Rostock, Germany.

A. Berardelli

Dipartimento di Scienze Neurologiche, Universita di Roma "La Sapienza", Viale Universita 30,00185 Rome, Italy.

B.R.Bloem

Department of Neurology, 326, University Medical Center St. Radboud, P.O. Box 9101, 6500 HB Nijmegen, The Netherlands.

C. Braun

MEG Center, Eberhard-Karls University Tubingen, Hoppe-Seyler Strasse 3, D-72076 Tubingen, Germany.

P. Brown

Sobell Department of Motor Neuroscience and Movement Disorders, Institute of Neurology, Queen Square, London WC1N 3BG, UK.

D. Burke

College of Health Sciences, The University of Sydney, Sydney, Australia.

J.N. Caviness

Department of Neurology, Mayo Clinic Scottsdale, 13400 East Shea Blvd., Scottsdale, AZ 85259, USA.

S. Chabardes

INSERM 318, Joseph Fourier University, Centre Hospitalier Universitaire, Pavillon B, BP 217,38043 Grenoble Cedex 9, France.

S. Chokroverty

Department of Neurology/Cronin 466, St. Vincents Hospital, 153 W 11th Street, New York, NY 10011, USA.

G. Croccu

Department of Neurological Sciences, University of Rome "La Sapienza", 00185 Rome, Italy.

viii

LIST OF CONTRIBUTORS

A. Curra

Istituto Neurologico Mediterraneo "Neuromed", Via Atinense 18, 86077 Pozzilli, IS, Italy.

G. Deuschl

Department of Neurology, Christian-Albrechts-Universitat Kiel, Niemannsweg 147, D-24lO5 Kiel, Germany.

C. Dohle

Department of Neurology, University Hospital Diisseldorf, Moorenstrasse 5, D-40225 Diisseldorf, Germany.

J.O. Dostrovsky

Department of Physiology, Faculty of Medicine, University of Toronto, Toronto, ON M5S lA8, Canada.

R.J. Elble

Department of Neurology, Southern Illinois University School of Medicine, P.O. Box 19643, Springfield, IL 62794-9643, USA.

U. Fietzek

Department of Neurology, Christian-Albrechts-Universitat Kiel, Niemannsweg 147, D-24lO5 Kiel, Germany.

M.K. Floeter

Electromyography Section, National Institute of Neurological Disorders and Stroke, NIH, 10 Center Drive, MSC 1404, Bethesda, MD 20892-1404, USA.

V. Fraix

INSERM 318, Joseph Fourier University, Centre Hospitalier Universitaire, Pavillon B, BP 217,38043 Grenoble Cedex 9, France.

H.-J. Freund

Department of Neurology, University Hospital Diisseldorf, Moorenstrasse 5, D-40225 Diisseldorf, Germany.

S.C. Gandevia

Spinal Injuries Research Centre, Prince of Wales Medical Research Institute, University of New South Wales, Sydney, Australia.

C. Gerloff

Cortical Physiology Research Group, Department of General Neurology, Eberhard-Karls University Tiibingen, Hoppe-Seyler Strasse 3, D-72076 Tiibingen, Germany.

J.-M. Grades

Department of Neurology, Mount Sinai Medical Center, 1 Gustave L. Levy Place, Annenberg 2/Box 1052, New York, NY 10029-6574, USA.

S.T. Grafton

Center for Cognitive Neuroscience, 6162 Moore Hall, Dartmouth College, Hanover, NH 03755, USA.

M. Hallett

Human Motor Control Section, National Institute of Neurological Disorders and Stroke, NIH, Building lO, Room 5N226, 10 Center Drive, MSC 1428, Bethesda, MD 20892-1428, USA.

M. Hayes

Department of Neurology, Concord Repatriation Hospital, Sydney, Australia.

B. Hellwig

Neurologische Universitatsklinik, Neurozentrum, Breisacher Strasse 64, D-79106 Freiburg, Germany.

W.Hening

Johns Hopkins Center for Restless Legs Syndrome, 5th Floor, Room 5B71C, Asthma & Allergy Center, 5501 Hopkins Bayview Circle, Baltimore, MD 21224, USA.

W.D. Hutchison

Department of Surgery, Division of Neurosurgery, Toronto Western HospitallMC 2-433, 399 Bathurst Street, East Wing 6-528, Toronto, ON M5T 2S8, Canada.

A. Ikeda

Department of Neurology, Kyoto University Graduate School of Medicine, Shogoin, Sakyo-ku, Kyoto 606-8507, Japan.

ix

LIST OF CONTRIBUTORS

M. Jahanshahi

Sobell Research Department of Motor Neuroscience and Movement Disorders, Institute of Neurology, Queen Square, London WCIN 3BG, UK.

R. Kaji

Department of Clinical Neuroscience, Hospital of the University of Tokushima, 2-Chome 5-1, Kuramotocho, Tokushima City, Tokushima 770-8503, Japan.

K.R. Kaufman

Biomechanics Laboratory, Charlton North L-IION, Mayo Clinic, 200 First Street SW, Rochester, MN 55905, USA.

S.Klebe

Department of Neurology, Christian-Albrechts-Universitat Kiel, Niemannsweg 147, D-24105 Kiel, Germany.

A. Koudsie

INSERM 318, Joseph Fourier University, Centre Hospitalier Universitaire, Pavillon B, BP 217,38043 Grenoble Cedex 9, France.

P. Krack

INSERM 318, Joseph Fourier University, Centre Hospitalier Universitaire, Pavillon B, BP 217,38043 Grenoble Cedex 9, France.

A.M. Lozano

Department of Surgery, Division of Neurosurgery, Toronto Western HospitallMC 2-433, 399 Bathurst Street, East Wing 6-528, Toronto, ON M5T 2S8, Canada.

C.H. Liicking

Neurologische Universitatsklinik, Neurozentrum, Breisacher Strasse 64, D-79106 Freiburg, Germany.

V.G. Macefield

Spinal Injuries Research Centre, Prince of Wales Medical Research Institute, University of New South Wales, Sydney, Australia.

M.-UManto

Charge de Recherches FNRS, Neurologie, 808 Route de Lennik, 1070 Brussels, Belgium.

F. Maugulere

Department of Functional Neurology and Epileptology, Neurological Hospital, 59 Boulevard Pinel, 69330 Lyon, France.

T.Mima

Department of Brain Pathophysiology, Human Brain Research Center, Kyoto University Graduate School of Medicine, Shogoin, Sakyo-ku, Kyoto 606-8507, Japan.

T. Nagamine

Human Brain Research Center, Kyoto University Graduate School of Medicine, Shogoin, Sakyo-ku, Kyoto 606-8507, Japan.

F. Pauri

AFaR, Dip. Neuroscienze, Osdpedale Fatebenefratelli, Isola Tiberina 39, 00186 Rome, Italy.

P.Poliak

INSERM 318, Joseph Fourier University, Centre Hospitalier Universitaire, Pavillon B, BP 217, 38043 Grenoble Cedex 9, France.

B.W. Ongerboer de Visser Department of Neurology/Clinical Neurophysiology Unit, Academic Medical Center, 1105 AZ Amsterdam, The Netherlands. P.M. Rossini

Direzione Scientifica AFaR, Associazione Fatebenefratelli per la Ricerca, Lungotevere degli Anguillara 12,00186 Rome, Italy.

J.C. Rothwell

Sobell Department, Institute of Neurology, Queen Square, London WCIN 3BG, UK.

H. Shibasaki

Human Brain Research Center and Department of Neurology, Kyoto University Graduate School of Medicine, Shogoin, Sakyo-ku, Kyoto 606-8507, Japan.

x

LIST OF CONTRIBUTORS

D.M. Simpson

Clinical Neurophysiology Laboratories, Mount Sinai Medical Center, 1 Gustave L. Levy Place, New York, NY 10029, USA.

L. Sudarsky

Department of Neurology, ASB 1-2, Brigham and Women's Hospital, 75 Francis Street, Boston, MA 02115, USA.

P.O. Thompson

Department of Neurology and University Department of Medicine, Royal Adelaide Hospital and University of Adelaide, Adelaide, SA 5000, Australia.

F.Valldeoriola

Unitat d'EMG, Servei de Neurologia, Hospital Clinic, Villarroel 170, Barcelona 08036, Spain.

J. Valls·Soh~

Unitat d'EMG, Servei de Neurologia, Hospital Clinic, Villarroel 170, Barcelona 08036, Spain.

J.E. Visser

Department of Neurology, 326, University Medical Center St. Radboud, P.O. Box 9101,6500 HB Nijmegen, The Netherlands.

J. Volkmann

Department of Neurology, Christian-Albrechts-Universitat Kiel, Niemannsweg 147, D-24105 Kiel, Germany.

U. Ziemann

Clinic of Neurology, Johann Wolfgang Goethe University of Frankfurt, TheodorStem-Kai 7, D-60590 Frankfurt am Main, Germany.

Movement Disorders Handbook of Clinical Neurophysiology, Vol. I M. Hallett (Ed.) © 2003 Elsevier B.V. All rights reserved

3 CHAPTER I

Movement disorders: overview Why have a book devoted to the Clinical Neurophysiology of movement disorders? A primary consideration is the growth of the field of movement disorders. Movement disorders is a relatively new field that grew out of an original interest in Parkinson's disease. The interest in Parkinson's disease itself blossomed after the finding that it could be well treated with levodopa. The field soon expanded to other disorders involving the basal ganglia, including dystonia and Huntington's disease. Myoclonus was added shortly after the finding that many cases were treatable with 5-hydroxytryptophan. Then other disorders were included where the motor system was impaired, and the field expanded to include the ataxias, spasticity and even paresis. The field got its name from Stan Fahn and C. David Marsden who helped popularize this way of organizing neurological disorders. The field has now been codified with the formation of an international society and journal. The common theme of movement disorders is the motor system and its diseases. A good deal of emphasis in the field has been placed on pharmacology as the mainline method of treatment. Recently, genetics and cell biology have been giving insights into the nature of the disease processes that cause movement disorders. Another critical area of interest has always been in the physiology and pathophysiology of the motor system, and this is the entry point for clinical neurophysiology. How is movement normally generated? What are the abnormalities underlying manifestations such as bradykinesia, tremor, chorea, tics, and myoclonus? Not only are these questions of interest by themselves, but the answers may point toward new therapeutic options. The first intersection of clinical neurophysiology and movement disorders is this research issue about the nature of motor disturbances. Clinical neurophysiology has always been a field that has contributed to the diagnosis of neurological disorders. As has been said often, it is an extension of the neurological examination. The second intersection of clinical neurophysiology and movement disorders is diagnosis. Clinical neurophysiologists

have traditionally been trained largely in EEG and EMG, focused largely on epilepsy and neuromuscular disorders, respectively. However, it is clear that clinical neurophysiology can contribute in other areas. In addition to movement disorders, for example, clinical neurophysiology can contribute to the fields of autonomic nervous system disorders, sleep disorders, and central nervous system monitoring during operations on the brain or spine. In movement disorders that look superficially similar, it is critical to make the right diagnosis because therapies might differ. Is a quick movement a tic, a myoclonic jerk or a voluntary movement? Studies of the surface EMG and the correlative EEG can give a definitive answer. Small differences in timing, easily measured with simple techniques, can be impossible to tell by eye. What is the burst duration of EMG underlying an involuntary movement? For example, is a myoclonic jerk due to a fragment of epilepsy or a fragment of a basal ganglia movement disorder? The EMG burst length is shorter in the former than the latter. What is the latency of a muscle jerk after a stimulus? Is it shorter than possible reaction time? If so, it cannot be voluntary or psychogenic. What is the frequency of tremor and how does it change with an intervention? Exaggerated physiological tremor should reduce in frequency with weighting of the limb by 1 or 2 Hz. This cannot be appreciated by visual inspection. Might there be two components of a tremor? Again, only physiological measurements will reveal this finding. Quantification of movement disorders is often useful for monitoring change over time including assessment of therapy. There are clinical scales that can be useful, but these are largely subjective and subtle changes over months or years might be missed. Physiological techniques can be valuable in this regard and can be used, for example, to monitor the amplitude of tremor or the magnitude of spastic tone. A new intersection of movement disorders and clinical neurophysiology is therapy. EMG guidance

4

improves the delivery of botulinum toxin to muscles with unwanted spasms. Neurophysiological monitoring is valuable in locating targets for deep brain stimulation. Transcranial magnetic stimulation is being explored for its utility in several movement disorders. This should be a valuable text for clinical neurophysiologists. The topic of movement disorders is usually treated superficially in even large textbooks of clinical neurophysiology. Useful clinical methods and research techniques are covered extensively. The book anticipates that the reader will have some basic knowledge of clinical neurophysiology, but then can be useful for the novice who wants an introduction and by the expert who is looking for details.

M. HALLETT

The book is arranged in two main parts. The first part deals with techniques. Here is where the reader can find exactly how to do a specific test and how to interpret the results. The application of the techniques to movement disorders is only briefly discussed. In the second part, individual movement disorders are the topics. Each chapter describes the disorder and its physiology, concentrating on the research and clinical methods that can be useful. It might be necessary when reading these chapters to refer back to the detailed technique in the first part of the book. At the end, there is a short look to the future with some guesses as to the direction of the field. Mark Hallett Bethesda, MD, USA June 2003

Movement Disorders Handbook of Clinical Neurophysiology, Vol. 1 M. Hallett (Ed.) © 2003 Elsevier B.V. All rights reserved

7 CHAPTER 2

Electromyography Mark Hallett* Human Motor Control Section. National Institute of Neurological Disorders and Stroke, NIH, Bethesda, MD 20892-/428. USA

Electromyography (EMG) is an important tool for the analysis of movement disorders (Hallett, 1999, 2000). Muscles are responsible for movements, and EMG is a direct measurement of the activity of muscle and a fairly direct measure of alpha motoneuron activity. EMG is most commonly used in clinical neurophysiology for the analysis of peripheral neuromuscular disease, answering the questions of whether there is a neuropathy, myopathy or neuromuscular junction disorder. In general, when dealing with movement disorders, it is assumed that all those aspects are normal, and the questions are different. In this context the questions asked include what muscles are active and what is the pattern of activation. The term kinesiological EMG is sometimes applied, and it is often apt, since the issue is what is the EMG responsible for the movement. Since numerous muscles act on any joint, it is typically necessary to record from at least two muscles with antagonist actions. The main information that can be extracted is the amplitude and the timing. EMG data can be measured with surface, needle, or wire electrodes (Hallett et al., 1994). Surface electrodes have the advantages that they are not painful and they record from a relatively large volume of muscle producing a good average of its activity. For these reasons, pairs of surface electrodes are ordinarily used for the analysis of movement disorders. The two surface electrodes are typically placed near the middle of the muscle belly about 2 or 3 em apart. Belly-tendon recording is often not optimal since recording volume would be too extensive and pickup would include unintended

* Dr. Mark Hallett, M.D., Human Motor Control Section, NIH, Building 10, Room 5N226, 10 Center Dr., MSC 1428, Bethesda, MD 20892-1428, USA. E-mail address: [email protected] Tel.: 3011496-9526; fax: 3011480-2286.

nearby muscles. Needle electrodes have the advantage that they are more selective, sometimes a necessity when recording from small or deep muscles. Traditional needle electrodes are stiff, however, and it is best to use them when recording from muscles during movements that are close to isometric. If there is substantial movement, needles will be very painful, in part because of the relative movement of the skin and muscle belly. Pairs of fine wire electrodes have the advantage of selectivity similar to that of needle electrodes and are flexible, permitting free movement with only minimal pain. There is slight movement of the wires with movement, but they do provide a reasonably stable recording. Regardless of the electrodes used, it is important to avoid movement artifact, which can contaminate the EMG signal. Wire movement should be limited. Low-frequency content of the EMG signal can be restricted with filtering, and this can remove much of the movement artifact. Movement artifact is largely in the range of DC to 10 or 20 Hz. Surface EMG has significant power in this range as well, but the peak power is at about 100 Hz, so the filtering of power below 20 Hz still leaves most of the EMG power. When surface electrodes are used, their impedance should be reduced to 10 kO or less. This will reduce electrogenesis at the electrode-skin interface caused by slight movements. In order to reduce the impedance sufficiently, it is usually necessary to abrade the skin (Fig. 1). If there is a question about the possible peripheral origin of the movement disorder, then needle EMG may be very helpful. For peripheral disorders, typically there are characteristic findings. For example, there might be fasciculations, myokymia, neuromyotonia or other high frequency bursts. The amplitude of EMG conveys information about the magnitude of the central nervous system output. For this purpose it is generally not useful to

M.HALLETT

8

A T R

C

• 2

,/·/'"\. ·

.

\

.

\

. , .1'' '\\, . .. .

~..r/"

· . \J'

. :r' 1)1 . .

.\ ' , . i/,~J·

4

)

r\....

i.,l·

.

Off

/,t\~

... r· .«. 500 juv 50.1. 5

,J~

Vl!I'ttv/'

T

.

,g

Amp

2

. Off

Fig. 1. Surface EMG recordings in a normal volunteer simulating a tremor. Muscles, from top down, are biceps, triceps, flexor carpi radialis and extensor carpi radialis. (A) The recordings are contaminated with movement artifact because the surface electrodes had high impedance. (B) The recordings here are much better after the skin was abraded and the impedance was reduced to less than 10 kD.

note the magnitude in mV. Since the electrodes are not in a "standard" position and there will be varying relationships between the skin surface and the muscle belly, the absolute measurement has ambiguous meaning. There are two ways of standardizing the measurements. One is in relation to maximal voluntary EMG output, and the other is with respect to a maximum compound muscle action potential (CMAP) produced by nerve stimulation. The former is more commonly used, and in many circumstances nerve stimulation would not be easy to do anyway.

EMG is occasionally used as a measure of force, but this should only be considered approximate since the relationship between EMG amplitude and force is not exact, may change in different circumstances and is often not linear (Perry and Bekey, 1981; Solomonow et al., 1986; Hof, 1997). In processing the EMG for amplitude it is often useful to rectify and smooth the recording (Fig. 2). This process yields an "envelope" of the EMG signal. The amplitude of the envelope can be used as the magnitude of the EMG, and for a burst of EMG,

ELECTROMYOGRAPHY

9

B

8

Off

Fig. 1. Continued.

the integrated area of the envelope is a good measure. If a tremor is being recorded, then the successive envelopes of EMG form a curve like a sinusoid, and the record can be subjected to frequency analysis to get information about the frequency content of the signals driving the tremor. The EMG patterns underlying voluntary movement are characteristic and vary with the speed of movement (Berardelli et al., 1996; Hallett, 1999, 2000) (Fig. 3). A slow, smooth movement is characterized chiefly by continuous activity in the agonist. A movement made as rapidly as possible, a so-called ballistic movement, has a triphasic pattern with a burst of activity in the agonist lasting 50-100

ms, a burst of activity in the antagonist lasting 50-100 ms, and return of activity in the agonist often in the form of a burst. In different disorders of voluntary movement, there are characteristic abnormalities. With cerebellar lesions, there is prolongation of the first agonist and/or antagonist EMG bursts. The prolongations can be marked, and there is a good correlation of the acceleration time of the movement with the duration of the first agonist burst. Unwanted prolongation of acceleration time should predispose to hypermetria. The antagonist burst can be delayed as well. With parkinsonian bradykinesia, there is abnormal patterning, with multiple bursts having the

10

M.HALLETT

A

500

IJ-V

Amp 2

B

100 IJ-V

Fig. 2. Surface EMG recordings in a normal volunteer simulating a tremor. (A) The top trace is the raw EMG from biceps and the bottom is the same trace rectified. (B) This is a record similar to that in (A), but the EMG has been smoothed as well as rectified.

Table I EMG appearance in different types of involuntary movements. Disorder

EMGpattem Reflex

Myoclonus

x

Examples/comment Ballistic

x

x

Tic Dystonia Chorea

X

X

(Modified from Hallett, 1999, with pennission.)

Tonic

x

Epileptic myoclonus Ballistic movement overflow myoclonus Dystonic myoclonus

x

Not fully involuntary

X

Also athetosis

X

Also dyskinesia, ballism

ELECTROMYOGRAPHY

A

BICEPS

TRIClPS

ARM

STEP POSITION

B BICEPS

TRICEPS

SUP

II

different types of movements. Specification of duration in the range of 30-300 ms merely by clinical inspection is virtually impossible due to the relative slowness of the mechanical events compared with the electrical events. Finally, antagonist muscle relationships can be specified as synchronous or asynchronous (reciprocal) by inspection of the EMG signal. In a tremor, asynchronous activity would be described as alternating (Fig. 4). There are three EMG patterns that may underlie involuntary movements (Figs. 5 and 6) (Hallett et al., 1987, 1994; Hallett, 1997, 1999,2000). One pattern, which can be called "reflex", resembles the burst occurring in many reflexes, including H-reflexes and stretch reflexes. The EMG burst duration is 10-30 ms, and EMG activity in the antagonist muscle is virtually always synchronous. Another pattern, which can be called "ballistic", resembles voluntary ballistic movements with a triphasic pattern; there is

AIlM POSITION

Fig. 3. EMG activity in biceps and triceps during (A) fast flexion of the elbow and (B) slow, smooth flexion. STEP indicates the target to be tracked and ARM POSITION is the actual elbow angle; reaction time and movement time information can be obtained from these records. The vertical calibration line corresponds to 500 mV for A and 20 mV for B. Modified from Hallett et al., 1975, with permission.

appearance of repetitrve cycles of the triphasic pattern to complete the movement. With dystonia and athetosis, there is excessive activity, including cocontraction activity, in the antagonist. Excessive activity also overflows into muscles not needed for the action. EMG burst length can be prolonged. In athetosis particularly, there are a variety of abnormal patterns of antagonist activity that appear to block the movement from occurring. Inspection of the EMG signal of an involuntary movement reveals, first, whether the movement is regular (usually a tremor) or irregular. There are sometimes surprises in such an analysis. Rhythmic EMG activity can appear irregular clinically if the amplitude varies; irregular EMG activity will sometimes appear rhythmic clinically if it is rapid. The duration of the EMG burst associated with an involuntary movement can also be measured; specific ranges of duration are associated with

A

EXT INn

B TIB ANT

GASTROC

Fig. 4. Recordings from pairs of antagonist muscles in different tremors. (A) Needle EMG recordings from the first lumbrical and the extensor indicis in a patient with essential tremor showing synchronous activation. (B) Surface EMG recordings in a patient with Parkinson's disease showing alternating activity in tibialis anteriorand gastrocnemius. From Sabra and Hallett, 1984, with permission.

12

Fig. 5. Comparison of (A) "reflex" and (B) "ballistic" EMG appearance underlying different types of myoclonus. Part A is from a patient with reticular reflex myoclonus, and part B is from a patient with ballistic movement overflow myoclonus. Vertical calibration is 1 mV for part A and 0.5 mV for part B. From Chadwick et al., 1977, with permission.

a burst of activity in the agonist muscle lasting 50-100 ms, a burst of activity in the antagonist muscle lasting 50-100 ms, and then return of activity in the agonist, often in the form of another burst. The third pattern, which can be called "tonic", resembles slow voluntary movements and is characterized by continuous or almost continuous EMG activity lasting for the duration of the movement, from 200-1000 ms or longer. Activity can be solely in the agonist muscle, or there can be some cocontraction of the antagonist muscle with the agonist.

M.HALLETT

Different types of myoclonus show one of the three types of patterns, and the EMG can be very helpful in making a diagnosis. Dystonia and athetosis show largely tonic patterns. Chorea is characterized by a wide variation of EMG burst durations encompassing all three patterns. In tic, there can be ballistic or tonic patterns. These data are summarized in Table 1. There can also be a brief lapse in tonic innervation that is clinically called asterixis or negative myoclonus (Shibasaki, 1995). Clinically, it appears as an involuntary jerk superimposed on a postural or intentional movement. Careful observation often reveals that the jerk is in the direction of gravity, but this can be difficult since the lapse is frequently followed by a quick compensatory antigravity movement to restore limb position. The involuntary movement is usually irregular, but when asterixis comes rapidly there may be the appearance of tremor. EMG analysis shows characteristic synchronous pauses in antagonist muscles (Fig. 7). In clinical practice, it is of course valuable to couple EMG studies with either kinesiologic or EEG observations or both. Nevertheless, the simple application of EMG can be extremely helpful as a first step.

Acknowledgment This review includes sections updated from earlier chapters (Hallett, 1999, 2000). Work of the U.S. government, it has no copyright.

Fig. 6. EMG recordings from a patient with focal hand dystonia when attempting hand writing. Recordings are from 4 muscles in the right arm during motor performance. From Cohen and Hallett, 1988, with permission.

ELECTROMYOGRAPHY

13

Fig. 7. EMG and accelerometric recording of asterixis. EMG is from flexors and extensors of the wrist and accelerometer was on the dorsum of the hand. From Hallett, 1999, with permission.

References Berardelli, A, Hallett, M, Rothwell, JC, Agostino, R, Manfredi, M and Thompson, PD et al. (1996) Review Article. Single-joint rapid arm movements in normal subjects and in patients with motor disorders. Brain, 119: 661-674. Chadwick, D, Hallett, M, Harris, R, Jenner, P, Reynolds, EH and Marsden, CD (1977) Clinical, biochemical, and physiological features distinguishing myoclonus responsive to 5-hydroxytryptophan, tryptophan with a monoamine oxidase inhibitor, and clonazepam. Brain, 100: 455-487. Cohen, LG and Hallett, M (1988) Hand cramps: clinical features and electromyographic patterns in a focal dystonia. Neurology, 38: 1005-1012. Hallett, M (1997) Myoclonus and myoclonic syndromes. In: JJ Engel and TA Pedley (Eds.), Epilepsy: A Comprehensive Textbook. Lippincott-Raven, Philadelphia, pp. 2717-2723. Hallett, M (1999) Electrophysiologic evaluation of movement disorders. In: MJ Arninoff (Ed.), Electrodiagnosis in Clinical Neurology. Churchill Livingstone, New York,pp.365-380. Hallett, M (2000) Electrodiagnosis in movement disorders. In: KH Levin and HO Liiders (Eds.), Comprehensive Clinical Neurophysiology. W.B. Saunders Company, Philadelphia, pp. 281-294.

Hallett, M, Shahani, BT and Young, RR (1975) EMG analysis of stereotyped voluntary movements in man. J. Neurol. Neurosurg. Psychiatry, 38: 1154-1162. Hallett, M, Marsden, CD and Fahn, S (1987) Myoclonus (Chapter 37). In: PJ Vinken, GW Bruyn and HL Klawans (Eds.), Handbook of Clinical Neurology. Elsevier Science Publishers, Amsterdam, pp. 609-625. Hallett, M, Berardelli, A, Delwaide, P, Freund, H-J, Kimura, J and Lucking, C et al. (1994) Central EMG and tests of motor control. Report of an IFCN Committee. Electroencephalogr. Clin. Neurophysiol., 90, 404-432. Hof, AL (1997) The relationship between electromyogram and muscle force. Sportverletz Sportschaden, 11, 79-86. Perry, J and Bekey, GA (1981) EMG-force relationships in skeletal muscle. Crit. Rev. Biomed. Eng., 7: 1-22. Sabra, AF and Hallett, M (1984) Action tremor with alternating activity in antagonist muscles. Neurology, 34: 151-156. Shibasaki, H (1995) Pathophysiology of negative myoclonus and asterixis. In: S Fahn, M Hallett, HO Luders and CD Marsden (Eds.), Negative Motor Phenomena. Lippincott-Raven Publishers, Philadelphia, pp. 199209. Solomonow, M, Baratta, R, Zhou, BH, Shoji, H and D'Ambrosia, R (1986) Historical update and new developments on the EMG-force relationships of skeletal muscles. Orthopedics, 9: 1541-1543.

Movement Disorders Handbook of Clinical Neurophysiology, Vol. I

M. Hallett (Ed.) © ZOO3 Elsevier B.V. All rights reserved

15 CHAPTER 3

EEG (MEG)/EMG correlation Hiroshi Shibasaki'<" and Takashi Nagamine" a

Human Brain Research Center and b Department of Neurology, Kyoto University Graduate School of Medicine, Shogoin, Sakyo-ku, Kyoto 606-8507, Japan

3.1. Introduction The cerebral cortex is involved in the control of voluntary movements to various extent depending on the kind of movement. In involuntary movements such as myoclonus and dystonia, the sensori-motor cortex in particular is involved either directly or indirectly in the pathophysiology. The functional relation between the cortical activities and the movements, whether voluntary or involuntary, can be studied non-invasively by using the functional neuroimaging techniques such as functional magnetic resonance imaging (fMRI) and positron emission tomography (PET), and with various electrophysiological methods. Among the latter techniques, the simultaneous recording of electroencephalogram (EEG) or magnetoencephalogram (MEG) with electromyograms (EMG) provides useful information about the temporal and spatial relation between the cortical activities and those movements (Shibasaki and Rothwell, 1999). Subcortical structures, especially basal ganglia and cerebellum, are extremely important for the control of voluntary motor execution and also in relation to the pathogenesis of certain involuntary movements. By using EEG or MEG, however, the functional states of these subcortical structures can only be estimated by measuring the cortical activities that might be influenced by the functional abnormalities of these subcortical structures. This is in strong contrast with neuroimaging techniques which enable us to visualize the activated areas, if any, also in these deep structures.

* Dr. Hiroshi Shibasaki, M.D., Ph.D., Department of Neurology, Kyoto University Graduate School of Medicine, Shogoin, Sakyo-ku, Kyoto 606-8507, Japan. E-mail address:[email protected] Tel.: +81-75-751-3770; fax: +81-75-761-9780.

Three kinds of electrophysiological methods can be used for analyzing the relation between the cortical electric activities and the movements: (1) recording of electric or magnetic fields associated with the movement; (2) analysis of change of cortical rhythmic oscillations related to the movement; and (3) coherence analysis between EEG or MEG and EMG. As the cortico-muscular coherence is dealt with in another chapter in this book, the first two techniques are described in this chapter.

3.2. EEG versus MEG With EEG and MEG, essentially the same neuronal activities can be recorded, namely the post-synaptic potentials of apical dendrites of large pyramidal neurons in the cerebral cortex. Theoretically EEG is formed by electric potential fields resulting from extra-cellular current of the apical dendrites whereas MEG is a magnetic field formed by the intra-cellular current generating the magnetic flow occurring around the apical dendrite. As the majority of those apical dendrites are oriented at a right angle to the cortical surface, and as the conventional MEG machine adopts sensors oriented parallel to the skull surface, only the current flow of apical dendrites tangentially oriented with respect to the head surface can be recorded by MEG. In other words, MEG is much more sensitive to the activities arising from the cortical sulci as compared to those arising from the crown of cortical gyri. This is in strong contrast with EEG that is derived from a summation of both tangentially and radially oriented current flows. Although both EEG and MEG have to depend on the estimation of the current sources by solving the inverse problem, the biggest advantage of MEG is its high spatial resolution. This is mainly because MEG is not affected by the 'shunt effect' caused by different electrical conductivity of the structures surrounding the brain surface such as

16

spinal fluid, skull and subcutaneous tissues whereas the EEG field is strongly influenced by the 'shunt effect'. Therefore, MEG provides more accurate information about the source of cortical electric activities as compared with EEG. 3.3. Electric or magnetic fields associated with movements 3.3.1. Self-initiated movements 3.3.1.1. Principle Among various methods used for the investigation of self-initiated movements, the EEG (MEG)/EMG correlation is studied mainly for understanding the cortical mechanisms underlying the central motor control and its disorders. Since the movementrelated cortical potentials (MRCP) (signal) are usually much smaller as compared to the background EEG or MEG activities (noise in this case), they cannot be identified just by visual inspection of the raw record, even if they might occur in close time relation to the movement. Therefore, the method for averaging the EEG or MEG with respect to the time of each movement is employed for increasing the signal-to-noise ratio. In contrast to the study of sensory or perceptive mechanisms, an emphasis is often placed on the epoch before the movement onset, which makes it necessary to average the data backward with respect to the actual fiducial time point; the movement onset in most cases. 3.3.1.2. Method of recording For recording MRCP, EEG and/or MEG and EMG are simultaneously recorded while the subject repeats a voluntary movement at a self-paced rate of once every 5 s or longer (Shibasaki, 1993). Any kind of movement can be studied, but simple movements such as finger extension, wrist extension, tongue protrusion and foot extension are commonly employed. MRCP can also be recorded in association with more natural movements such as gait initiation (Vidailhet et al., 1993) and chewing (Yoshida et al., 2000). Magnitude and distribution of MRCP is slightly different depending on the kind of movements. Usually a complex movement is associated with a larger pre-movement negative slope than a simple one (Kitamura et al., 1993b), and a discrete movement of an isolated finger produces a larger slope than the simultaneous movement of two fingers (Kitamura et al., 1993a).

H. SHIBASAKI AND T. NAGAMINE

EEG electrodes should be placed at least over the central areas, approximately C3 and C4 for hand movement and Cz for foot movement according to the International 10-20 system. Use of additional electrodes, if possible, is helpful not only for studying the distribution of potentials over the head, but also, perhaps more importantly, for identifying each subcomponent of MRCP. Linked ear electrodes are commonly used as the reference for common referential derivation. Electrode impedance must be kept below 5 ko. It is important to use a long time constant for recording the slow components, and thus the filter setting of amplifiers commonly used is 0.05-120 Hz (-3 dB). EMG is recorded by a pair of disk electrodes placed over the contracting muscle with a filter setting of 30-120 Hz, and rectified. Accordingly, sampling rate of 400 Hz or higher is preferable. A fiducial point for averaging is determined when the rectified EMG exceeds a preset threshold for an on-line analysis, but an off-line analysis by visually determining the precise onset of each EMG discharge as well as by excluding the trials containing various artifacts gives rise to better waveforms (Barrett et al., 1985). The analysis window should preferably cover 3.0 s before and 0.5 s after the movement onset. For monitoring blink artifacts, electro-oculograms (EOGs) should be simultaneously recorded and averaged by using the same method as used for processing the EEG. Two or more sessions should be repeated for each kind of movement in order to confirm reproducibility of the waveform, and each session commonly consists of about 50 trials. 3.3.1.3. Normalfindings In normal subjects, self-initiated finger movement is preceded by a slow negative potential shift starting at about 2.0 s before the movement onset (Fig. 1). This early component called either Bereitschaftspotential (BP), or the early component of readiness potential (RP) or NS1, is maximal at the midline central region and distributed widely and symmetrically over the scalp whatever movement is employed as the task (Shibasaki et al., 1980). Approximately 400 ms before the movement onset, a sharper negative slope, called Negative Slope (NS'), the late component of RP or NS2, appears over the central region contralateral to the movement (Cl for the right and C2 for the left hand movement). Immediately before the movement, a small negative

17

EEG (MEG)/EMG CORRELATION

Vol. Lt Wrist Extension Ref:A1A2

BP

BP

F3 F1

Fz

F2 F4 C3

C1

Cz C2 C4 P3 P1 pz P2 P4 EOG EMG

-"-~+--~~~~~~::b'6"'--fO';;""-~

NS' _~=:..o!::~~~:::::::::~:::+

1 2 0.uv

-~::=-o~~~~~~:::-+A:-=;~~'1:M'

+

-~-----""""",,,------'r--~~~~~ ---+------~~-+--

(n=98)

SOOms

Fig. 1. Movement-related cortical potentials (MRCP) recorded from a normal subject and schematic illustration of the scalp distribution of Bereitschaftspotential (BP) and negative slope (NS'). MP: motor potential. Self-paced, voluntary (Vol.) left (Lt) wrist extension. Average of 98 trials.

potential is identified in some subjects, which is localized to the contralateral central electrode (motor potential, MP). This potential continues to increase in negativity and moves anteriorly and to the midline where it peaks about 100 ms after the movement onset. According to the results of direct cortical recording of these potentials with subdurally implanted electrodes, the BP preceding unilateral hand movement mainly arises from activities in the hand area of the primary sensori-motor cortex (S1M!) and the supplementary motor areas (SMAs) both bilaterally, and the NS' reflects activities arising from the same areas but with much more contralateral predominance (Ikeda and Shibasaki, 1992; Ikeda et al., 1992). During the early pre-movement period, the averaged magnetic field is rather silent while BP is observed on the averaged EEG, and in the later period while NS' is seen bilaterally with the

contralateral predominance, the magnetic field is seen almost exclusively at the contralateral central region (Nagamine et al., 1996). In view of the fact that MEG is almost exclusively composed of tangentially oriented current sources (see 3.2), it can be postulated that initially the crown of the precentral gyrus (probably the premotor area for hand movements) is activated bilaterally, and the anterior bank of the contralateral central sulcus (area 4) is activated closer to the movement onset. How much the activities of SMA contribute to the scalprecorded BPINS' has not yet been clarified. In the recent subdural recording in epilepsy patients, it has been shown that not only the SMA proper but also the rostral part of SMA, probably corresponding to the pre-SMA of monkey, produces BPINS' (Kunieda et al., 2000). The SMA, especially the pre-SMA, was significantly less active with the rapid rate of movement repetition (2 Hz) than with the slower rate

18

while the Sl-MI remained active regardless of the movement rate (Kunieda et al., 2000). 3.3.1.4. Muscle relaxation MRCP is usually recorded by employing the movements caused by muscle contraction, but the negative movement caused by voluntary termination of muscle contraction, or muscle relaxation, can be also subjected to the MRCP recording. For this purpose, the onset of the negative movement can be identified by visual inspection of the silent period of EMG discharges or by an accelerometer and is used as a fiducial point for averaging the simultaneously recorded EEG or MEG. BPINS' preceding the movement induced by muscle relaxation appears quite similar to the one preceding the muscle contraction (Terada et al., 1995, 1999). In isometric muscle relaxation, however, a smaller BPINS' was observed in the lateral central area as compared with muscle contraction (Rothwell et al., 1998). Ikeda et al. (2000) reported, based on the presurgical study of medically intractable epilepsy patients, that stimulation of some parts of the SI-Ml with a single electric shock produced a silent period in the on-going EMG discharges without any preceding motor evoked potential (MEP). In these areas, the stronger the stimulus intensity, the longer was the induced silent period. This finding suggests the presence of inhibitory areas in human SI-Ml. By contrast, Liiders et al. (1995) proposed two distinct negative motor areas in human also based on the study of epilepsy patients; one rostral to the face area of Ml at the lateral frontal convexity (primary negative motor area, PNMA) and the other rostral to SMA (supplementary negative motor area, SNMA). Stimulation of these areas with high frequency (50 Hz) pulses was found to interfere with repetitive finger movements and sustained muscle contraction. BPINS' can be recorded from the SNMA not just before self-paced muscle relaxation but also before muscle contraction (Yazawa et aI., 1998). Furthermore, the SNMA was found to generate BP regardless of the site of movement (Yazawa et aI., 2000). Thus, the mechanism of voluntary motor inhibition remains to be elucidated further. 3.3.1.5. Clinical application MRCP can be applied to the investigation of pathophysiology of primary as well as non-primary motor cortices in various kinds of movement dis-

H. SHIBASAKI AND T. NAGAMINE

orders. For example, BPINS' is often absent in patients with cerebellar lesions, especially those involving the dentate nucleus and its efferent pathway or the dentato-thalamo-cortical pathway (Shibasaki et al., 1986; Ikeda et al., 1994; Kitamura et aI., 1999). BP was reported to be smaller in patients with Parkinson's disease than in agematched control subjects while NS' was normal or even larger in the patient group, and administration of L-DOPA to these patients was shown to increase the amplitude of BP along with the clinical improvement (Dick et aI., 1989). NS' was reported to be smaller than normal at the contralateral central region in patients with writer's cramp; a kind of focal dystonia (Deuschl et aI., 1995). Yazawa et aI. (1999) found that, in patients with writer's cramp, the BPINS' preceding self-initiated muscle relaxation of the right hand was smaller at the left central area whereas that preceding muscle contraction did not show significant difference from the normal control. It was interpreted to indicate dysfunction of the inhibitory system in the motor cortex in focal hand dystonia. MRCP can also be applied to the clinical study of the mechanism underlying motor recovery of a paralyzed limb. Investigating two patients, who recovered from hemiparesis due to stroke, by using the blood flow activation study with PET and by recording MRCP both associated with hand movement, Honda et aI. (1997) showed evidence suggesting a participation of the ipsilateral hemisphere to motor recovery. 3.3.1.6. Steady-state movement-related fields Recording of MRCP or movement-related magnetic fields with long inter-trial intervals requires strong motivation and cooperation of the subjects, which makes it difficult to apply this technique to patients with various movement disorders. In such cases, fast repetitive movements, at a rate of 2 Hz for example, will provide enough information at least as to the location of the precentral and postcentral gyrus (Fig. 2) (Gerloff et aI., 1998). 3.3.2. Cued movements

In a reaction time paradigm in which a pair of stimuli (S1 and S2) are presented with a fixed interstimulus interval of, for example, 2.0 s, and the subject is asked to make a motor task soon after S2,

19

EEG (MEG)/EMG CORRELATION

2 Hz movement

Fig. 2. (a) Steady-state movement-related magnetic fields associated with self-paced abduction of the right thumb in a normal subject. The first component (motor field, MF; asterisk) peaks immediately after the EMG onset (vertical line), and the second component (post-movement field, post-MF; closed circle) peaks after the EMG peak. The analysis window covers 300 ms before and 200 ms after the EMG onset. (b) Co-registration of equivalent current dipole (ECD) solutions with anatomical MRls of two subjects. Self-paced abduction of the right thumb. The ECD for the MF (red) is located in the precentral gyrus and that for the post-MF (yellow) is in the postcentral gyrus. White arrows indicate the left central sulcus. (Cited from Gerloff et a\., 1998 with permission).

a slowly rising negative potential shift called contingent negative variation (CNV) is recorded between the two stimuli. By subdural recording, the late component of CNV was found to be generated, in addition to the SMA and SI-Ml, in many cortical areas including the prefrontal cortex (Ikeda et aI., 1996; Hamano et aI., 1997). The late CNV was

found to be abnormal especially in patients with basal ganglia disorders such as Parkinson's disease and focal dystonia while it was well preserved in patients with cerebellar disorders (Ikeda et aI., 1994; Kaji et aI., 1995; Ikeda et aI., 1997). CNV can be also applied to the investigation of daily movement like gait initiation. In this case, gait initiation is

20

subject 1

H. SHIBASAKI AND T. NAGAMINE

subject 5

Fig. 2. Continued.

employed as the motor task immediately after the S2. By this method, Yazawa et al. (1997) found a larger amplitude of the late CNV in gait initiation compared with simple foot dorsiflexion employed as a control task. This observation probably suggests a greater participation of SMA in gait initiation than in the simple foot movement.

3.3.3. Involuntary movements 3.3.3.1. EEG (MEG)/EMG polygraphic recording The use of EEG (MEG)/EMG correlation for the study of involuntary movements serves as a supplement to clinical diagnosis as well as a method for clarifying the physiological mechanism underlying

EEG (MEG)/EMG CORRELATION

the generation of each involuntary movement. The most fundamental method for studying EMG-EEG correlation in spontaneous involuntary movements like myoclonus is a polygraphic recording of surface EMG and EEG, by using either an electroencephalograph or a cathode ray oscillograph. The same electrode placement as used for the recording of MRCP is applicable to this study. Filter setting of amplifiers used for recording EEG is usually 1-300 Hz, and that for EMG is 30-300 Hz. By contrast to MRCP, higher sampling rate such as 1000 Hz is preferable. Large EEG activities like periodic synchronous discharges seen in association with periodic myoclonus in patients with CreutzfeldtJakob disease or subacute sclerosing panencephalitis are easily identified on the conventional polygraphic records without averaging (Oga et aI., 2000). 3.3.3.2. Jerk-locked back averaging On the conventional polygraph, however, it is usually difficult to study a precise relation between the EMG and EEG activities. In such cases, EEG or MEG can be back averaged with respect to the onset of the EMG discharge associated with involuntary movements by adopting the same principle as used for recording the MRCP (jerk-locked back averaging) (Shibasaki and Kuroiwa, 1975; Shibasaki, 1993, 2000). The analysis window commonly used for the investigation of myoclonus is from 400 ms before to 200 ms after the myoclonus onset. By this method, the temporal and spatial relation between myoclonus and the EEG or MEG activity can be studied more precisely. In cases of cortical reflex myoclonus, a myoclonic jerk of the hand is preceded by a positivenegative cortical spike by about 20 ms and the latter is localized to the hand area of the contralateral central region. As for the generator site of the myoclonus-related cortical activity, the jerk-locked averaging of MEG confirmed the source in the contralateral precentral gyrus corresponding to the involved muscle (Figs. 3 and 4) (Mirna et aI., 1998a). Even in cases where the conventional polygraph fails to show EEG spikes in association with myoclonus, this technique may disclose a spike preceding the myoclonus if that is of cortical origin. Furthermore, in some cases the jerk-locked averaging of MEG may detect the myoclonus-related cortical activity that is not detectable on the jerklocked average of EEG (Mirna et al., I998a).

21

Furthermore, if the aim of the study is to distinguish psychogenic movement disorders such as psychogenic myoclonus from "organic" involuntary movements, the use of low frequency filter of approximately 0.05 Hz is necessary, so that slow activity like BPINS' could be recorded to see if there was any such activity preceding the movement in question (Terada et aI., 1995).

Fig. 3. Myoclonus-related magnetic fields obtained by jerk-locked back averaging in a patient with cortical myoclonus. (a) Waveforms from two different blocks of averages are superimposed. The vertical line corresponds to the onset of myoclonic EMG discharge recorded from the left wrist extensor muscle. Biphasic activity preceding the myoclonus is localized to the right central area. (b) Magnified waveform from a selected channel (rectangle in a). (c) Contour map of the earliest peak preceding the myoclonus onset. The white arrow indicates the site of the ECD and its direction. (Cited from Mirna T et al., 1998a with permission).

22

H. SHIBASAKI AND T. NAGAMINE

Fig. 4. ECDs of myoclonus-related cortical activities superimposed on the axial MRI in a patient with cortical reflex myoclonus. Yellow spot shows ECD of the initial negative peak (N20m) of somatosensory evoked magnetic field in response to electric stimulation of the left median nerve at wrist, and the blue spot the following peak that is abnormally large (P25m). The ECD of the main peak of the pre-myoclonus activity demonstrated by jerk-locked back averaging with respect to the spontaneous myoclonus of the left hand is shown in red. Green arrowhead indicates the right central sulcus. (Recorded by Mirna T).

3.3.3.3. Cortical reflex myoclonus Some involuntary movements, especially myoclonus, are often induced by external stimuli, thus called stimulus-sensitive or reflex myoclonus. Stimulus-sensitive myoclonus can be studied by the

simultaneous recording of the short latency cortical evoked potentials and the transcortical long loop reflexes. Cortical reflex myoclonus is the best clinical indication for this study. The most appropriate modality of stimulus should be chosen

EEG (MEG)/EMG CORRELATION

depending on the specificity of stimulus sensitivity in each individual case, but electric stimulation of the peripheral nerve is most commonly used for recording the somatosensory evoked potentials (SEP) and the long loop reflex simultaneously. EMG is recorded from a pair of electrodes placed over the corresponding muscles of the stimulated or other extremities, and rectified. Then the rectified EMG and the simultaneously recorded EEG or MEG are averaged by using the stimulus onset as a trigger. In cortical reflex myoclonus, the EMG response of a hand muscle at a latency of about 45 ms is extremely enhanced (C reflex), which corresponds to an evoked myoclonic jerk itself. Cortical reflex myoclonus often shows a rapid spread of the reflex myoclonus from the proximal to distal muscles within the stimulated limb and also to the muscles of the contralateral limb, the latter most likely being mediated by transcallosal conduction (Brown et aI., 1991). In most cases of cortical reflex myoclonus, early cortical components of SEP, except for the initial peak (N20), are extremely enhanced (giant SEP). By using somatosensory evoked magnetic fields, Mima et al. (I 998b ) identified the source of the enhanced P25 component in the precentral gyrus, just near the area where the cortical activity preceding the spontaneous myoclonus occurred (Fig. 4). 3.3.3.4. Negative myoclonus Negative myoclonus (asterixis) of cortical origin can be studied also by using the back averaging method. In this case, the onset of the EMG silent period is used as the fiducial point for averaging (silent period-locked averaging) (Ugawa et aI., 1989). The precise origin of the cortical negative myoclonus has not been determined, but some of them are sensitive to the somatosensory stimulus just like cortical reflex positive myoclonus (cortical reflex negative myoclonus) (Shibasaki et aI., 1994).

3.4. Event-related desynchronization and synchronization 3.4.1. Background EEG and MEG signals are composed of activities belonging to several frequency bands, and some of

23 them contain strong rhythmic oscillations. Oscillations of different frequencies are thought to be produced by activities arising from different physiological circuits within the brain. Oscillation of brain activity, such as the mu rhythm, has been estimated to have functional significance from its topographical distribution and its characteristic response to external stimuli. However, temporal profile of the change in response to external stimuli or in association with internal drive for action cannot be obtained by the mere inspection of the raw activity. In order to detect relations between discrete EMG events and alteration of oscillatory activities in the EEG/MEG, the technique of event-related desynchronization/synchronization (ERD/ERS) can be used. For extracting the desired activity embedded in the background activity, the averaging technique has been widely adopted, in which a specific event such as EMG onset is used as the fiducial point for averaging. However, this technique is applicable only when the desired activity (signal) is timelocked as well as phase-locked to each event. Since the palarity of oscillatory activity is not necessarily phase-locked among different epochs, it has to be expressed as the change of net power of the activity (event-induced activity) (Lopes da Silva and Pfurtscheller, 1999). During the resting state, electric activity within a local area shares a common phase of oscillation, like a resonance, which is called an idling state. Once the local area becomes activated, each electric activity within the restricted area becomes detached from the resonant state and behaves out of phase to various extents. Due to the phase cancellation of the electrical activity within the local area in spite of the maintenance of the frequency range, the net power of the electric events within the local area is reduced (desynchronization). When the activation ceases, oscillation of the circuit is reinforced, sometimes resulting in transient hyperactivity (hypersynchronization), and finally returns to the original state. Alteration of oscillatory activity partly reflects a non-specific general reaction, but it also reflects a more specific function of the restricted cortical area. Among them, ERD/ERS in the sensorimotor area has been shown to have distinct functional relevance with motor preparation and execution.

24

H. SHIBASAKl AND T. NAGAMINE

3.4.2. Method of recording

3.4.3. Normal findings

For investigating ERDIERS in relation to voluntary movements, EEG/MEG and EMG signals are usually recorded continuously and stored on a computer while the subject repeats a voluntary movement similar to the recording of the movementrelated slow potentials. An interval of 4 s or longer is preferable to detect pre-movement dampening. These signals are bandpass filtered depending on the frequency range of interest by digital filtering, and then squared (Pfurtscheller and Aranibar, 1977) or rectified (Salmelin and Hari, 1994) in order to avoid phase cancellation at the time of averaging across epochs. Manual or automatic methods are then used to identify particular events in the EMG that can be used as fiducial points about which averaging can be performed with certain duration of time window. Instead of squaring or rectification, the envelope of the signal can also be used (Ciochon et aI., 1996). In order to normalize the different power of the resting activity, averaged responses are often expressed as the relative magnitude to that of the initial segment of the analysis time window. As for motor function, since 10Hz and 20 Hz frequency ranges are considered to reflect different functions, most analyses utilize bandpass filters centering around each of these two frequency ranges (Pfurtscheller and Klimesch, 1991). In some special conditions, 40 Hz activity also depicts the motor related modulation. Digital methods are used to filter the EEG. In addition to the center frequency of bandpass filter, the filter width is another concern. Since a narrower frequency width requires a longer analysis window, frequency resolution and temporal resolution have reciprocal interaction. Derived from the compromise between these two factors, 2 to 4 Hz are recommended as the filter width for motor function (Salmelin and Hamalainen, 1995; Pfurtscheller and Lopes da Silva, 1999). Pfurtscheller and Lopes da Silva (1999) recommend dividing the alpha frequency range into two, low alpha and high alpha. Another way to delineate the temporal evolution of oscillations is the use of the power spectrum of short-time periods. In this case, since four dimensional data of time, frequency, power and location is obtained, the conventional display reduces the results into three dimension by selecting a certain location and shows power by color coding (Baker et aI., 1997).

A typical use of ERDIERS techniques is to demonstrate the attenuation of the 10Hz rhythm (mu or alpha rhythm) preceding and accompanying voluntary movement, which is commonly followed by the rebound of the 20 Hz rhythm (beta rhythm). Usually the 10 Hz rhythm starts dampening (ERD) around 2 s prior to the movement onset over the sensorimotor area contralateral to the movement. The ERD continues during the ongoing muscle contraction and recovers to the original resting state I s after the cessation of the movement. The ipsilateral sensorimotor area also demonstrates 10Hz ERD with later initiation compared to the contralateral side. The 20 Hz rhythm shows later initiation of dampening compared to the 10Hz ERD, actually starting around 1 s prior to the movement onset. It also shows rebound peaking around 1 s after the cessation of movement (ERS). As for the spatial distribution, the 10 Hz ERD has bilateral sensorimotor foci around the movement onset. In contrast, the 20 Hz ERD is usually more localized and located anterior to that of the 10Hz ERD with contralateral predominance. The focus of 20 Hz ERD corresponds to the respective cortical representation of the movements (Salmelin et al., 1995; Pfurtscheller et aI., 1997). This distribution is in good conformity with the results of subdural recording of ERDIERS, which is restricted to the contralateral side and located anterior to that of the 10 Hz ERD (Crone et al., 1998; Ohara et aI., 2000). In MEG studies, similar reactivity patterns of 10 Hz and 20 Hz rhythms are observed (Fig. 5). Contralateral predominance of 20 Hz ERDIERS of MEG is more conspicuous compared to that of EEG (Nagamine et al., 1996). The 20 Hz ERDIERS also discloses somatotopic representation and anterior localization compared to the 10Hz ERD (Salmelin et aI., 1995). The 10Hz ERD in MEG starts as early as 3 to 4 s prior to the movement onset, whereas the readiness field of MEG starts around 1 s before the movement initiation (Fig. 5). This is in clear contrast to the relation of MRCP with ERDIERS in EEG (Nagamine et aI., 1996). The 20 Hz ERD in MEG is seen a few seconds before the actual event in both muscle contraction and muscle relaxation, which suggests participation of similar preparatory mechanisms in these two kinds of movements (Toma et al., 2000).

25

EEG (MEG)/EMG CORRELATION

Somatomotor area

Rt

Lt

18·"

Slow

We~~netic

~±

+ -i:RF

'».

MF

MEFI

-4

o

j

100fT/em

0 -100

4 sec

Fig. 5. Event-related desynchronization (ERD) and eventrelated synchronization (ERS) in association with self-initiated movement in comparison with slow magnetic fields (movement-related cortical magnetic fields) in a normal subject. Channels corresponding to the sensorimotor areas from each hemisphere are shown. Magnetic fields were recorded by a first-order planar gradiometer that detects the largest signal just above the generator currents. Average of 74 trials of right index finger extension.

Comparison between the slow potential and ERD was performed in contingent negative variation task (Filipovic et al., 200 I). Contrary to the almost zero reaction in the no-go condition during the preparatory phase between S I (go/no-go) and S2, alphaERD in no-go condition showed dampening to similar degree to that in the go condition.

3.4.4. Clinical application Since ERDIERS cannot provide high spatial resolution comparable to multichannel recording of slow potential fields due to methodological limitations, it is mainly applied for the investigation of pathophysiology of abnormal motor function. Focusing on akinesia in patients with Parkinson's disease, self-initiated unilateral hand movement has been analyzed, mainly focusing on pre-movement ERD of the mu rhythm. Comparison between the age-matched controls and chronic Parkinson patients

under medication demonstrated several differences. Delayed initiation of contralateral ERD of the mu rhythm, increased ipsilateral ERD of the mu rhythm with similar onset latency, and delayed bilateralization were the main features seen in Parkinson patients (Defebvre et al., 1993, 1994). Further studies evaluated untreated akinetic Parkinson patients and the influence of chronic administration of L-DOPA. The delayed contralateral ERD of the mu rhythm was also noted in patients of early stages, and ERD occurred earlier after chronic administration of L-DOPA (Defebvre et al., 1998). A similar effect was also noted after acute administration of LDOPA in patients under chronic L-DOPA treatment (Magnani et aI., 1999). ERS of the beta rhythm induced by simple voluntary movement was investigated in patients with Parkinson's disease, and the ERS was found to be reduced compared to normal subjects. Another study also showed higher peak frequency of ERS in patients with Parkinson's disease compared to that in normal subjects (Pfurtscheller et al., 1998). Another experiment using contingent negative variation (CNV) has focused on the preparatory period. Similar to the ERD associated with simple self-initiated movement, contralateral alpha band ERD started around 1.5 s prior to the movement triggered by the go signal. In Parkinson patients, this contralateral ERD of the alpha band started later, similar to the latency of the self-initiated movement. This indicates that an external cue does not ameliorate the late onset ERD of the alpha band in Parkinson's disease (Magnani et al., 1998). Investigation of ERD of the mu rhythm in patients with progressive supranuclear palsy (PSP) suggested more severely affected motor programming than in patients with Parkinson's disease. In case of selfinitiated movement, initiation of ERD of the mu rhythm over the contralateral sensorimotor area was delayed (around 0.4 s) in PSP patients compared to Parkinson patients (I s) and normals (1.7 s). Bilateralization of ERD was similar to that of Parkinson patients (Defebvre et al., 1999). In patients with focal dystonia, Toro et aI. (2000) found reduced 20-30 Hz ERD over the contralateral motor area in movements of both affected and unaffected hands. This reduction was prominent both before and after the movement. This finding supports the notion of abnormal command at the cortical level in dystonic patients.

26

Reorganization of the motor cortex was investigated in patients with ischemic supratentorial stroke during self-initiated movement (Platz et al., 2000). There was increased beta-ERD in patients suffering from central arm paresis and reduced alpha-ERD and beta-ERD in patients with somatosensory deficits. Patients with ideomotor apraxia showed reduced beta-ERD during movement preparation. Abnormal ERD was also demonstrated in patients with focal motor seizures (Derambure et al., 1997). Contralateral ERD of the mu rhythm accompanying self-initiated movement was delayed in patients with frequent motor seizures compared to normals and patients with temporal lobe seizures, indicating functional alterations of the motor areas in those patients. Related to the development of implicit and explicit learning, Zhuang et aI. (1997) investigated alpha-ERD during a variation of the serial reaction time task. During the initial learning, maximal alpha-ERD over the sensorimotor area contralateral to the hand movement was gradually enhanced and reached its peak when the subjects acquired full knowledge of the sequence explicitly. Thereafter, alpha-ERD became smaller. This finding also supports the concept of modulation of motor area activation over the course of learning.

References Baker, SN, Olivier, E and Lemon, RN (1997) Coherent oscillations in monkey motor cortex and hand muscle EMG show task-dependent modulation. J. Physiol., 501: 225-241. Barrett, G, Shibasaki, Hand Neshige, R (1985) A computer-assisted method for averaging movementrelated cortical potentials with respect to EMG onset. Electroenceph. Clin. Neurophysiol., 60: 276-281. Brown, P, Day, BL, Rothwell, JC, Thompson, PD and Marsden, CD (1991) Intrahemispheric and interhemispheric spread of cerebral cortical myoclonic activity and its relevance to epilepsy. Brain, 114: 2333-2351. CIochon, P, Fontbonne, J, Lebrun, Nand Etevenon, P (1996) A new method for quantifying EEG eventrelated desynchronization: amplitude envelope analysis. Electroenceph. Clin. Neurophysiol., 98: 126-129. Crone, NE, Miglioretti, DL, Gordon, B, Sieracki, JM, Wilson, MT, Uematsu, S and Lesser, RP (1998) Functional mapping of human sensorimotor cortex with electrocorticographic spectral analysis. I. Alpha

H. SHIBASAKl AND T. NAGAMINE

and beta event-related desynchronization. Brain, 121: 2271-2299. Defebvre, L, Derambure, P, Bourriez, JL, Jacquesson, JM, Dujardin, K, Destee, A and Guieu, JD (1993) Spatiotemporal study of event-related desynchronization in idiopathic Parkinson's disease. Adv. Neurol., 60: 422-428. Defebvre, L, Bourriez, JL, Dujardin, K, Derambure, P, Destee, A and Guieu, JD (1994) Spatiotemporal study of Bereitschaftspotential and event-related desynchronization during voluntary movement in Parkinson's disease. Brain Topogr., 6: 237-244. Defebvre, L, Bourriez, JL, Derambure, P, Duhamel, A, Guieu, JD and Destee, A (1998) Influence of chronic administration of L-DOPA on event-related desynchronization of mu rhythm preceding voluntary movement in Parkinson's disease. Electroencephalogr. Clin. Neurophysiol., 109: 161-167. Defebvre, LJ, Derambure, P, Bourriez, JL, Cassim, F, Guieu, JD and Destee, A (1999) Motor programming is more affected in progressive supranuclear palsy than in Parkinson's disease: a spatiotemporal study of eventrelated desynchronization. Mov. Disord., 14: 634-641. Derambure, P, Bourriez, JL, Defebvre, L, Cassim, F, Josien, E, Duhamel, A, Destee, A and Guieu, JD (1997) Abnormal cortical activation during planning of voluntary movement in patients with epilepsy with focal motor seizures: event-related desynchronization study of electroencephalographic mu rhythm. Epilepsia, 38: 655-662. Deuschl, G, Toro, C, Matsumoto, J and Hallett, M (1995) Movement-related cortical potentials in writer's cramp. Ann. Neurol., 38: 862-868. Dick, JPR, Rothwell, JC, Day, BL, Cantello, R, Buruma, 0, Gioux, M, Benecke, R, Berardelli, A, Thompson, PD and Marsden, CD (1989) The Bereitschaftspotential is abnormal in Parkinson's disease. Brain, 112: 233-244. Filipovic, SR, Jahanshahi, M and Rothwell, JC (2001) Uncoupling of contingent negative variation and alpha band event-related desynchronizaiton in a go/no-go task. Clin. Neurophysiol., 112: 1307-1315. Gerloff, C, Uenishi, N, Nagamine, T, Kunieda, T, Hallett, M and Shibasaki, H (1998) Cortical activation during fast repetitive finger movements in humans; steadystate movement-related magnetic fields and their cortical generators. Electroencephalogr. Clin. Neurophysiol., 109: 444-453. Hamano, T, Luders, HO, Ikeda, A, Collura, TF, Comair, YG and Shibasaki, H (1997) The cortical generators of the contingent negative variation in humans: a study with subdural electrodes. Electroencephalogr. Clin. Neurophysiol., 104: 257-268.

EEG (MEG)/EMG CORRELATION

Honda, M, Nagamine, T, Fukuyama, H, Yonekura, Y, Kimura, J and Shibasaki, H (1997) Movement-related cortical potentials and regional cerebral blood flow change in patients with stroke after motor recovery. J. Neurol. Sci., 146: 117-126. Ikeda, A and Shibasaki, H (1992) Invasive recording of movement-related cortical potentials in humans. J. Clin. Neurophysiol., 9: 509-520. Ikeda, A, Luders, HO, Burgess, RC and Shibasaki, H (1992) Movement-related potentials recorded from supplementary motor area and primary motor area. Role of supplementary motor area in voluntary movements. Brain, 115: 1017-1043. Ikeda, A, Shibasaki, H, Nagamine, T, Terada, K, Kaji, R, Fukuyama, H and Kimura, J (1994) Dissociation between contingent negative variation and Bereitschaftspotential in a patient with cerebellar efferent lesion. Electroencephalogr. Clin. Neurophysiol., 90: 359-364. Ikeda, A, Luders, HO, Collura, T, Burgess, RC, Morris, HH, Hamano, T and Shibasaki, H (1996) Subdural potentials at orbitofrontal and mesial prefrontal areas accompanying anticipation and decision making in humans: a comparison with Bereitschaftspotential. Electroencephalogr. Clin. Neurophysiol., 98, 206-212. Ikeda, A, Shibasaki, H, Kaji, R, Terada, K, Nagamine, T, Honda, M and Kimura, J (1997) Dissociation between contingent negative variation (CNV) and Bereitschaftspotential (BP) in patients with parkinsonism. Electroencephalogr. Clin. Neurophysiol., 102: 142-151. Ikeda, A, Ohara, S, Matsumoto, R, Kunieda, T, Nagamine, T, Miyamoto, S, Kohara, N, Taki, W, Hashimoto, N and Shibasaki, H (2000) Role of primary sensorimotor cortices in generating inhibitory motor response in humans. Brain, 123: 1710-1721. Kaji, R, Ikeda, A, Ikeda, T, Kubori, T, Mezaki, T, Kohara, N, Kanda, M, Nagamine, T, Honda, M, Rothwell, JC, Shibasaki, H and Kimura, J (1995) Physiological study of cervical dystonia. Task-specific abnormality in contingent negative variation. Brain, 118: 511-522. Kitamura, J, Shibasaki, H and Kondo, T (1993a) A cortical slow potential is larger before an isolated movement of a single finger than simultaneous movement of two fingers. Electroencephalogr. Clin. Neurophysiol., 86: 252-258. Kitamura, J, Shibasaki, H, Takagi, A, Nabeshima, Hand Yamaguchi, A (1993b) Enhanced negative slope of cortical potentials before sequential as compared with simultaneous extensions of two fingers. Electroencephalogr. Clin. Neurophysiol., 86: 176-182. Kitamura, J, Shibasaki, H, Terashi, A and Tashima, K (1999) Cortical potentials preceding voluntary finger

27 movement in patients with focal cerebellar lesion. Clin. Neurophysiol., 110: 126-132. Kunieda, T, Ikeda, A, Ohara, S, Yazawa, S, Nagamine, T, Taki, W, Hashimoto, Nand Shibasaki, H (2000) Different activation of pre-supplementary motor area, supplementary motor area proper, and primary sensorimotor area, depending on the movement repetition rate in humans. Exp. Brain Res., 135: 163-172. Lopes da Silva, FH and Pfurtscheller, G (1999) Basic concepts on EEG synchronization and desynchronization. In: G Pfurtscheller and FH Lopes da Silva (Eds.), Event-Related Desynchronization. Handbook of Electroencephalography and Clinical Neurophysiology., Elsevier, Amsterdam, revised series, Vol. 6, pp. 3-11. Ltiders, HO, Dinner, DS, Morris, HH, Wyllie, E and Comair, YG (1995) Cortical electrical stimulation in humans. The negative motor areas. Adv. Neurol., 67: 115-129. Magnani, G, Cursi, M, Leocani, L, Volonte, MA, Locatelli, T, Elia, A and Comi, G (1998) Event-related desynchronization to contingent negative variation and self-paced movement paradigms in Parkinson's disease. Mov. Disord., 13: 653-660. Magnani, G, Leocani, L, Cursi, M and Comi, G (1999) ERD of the mu rhythm in self-paced and externally triggered movements in idiopathic Parkinson's disease and effect of a single dose of L-DOPA on self-paced movement. In: G Pfurtscheller and FH Lopes da Silva (Eds.), Event-Related Desynchronization. Handbook of Electroencephalography and Clinical Neurophysiology. Elsevier, Amsterdam, revised series, Vol. 6, pp.371-382. Mirna, T, Nagamine, T, Ikeda, A, Yazawa, S, Kimura, J and Shibasaki, H (l998a) Pathogenesis of cortical myoclonus studied by magnetoencephalography. Ann. Neurol., 43: 598-607. Mirna, T, Nagamine, T, Nishitani, N, Mikuni, N, Ikeda, A, Fukuyama, H, Takigawa, T, Kimura, J and Shibasaki, H (I 998b) Cortical myoclonus. Sensorimotor hyperexcitability. Neurology, 50: 933-942. Nagamine, T, Kajola, M, Salmelin, R, Shibasaki, Hand Hari, R (1996) Movement-related slow cortical magnetic fields and changes of spontaneous MEG- and EEG-brain rhythms. Electroencephalogr. Clin. Neurophysiol., 99, 274-286. / Oga, T, Ikeda, A, Nagamine, T, Sumi, E, Matsumoto, R, Akiguchi, I, Kimura, J and Shibasaki, H (2000) Implication of sensorimotor integration in the generation of periodic dystonic myoclonus in subacute sclerosing panencephalitis (SSPE). Mov. Disord., 15: 1173-1183. Ohara, S, Ikeda, A, Kunieda, T, Yazawa, S, Baba, K, Nagamine, T, Taki, W, Hashimoto, N, Mihara, T and Shibasaki, H (2000) Movement-related change of

28 electrocorticographic activity in human supplementary motor area proper. Brain, 123: 1203-1215. Pfurtscheller, G and Aranibar, A (1977) Event-related cortical desynchronization detected by power measurements of scalp EEG. Electroencephalogr. Clin. Neurophysiol., 42: 817-826. Pfurtscheller, G and Klimesch, W (1991) Event-related desynchronisation during motor behaviour and visual information processing. Electroencephalogr. Clin. Neurophysiol., Suppl. 42: 58-65. Pfurtscheller, G, Neuper, C, Andrew, C and Edlinger, G (1997) Foot and hand area mu rhythms. Int. J. Psychophysiol.,26: 121-135. Pfurtscheller, G, Pichler-Zalaudek, K, Ortmayr, B, Diez, J and Reisecker, F (1998) Postmovement beta synchronization in patients with Parkinson's disease. J. Clin. Neurophysiol., 15: 243-250. Pfurtscheller, G and Lopes da Silva, FH (1999) Functional meaning of event-related desynchronization (ERD) and synchronization (ERS). In: G Pfurtscheller and FH Lopes da Silva (Eds.), Event-Related Desynchronization. Handbook of Electroencephalography and Clinical Neurophysiology. Elsevier, Amsterdam, revised series, Vol. 6, pp. 51-63. Platz, T, Kim, IH, Pintschovius, H, Winter, T, Kieselbach, A, Villringer, K, Kurth, R and Mauritz, KH (2000) Multimodal EEG analysis in man suggests impairmentspecific changes in movement-related electric brain activity after stroke. Brain, 123: 2475-2490. Rothwell, JC, Higuchi, K and Obeso, JA (1998) The offset cortical potential: an electrical correlate of movement initiation in man. Mov. Disord., 13: 330-335. Salmelin, Rand Hari, R (1994) Spatiotemporal characteristics of sensorimotor neuromagnetic rhythms related to thumb movement. Neuroscience, 60: 537-550. Salmelin, RH and Harnalainen, MS (1995a) Dipole modelling of MEG rhythms in time and frequency domains. Brain Topogr., 7: 251-257. Salmelin, R, Hamalainen, M, Kajola, M and Hari, R (1995b) Functional segregation of movement-related rhythmic activity in the human brain. Neuroimage, 2: 237-243. Shibasaki, H (1993) Movement-related cortical potentials. In: A Halliday (Ed.), Evoked Potentials in Clinical Testing. Churchill Livingstone, Edinburgh, 2nd ed., pp.523-537. Shibasaki, H (2000) Electrophysiological studies of myoclonus. Muse. Nerve, 23: 321-335. Shibasaki, H and Kuroiwa, Y (1975) Electroencephalographic correlates of myoclonus. Electroencephalogr. Clin. Neurophysiol., 39: 455-463. Shibasaki, H and Rothwell, JC (1999) EMG-EEG correlation. In: G Deuschl and A Eisen (Eds.), Recommendations for the Practice of Clinical Neurophysiology:

H. SHIBASAKI AND T. NAGAMINE

Guidelines of the International Federation of Clinical Neurophysiology. Electroencephalogr. Clin. Neurophysiol., Suppl, 52, pp. 269-274. Shibasaki, H, Barrett, G, Halliday, E and Halliday, A (1980) Components of the movement-related cortical potential and their scalp topography. Electroencephalogr. Clin. Neurophysiol., 49: 213-226. Shibasaki, H, Barrett, G, Neshige, R, Hirata, I and Tomoda, H (1986) Volitional movement is not preceded by cortical slow negativity in cerebellar dentate lesion in man. Brain Res., 368: 361-365. Shibasaki, H, Ikeda, A, Nagamine, T, Mirna, T, Terada, K, Nishitani, N, Kanda, M, Takano, S, Hanazono, T, Kohara, N, Kaji, R and Kimura, J (1994) Cortical reflex negative myoclonus. Brain, 117: 477-486. Terada, K, Ikeda, A, Nagamine, T and Shibasaki, H (1995a) Movement-related cortical potentials associated with voluntary muscle relaxation. Electroencephalogr. Clin. Neurophysiol., 95: 335-345. Terada, K, Ikeda, A, Van Ness, PC, Nagarnine, T, Kaji, R, Kimura, J and Shibasaki, H (1995b) Presence of Bereitschaftspotential preceding psychogenic myoclonus: clinical application of jerk-locked back averaging. J. Neurol. Neurosurg. Psychiatry, 58: 745-747. Terada, K, Ikeda, A, Yazawa, S, Nagamine, T and Shibasaki, H (1999) Movement-related cortical potentials associated with voluntary relaxation of foot muscles. Clin. Neurophysiol., 110: 397-403. Toma, K, Nagamine, T, Yazawa, S, Terada, K, Ikeda, A, Honda, M, Oga, T and Shibasaki, H (2000) Desynchronization and synchronization of central 20 Hz rhythms associated with voluntary muscle relaxation: a magnetoencepha1ographic study. Exp. Brain Res., 134: 417-425. Toro, C, Deusch1, G and Hallett, M (2000) Movementrelated electroencephalographic desynchronization in patients with hand cramps: evidence for motor cortical involvement in focal dystonia. Ann. Neurol., 47: 456461. Ugawa, Y, Shimpo, T and Mannen, T (1989) Physiological analysis of asterixis: silent period locked averaging. J. Neurol. Neurosurg. Psychiatry, 52: 89-93. Vidailhet, M, Stocchi, F, Rothwell, JC, Thompson, PD, Day, BL, Brooks, DJ and Marsden, CD (1993) The Bereitschaftspotential preceding simple foot movement and initiation of gait in Parkinson's disease. Neurology, 43: 1784-1788. Yazawa, S, Shibasaki, H, Ikeda, A, Terada, K, Nagamine, T and Honda, M (1997) Cortical mechanism underlying externally cued gait initiation studied by contingent negative variation. Electroencephalogr. Clin. Neurophysiol., 105: 390-399. Yazawa, S, Ikeda, A, Kunieda, T, Mirna, T, Nagamine, T, Ohara, S, Terada, K, Taki, W, Kimura, J and Shibasaki,

EEG (MEG)/EMG CORRELATION

H (1998) Human supplementary motor area is active in preparation for both voluntary muscle relaxaion and contraction: subdural recording of Bereitschaftspotential. Neurosci. Lett., 244: 145-148. Yazawa, S, Ikeda, A, Kaji, R, Terada, K, Nagamine, T, Toma, K, Kubori, T, Kimura, J and Shibasaki, H (1999) Abnormal cortical processing of voluntary muscle relaxation in patients with focal hand dystonia studied by movement-related potentials. Brain, 122: 13571366. Yazawa, S, Ikeda, A, Kunieda, T, Ohara, S, Mirna, T, Nagamine, T, Taki, W, Kimura, J, Hori, T and Shibasaki, H (2000) Human supplementary motor area

29 is active before voluntary movement: subdural recording of Bereitschaftspotential from medial frontal cortex. Exp. Brain Res., 131: 165-177. Yoshida, K, Kaji, R, Hamano, T, Kohara, N, Kimura, J, Shibasaki, H and Iizuka, T (2000) Cortical potentials associated with voluntary mandibular movements. J. Dent. Res., 79: 1514-1518. Zhuang, P, Toro, C, Grafman, J, Manganotti, P, Leocani, L and Hallett, M (1997) Event-related desynchronization (ERD) in the alpha frequency during development of implicit and explicit learning. Electroencephalogr. Clin. Neurophysiol., 102: 374-38 I.

Movement Disorders Handbook of Clinical Neurophysiology, Vol. 1 M. Hallett (Ed.) © 2003 Elsevier B.v. All rights reserved

31 CHAPTER 4

Electrocorticography in motor control and movement disorders Akio Ikeda* Department of Neurology. Kyoto University Graduate School of Medicine. Shogoin, Sakyo-ku, Kyoto 606, Japan

4.1. Introduction

Direct recording from the brain surface in humans is not a routine part of clinical EEG. It was initially performed in the 1940s intraoperatively by using electrodes placed on the cortical surface (electrocorticography or ECoG) (Penfield and Jasper, 1954). Chronic intracranial recordings were subsequently obtained by using flexible strips of electrodes inserted into the epidural or subdural space through a burr hole (Penfield and Jasper, 1954). Stereotactic implantation of rigid electrodes (depth electrodes) directly into specific intracerebral structures was later developed in order to obtain more anatomically precise localization of epileptiform discharges (stereotactic EEG or SEEG) (Talairach and Bancaud, 1974). On the other hand, since the 1970s, chronic implantation of large multi-electrode grids over a wide surface of the cerebral cortex exposed by craniotomy has enabled us to obtain well organized cortical activity almost continuously in awake patients without much restriction for up to several weeks (chronic subdural EEG). These invasive recording techniques had been developed exclusively for epilepsy surgery, but recently chronic subdural EEG has also been applied for functional surgery other than epilepsy, for example, when a brain tumor is located at or very close to the functional cortices such as motor, sensory, visual and language areas (Morris et aI., 1986). Chronic subdural EEG recorded by electrode grids provides us with the following two advantages to SEEG: (1) The cortical surface is widely covered

* Dr. Akio Ikeda, M.D., Ph.D., Department of Neurology, Kyoto University Graduate School of Medicine, Shogoin, Sakyo-ku, Kyoto 606, Japan. E-mail address:[email protected] Tel.: +81-75-751-3772; fax: +81-75-751-9416.

by electrode grids spaced 1 em apart. These grids are able to localize the recorded activity such as epileptiform discharges, mu rhythm, evoked cortical potentials to peripheral sensory stimulation and so on. (2) Electric currents applied on each grid electrode can be used for functional cortical mapping. In studies of normal motor control or in patients with movement disorders of central origin, the Bereitschaftspotential (BP) or readiness potential, the slow brain potential occurring before voluntary movements, may provide important clues to the understanding of preparatory cortical functions in association with voluntary movements in normal subjects and in patients with movement disorders like dystonia, parkinsonism, and psychogenic movement disorders. Since the initial report of BP in scalp recording (Komhuber and Deecke, 1965), its complicated scalp distribution has impeded the understanding of its cortical generators. Invasive studies in animals have partly revealed the cortical generators of the BP (Sasaki and Gemba, 1991), but because the BP reflects preparatory brain functions in voluntary movements, animal studies are limited. Recent study by means of magnetoencephaIography with whole head coverage has revealed the cortical generators partly, but deeper sources like in the mesial frontal surface could not be completely investigated, and furthermore, sources producing radial dipoles are poorly recorded (Erdler et aI., 2000). Chronic subdural EEG recording in epilepsy surgery provides the opportunity to investigate cortical generators of the BP to clarify the central cortical mechanism of voluntary movements (Ikeda et aI., 1992; Shibasaki and Ikeda, 1997; Yazawa et aI.,2000). Because subdural EEG recording is an invasive, diagnostic technique to localize cortical functions

32

and abnormal cortical areas prior to surgical intervention (Ltiders et al., 1987; Ajmone-Marsan, 1990), there is little indication in patients with movement disorders. However, if noninvasive techniques fail to differentiate severe movement disorders like paroxysmal dyskinesias from intractable, focal epilepsy, then chronic subdural EEG would be clinically useful as previously reported (Lombroso, 1995, 2000). 4.2. Recording technique In chronic implantation for several days to weeks, commercially available subdural electrodes are used consisting of electrode grids or strips with multiple contacts embedded at uniform intervals (such as 1 em) in a silicon rubber sheet. Electrodes are made of platinum-iridium or stainless steel. 4.2.1. Electrodes The size and shape of grids used clinically depend on where they are placed over the cortical surface. Large grids (8 x 8, 6 x 5 or 5 x 4 electrode contacts) are usually placed over the lateral convexity where a large functional area is expected to be covered such as the perirolandic area or perisylvian area (Fig. 1). Small grids (8 x 2, 5 x 2) or strips (1 x 4, 1x 5, 1 x 8) are prepared for investigating the interhemispheric zone (supplementary motor area (SMA), cingulate motor area or mesial occipital area), basal temporal area or basal frontal area. Bidirectional electrode grids (two identical sheets of silicon rubber containing 2 x 5 set of electrodes glued together back to back so that the recording surface of each set of 10 electrodes faces in opposite directions) are often used to investigate the bilateral mesial frontal areas simultaneously by inserting only one in one side of the interhemispheric fissure (Fig. 2). It records subdural EEG from one side, and epidural EEG from the opposite side through the falx cerebri. Electrodes are typically 3 mm in diameter, and commonly the center-to-center electrode distance is set to 1 em. Depth electrodes have much smaller recording surfaces. When an electrode is attached on the scalp or brain surface, it can be represented by a resistance (R) and capacitance (C) in series (R-C circuit). Therefore, it has the property of a low frequency filter (LFF), and acts together with the input impedance of the amplifier (Tyner et al., 1983). Since the capacitance is proportional to the surface

A. IKEDA

area of an electrode, an electrode of large surface such as the subdural contact with a diameter of 3 mm, attenuates slow potentials minimally. Depth electrodes with small contact surfaces can not record slow potentials well, and thus, subdural electrodes should record the BP much better than depth electrodes. However, in published reports, depth electrodes inserted in patients with intractable partial epilepsy recorded the BP at the primary motor area (MI) as nicely as subdural electrodes did (Rector et al., 1994, 1998). The effects of electrode size on slow subdural EEG were also compared between electrode contacts of 3 mm diameter and 5 mm diameter, and no significant difference was observed between the two (Ikeda et al., 1999a). This was probably because modem amplifiers with large input impedance more than 50 megaohm can overcome the negative effects of recording surface size. It is difficult to decide optimal inter-electrode distance, but in the clinical situation decreasing electrode interval below 5 mm usually yields little improvement or difference in the information obtained (Ajmone-Marsan, 1990). This can be explained by the experimental results that two epileptic spike foci separated by 4 mm on the same gyrus produced by penicillin solution were consistently dependent whereas spike foci separated by 6 mm were always independent (Lueders et al., 1981). "Cross-talk" between cortical columns is limited to columns 1 to 2 mm apart, and the largest horizontal branches extend approximately 2.5 mm in the superficial cortical layer (Szentagothai, 1969). An interelectrode interval of 5 to 10 mm, mostly 10 mm, is clinically used. On special occasions, an interelectrode distance of 5 mm has been used to map the detailed topography of short latency, cortical components of somatosensory evoked potentials (Allison et al., 1990). Metals for invasive electrodes should not be toxic to brain tissue, and thus stainless steel or platinum are widely used. Electrodes made of stainless steel produce baseline fluctuations large enough to obscure slow potentials. These large electrode potentials arise because stainless steel electrodes are polarized, and thus they also theoretically have problems recording slow EEG components (Cooper, 1963, 1980). However there are several factors that may compensate: (i) a large input impedance would reduce voltage "loading" at the electrode level, resulting in less electrode potentials; (ii) subdural

33

ELECTROCORTICOGRAPHY IN MOTOR CONTROL AND MOVEMENT DISORDERS

;J\.

A1

B1

C1 ::

~~ - -~

--..+-

;

03

~

04

-2

-1

C6

iii i

F~~~+!1 -100

-2

Sec.

05 :•~ ~.:. -:1--

i : :

~ t:,

-1

0

+1

+250

msec.

.

~ C5

ii' :

.. '

~

----*-

B6

~

~

;I~'¥

~ B4

01

....

~

- ,lp,. --

' ..

~

Iv GKP (EMG)

GKP

-2

06

-1

Sec.

0

+1

150 IJV

+

(EEG)

-2

-1

Sec.

0

+1

Fig. 1. Movement-related cortical potentials in association with self-paced tongue protrusion recorded from the subdural electrodes placed across the left lower central sulcus in an epilepsy patient. Large negative BP, NS' and MP are seen in the tongue motor area (maximum at Al and B1), defined by high frequency electric cortical stimulation. Symbols in the figure show the results of electric cortical stimulation (AI, A2, B3, B4 and C4=positive motor response of the right mouth and tongue; Bland C3 = positive motor and sensory responses of the right mouth and tongue; B2 and C2=positive sensory responses of the right mouth and tongue; D4, E3 and E4= sensory language function). (From Ikeda et aI., 1995a with permission.)

electrodes fixed on the brain surface nunmuze baseline fluctuations of electrode potentials arising from unstable contact between electrode and recording surface; and (iii) subdural EEG potentials are recorded with the amplifier sensitivity of 70 to 100 f1 V/mm in contrast with that of 10 f1 V/mm in usual scalp recording. Therefore subdural recording lessens the effect of electrode potentials by 1/7 to 1/10 times. Taking these three factors into account, it is most likely that slow brain potentials such as BP, contingent negative variation (CNV) or ictal DC

shifts are well recorded by using subdural electrodes made even of stainless steel (Ikeda et al., 1992, 1996c, 1998). Since platinum can record slow potentials better than stainless steel (Cooper, 1963), subdural electrodes made of platinum are more effective. 4.2.2. Amplifier and recording conditions

Currently available amplifiers have a large input impedance of 50-200 Mf], and thus, as discussed

A. IKEDA

34

seizure discharges or significant slow activity. Alternately, one of the scalp electrodes placed on the mastoid process or earlobe contralateral to the side of craniotomy can also be used. Since all the EEG signal from the brain surface is recorded with the sensitivity of 70 I-lVImm, small artifacts from the scalp electrode used as the reference is usually negligible. 4.2.3. Electrode placement Fig. 2. Schematic representation of the coronal view of an interhemispheric electrode plate. Two of 2 x 5 electrode grids are attached to each other and face in opposite direction, so as to record subdural potentials directly from one side and epidural potentials from the other side through the falx cerebri. (From Ikeda et aI., 1992 with permission.)

above, they minimize the adverse effects of electrode potentials on slow potential recording. In conventional chronic subdural EEG monitoring for several days, LFF and high frequency filter (HFF) are set to the same condition as that of scalp EEG recording, but opening of LFF down to 0.03 or even 0.01 Hz can record very slow subdural EEG potentials satisfactorily without artifacts as opposed to the scalp EEG even though the patients do not necessarily stay still on the bed. EEG signals are continuously recorded with the sampling rate of at least 200 Hz, preferentially 500 Hz. When fast activity is to be analyzed, larger sampling rates such as 1000 Hz are recommended. Scalp EEG reflects the brain activity occurring simultaneously at the cortical surface of at least 6 em', and the brain activity smaller than that is little recorded because of general attenuation of EEG signal mainly caused by the bone (Cooper et aI., 1965). On the other hand, sensitivity for subdural EEG is usually set to 70 I-lVImm, one-seventh of that of scalp EEG. Fast activity more than 15 Hz is obviously better recorded as compared with scalp EEG, because the scalp and the bone also act as an HFF so as to attenuate further selectively faster frequencies in scalp EEG (Pfurtscheller and Cooper, 1975). A referential montage is better than a bipolar montage to map any focal cortical activity within the area of the grid. As a reference electrode, a subdural electrode contact is chosen which on cortical stimulation does not elicit any symptom or sign, and from which subdural EEG does not show spikes,

Electrode grids of appropriate size are placed during a first craniotomy and removed during a second craniotomy. Detailed surgical techniques are beyond this chapter. When targeting a functional area for mapping, it is important to cover the whole functional area as much as possible because the information of the accurate boundary of the functional area is clinically required. Therefore, in addition to the anatomical landmarks, intraoperative recording of short latency evoked potentials or the mu rhythm immediately before implantation is clinically useful to decide which area should be mainly covered by the electrode grids. 4.2.4. Patient condition for recording and safety

Recording can be started soon after the patients recover from the first craniotomy to place the subdural electrode grids. During the initial several days, intermittent focal or widely distributed slow or transient spikes are observed, but they usually disappear or become less. Conventional subdural EEG can record fast activity more than scalp EEG as already discussed. Surface EMG and EOG are sometimes recorded simultaneously as done in a scalp EEG study. In a scalp EEG study, EOG is monitored in order to exclude the possibility of EOG artifacts on the scalp EEG. Subdural EEG is almost free from artifacts of extracranial origin such as EMG or EOG. Therefore, EOG is not necessarily monitored in this respect (Ikeda et al., 1996a). For obtaining a BP, voluntary movements of any part of the body can be examined as long as fiducial points can be accurately identified with the movement onset. Each movement should be identical and self-paced at intervals of 5-10 s. Subdural EEGs are finally averaged backward and forward time-locked to the trigger pulse, for which the onset of surface EMG discharge is usually used. Subdural EEG has a

35

ELECTROCORTICOGRAPHY IN MOTOR CONTROL AND MOVEMENT DISORDERS

Precentral (A1)

--100 msec. (BP, NS')

Postcentral (C2)

-+50 msec. (MP)

-+250 msec. (RAP)

Fig. 3. Subdurally recorded movement-related cortical potentials across the central sulcus in tongue protrusion, and assumed cortical generator in an epilepsy patient shown in Fig. I. Surface-negative BP at the precentral and positive BP at the postcentral area are seen. (From Ikeda et aI., 1995a with permission.)

better SIN ratio than scalp EEG, but at least 50 trials should be averaged for one session. A rapid-rate movement task (1-2 Hz) is also possible to locate the cortical generator corresponding to MI (Gerloff et aI., 1997; Kunieda et aI., 2000). Patients with intracranial electrodes should be considered in the high risk category for safety. Current leakage of any devise attached to the intracranial electrodes should be less than 10 fLA (Engel et aI., 1986). When patients suffer from intractable epilepsy and antiepileptic medication is also reduced during invasive recording to catch habitual seizures, precautions against accidental head trauma by falling with seizures should be taken. Attenuation of the amplitude of all frequencies or selective attenuation of the fast activity suggests certain pathological conditions like development of acute subdural hematoma or increased intracranial pressure leading to compressing of the brain surface by the grid and causing dysfunction of the cortex.

4.2.5. Data analysis and interpretation

Besides BP, any kind of study done by scalp EEG (sensory evoked potentials, CNV, event-related potentials, event-related desynchronization, EEGEMG coherence and EEG-EEG coherence, etc.) are all possible. Since a large grid can delineate the distribution of the potentials, application of the current source density method is more accurate than for scalp EEG (Nagamine et al., 1992). Subdural electrodes placed on the cortical surface theoretically record both tangential and radial dipole components, arising from the crown of a gyrus and from the sulcus, respectively (Fig. 3), according to solid angle theory (Gloor, 1985). However, since an electrode records potentials from both the components simultaneously, it is usually difficult to decompose the two. Partial coherence analysis methods can differentiate the two using phase analysis (Towle et aI., 1999).

36

A. IKEDA

4.2.6. Intraoperative ECoG This refers to acute recording of epicortical EEG during the exposure of the cortical surface in the course of surgical treatment. If the patient's cooperation is needed, recording is done without general anesthesia, but the time of recording and patient cooperation is limited because of a less comfortable situation. Otherwise, all the recording equipment and settings are the same as that of chronic subdural EEG.

4.3. Cortical generators of BP 4.3.1. Scalp-recorded SP It is well accepted, regardless of movement sites of the body, that pre-movement potentials consist of

three components; BP or readiness potential (Komhuber and Deecke, 1965), negative slope (NS') (Shibasaki et al., 1980), and motor potential (MP) (Komhuber and Deecke, 1965) (Fig. 4). They are followed by post-movement potentials called Reafferente Potentiale (RAP) (Kornhuber and Deecke, 1965). In the case of simple limb movements, the BP starts as early as 1.5-2 s before movement onset. It is maximal at the central midline and is symmetrically and widely distributed over the scalp irrespective of the side of the movements. At 300-400 ms before movement onset, the slow negative shift becomes steeper (NS'), especially at the centroparietal region contralateral to the movement side. The negative slope becomes much steeper immediately before movement onset and then peaks at or just after the movement onset (MP), followed by large, mainly positive transients (RAP). BPINS' is strongly related to voluntary movements, because it is not recorded before passive movements or involuntary movements such as simple tics, myoclonus and chorea in Huntington's disease (Shibasaki and Ikeda, 1996). BPINS' becomes smaller in patients with Parkinson's disease and with focal dystonia associated with a lesion of the basal ganglia.

4.3.2. Multiple cortical generators of SP EMG

_ I

I

I

-1,500 -900

BP

NSf

o 300 msec

MP

@~ Fig. 4. Scalp-recorded wave forms of BP, NS', MP and RAP with voluntary, self-paced extensions of the right hand(upperhalf), and schematic scalptopography of each component (BP, NS' and MP) with right hand voluntary movements viewed from the top (lower half). LHM=left hand motor area, RHM=right hand motor area. (Upper half was modified from Fig. 4 of Shibasaki et aI., 1980 with permission: lower half modified from Fig. 5 of Neshige et al., 1988 with permission.)

Based on the distribution of each component on the scalp, the BP was originally thought to be represented by the bilateral SMA activities, and the NS' and MP were mainly represented by the contralateral primary motor and sensory cortices (MI-SI). So far, subdural EEG studies have revealed that MI-SI, SMA proper and pre-SMA are cortical generators of pre-movement potentials (BP, NS' and MP) (Fig. 5) (Ikeda and Shibasaki, 2002).

4.3.2.1. Precentral gyrus (positive motor area) A somatotopy of the positive motor area in the precentral gyrus as revealed by electric cortical stimulation is also consistent with that by BP (Fig. 6) (Neshige et al., 1988; Ikeda et al., 1992), but it seems that BPs for each task are often distributed wider than the motor mapping made by cortical stimulation. It may be partly because BP is generated not only by the anterior bank but also by the crown part of the precentral gyrus (Ikeda et al., 1995a), from both of which corticospinal pathways originate.

ELECTROCORTICOGRAPHY IN MOTOR CONTROL AND MOVEMENT DISORDERS

Contralateral

37

Ipsilateral

Fig. 5. Schematic representation of the cortical generators of BP, NS' and MP for hand movements viewed from the top. The degree of darkness of the shading in the pre-SMA, SMA proper and MI-SI is approximately proportional to the amplitude of the corresponding potentials. MI=primary motor area, Sl eprimary sensory area. (Modified from Fig. 14 of Ikeda et al., 1992 with permission.)

4.3.2.2. Supplementary motor area (SMA) SMA is subdivided into caudal (SMA proper) and rostral (pre-SMA) parts in humans as well as in monkeys at the level of the VAC line (Picard and Strick, 1996); both SMA proper and pre-SMA are generators of the BP. SMA proper, like the precentral gyrus, has a somatotopic organization demonstrated by both electric cortical stimulation and BP (Fig. 7) (Ikeda et al., 1992, 1993, 1995b). Pre-SMA generates a similar BP invariably regardless of the sites of the voluntary movements in the body (Fig. 8) (Yazawa et al., 2000). Furthermore, the following findings may aid in differentiating the motor cortices located in the mesial frontal area.

proper could not be defined only by anatomical information. (2) The pre-SMA and SMA proper both generate a clear BP with slow-rate repetition of voluntary movements, but little BP in the rapid-rate repetition of movements, whereas MI generates a clear BP for slow- and rapid-rate repetition of movements (Kunieda et al., 2000). In the rapidrate repetition of movements, voluntary movements may be conducted as so called "automatic" movements, while in the slow-rate repetition of movements each movement is regarded as a discrete, individual execution, each of which may involve different motor control mechanisms.

(1) Onset time of BP between the foot MI and foot

4.4. Motor mapping by BP in functional neurosurgery

area at the SMA proper has no significant difference, but movement-related desynchronization before movement onset starts earlier at the SMA proper than at MI (Ohara et al., 2000). The foot area of the SMA proper generates BP not only for contralateral but also for ipsilateral movements whereas the foot MI generates BP exclusively for contralateral foot movements. Additionally, the latter generates a clear RAP immediately after movement onset whereas the former does not (Ikeda et al., 1992) (Fig. 9). Since the SMA proper is also often found in the lower half of the paracentral lobule by both electric cortical stimulation (Lim et al., 1996) and the BP (Ikeda et al., 1992), the boundary between the foot MI and foot area at the SMA

For mapping SI, SEPs and high frequency cortical stimulation are equally useful. The central sulcus can be delineated by SEPs. However, it does not directly represent the location of MI, but either the radial or tangential dipole arising from the postcentral gyrus. Like SEPs to define postcentral gyrus, the BP is clinically useful to localize the motor cortices. With regard to the mesial frontal cortices, functional delineation between the foot MI and the foot area at the SMA proper, or between the SMA proper and the pre-SMA, is made exclusively by high frequency cortical stimulation or just by anatomical approach. The BP can also be clinically useful to differentiate the mesial motor cortices, at least by complementing

38

A. IKEDA

R MF Ext.

STIMULUS EFFECT I

None

"'1lI.~'''''''''~~~()<,.'.'''''I).(I,~ I

I

I

I

I

'Jl<x.~~\,;:~

None I

.~~~~.::f'~:::---Ll--~

~~

M:Finger (I II, III)

I

.......

I

BP ._~~.Y~

,,

:,1

•

,

M: Finger (II, III)

I

I

-\r

_

I

I I

s:

r

··,.,. " :'/ •

•!

t

........-.-.-..."'-'l_.-,."""Il~_?:_:?:_~_-l

_

Finger (I)

s: Palm

I

I

-I~OO

o

+500

ms

Fig. 6. Movement-related cortical potentials in association with self-paced extension of the right middle finger recorded from the subdural electrodes placed across the left central sulcus in an epilepsy patient. BP, NS' and MP are well localized to the finger motor area, and to a lesser degree, to the sensory area. Me motor, S=sensory. (From Neshige et al., 1988 with permission.)

the result of cortical stimulation (Ikeda and Shibasaki, 1992; Allison et aI., 1996; Yazawa et al., 1997). Motor mapping by BP has the following four advantages: (1) it reflects the activity of the motor cortices which are actually involved in preparation for and execution of voluntary movements; (2) it is present for voluntary movements of all parts of the body; (3) cortical activity in the bank of the sulcus (like the anterior bank of the central sulcus) may be recorded as a tangential dipole; (4) it is not associated with the risk of seizure induction in contrast with high frequency cortical electric stimulation. The BP method has the following two

disadvantages as compared with cortical electric stimulation: (1) voluntary movements of one kind need to be repeated at least 50 times to obtain averaged wave forms; (2) the patient's cooperation is needed to conduct the task of voluntary movements and to obtain satisfactory results (Ikeda et al., 2002). 4.5. Contingent negative variation (CNV) CNV is one of the event-related potentials initially reported by Walter et al. (1964). It is a slow negative brain potential occurring between two successive

ELECTROCORTICOGRAPHY IN MOTOR CONTROL AND MOVEMENT DISORDERS (1) Right Ankle Extension

~ir-=-

85_-t.,.f,---= }

A6

86

A7

87

~

----v

_1.

-~

---"'It"l~

(2) Right Finger Extension

AS

-+--_

~,:;

"p'"'

-.,-.rA7

85

.

J 1-=-....

".

...of'==

~.

~6

}

87

--~

(3) Tongue Protrusion

·2

·1

+1

·2

Sec.

·1

+1

~I

50~V

(EEG)

Fig. 7. Movement-related cortical potentials in association with self-paced movements of the right finger, right foot and tongue recorded from the left SMA in an epilepsy patient. The figure illustrates the somatotopic distribution of the BP within the SMA proper, consistent with the results of cortical stimulation. The broken line shows the boundary between the foot MI posteriorly (lower part in the right figure) and the SMA proper anteriorly (middle part in the right figure). Symbols in the figure show the results of electric cortical stimulation (A5 and B5 =positive motor response in the face and negative motor response of the bilateral hands; A6=rhythmic vocalization; B6=tonic motor response of the right hand; A7 and B7=positive motor response of the right hand and foot; A8, B8, A9 and B9=clonic motor response of the right foot). (From Ikeda et al., 1992 with permission.)

stimuli only when the two stimuli are associated with or contingent to each other, but not when the two are merely a simple pairing of stimuli. The basic paradigm of CNV recording is a constant foreperiod, reaction-time paradigm. In general, the first stimulus (SI) serves as a preparatory "warning" signal for the "imperative" stimulus (S2), to which a motor response is usually made. From the viewpoint of motor control and movement disorders, the CNV should represent the neuronal activity necessary for sensorimotor integration or association, and, there-

39

fore, it is related to planning or execution of externally-paced, voluntary movements. On the other hand, BPs represent the neuronal activity prior to self-paced voluntary movement (Tecce and Cattanach, 1993). The CNV is a composite wave form consisting of at least two components when the interval between S1 and S2 is set to 1.5 s or longer (Rohrbaugh et al., 1976). These are: (i) an early CNV (initial epoch of 300-500ms); and (ii) a late CNV (final epoch of 300-500 ms). The early CNV is thought to represent a reactive, arousal process associated with S1 and partly an anticipatory process for S2, and the late CNV is thought to represent an anticipatory process for S2 and attention to the motor response (Simson et al., 1977). The early CNV shows a frontal dominant distribution maximal at the midline. The late CNV is centrally dominant with little or no laterality, as opposed to the later part of the BP which is larger contralateral to the movement side (Grunewald et al., 1979). The late CNV is diminished in patients with focal dystonia, suggesting impairment of sensorimotor integration in a reaction-time paradigm (Kaji et al., 1995; Ikeda et al., 1996b; Hamano et al., 1999). Subdural EEG recording has revealed the cortical generators of CNV. The late CNV originates at nonprimary motor cortices (pre-SMA, SMA proper, lateral premotor area), prefrontal area (mesial and orbitofrontal areas) and MI (Lamarche et al., 1995; Ikeda et al., 1996a, d, 1999b; Hamano et al., 1997). The early CNV, if related to discrimination of external dichotic stimuli and selection of the movement, was observed principally at the pre-SMA, and its onset and peak latencies were 200 and 600 ms, respectively, after S1 in an "S l-choice of GolNoGo and S2-reaction-time" paradigm (Ikeda et al., 1999b). It represents the characteristic function of pre-SMA in cognitive motor control. 4.6. Electrocorticography in patients with movement disorders Paroxysmal dyskinesias are a heterogeneous group of movement disorders whose main features are intermittent attacks of dystonic, choreoathetotic and ballistic movements, and the clinical semiology is often indistinguishable from seizures arising from the SMA. Nocturnal paroxysmal dystonia has clin-

40

A. IKEDA

Fig. 8. Movement-related cortical potentials in association with self-paced extension of the left middle finger (B) and the right middle finger (C), and dorsiflexion of the left foot (D) and the right foot (E), recorded from the right mesial frontal cortex in an epileptic patient. In all tasks, negative BP is observed at F4, located anterior to the VAC line, being most likely pre-SMA. F2 just on the paracentral lobule (most likely foot MI area) generates BP only in association with the left foot movements. VAC=a line on the anterior commissure perpendicular to AC-PC line, VPC =a line on the posterior commissure perpendicular to AC-PC line. (From Yazawa et al., 2000 with permission.)

ical semiology suggesting paroxysmal dysfunction of the basal ganglia, but it is also indistinguishable from epileptic seizures arising from SMA or other frontal lobe structure. Based on the assumption that those movement disorders are caused by paroxysmal activity at either frontal lobe or basal ganglia following noninvasive EEG and SPECT studies, Lombroso and colleague decided to implant subdural electrode grids chronically over the mesial and lateral frontal cortices and depth electrodes at the caudate and amygdala nuclei in 3 patients (Lombroso, 1995, 2000; Lombroso and Fischman, 1999). Among 3 patients, 1 patient had ictal onset activity at the SMA, rapidly spreading to the basal ganglia; 1 patient had ictal onset at the orbitofrontal area, rapidly spreading selectively to the SMA; and 1 patient showed sustained 7-12 Hz discharges only at the caudate nucleus during habitual attacks. Invasive recording techniques can provide important informa-

tion with regard to pathogenesis in patients with movement disorders, and indications should be clarified based on the results of noninvasive, comprehensive investigations done beforehand. 4.7. Future consideration

Multiple cortical generators of BP and eNV have been revealed by invasive studies, but it is not certain which are the most important generators, or how they are functionally related to each other. Inter-areal functional coupling among the areas is essential for studying more sophisticated higher cortical functions. Furthermore, the presence of multiple cortical areas does not necessarily mean that surgical resection of any participating area causes dysfunction, and thus a future study combined with transcranial magnetic stimulation (TMS) which

ELECTROCORTICOGRAPHY IN MOTOR CONTROL AND MOVEMENT DISORDERS

41

Acknowledgments This study was supported by Grants-in-Aid for Scientific Research on Priority area (C)-Advanced Brain Science Project 12210012 from the Japan Ministry of Education, Culture, Sports, Science and Technology (MEXT); Research Grants (B2)13470134 and (C2)-13670460 from the Japan Society for Promotion of Sciences; Research Grant from Brain Science Foundation.

85

AS

A6 A7 -oS

,

~,

.... 1-

-

~

_89_ _.:-.-_ I

~I SOI'V

References

(EEG)

'-r ~

85

A6

B6

t......,....

A7

87

~ ~

It

..w==

o

+1

AS

1'\1

-2

,..===

-1

Ba B9

·2

Sec.

·1

o

+1

·2·1

0

+1

Sec.

Fig. 9. Movement-related cortical potentials associated with voluntary, self-paced foot dorsiflexion of the left (upper halt) and the right (lower halt) side recorded from the left mesial brain surface subdurally in the same patient as Fig. 7. Positive BPs are of approximately equal amplitude at both SMA proper (A7) and in the contralateral foot MI (A8). A negative transient at + 145 ms was seen only at the contralateral foot MI (A8, A9), but not in the SMA proper (A7). The broken line shown the boundary between the foot MI posteriorly (lower part in the figure) and the SMA proper anteriorly (upper part in the figure). Symbols in the figure are explained in Fig. 7. (From Ikeda et aI., 1992 with permission.) produces transient dysfunction may be helpful. "Functional plasticity" associated with pathological conditions such as epilepsy or intraparenchymal tumor could produce a large variability in the functional mapping, and thus a careful investigation and interpretation of the results is needed for each patient. How well cortical mapping by BP or CNV reflects "functional plasticity" as compared with mapping by high frequency electric cortical stimulation should also be studied.

Ajmone-Marsan, C (1990) Chronic intracranial recording and e1ectrocorticography. In: DD Daly and TA Pedley (Eds.), Current Practice of Clinical Electroencephalography. Raven Press, New York, pp. 535-560. Allison, T, McCarth,y G, Wood, CC, Darcey, TM, Spencer, D and WilIIiamson, PD (1989) Human cortical potentials evoked by stimulation of the median nerve. I. Cytoarchitecture areas generating shortlatency activity. J. Neurosci., 62: 694-710. Allison, T, McCarthy, G, Luby, M, Puce, A and Spencer, DD (1996) Localization of functional regions of human mesial cortex by somatosensory evoked potential recording and by cortical stimulation. Elecroencephalogr. Clin. Neurophysiol., 100: 126-140. Cooper, R (1963) Electrodes. Amer. J. EEG Technol., 3: 91-101. Cooper, R, Winter, AL, Crow, HJ and Walter, WG (1965) Comparison of subcortical, cortical, and scalp activity using chronically indwelling electrodes in man. Elecroencephalogr. Clin. Neurophysiol., 18: 217-228. Cooper, R, Osselton, JW and Shaw, JC (1980) Electrodes. In: EEG Technology. Butterworth, London, 3rd ed., pp.15-31. Engel, J, Jr and Crandall, PH (1986) Intensive neurodiagnostic monitoring with intracranial electrodes. In: RJ Gumnit (Ed.), Intensive Neurodiagnostic Monitoring. Advances in Neurology. Raven Press, New York, Vol. 46, pp. 85-106. Erdler, M, Beisteiner, R, Mayer, D, Kaindl, T, Edward, V, Windischberger, C, Lindinger, G and Deecke, L (2000) Supplementary motor area activation preceding voluntary movement is detectable with a whole-scalp magnetoencephalography system. Neuroimage, II: 697-707. Gerloff, C, Toro, C, Uenishi, N, Cohen, LG, Leocani, L and Hallett, M (1997) Steady-state movement-related cortical potentials: a new approach to assessing cortical activity associated with fast repetitive finger move-

42 ments. Elecroencephalogr. Clin. Neurophysiol., 102: 106-113. Gloor, P (1985) Neuronal generators and the problem of localization in electroencephalography: application of volume conductor theory to electroencephalography. J. cu« Neurphysiol., 2: 327-354. Grunewald, G, Grunewald-Zuberbier, E, Netz, J, Homberg, V and Sander, G (1979) Relationship between the late component of the negative contingent variation and the bereitschatspotential. Elecroencephalogr. Clin. Neurophysiol., 46: 538-545. Hamano, T, Liiders, HO, Ikeda, A, Collura, T, Comair, YG and Shibasaki, H (1997) The cortical generators of the contingent negative variation in humans: a study with subdural electrodes. Elecroencephalogr. Clin. Neurophysiol., 104: 257-268. Hamano, T, Kaji, R, Katayama, M, Kubori, T, Ikeda, A, Shibasaki, H and Kimura, J (1999) Abnormal contingent negative variation in writer's cramp. Clin. Neurophysiol., 110: 508-515. Ikeda, A and Shibasaki, H (1992) Invasive recording of movement-related cortical potentials in humans. 1. cu« Neurophysiol., 9: 509-520. Ikeda, A and Shibasaki, H (2002) Intracranial recording of Bereitschaftspotentials in patients with epilepsy. In: Jahanshai and M Mallett (Eds.), The Bereitschaftspotentials: Movement-related cortical potentials, Kluwer AcaderniclPlenum, London, pp. 45-59. Ikeda, A, Liiders, HO, Burgess, RC and Shibasaki, H (1992) Movement-related potentials recorded from supplementary motor area and primary motor area: role of supplementary motor area in voluntary movements. Brain, 115: 1017-1043. Ikeda, A, Liiders, HO, Burgess, RC and Shibasaki, H (1993) Movement-related potentials associated with single and repetitive movements recorded from human supplementary motor area. Elecroencephalogr. Clin. Neurophysiol., 89: 269-277. Ikeda, A, Luders, HO, Burgess, RC, Sakamoto, A, Klem, GH, Morris, HH and Shibasaki, H (1995a) Generator locations of movement-related potentials with tongue protrusions and vocalization: subdural recording in human. Elecroencephalogr. Clin. Neurophysiol., 96: 310-328. Ikeda, A, Ltiders, HO, Shibasaki, H, Collura, TF, Burgess, RC, Morris, HH and Hamano, T (1995b) Movementrelated potentials associated with bilateral simultaneous and unilateral movement recorded from human supplementary motor area. Elecroencephalogr. Clin. Neurophysiol., 95: 323-334. Ikeda, A, Luders, HO, Collura, TF, Burgess, RC, Morris, HH, Hamano, T and Shibasaki, H (1996a) Subdural potentials at orbitofrontal and mesial prefrontal areas

A. IKEDA

accompanying anticipation and decision making in humans: a comparison with Bereitschaftspotential. Elecroencephalogr. Clin. Neurophysiol., 98: 206212. Ikeda, A, Shibasaki, H, Kaji, R, Terada, K, Nagarnine, T, Honda, M, Hamano, T and Kimura, J (1996b) Abnormal sensorimotor integration in writer's cramp: study of contingent negative variation. Mov. Disord., 11: 691-700. Ikeda, A, Terada, K, Mikuni, N, Burgess, R, Comair, Y, Taki, W, Hamano, T, Kimura, J, Liiders, HO and Shibasaki, H (1996c) Subdural recording of ictal DC shifts in neocortical seizures in human. Epilepsia, 37: 662-674. Ikeda, A, Liiders, HO and Shibasaki, H (1996d) Generation of contingent negative variation (CNV) in the supplementary sensorimotor area. In: HO Ltiders (Ed.), Supplementary Sensorimotor Area. Advances in Neurology. Lippincott-Raven, New York, Vol. 70, pp. 153159. Ikeda, A, Nagarnine, T, Yarita, M, Terada, K, Kimura, J and Shibasaki, H (1998) Reappraisal of the effects of electrode property on recording slow potentials. Elecroencephalogr. Clin. Neurophysiol., 107: 59-63. Ikeda, A, Taki, W, Kunieda, T, Terada, K, Mikuni, N, Nagamine, T, Yazawa, S, Ohara, S, Hori, T, Kaji, R, Kimura, J and Shibasaki, H (1999a) Focal ictal DC shifts in human epilepsy as studied by subdural and scalp recording. Brain, 122: 827-838. Ikeda, A, Yazawa, S, Kunieda, T, Ohara, S, Terada, K, Mikuni, N, Nagarnine, T, Taki, W, Kimura, J and Shibasaki, H (1999b) Cognitive motor control in human pre-supplementary motor area studied by subdural recording of discrimination! selection-related cortical potentials. Brain, 122: 915-931. Ikeda, A, Miyamoto, Sand Shibasaki, H (2002) Cortical motor mapping in epilepsy patients: information from subdural electrodes in presurgical evaluation. Epilepsia, 43 (Suppl. 9): 56-60. Kaji, R, Ikeda, A, Ikeda, T, Kubori, T, Mezaki, T, Kohara, N, Kanda, M, Nagarnine, T, Honda, M, Rothwell, J, Shibasaki, H and Kimura, J (1995) Physiological study of cervical dystonia: task-specific abnormality in contingent negative variation. Brain, 118: 511-522. Kornhuber, HH and Deecke, L (1965) Hirnpotentialanderungen bei Willkiirbewegungen und passiven Bewegungen des Menschen: Bereitschaftspotential und reafferente Potentiale. Pfiugers Archiv., 284: 1-17. Kunieda, T, Ikeda, A, Ohara, S, Yazawa, S, Nagarnine, T, Taki, W, Hashimoto, Nand Shibasaki, H (2000) Different activation of pre-supplementary motor area (pre-SMA), SMA-proper and primary sensorimotor area depending on the movement repetition rate in humans. Exp. Brain Res., 135: 163-172.

ELECTROCORTICOGRAPHY IN MOTOR CONTROL AND MOVEMENT DISORDERS

Lamarche, M, Louvel, J, Buser, P and Rector, I (1995) Intracerebral recording of slow potentials in a contingent negative variation paradigm: an exploration in epileptic patients. Elecroencephalogr. Clin. Neurophysiol., 95: 268-276. Lim, SH, Dinner, DS and Liiders, HO (1996) Cortical stimulation of the supplementary sensorimotor area. In: H Liiders (Ed.), Advances in Neurology. Supplementary sensorimotor Area. Lippincott-Raven, Philadelphia, Vol. 70, pp. 187-197. Lombroso, CT (1995) Paroxysmal kinesigenic choreoathetosis: an epileptic or non-epileptic disorder? Ital. J. Neurosci., 16: 271-277. Lombroso, CT (2000) Nocturnal paroxysmal dystonia due to a subfrontal cortical dysplasia. Epileptic Disord., 2: 15-20. Lombroso, CT and Fischman A (1999) Paroxysmal nonkinesigenic dyskinesia: pathophysiological investigation. Epileptic Disord., 2: 187-193. Lueders, H, Bustamante, L, Zablow, Land Goldensohn, ES (1981) The independence of closely spaced discere experimental spike foci. Neurology, 31: 846-851. Liiders, H, Lesser, RP, Dinner, DS, Morris, HH, Hahn, JF, Friedman, L, Skipper, G, Wyllie, E and Friedman, D (1987) Commentary: Chronic intracranial recording and stimulation with subdural electrodes. In: J Engel Jr (Ed.), Surgical Treatment of the Epilepsies. Raven Press, New York, pp. 297-321. Morris, HH III, Liiders, H, Hahn, JF, Lesser, RP, Dinner, DS and Estes, ML (1986) Neurophysiological techniques as an aid to surgical treatment of primary brain tumors. Ann. Neurol., 19: 559-567. Nagamine, T, Kaji, R, Suwazono, S, Hamano, T, Shibasaki, H and Kimura, J (1992) Current source density mapping of somatosensory evoked responses following median and tibial nerve stimulation. Elecroencephalogr. Clin. Neurophysiol., 84: 248-256. Neshige, R, Liiders, H and Shibasaki, H (1988) Recording of movement-related potentials from scalp and cortex in man. Brain, 111: 719-736. Ohara, S, Ikeda, A, Kunieda, T, Yazawa, S, Baba, K, Nagamine, T, Taki, W, Hashimoto, N, Mihara, T and Shibasaki, H (2000) Movement-related change of electrocorticographic activity in human SMA proper. Brain, 123: 1203-1215. Penfield, Wand Jasper, H (1954) Epilepsy and functional anatomy of the human brain. Little, Brown and Company, Boston, 896 pp. Pfurtscheller, G and Cooper, R (1975) Frequency dependence of the transmission of the EEG from cortex to scalp. Elecroencephalogr. Clin. Neurophysiol., 38: 93-96.

43

Picard, N and Strick, P (1996) Motor area of the mesial wall: a review of their location and functional activation. Cerebr. Cortex, 6: 342-353. Rector, I, Feve, A, Buser, N, Bathien, N and Lamarche, M (1994) Intracranial recording of movement-related readiness potentials: an exploration in epileptic patients. Elecroencephalogr. Clin. Neurophysiol., 90: 273-283. Rector, I, Louvel, J and Lamarche, M (1998) Intracranial recording of potentials accompanying simple limb movements: a SEEG in epileptic patients. Elecroencephalogr. Clin. Neurophysiol., 107: 277-286. Rohrbaugh, JW, Syndulko, K and Lindsley, DB (1976) Brain wave components of the contingent negative variation in human. Science, 191: 1055-1057. Sasaki, K and Gemba, H (1991) Cortical potentials associated with voluntary movements in monkeys. Elecroencephalogr. Clin. Neurophysiol., Suppl. 42: 80-96. Shibasaki, H, Barrett, G, Halliday, E and Halliday, AM (1980) Components of the movement-related cortical potentials and their scalp topography. Elecroencephalogr. Clin. Neurophysiol., 49: 213-226. Shibasaki, H and Ikeda, A (1996) Generation of movement-related potentials in the supplementary sensorimotor area. In: HO Liiders (Ed.), Supplementary SensorImotor Area. Advances in Neurology. Lippincott-Raven, New York, Vol. 70, pp. 117-125. Simson, R, Vaughn, HG Jr and Ritter, W (1977) The scalp topography of potentials in auditory and visual go/nogo tasks. Elecroencephalogr. Clin. Neurophysiol., 43: 864-875. Szentagothai, J (1969) Architecture of the cerebral cortex. In: HH Jasper, AA Ward and A Pope (Eds.), Basic Mechanisms of the Epilepsies. Little, Brown, Boston, pp.13-28. Talairach, J and Bancaud, J (1974) Stereotactic exploration and therapy in epilepsy. In: PJ Vinken and GW Bruyer (Eds.), Handbook of Clinical Neurology. NorthHolland, Amsterdam, Vol. 15, pp. 758-782. Tecce, 11 and Cattanach, L (1993) Contingent negative variation. In: E Niedermeyer and F Lopes da Silva (Eds.), Electroencephalography. Basic Principles, Clinical Applications and Related Fields. Wiliams & Wilkins, Baltimore, pp. 887-910. Towle, VL, Carder, RK, Khorasani, L and Lindberg, D (1999) Electrocorticographic coherence pattern. J. Clin. Neurpohysiol., 16: 528-547. Tyner, FS, Knott, JR and Mayer, WB Jr (1983) Electrodes. In: Fundamentals of EEG Technology. Basic Concepts and Methods. Raven Press, New York, Vol. 1, pp. 126-135. Walter, WG, Cooper, R, Aldridge, VJ, McCallum, WC and Winter, AL (1964) Contingent negative variation: an

44

electrical sign of sensorimotor association and expectancy in the human. Nature, 203: 380-384. Yazawa, S, Ikeda, A, Terada, K, Mirna, T, Mikuni, N, Kunieda, T, Taki, W, Kimura, J and Shibasaki, H (1997) Subdural recording of Bereitschaftspotential is useful for functional mapping of the epileptogenic motor area: a case report. Epilepsia, 38: 245-248.

A. IKEDA

Yazawa, S, Ikeda, A, Kunieda, T, Ohara, S, Mirna, T, Nagamine, T, Taki, W, Hori, T and Shibasaki, H (2000) Human pre-supplementary motor area is active before voluntary movement: subdural recording of Bereitschaftspotentials from mesial frontal cortex. Exp. Brain Res., 131: 165-177.

Movement Disorders Handbook of Clinical Neurophysiology, Vol. 1 M. Hallett (Ed.) © 2003 Elsevier B. V. All rights reserved

45 CHAPTER 5

Somatosensory evoked responses Francois Mauguiere" Department of Functional Neurology and Epileptology, Neurological Hospital, 59 Boulevard Pinel, 69330 Lyon, France

Modem history of clinical somatosensory evoked potentials (SEPs) testing began fifty years ago with George Dawson's recordings, in patients with myoclonus, of what is known today as a 'giant' somatosensory cortical response (Dawson, 1947). Thus, the first clinical application of SEP was to elucidate the mechanism of a movement disorder. Cortical SEPs were the first studied in normal subjects and patients, while spinal and sub-cortical far-field potentials were identified in the seventies and early eighties. The most recent advances in our knowledge of cortical responses to somatosensory stimulation are issued from development of multichannel recordings of SEPs and somatosensory evoked magnetic fields (SEFs) coupled with source localization in the 3D images of brain volume, as provided by magnetic resonance imaging (MRI). Studies based on electro-clinical correlations in patients with focal lesions have identified a single generator for each SEP component, a view compatible with the concept that the processing of somatosensory inputs by the nervous system is based on a sequential activation of fibers and synaptic relays. Source localization studies rather suggest that the evoked field recorded on the scalp surface at a given moment often results from activities of multiple distributed sources overlapping in time (Mauguiere et aI., 1997). This model fits better with the parallel activation and feedback loops characterizing the process of sensory inputs at the cortical level. In this chapter, we will focus on the early SEP components that can be helpful for analyzing the somatosensory control of movement in normal subjects. Similarly we will consider only SEPs' abnormalities observed in patients presenting with movement disorders.

* E-mail address:[email protected]

Tel.: 33 (0) 472 35 7106; fax: 33 (0) 472 35 7105.

5.1. Recording procedures Only the essential technical requirements for SEPs recording will be given here. More technical details concerning electrodes, amplifiers, stimulators, and safety can be found in several book chapters (Spehlmann, 1985; Chiappa, 1990; Halliday, 1993; Mauguiere, 1995, 1999) and in guidelines published by the International Federation of Clinical Neurophysiology (Mauguiere et aI., 1999). 5.1.1. Stimulus types and peripheral fibers involved in response

SEPs can be evoked either by electric stimulation of different categories of nerves according to their content in skin, joint, muscle afferent fibers, or by natural stimuli. Since one of the applications of SEPs in movement analysis consists in assessing somatosensory feedback control mechanisms of movement, it is of uppermost importance to know which type of fibers is involved in producing the response. Furthermore, when stimulating a nerve containing several fibers of different origins, diameters and conduction velocities, interference and occlusion phenomena may occur. Lastly the different types of somatosensory inputs will not necessarily produce all of the SEP components. This last point will be detailed in the description of the normal components of the response (see below). 5.1.1.1. Electrical stimuli Brevity and strict control in time of stimulus onset and cutoff, and thus of averaging trigger, are the major advantages of electrical stimulation over any other types of stimulus. SEPs are usually evoked by bipolar transcutaneous electrical stimulation applied on the skin over the trajectory of peripheral nerves. Monophasic electrical pulses of 100-300 u.s can

46

be delivered through rectangular, disk or needle electrodes connected respectively to the negative (cathode) or positive (anode) pole of the stimulator. Electrical stimuli thus by-pass the peripheral encoding of natural stimuli (pressure, vibrations, joint movements) by the receptors. Electric simulation of pure sensory nerves, such as digital nerves for the upper limb and sural nerve for the lower limb activates exclusively skin and joint peripheral and dorsal column fibers. It is advisable to use this type of stimulus in any attempt to correlate the quality of perception with SEP data in normal subjects as well as in patients with impaired sensation (Mauguiere et al., 1983a). At the upper limb stimulation of the finger tip (Desmedt and Osaki, 1991) or distal phalanx of fingers (Restuccia et al., 1999) activates selectively skin fibers, while stimulation of digital nerves at the level of the first or second phalanx concern both joint and skin afferents. In most clinical applications the electric stimulus is applied to mixed nerve at intensities equivalent to 3 to 4 times the sensory threshold. This type of stimulus produces a twitch in the muscles innervated by the stimulated nerve when it contains a contingent of motor fibers. At these stimulus intensities all rapidly conducting large myelinated fibers, including fibers subserving touch and joint sensation but also muscle afferents, are activated. The contribution of muscle afferents to cortical SEPs after stimulation of upper limb mixed nerves remains a matter of controversy. It is considered as negligible by Halonen et al. (1984) and Kunesch et al. (1995). However selective intrafascicular stimulation of hand muscle afferents at the wrist and motor-point stimulation were shown to produce short-latency cortical SEPs with the same waveforms (Gandevia et al., 1984; Gandevia and Burke, 1988) and modeled dipolar sources (Restuccia et al., 2002) as those obtained by stimulating the trunk of a mixed nerve. These cortical responses were obtained also after motor point stimulation of proximal upper limb muscles (Gandevia and Burke, 1988) and muscles of the trunk (Gandevia and Macefield, 1989). They are relatively small as compared with potentials evoked from the whole mixed nerve or digital nerves because only relatively few afferents are activated. Their first cortical' component peaks later than for median nerve stimulation at the wrist and at approximately the same latency as for

F. MAUGUIERE

stimulation of the digital nerves (Gandevia and Burke, 1990). Therefore the possibility remains that, when stimulating a mixed nerve of the upper limb such as the median nerve, the cortical response to muscle afferents inputs could be gated by the response to cutaneous and joint inputs. To conclude, the contribution of muscle afferents to median nerve SEPs can be considered as weak as compared to that of cutaneous afferents. The situation is quite different after electric stimulation of lower limb sensory-motor nerves such as the posterior tibial nerve for which muscle afferents have been shown to have a major contribution to cortical SEPs (Burke et al., 1981, 1982: Gandevia et al., 1984; Macefield et al., 1989). The latency of the earliest cortical response is shorter after stimulation of abductor hallucis muscular fascicle than after stimulation of the posterior tibial nerve and its amplitude is approximately half that of the tibial nerve response (Burke et al., 1981). It has been shown that, after stimulation of the tibial nerve, inputs conveyed by muscle afferents, which have the fastest conduction velocity, are able to occlude the response to cutaneous inputs (Burke et al., 1982). Because of this gating the cutaneous afferent volley may make little, or no contribution at all, to the cerebral potential evoked by electric stimulation of lower limb mixed nerves. Paired electrical stimuli delivered to the same nerve at various interstimulus intervals can be used to characterize distinct components of a given response according to their refractoriness after the response to the first stimulus of the pair (conditioning stimulus), and also to combine SEP testing with psycho-physical evaluation of time discrimination performances in detecting the second stimulus of the pair (El Kharroussi et al., 1996). The voltage of the response to the second stimulus of the pair is evaluated by subtracting the response to the test stimulus delivered alone in a separate session. The effect on SEPs of spatial interference between two electrical stimuli delivered simultaneously but at two distinct sites has been first described by Burke et al. (1982) and Gandevia et al. (1983). These authors showed that after simultaneous stimulation of both index and middle fingers of the same hand the size of the early component of the cerebral potential is less than predicted by simple addition of the potentials produced by stimulation of the fingers individually. A procedure based on this observation has been

47

SOMATOSENSORY EVOKED RESPONSES

recently applied to the study of somatosensory processing in patients with dystonia (Tinazzi et al., 2000). The interfering influences of sensory and motor events on electrically evoked SEPs have been widely used in the context of gating paradigms (see Cheron et al., 2000 for a review). In particular, models of centrifugal and centripetal SEP gating during movement programming or execution are based on the results from these studies (see below). 5.1.1.2. Natural stimuli Physiological stimuli have a higher selectivity than that of electrical stimulation of sensory nerves. SEPs and somatosensory evoked magnetic fields (SEFs) can be obtained in response to a brief mechanical impact on the finger tip (Debecker and Desmedt, 1964; Halliday and Mason, 1964; Nakanishi et al., 1973; Pratt and Starr, 1981; Kakigi and Shibasaki, 1984) or air puffs (Schieppati and Ducati, 1984; Hashimoto, 1987; Forss et al., 1994). For skin stimulation there is some difficulty in getting a consistent quick and well defined movement of the striker using an electro-mechanical device; this is why stimulators producing fast rise-time (1 ms) and short duration (4-5 ms) air puffs have been developed. However, due to the small population of excited fibers, these mechanical stimuli produce low voltage responses and are not used routinely in the clinical setting, in spite of their potential advantage over conventional electrical stimuli. Selective stimulation of joint afferent fibers using passive movements of fingers is difficult to achieve mostly because of possible interference between activations of joint, tendinous and muscle stretch receptors. Dedicated devices have been developed in order to activate selectively finger joint afferents (Desmedt and Osaki, 1991) or muscle afferents (Mirna et al., 1996). These stimulators produce a small and brisk passive flexion movement of finger II, or fingers II and III, proximal interphalangeal joint, with an amplitude 2° (Desmedt and Osaki, 1991) or 4 ° (Mirna et al., 1996) in 25 ms. In spite of close similarities between stimulation paradigms and evoked responses in these two studies, authors are diverging about the nature of afferent peripheral fibers involved in cortical response genesis, although they discard the hypothesis of a significant participation of cutaneous fibers. Desmedt and Osaki (1991) considered that the small 2° flexions they applied to

the finger with the wrist in slight extension were unlikely to activate the spindle afferents of finger extensor muscles. Mirna et a1. (1996) opposed the argument that joint afferents discharge only in extreme flexion or extension positions while a 4° flexion of the finger produces a muscle stretch in extensor muscle far larger than that needed to record evoked cortical response in monkeys (Phillips et al., 1971). Moreover they observed that responses to finger flexion were preserved after anoxic ischemia blocking the conduction of joint and cutaneous fibers from the moved finger, and concluded that muscle afferents from finger extensor muscles were responsible for cortical evoked responses. Lastly this latter observation is in partial contradiction with that of an abolition of cortical evoked responses to larger finger displacements (up to 45°) after anoxia of the hand reported earlier by Papakostopoulos et a1. (1974). Thus, although it is obviously difficult to be sure of the fibers involved during passive finger movements, this natural stimulus has the clear advantage over electrical stimulation to reproduce the physiological condition of the proprioceptive control of movement and would merit a larger utilization for exploring the pathophysiology of movement disorders. As performed routinely in Clinical Neurophysiology laboratories, SEPs do not explore the small myelinated Ao or unmyelinated C afferents to spinothalamic tracts, which subserve temperature and pain sensation. Since the report by Halliday and Wakefield (1963) it has been known that SEPs to electrical stimulation of the median nerve are normal in patients with lesions of the spino-thalamic tract and loss of pain and temperature sensation. Selective activation of Ao and C fibers can be achieved by brief heat pulses delivered by a CO 2 laser beam applied to the skin surface (Bromm and Treede, 1987). Since there is no evidence that slow conducting peripheral fibers are involved in movement control, we will not report on these potentials in this chapter. 5.1.2. Stimulus rate Stimulus rates of 1 to 10 s are commonly used in clinical neurophysiology laboratories with no significant changes in the amplitudes or latencies of subcortical or primary sensory cortex (SI) responses. These rates are considered as acceptable for clinical

48

studies in most textbooks (Chiappa, 1990; Halliday, 1993). However, lower rates are recommended to study the amplitudes of cortical potentials peaking more than 25 ms after stimulation of the upper limb. At rates over 3 s even early cortical potentials peaking after the initial SI response can be depressed, in particular the frontal N30 potential which is one of the SEPs most often reported as affected in movement disorders. The amplitudes of middle and long latency responses, in particular those generated in the SII area, reach their saturation level at stimulus rates far below 1 s (Hari et aI., 1984). The effect of stimulus rates on the amplitude of cortical SEPs can be used as a means to distinguish the components one with another, this approach has been applied to identify short latency SEPs to median nerve (Garcia-Larrea et al., 1992) and tibial nerve (Tinazzi et al., 1996a) stimulation.

5.1.3. Analysis time, sampling rate and filters

Most of the SEP components which have proved their clinical utility peak before 50 and 100 ms respectively for upper and lower limb stimulation. Consequently there is no necessity to record SEPs over analysis times longer than 100 ms in routine except for the recording of middle latency responses generated in SII (Hari et al., 1983, 1984), posterior parietal (Forss et al., 1994) and mesial cortical areas (Forss et al., 1996) Most commercially available devices can calculate 512 sampling points over analysis times in the 10-100 ms range. This gives bin widths of 97 and 195 f..LS for analysis times of 50 and 100 ms, which are appropriate to obtain far-field and short-latency cortical SEPs without aliasing, respectively (see Spehlmann, 1985). A band-pass, with a high-pass filter set at less than 3 Hz and a low pass filter set over 2000 Hz, is optimal to record without distortion all early SEP components including fast far-field potentials. A lower high cut-off frequency down to 200 Hz is acceptable for the recording of SEPs peaking later than 50 ms for upper limb stimulation. Analog filter rolls off should not exceed 12 and 24 dB/octave for low and high frequency filters respectively. Digital filtering of responses acquired with a broad band pass filter offers the possibility of selecting off-line frequencies of clinical interest.

F. MAUGUIERE

5.1.4. Gain and number of sweeps

Gains between 100,000 and 200,000 are adequate. The number of sweeps to be averaged is variable according to the signal/noise ratio of the different SEP components. In practice 500 sweeps are enough, in most instances, to obtain the responses at Erb's point, as well as to extract the main early cortical components. For the recording of spinal and scalp far-field EPs often up to 1000-2000 sweeps must be averaged. 5.1.5. Placement of the recording electrodes

When recording SEP, one usually aims to study during the same run peripheral, spinal, brainstem and early cortical SEPs. Peripheral nerve volleys of action potentials as well as conducted and segmental spinal SEPs can be recorded by placing cutaneous electrodes along the route of the fibers tracts or close to the fixed generators of segmental spinal responses (near-field recording). Electrodes placed on the scalp pick up both SEPs generated in the cortex which are picked up as near-field responses located in restricted areas and far-field positivities reflecting the evoked activity generated in peripheral, spinal and brainstem somatosensory fibers. Standard disk electrodes, well attached on the skin and scalp surface, are suitable for most types of recordings; needle electrodes can also be used. The routine four-channels montages proposed in the IFCN guidelines (Mauguiere et al., 1999) explore the afferent peripheral volley (in the supra-clavicular and popliteal fossa for upper and lower limb, respectively), the segmental spinal responses at the neck and lumbar spine levels, as well as the subcortical far-field and early cortical SEPs using scalp electrodes placed in the parietal and frontal regions for upper limb SEPs and at the vertex for lower limb SEPs. MUltiple channel recordings (24, 32, 64 or more) permit to assess the scalp topographic distribution of SEPs; they can be obtained using a cap with electrodes placed according to the modified tentwenty electrode system for SEPs, or whole-head magneto-encephalography (MEG) devices for SEFs. These multiple-channel recordings, eventually assisted by mapping based on computerized voltage or current fields interpolation, have brought invaluable information on the spatial distribution of each component of the response; moreover they are indispensable for calculating the evoked fields

49

SOMATOSENSORY EVOKED RESPONSES

sources coordinates in a spherical or MRI-based realistic model of the head. The SEP literature is full of discussions about the most appropriate site for the reference electrode to record each of the components of the response; in fact the knowledge of the field distributions and of supposed source locations and orientations is enough for a correct choice of the recording derivation. Considering the field distribution, the optimal recording condition is in theory that where the reference is not influenced by the activity under study. Most of the far-field potentials that have proved to be clinically useful are widely distributed over the scalp. Consequently they reach their maximal amplitude in referential recordings in which the reference electrode is placed in a noncephalic position and they are cancelled, or drastically reduced, when both recording electrodes are placed on the scalp. Furthermore a non-cephalic reference common to all channels is also adequate for all near-field recordings and there is no theoretical argument against the recommendation of using this type of montage in all situations. However the electrical and physiological (ECG, EMG ...) noise level may be increased by the long distance between the active and reference electrode in non-cephalic reference montages. Conversely, differential amplification between the two maxima of a dipolar potential field on the scalp produced by an identified single generator increases the signal voltage and reduces the noise level. Therefore montages proposed for routine recordings often stem from a compromise between: (i) the theoretical optimum represented by a common non-cephalic reference; and (ii) practical considerations in terms of pertinent components to be studied in a given clinical condition and electrical environment. 5.1.6. Placement of the ground electrode

To minimize the electrical artifact produced by the stimulation, the ground electrode should be placed on the stimulated limb between the stimulation site and the recording electrodes. Flexible metal strips covered with a saline-soaked cloth wrapped around the limb close to the stimulus site are recommended. Electrically isolated stimulators allow the use of a ground electrode on the head which is adequate for eliminating artifact related to the electrical main and radiofrequency interference.

5.2. Normal responses Short latency SEPs to upper limb stimulation (electrical stimulation of median, ulnar, radial or finger nerves)

The main early SEP components will be described as they appear on recordings using a non-cephalic reference electrode placed at the shoulder on the non-stimulated side after stimulation of the median nerve at the wrist (Fig. 1). When necessary we will discuss the modifications of waveform and latency related to the use of other reference electrode sites.

7

6

5

4

3

, P13 "spinal"

~ Nl-o-- N11

2_~

~ tN9

L-

~35msK

Fig. 1. Normal median nerve SEPs. This figures illustrates normal responses recorded using a non-cephalic reference montage. Note that the segmental response peaks as a negative N13 at Cv6 spinous process (channel 2) and as a positive spinal Pl3 at the anterior cervical recording site (channel 3). In this subject the four far-field scalp positive potentials P9, PH, Pl3 and Pl4 are well individualized (channels 4-7); the Nl8 potential recorded in the parietal region ipsilateral to stimulus looks biphasic because of the large diffusion of the parietal P27 potential (channel 4) ; the N24 potential is hardly visible on the ascending phase of the frontal N30 (channel 7) between the P20 and N30 peaks.

50

5.2.1. Peripheral, spinal and sub-cortical potentials 5.2.1.1. The peripheral N9 potential Stimulation of a mixed nerve such as the median nerve elicits a compound action potential (CAP) reflecting the peripheral ascending volley that can be recorded at different levels of forearm and arm. In most routine SEP recordings the peripheral ascending volley is recorded only at Erb's point in the supraclavicular fossa ipsilateral to stimulation. The CAP appears as a triphasic positive-negative-positive waveform with a negative peak culminating at about 9 ms in normal subjects (N9). The reference electrode can be placed at the contralateral Erb's point (EPc) or at Fz site in the frontal region of the scalp. With the EPc reference, only the N9 triphasic wave is recorded, with the Fz reference it is followed by a Nl4 negativity and a P30 positivity which correspond to the P14 and N30 potentials recorded at Fz, respectively (see below). When a limited number of channels is available the EPi-Fz derivation thus permits to assess the far-field P14 and frontal N30 potentials. Median nerve CAP are a mixture of motor antidromic and sensory orthodromic responses, and thus are qualitatively different from SEPs generated in the somatosensory pathways. The amplitude of peripheral CAPs are unaffected by interfering stimuli applied on the stimulated hand or when the stimulus rate is increased up to 50 Hz (Ibanez et aI., 1989). 5.2.1.2. The spinal segmental N13 potential Electrical stimulation of large diameter myelinated fibers of peripheral nerves produces a dorsal horn (DH) potential with a posterior-negative and anterior positive dipolar field perpendicular to the cord axis. This field distribution ofDH potentials has been studied by direct spinal recordings at all segmental levels of the cord. A comprehensive atlas of these responses is now avalaible (Dimitrijevic and Halter, 1995). The cervical N13 potential is recorded at the posterior neck, with a maximum voltage at the level of Cv5-Cv7 spinous processes and decreases in amplitude at more rostral or caudal electrode positions. N 13 does not show any latency shift between lower and upper cervical levels after median nerve stimulation in' normal subjects (Desmedt and Cheron, 1980a; Mauguiere, 1983; Desmedt and Nguyen, 1984). When recorded ante-

F. MAUGUIERE

rior to the cord by an electrode placed at the anterior aspect of the neck (Desmedt and Cheron, 1980a; Mauguiere and Ibanez, 1985) the cervical DH response is recorded as a spinal P13 positivity. This polarity reversal has been demonstrated with esophageal (Desmedt and Cheron, 1981a) and epidural (Cioni and Meglio, 1986) recordings and also by direct recordings from the surface of the cervical cord (Jeanmonod et aI., 1989, 1991). Both N13 and P13 spinal components are reduced in amplitude to the same degree when the stimulation rate is increased over 10 s (Ibanez et aI., 1989), both persist in brain death or in cervico-medullary lesions (Mauguiere et aI., 1983b; Buchner et aI., 1988) and both are selectively reduced or abolished in lesions of the cervical cord gray matter sparing the spinal somatosensory tracts (Restuccia and Mauguiere, 1991; Ibanez et aI., 1992). Thus, the most likely generator of the N13/P13 cervical potentials is the compound segmental post-synaptic potential triggered in the dorsal horn gray matter by the afferent volley in fast conducting myelinated fibers. A cervical transverse derivation between two electrodes located respectively over the Cv6 spinal process and at the anterior aspect of the neck (anterior cervical reference, AC) above the laryngeal cartilage is the most adapted for recording the spinal N13/P13 segmental response. Subtraction of the anterior cervical positivity from the posterior neck negativity increases the signal/noise ratio and has the advantage, over the more conventional Cv6-Fz montage, to record a segmental DH response that is not contaminated by SEP components generated above the foramen magnum (Mauguiere and Restuccia, 1991). Apart from its field distribution a second characteristic of the segmental N13 potential is that it shows amplitude reduction, rather than latency prolongation, when the dorsal afferent volley is time dispersed, or when the number of responding DH neurons is reduced. For diagnostic purposes the limitation of N13 amplitude measurements is that distribution of this parameter in a group of normal subjects does not show a Gaussian profile. Therefore, the lower limit of normal N13 amplitude values estimated by the mean minus 2 to 3 standard deviations (SD) is statistically meaningless. The best way to evaluate the N13 amplitude is to calculate the ratio between the N13 amplitude and that of the dorsal root N9/P9 deflection,which immediately

SOMATOSENSORY EVOKED RESPONSES

precedes N13 in transverse cervical montage recordings (Restuccia and Mauguiere, 1991). The distribution of the loglO(N131N9) amplitude ratio has a Gaussian profile in normals. The normal lower limit of the normal N131N9 ratio, defined as the mean value minus 2.5 SD, is of about 1.2. The N91P9 deflection shows a good test-retest and its amplitude closely reflects the incoming volley in the cervical roots. Moreover the N131P9 amplitude ratio remains stable when tested at stimulus intensities between motor threshold and twice the motor threshold. Lastly normal values of the amplitude ratio between N13 and the preceding N9-P9 have been published for median, ulnar and radial nerve stimulation (Restuccia and Mauguiere, 1991; Restuccia et aI., 1992) with this derivation. 5.2.1.3. The upper cervical NJ3 potential It has been known for many years that a negative potential peaking at a latency of about 13 ms can be recorded intraoperatively at the dorsal aspect of the cervico-medullary junction (Andersen et aI., 1964; Allison and Hume 1981; Lesser et al 1981; Meller et aI., 1986; Morioka et aI., 1991). On the skin surface however, it is uneasy after median nerve stimulation to separate this potential at upper cervical level from the segmental cervical N13, so that the former can be interpreted as a rostral spread of the latter. In particular there is no evidence of a latency increase of the median nerve N13 negativity from Cv6 to Cv2 level in non-cephalic reference recordings, although in bipolar recordings along the neck the mean N13 peak latency was found to be longer at rostral than at caudal levels (Kaji and Sumner, 1987). An elegant insight to this latency problem was provided by Zanette et al. (1995) who showed that the N13 recorded at upper cervical level peaks 0.8 ms later than the Ic-N13 after ulnar nerve stimulation, but not after median nerve stimulation, in non-cephalic reference recordings. The greater distance from the dorsal root entry zone to the cuneate nucleus for ulnar nerve than for median nerve afferent fibers was proposed to explain this latency shift. More recently Araki et al. (1997) have addressed the question of separate generators for the lower and upper cervical N13 potentials by using a conditioning-test paired stimulus paradigm. They confirmed that at interstimulus intervals of 4-18 ms the test median nerve lower cervical N13 to the second stimulus was attenuated by 2-34% when compared to the control

51

response, as previously reported by Iragui (1984) after median nerve stimulation and by EI-Kharoussi et al. (1996) after stimulation of the 2nd and 3rd fingers. More unexpected was the finding that, while the scalp P13-P14 showed the same attenuation as the lower cervical N13, the N13 recorded at the Cv2 level showed an increase of 4-25% at these lSI. The upper cervical N13 is supposed to be generated by the pre-synaptic volley in the DC fibers close to the cuneate nucleus, but the difficulty for separating it from the dorsal hom N13 lessens its clinical reliability in routine clinical recordings (see Mauguiere, 2000, for a review). 5.2.1.4. The conducted cervical N11 potential The NIl potential is recorded all along the posterior aspect of the neck, where it usually appears to encroach upon the ascending slope of N13. In non-cephalic reference recordings, its onset latency increases from Cv6 to Cv 1 spinal processes by 0.9±0.15 ms (Desmedt and Cheron, 1980a; Mauguiere, 1983). This bottom-up shift of NIl onset latency suggests that Nil is generated by the ascending volley of action potentials in the dorsal columns of the cervical spinal cord. As first reported by Cracco (1973), it is often difficult to differentiate the Nil from the following N13 component. This seriously hampers the use of NIl in clinical practice, which is limited to cervical cord lesions obliterating selectively the N13 potential. In direct recordings on the cervical cord surface during surgery, Nil appears as a fast polyphasic component which overlaps in time with the slower N13 segmental post-synaptic potential. 5.2.1.5. The far-field positive scalp potentials (P9, PlJ, PJ3, P14) On scalp non-cephalic reference recordings with a high frequency filter over 1000 Hz and a fast sampling rate (bin width ~200 us), three or four stationary positive potentials are consistently observed with a wide distribution and a mid-frontal predominance. These potentials have in common to be generated by sub-cortical structures (Craceo and Cracco, 1976; Nakanishi et aI., 1978; Anziska and Cracco, 1980; Desmedt and Cheron, 1980a; Yamada et al., 1980; Mauguiere and Courjon, 1981). In normal adults these potentials peak with mean latencies of 9, 11, 13 and 14 ms respectively and are labeled P9, P11, P13 and P14 according to the

52 polarity-latency nomenclature. There is some interindividual variation in the waveform of the two latest Pl3 and P14 potentials since either the Pl3 or P14 peaks can predominate. The amplitude of these farfield positive potentials is not affected when the stimulation rate is increased up to 50 Hz (Ibanez et al., 1989). Due to their widespread distribution on the scalp, far-field positive SEPs must be recorded with a noncephalic reference electrode. The best location for the active electrode is the medio-frontal region; however, when only a limited number of channels is available, it is recommended to place the active electrode in the parietal region contralateral to stimulation in order to record also the early parietal SEPs. The P9 potential is picked up at the neck as well as on the scalp. It reflects the afferent volley in the trunks of the brachial plexus in axilla and supra-clavicular fossa (Cracco and Cracco, 1976; Nakanishi et al., 1983). Its peaking latency varies with arm length and its onset latency, but not its peaking latency, is influenced by the respective positions of and trunk (Desmedt et al., 1983). The P11 potential reflects the ascending volley in the fibers of dorsal columns at the cervical level. Desmedt and Cheron (l980a) first reported that, after median nerve stimulation at the wrist, the P11 potential begins in synchrony with the cervical NIl potential at the Cv6 level, which is close to the dorsal root entry zone in the cervical cord. The P11 potential is not recorded in about 20% of normal controls and, thus the clinical significance of its absence in patients is questionable. Conversely its persistence when later scalp components are abnormal is a reliable indicator of preserved dorsal column function in patients with lesions located in the medulla oblongata or at the cervico-medullary junction. The P13-P14 far-field potentials are consistently recorded in normal subjects (see Restuccia, 2000, for a recent review). Both may be of similar amplitude, or either of them may be the larger of the two. In some subjects P14 is hardly visible as a notch on the ascending phase of Pl3, in others Pl3 and P14 cannot be differentiated. In the same individual the morphology of the Pl3-P14 complex can display some degree of side-to-side difference, which creates difficulties for interpreting left-right asymmetries in patients. P14, but not Pl3, always peaks

F. MAUGUIERE

later than the cervical segmental Nl3 potential (Mauguiere, 1987). Since the Pl3-P14 complex is picked up at the earlobe with a lesser amplitude than in the frontal region of the scalp, it can be recorded with a scalp-earlobe montage (Nakanishi et al., 1978). The P13-P14 complex is the only scalp far-field SEP which is reduced by interfering stimulations, such as vibrations, applied to the hand on the stimulated side (Ibanez et al., 1989). This suggests that it is generated after the synaptic relay in the nucleus cuneatus (NC). Patients with thalamic lesions usually show a normal P14 suggesting that this component originates at sub-thalamic level (Nakanishi et al., 1978; Anziska and Cracco, 1980; Mauguiere et al., 1982). Conversely in cervicomedullary lesions (Mauguiere et al., 1983b; Mauguiere and Ibanez, 1985; Yamada et al., 1986) or in brain-dead patients (Buchner et al., 1988) the P14 potential is absent. These findings suggest that the generator of P14 is situated above the level of the foramen magnum. Simultaneous scalp and naso-pharyngea1 SEP recording in deeply comatose and brain-dead patients brought some information concerning the origins of the P14 potential (Wagner, 1991). In patients with signs of upper brainstem dysfunction this author observed that the P14 recorded between scalp and naso-pharynx is lost while an earlier P13 positivity persists in scalp to shoulder traces. This observation suggests that the P14 recorded between the scalp and the naso-pharynx reflects the activity of medial lemniscus pathways at the upper part of the brainstem whereas the P13 might be generated at the cervico-medullary junction. A normal Pl3 coexisting with an abnormal P14 has been observed in patients with pontine or mesencephalic lesions (Nakanishi et al., 1983; Delestre et al., 1986; Kaji and Sumner, 1987; Mavroudakis et al., 1993), supporting the hypothesis that Pl3 and P14 potentials could be generated at different anatomical levels in the cervico-medullary and brainstem somatosensory pathways. This view is reinforced by the observation, in non-cephalic reference recordings, that the P13 positivity recorded at the naso-pharynx can be preserved while the scalp P14 is abnormal in lesions of the lower brainstem (Restuccia et al., 1995). The scalp and naso-pharynx Pl3 potential is different from the segmental spinal Pl3 recorded at the anterior aspect of the neck (see above). In spite

SOMATOSENSORY EVOKED RESPONSES

of their similar latencies and polarities, these two components have different distributions and origins and the segmental spinal Pl3 is preserved in all patients with upper cervical or cervico-medullary lesions. Thus, to our present knowledge the most probable generators of the P13-P14 potentials are: (i) the ascending volley in upper cervical DC fibers at the cervico-medullary junction or in medial lemniscus fibers in the brainstem; (ii) the post-synaptic response of NC neurons; (iii) junctional stationary potentials related with the moving action potential volley across the border between neck and posterior fossa (Kimura et al., 1983, 1984). For sake of clarity the P13-P14 complex is often labeled as "P14." It can be useful to calculate the P91 P14 amplitude ratio in patients. As the IIV amplitude ratio of Brainstem Auditory EPs the P9/P14 ratio compares the amplitude of a peripheral potential (P9) with that of a brainstem potential (PI4). In normal subjects the P9/P14 ratio is less than 1 when recorded with a broad band-pass of 1.6 to 3200 Hz, and its distribution has a Gaussian profile (GarcfaLarrea and Mauguiere, 1988). 5.2.1.6. The N18 scalp potential This potential identified by Desmedt and Cheron (1981b) is a long-lasting scalp negative shift that immediately follows P14 in non-cephalic reference scalp recordings. In normal subjects N18 can only be identified in the posterior parietal region ipsilateral to the stimulation, where there is no or minimal interference with cortical potentials. The N18 is a long lasting component (about 20 ms when recorded with a 1-3000 Hz band pass). The N18 negativity may contain several sub-components considering that several wavelets are superimposed on its plateau at fixed latencies in the same individual. In normal subjects non-cephalic reference recordings of scalp SEPs show two positive cortical potentials, which are superimposed upon N18 in the central and frontal regions contralateral to stimulation, namely central P22 and frontal P20 (see below). Thus, the early portion of N18 appears as a negativity preceding the onsets of P20 and P22 in the midfrontal and central regions respectively. These frontal and central negativities have been considered as genuine cortical responses and labeled as "NI5", "NI6", "NIT' or "NI9" components (Yamada et al., 1984; Iwayama et al., 1988). However super-

53

imposition of traces recorded in the posterior parietal region ipsilateral to stimulation, where Nl8 is not contaminated by contralateral cortical components, clearly shows that there is no scalp negativity other than N18 and N20 within the first 22 ms following stimulation. When all cortical responses are eliminated by deafferentation due to a lesion of the ventro-posterolateral (VPL) thalamic nucleus (Mauguiere et al., 1983c) or by a direct lesion of the centro-parietal cortex (Mauguiere, 1987), N18 can be recorded on the whole surface of the scalp. After hemispherectomy, which entails retrograde degeneration of the thalamo-cortical neurons, the N18 potential is most often preserved (Mauguiere and Desmedt, 1989). However it can be missing, as well as the P14 potential, long after hemispherectomy including the thalamus (Restuccia et al., 1996) in relation with retrograde degeneration of the lemniscal fibers projecting on the thalamus. These observations rule out the possibility that N18 could reflect a cortical or a thalamic response to lemniscal or extra-lemniscal inputs. Since N18 is absent, or grossly abnormal, in cervico-medullary lesions it is assumed that its generator is situated below the thalamus and above the foramen magnum. The earliest hypothesis, concerning the generation of N18, was that of a post-synaptic response of the brainstem nuclei connected with the dorsal column (DC) nuclei to ascending inputs conveyed by ancillary fibers of the medial lemniscus. This hypothesis has been supported by intra cerebral SEP recordings in man showing a stationary negativity between the upper pons and the midbrain, peaking at the same latency as the scalp N18 (Urasaki et al., 1990). However naso-pharyngeal recordings (Tomberg et al., 1991) rather suggested an N18 origin in the medulla oblongata and, more precisely in the nucleus cuneatus (NC). Potentials recorded dorsal and ventral to NC in humans (Morioka et al., 1991) show a biphasic waveform, which is very similar to that of NC responses to peripheral nerve stimulation in cats. The first phase of this response in cats represents the post-synaptic potential of NC relay neurons to the DC afferent volley, while the second one reflects the pre-synaptic inhibition of DC fibers terminals in NC (Andersen et al., 1964). This inhibition consists in a feedback depolarization of DC terminals, triggered by the afferent volley in DC fibers collaterals, and mediated by NC inter-neurons.

54

To date it is accepted that N18 reflects a similar phenomenon in the human brainstem (see Sonoo, 2000, for a review). Several arguments lend substance to this analogy: (i) The distribution and polarity are similar in cat and human recordings; (ii) The pre-synaptic inhibition of DC terminals in cats, like the human N18, lasts for several tens of milliseconds, probably because it is mediated via a poly-synaptic chain of inter-neurons; and (iii) The N18 voltage is unaffected by interfering vibrations applied on the hand on the stimulated side (Manzano et al., 1998), contrary to the P14 potential, which is attenuated (Ibanez et al., 1989); this finding fits with the inference that N18 reflects inhibitory activity upon DC terminals. 5.2.2. Early cortical potentials

These cortical SEPs are peaking in the 18-35 ms latency range and are all obtained with optimal voltage in response to stimuli delivered at a slow rate (less than 2 per second). They are recorded on the scalp in the parietal region contralateral to stimulation and in a large fronto-central area, mostly contralateral to stimulation. Though studies in normal subjects and patients converge on the conclusion that responses recorded in the parietal region originate in the somatosensory area SI, there is still some controversy as to whether some of the early cortical SEPs recorded in the frontal region might be generated in the precentral cortex. 5.2.2.1. The N20-P20 and P22 potentials Two sets of early cortical potentials are consistently recorded in normal subjects on the scalp contralateral to stimulation. The first one is made of a parietal N20 and frontal P20 dipolar field; the second is composed of a central P22 positive potential. These components are also present after finger stimulation with a latency delay of 2-3 ms, as compared with median nerve SEPs, because of finger to wrist conduction time. There is a consensus to consider that parietal N20 and frontal P20 potentials represent the earliest cortical potential elicited by median nerve stimulation and reflect the activity of a dipolar generator in Brodmann's area 3b, tangent to scalp surface and situated in the posterior bank of the rolandic fissure (Broughton, 1969; Goff et al., 1977; Allison et al., 1980). This dipolar field distribution is also observed in magnetic recordings.

F. MAUGUIERE

On scalp and some direct cortical recordings a P22 recorded in the central region was found to peak 1-2 ms later than the N20-P20 potentials (Desmedt and Cheron, 1980b, 1981b; Papakostopoulos and Crow, 1980). This has been taken as an argument to consider P22 as a genuine component generated by a source distinct from that of N20-P20 (Desmedt and Cheron, 1981b), while others denied the existence of an independent P22 (Allison, 1982). Sequential spatial maps of scalp recorded SEPs have shown that the N20-P20 dipolar field is followed in the central region contralateral to stimulation by a positive P22 field peaking 1.0-2.5 ms with little or no negative counterpart on the scalp (Deiber et al., 1986). This spatial distribution of P22 suggests that its source is radial to the scalp surface. Most studies of somatosensory evoked magnetic fields (SEFs) failed to confirm such a source in the central region probably because magnetic recordings are blind to magnetic fields produced by dipolar sources radial to scalp surface (Brenner et al., 1978; Okada et al., 1984; Wood et al., 1985). The question whether the radial source of the central P22 is located behind the central sulcus, in the primary somato-sensory area, or in front of it, in the primary motor area, is not easy to address by scalp recordings in normal subjects, even when assisted by dipole modeling techniques based on a spherical head-model (Franssen et al., 1992; Buchner et al., 1995). One argument in favor of a pre-central origin of the P22 potential is that it may be selectively lost in patients with hemispheric lesions who have normal sensations but signs of upper motor neuron dysfunction, and selectively preserved in patients with hemianesthesia or astereognosis and no motor deficit (Mauguiere et al., 1983a; Slimp et al., 1986; Tsuji et al., 1988; Mauguiere and Desmedt, 1991). However, some direct cortical recordings in monkeys and humans failed to identify a P22 in the precentral cortex (Allison et al., 1989, 1991; McCarthy et al., 1991) but individualized a surface positive potential that they labeled as P25, with a source radial to the scalp surface and located at the posterior edge of the rolandic fissure in Brodmann's area 1. Dipole modeling studies of SEPs using realistic head models and constraining the solution in the individual brain MRI volume supported the view that the P22 source is located at the crown of the post-central gyrus (Buchner et al., 1996). These data from

SOMATOSENSORY EVOKED RESPONSES

cortical recordings and modeling of scalp responses are challenged by other experimental data in monkeys; (i) short-latency unit response to finger stimulation has been recorded in the motor cortex (Lemon, 1981; Tanji and Wise, 1981); and (ii) an early median nerve potential has been recorded in the motor cortex (Brodmann's area 4) as well as in area 1 (Nicholson-Peterson et aI., 1995). According to the data of Nicholson-Peterson et al. (1995), the earliest cortical responses to upper limb stimulation are likely to result from the approximately coincident activation of at least three dipolar sources in areas 3b, 4 and 1 with orientations both opposing and orthogonal to each other. Sources in area 3b and 4 have opposite orientations tangent to the scalp surface, while the source located in area 1, at the crown of the post-central gyrus, is oriented radial to the scalp surface. Thus, even if it would be hazardous to imagine a strict homology between SEP components in monkey and man, due to the considerable anatomical differences between the two species, these experimental data support the contribution of the motor cortex to the N20-P20 and P22 scalp potentials. 5.2.2.2. The parietal P24 and P27 potentials These potentials are recorded in the parietal region contralateral to stimulation, their peaking latencies show large inter-individual variations between 24 and 27 ms. In some subjects two distinct P24 and P27 can be identified while only either of the two peaks is observed in others. This explains why, according to the polarity-latency nomenclature, the first parietal positive potential following N20 has received various labels in literature (P24, P25 or P27). These variations reflect the fact that the activities of several parietal sources overlap in time in this latency range. The P27 potential was found to be abnormal in patients with focal lesions of the parietal cortex, presenting with a astereognosis and normal N30 frontal responses (Mauguiere et aI., 1982; Mauguiere and Desmedt 1983a). No abnormality of these potentials has been specifically related to movement disorders, except their exaggeration in cortical myoclonus (see below). 5.2.2.3. Thefrontal "N30" potential (N24-N30 complex) The frontal potential, labeled as "N30" in most clinical studies, is picked up in the frontal region

55

contralateral to stimulation, however it often spreads to the mid-frontal region and to the frontal region ipsilateral to stimulus. Its waveform show two distinct components of which the earlier one, peaking at about 24 ms (N24), appears as notch on the ascending slope of the later one, which peaks at 30 ms (N30) and has the larger voltage of the two in normal young adults. There is presently some consensus on the conclusions that the frontal negativity with its two N24 and N30 peaks cannot be generated by a single source (Delberghe et aI., 1990; Garcia-Larrea et aI., 1992; Osaki et aI., 1996), and that the early part (N24) of the frontal negativity corresponds to the polarity reversal of a parietal P24 potential across the central sulcus. The latter statement is supported by several observations issued from intra-cranial and scalp recordings: (1) In the latency range of the scalp "N30" Allison et al. (1989), using intra-cranial recordings during surgery, have identified a dipolar field, frontal negative and parietal positive, and considered it as generated by a source in area 3b, tangent to the scalp surface. Probably because of anesthesia this field reached its maximum at 30 ms and was labeled N30-P30 by these authors. (2) On the scalp the field distribution is clearly dipolar at the peaking latency of the N24 with a parietal positivity and a mid-frontal negativity spreading to the fronto-central region. This dipolar potential field has been labeled N24-P24 (Garcia-Larrea et aI., 1992) or N27-P27 (Osaki et aI., 1996), it corresponds to the N30-P30 of Allison et al. (1989) and its distribution fits well with a dipolar source perpendicular to the rolandic fissure and tangent to the scalp. (3) The two negative frontal N24 and N30 components react differently to changes in stimulus rate (Delberghe et aI., 1990; Garcia-Larrea et aI., 1992; Valeriani et aI., 1998a). The early N24 remains unaffected if the stimulus frequency is increased up to 10Hz, while the later N30 decreases at stimulus rates higher than 1 per second and is virtually absent at 10 per second. (4) Source modeling studies of scalp responses converge to the conclusion that the distribution of the scalp N24-P24 can be explained by a single source in the posterior bank of the central

56

sulcus (Buchner et aI., 1995; Valeriani et aI., 1998b). (5) The administration of tiagabine, which enhances GABA inhibition, increases selectively the voltage of the N24-P24 dipolar field suggesting that it could reflect the repolarization of the neuronal population that produces the N20-P20 field in area 3b (Restuccia et aI., 2002). Conversely the origin of the main N30 negativity of the frontal N24-N30 is still a debated issue. Since it has been argued that N30 might reflect the pre-motor cortex participation in the long loop of trans-cortical reflexes, this question is of importance regarding the potential use of SEPs in movement disorders. Three early observations have supported the hypothesis that, contrary to the N241P24 dipolar field, N30 does not represent the frontal counterpart of the parietal P27: (1) N30 amplitude decreases with ageing whilst that

of P27 increases (Desmedt and Cheron, 1980b). (2) N30 can be selectively absent, or decreased, with preserved parietal P27 in focal hemispheric lesions causing pure motor symptoms with normal sensation (Mauguiere et aI., 1983a; Mauguiere and Ibanez, 1990). (3) N30 was found to be selectively reduced in Parkinson's disease with a decrease of the N30/ P27 amplitude ratio (see below). Based on these findings it has been proposed that the N30 and P27 potentials originate in the pre-central cortex and in SI (Brodmann's area I), respectively; each of these two dipolar sources being oriented perpendicular to the scalp surface. Some observations (see Cheron et aI., 2000, for a review) are compatible with the view that the N30 potential could be related with motor programming and generated in the pre-motor cortex of area 6, and more precisely in the supplementary motor area (SMA): (1) N30 was reported to be absent in a patient with a falx meningioma compressing the inner aspect of the pre-central frontal cortex (Rossini et aI., 1989). (2) Voluntary movements of the fingers on the stimulated side was shown to exert a gating effect on frontal SEPs by reducing selectively

F. MAUGUIERE

the amplitude of N30, without effect on parietal potentials (Cohen and Starr, 1985; Cheron and Borenstein, 1987, 1991; Rossini et aI., 1989; Rossi et aI., 2002). This effect, which had been observed and to a certain extent overviewed by Papakostopoulos et aI. in 1975, is less for simple than for complex fingers movements. Two mechanisms have been proposed to explain this selective gating of frontal SEPs by active movement. First, cortical motor neurons involved in fingers movements could also account for part of the frontal SEPs and would be unable to respond simultaneously to the afferent volley triggered by the electric stimulation. The second mechanism is that of a cortico-cortical inhibition by which the motor neurons active during movements would suppress the response of cortical neurons involved in the generation of the N30 potential. (3) A gating of N30, but also of N24, occurs during movement preparation in reaction time paradigms where SEPs are recorded before movement onset (Shimazu et aI., 1999). (4) The gating of N30 does not require that the intended movement be executed (Fig. 2) and also occurs when the subject mentally simulates a complex sequence of finger movements on the stimulated side (Cheron and Borenstein, 1992; Rossini et aI., 1996). When the same mental movement simulation (MMS) concerns the nonstimulated hand, or when the subject performs a mental task other than motor simulation, this gating effect does not occur. It has been known for long that this MMS task selectively increases the cerebral blood flow in the SMA (Roland et aI., 1980); this observation has been considered as indirect evidence that SMA might be the generator of the N30 potential. (5) The N30 potential, as well as the homologue M30 magnetic field, is enhanced when subject observes repetitive grasping movements or complex sequences of fingers movements performed by an examiner during stimulation. This effect is independent of the complexity of the observed movements; it has been interpreted as reflecting storage of subject's somatosensory information connected with the observed movements (Rossi et aI., 2002), and related to the SMA activation observed in PET studies when subjects observe meaningful actions (Grezes et aI., 1998).

SOMATOSENSORY EVOKED RESPONSES

.....J

~ z

o

0:::

LL

Fig. 2. Gating of frontal SEPs during mental movement simulation (MMS). These responses were obtained in two distinct normal subjects (right and left traces) before (A, D, G, J), during (B, E, H, K) and after (C, F, I, L) mental simulation of movement. The dotted areas indicate the N30 potential surface above the baseline. Note that the N24 frontal potential, labeled N23 by the authors, and the parietal responses are not affected by mental imagery of movement. (From Cheron and Borenstein, 1992, with permission.) However, if there is little doubt that the N30 amplitude is linked to motor programming, its origin in the SMA is challenged by data from direct cortical recordings, which favor an N30 dipolar source radial to the scalp surface, located in Brodmann's area 1 of the SI cortex, (Allison et al., 1989, 1991), and do not show a N30 potential in this area (Barba et aI., 2001). Moreover, after subtraction of the N24-P24 field, the maximum of the remaining scalp N30 field has the same location as that of the P22 potential in the central region (Osaki et aI., 1996). This suggests

57

that the N30 potential is generated close to the central sulcus and not in the pre-motor cortex, a view supported by most of the source modeling studies of early SEPs (Valeriani et aI., 2000). Lastly, functional magnetic resonance imaging studies have shown that all of cortical areas involved in actual movements, and not only the SMA, are activated during mental simulation of movements (Stephan et aI., 1995). To date the hypothesis that N30 could be generated in the primary motor area has not been ruled out by direct recordings, or source modeling, of SEPs or SEFs. Conversely recent source modeling studies rather supported this hypothesis (Waberski et aI., 1999). The movement related gating of N30 could therefore reflect a similar change in motor cortex excitability both during movement and mental motor imagery. Studies using transcranial magnetic stimulation (TMS) converge to the conclusion that the excitability of the motor cortex is increased during actual movement and mental movement simulation (Ridding et aI., 1995; Kasai et aI., 1997), so that N30 amplitude would reflect the level of motor cortex inhibition. However, the fact that N30 and motor cortex inhibition are decreased during movement and motor imagery does not imply that the underlying mechanisms are the same in the two conditions. A recent TMS study by Ridding and Rothwell (1999) suggests that, in spite of a similar effect of movement and motor imagery on N30 amplitude, activity in the circuits responsible for GABAergic inhibition of the motor cortex is reduced during voluntary contraction while motor imagery, which activates similar brain regions as overt movement but does not result in afferent input, does not produce significant changes in intracortical inhibition. 5.3. Short latency SEPs to lower limb stimulation (electric stimulation of tibial, sural and peroneal nerves) Electrical stimulation of the tibial nerve at the ankle is adopted by most authors for the testing of the sensory pathways of the lower limb. However the electrical stimulation can also be applied to the sural nerve at the ankle, or to the peroneal nerve at the knee without major changes in the general waveform of the spinal or scalp responses. As compared with tibial nerve SEPs only the peaking latencies are modified; they are delayed by about 3 ms, or

58

F. MAUGUrERE

shortened by about 5-6 ms, respectively for sural and peroneal nerve stimulation. In what follows tibial nerve SEPs will be taken as the reference for describing the normal waveforms (Fig. 3). 5.3.1. Peripheral, spinal and subcortical potentials

5.3.1.1. The N7 potential A compound action potential corresponding to the activation of tibial nerve fibers is recorded at the posterior aspect of the knee using a bipolar montage. The latency of this near-field potential is of about 7 ms in adults and varies with the length of the lower limb. N7 reflects a mixed response of motor and sensory fibers, which is clinically useful to assess the function of the peripheral segment of the pathway. The afferent volley in cauda equina roots can be recorded using skin electrodes placed at the L5-Sl level and a distant reference site, for instance at the knee opposite to the stimulation. 5.3.1.2. The spinal potentials A montage between an electrode situated on the spinal process of TI2 or Ll vertebrae and distal reference electrode records a spinal negative potential peaking at 21-24 ms in normal subjects. According to different authors this segmental

response has been labeled N20 (Tsuji et aI., 1984), N2l (Small and Matthews, 1984), N22 (Lastimosa et al., 1982; Riffel et aI., 1984), N23 (Yamada et al., 1982) or N24 (Desmedt and Cheron, 1983). These differences in nomenclature are mostly due to differences in the mean body height of the subjects sampled for normative studies, In what follows this response will be labeled N22. After tibial nerve stimulation the lumbar N22 originates mostly from the spinal segment receiving fibers from the S1 root (Delbeke et aI., 1978; Dimitrijevic et aI., 1978; Jones and Small, 1978; Lastimosa et aI., 1982; Desmedt and Cheron, 1983; Small and Matthews, 1984; Tsuji et al., 1984; Delwaide et aI., 1985). As all segmental dorsal hom responses (see Dimitrijevic and Halter, 1995, for a review) the N22 potential demonstrates a polarity reversal when recorded anterior to the cord, a field distribution consistent with a horizontal dipolar source reflecting the postsynaptic response of dorsal hom neurons to incoming inputs. In direct spinal cord and roots recordings the N22 amplitude is maximal close to the entry zone of the S1 root in cord surface recordings and decreases steeply without any latency shift at more rostral or caudal electrode sites (Jeanmonod et aI., 1991). The N22 potential is followed by a slow positivity, known as the "P" wave, which reverses into a negativity when recorded at the anterior aspect of the cord (Shimoji et al., 1977; Jeanmonod et al., 1989) and may reflect

Left Tibial nerve

Cz'-A2

P39!f\ Pro JW~ ~

:>

so

P~O

Fpz-Cv6 N37

Ll-Abd '--=----' l"22

Po. F

100 ms

Fig. 3. Normal tibial nerve SEPs (see text for description).

59

SOMATOSENSORY EVOKED RESPONSES

the processes of pre-synaptic inhibition affecting the primary afferent fibers in the dorsal hom. In peroperative direct recording of the cord, several fast negativities are superimposed to the ascending slope of the N22, which reflect the action potentials ascending in the dorsal columns. In most skin surface recordings these negativities cannot be individualized. When the reference electrode is not situated on the axis of propagation of the peripheral ascending volley, the N22 potential is preceded by a small positivity peaking around 17 ms. This P17 potential is a far-field potential originating in lumbo-sacral plexus trunks, which can also be recorded occasionally on the scalp when the reference electrode is placed at the knee (or iliac crest) on the nonstimulated side (Yamada et al., 1982; Desmedt and Cheron, 1983).

5.3.1.3. The scalp far-field P30 potential With appropriate non-cephalic reference montages widespread far-field potentials, other than the P17 described above, can be recorded on the scalp which peak before the onset of the earliest cortical potential in the 25-32 ms latency range (Yamada et al., 1982; Desmedt and Cheron, 1983; Kakigi and Shibasaki, 1983). Only the latest of these potentials, identified by Yamada et al. (1982) as "P31", is consistently recorded in normal subjects after tibial nerve stimulation at the ankle. This observation was confirmed by several investigators who labeled this potential as P28 (Chiappa and Ropper, 1982; Kakigi et aI., 1982), P30 (Vera et al., 1983; Desmedt and Bourguet, 1985; Guerit and Opsomer, 1991; Urasaki et aI., 1993) or P31 (Desmedt and Cheron, 1983; Seyal et al., 1983). The latency of this potential varies according to body height, and will be labeled P30 in this chapter, according to the author's normative data. The utility of the P30 potential has been validated for clinical applications of tibial SEPs in patients with spinal cord and brainstern lesions (Tinazzi and Mauguiere, 1995; Tinazzi et aI., 1996b). The P30 potential is widely distributed on the scalp but predominates in the frontal region (Desmedt and Bourguet, 1985; Guerit and Opsomer, 1991, therefore it is drastically reduced in scalpreference recordings (Seyal et aI., 1983). When recorded with electrodes located in the fourth ventricle during surgery, this potential shows the same intra-cranial spatio-temporal distribution as the

Pl4 component of median nerve SEPs (Urasaki et aI., 1993). Therefore the P30 potential is likely to be generated in the lower brainstem (Yamada et al., 1982; Desmedt and Cheron, 1983) and can be viewed as the homologue, for the lower limb, of the far-field P14 recorded on the scalp after median nerve stimulation. Recording of P30 is made easier if the scalp electrode is placed in the mid-frontal region at Fz or Fpz, where P30 has its maximal amplitude with minimal contamination by subsequent cortical potentials. The most practical reference site for recording of P30 is the spinous process of the 6th cervical vertebra (Seyal et al., 1983; Tinazzi and Mauguiere, 1995). Since no segmental response is evoked by lower limb stimulation at the cervical cord level, the neck can be considered as virtually inactive at the peaking latency of P30. The ascending volley in spinal somatosensory pathways reaches the cervical level at a latency of about 27 ms but, due to the time dispersion of afferent impulses, there is no consistent potential recorded by a skin cervical electrode at this latency. In some normal subjects a scalp far-field P27 potential distinct from P30 can be recorded in the frontal region, which reflects the ascending volley in the cervical cord (Yamada et aI., 1982; Desmedt and Cheron, 1983).

5.3.1.4. The N33 scalp potential In non-cephalic reference recordings a small N33 negativity (Desmedt and Bourguet, 1985; Guerit and Opsomer, 1991) may precede the P39 potential. This negativity is widely distributed over the scalp; part of it is also picked at the earlobe so that it is drastically reduced in ear lobe reference recordings. Because some of the cortical potentials are picked up at the earlobe contralateral to stimulation, it is preferable to use the earlobe ipsilateral to stimulation as reference site for the recording of cortical tibial nerve SEPs (Tinazzi et aI., 1997a). N33 is considered as the homologue of the N18 upper limb SEP and is likely to have the same origin in the brainstem (see above). 5.3.2. Early cortical potentials 5.3.2.1. The P39 potential The earliest cortical potential elicited by the stimulation of tibial nerve at the ankle is a positive

60 potential usually labeled P37, P39, or P40 for it peaks at a mean latency of 37-40 ms in normal subjects. In what follows this potential will be labeled as P39. This potential peaks 6 to 7 ms earlier after stimulation of the tibial nerve at the popliteal fossa, and about 3 ms later after stimulation of the sural nerve at the ankle. In normal subjects P39 is recorded on the vertex and can be reliably obtained at midway between Pz and Cz using a scalp to earlobe montage (see Yamada et al., 2000, for a recent review). Frequently the maximum of P39, although close to the midline, is slightly shifted on the side of the scalp ipsilateral to the stimulation (Cruse et al., 1982). The reason for this paradoxicallateralization, documented by source modeling of SEPs (Baumgartner et al., 1998), is that the somatotopic representation of the lower limb in the somatosensory SI, and in particular that of the foot, is situated at the inner aspect of the hemisphere. This paradoxical scalp distribution of P39 is not observed when the proximal leg is stimulated for instance: after stimulation of the femoral nerve or lateral femoral cutaneous nerve (Wang et al., 1989; Yamada et al., 1996). SEP mapping in normal subjects often shows a dipolar field distribution for P39, with a maximum of positivity ipsilateral to the stimulation in the parietal region and a maximum of negativity (N39) in the contralateral fronto-central region (Cruse et al., 1982; Seyal et al., 1983; Desmedt and Bourget, 1985; Kakigi and Jones, 1986; Tinazzi et al., 1996a). Magnetic fields evoked by stimulation of the lower limb also show this type of distribution (Hari et al., 1984; Huttunen et al., 1987). The negative N39 maximum of this dipole is not consistently obtained, probably because of differences between subjects in the orientation of the leg area in SI with respect to the scalp surface, so that all intermediates are possible between a vertical radial dipole with a positive maximum on the vertex, and a nearly tangential one with a P39 and N39 culminating respectively in the ipsi- and contralateral parietal regions. As for median nerve SEPs the question is still debated whether all of these potentials reflect the activity of the lower limb area in SI, or whether some could be generated in other cortical areas. Another question is to determine whether the P39N39 dipolar field has a single source in SI tangent to the scalp (Tsumoto et al., 1972; Cruse et al., 1982;

F. MAUGUIERE

Seyal et al., 1983; Kakigi and Jones, 1986; Kakigi and Shibasaki, 1992), or several (Vas et al., 1981; Desmedt and Bourguet, 1985; Pelosi et al., 1988). Recent studies have rather favored the latter hypothesis by showing that: (1) N39 and P39 can show some difference in their respective latencies, the former peaking earlier than the latter, this explains why some authors have proposed to label these potentials N37 and P40, respectively (Yamada et al., 2000). (2) P39 is selectively reduced with preserved N39 in patients with motor neuron disease (Zannette et al., 1996). (3) P39 is reduced at high stimulation rates (Chiappa and Ropper, 1982; Chiappa, 1990), while the frontal N39 amplitude remains unaffected at stimulus rates up to 7.5 Hz (Tinazzi et al., 1996a). (4) Recent source modeling studies of scalp responses (Baumgartner et al., 1998) and direct intra-cerebral recordings (Valeriani et al., 2001) have suggested that two sources located in the central region are quasi-simultaneously involved in the generation of the N39-P39 field on scalp surface. One is indeed tangential to scalp surface and responsible for a dipolar field with a positive maximum in the centro-parietal region ipsilateral to stimulus, and a negative maximum in the fronto-temporal region contralateral to stimulus. The other is perpendicular to scalp surface and responsible for part of the vertex P39 potential. Recent studies of the effects of movement on cortical tibial SEPs (Fig. 4) have shown that P39, but not N39, is gated during voluntary movement of the stimulated foot (Tinazzi et al., 1997b; Valeriani et al., 1998b, 2001). During tonic contraction this effect is proportional to the level of muscle strength maintained by the subjects. These findings have been interpreted in different ways according to the different source models. Tinazzi et al. (1998) proposed that P39 could be generated in the motor cortex, while N39 could be the equivalent of the median nerve N20 and generated in area 3b in the SI area. Valeriani et al. (1998b, 2001) considered that only the radial central source of the vertex P39, possibly located rostral to the central sulcus, is sensitive to voluntary movements.

61

SOMATOSENSORY EVOKED RESPONSES

with distinct orientations (Desmedt and Bourguet, 1985; Guerit and Opsomer, 1991»However, there is no clinical study demonstrating that N5Qor P60 can be selectively affected by a single focal lesion, and these two potentials show the same attenuation as P39 during active movement (Tinazzi et aI., 1997b). 5.4. Abnormal SEPs in central motor deficits and movement disorders 5.4.1. Motor deficits related with hemispheric lesions

Fig. 4. Gating of tibial nerve SEPs during movement. (Grand average waveforms in 10 normal subjects.) The vertex P37 ipsilateral to stimulus (right traces) and the contralateral frontal N37 (left traces) correspond to the P39 and N39 potentials described in the text, respectively. Note that the gating of the P37 is observed only during active isometric contraction or active movement of the foot on the stimulation side. Conversely movement of the foot contralateral to stimulus and passive foot movement on the stimulated side do not affect the P37 potential. No change in N37 potential amplitude is observed in any of the experimental conditions. (From Tinazzi et aI., I997b, modified, with permission.)

5.3.2.2. The N50 and P60 potentials The "W" profile of the cortical response recorded on the vertex and in the centro-parietal region is made of three sequential responses P39, N50 and P60. This waveform has been reported in all studies of tibial nerve SEPs (see Tinazzi et aI., 1997a, for a recent review). Electrical stimulation of the L5 and S1 dermatomes also evokes a consistent, W-shaped response on the scalp (Katifi and Sedgwick, 1986). The N50 and P60 potentials culminate at the vertex (Cz) and do not show the same clear dipolar distribution as the P39-N39 response on the scalp, thus suggesting that the W shaped P39-N50-P60 waveform could be generated by several sources

In our series of 241 patients with a focal hemispheric lesion documented by neuroimaging (Mauguiere and Ibanez, 1990), SEPs to median nerve stimulation were abnormal in more than 70% of capsulo-thalamic and in nearly 90% of posterior thalamic lesions. After stimulation of the affected side amplitude reduction or absence of both parietal N20-P27 and frontal P22-N30 potentials with preserved scalp far-field potentials, including P14 and N18, represent the most frequent SEP abnormalities in thalamic or capsular lesions interrupting the somatosensory pathways (Nakanishi et aI., 1978; Mauguiere et aI., 1982, 1983c; Stohr et aI., 1983; Graff-Radford et aI., 1985; Yamada et al., 1985; Mauguiere and Ibanez, 1990). In large cortical lesions involving the central and parietal areas SEP abnormalities are very similar to those encountered in posterior thalamic lesions. When pooling all patients (n= 109) with cortical lesions of our series, there was globally a highly significant correlation between abnormal SEPs and the presence of sensory and/or motor deficits. Astereognosis of the hand opposite to the cortical lesion, eventually combined to hypesthesia for some elementary sensory modalities, was consistently associated with abnormal N20 and/or P27 abnormality (Mauguiere et al., 1983a). Conversely precentral lesions eliminating specifically the P22 and N30 potentials were usually associated to central hemiparesis or hemiplegia. Similarly patients with a selective loss of frontal P22-N30 SEPs in relation with a sub-cortical lesion present with various types of motor deficits including; hemiplegia, ataxic hemiparesis and motor neglect. However hemiparesis, or hemiplegia, can be associated with preserved frontal P22 and N30 potentials in sub-cortical lesions interrupting selectively the efferent motor pathways.

62

Less than 10% of patients with focal hemispheric lesions present with dissociated abnormalities of frontal and parietal early cortical SEPs showing either reduced or absent N20-P27 with preserved P22-N30, or the reverse (Mauguiere and Ibanez, 1990). 5.4.2. Myoclonus

Since the first description by Dawson in 1947, it has been known that giant SEPs can occur in association with the cortical reflex myoclonus observed in progressive myoclonic epilepsies (PME) and with focal motor seizures in lesions of the peri-rolandic area (Dawson, 1947; Shibasaki and Kuroiva, 1975; Mauguiere and Courjon, 1980; Rothwell et al., 1984; Obeso et al., 1985; Shibasaki et al., 1985a, b; Kakigi and Shibasaki, 1987; Seyal, 1991; Ikeda et al., 1995; Valeriani et al., 1997a, b). The giant SEP in PME was first described as a "P25-N33" complex by Shibasaki and Kuroiwa (1975), the amplitude of which, when measured from peak to peak, can reach 50 f.LV or more and is usually of more than 15 f.L V. Giant SEPs show a large degree of inter- and intra-individual variation, nevertheless most of PME patients with cortical myoclonus show a parietal positive and frontal negative dipolar field peaking at about 25 ms contralateral to median nerve stimulation. Jerklocked back averaging the EEG prior to the myoclonus shows that spontaneous myoclonus can be associated with an abnormal cortical spike in some patients with cortical myoclonus (Shibasaki and Kuroiwa, 1975). The interval between the peak of the back-averaged cortical spike and the onset of the myoclonus is in the order of 20 ms in the arm, consistent with rapid conduction from motor cortex to muscle down to a direct cortico-spinal pathway. This observation suggests that giant SEPs and myoclonus-related spikes reflect a cortical hyperexcitability to afferent impulses which is the mechanism underlying the "pyramidal myoclonus" (Halliday, 1967). When identifiable the parietal N20 potential to median nerve stimulation has normal latency and amplitude in most patients with cortical myoclonus (Mauguiere et al., 1981; Rothwell et al., 1984; Obeso et al., 1985; Shibasaki et al., 1985a). Similarly, P14 is normal in such patients (Mauguiere et al., 1981). The normal size of the N20 component

F. MAUGUIERE

indicates that the sensory input into the cortex as well as the primary cortical response are not grossly abnormal. It has been suggested on the basis of waveform decomposition and computerized modeling that the giant SEP results from an abnormal enhancement of certain components of the normal SEPs (Shibasaki et al., 1990). This has been confirmed by dipole modeling showing that the scalp distribution of both normal and giant SEPs can be explained by the same cortical sources and suggesting that giant SEPs mostly reflect an enhanced response of the N24-P24 generator in area 3b (Valeriani et al., 1997a). Enhanced cortical response to proprioceptive inputs can be observed in some of the PME patients with giant median nerve SEPs, in association with exaggerated EMG response to passive movements (Mima et al., 1997). Thus, cortical hyper-responses to either cutaneous inputs alone or to both cutaneous (area 3b) and proprioceptive (areas 3a or 2) inputs can occur in PME with cortical reflex myoclonus. The various combinations of these two abnormalities, in terms of amplitude, may account for the waveform variations reported in early literature on this topic. Intravenously (LV) injected benzodiazepines reduce spontaneous myoclonus and reflex muscle jerking, but their effect on giant SEPs is a matter of controversy. Both reduction (Mauguiere and Courjon, 1980; Shibasaki et al., 1985a) and paradoxical enhancement (Rothwell et al., 1984) of SEPs have been reported under the effect of benzodiazepines. Giant SEPs can be recorded in the absence of myoclonus-related spikes and vice versa. Thus, patients who apparently have similar clinical symptoms can show different SEP abnormalities (see Tassinari et al., 1998, for a review). Complete electrophysiological investigation of patients with myoclonus includes jerk-locked EEG back-averaging (Shibasaki and Kuroiwa, 1975), conventional SEP recordings to median nerve and/or finger stimulation and the recording of the long latency myogenic reflex activity (C reflex) to median nerve stimulation, which reflects the myoclonus itself. When both giant SEPs and jerk-locked spikes are absent in a patient with myoclonus, the efferent impulses generating myoclonus in response to afferent inputs is likely to be generated in subcortical reticular structures (reticular reflex myoclonus of Hallett et al., 1977).

63

SOMATOSENSORY EVOKED RESPONSES

Not all of the patients with myoclonus have giant SEPs. In particular enhanced SEPs are usually not observed in benign forms of juvenile myoclonic epilepsies and in essential non-progressive isolated myoclonus. They are inconstant in CreutzfeldtJakob disease and post-hypoxic myoclonus. In the two latter conditions the existence of giant ·SEPs probably depends on the evolution stage of the disease. Conversely giant SEPs are an almost constant feature in the various forms of Progressive Myoclonic Epilepsies (PME). At the end of the evolution amplitude of SEPs may decrease and even return to normal range in PMEs (Mervaala et aI., 1984). Giant SEPs can also be observed over the damaged hemisphere in patients with supra-tentorial tumors, post-traumatic cortical atrophies or long after an ischemic or hemorrhagic stroke (Laget et al., 1967; Furlong et al., 1993; Valeriani et aI., 1997b). In these patients the enhancement of the cortical response is often less than that observed in PME patients with cortical reflex myoclonus. Loss of inhibitory control and post-lesional collateral sprouting of cortical afferents could be responsible for such SEP abnormalities. Various components of the normal cortical SEP waveform can be increased and the resulting aspect often differs from that of the giant P25-N33 observed in cortical myoclonus. A selective increase of the N20-P22 deflection strictly localized in the central region has been reported in a patient with a rolandic tumor (Valeriani et al., 1997b). The scalp topography of this abnormal response was compatible with a dipolar source radial to the scalp surface, and interpreted as an abnormal response of Brodmann's area 3a, which is known to receive proprioceptive inputs from muscle afferents. In focal lesions giant SEPs are often observed in the absence of myoclonus triggered by somatosensory stimuli, but the occurrence of focal motor seizures is frequent in the history of such patients. In some children aged 3-13 years with normal neurological status vertex and parietal EEG spikes, corresponding to high voltage SEPs (up to 400 /-LV), can be evoked by a single tactile stimulation (De Marco and Negrin, 1973). The presence of these "Extreme SEPs" might forecast the possible occurence of partial motor seizures with benign outcome (De Marco and Tassinari, 1981). Median nerve cortical SEPs in these children are usually normal up to a latency of 60 ms and then show a large central

negativity, which reaches its maximum only for low stimulation frequencies of 0.2 to 0.5 Hz (Plasmati et aI., 1989). 5.4.3. Huntington's disease Since the first study by Oepen et aI. (1981), several authors have reported abnormal early cortical N20-P27 potentials in scalp reference recordings of patients with Huntington's chorea (Bollen et aI., 1984; Ehle et aI., 1984; Noth et al., 1984). Topper et al. (1993) have shown that the frontal P22 and N30 potentials are selectively depressed in this disease. However these authors did not find any significant correlation between the severity of the movement disorder and that of the SEP abnormalities. Moreover abnormal SEPs have also been reported in clinically asymptomatic subjects at risk for the disease, as well as in symptomatic patients during sedation of the choreic movements. Thus, there is no proven link between the movement disorder and SEP abnormalities in this pathology.

5.4.4. Parkinson's disease Rossini et aI. (1989) were the first to report a reduction of the frontal N30 median nerve SEP in patients with Parkinson's disease (PD), in association with parietal SEPs of normal amplitude. These authors hypothesized that the rigidity and/or the akinesia of PD patients could be related to this SEP abnormality, which was confirmed after MPTP treatment in monkeys by the recording of the frontal N15, which is the homologue of the human N30 in monkeys (Onofrj et al., 1994). Statistical analysis of SEP topography showed that differences in temporal and power spectrum distributions of SEPs between normal subjects and PD patients are confined to frontal scalp areas (Babiloni et aI., 1994). Moreover the amplitude of the N30 potential was found to be inversely correlated with that of the second component (latency range 50-60 ms) of the long latency reflexes (LLRs) evoked by electrical stimulation of the median nerve (Rossini et aI., 1991), which are commonly increased in PD. It has also been shown by Cheron et al. (1994) that, though reduced, the N30 potential in PD shows the same amplitude decrease as in normal subjects during voluntary finger movements, and that this gating effect is not modified by apomorphine, a dopamine receptors agonist.

64

There are however conflicting reports concerning attenuation of the frontal N30 potential in patients with PD. The N30 reduction was confirmed in two successive studies respectively in 69% and 71% of PD patients (Rossini et al., 1993, 1995) emphasizing the utility of evaluating the decrease of the amplitude ratio between the frontal and parietal SEPs, the latter being preserved in such patients. Conversely a reduced N30 was observed in only 32.5% of PD patients by Onofrj et al. (1995), with no clear relation between N30 amplitude and motor performances and several authors failed to confirm this N30 reduction, either when comparing PD patients with normal subjects matched for age, or when comparing the SEPs to stimulation of the most affected side with those to stimulation of the less affected side in patients with hemiparkinsonism (Huttunen and Teravainen, 1993; Mauguiere et al., 1993; Garcia et al., 1995). SEPs to median nerve stimulation have also been reported as normal in multiple system atrophy (Abbruzzese et al., 1997). Another open question is whether the amplitude of the N30 potential actually correlates with motor performances in Parkinson's disease and could be used as an objective marker of the evolution state of the disease. Motor fluctuations with "off' periods of severe parkinsonian disability alternating with "on" periods of relative mobility marked by dyskinetic involuntary movements offer the possibility to address directly this question in test-retest studies in individual patients. "On" periods can be obtained by subcutaneous (s.c) injection of apomorphine which is an agonist of D1 and D2 dopamine receptors. The effects of subcutaneous apomorphine injection on the amplitude of the frontal N30 have been studied with controversial results. Rossini et al. (1993, 1995), Cheron et al. (1994) and Stanzione et al. (1997) reported a clear-cut and selective amplitude increase of N30 in association with clinical improvement, while other authors did not observe this effect on frontal SEPs, in spite of a clear improvement of motor performances (Mauguiere et al., 1993). In line with this latter negative result, acute or chronic administration of L-DOPA and bromocriptine did not modify the SEP amplitude in PD patients, in spite of positive clinical effects in the study by Onofrj et al. (1995). Differences in the severity of the disease between patients cohorts included in these studies may account for such divergences. Pierantozzi et al. (1999) provided the most recent evidence in favor of

F. MAUGUIERE

a link between N30 amplitude and motor performances in PD patients treated deep brain stimulation of the internal globus pallidus or of the nucleus subthalamicus. These authors reported that during stimulation, the N30 amplitude increases in correlation with the positive effects on motor abilities, conversely after interruption of stimulation the N30 enhancement fades nearly in parallel with the clinical effects.

5.4.4.1. Dystonia Abnormal central integration of afferent somatosensory inputs is, among others, a possible causative mechanism of dystonia (Hallett, 1995). The first report on SEPs in patients with focal or generalized dystonia is that of Reilly et al. (1992). These authors reported a selective increase of the N30 potential, which could reflect hyperactivity of the striatocortical loops. However, in this study, N30 abnormalities were similar after stimulation of either the affected or unaffected side in patients with unilateral dystonia. This result has been replicated by Kanovsky et al. (1997, 1998) in patients with spasmodic tortiscollis, but a decreased N30 has also been reported in this condition (Mazzini et al., 1994) and in patients with hand dystonia (Grissom et al., 1995). More recently Tinazzi et al. (1999) reported an increase of tibial nerve P39 and N50 potentials that was unrelated to the severity of the disease and present on both sides in patients with unilateral dystonia. This finding can be interpreted as reflecting an abnormal central processing of somatosensory inputs related to increased excitability of the motor cortex and is quite coherent with the early findings of Reilly et al. (1992). This interpretation makes sense if one assumes that P39 and N50 are generated in the motor cortex, a view that is plausible (see above), and that the decreased activity in the putamen, which is the most commonly affected in secondary dystonia, increases the excitability of the motor cortex. Responses to paired interfering stimuli have been studied recently by comparing the amplitudes of SEPs to simultaneous stimulation of the median and ulnar nerve on the same side, with that of the sum of SEPs amplitudes after stimulation of each nerve, individually. In normal subjects the former are smaller than the latter for N13, P14, N20, P27 and N30 potentials, while this does not occur for the peripheral N9 potential. This gating phenomenon,

65

SOMATOSENSORY EVOKED RESPONSES

which can be interpreted as reflecting a central surround inhibition of incoming volleys from neighboring territories, was found to be less efficient in dystonic patients than in normal subjects (Tinazzi et aI., 2000). This sensory overflow could playa role in the distortion of sensory and motor cortical maps of dystonic patients. Abnormal movement related gating of SEPs has been carefully investigated in patients with writer's cramp using a reaction time paradigm (Murase et aI., 2000). The main finding was that pre-movement gating of N24 and N30 frontal SEPs, usually observed in normal subjects with this paradigm is lacking, or reduced, in patients with writer's cramp (Fig. 5). This abnormality was identical in dystonic as well as in simple writer's cramp. Conversely there was no difference in SEPs gating during voluntary movements between patients and normal subjects.

These authors considered that, during movement preparation, changes in SEPs are due mainly to changes in central sensory transmission produced by the intention to move, and unrelated to a centrifugal gating produced by an efferent copy of the motor command, which is preserved in writer's cramp. Thus, dystonia could be related with some defect in a pre-motor subroutine (Kaji et aI., 1995), which includes the specification of motor commands for a forthcoming action as well as the specific setting of cortical responsiveness to afferent somatosensory inputs.

5.5. Conclusion For the past ten years there has been a revival of SEP studies in movement disorders in parallel with new concepts on motor processes, which emphasizes

Fig. 5. Pre-movement gating of frontal median nerve SEPs. Responses recorded during movement preparation are represented in thick traces and responses at rest are in thin traces. Left traces are grand average responses in 11 normal subjects. Right traces are grand average responses in 10 patients with writer's cramp. Note that the N30b corresponds to the frontal N30 potential, while the peak labeled N30a corresponds to the N24 potential described in the text. These two potentials are reduced in the pre-movement phase in normal subjects, while this gating does not occur in patients (from Murase et al., 2000, with permission.)

66

the role of somatosensory information processing in the preparation and execution of movements. Most of these recent advances have no direct diagnostic application but proved useful to assess the pathophysiology of motor disorders. At the present time only a few research laboratories have developed experimental paradigms where all of the physiological parameters influencing cortical SEPs are adequately controlled. Persisting controversies concerning the source locations of cortical potentials sensitive to motor programming are another limiting factor for a routine use of SEPs in patients with motor disorders. However in clinical research, SEPs have a potential utility for monitoring the effects of medical and surgical treatments, as well as post lesion reorganization of cortical functional representations.

References Abbruzzese, G, Marchese, Rand Trompetto, C (1997) Sensory and motor evoked potentials in multiple system atrophy: a comparative study with Parkinson's disease. Mov. Disord., 12: 315-321. Allison, T and Hume, AL (1981) A comparative analysis of short latency somatosensory evoked potentials in man, monkey, cat, and rat. Exp. Neurol., 72: 592-611. Allison, T, Goff, WR, Williamson, PD and Vangilder, JC (1980) On the neural origin of early components of the human somatosensory evoked potentials. In: JE Desmedt (Ed.), Clinical Uses of Cerebral, Brainstem and Spinal Somatosensory Evoked Potentials. Progr. Clin. Neurophysiol. Karger, Basel, Vol. 7, pp. 51-68. Allison, T, McCarthy, G, Wood, CC, Darcey, TM, Spencer, DD and Williamson, PD (1989) Human cortical potentials evoked by stimulation of the median nerve. I. Cytoarchitectonic areas generating short-latency potentials.1. Neurophysiol., 62: 694-710. Allison, T, McCarthy, G, Wood, CC and Jones, SJ (1991) Potentials evoked in human and monkey cerebral cortex by stimulation of the median nerve: a review of scalp and intracranial recordings. Brain, 114, 2465-2503. Andersen, P, Eccles, JC, Schmidt, RF and Yokota, T (1964) Slow potential waves produced in the cuneate nucleus by cutaneous volley and cortical stimulation. J. Neurophysiol., 27: 78-91. Anziska, A and Cracco, RQ (1980) Short-latency somatosensory evoked potentials: studies in patients with focal neurological disease. Electroencephalogr. Clin. Neurophysiol., 49: 227-239. Araki, A, Yamada, T, Ito, T, Urushibara, N, Kohira, R, Hsu, SP and Yeh, M (1997) Dissociation between upper and lower neck N13 potentials following paired

F. MAUGUIERE

median nerve stimuli. Electroencephalogr. Clin. Neurophysiol., 104: 68-73. Babiloni, F, Babiloni, C, Cecchi, L, Onorati, P, Salinari, S and Urbano, A (1994) Statistical analysis of topographic maps of short-latency somatosensory evoked potentials in normal and parkinsonian subjects. IEEE Trans. Biomed. Eng., 41: 617-624. Barba, C, Frot, M, Guenot, M and Mauguiere, F (2001) Stereotactic recordings of median nerve SEPs in the human supplementary motor area. Europ. J. Neurosci., 13: 347-356. Baumgartner, V,Vogel, H, Ellrich, J, Gawehn, J, Stoeter, P and Treede, RD (1998) Brain electrical source analysis of primary cortical components of the tibial nerve somatosensory evoked potential using regional sources. Electroencephelogr. Clin. Neurophysiol., 108: 588-599. Bollen, EL, Arts, RI, Roos, RA, Van der Velde, EA and Buruma, OJ (1985) Somatosensory evoked potentials in Huntington's chorea. Electroencephalogr. Clin. Neurophysiol., 62: 235-240. Brenner, D, Lipton, J, Kaufman, L and Williamson, SJ (1978) Somatically evoked magnetic fields of the human brain. Science, 199: 81-83. Bromm, Band Treede, RD (1987) Human cerebral potentials evoked by CO2-laser stimuli causing pain. Exp. Brain. Res., 67: 153-162. Broughton, RJ (1969) In: E Donchin and DB Lindsley (Eds.), Average Evoked Potentials. U.S. Government Printing Office, Washington, D.C., pp. 79-84. Buchner, H, Ferbert, A and Hacke, W (1988) Serial recording of median nerve stimulated subcortical somatosensory evoked potentials (SEPs) in developing brain death. Electroencephalogr. Clin. Neurophysiol., 69; 14-23. Buchner, H, Adams, L, Miiller, A, Ludwig, I, Knepper, A, Thron, A, Niemann, K and Scherg, M (1995) Somatotopy of human hand somatosensory cortex revealed by dipole source analysis of early somatosensory evoked potentials and 3D-NMR tomography. Electroencephalogr. Clin. Neurophysiol., 96: 121-134. Buchner, H, Waberski, TD, Fuchs, M, Drenckhahn, R, Wagner, M and Wischmann, HA (1996) Postcentral origin of P22: evidence from source reconstruction in a realistically shaped head model and from a patient with a post-central lesion. Electroencephalogr. Clin. Neurophysiol., 100: 332-342. Burke, D, Skuse, NF and Lethlean, K (1981) Cutaneous and muscle afferent components of the cerebral potential evoked by electrical stimulation of human peripheral nerves. Electroencephalogr. Clin. Neurophysiol., 51: 579-588. Burke, D, Gandevia, SC, McKeon, Band Skuse, NF (1982) Interactions between cutaneous and muscle

SOMATOSENSORY EVOKED RESPONSES

afferent projections to cerebral cortex in man. Electroencephalogr. Clin. Neurophysiol., 53: 349-360. Cheron, G and Borenstein, S (1987) Specific gating of the early somatosensory evoked potentials during active movement. Electroencephalogr. Clin. Neurophysiol., 67: 537-548. Cheron, G and Borenstein, S (1992) Mental movement stimulation affects the N30 frontal component of the somatosensory evoked potential. Electroencephalogr. Clin. Neurophysiol., 84: 288-292. Cheron, G, Piette, T, Thiriaux, A, Jacquy, J and Godaux, E (1994) Somatosensory evoked potentials at rest and during movement in Parkinson's disease: evidence for a specific apomorphine effect on the frontal N30 wave. Electroencephalogr. Clin. Neurophysiol., 92: 491-501. Cheron, G, Dan, B and Borenstein, S (2000) Sensory and motor interfering influences on somatosensory evoked potentials. J. Clin. Neurophysiol., 17: 280-294. Chiappa, KH (1990) Evoked potentials in clinical medicine (2nd ed.). Raven Press, New York. Chiappa, KH and Ropper, AH (1982) Evoked potentials in clinical medicine. New Engl. J. Med., 306: 12051211. Cioni, B and Meglio, M (1986) Epidural recordings of electrical events produced in the spinal cord by segmental, ascending and descending volleys. Appl. Neurophysiol., 49: 315-326. Cohen, LG and Starr, A (1985) Specific gating of the early somatosensory evoked potentials. Neurology, 35: 691698. Cracco, RQ (1973) Spinal evoked response: peripheral nerve stimulation in man. Electroencephalogr. Clin. Neurophysiol., 35: 379-386. Cracco, RQ and Cracco, JB (1976) Somatosensory evoked potential in man: Far-field potentials. Electroencephalogr. Clin. Neurophysiol., 41: 460-466. Cruse, R, Klem, G, Lesser, Rand Luders, H (1982) Paradoxical lateralization of cortical potentials evoked by stimulation of posterior tibial nerve. Arch. Neurol., 39: 222-225. Dawson, GD (1947) Investigations on a patient subject to myoclonic seizures after sensory stimulation. J. Neurol. Neurosurg. Psychiat., 10: 141-162. Debecker, J and Desmedt, JE (1964) les potentials evoques cerebraus et les potentials de nerf sensible chez l'homme. Acta Neurol. Belg., 64: 1212-1248. Deiber, MP, Giard, MH and Mauguiere, F (1986) Separate generators with distinct orientations for N20 and P22 somatosensory evoked potentials to finger stimulation. Electroencephalogr. Clin. Neurophysiol., 65: 321-334. Delbeke, EJ, McComas, AJ and Kopec, SJ (1978) Analysis of lumbosacral evoked potentials in man. J. Neurol. Neurosurg. Psychiat., 41: 293-302.

67 Delberghe, X, Mavroudakis, N, Zegers de Beyl, D and Brunko, E (1990) The effect of stimulus frequency on post- and precentral short latency somatosensory evoked potentials (SEPs). Electroencephalogr. Clin. Neurophysiol., 77: 96-92. Delestre, F, Lonchampt, P and Dubas, F (1986) Neural generator ofP14 far-field somatosensory evoked potential studied in a patient with a pontine lesion. Electroencephalogr. Clin. Neurophysiol., 65: 227-230. Delwaide, PJ, Schoenen, J and De Pasqua, V (1985) Lumbosacral spinal evoked potentials in patients with MS. Neurology, 35: 174-179. De Marco, P and Negrin, P (1973) Parietal focal spike evoked by contralateral tactile somatotopic stimulations in four non-epileptic subjects. Electroencephalogr. Clin. Neurophysiol., 24: 308-312. De Marco, P and Tassinari, CA (1981) Extreme somatosensory evoked potential (ESEP): an EEG sign forecasting a possible occurrence of seizures in children. Epilepsia, 22: 569-575. Desmedt, JE and Bourguet, M (1985) Color imaging of parietal and frontal somatosensory potential fields evoked by stimulation of median or posterior tibial nerve in man. Electroencephalogr. Clin. Neurophysiol., 62: 1-17. Desmedt, JE and Cheron, G (1980a) Central somatosensory conduction in man: Neural generators and interpeak latencies of the far-field components recorded from neck and right or left scalp or earlobes. Electroencephalogr. Clin. Neurophysiol., 50: 382-403. Desmedt, JE and Cheron, G (1980b) Somatosensory evoked potentials to finger stimulation in healthy octogenarians and in young adults: waveforms, scalp topography and transit times of parietal and frontal components. Electroencephalogr. Clin. Neurophysiol., 50: 404-425. Desmedt, JE and Cheron, G (1981a) Prevertebral (oesophageal) recording of subcortical somatosensory evoked potentials in man: the spinal PI3 component and the dual nature of the spinal generators. Electroencephalogr. Clin. Neurophysiol., 52: 257-275. Desmedt, JE and Cheron, G (l981b) Non-cephalic reference recording of early somatosensory potentials to finger stimulation in adult or aging man: differentiation of widespread NI8 and contralateral N20 from the prerolandic P22 and N30 components. Electroencephalogr. Clin. Neurophysiol., 52: 553-570. Desmedt, JE and Cheron, G (1983) Spinal and far-field components of human somatosensory evoked potentials to posterior tibial nerve stimulation analyzed with eesophageal derivations and non-cephalic reference recording. Electroencephalogr. Clin. Neurophysiol., 56: 635-651.

68 Desmedt, IE and Nguyen, TH (1984) Bit-mapped colour imaging of the potential fields of propagated and segmental sub-cortical components of somatosensory evoked potentials in man. Electroencephalogr. Clin. Neurophysiol., 58: 481-497. Desmedt JE and Osaki, I (1991) SEPs to finger joint input lack the N20-P20 response that is evoked by tactile inputs: contrast between cortical generators in areas 3b and 2 in humans. Electroencephalogr. CUn. Neurophysiol., 80: 513-521. Desmedt, IE, Tran Huy, N and Carmelier, J (1983) Unexpected latency shifts of the stationary P9 somatosensory evoked potential far-field with changes in shoulder position. Electroencephalogr. CUn. Neurophysiol., 56: 628-634. Dimitrijevic, MR and Halter, JA (Eds.) (1995) Atlas of Human Cord Evoked Potentials. Butterworth-Heinemann, Boston. Dimitrijevic, MR, Larsson, LE, Lehmkuhl, D and Sherwood, A (1978) Evoked spinal cord and nerve root potentials in humans using a non-invasive recording technique. Electroencephalogr. CUn. Neurophysiol., 45: 331-340. Ehle, AL, Stewart, RM, Lellelid, NA and Leventhal, NA (1984) Evoked potentials in Huntington's disease: a comparative longitudinal study. Arch. Neurol., 41: 379-382. El Kharoussi, M, Ibanez, V, Ben Jelloun, W, Hugon, M and Mauguiere, F (1996) Potentiels evoques somesthesiques: Interferences et masquage perceptif des afferences cutanees chez l'Homme. Neurophysiol. CUn., 26: 85-101. Forss, N, Hari, R, Salmelin, R, Ahonen, A, Hamalainen, M, Kajola, M, Knuutila, J and Simola, J (l994a) Activation of the human posterior parietal cortex by median nerve stimulation. Exp. Brain Res., 99: 309-315. Forss, N, Salmelin, Rand Hari, R (l994b) Comparison of somatosensory evoked fields to airpuff and electric stimuli. Electroencepalogr. Clin. Neurophysiol., 92: 510-517. Forss, N, Merlet, I, Vanni, S, Hamalainen, M, Mauguiere, F and Hari, R (1996) Activation of human mesial cortex during somatosensory attention task. Brain Res., 734: 229-235. Franssen, H. Stegeman, DF, Molemen, J and Schoobaar, RP (1992) Dipole modelling of median nerve SEPs in normal subjects and patients with small subcortical infarcts. Electroencephalogr. CUn. Neurophysiol., 84: 401-417. Furlong, PL, Wimalaratna, S and Harding, GFA (1993) Augmented P22-N31 SEP component in a patient with unilateral space occupying lesion. Electroencephalogr. Clin. Neurophysiol., 88: 72-76.

F. MAUGUIERE

Gandevia, SC and Burke, D (1988) Projection to the cerebral cortex from proximal and distal muscles in the human upper limb. Brain, 111: 389-403. Gandevia, SC and Burke, D (1990) Projection of thenar muscle afferents to frontal and parietal cortex of human subjects. Electroencephalogr. CUn. Neurophysiol., 77: 353-361. Gandevia, SC and Burke, D (1990) Projection of thenar muscle afferents to frontal and parietal cortex of human subjects. Electroencephalogr. CUn. Neurophysiol., 77: 353-361. Gandevia, SC and Macefield, G (1989) Projection of lowthreshold muscle afferents from intercostal muscles to the cerebral cortex. Resp. Physiol., 77: 203-214. Gandevia, SC, Burke, D and McKeon, BB (1983) Convergence in the somatosensory pathway between cutaneous afferents from the index and middle fingers in man. Exp. Brain. Res., 50: 415-425. Gandevia, SC, Burke, D and McKeon, BB (1984) The projection of muscle afferents from hand to cerebral cortex in man. Brain, 107: 1-13. Garcia, PA, Aminoff, MJ and Goodin, DS (1995) The frontal N30 component of the median-derived SEP in patients with predominantly unilateral Parkinson's disease. Neurology, 45: 989-992. Garcia-Larrea, Land Mauguiere, F (1988) Latency and amplitude abnormalities of the scalp far-field P14 to median nerve stimulation in multiple sclerosis. Electroencephalogr. Clin. Neurophysiol., 71: 180-186. Garcia-Larrea, L, Bastuji, H and Mauguiere, F (1992) Unmasking of cortical SEP components by changes in stimulus rate: a topographic study. Electroencephalogr. Clin. Neurophysiol., 84: 71-83. Goff, GD, Matsumiya, Y, Allison. T and Goff, WR (1977) The scalp topography of human somatosensory and auditory evoked potentials. Electroencephalogr. Clin. Neurophysiol., 42: 57-76. Graff-Radford, NR, Damasio, H, Yamada, T, Eslinger, PJ and Damasio, AR (1985) Nonhaemorrhagic thalamic infarction: clinical, neuropsychological and electrophysiological findings in four anatomical groups defined by computerized tomography. Brain, 108: 485-516. Grezes, J, Costes, Nand Decety, J (1998) Top-down effect of the strategy on the perception of biological motion: a PET investigation. Cognit. Neuropsychol., 15: 553582. Grissom, JR, Toro, C, Tretteau, J and Hallett, M (1995) The N30 and N140-P190 median nerve somatosensory evoked potentials waveforms in dystonia involving the upper extremity (Abstract). Neurology, 45 (Suppl, 4): A458. Guerit, 1M and Opsomer, RJ (1991) Bit-mapped imaging of somatosensory evoked potentials after stimulation of

SOMATOSENSORY EVOKED RESPONSES

the posterior tibial nerves and dorsal nerve of the penisl clitoris. Electroencephalogr. Clin. Neurophysiol., 80: 228-237. Hallett, M (1995) Is dystonia a sensory disorder? Ann. Neurol., 38: 139-140. Hallett, M, Chadwick, D, Adam, J and Marsden, CD (1977) Reticular reflex myoclonus; a physiological type of human post-hypoxic myoclonus. 1. Neurol. Neurosurg. Psychiat., 40: 253-264. Halliday, AM (1967) The electrophysiological study of myoclonus in man. Brain, 90: 241-284. Halliday, AM (1993) Evoked potentials in clinical testing. Churchill Livingstone, Edinburgh. Halliday, AM and Mason, AA (1964) The effect of hypnotic anesthesia on cortical responses. 1. Neurol. Neurosurg. Psychiat., 27: 300-312. Halliday, AM and Wakefield, GS (1963) Cerebral evoked potentials in patients with dissociated sensory loss. J. Neurol. Neurosurg. Psychiat., 26: 211-219. Halonen, JP, Jones, Sand Shawkat F (1988) Contribution of cutaneous and muscle afferent fibers to cortical SEPs following medial and radial nerve stimulation in man. Electroencephalogr. Clin. Neurophysiol., 71: 331-335. Hari, R, Hamalainen, M, Kaukoranta, E, Reinikainen, K and Teszner, D (1983) Neuromagnetic responses from the second somatosensory cortex in man. Acta Neurol. Scand., 68: 207-212. Hari, R, Reinikainen, K, Kaukoranta, E, Hamalainen, M, Ilmoniemi, R, Penttinen, A, Salminen, J and Teszner, D (1984) Somatosensory evoked cerebral magnetic fields from SI and SIT in man. Electroencephalogr. Clin. Neurophysiol., 57: 254-263. Hashimoto, I (1987) Somatosensory evoked potentials elicited by air-puff stimuli generated by a new highspeed air control system. Electroencephalogr. CUn. Neurophysiol., 67: 231-237. Huttunen, J and Teravainen, H (1993) Pre- and postcentral cortical somatosensory evoked potentials in hemiparkinsonism. Mov. Disord., 8: 430-436. Huttunen, J, Kaukoranta, E and Hari, R (1987) Cerebral magnetic responses to stimulation of tibial and sural nerves. J. Neurol. Sci., 79: 43-54. Ibanez, V, Deiber, MP and Mauguiere, F (1989) Interference of vibrations with input transmission in dorsal hom and cuneate nucleus in man: A study of somatosensory evoked potentials (SEPs) to electrical stimulation of median nerve and fingers. Exp. Brain Res., 75: 599--610. Ibanez, V,Fischer, G and Mauguiere, F (1992) Dorsal hom and dorsal column dysfunction in intramedullary cervical cord tumors: A SEP study. Brain, 115: 1209-1234.

69 Ikeda, A, Shibasaki, H, Nagarnine, T, Xu, X, Terada, K, Mirna, T, Kaji, R, Kawai, I, Tatsuoka, Y and Kimura, J (1995) Peri-rolandic and fronto-parietal components of scalp recorded giant SEPs in cortical myoclonus. Electroencephalogr. Clin. Neurophysiol., 96: 300-309. Iragui,VJ (1984) The cervical somatosensory evoked' potential in man, far-field conducted and segmental components. Electroencephalogr. CUn. Neurophysiol., 57: 228-235. Iwayama, K, Mori, K, Iwamoto, K, Yamauchi, T and Masago, M (1988) Origin of frontal Nl5 component of SEP in man. Electroencephalogr. CUn. Neurophysiol., 71: 125-132. Jeanmonod, D, Sindou, M and Mauguiere, F (1989) Three transverse dipolar generators in the human cervical and lumbo-sacral dorsal hom: evidence from direct intraoperative recordings on the spinal cord surface. Electroencephalogr. CUn. Neurophysiol., 74: 236-240. Jeanmonod, D, Sindou, M and Mauguiere, F (1991) The human cervical and lumbo-sacral evoked electrospinogram. Data from intra-operative spinal cord surface recordings. Electroencephalogr. CUn. Neurophysiol., 80: 477-489. Jones, SJ and Small, M (1978) Spinal and subcortical evoked potentials following stimulation of the posterior tibial nerve in man. Electroencephalogr. Clin. Neurophysiol., 44: 299-306. Kaji, R and Sumner, AJ (1987) Bipolar recording of shortlatency somatosensory after median nerve stimulation. Neurology, 37: 410-418. Kaji, R, Shibasaki, H and Kimura, J (1995) Writer's cramp: a disorder of motor subroutine? Ann. Neurol., 38: 837-838. Kakigi, R and Jones, SJ (1986) Influence of concurrent tactile stimulation on somatosensory evoked potentials following posterior tibial nerve stimulation in man. Electroencephalogr. Clin. Neurophysiol., 65: 118-129. Kakigi, Rand Shibasaki, H (1983) Scalp topography of the short-latency somatosensory evoked potentials following posterior tibial nerve stimulation in man. Electroencephalogr. Clin. Neurophysiol., 56: 430-437. Kakigi, R and Shibasaki, H (1984) Scalp topography of mechanically and electrically evoked somatosensory potentials in man. Electroencephalogr. Clin. Neurophysiol., 59: 44-56. Kakigi, Rand Shibasaki, H (1987) Generator mechanisms of giant somatosensory evoked potentials in cortical reflex myoclonus. Brain, 110: 1359-1373. Kakigi, Rand Shibasaki, H (1992) Effects of age, gender and stimulus side on the scalp topography of somatosensory evoked potentials following posterior tibial nerve stimulation. J. Clin. Neurophysiol., 9: 431-440. Kakigi, R, Shibasaki, H, Hashizume, A and Kuroiwa, Y (1982) Short latency somatosensory evoked spinal and

70 scalp-recorded potentials following potential tibial nerve stimulation in man. Electroencephalogr. Clin. Neurophysiol., 53: 602-611. Kanovsky, P, Streitova, H, Dufek, J, and Rektor, I (1997) Lateralization of the P221N30 component of somatosensory evoked potentials of the median nerve in patients with cervical dystonia. Mov. Disord., 12: 553-560. Kanovsky, P, Streitova, H, Dufek, J, Znojil, V, Daniel, P and Rektor, I (1998) Changes in lateralization of the P221N30 component of median nerve somatosensory evoked potentials in patients with cervical dystonia after successful treatment with botulinum toxin. Mov. Disord., 13: 108-117. Kasai, T, Kawai, S, Kawanishi, M and Yahagi, S (1997) Evidence for facilitation of motor evoked potentials (MEPs) induced by motor imagery. Brain Res., 744: 147-150. Katifi, HA and Sedgwick, EM (1986) Somatosensory evoked potentials from posterior tibial nerve and lumbosacral dermatomes. Electroencephalogr. Clin. Neurophysiol., 65: 249-259. Kimura, J, Mitsudome, A, Beck, DO, Yamada T and Dickins, QS 1983) Field distribution of antidromically activated digital nerve potentials. Model for far-field recording. Neurology, 33: 1164-1169. Kimura, J, Mitsudome, A, Beck, DO, Yamada, T and Dickins, QS (1984) Stationary peaks from a moving source in far-field recording. Electroencephalogr. Clin. Neurophysiol., 58: 351-361. Kunesch, E, Knecht, S, Schnitzler, A, Tyercha, C, Schmitz, F and Freund, HJ (1995) Somatosensory evoked potentials elicited by intraneural microstimulation of afferent nerve fibers. J. Clin. Neurophysiol., 12: 476-487. Laget, P. Mamo, H and Houdart, R (1967) De l'interet des potentiels evoques somesthesiques dans I'etude des lesions du lobe parietal de I'homme. Etude preliminaire. Neurochirurgie, 13: 841-853. Lastimosa, A.CB, Bass, NH, Stanback, K and Norwell, EE (1982) Lumbar spinal cord and early cortical evoked potentials after tibial nerve stimulation: effects of stature on normative data. Electroencephalogr. Clin. Neurophysiol., 54: 499-507. Lemon, RN (1981) Functional properties of monkey motor cortex neurons receiving afferent inputs from the hand and fingers. J. Physiol., 311: 313-339. Lesser, RP,Lueders, H, Hahn, J and Klem, G (1981) Early somatosensory evoked potentials by mesian nerve stimulation: intraoperative monitoring. Neurology, 31: 1519-1523. Macefield, G, Burke, D and Gandevia, SC (1989) The cortical distribution of muscle and cutaneous afferent

F. MAUGUIERE

projections from the human foot. Electroencephalogr. Clin. Neurophysiol., 72, 518-528. Manzano, GM, Negrao, ? and Nobrega JAM (1998) The N18 component of the median nerve SEP is not reduced by vibrations. Electroencephalogr. Clin. Neurophysiol., 108: 259-266. Mauguiere, F (1983) Les potentiels evoques somesthesiques cervicaux chez Ie sujet normal. Analyse des aspets obtenus selon Ie siege de I'electrode de reference. Rev. EEG. Neurophysiol., 13: 259-272. Mauguiere, F (1987) Short-latency somatosensory evoked potentials to upper limb stimulation in lesions of brainstem, thalamus and cortex. In: RJ Ellingson, NMF Murray and AM Halliday (Eds.), The London Symposia. Electroencephalogr. Clin. Neurophysiol. Elsevier, Amsterdam, Suppl. 39, pp. 302-309. Mauguiere, F (1995) Evoked potentials. In: JW Osselton, CD Binnie, R. Cooper, CJ Fowler, F. Mauguiere and PF Prior (Eds.), Clinical Neurophysiology: EMG, Nerve Conduction and Evoked Potentials. Butterworth-Heinemann, Oxford, pp. 325-563. Mauguiere F (1999) Somatosensory evoked potentials: Normal responses, abnormal waveforms and clinical applications in neurological diseases. In: E. Niedermeyer and F. Lopez da Silva (Eds.), Electroencephalography, Basic principles. Clinical Applications and Related Fields. William & Wilkins, New York, Nth ed., pp. 1014-1058. Mauguiere, F (2000) Anatomical origin of the cervical N13 potential evoked by upper extremity stimulation. J. Clin. Neurophysiol., 17: 236-245. Mauguiere, F and Courjon, J (1980) Effects of intravenously injected clonazepam on the cortical somatosensory evoked response in dyssynergia cerebellaris myoclonica. In: H Lechner and A Aranibar (Eds.), EEG and Clinical Neurophysiology. Excerpta Medica, Amsterdam, pp. 433-444. Mauguiere, F and Courjon, J (1981) The origins of shortlatency somatosensory evoked potentials in humans. A clinical contribution. Ann. Neurol., 9: 607-611. Mauguiere, F and Desmedt, JE (1991) Focal capsular vascular lesions can selectively deafferent the prerolandic or the parietal cortex: Somatosensory evoked potentials evidence. Ann. Neurol., 30: 71-75. Mauguiere, F and Ibanez, V (1985) The dissociation of early SEP components in lesions of the cervicomedullary junction: a cue for routine interpretation of abnormal cervical responses to median nerve stimulation. Electroencephalogr. Clin. Neurophysiol., 62: 406-420. Mauguiere, F and Ibanez, V (1990) Loss of parietal and frontal somatosensory evoked potentials in hemispheric .deafferentation. In: PM Rossini and F. Mauguiere (Eds.), New Trends and Advanced Tech-

SOMATOSENSORY EVOKED RESPONSES

niques in Clinical Neurophysiology. EEG J. Elsevier, Amsterdam, Supp!. 41, pp. 274-285. Mauguiere, F and Restuccia, D (1991) Inadequacy of the forehead reference montage for detecting spinal N13 potential abnormalities in patients with cervical cord lesion and preserved dorsal column function. Electroencephalogr. Clin. Neurophysiol., 79: 448-456. Mauguiere, F, Bard, J and Courjon, J (1981) Les potentiels evoques somesthesiques dans la dyssynergie cerebelleuse myoclonique progressive. Rev. EEG. Neurophysiol., 11: 174-182. Mauguiere, F, Brunon, AM, Echallier, JF and Courjon J (1982) Early somatosensory evoked potentials in thalamocortical lesions of the lemniscal pathways in humans. In: 1. Courjon, F. Mauguiere and M. Revol (Eds.), Clinical Applications of Evoked Potentials in Neurology, Advances in Neurology. Raven Press, New York, Vo!. 32, pp. 321-338. Mauguiere F, Desmedt JE and Courjon J (1983a) Astereognosis and dissociated loss of frontal or parietal components of somatosensory evoked potentials in hemispheric lesions. Brain, 106: 271-311. Mauguiere, F, Schott, Band Courjon, J (1983b) Dissociation of early SEP components in unilateral traumatic section of the lower medulla. Ann. Neurol., 13: 309-313. Mauguiere, F, Desmedt, 1£ and Courjon, J (1983c) Neural generators of N18 and P14 far-field somatosensory evoked potentials studied in patients with lesions of thalamus or thalamo-cortical radiations. Electroencephalogr. Clin. Neurophysiol., 56: 283-292. Mauguiere, F, Broussolle, E and Isnard, J (1993) Apomorphine induced relief of the akinetic-rigid syndrome and early median nerve somatosensory evoked potentials in Parkinson's disease. Electroencephalogr. Clin. Neurophysiol., 88: 243-254. Mauguiere, F, Merlet, I, Forss, N, Vanni, S, Jousmiiki, V, Adeleine, P and Hari, R (1997) Activation of a distributed cortical network in the human brain. A dipole modelling study of magnetic fields evoked by median nerve stimulation. Part I: Location and activation timing of SEF sources. Electroencephalogr. Clin Neurophysiol., 104: 281-289. Mauguiere, F, Allison, T, Babiloni, C, Buchner, H, Eisen, AA, Goodin, DS, Jones, SJ, Kakigi, R, Matsuoka, S, Nuwer, M, Rossini, PM and Shibasaki, H (1999) Somatosensory evoked potentials. In: G Deuschl and A Eisen (Eds.), Recommendations for the Practice of Clinical Neurophysiology. Electroencephalogr. Clin. Neurophysiol.. Elsevier Science BV Amsterdam, Supp!. 52, pp. 79-90. Mazzini, L, Zaccala, M and Balzarini, C (1994) Abnormalities of somatosensory evoked potentials in spasmodic torticollis. Mov. Disord., 9: 426-430.

71 McCarthy, G, Wood, CC and Allison, T (1991) Cortical somatosensory evoked potentials: 1. Recordings in the monkey Macaca fascicularis. J. Neurophysiol., 66: 53-63. Mervaala, E, Partanen, JV, Keranen, T, Penttila, M and Riekkinen, P (1984) Prolonged cortical somatosensory evoked potential latencies in progressive myoclonus epilepsy. J. Neurol. Sci., 64: 131-135. Mirna, T, Terada, K, Maekawa, M, Nagamine, T, Ikeda, A and Shibasaki, H (1996) Somatosensory evoked potentials following proprioceptive stimulation of fingers in man. Exp. Brain Res., 29: 440-443. Mirna, T, Terada, K, Ikeda, H, Fukuyama, T, Takigawa, J, Kimura, J and Shibasaki, H (1997) Afferent myoclonus studied by proprioception-related SEPs. Electroencephalogr. Clin. Neurophysiol., 104: 51-59. Meller, AR, Jannetta, PJ and Burgess, 1£ (1986) Neural generator of the somatosensory evoked potentials recording from the cuneate nucleus in man and monkeys. Electroencephalogr. Clin. Neurophysiol., 65: 241-248. Morioka, T, Shima, F, Kato, M, Fukui, M (1991) Direct recordings of the somatosensory evoked potentials in the vicinity of the dorsal column nuclei in man: their mechanisms and contribution to the scalp far-field potentials. Electroencephalogr. Clin. Neurophysiol., 80: 221-227. Murase, N, Kaji, R, Shimazu, H, Katayama-Hirota, M, Ikeda, A, Kohara, N, Kimura, J, Shibasaki, H and Rothwell, JC (2000) Abnormal premovement gating of somatosensory input in writer's cramp. Brain, 123: 1813-1829. Nakanishi, T, Takita, K and Toyokura, Y (1973) SEPs to tactile tap in man. Electroencephalogr. Clin. Neurophysiol., 34: 1-6. Nakanishi, T, Shimada, Y, Sakuta, M and Toyokura, Y (1978) The initial positive component of the scalp recorded somatosensory evoked potential in normal subjects and in patients with neurological disorders. Electroencephalogr. Clin. Neurophysiol., 45: 26-34. Nakanishi, T, Tamaki, M, Ozaki, Y and Arasaki, K (1983) Origins of short latency somatosensory evoked potentials to median nerve stimulation. Electroencephalogr. Clin. Neurophysiol., 56: 74-85. Nicholson-Peterson, N, Schroeder, CE and Arezzo, JC (1995) Neural generators of the early cortical somatosensory evoked potentials in the awake monkey. Electroencephalogr. Clin. Neurophysiol., 96: 248-260. Noth, J, Engel, L, Friedemann, HH and Lange, HW (1984) Evoked potentials in patients with Huntington's disease and their offspring. Electroencephalogr. Clin. Neurophysiol., 59: 134-141.

72 Obeso, JA, Rothwell, JC and Marsden, CD (1985) The spectrum of cortical myoclonus: from focal reflex jerks to spontaneous motor epilepsy. Brain, 108: 193-224. Oepen, G, Doerr, M and Thoden, U (1981) Visual (YEP) and somatosensory (SSEP) evoked potentials in Huntington's chorea. Electroencephalogr. Clin. Neurophysiol., 51: 666-670. Okada, YC, Tanenbaum, R, Williamson, SJ and Kaufman, L (1984) Somatotopic organization of the human somatosensory cortex revealed by neuromagnetic measurements. Exp. Brain Res., 56: 197-205. Onofrj, M, Ferracci, F, Fulgente, T, Malatesta, G and Ghilardi, MF (1994) Effects of drug manipulation on anterior components of somatosensory evoked potentials in a parkinsonian animal model. Drugs Exp. Clin. Res., 20: 29-36. Onofrj, M, Fulgente, T, Malatesta, G, Ferracci, F, Thomas, A, Curatola, L, Bolletini, F and Ragno, M (1995) The abnormality of N30 somatosensory evoked potential in idiopathic Parkinson's disease is unrelated to disease stage or clinical scores and insensitive to dopamine manipulations. Mov. Disord., 10: 71-80. Ozaki, I, Shimamura, H, Baba, M and Matsunaga, M (1996) N30 in Parkinson's disease. Neurology, 47: 303-305. Papakostopoulos, D and Crow, HJ (1980) Direct recording of the somatosensory evoked potentials from the cerebral cortex of man and the difference between precentral and postcentral potentials. In: JE Desmedt (Ed.), Clinical Uses of Cerebral, Brainstem and Spinal Somatosensory Evoked Potentials. Progr. Clin. Neurophysiol. Karger, Basel, Vol. 7, pp. 15-26. Papakostopoulos, D, Cooper, R and Crow, HJ (1974) Cortical potentials evoked by finger displacement in man. J. Physiol., 245: 70P-72P. Papakostopoulos, D, Cooper, R and Crow, HJ (1975) Inhibition of cortical evoked potentials and sensation by self-initiated movement in man. Nature, 258: 321-324. Pelosi, L, Cracco, JB, Cracco, RQ and Hassan, NF (1988) Comparison of scalp distribution of short latency somatosensory evoked potentials (SSEPs) to different stimulation of different nerves in the lower extremity. Electroencephalogr. Clin. Neurophysiol., 71: 422-428. Phillips, CG, Powell, TP and Wiesendanger, M (1971) Projection from low-threshold muscle afferents of hand and forearm to area "a" of the baboon cortex. J. Physiol., 217: 419-446. Pierantozzi, M, Mazzone, P, Bassi, A, Rossini, PM, Peppe, A, Altibrandi, MG, Stefani, A and Stanzione, P (1999) The effect of deep brain stimulation on the frontal N30 component of somatosensory evoked potentials in advanced Parkinson's disease patients. Electroencephalogr. CUn. Neurophysiol., 110: 1700-1707.

F. MAUGUIERE

Plasmati, R, Blanco, M, Michelucci, R, Rubboli, G, Salvi, F and Tassinari, CA (1989) SEPs study in idiopathic infantile epilepsies. In: A. Beaumanoir, H. Gastaut and R. Naquet (Eds.), Reflex Seizures and Epilepsies, Editions Medecine et Hygiene. Geneva, pp.75-81. Pratt, H and Starr, A (1981) Mechanically and electrically evoked somatosensory potentials in humans; scalp and neck distribution of short latency components. Electroencephalogr. Clin. Neurophysiol., 51: 138-147. Reilly, JA, Hallett, M, Cohen, LG, Tarkka, 1M and Dang, N (1992) The N30 component of somatosensory evoked potentials in patients with dystonia. Electroencephalogr. Clin. Neurophysiol., 84: 243-247. Restuccia, D (2000) Anatomic origin of P13 and P14 scalp far-field potentials. J. Clin. Neurophysiol., 17: 246257. Restuccia, D and Mauguiere, F (1991) The contribution of median nerve SEPs in the functional assessment of the cervical spinal cord in syringomyelia. A study of 24 patients. Brain, 114: 361-379. Restuccia, D, Di Lazzaro, V, Valeriani, M, Tonali, P and Mauguiere, F (1992) Segmental dysfunction of the cervical cord revealed by abnormalities of the spinal N13 potential in cervical spondylotic myelopathy. Neurology, 42: 1054-1063. Restuccia, D, Di Lazzaro, V, Valeriani, M, Conti, G, Tonali, P and Mauguiere, F (1995) Origin and distribution of P13 and Pl4 far-field potentials after median nerve stimulation. Scalp, nasopharyngeal and neck recordings in healthy subjects and in patients with cervical and cervico-medullary lesions. Electroencephalogr. Clin. Neurophysiol., 96: 371-384. Restuccia, D, Di Lazzaro, V, Valeriani, M, Mariotti, MG, Torrioli, MG, Tonali, P and Mauguiere, F (1996) Brainstem somatosensory dysfunction in a case of long-standing left hemispherictomy with removal of the left thalamus: a nasopharyngeal and scalp SEP study. Electroencephalogr. CUn. Neurophysiol., 100: 184-188. Restuccia, D, Valeriani, M, Barba, C, Le Pera, D, Tonali, P and Mauguiere, F (1999) Different contribution of joint and cutaneous inputs to early scalp somatosensory evoked potentials. Muse. Nerve, 22: 910--919. Restuccia, D, Valeriani, M, Grassi, E, Gentili, G, Mazza, S, Tonali, P and Mauguiere, F (2oo2a) Contribution of gabaergic circuitry in shaping scalp somatosensory evoked responses in humans: specific changes after single-dose administration of tiagabine. Clin. Neurophysiol., 113: 656-671. Restuccia, D, Valeriani, M, Insola, A, Lo Monaco, M, Grassi, E, Barba, C, Le Pera, D, Tonali, P and Mauguiere, F (2002b) Modality-related scalp responses after selective electrical stimulation of cutaneous and muscular upper limb afferents in

SOMATOSENSORY EVOKED RESPONSES

humans: specific and unspecific components. Muse. Nerve, 26: 44-54. Ridding, MC and Rothwell, JC (1999) Afferent input and cortical organization: a study with magnetic stimulation. Exp. Brain Res., 126: 536-544. Ridding, MC, Taylor, JL and Rothwell, JC (1995) The effect of voluntary contraction on cortico-cortical inhibition in human motor cortex. J. Physiol., 487: 541-548. Riffel, B, Stohr, M and Komer, S (1984) Spinal and cortical potentials following stimulation of posterior tibial nerve in the diagnosis and localization of spinal cord diseases. Electroeneephalogr. Clin. Neurophysiol., 58: 400-407. Roland, PE, Larsen, B, Larsen, NA and Skinhej, E (1980) Supplementary motor area and other cortical areas in organization of voluntary movements in man. J. Neurophysiol., 43: 118-136. Rossi, S, Tecchio, F, Pasqualetti, P, Ulivelli, M, Pizzella, V, Romani, GL, Passero, S, Battistini, N and Rossini, PM (2002) Somatosensory processing during movement observation in humans. Clin. Neurophysiol., 113, 16-24. Rossini, PM, Babiloni, F, Bernardi, G, Cecchi, L, Johnson, PB, Malentacca, A, Stanzione, P and Urbano, A (1989) Abnormalities of short-latency somatosensory evoked potentials in parkinsonian patients. Electroencephalogr. Clin. Neurophysiol., 74: 277-289. Rossini, PM, Paradiso, C, Zarola, F, Bernardi, G, Caramia, MD, Margari, L and Ferrari, E (1991) Brain excitability and long-latency muscular arm responses: noninvasive evaluation in healthy and parkinsonian subjects. Electroencephalogr. Clin. Neurophysiol., 81: 454-465. Rossini, PM, Traversa, R, Boccasena, P, Martino, G, Passarelli, F, Pacifici, L, Bernardi, G and Stanzione, P (1993) Parkinson's disease and somatosensory evoked potentials: apomorphine-induced transient potentiation of frontal components. Neurology, 43: 2495-2500. Rossini, PM, Bassetti, MA and Pasqualetti, P (1995) Median nerve somatosensory evoked potentials. Apomorphine-induced transient potentiation of frontal components in Parkinson's disease and in parkinsonism. Electroencephalogr. Clin. Neurophysiol., 96: 236-247. Rossini, PM, Cararnia, D, Bassetti, MA, Pasqualetti, P, Tecchio, F and Bernardi, G (1996) Somatosensory evoked potentials during the ideation and execution of individual finger movements. Muse. Nerve, 19, 191-202. Rothwell, LJC, Obeso, JA and Marsden, CD (1984) On the significance of giant somatosensory evoked potentials in cortical myoclonus. J. Neurol. Neurosurg. Psychiat., 47: 33-42.

73 Schieppati, M and Ducati, A (1984) Short latency cortical potentials evoked by tactile air-jet stimulation of body and face in man. Electroencephalogr. Clin. Neurophysiol., 58: 418-425. Seyal, M (1991) Cortical reflex myoclonus. A study of the relation between giant somatosensory evoked potentials and motor excitability. J. Clin. Neurophysiol., 8: 95-101. Seyal, M, Emerson, RG and Pedley, TA (1983) Spinal and early scalp-recorded components of the somatosensory evoked potential following stimulation of the posterior tibial nerve. Electroencephalogr. Clin. Neurophysiol., 55: 320-330. Shibasaki, Hand Kuroiwa, Y (1975) Electroencephalographic correlates of myoclonus. Electroencephalogr. Clin. Neurophysiol., 39: 455-463. Shibasaki, H, Yamashita, Y, Neshige, R, Tobimatsu, Sand Fukui, R (1985a) Pathogenesis of giant somatosensory evoked potentials in progressive myoclonic epilepsy. Brain, 108: 225-240. Shibasaki, H, Neshige, R and Hashiba, Y (1985b) Cortical excitability after myoclonus. Jerk-locked somatosensory evoked potentials. Neurology, 35: 36-41. Shibasaki, H, Nakamura, M, Nishida, S, Kakigi, R and Ikeda, A (1990) Waveform decomposition of 'giant SEP' and its computer model for scalp topography. Electroencephalogr. Clin. Neurophysiol., 77: 286-294. Shimazu, H, Kaji, R, Murase, N, Kohara, N, Ikeda, A, Shibasaki, H, Kimura, J and Rothwell, JC (1999) Premovement gating of short-latency somatosensory evoked potentials. Neuroreport, 10: 2457-2460. Shimoji, K, Matsuki, M and Shimizu, H (1977) Waveform characteristics and spatial distribution of evoked spinal electrogram in man. J. Neurosurg., 46: 304-313. Slimp, JC, Tamas, LB, Stolow, WC and Wyler, AR (1986) Somatosensory evoked potentials after removal of somatosensory cortex in man. Electroencephalogr. Clin. Neurophysiol., 65: 111-117. Small, M and Matthews, WB (1984) A method of calculating spinal cord transit time from potentials evoked by tibial nerve stimulation in normal subjects and patients with spinal cord disease. Electroencephalogr. Clin. Neurophysiol., 59: 156-164. Sonoo, M (2000) Anatomic origin and clinical application of the widespread N18 potential in median nerve somatosensory evoked potentials. J. Clin. Neurophysiol., 17: 258-268. Spehlmann, R (1985) Evoked Potential Primer. Visual, Auditory and Somatosensory Evoked Potentials in Clinical Diagnosis. Butterworth, Boston. Stanzione, P, Traversa, R, Pierantozzi, M, Semprini, R, Loberti, M, Peppe, A, Santilli, AM and Bernardi, G (1997) SEPs N30 amplitude in pharmacologically

74 induced rigidity: relationship with the clinical status. Eur. J. Neurol., 4: 24-38. Stephan, KM, Fink, GR, Passingham, RE, Silbersweig, D, Ceballos-baumann, AO, Frith, CD and Frackowiak, RS (1995) Functional anatomy of the mental representation of upper extremity movements in healthy subjects. J. Neurophysiol., 73: 373-386. Stohr, M, Dichgans, J, Voigt, K and Buettner, UW (1983) The significance of somatosensory evoked potentials for localization of unilateral lesions within the cerebral hemispheres. J. Neuro. Sci., 61: 49--63. Tanji, J and Wise, SP (1981) Submoda1ity distribution in sensorimotor cortex of the unanesthetized monkey. J. Neurophysiol. 45: 467-481. Tassinari, CA, Rubboli, G and Shibasaki, H (1998) Neurophysiology of positive and negative myoclonus. Electroencephalogr. Clin. Neurophysiol., 107: 181195. Tinazzi, M and Mauguiere, F (1995) Assessment of intraspinal and intracranial conduction by P30 and P39 Tibial nerve somatosensory evoked potentials in cervical cord and hemispheric lesions. J. Clin. Neurophysiol., 12: 237-253. Tinazzi, M, Zanette, G, Fiaschi, A and Mauguiere, F (1996a) Effects of stimulus rate on the cortical posterior tibial nerve SEPs; A topographic study. Electroencephalogr. Clin. Neurophysiol., 100: 210219. Tinazzi, M, Zanette, G, Bonato, C, Manganotti, P, Polo, A, Fiaschi, A and Mauguiere, F (1996b) Neural generators of tibial nerve P30 somatosensory evoked potential studied in patients with a focal lesion of the cervicomedullary junction. Muse. Nerve, 19: 1538-1548. Tinazzi, M, Zanette, G, Manganotti, B, Bonato, C, Polo, A, Fiaschi, A and Mauguiere, F (1997a) Amplitude changes of tibial nerve cortical SEPs when using ipsilateral or contralateral ear as reference. J. Clin. Neurophysiol., 14: 217-225. Tinazzi, M, Zanette, G, La Porta, F, Polo, A, Volpato, D, Fiaschi, A and Mauguiere, F (1997b) Selective gating of lower limb cortical somatosensory evoked potentials (SEPs) during passive and active foot movements. Electroencephalogr. Clin. Neurophysiol., 104: 312-321 Tinazzi, M, Fiaschi, A, Mauguiere, F, Manganotti, P, Polo, A, Bonato, C and Zanette, G (1998) Effects of voluntary contraction on tibial nerve somatosensory evoked potentials: gating of specific cortical responses. Neurology, 50: 1655-1661. Tinazzi, M, Frasson, E, Polo, A, Tezzon, F, Bovi, P, Deotto, L, Mauguiere, F, Fiaschi, A and Ferrari, G (1999) Evidence for an abnormal cortical sensory processing in dystonia: selective enhancement of lower

F. MAUGUIERE

limb P37-N50 somatosensory evoked potential. Mov. Disord., 14: 473-480. Tinazzi, M, Priori, A, Bertolasi, L, Frasson, E, Mauguiere, F and Fiaschi, A (2000) Abnormal central integration of a dual somatosensory input in dystonia: Evidence for sensory overflow. Brain, 123: 42-50. Tomberg, C, Desmedt, JE, Ozaki, I and Noel, P (1991) Nasopharyngeal recordings of somatosensory evoked potentials document the medullary origin of the Nl8 far-field. Electroencephalogr. Clin. Neurophysiol., 80: 496-503. Topper, R, Schwarz, M, Podoll, K, Domges, F and Noth, J (1980) Absence of frontal somatosensory evoked potentials in Huntington's disease. Brain, Ill: 87101. Tsuji, S, Liiders, H, Lesser, RP, Dinner, DS and Klem, G (1984) Subcortical and cortical somatosensory potentials evoked by posterior tibial nerve stimulation: normative values. Electroencephalogr. Clin. Neurophysiol., 59: 214-228. Tsuji, S, Muray, Y and Kadoya, C (1988) Topography of somatosensory evoked potentials to median nerve stimulation in patients with cerebral lesions. Electroencephalogr. Clin. Neurophysiol., 71: 280-288. Tsumoto, T, Hirose, N, Nonaka, S and Takahashi, M (1972) Analysis of somatosensory evoked potentials to lateral popliteal nerve stimulation in man. Electroencephalogr. Clin. Neurophysiol., 33: 379-388. Urasaki, E, Wada, S, Kadoya, C, Yokota, A, Matsuoka, S and Shima, F (1990) Origin of scalp far-field N18 of SSEPs in response to median nerve stimulation. Electroencephalogr. Clin. Neurophysiol., 77: 39-51. Urasaki, E, Tokimura, T, Yasukouchi, H, Wada, S and Yokota, A (1993) P30 and N33 of posterior tibial nerve SSEPs are analogous to Pl4 and Nl8 of median nerve SSEPs. Electroencephalogr. Clin. Neurophysiol., 88: 525-529. Valeriani, M, Restuccia, D, Di Lazzaro, V, Le Pera, D and Tonali, P (1997a) The pathophysiology of giant SEPs in cortical myoclonus: a scalp topography and dipolar source modelling study. Electroencephalogr. Clin. Neurophysiol., 104: 122-131. Valeriani, M, Restuccia, D, Di Lazzaro, V, Le Pera, D, Scerrati, M, Tonali, P and Mauguiere, F (1997b) Giant central N20-P22 with normal area 3b N20-P20: an argument in favour of an area 3a generator of early median nerve cortical SEPs? Electroencephalogr. Clin. Neurophysiol., 104: 60--67. Valeriani, M, Restuccia, D, Di Lazzaro, V, Le Pera, D, Carba, C, Tonali, P and Mauguiere, F (1998a) Dipolar sources of the early scalp SEPs to upper limb stimulation. Effects of increasing stimulus rates. Exp. Brain Res., 120: 306-315.

SOMATOSENSORY EVOKED RESPONSES

Valeriani, M, Restuccia, D, Di Lazzaro, V, Barba, C, Le Pera, D and Tonali, P (1998b) Dissociation induced by voluntary movement between two different components of the centro-parietal P40 SEP to tibial nerve stimulation. Electroencephalogr. Clin. Neurophysiol., 108: 190-198. Valeriani, M, Restuccia, D, Barba, C, Tonali, P and Mauguiere, F (2000) Central scalp projection of the N30 SEP source activity after median nerve stimulation. Muse. Nerve, 23: 353-360. Valeriani, M, Insola, A, Restuccia, D, Le Pera, D, Mazzone, P, Altibrandi MG and Tonali, P (2001) Source generators of the early somatosensory evoked potentials to tibial nerve stimulation: an intracerebral and scalp recording study. Clin. Neurophysiol., 112: 1999-2006. Vas, GA, Cracco, JB and Cracco, RQ (1981) Scalp recorded short latency cortical and subcortical somatosensory evoked potentials in peroneal nerve stimulation. Electroencephalogr. Clin. Neurophysiol., 51: 1-8. Vera, CL, Perot, PL, and Fountain, EI (1983) Scalp recorded somatosensory evoked potentials after stimulation of the tibial nerve. Electroencephalogr. Clin. Neurophysiol., 56: 159-168. Wabersky, TD, Buchner, H, Perkuhn, M, Gobbele, R, Wagner, M, Kueker, Wand Silny, J (1999) N30 and the effect of explorative finger movements: a model of the contribution of the motor cortex to early somatosensory potentials. Clin. Neurophysiol., 110: 1589-1600. Wagner, W (1996) Scalp, earlobe and nasopharyngeal recordings of the median nerve somatosensory evoked P14 potential in coma and brain death. Brain, 119: 1507-1521. Wang, J, Cohen, LJ and Hallett, M (1989) Scalp topography of somatosensory evoked potentials fol-

75 lowing electrical stimulation of femoral nerve. Electroencephalogr. Clin. Neurophysiol., 74: 112-123. Wood, CC, Cohen, D, Cuffin, BN, Yarita, M and Allison, T (1985) Electrical sources in human somatosensory cortex: identification by combined magnetic and potential recordings. Science, 227: 1051-1053. Yamada, T (2000) Neuroanatomic substrates of lower extremity somatosensory evoked potentials. J. Clin. Neurophysiol., 17: 269-279. Yamada, T, Kimura, J and Nitz, DM (1980) Short latency somatosensory evoked potentials following median nerve stimulation in man. Electroencephalogr. Clin. Neurophysiol., 48: 367-376. Yamada,T, Machida, M and Kimura, J (1982) Far-field somatosensory evoked potentials after stimulation of the tibial nerve. Neurology, 32: 1151-1158. Yamada, T, Kayamori, R, Kimura, J and Beck, DO (1984) Topography of SEPs after stimulation of the median nerve. Electroencephalogr. Clin. Neurophysiol., 59: 29-43. Yamada, T, Ishida, T, Kudo, Y, Rodnitzi, RL and Kimura, J (1986) Clinical correlates of abnormal PI4 in median SEPs. Neurology, 36: 765-771. Yamada, T, Matsubara, M, Shiraishi, M, Yeh, M and Kawasahi, T (1996) Topographic analysis of somatosensory evoked potentials following stimulation tibial, sural and lateral femoral cutaneous nerve. Electroencephalogr. Clin. Neurophysiol., 100: 33-43. Zanette, G, Tinazzi, M, Manganotti, P, Bonato, C and Polo, A (1995) Two distinct cervical N13 potentials are evoked by ulnar nerve stimulation. Electroencephalogr. Clin. Neurophysiol., 96: 114-120. Zanette, G, Tinazzi, M, Polo, A, Rizzuto, N (1996) Motor neuron disease with pyramidal tract dysfunction involves the cortical generators of the early somatosensory evoked potential to tibial nerve stimulation. Neurology, 47: 932-938.

Movement Disorders Handbook of Clinical Neurophysiology, Vol. I M. Hallett (Ed.) © 2003 Elsevier B.Y. All rights reserved

77 CHAPTER 6

Coherence, cortico-cortical Christian Gerloff":", Christoph Braun" and Mark Hallett" a

Cortical Physiology Research Group, Department of General Neurology, Hertie Institute for Cl~~i~'al Brain Research, and h MEG Center. Eberhard-Karls University Tlibingen, Hoppe-Seyler Strasse 3, D-72076 Tiibingen. Germany C HMCS, MNB, NINDS, National Institutes of Health, Bethesda, MD 20892, USA

6.1. Introduction Synchronization (coupling) of distant brain areas provides the basis for integration of complex information and is thus a fundamental element in the neuronal representation and control of behavior. Perceptions or actions can be represented in the brain by large numbers of distributed neurons firing in synchrony (Singer, 1994). In the visual system, this has been referred to as 'binding' (Singer and Gray, 1995). From correlated single unit discharges, the 'binding' concept has been extended to coherent oscillations of local field potentials (LFPs) (Singer, 1993) and from high-frequency (gamma) (Miltner et aI., 1999; Rodriguez et aI., 1999) to oscillatory activity in lower frequency bands (below 30 Hz; alpha, beta; (Classen et aI., 1998; Gerloff et aI., 1998; Mirna et aI., 2000a» in the surface EEG. It has been shown in laboratory animals (Bressler et aI., 1993; Singer and Gray, 1995) and in humans (Classen et aI., 1998; Miltner et aI., 1999; Rodriguez et aI., 1999) that the presence of interregionally coherent oscillations can represent relevant aspects of behavior, e.g. expectation of stimuli after associative learning, perception of faces, or coordination of motor reactions in reaction time experiments. In addition, it is thought that interregional synchronization plays an important role in any form of synaptic modulation, e.g. during learning or remodeling of neuronal representations after lesions. Disturbance of interregional coupling causes disconnection syndromes like associative object agnosia (Carlesimo et aI., 1998), left hand apraxia

* Correspondence to: Dr. Ch. Gerloff, Dept. of Neurology, University of Tiibingen, Hoppe-Seyler Strasse 3, 0-72076 Tiibingen, Germany. E-mail address:[email protected] Tel: +497071 298 6525.

(Tanaka et aI., 1996; Marangolo et aI., 1998) or alien hand syndrome (Geschwind et aI., 1995). The clinical observations suggest that functional decoupling could be a mechanism underlying several neurological syndromes. Of note, modulation of interregional synchronization may occur in the absence of changes in mean neuronal firing rates or LFP amplitudes, and without significant alterations of other markers of regional cerebral activation (evoked potential amplitudes, local spectral power changes in EEG or MEG; regional cerebral blood flow or blood oxygenation levels in positron emission tomography or functional magnetic resonance imaging). Any mathematical model applied needs to incorporate this physiological feature. After covariance methods had been used in the context of time-averaged evoked responses in the 80s and 90s (Gevins et aI., 1987, 1989), the focus has lately shifted to methods allowing for correlation analyses in the frequency domain based on single trials. The standard approach which will be outlined in detail below is the analysis of cortico-cortical coherence. Coherence analysis has the advantage that it does not make assumptions about the phase relationship between two signals of concern. While time-domain averaging and subsequent covariance analysis requires the signals to be phase-locked to a certain event, coherence analysis in the frequency domain can determine synchronization independent of phase. Phase (tp) information, however, is also available from ordinary coherence analysis. It corresponds to the relation of imaginary (Im(x» to real part (Re(x» of the complex cross-spectrum and can be analyzed separately (in addition to coherence magnitude). Another advantage of coherence relates to the theory of frequency coding. This concept allows for a virtually unlimited number of parallel,

78

C. GERLOFF ET AL.

independent encoding processes within the same neuronal assembly or inside the same distributed network in different frequency bands.

6.2. Methodology 6.2.1. The basic approach: coherence estimation

The most commonly used approach is based on discrete Fourier transforms, usually Fast Fourier Transformation (FFT). The general analytical approach can be separated in 5 steps: (1) Segmentation of raw data into epochs (2) Minimization of spectral leakage (e.g. Hamming window) (3) Computation of coherence (Coh) based on FFT (4) Computation of Task-Related (or event-related) Coherence (TRCoh or ERCoh) by comparing coherence values during a given task with coherence values during rest (for details of TRCoh and ERCoh computation see below) (5) Statistical evaluation (e.g. ANOVA) after stabilizing the variances (e.g. inverse hyperbolic tangent transformation, tanh-I) Depending on the statistical model used, steps 4 and 5 can be merged and the task dependency of coherence changes can be tested by contrast analyses, post-hoc tests or other forms of determining significance criteria (e.g. confidence intervals, statistical probability mapping based on permutation analyses as described below). The FFT approach implies that all information is collapsed within one epoch, and it is assumed that the signal is stationary within the epoch. For nonoverlapping epochs, the length (1) of the epoch is the temporal resolution (Res(t)=T [s]). If T is in seconds, the frequency resolution Res(F) =liT (Hz). For facilitation of statistical analysis, it is recommended to use non-overlapping epochs. However, if higher temporal resolution is attempted, overlapping (sliding) windows can be used. The use of overlapping windows increases the risk of broadening a temporal event, but can improve resolution of the peak. The coherence values are calculated for each frequency bin>. according to Eq. (1). 2

Coh (>.)=IR (>')12 Ifxi>') 1 xy x)' fxx(>')fvi>')

(1)

Equation (1) is the extension of Pearson's correlation coefficient to complex number pairs (Papoulis, 1984; Bronstein and Semendjajew, 1987). In this equation, f denotes the spectral estimate of two EEG signals x and y for a given frequency bin (X), The numerator contains the cross-spectrum for x and y (!xy), the denominator the respective auto-spectra for x (fxx) and y (1;)'). For each frequency, the coherence value (Cohxy) is obtained by squaring the magnitude of the complex correlation coefficient R, and is a real number between 0 and I. Some authors prefer to use the square-root of coherence, which is termed coherency.

6.2.2. Functional measures: task-related coherence and event-related coherence

With few exceptions, functional changes of cortico-cortical coherence in relation to behavioral tasks are evaluated and compared with coherence estimates during baseline or control conditions (Rappelsberger et al., 1994; Leocani et aI., 1997; Classen et aI., 1998; Gerloff et aI., 1998; Miltner et al., 1999; Mima et al., 200 I b). This reduces effects of inter-subject and inter-electrode-pair variability of absolute coherence values. If a continuous task is under study (i.e. continuous movement or force generation), task-related coherence (TRCohxy) is obtained by subtracting rest (Cohxy rest) from corresponding activation conditions (Cohxy activation), according to Eq. (2).

In this equation, coherence magnitude increments are expressed as positive values, and coherence decrements are expressed as negative values. Coherence increments or decrements between baseline and movement conditions for each pair of electrodes can be displayed as color-coded "link" plots, which permit the inspection of the magnitude and spatial patterns of TRCoh (Fig. I). This method also eliminates the bias in the absolute coherence introduced by the reference electrodes (Fein et aI., 1988; Rappelsberger and Petsche, 1988; Classen et aI., 1998; Gerloff et aI., 1998). If coherence changes are expected to occur as a phasic change, tightly bound to a single event, then it might be more appropriate to analyze event-related

79

COHERENCE,CORTICO-CORTICAL

9-11 Hz

20-22 Hz

Fig. 1. Color-coded "link" plot of TRCoh during simple finger movements. EEG data. Red to yellow lines indicate taskrelated increases of coherence (positive values, "coupling"). Blue and green colors indicate task-related decreases of coherence (negative values, "decoupling"), In this example, the major network nodes lie in the bilateral central region. The links converge predominantly on electrodes C3 and CP3, known to overlie roughly the primary motor, primary somatosensory and parietal cortices.

coherence (ERCoh). There are several ways to compute ERCoh. Example I. The mathematics of ERCoh can be identical with those of TRCoh, but ERCoh is computed on shorter epochs (i.e. 256 ms), if necessary with 50% (or 75%) overlap in order to arrive at (smoothed) 128 ms (or 64 ms) temporal resolution without reducing the frequency resolution further. This procedure has the advantage of not forcing the experimenter to make a priori assumptions about reactive frequency bands. The temporal evolution of coherence is automatically available across the full frequency range. This however is achieved at the expense of frequency resolution Res(F). According to Res(F) = liT (Hz), with T=0.256 s, Res(F)=3.9 Hz. Example 2. If the frequency range of interest is clear, then an approach similar to the one of eventrelated desynchronization can be used. The raw data is band-pass filtered for the frequency range of interest (a filter without phase distortion should be used). Then covariances are computed on the basis of single trials between electrodes of interest. Correlation in the time domain is identical with the real part of coherency (the imaginary part can be expressed as the correlation of the time-lagged signals). With this approach, the covariances can be plotted for each sampling point. However, this does not correspond to a temporal resolution of ERCoh that is only determined by the sampling rate. The trade-off between temporal and spectral resolution is similar to example 1. This is due to filtering. In order

to arrive at correlation values for certain frequency ranges, filtering is mandatory before covariances are computed. A filter of specific width in the frequency domain must have a minimum length in the time domain. Therefore, the filtered data are implicitly smoothed. For example, a steep band-pass filter for 10-10.9 Hz will cause smoothing of the data over several hundred milliseconds. This makes the results of example 2 similar to using FFT-based coherences in overlapping epochs as described in example I. Because of boundary effects, however, the results of examples I and 2 will never be absolutely identical. Example 3. In addition to ordinary coherence and covariance methods, wavelet analysis has been used to determine event-related changes of interregional coupling in the frequency domain. Wavelet analysis has a number of mathematical advantages and can attenuate the dilemma of the trade-off between Res(F) vs. Res(t). However, it has the inherent problem of determining physiologically valid wavelet shapes a priori. At this point, it is not clear how much of a bias can arise from this a priori decision, and thus, we prefer ordinary coherence analysis or covariance analysis on band-pass filtered data at this point. 6.2.3. Special aspects: multiple frequencies, statistical evaluation

Coherence values have certain properties which make it necessary to transform them prior to further statistical evaluation (Rosenberg et aI., 1989;

80

C. GERLOFF ET AL.

Farmer et a!., 1993; Halliday et a!., 1995; Amjad et al., 1997). For coherence values, which are real numbers between 0 and 1, the variance decreases for values at both ends of the range (Brockwell and Davis, 1991). For coherence estimates, the inverse hyperbolic tangent (tanh:': synonym arc-hyperbolic tangent) transformation of the square root of coherence cVCohx.P,)=RxvCA))) is typically used to normalize the underlying distribution of correlation coefficients and to stabilize the variances (var) of the distributions (Rosenberg et al., 1989; Farmer et al., 1993) (Eq. (3». A limitation of this transformation is related to its monotonically increasing slope (0--+ infinity). It stretches intervals close to 1 more than those close to O. The transformation is nearly linear when coherence is less than -0.6, so it has little effect on the interpretation or statistical analysis in this range. In the range of (non-subtracted) coherence magnitudes that are commonly observed between EEG or MEG signals, the transformed coherences have approximately constant variance. Although the transformed coherence is slightly less easily interpretable than raw coherence, it is more suitable for analyses of variance (ANOVA). 1

varitanh:'(RxvC A))) =-

.

(3)

2n

In this equation, var denotes the variance of the transformed coherence measure between two signals x and y at a frequency A, and n the number of trials entered into the coherence calculation. The tanh:' transformation is applied to each individual data set prior to subtraction of the coherence estimates in order to compute the task-related coherence TRCoh. Thus, for the statistical analysis, Eq. (2) becomes Eq. (4).

TRtanh-1Rxi A)

=tanh:' [Rxy(A)activation] -

tanh:' [Rxy(A)resl]

(4)

The TRtanh-'RxiA) values can be entered into a factorial ANOVA to assess the significance of TRCoh differences between different conditions (Gerloff et al., 1998). Alternatively, as mentioned above, the unsubtracted tanh-IRxy(A) values for each condition can be entered directly into the ANOVA including tanh-1Rxy(A)resl and significant task-related (TR) changes can be determined by contrast or posthoc analysis (Andres et al., 1999).

TRCoh is usually analyzed separately in different frequency bins. If the TRCoh properties across frequency bins are to be tested, a repeated-measures ANOVA design with frequency as a within-subjects factor is advisable. To account for violation of sphericity, the Greenhouse-Geisser or Huynh-Field correction should then be used. For appropriate statistical testing, data reduction can be necessary. With respect to the topographic distribution of coherences, a "region-of-interest" or "electrode pair/sensor pair-of-interest" (POI) approach is acceptable if clear a priori hypotheses can be formulated. Similarly, frequencies of interest can be defined a priori if the functional relevance of the respective frequency ranges can be determined on the basis of prior knowledge. Mere statistical parametric mapping, i.e. entering all possible combinations of signal pairings and frequency bins into the analysis is mostly inappropriate because of lack of statistical power and the need of corrections for multiple comparisons in the order of several thousands (e.g. 64 channels --+2,048 pairs --+ x 50 frequency bins from 1-50 Hz-e102,400 pairs). In addition to the "pair-of-interest" approach there are hypothesis-free techniques which can be used to get around this dilemma, for example, principal component analysis (PCA) (Andres et al., 1999) and "statistical probability mapping" (Lutzenberger et aI., 2002). In the PCA, the coherence values for all conditions are entered and the eigenvalues for each main component and the loading factors for each electrode pair are determined. The loading factors are ranked, plotted as a function of electrode pair, and the cut-off point is defined, for example visually determined as the first prominent change in slope when starting from the maximum value. The PCA can also be used to test the plausibility of POI that have been selected a priori on the basis of physiological and anatomical knowledge. To reduce the risk of false positives further, it should be considered to only accept electrodes with at least 2 links to other electrodes as potential nodes of a coherence network (Andres et al., 1999). We have used the PCA approach usually in order to determine the POI within selected frequency ranges. An interesting technique to select reactive frequency bins without making a priori assumptions has been referred to as "statistical probability mapping" (Lutzenberger et al., 2002). This method

81

COHERENCE, CORTICO-CORTICAL

has orginally been applied to spectral power values but can be extended to coherence estimates. The statistical probability mapping approach is based on permutation tests (Blair and Karniski, 1993). It includes corrections for multiple comparisons and for possible correlations between data from neighboring frequency bins. The starting point is the observed distribution of the t values for all frequency bins x coherence estimates. To ensure that tests for two consecutive frequency bins are significant, a new distribution of the minimal t values, tm, is computed for all pairs of neighboring frequency bins (time points) i and sensors}, according to Eq. (5). (5)

The t value tm and its corresponding p value Po.os are determined for which 5% of the observed tmi,j are greater. In the case of highly correlated data, Po,os is ::;0.05, for highly independent data, Po,os is >0.05. Then the distribution of the maximal t values, tmax' is computed for each of the n rand=2n permutations across n subjects of the observed values for the two conditions for all frequency bins i x coherence estimates j, according to Eq. (6).

(6) The critical t value t erit is now defined as the value where Po.os x nrand of the obtained t max are greater. This critical t value t erit is then applied as significance criterion to the observed data. Once the frequency bins of interest are defined, broad-band coherence values across multiple frequencies can be computed. For broad-band coherence, Cohxi~), Cohxi}..) is averaged over frequency bins j = }..min to }..max (with }..min and }..max corresponding to the lower and upper frequency bins in the chosen frequency band). To average the frequency bins the concept of pooled coherence can be used as described by Amjad et aI. (1997). For all subjects, electrode pairs, and conditions, the number of frequency bins pooled should be equal (Amjad et al., 1997). 6.2.4. Advanced computations: phase analysis, partial coherence analysis, source modeling

In the coherence estimate, phase P xy is delineated from the relation of imaginary (1m(x» to real part (Retx) of the complex cross-spectrum f .ry It can also

be expressed as the argument of the cross-spectrum, according to Eq. (7). (7)

This equation provides information about the temporal relationship (phase lag) of the 2 signals of interest. However, it is not possible to decide which of the signals leads or lags. One approach to solve this dilemma is the "constant phase shift and constant time lag model" (Mirna et aI., 2000b). This model is based on regression analysis of constant or frequency-independent time lags in multiple frequencies and is described in more detail in Chapter 6 of this volume. The model has been successfully applied to EEG-EMG coherence, but its application to cortico-cortical coherence is more difficult. While EEG-EMG (or MEG-EMG) coherences are raw coherence data, cortico-cortical coherences have to be corrected for influences of volume conduction and variable inter-electrode distances. As described above, this is done by computing task-related coherence, TRCoh. Accordingly, this approach will provide us with information that resembles a "taskrelated change of the phase lag between two signals". This information cannot simply be translated into conduction times or numbers of synapses involved. This information might be helpful if during the task, zero-phase lag synchronization is present, but not during rest. However, the physiological meaning of the magnitude of task-related changes of phase lags yet needs to be determined. An alternative approach to describing the "direction of information flow" between multivariate time series is "directed coherence" (Wang and Takigawa, 1992; Baccala and Sameshima, 2001; Mirna et aI., 2001a). The principle of this method is the decomposition of multivariate partial coherences computed from multivariate autoregressive models into two directed coherences rather than into one "ordinary" coherence (Eq. (1». Directed coherence is thought to reflect a frequency-domain representation of the concept of Granger causality (Baccala and Sameshima, 2001). It stresses the "lead" or "lag" of a signal by decomposing their interactions into "feedforward" and "feedback" aspects. So far, little attention has been paid to this method in respect to the analysis of task-related modulations of corticocortical interactions, although the method has been applied sporadically in this context for already more than one decade (Wang and Takigawa, 1992).

82 Directed coherence between two time series X and Y at a given frequency (A) is usually referred to as 'Yx/A) and is similar to the estimator called "directed transfer function" (DTF) (Kaminski et al., 2001). Of note, any phase computations and interpretations in the context of cortico-cortical coherences are only valid, if "reference-free" signals are used. If two signals with a common reference are compared, "zero-phase lag synchronization" might erroneously be introduced by the relative contribution of the reference signal. Bipolar leads or source derivations (e.g. Laplacian) must be computed if phase information is to be addressed. The general question of which reference should be used for cortico-cortical coherence analyses has been much debated. For coherence magnitude, in addition to bipolar and Laplacian signals, common average or monopolar recordings can also be used. The crucial point is that the reference signal must not be modulated as a function of the behavioral task. For example, if the primary sensorimotor cortex is the region of interest and the task is visually cued, significant task-related influences of earlobe electrodes are quite unlikely. This would be very different for an auditory task. The choice of the reference must depend on the physiological paradigm under study. Recently, the method of partial coherence analysis has been introduced into EEG spectral analysis. Most importantly, regarding coherence results in the alpha band partial coherence allows for assessing potential influences of the reference electrode(s). This could be the case with false high interhemispheric coherence between sensorimotor regions because of contamination of the reference electrode signal (earlobe or mastoid electrodes) by the occipito-parietal alpha rhythm (Mima et al., 2000a). Partial coherence is computed according to Eq. (8).

In this equation, ijA), J;,/A), andlziA) are autospectra of EEG (or MEG) signals X, Y, and Z for a given frequency (A). The terms fx/A),frz(A), and Iz/A) are

C. GERLOFF ET AL.

the cross-spectra between two of them for a given frequency (A). In this example, Z would be the signal corresponding to the occipito-parietal alpha rhythm, X would correspond to the signal over the left, Y over the right sensorimotor cortex. Equation (8) then provides the coherence between the two sensorimotor regions after removal of any potential linear influence of the occipito-parietal alpha rhythm. Figure 2 illustrates how task-related coherence patterns may change after eliminating the influence of the occipito-parietal alpha rhythm. The question of which sources are generating the coherence estimates can be addressed by using source models, i.e. considering brain regions, rather than raw data or digitally re-referenced or otherwise

Fig. 2. Task-related coherence (TRCoh) during a simple finger extension task. Epoch length, I s. Top, standard computation of TRCoh (movement minus baseline). Note the similar amplitudes of TRCoh between left sensorimotor region (LSM) and frontocentral midline cortex (ML) and between LSM and the right sensorimotor region (RSM). After removal of linear influences of the occipitoparietal alpha rhythm by partial coherence analysis, interhemispheric coherence (LSM-RSM) is relatively reduced as compared with intrahemispheric coherence (e.g. LSM-ML). The orange bar indicates the usual range of frequencies of interest (alpha, beta) in this type of paradigm with EEG. Atanh = tanh:', inverse hyperbolic tangent. The horizontal dotted line indicates minimum level for significant change.

COHERENCE, CORT/CO-CORT/CAL

spatially processed (e.g. Laplacian) data from multiple recording channels. The basis for coherence calculation is then the source waveform which corresponds to the (modelled) time course of oscillatory activity of the respective source (Thatcher, 1995; Gross et aI., 2001). These approaches are important extensions of the coherence method and are particularly useful, if intrasulcal activity is investigated. Electric or magnetic sources that are located inside a sulcus produce predominantly dipolar field distributions. For the primary sensorimotor region, this can result in two maxima, a frontal and a parietal one which are 180 deg out of phase but highly coherent. Source reconstructions can here be used to image the most likely origin of the signal. Naturally, a source model can only be as good as its stability and the plausibility of the assumptions that the model is based on. Spurious constraints, biases by a priori channel selections or selection of an unfortunate starting point (for a moving dipole fit), or solutions contaminated by poor signal-to-noise ratios will result in meaningless coherence data. Gross et al. (2001) have proposed a signal-to-noise ratio of 5-12 for their method of "dynamic imaging of coherent sources" (DICS). 6.2.5. Data acquisition: comments on recording technique

Beyond the usual requirements for high-quality EEG or MEG recordings (or recordings of firing rates or LFPs with invasive methods), a particular point of concern for coherence data is the number of single trials (epochs). The variance of the squareroot of coherence is solely determined by the n of epochs (see Eq. (3». There is no mathematically defined range for the number of trials necessary to produce reliable coherence values. However, in our experience one should obtain approximately n= 100 artifact-free, non-overlapping epochs, corresponding to an improvement of the signal-to-noise ratio of 10 (=Yn).

For the experimental design, it is important to decide the desired frequency resolution. For example, a standard task-related approach aimed at (hypothesis-free) I-Hz frequency resolution necessitates an epoch length of at least 1000 ms for each target condition.

83

6.3. Applications The analysis of task- and event-related coherence has advanced our knowledge of neurophysiological processes underlying the performance and learning of skilled finger movements (Rappelsberger et al., 1994; Andrew and Pfurtscheller, 1996; Leocani et al., 1997; Gerloff et al., 1998; Manganotti et al., 1998). It has become clear that various types of higher task demands are reflected by changes in the functional coupling of different cortical areas and not only by changes in regional activation (Gerloff et al., 1998; Manganotti et al., 1998). It has been demonstrated that the learning of bimanual coordination is associated with modulation of taskrelated coherence between the motor areas of both cerebral hemispheres (establishment of efficient bimanual 'motor routines') (Andres et al., 1999; Gerloff and Andres, 2002). In paradigms involving the functional interaction between visual and sensorimotor areas, corticocortical coherence has been demonstrated to be elevated specifically in situations in which signal processing in both regions needs to be integrated (Classen et al., 1998; Miltner et al., 1999; Chen et al., 2003). The amplitude of task-related coherence between visual and sensorimotor regions correlated positively with the behavioral efficacy in a visuotactile matching task (Hummel and Gerloff, 2001). In particular, the studies on visuomotor and visuotactile integration suggest that cortico-cortical coherence reflects meaningful and functionally relevant coupling of distant brain areas in large-scale networks. Cortico-cortical coherence has also been used in a number of neurocognitive paradigms, such as language processing (Weiss and Rappelsberger, 1996), associative learning (Miltner et al., 1999), processing of pain (Chen and Rappelsberger, 1996), mental rotation (Rappelsberger and Petsche, 1988), and object recognition (Neven and Aertsen, 1992; Mima et al., 200lb). Recently, the influence of the dopaminergic neurotransmitter system on task-related cortico-cortical coherence has been addressed in parkinsonian patients during a visuomotor tracking task (Cassidy and Brown, 2001). After levodopa the amplitude of broad-band interregional coherence was significantly increased (compared with off levodopa). This led the authors to the conclusion that ascending

84

dopaminergic projections from the ventral mesencephalon may be important in determining the pattern and extent of corticocortical coupling during executive tasks. The relative contributions of corticocortical connectivity as self-emerging property of the neocortex and of third pacemaker structures to cortico-cortical coherence are yet to be determined. Cortico-cortical coherence of the resting EEG (not task-related) has furthermore been used to assess cortical connectivity in several neurological (e.g. multiple sclerosis (Leocani and Comi, 1999)) and psychiatric conditions (e.g. dementia (Stevens et al., 2001».

References Amjad, AM, Halliday, DM, Rosenberg, JR and Conway, BA (1997) An extended difference of coherence test for comparing and combining several independent coherence estimates: theory and application to the study of motor units and physiological tremor. J. Neurosci. Meth., 73: 69-79. Andres, FG, Mirna, T, Schulman, AE, Dichgans, J, Hallett, M and Gerloff, C (1999) Functional coupling of human cortical sensorimotor areas during bimanual skill acquisition. Brain, 122: 855-870. Andrew, C and Pfurtscheller, G (1996) Event-related coherence as a tool for studying dynamic interaction of brain regions. Electroencephalogr. Clin. Neurophysiol., 98: 144-148. Baccala, LA and Sameshima, K (200 I) Partial directed coherence: a new concept in neural structure determination. Bioi. Cybern., 84: 463-474. Blair, RC and Karniski, W (1993) An alternative method for significance testing of waveform difference potentials. Psychophysiology, 30: 518-524. Bressler, SL, Coppola, R and Nakamura, R (1993) Episodic multiregional cortical coherence. Nature, 366: 153-156. Brockwell, PJ and Davis, RA (1991) Time series: theory and methods. New York: Springer Verlag. Bronstein, IN and Semendjajew KA (1987) In: G Grosche, V Ziegler and G Ziegler (Eds.), Taschenbuch der Mathematik. Frankfurt am Main: Verlag Harry Deusch. Carlesimo, GA, Casadio, P, Sabbadini, M and Caltagirone C (1998) Associative visual agnosia resulting from a disconnection between intact visual memory and semantic systems. Cortex, 34: 563-576. Cassidy, M and Brown, P (2001) Task-related EEG-EEG coherence depends on dopaminergic activity in Parkinson's disease. Neuroreport, 12: 703-707.

C. GERLOFF ET AL.

Chen, ACN and Rappelsberger, P (1996) Brain and human pain: topographic EEG amplitude and coherence mapping. Brain Topogr., 7: 129-140. Chen, Y,Ding, M and Kelso, JA (2003) Task-related power and coherence changes in neuromagnetic activity during visuomotor coordination. Exp. Brain Res., 148: 105-116. Classen, J, Gerloff, C, Honda, M and Hallett, M (1998) Integrative visuomotor behavior is associated with interregionally coherent oscillations in the human brain. J. Neurophysiol., 79: 1567-1573. Farmer, SF, Bremner, FD, Halliday, DM, Rosenberg, JR and Stephens, JA (1993) The frequency content of common synaptic inputs to motoneurons studied during voluntary isometric contraction in man. J. Physiol. (Lond.), 470: 127-155. Fein, GJR, Brown, F and Merrin, E (1988) Common reference coherence data are confounded by power and phase effects. Electroencephalogr. Clin. Neurophysiol., 69: 581-584. Gerloff, C and Andres, FG (2002) Bimanual coordination and interhemispheric interaction. Acta Psychol. (Amst.), 110: 161-186. Gerloff, C, Hadley, J, Richard, J, Uenishi, N, Honda, M and Hallett, M (1998) Functional coupling and regional activation of human cortical motor areas during simple, internally paced and externally paced finger movements. Brain, 121: 1513-1531. Geschwind, DH, Iacoboni, M, Mega, MS, Zaidel, DW, Cloughesy, T and Zaidel, E (1995) Alien hand syndrome: interhemispheric motor disconnection due to a lesion in the midbody of the corpus callosum. Neurology, 45: 802-808. Gevins, AS, Cutillo, BA, Morgan, NH, Bressler, SL, Illes, J and White, RM et al. (1987) Event-related covariances of a bimanual visuomotor task. Electroencephalogr. Clin. Neurophysiol., Suppl. 40: 31-40. Gevins, AS, Bressler, SL, Morgan, NH, Cutillo, BA, White, RM and Greer, DS et al. (1989) Event-related covariances during a bimanual visuomotor task. I. Methods and analysis of stimulus- and response-locked data. Electroencephalogr. Clin. Neurophysiol., 74: 58-75. Gross, J, Kujala, J, Hamalainen, M, Timmermann, L, Schnitzler, A and Salmelin, R (2001) Dynamic imaging of coherent sources: Studying neural interactions in the human brain. Proc. Natl. Acad. Sci. USA, 98: 694-699. Halliday, DM, Rosenberg, JR, Amjad, AM, Breeze, P, Conway, BA and Farmer, SF (1995) A framework for the analysis of mixed time series/point process data theory and application to the study of physiological tremor, single motor unit discharges and electromyograms. Prog. Biophys. Mol. Biol., 64: 237-278.

COHERENCE, CORTICO-CORTICAL

Hummel, F and Gerloff, C (200 I) Better swing together. Soc. Neurosci. Abstr., 456(1). Kaminski, M, Ding, M, Truccolo, WA and Bressler, SL (2001) Evaluating causal relations in neural systems: Granger causality, directed transfer function and statistical assessment of significance. Biol. Cybern., 85: 145-157. Leocani, Land Comi, G (1999) EEG coherence in pathological conditions. J. Clin. Neurophysiol., 16: 548-555. Leocani, L, Toro, C, Manganotti, P, Zhuang, P and Hallett, M (1997) Event-related coherence and event-related desynchronization/synchronization in the 10Hz and 20 Hz EEG during self-paced movements. Electroencephalogr. Clin. Neurophysiol., 104: 199-206. Lutzenberger, W, Ripper, B, Busse, L, Birbaumer, N and Kaiser, J (2002) Dynamics of gamma-band activity during an audiospatial working memory task in humans. J. Neurosci., 22: 5630-5638. Manganotti, P, Gerloff, C, Toro, C, Katsuta, H, Sadato, N and Zhuang, P et al. (1998) Task-related coherence and task-related spectral power changes during sequential finger movements. Electroencephalogr. Clin. Neurophysiol., 109: 50-62. Marangolo, P, De Renzi, E, Di Pace, E, Ciurli, P and Castriota-Skandenberg, A. Let not thy left hand know what thy right hand knoweth. The case of a patient with an infarct involving the callosal pathways. Brain, 121: 1459-1467. Miltner, WH, Braun, C. Arnold, M, Witte, Hand Taub, E (1999) Coherence of gamma-band EEG activity as a basis for associative learning. Nature, 397: 434-436. Mirna, T, Matsuoka, T and Hallett, M (2000a) Functional coupling of human right and left cortical motor areas demonstrated with partial coherence analysis. Neurosci. Lett., 287: 93-96. Mirna, T, Steger, J, Schulman, AE, Gerloff, C and Hallett, M (2000b) Electroencephalographic measurement of motor cortex control of muscle activity in humans. Clin. Neurophysiol., Ill: 326-337. Mirna, T, Matsuoka, T and Hallett, M (2001a) Information flow from the sensorimotor cortex to muscle in humans. Clin. Neurophysiol., 112: 122-126. Mirna, T, Oluwatimilehin, T, Hiraoka, T and Hallett, M (200 Ib) Transient interhemispheric neuronal synchrony correlates with object recognition. J. Neurosci., 21: 3942-3948. Neven, Hand Aertsen, A (1992) Rate coherence and event coherence in the visual cortex: a neuronal model of

85 object recognition. Biol. Cybern., 67: 309-322. Papoulis, A (1984) Probability, random variables and stochastic processes. New York: McGraw-Hill. Rappelsberger, P and Petsche, H (1988) Probability mapping: power and coherence analyses of cognitive processes. Brain Topogr., I: 46-54. Rappelsberger, P, Pfurtscheller, G and Filz, 0 (1994) Calculation of event-related coherence - a new method to study short-lasting coupling between brain areas. Brain Topogr., 7: 121-127. Rodriguez, E, George, N, Lachaux, JP, Martinerie, J, Renault, B and Varela, FJ (1999) Perception's shadow: long-distance synchronization of human brain activity. Nature, 397: 430-433. Rosenberg, JR, Amjad, AM, Breeze, P, Brillinger, DR and Halliday, DM (1989) The Fourier approach to the identification of functional coupling between neuronal spike trains: Maximum likelihood analysis of spike trains of interacting nerve cells. Prog. Biophys. Mol. Biol., 53: 1-31. Singer, W (1993) Neuronal representations, assemblies and temporal coherence. Prog. Brain Res., 95: 461474. Singer, W (1994) Coherence as an organizing principle of cortical functions. Int. Rev. Neurobiol., 37: 153-183. Singer, Wand Gray, CM (1995) Visual feature integration and the temporal correlation hypothesis. Annu. Rev. Neurosci., 18: 555-586. Stevens, A, Kircher, T, Nickola, M, Bartels, M, Rosellen, Nand Wormstall, H (200 I) Dynamic regulation of EEG power and coherence is lost early and globally in probable DAT. Eur. Arch. Psychiatry Clin. Neurosci., 251: 199-204. Tanaka, Y, Yoshida, A, Kawahata, N, Hashimoto, Rand Obayashi, T (1996) Diagonistic dyspraxia. Clinical characteristics, responsible lesion and possible underlying mechanism. Brain, 119: 859-873. Thatcher, RW (1995) Tomographic electroencephalography/magnetoencephalography. Dynamics of human neural network switching. J. Neuroimaging, 5: 35-45. Wang, G and Takigawa, M (1992) Directed coherence as a measure of interhemispheric correlation of EEG. Int. J. Psychophysiol., 13: 119-128. Weiss, Sand Rappelsberger, P (1996) EEG coherence within the 13-18 Hz band as a correlate of a distinct lexical organisation of concrete and abstract nouns in humans. Neurosci. Lett., 209: 17-20.

Movement Disorders Handbook of Clinical Neurophysiology, Vol. 1 M. Hallett (Ed.) © 2003 Elsevier B.v. All rights reserved

87 CHAPTER 7

Coherence, cortico-muscular Tatsuya Mirna":", Mark Hallettb and Hiroshi Shibasaki" a

Department of Brain Pathophysiology, Human Brain Research Center, Kyoto University Graduate School of Medicine, Shogoin, Sakyo-ku, Kyoto 606-8507, Japan b HMCS, NINDS, National Institutes of Health, Bethesda, MD 20892, USA

7.1. Introduction

understanding the cortical control of muscle activity.

One of the fundamental questions in motor physiology is the relationship between cortical activity and the motor output. In some pathological conditions, the electroencephalographic (EEG) correlate of electromyographic (EMG) activity can be recognized in the conventional EEG-EMG polygraph. For example, an EMG jerk can be preceded by a positive spike at the contralateral central area in some patients with cortical myoclonus. Even a small EEG activity can be detected by applying an averaging technique with respect to the EMG onset (jerk-locked back averaging). For voluntary ballistic movements, the same back averaging technique can reveal a cortical slow negative potential preceding the movement (movement-related cortical potentials, Bereichtshaftspotential) (Komhuber and Deecke, 1965; Shibasaki et aI., 1980). In addition to this type of time-domain analysis for the EEG-EMG correlation, there have been considerable interests in cortico-muscular coherence, starting with the report using magnetoencephalography (MEG) (Conway et aI., 1995). Coherence is also a measure of correlation but in the frequency domain. Coherence computation enables us to compute the normalized magnitude of correlation as a function of frequency (coherence spectra). Since the oscillatory nature of EEG/MEG and EMG signals has been well-recognized for many years, this approach might provide us a new useful tool for

* Correspondence to: Dr. T. Mirna, Department of Brain Pathophysiology, Human Brain Research Center, Kyoto University Graduate School of Medicine, Shogoin, Sakyoku, Kyoto 606-8507, Japan. E-mail address:[email protected] Fax: +81-75-751-3202.

At first, we will discuss the computational and statistical evaluation of EEG-EMG coherence. Secondly, we will overview the normal findings and the possible generator mechanisms of corticomuscular coherence. Thirdly, clinical applications of this method will be introduced. Finally, recent developments in computational technique will be reviewed. 7.2. Methodology

7.2.1. Coherence estimation Coherence is a measure of linear correlation in the frequency domain and defined as cross-spectra divided and normalized by auto-spectra. Some researchers prefer to use the square-root of coherence, which is termed as coherency. Mathematical formula of coherence is as follows.

In this equation, Ix/i), J;;ii) and jxy(i) are values of auto- and cross-spectra at a specific frequency i. Since the coherence is a normalized value, it distributes from 0 to 1, with 1 indicating the perfect linear correlation and 0 indicating the lack of linear association. To perform a spectral analysis of the signal, a fast Fourier transform method (FFT) is commonly used. To test whether the observed coherence value is statistically significant or not, the 95% confidence value based on the assumption of linearly independent signals is usually used. Confidence limit at X%

88

T. MIMA ET AL.

Table I Confidence level of coherence as a function of sample numbers. Number of samples 95% confidence 99% confidence 0.146 0.059 0.030 0.015 0.010 0.006

20 50 100 200 300 500

0.215 0.090 0.045 0.023 0.Q15 0.009

can be approximated by the equation as follows (Rosenberg et aI., 1989):

CL(X)=I-

(

)lI(n-IJ

X I--

lOa

This is relatively a simple equation, and it should be noted that the number of independent samples (n) influences the confidence limit (Table I). Although statistically significant, the actual magnitude of cortico-muscular coherence is not very large and is usually smaller than 0.3. To compare the magnitude of coherence, the archyperbolic tangent transformation of the square root of coherence is used to normalize the data, which stabilizes the variance as a function of the number of independent samples (Rosenberg and Amjad, 1989): 1 var(tanh-I(Rxy(i)))=-

2n

Therefore, if the number of samples for the computation of coherence is approximately the same, a general linear statistical approach, such as t-test or analysis of variance can be applied to the normalized coherence. 7.2.2. Phase analysis

Since the cross-spectra are complex numbers, their argument gives us the phase information between two signals at a given frequency:

PX),(j) =arg(f.i i» Although the phase reflects the temporal relationship between two signals, it is impossible to determine whether the phase lag reflects the delay or advance

in the time domain. However, if multiple frequencies have a constant or frequency-independent time lag, a regression analysis can reveal the temporal relationship between the two signals ("constant phase shift and constant time lag model") (Mima et aI., 2000):

360 Px/i) = 1000 i.i:« where t is the frequency-independent time lag in milliseconds and K is the constant phase shift. In the case of cortico-muscular coherence, the EEG leads the EMG with a short time delay at the frequency around 20 Hz (Mima et aI., 2000). 7.2.3. Recording technique

To extract the timing information of grouped motor unit activity from the surface EMG, the EMG signal is full-wave rectified and used for further analysis. It seems that the surface EMG is better than the needle EMG for the purpose of coherence estimation (unpublished observation). Since the polarity of the recorded motor unit activity can vary depending on the location of the active electrode, it is necessary to rectify the EMG signal before coherence computation. Spectral analysis of the rectified EMG typically shows 10, 20 and 40 Hz peaks, which might partly reflect the centrally generated frequencies (McAuley et aI., 1997). If the EEG is used for recording the cortical activity, it should be noted that the magnitude and distribution of cortico-muscular coherence heavily depends on the EEG derivation type (Mima and Hallett, 1999). The best results in terms of coherence value and spatial resolution can be achieved by the current source density (CSD) estimation, including Hjorth transformation. If the earlobe electrode or common average reference is used, the measured EEG-EMG coherence is smaller compared to CSD method. Moreover, possibly artifactual EEG-EMG coherence over the medial frontal area can be observed in these EEG derivations because this coherence can be abolished by CSD computation. However, a contribution of the medial frontal area for the generation of cortico-cortical coherence has been demonstrated with subdural recording in patients with intractable seizures (Ohara et aI., 2000). Bipolar EEG might distort the phase spectra because the phase can show the pi-phase delay (polarity inversion) depending on the relative loca-

89

COHERENCE, CORTICO-MUSCULAR

tion of the active and reference electrode with respect to the coherent EEG, which might produce frequency-independent constant phase spectra (Mirna and Hallett, 1999).

tion time (Salenius et aI., 1997; Mirna et aI., 2000). It is strongly suggested that the coherence reflects the motor command from the cortex to the muscle. alphabeta

7.3. Normal findings and possible generator mechanisms Cortico-muscular coherence measured by EEG, electrocorticogram (ECoG) or MEG shows significant value over the primary sensorimotor area during a weak tonic contraction task mainly at 15-35 Hz (Conway et aI., 1995; Salenius et aI., 1996, 1997; Baker et aI., 1997; Mirna et aI., 1998; Marsden et aI., 2000; Mirna et aI., 2000). Using a whole-head MEG system coregistered with brain MRI, the generator source of the EMG-correlated cortical activity can be localized at the anterior bank of the central sulcus, that is the primary motor cortex (Salenius et aI., 1997). The spatial distribution of corticomuscular coherence over the scalp demonstrates a somatotopic pattern; the hand and arm over the contralateral lateral convexity and the foot over the midline (Salenius et aI., 1997; Brown et al., 1998; Mirna et aI., 2000). In addition to the primary sensorimotor area, subdural EEG recording in epileptic patients revealed involvement of contralateral premotor cortex, supplementary motor areas and cingulate cortex (Ohara et aI., 2000). It is possible that the coherence between the activities at those motor-related areas and muscle is associated with direct cortical projections to the spinal motoneuron (Dum and Strick, 1991; Rizzolatti et al., 1998). Not only the spatial localization but also the spectral structure can be studied by using coherence analysis. During a weak tonic contraction, corticomuscular coherence demonstrates the highest peak usually within the beta band (13-30 Hz) (Conway et aI., 1995; Baker et al., 1997; Salenius et aI., 1997; Mirna et aI., 2000). However, some normal subjects show the significant coherence in a lower frequency band (3-13 Hz) (Mirna et al., 1999; Mirna et aI., 2000; Marsden et al., 200 I). Phase analysis revealed that the generator mechanism of coherence in the beta band might be different from that in the lower frequency band. For the beta band, both phase analysis and cross-correlogram showed that the EEG leads the EMG by a short time lag which is similar or slightly shorter than the cortico-muscular conduc-

gamma

0.11------------'

b

\

100

~

2:

\....f\....

j10

C

EMG

-

......... ...........~ . , -'

~

0.3

e

~ 0.2

EEG-EMG coherence

~

o 0.1

d

o

95% confidence u.;;;;-.c.::.:':":"::':":":~ioOiQ,i!:s::..;;;;"""a;&J

180r------------,

1120 ~ 60

~

/

:l:Or-------t'-----l

iii

,/

III -60 as '&. -120

rJ.,.-.r! phase

-18id.O~~~~~~~~~

5 101520253035404550 Frequency(Hz)

Fig. 1. (a) EEG power spectra at the left central area, showing a peak at 9 Hz. (b) EMG power spectra of a right hand muscle (abductor poIIicis brevis), showing a peak at 20 Hz. (c) EEG-EMG coherence spectra computed from the data shown in Figs. la and lb (n=280). Significant coherence can be seen at 8-33 Hz. 95% confidence limit is shown as the dotted line. (d) EEG-EMG phase spectra. Phase spectra is linear >20 Hz (modified from Mirna et aL,2000).

90

T. MIMA ET AL.

Fig. 3. EEG-EMG coherence as a function of produced force. Coherence between the EMG at the APB muscle and the EEG at C3 (thick line) and that at FC3 (thin line). During the very strong contraction (80% of the maximal force), gamma band coherence can be seen at FC3. In contrast, EEG-EMG coherence within the beta range was almost invariant for the force change from 10 to 60% of the maximal contraction (modified from Mima et a!., 1999).

Fig. 2. Scalp topography of the significant EEG-EMG coherence for the hand (abductor pollicis brevis) and foot (abductor hallucis) muscles. Topography was constructed at the peak frequency of the EEG-EMG coherence.

In contrast, cortico-muscular coherence within the lower frequency band shows the near-zero time lag (Mirna et aI., 2000). Thus, it is possible that this coherence might be associated with a subcortical rhythm generator which affects both the cortical and muscle oscillatory activities. A possible association between physiologic tremor and this coherence has been suggested (Marsden et aI., 2001). It is unlikely, however, that this coherence has relevance for the generation of physiologic tremor because the coherence in this lower frequency band is generally smaller than that in the beta band even in normal subjects with enhanced physiologic tremor (Mirna et aI., in press). Cortico-muscular coherence in the gamma band can be seen during a very strong contraction task

(Brown et aI., 1998; Mirna et aI., 1999). It has been reported that this coherence is associated with the motor unit grouping at around 40 Hz (Piper rhythm). Even when the subjects performed a gradual increase in the force level, there is no gradual shift in the coherence from the beta to the gamma band (Brown et aI., 1998; Mirna et aI., 1999). Brown and colleagues showed that the peak coherence frequency in the gamma band for the arm and foot muscle are different, which suggests different neural circuits for the generation of coherence at different somatotopic cortical areas (Brown et aI., 1998). . These findings might indicate that the corticomuscular coherence in different frequency bands reflects the different coupling mechanisms between cortex and muscle.

7.4. Factors that will affect the cortico-muscular coherence For physiological experiments or clinical application of cortico-muscular coherence, one should

COHERENCE, CORTICO-MUSCULAR

91

EEG·EMG coherence in Parkinson tremor 0.16 . , - - - - - - - - - - - - -

consider the following factors that have considerable effects on the coherence estimate: (1) Produced force level As we have discussed earlier, the corticomuscular coherence during very strong contraction (more that 60% of maximal strength) might show a peak frequency at around 40 Hz. However, for weak to moderate isometric tonic contractions, the coherence in the beta band is generally insensitive to the produced force (Brown et al., 1998; Mima et al., 1999). In a precision grip task against a spring-like load, there is a positive relationship between the coherence magnitude and the level of object compliance (Kilner et al., 1999). (2) Phasic vs. tonic movements Significant cortico-muscular coherence can be observed during continuous tonic contraction tasks or the tonic hold phase of precision grip tasks, but is absent or very small during phasic movements (Baker et al., 1997; Kilner et al., 1999). (3) EEG derivation type As we have mentioned earlier, reliable EEGEMG coherence can be computed by the current source density or its approximation (Hjorth transformation). If bipolar or ear-lobe reference EEG is used, EEG-EMG coherence might show a misleading spatial distribution and phase estimate as well as an apparently small value (Mirna and Hallett, 1999).

7.5. Cortico-muscular coherence in pathological conditions Since cortico-muscular coherence is a noninvasive measure of cortico-motoneuronal coupling as a function of frequency, this method can be applied clinically to various motor dysfunctions or movement disorders. One possible application is to investigate the cortico-spinal function in motor deficits. For example, in patients with subcortical stroke who manifested pure motor paresis and recovered well, this coherence for the affected side was significantly smaller than that for the healthy side (Mima et al., 2002). It is suggested that the clinical recovery in these patients might not be associated with the direct

~

~

0.1

Q)

:3 \A

10

A }95% confidence L--'

20

30

40

50

Frequency (Hz) Fig. 4. EEG-EMG coherence in a Parkinson patient during a rest condition. EMG was recorded from the left extensor carpi radialis muscle. Significant coherence is observed at the resting tremor frequency (6 Hz). This observation may indicate the involvement of the primary sensorimotor cortex for the generation of tremor.

cortico-motoneuronal pathway but with an indirect one. Another challenging field of clinical application is movement disorder research. Existence of corticomuscular coherence might suggest the involvement of the primary sensorimotor area for the generation of involuntary movement. Volkmann and colleagues showed cortico-muscular coherence at the resting tremor frequency over the contralateral premotor and primary sensorimotor area in Parkinsonian patients (Volkmann et al., 1996). For essential tremor, there are conflicting results on the existence of corticomuscular coherence at tremor frequency (Halliday et al., 2000; Hellwig et al., 2000, 2001). The negative results might well be false negatives due to low amplitude tremors. These studies of tremor-related coherence indicate the possible important role of the cortex or cortico-subcortical loop for the generation of tremors. Recent study of the local field potential oscillation recorded from the nucleus ventralis intermedius of the thalamus also supports this idea (Marsden et al., 2000). Cortico-muscular coherence can be applicable for other types of movement disorders, such as cortical myoclonus and dystonia (Brown et al., 1999; Tijssen et al., 2000). 7.6. Newly developed computational techniques To measure the correlation between cortical oscillatory activity and EMG quantitatively, several

T. MIMA ET AL.

92

0.2

Coherence

r------......----~--__,

0.1

10

20

30

40

50

methods to measure phase coherence that focuses on the phase but not on the amplitude information (Tass et aI., 1998; Feige et aI., 2000; Gross et aI., 2000, 2001). Another approach is the use of directed coherence or directed transfer function (Mirna et al., 2001). Conventional coherence is a measure of correlation and does not account for the temporal relationship between two signals. By using a multivariate autoregression model for the analysis of cortical and muscle activities, it is possible to detect the normalized value of information flow from cortex to muscle and from muscle to cortex separately.

Frequency (Hz) 7.7. Conclusion

0.2

Directed transfer function

Functional coupling between the cortical oscillatory activity and EMG can be quantitatively measured by cortico-muscular coherence, using EEG, MEG or electrocorticogram. This method can be a useful tool for understanding the pathophysiology of motor deficits and movement disorders as well as normal motor control.

From EEG to EMG 0.1 From EMG to EEC\••

•• •

••

••••••••

••••••••• 0 ......._ ........_ _.........._ _""-10

20

30

"___....:...:1

40

References

50

Frequency (Hz) Fig. 5. Coherence spectra and directed transfer function computed by the multivariate autoregressive model (n = 200). In this subject, DTF from the EEG to EMG (thick line) was larger than that from EMG to EEG within 0-30 and 40-50 Hz. This finding suggests an asymmetric information flow from the cortex to the muscle in these frequencies.

new methods other than conventional coherence have been introduced. One possible way is to measure the phase coherence. Since conventional coherence is a normalized cross-spectra, it reflects both the amplitude covariation and the phase consistency. In the case of functional coupling between cortex and muscle, the latter factor might be more physiologically important for the cortical control of muscle activity. Several groups have introduced

Baker, SN, Olivier, E and Lemon, RN (1997) Coherent oscillations in monkey motor cortex and hand muscle EMG show task-dependent modulation. J. Physiol., SOl: 225-241. Brown, P, Salenius, S, Rothwell, JC and Hari, R (1998) Cortical correlate of the Piper rhythm in humans. J. Neurophysiol., 80: 2911-2917. Brown, P, Farmer, SF, Halliday, DM, Marsden, J and Rosenberg, JR (1999) Coherent cortical and muscle discharge in cortical myoclonus. Brain, 122: 461-472. Conway, BA, Halliday, DM, Farmer, SF, Shahani, D, Maas, P, Weir, AI and Rosenberg, JR (1995) Synchronization between motor cortex and spinal motoneuronal pool during the performance of a maintained motor task in man. J. Physiol., 489: 917-924. Dum, RP and Strick, PL (1991) The origin of corticospinal projections from the premotor areas in the frontal lobe. J. Neurosci., 11: 667-689. Feige, B, Aertsen, AD and Kristeva-Feige, R (2000) Dynamic synchronization between multiple cortical motor areas and muscle activity in phasic voluntary movements. J. Neurophysiol., 84: 2622-2629.

COHERENCE, CORTICO-MUSCULAR

Gross, J, Tass, PA, Salenius, S, Hari, R, Freund, H-J and Schnitzler, A (2000) Cortico-muscular synchronization during isometric muscle contraction in humans as revealed by magnetoencephalography. J. Physiol., 527: 623-631. Gross, J, Kujala, J, Harnalainen, M, Timmermann, L, Schnitzler, A and Salmelin, R (200 I) Dynamic imaging of coherent sources: Studying neural interactions in the human brain. Proc. Nat. Acad. Sci. USA, 98: 694-699. Halliday, DM, Conway, BA, Farmer, SF, Shahani, U, Russell, AJC and Rosenberg, JR (2000) Coherence between low-frequency activation of the motor cortex and tremor in patients with essential tremor. Lancet, 355: 1149-1153. Hellwig, B, HauBler, S, Lauk, M, Guschlbauer, B, Koster, B, Kristeva-Feige, R, Timmer, J and Liicking, CH (2000) Tremor-correlated cortical activity detected by electroencephalography. Clin. Neurophysiol., 111: 806-809. Hellwig, B, Haufsler, S, Schelter, B, Lauk, M, Guschlbauer, B, Timmer, J and Liicking, CH (2001) Tremor-related cortical activity in essential tremor. Lancet, 357: 519-523. Kilner, lM, Baker, SN, Salenius, S, Jousmaki, V, Hari, R and Lemon, RN (1999) Task-dependent modulation of 15-30 Hz coherence between rectified EMGs from human hand and forearm muscles. J. Physiol., 516: 559-570. Kornhuber, HH and Deecke, L (1965) Hirnpotentialanderungen bei Willkiirbewegungen und passiven Bewegungen des Menschen: Bereitschaftspotential und reafferente Potentiale. Pfliigers. Arch. Ges. Physiol., 284: 1-17. Marsden, JF, Ashby, P, Limousin-Dowsey, P, Rothwell, JC and Brown, P (2000a) Coherence between cerebellar thalamus, cortex and muscle in man: cerebellar thalamus interactions. Brain, 123: 1459-1470. Marsden, JF, Werhahn, KJ, Ashby, P, Rothwell, JC, Noachtar, S and Bown, P (2000b) Organization of cortical activities related to movement in humans. J. Neurosci., 20: 2307-2314. Marsden, JF, Brown, P and Salenius, S (200 I) Involvement of the sensorimotor cortex in physiological force and action tremor. Neuroreport, 12: 1937-1941. McAuley, JH, Rothwell, JC and Marsden, CD (1997) Frequency peaks of tremor, muscle vibration and electromyographic activity at 10 Hz, 20 Hz and 40 Hz during human finger muscle contraction may reflect rhythmicities of central neural firing. Exp. Brain Res., 114: 525-541. Mirna, T and Hallett, M (l999a) Cortico-muscular coherence: a review. J. CUn. Neurophysiol., 16: 501-511. Mirna, T and Hallett, M (1999b) Electroencephalographic analysis of cortical-muscular coherence: reference

93 effect, volume conduction and generator mechanism. Clin. Neurophysiol., 110: 1892-1899. Mirna, T, Gerloff, C, Steger, J and Hallett, M (1998) Frequency-coding of motor control system - coherence and phase estimation between cortical rhythm and motoneuronal firing in humans. Soc. Neurosci. Abstr., 24: 1768. Mirna, T, Simpkins, N, Oluwatimilehin, T and Hallett, M (1999) Force level modulates human cortical oscillatory activities. Neurosci. Lett., 275: 77-80. Mirna, T, Steger, J, Schulman, AE, Gerloff, C and Hallett, M (2000) Electroencephalographic measurement of motor cortex control of motoneuronal firing in humans. CUn. Neurophysiol., 111: 326-337. Mirna, T, Matsuoka, T and Hallett, M (200 I) Information flow from the sensorimotor cortex to muscle in humans. CUn. Neurophysiol., 112: 122-126. Mirna, T, Toma, K, Koshy, B and Hallett, M (2002) Coherence between cortical and muscular activities following subcortical stroke. Stroke, 32: 2597-2601. Mirna, T, Ohara, Sand Nagamine, T (in press) Corticalmuscular coupling. In: H Shibasaki (Ed.), Inter-areal Functional Coupling of the Human Brain. Elesevier Co. Ohara, S, Nagamine, T, Ikeda, A, Kunieda, T, Matsumoto, R, Taki, W, Hashimoto, N, Baba, K, Mihara, T, Salenius, Sand Shibasaki, H (2000) Electrocorticogram-electromyogram coherence during isometric contraction of hand muscle in human. CUn. Neurophysiol., Ill: 2014-2024. Rizzolatti, G, Luppino, G and Matelli, M (1998) The organization of the cortical motor system: new conElectroencephalogr. CUn. cepts Review. Neurophysiol., 106: 283-296. Rosenberg, JR, Amjad, AM, Breeze, P, Brillinger, DR and Halliday, DM (1989) The Fourier approach to the identification of functional coupling between neuronal spike trains. Prog. Biophys. Molec. Biol., 53: 1-31. Salenius, S, Salmerin, R, Neuper, C, Pfurtscheller, G and Hari, R (1996) Human cortical40-Hz rhythm is closely related to EMG rhythmicity. Neurosci. Lett., 213: 75-77. Salenius, S, Portin, K, Kajola, M, Salmelin, Rand Hari, R (1997) Cortical control of human motoneuron firing during isometric contraction. J. Neurophysiol., 77: 3401-3405. Shibasaki, H, Barrett, G, Halliday, E and Halliday, AM (1980) Components of the movement-related potentials and their scalp topography. Electroencephalogr. Clin. Neurophysiol., 1980: 213-226. Tass, P, Rosenblum, MG, Weule, J, Kurths, J, Pikovsky, A, Volkmann, J, Schnitzler, A and Freund, H-J (1998) Detection of n: m phase locking from noisy data:

94 Application to magnetoencephalography. Physic. Rev. Lett., 81: 3291-3294. Tijssen, MA, Marsden, JF and Brown, P (2000) Frequency analysis of EMG activity in patients with idiopathic torticollis. Brain, 123: 677-686.

T. MIMA ET AL.

Volkmann, J, Joliot, M, Mogilner, A, Ioannides, AA, Lado, F, Fazzini, E, Ribary, U and Llinas, R (1996) Central motor loop oscillations in parkinsonian tremor revealed by magnetoencephalography. Neurology, 46: 1359-1370.

Movement Disorders Handbook of Clinical Neurophysiology, Vol. I M. Hallett (Ed.) © 2003 Elsevier B.V. All rights reserved

95 CHAPTER 8

Transcranial magnetic stimulation Vif Ziemann* Clinic of Neurology, Johann Wolfgang Goethe University of Frankfurt, Theodor-Stem-Kai 7. D-60590 Frankfurt am Main, Germany

This chapter surveys the techniques and principles of transcranial magnetic stimulation (TMS) to study the corticospinal tract, motor cortex excitability and motor cortex connectivity. Clinical applications are usually only hinted at but will be dealt with in the disorders section of the book.

8.1. Technical principles: modes of corticospinal tract activation TMS was introduced in its contemporary form in 1985 (Barker et aI., 1985a, b; Barker, 1999). It was originally designed to assess non-invasively the human corticospinal tract and to do this with much less discomfort compared to transcranial electrical stimulation (Merton and Morton, 1980). A magnetic stimulator works by discharging a capacitance through a copper-wire stimulating coil. A large current (current peak -5000 A) flows and induces a transient magnetic field (rise time 50-200 us, maximum field strength 1-2.5 Tesla) about the coil. Because the magnetic field changes rapidly, it can induce a (much smaller) current in other conductors, such as the human brain. In contrast to electrical currents, the magnetic field passes through electrical insulators, like the scalp and skull, without significant attenuation. The depth of penetration of the magnetic field is limited to a few ems because the magnetic field strength decays rapidly as a quadratic function of the distance from the coil (Epstein et aI., 1990; Roth et aI., 1991). Magnetic pulse characteristics are either near-monophasic, biphasic or polyphasic (damped cosine). These different waveforms may have complex effects on the exact way how neural tissue is activated (Maccabee et aI., 1998; Corthout et aI., 2001; Kammer et aI., 2001). Various geometries of stimulating coils are available

* E-mail address:[email protected] Tel.: +49696301 5739; fax: +49696301 6842.

for non-focal (circular, cap-shaped) and focal (eightshaped, double-cone) TMS (Roth et aI., 1991; Kraus et aI., 1993). How does TMS excite the corticospinal tract? A single high-intensity TMS pulse produces multiple descending corticospinal discharges (Fig. 1). The first wave is called D-wave, later waves are called I-waves (Patton and Amassian, 1954; Arnassian et aI., 1987). In humans, the D-wave very likely reflects direct excitation of the cortico-spinal axon at the initial axon segment (Day et aI., 1989). In contrast, I-waves reflect indirect (i.e. transsynaptic) excitation of corticospinal cells through excitation of tangentially oriented cortico-cortical or thaIamo-cortical axons in the deep cortical layers or in the superficial white matter at a depth of 1.5-2 em beneath the scalp surface (Day et aI., 1989; Epstein et aI., 1990; Rudiak and Marg, 1994). The I-waves follow the Dwave at regular intervals of about 1.5 ms. All waves are rapidly conducted at 60-70 rnIs (Arnassian et aI., 1987; Di Lazzaro et aI., 1999b). The first wave recruited depends on the direction of the current induced in the brain. With posterior-to-anterior direction, the lowest threshold wave usually is the 1,wave (Fig. 1), with lateral-to-medial direction the D-wave and with anterior-to-posterior direction the 13-wave (Werhahn et aI., 1994; Kaneko et aI., 1996a; Nakamura et aI., 1996; Sakai et aI., 1997; Di Lazzaro et aI., 1998a) (Fig. 1). Increase in stimulus intensity leads to the recruitment of other waves. The descending corticospinal volley is transmitted mainly or exclusively monosynaptically to the motoneurons in the spinal cord (Zidar et aI., 1987; Mills, 1991; Maertens de Noordhout et aI., 1999). Some evidence was given for the additional activation of a disynaptic pathway to forearm muscles via propriospinal-like intemeurons in the cervical spinal cord (Pierrot-Deseilligny, 1996), or for the activation of cortico-reticulospinal projections (Ziemann et aI., 1999). In adults, the motor response occurs usually

V.ZIEMANN

96

exclusively in muscles contralateral to the stimulated motor cortex. One exception are facial and axial PO

P1 P2 P3

I

AnodelSS %

Clockwise magnetic 35%

,

Antlclockwise magnetic35%

, 40

Fig. 1. Post-stimulus time histograms from a single motor unit in the first dorsal interosseous muscle studied using anodal TES (top), clockwise TMS (round coil, current in the coil view from above, middle) and anticlockwise TMS (bottom). Each histogram was constructed from the responses to 100 stimuli given 10 ms before the start of each record. Stimulus intensity is given as a percentage of the maximum output of each stimulator. The histogram peaks have been labeled PO, PI, P2 and P3 (continuous vertical lines) according to their latency after the stimulus. The discrete sub-peaks are called D-wave (PO), and I" 12 , Irwaves (PI, P2, P3). The single motor unit data suggest that all forms of cortical stimulation produce a complex corticospinal discharge with multiple peaks, but that anodal TES preferentially produces a D-wave, while TMS preferentially produces an I-wave, either an I,-wave if the induced current in the brain is directed from posterior to anterior, or an I}-wave if the induced current is directed from anterior to posterior (with kind permission, modified from Day et al., 1989).

muscles which often receive corticospinal projections from either hemisphere (Benecke and Meyer, 1991; Ferbert et al., 1992a; Hamdy et al., 1996). Another exception are ipsilateral motor responses in limb muscles which, however, are only obtainable during muscle contraction and usually much smaller and delayed by some 5-10 ms when compared to size-matched contralateral responses (Wassermann et al., 1994; Ziemann et al., 1999; Eyre et al., 2001). In summary, for neurophysiological routine purposes it can be taken that TMS assesses the motor system via trans synaptic activation of fast-conducting cortico-spinal tract fibers that project monosynaptically on alpha-motoneurons in the contralateral spinal cord. Finally, many studies used transcranial electrical stimulation (TES) in addition to TMS to sort out whether a physiological process occurs at the cortical or at a subcortical level (see below). This opportunity relies on the fact that low-intensity anodal TES activates the corticospinal tract directly (Fig. 1), probably close to the axon initial segment, and therefore does not detect changes that occur at the level of the corticospinal neuron or further upstream (Day et al., 1989; Rothwell et al., 1994; Di Lazzaro et al., 1998a).

8.2. Assessment of the cortico-spinal tract A recipe for how to perform single steps of TMS in clinical practice, including safety issues, exact placement of EMG electrodes and stimulating coil etc. is given in the IFCN recommendations for the practice of clinical neurophysiology (Rossini et al., 1999) and will not be repeated here. The following sections are devoted to details of how the integrity of the corticospinal tract can be assessed by central motor conduction time and the amplitude of the motor evoked potential.

8.2.1. Central motor conduction time 8.2.1.1. Techniques and principles The conduction time from motor cortex to alphamotoneurons in the spinal cord is referred to as the central motor conduction time (CMCT). Several methods have been proposed for measuring CMCT. All of them compute the difference between the conduction time from cortex to muscle (cortico-

TRANSCRANIAL MAGNETIC STIMULATION

muscular latency, CML) and the peripheral motor conduction time (PMCT). The CML equals the onset latency of the motor evoked potential (MEP). The PMCT can be measured by five different methods: F-wave latency (Rossini et aI., 1987; Robinson et aI., 1988), tendon reflex latency (Ofuji et aI., 1998), electrical needle stimulation of the spinal nerve roots (Evans et aI., 1990), transcutaneous electrical stimulation of the spinal nerve roots (Merton et aI., 1982; Rossini et al., 1985; Schmid et aI., 1991), and transcutaneous magnetic stimulation of the spinal nerve roots (Epstein et aI., 1991). For the F-wave method, PMCT is calculated by PMCT=(F+M -1)/2 where F is the shortest of (usually) 20 F-wave latencies, M is the M-wave latency and 1 (in ms) is the estimated delay time for antidromic activation of the spinal alpha-motoneuron. For the tendon reflex method, PMCT is calculated by PMCT=(T-1)12 where T is the tendon reflex latency and 1 (in ms) is the delay time at the spinal alpha-motoneuron. For all other methods, PMCT equals the onset latency of the compound muscle action potential (CMAP) evoked by stimulation of the spinal nerve roots. Spinal nerve root stimulation implies that the CMCT includes the proximal segment of the spinal motor nerve because excitation takes place not at the spinal alpha-motoneuron directly but in the region of the intervertebral foramen (Epstein et aI., 1991; Maccabee et al., 1991). Therefore, the CMCT obtained by the F-wave and tendon reflex methods is shorter by 0.5-1.4 ms for the cervical roots (Ugawa et aI., 1989; Britton et aI., 1990; Chokroverty et aI., 1991; Samii et aI., 1998) and by 3.~.1 ms for the lumbosacral roots (Ugawa et aI., 1989; Britton et al., 1990; Chokroverty et aI., 1993) compared to the CMCT obtained by one of the spinal nerve root stimulation methods. CMCT measured by the Fwave or stretch reflex methods comprises the conduction time from motor cortex to spinal cord plus one synaptic delay (synapse between the corticospinal axon and the spinal alpha-motoneuron) plus temporal summation time at the spinal alphamotoneuron to reach action potential threshold. This spinal delay time was estimated as 0.5 ±0.3 ms in active target muscle (Ugawa et aI., 1995a).

97

It is recommended to measure CMCT in the isometrically contracting target muscle (5-20% of maximum voluntary contraction) using stimulus intensities of 120-150% above motor threshold (Rossini et aI., 1994, 1999; Rothwell et aI., 1999). The CMCT measured in active muscle is shorter by some 2-3 ms (sometimes up to 6 ms) compared to the CMCT measured in resting muscle (Hess et aI., 1986; Rothwell et aI., 1987). This is explained by elimination of the temporal summation to reach action potential threshold of some spinal alphamotoneurons at motor response onset, and by recruitment of larger, more rapidly conducting spinal alpha-motoneurons during muscle contraction. The CMCT is shorter by 0-3 ms if the current induced in the brain is directed lateral-to-medial rather than posterior-to-anterior due to the preferential activation of the D-wave with lateral-to-medial directed currents compared with preferential activation of I-waves with posterior-to-anterior directed currents (see above) (Werhahn et aI., 1994). These sources of variance underline the necessity to adhere always to a standardized protocol for CMCT measurements (e.g. Rossini et aI., 1994, 1999; Rothwell et al., 1999). Normative CMCT data in adults are available for many muscles of the upper and lower limb, axial muscles and cranial muscles (e.g. Rothwell et al., 1991). CMCT to lower but not upper limb muscles requires correction for body height (Claus, 1990). Finally, CMCT changes with maturation and aging of the corticospinal tract (Muller et al., 1991; Eyre and Miller, 1992; Kloten et aI., 1992; Caramia et aI., 1993). Conduction time measurements of the corticospinal tract can be segmented. The CMCT from motor cortex to the pyramidal tract decussation is the latency difference of the motor evoked potentials elicited by motor cortex stimulation and by either magnetic (Ugawa et aI., 1994b; Ugawa, 1999) or electrical direct stimulation (Ugawa et aI., 1991b) of the descending corticospinal tract at the level of the foramen magnum. Furthermore, the corticospinal tract can be activated by transcutaneous electrical stimulation at several levels further downstream (Ugawa et aI., 1995a). The conduction velocity is 68±5 mls along the corticospinal tract but lowers to 40±7 mls at the cauda equina (Ugawa et aI., 1995a). Conduction time within the cauda equina can be measured as the latency difference of motor evoked

98

potentials elicited by magnetic stimulation with a large eight-shaped coil oriented vertically over the proximal cauda and horizontally over the distal cauda (Maccabee et al., 1996).

8.2.1.2. Applications CMCT prolongation may be caused by: (i) demyelination (and therefore slowing of conduction) of the fastest-conducting corticospinal fibers; (ii) degenerative or ischemic damage of the fastestconducting fibers; and (iii) increased dispersion of the multiple corticospinal discharge, resulting in less effective temporal summation at the spinal alphamotoneuron. Consequently, CMCT measurements are useful in central demyelinating disease (e.g. multiple sclerosis), cerebral ischemic stroke, myelopathies, neurodegenerative disease affecting the corticospinal tract (e.g. ALS, hereditary spastic paraplegia, HMSN with pyramidal signs, Friedreich's ataxia, SCA-1), some movement disorders (e.g. Wilson's disease, MSA, PSP), mitochondrial disorders, and myotonic dystrophy. 8.2.2. Motor evoked potential amplitude 8.2.2.1. Techniques and principles The amplitude of the motor evoked potential (MEP) elicited by motor cortex stimulation should be related to the amplitude of the maximum M wave obtained by supramaximal stimulation of the peripheral nerve (Rossini et al., 1994): MEP%=MEPlMmaxx 100 where MEP and M max are the peak-to-peak amplitude (Rossini et al., 1994) or area under the curve (Kiers et al., 1995) of the motor evoked potential and the maximum M wave, respectively. This ratio estimates the portion of the spinal alpha-motoneuron pool activated by TMS (Rossini et al., 1994). MEP amplitude and MEP area increase with stimulation intensity, usually in a non-linear, sigmoid fashion (Hess et al., 1987a; Rothwell et al., 1987; Kiers et al., 1995; Devanne et al., 1997). Intrinsic hand muscles show the steepest input-output relation and highest MEP% levels (Rothwell et al., 1987; Chen et al., 1998b) while proximal arm muscles, in particular the biceps (Rothwell et al., 1987; Chen et al., 1998b), trunk muscles (Chen et al., 1998b) and lower limb muscles (Rothwell et al., 1987; Chen et al., 1998b) have lower figures. These differences are

V.ZIEMANN

explained by the highest density and largest number of corticomotoneuronal fibers, and the highest excitatory descending drive onto spinal alpha-motoneurons of intrinsic hand muscles compared to other muscles (Cowan et al., 1986; Palmer and Ashby, 1992a; Maertens de Noordhout et al., 1999). MEP% increases with voluntary contraction of the target muscle (Hess et al., 1986, 1987a; Ravnborg et al., 1992; Kischka et al., 1993) or, to a lesser degree, with contraction of other muscles (Hess et al., 1986; Pereon et al., 1995; Stedman et al., 1998). Most likely, both cortical (Hess et al., 1986; Baker et al., 1995; Lemon et al., 1995; Pereon et al., 1995; Mills and Kimiskidis, 1996; Di Lazzaro et al., 1998b) and spinal mechanisms (Maertens de Noordhout et al., 1992a; Kaneko et al., 1996b; Di Lazzaro et al., 1998b) contribute this MEP facilitation induced by contraction. Measurements of MEP% suffer from two important limitations. First, MEP% shows a considerable trial-to-trial variability the causes of which are not entirely understood (Hess et al., 1987a; Rothwell et al., 1987; Kiers et al., 1993; Nielsen, 1996; Van der Kamp et al., 1996; Ellaway et al., 1998; Gugino et al., 2001). This variability can be reduced by high stimulus intensity, controlled muscle contraction, or paired stimulation (Nielsen, 1996; Van der Kamp et al., 1996; Kiers et al., 1993). Second, even with maximum stimulus intensity and strong muscle contraction, MEP% is usually considerably less than 100%. This is mainly due to phase cancellation of muscle action potentials, caused in turn by the desynchronized activation of the spinal alpha-motoneurons (Magistris et al., 1998). For intrinsic hand muscles, maximum MEP% values are, on average, around 50-60% with a large inter-individual variability (Hess et al., 1987a, b; Rothwell et al., 1987). Therefore, MEP% in hand muscles can be considered abnormal only when < 15%.

8.2.2.2. Applications MEP% reduction may be caused by: (i) increased dispersion of the multiple corticospinal discharge, resulting in increased desynchronicity of alphamotoneuron discharge and increased phase cancellation; (ii) complete or incomplete failure of conduction due to interruption or degeneration of the corticospinal tract; (iii) severe depression of excitability of the corticospinal neuron and/or the spinal motoneuron; and (iv) intracortical conduction block

TRANS CRANIAL MAGNETIC STIMULATION

along the neural elements responsible for the generation of l-waves. MEP% measurements are useful in the same diseases given above for CMCT measurements. 8.2.3. Triple stimulation technique 8.2.3.1. Techniques and principles The problem of phase cancellation in the conventional MEP% measurement can be overcome by a recently developed triple stimulation technique (TST) (Magistris et al., 1998) which, through two collisions, suppresses MEP desynchronization. TST was first introduced and will be described here for the abductor digiti minimi muscle (Magistris et al., 1998) (Fig. 2). The first pulse is TMS applied to the hand area of the motor cortex. The second pulse is supramaximal electrical stimulation to the ulnar nerve at the wrist after Delay I (equal to the minimal latency of the MEP, rounded down to the nearest millisecond, minus the latency of the maximum M wave, rounded up to the nearest millisecond). The third pulse is supramaximal electrical stimulation of Erb's point after Delay II (equal to the CMAP Erb latency rounded down to the nearest millisecond, minus the latency of the maximum M wave, rounded up to the nearest millisecond). The first collision occurs along the peripheral nerve between the descending volley induced by TMS and the antidromically ascending volley induced by the distal ulnar nerve stimulation. The second collision occurs between the descending volley induced by stimulation of Erb's point and that part of the ascending volley produced by distal ulnar nerve stimulation which had not been collided in the first collision. Two extreme scenarios are possible. If TMS results in 0% activation of the alpha-motoneuron pool (due to, e.g. subthreshold TMS or complete cortico-spinal conduction block) then no first collision takes place because there is no descending volley. The second collision will be complete so that the resulting CMAP Erb after TST will be 0% of an unconditioned CMAPErb control response. In contrast, if TMS results in 100% activation of the alpha-motoneuron pool then the first collision will be complete. No second collision will occur because there is no remaining part of the ascending volley from the distal ulnar nerve stimulation to collide with. Therefore, CMAP Erb after TST will be 100% of the unconditioned CMAP Erb control response. The larg-

99

est obtainable ratio of TST CMAPErJcontrol CMAP Erb was, on average, 99±2% in healthy subjects (Magistris et al., 1998). To achieve this, slight contraction of the target muscle was required in most instances. The intensity of TMS was on average 80% of maximum stimulator output. TST values <93% were considered abnormal (Magistris et al., 1998). TST was also validated for a foot muscle (Buhler et al., 2001). Limitations of TST are that it is more complex than MEP% measurements and cannot be applied to proximal muscles. 8.2.3.2. Applications TST shows promise to detect and quantify with greater sensitivity and reliability conduction failure along the corticospinal tract than was possible with the conventional MEP% technique (Magistris et al., 1999; Rosier et al., 2000; Truffert et al., 2000). Consequently, TST is useful in the diseases given above for CMCT and MEP%. 8.2.4. Single motor unit recordings 8.2.4.1. Techniques and principles In the routine neurophysiological setting, the motor response induced by TMS is recorded most often by surface electromyography (EMG). In addition, or alternatively, single motor units (SMU) can be recorded with needle EMG. This may reveal additional information that cannot be disclosed with surface EMG (Mills, 1991). In hand muscles of healthy subjects, peri-stimulus time histograms (PSTH) show an increased firing probability of a voluntary activated SMU 17-30 ms after the TMS pulse (Day et al., 1989; Boniface et al., 1991; Mills, 1991) (Fig. 1). The area of this 'primary peak' can be used to estimate the amplitude of the composite excitatory post-synaptic potential (EPSP) at the spinal alpha-motoneuron, and the duration of the primary peak reflects the rise time of this EPSP (Ashby and Zilm, 1982; Fetz and Gustafsson, 1983). The short duration of the primary peak of 4.6± 1.7 ms suggests that the corticospinal neurons activated by TMS project largely mono-synaptically onto the spinal alpha-motoneurons (Day et al., 1989). The primary peak is largest for intrinsic hand muscles (estimated mean EPSP for the first dorsal interosseus 2.9±0.2 mY) while forearm muscles and the biceps brachii receive weaker net facilitation (Palmer and Ashby, 1992a). Many triceps and deltoid SMU are

100

V.ZIEMANN

2

(A) Brain ....,4A

3

4

1~

(\ multiple J \\ ~~~YS ~

r--.- \

Erb

Wrist

T(?st curve

wtt t

t' n

....

_~

---

V ""

br~t stirn.

J \ »;: .__ \

V

stirn.

Erb's

~-_.

stirn.

(8)

1

2

3

11

Erb

r-~

I,

TST control curve

t

t t\.

Erb's wrist stlm. sum.

Wrist =-~

(C) Wrist

(D)

Er 's

snm

lill

Superimposition

L

TRANS CRANIAL MAGNETIC STIMULATION

inhibited by TMS so that there is no net facilitation (Palmer and Ashby, 1992a). SMU are recruited by TMS in the same order as by voluntary contraction, following the Henneman size principle (Hess et aI., 1987a; Awiszus and Feistner, 1994; Awiszus et aI., 1999). The primary peak is divided into subpeaks, some 1.5 ms apart (Day et aI., 1989; Mills, 1991) (Fig. 1). Very likely, these subpeaks reflect the Dand I-waves (see above). With different current directions in the motor cortex it is possible to activate selectively single subpeaks (Day et al., 1989; Sakai et aI., 1997; Terao et aI., 2001) (Fig. 1).

8.2.4.2. Applications Multiple sclerosis (Boniface et aI., 1991), ALS (e.g. Awiszus and Feistner, 1995; Weber et aI., 2000), Parkinson's disease (Kleine et aI., 2001), spinal cord injury (Smith et aI., 2000). 8.3. Motor cortex excitability 8.3.1. Motor threshold 8.3.1.1. Techniques and principles Motor threshold (MT) has been defined in different ways in the literature. According to IFCN recommendation, MT is the minimum intensity that is sufficient to produce a small MEP (> 50 IJo V) in at least half of the trials in the resting (resting MT) or contracting (active MT) target muscle (Rossini et aI., 1994, 1999). A slightly different approach deter-

101

mines the maximum stimulus intensity with a probability of zero to elicit an MEP in the target muscle (lower threshold) and the minimum stimulus intensity with a probability of one (upper threshold) (Mills and Nithi, 1997b). Finally, threshold can be defined as the x-axis intercept of the peak slope tangent of the MEP intensity curve (Carroll et aI., 2001). Within-subject variability of MT is low so that longitudinal measurements are reasonable (Ziemann et aI., 1996a; Mills and Nithi, 1997b; Carroll et aI., 2001). In contrast, between-subject variability is high for a largely unexplained reason (Cicinelli et aI., 1997; Mills and Nithi, 1997b). The active MT is on average 25% lower than the resting MT. Therefore, it is important to monitor complete voluntary muscle relaxation for measurements at muscle rest. This can be done by high gain (50 IJo V/Div) EMG audiovisual feedback, and automatic trial rejection whenever significant voluntary EMG activity contaminated the pre-stimulus EMG period (Kaelin-Lang and Cohen, 2000). MT is lowest for hand muscles and higher for proximal muscles of the arm (Brouwer and Ashby, 1990; Macdonell et aI., 1991; Chen et aI., 1998b), trunk and lower limb (Chen et aI., 1998b). These differences may reflect the density of the corticomotoneuronal projection which is highest for intrinsic hand muscles. MT reflects the excitability of the most excitable corticospinal neurons (Hallett et al., 1999). Insight into the physiology of the MT and other motor cortex excitability measures (see below)

Fig. 2. The triple stimulation technique (TST) principle. On the left, a schematic diagram of the motor tract is simplified to four corticospinal axons with monosynaptic connections to four spinal motoneurons. Horizontal lines represent muscle fibers of four motor units. EMG recordings in the abductor digiti minimi muscle are shown on the right. (A) TST test, (H) TST control, (C) response to a single stimulus at wrist and (D) superimposition of recordings A-C. In this example, submaximal TMS excites 75% of the axons (three axons out of four). Desynchronization of the three action potentials is assumed to occur within the corticospinal tract (or possibly at the spinal motoneuron level). (A, 1) TMS excites three out of four axons. (A, 2) After a delay, a supra-maximal stimulus applied to the wrist evokes the first negative (upward) deflection in the TST test trace, this response is followed by that of the multiple-discharge volley of spinal motoneurons (not figured on the left scheme). (A, 3) After a delay, a supra-maximal stimulus is applied to Erb's point; (A, 4) a synchronized response from the three axons excited initially by TMS is recorded as the second large deflection of TST. (H, 1) A supra-maximal stimulus is applied to Erb's point; (H, 2) after a delay, a supra-maximal stimulus applied to the wrist evokes the first deflection of TST control trace; (H, 3) after a delay, a supra-maximal stimulus is applied to Erb's point; (H, 4) a synchronized response from the four axons is recorded as the second deflection ofTST control trace. (C) The response evoked by stimulating the wrist serves as a baseline for measurement of the amplitude and area of the second deflection of the TST curves. (D) On the superimposed traces, the smaller size of the second deflection on the TST test trace, compared with that of the TST control trace, demonstrates that not all spinal axons of the target muscle were excited by TMS (in this example both amplitude and area ratios should be 75% if the four individual motor units had identical sizes). Calibrations: 2 mV and 5 ms (with kind permission, from Magistris et al., 1998).

V.ZIEMANN

102

was fostered by the combination of TMS and neuropharmacology. Voltage-gated sodium and calcium channel blocking drugs, such as carbamazepine or phenytoin elevate MT (Mavroudakis et al., 1994; Ziemann et al., 1996a; Chen et al., 1997b; Boroojerdi et al., 2001), whereas a single loading dose of neurotransmitter (GABA, glutamate, dopamine, serotonin, norepinephrine, acetylcholine) modulating drugs does usually not affect MT (for review, Ziemann and Hallett, 2000). These findings suggest that MT reflects primarily membrane-related ionchannel dependent excitability ofaxons excited by TMS. 8.3.1.2. Applications Increased MT may indicate significant damage of the corticospinal tract after stroke (Homberg et al., 1991; Catano et al., 1996) or spinal cord lesion (Linden and Berlit, 1994; McKay et al., 1997). Decreased MT suggests increased corticospinal tract excitability in untreated patients with idiopathic generalized epilepsy (Reutens et al., 1993a), in nonwasted muscles of patients with early ALS (Mills and Nithi, 1997a), in Alzheimer's disease (Pepin et al., 1999) and in patients with obsessive-compulsive disorder (Greenberg et al., 2000). MT is of limited use as a one-point observation in a single patient due to large between-subject variability, but longitudinal measurements are feasible due to low within-subject variability. 8.3.2. Motor evoked potential intensity curve 8.3.2.1. Techniques and principles The MEP intensity curve is defined as the increase in MEP amplitude (or MEP area) with stimulus intensity. This increase is typically non-linear (Hess et al., 1987a; Kiers et al., 1995) and described best by a sigmoid function (Devanne et al., 1997; Carroll et al., 2001)

MEP(s)=MEPrna/I +e m(S50-s) where MEP(s) is the MEP amplitude (or MEP area) at a given stimulus intensity (s), MEP max is the maximum MEP defined by the function, m is the slope parameter of the function, and S50 is the stimulus intensity at which the MEP size is 50% of MEPmax' Stimulus intensity can be set as percentage of maximum stimulator output irrespective of motor threshold. This facilitates longitudinal observations

where the primary interest is to compare MEP elicited by the same physical stimulus intensities. Another possibility is to set stimulus intensity as percentage of motor threshold. This would facilitate longitudinal observations corrected for change in threshold. Voluntary contraction of the target muscle shifts the MEP intensity curve toward lower intensities (decrease in motor threshold) and increases the slope (Devanne et al., 1997). The high-intensity part of the MEP intensity curve reflects most likely the excitability of corticospinal neurons with intrinsically high threshold (Thickbroom et al., 1998; Hallett et al., 1999). Several neurotransmitter-modulating drugs decrease (lorazepam, levetiracetam) or increase (d-amphetamine) slope and MEP max (Boroojerdi et al., 2001; Sohn et al., 2001). 8.3.2.2. Applications Movement disorders (Valls-Sole et al., 1994; Ikoma et al., 1996), reorganization of the motor cortex after limb amputation (Cohen et al., 1991a; Kew et al., 1994; Chen et al., 1998a). 8.3.3. Motor evoked potential mapping 8.3.3.1. Techniques and principles An MEP map is the area on the scalp surface from which MEPs can be elicited in a given target muscle. To obtain an MEP map, multiple scalp sites are stimulated by moving the stimulating coil (ideally a small eight-shaped coil) along a grid. The coordinates of the grid should be referenced relative to standard landmarks, such as C, according to the International 10-20 system. Mapping can be done either along grids with co-ordinates 0.5-2 em apart (Cartesian co-ordinate system) (Brasil-Neto et al., 1992; Wassermann et al., 1992; Classen et al., 1998), or grids with co-ordinates based on a latitudelongitude system, which forms a more general frame of reference by taking into account head curvature (Wilson et al., 1993b; Thickbroom et al., 1998). MEP mapping guided by online co-registered MRI shows significantly improved precision of coil placement compared with conventional 'blind' MEP mapping (Gugino et al., 2001). Mapping should be performed with a moderate stimulation intensity of 110-120% MT, as determined at the optimal site for eliciting MEP in the target muscle. To account for the considerable trial-to-trial variability in MEP amplitude which even increases with distance from

TRANS CRANIAL MAGNETIC STIMULATION

the optimal site (Brasil-Neto et aI., 1992), about 10 trials need to be applied to a given co-ordinate (Classen et aI., 1998). Mapping should be continued until effective sites are completely surrounded by non-effective sites. A restriction to those sites in proximity to the optimal site may degrade map accuracy considerably (Classen et aI., 1998). With an optimal mapping technique, good reliability (Mortifee et aI., 1994) and a spatial resolution in the order of 0.5 em (Brasil-Neto et aI., 1992) can be achieved. Any map is characterized by three features: extent, location and shape. Map extent is usually expressed by the number of effective stimulation sites (Wassermann et aI., 1992; Classen et aI., 1998). Map extent is a direct function of the excitability of the stimulated corticospinal neurons as shown by a close correlation between map extent and the slope of the MEP intensity curve (Ridding and Rothwell, 1997). Current spread and the depth of the stimulated corticospinal neurons with respect to scalp surface contribute to map extent, i.e. the map is always larger than the actual extent of the population of stimulated corticospinal neurons (Thickbroom et aI., 1998). Map location is often expressed by the site of the maximum MEP, but it is better to use the center of gravity (COG) which is the sum of all map coordinates weighted by MEP amplitude at those co-ordinates, divided by the sum of all MEP amplitudes (map volume) (Wassermann et aI., 1992). The weight at any scalp position can be interpreted as the proportion of the map volume contributed by that location. Map location corresponds to the scalp location at which the greatest number of the most excitable corticospinal neurons can be stimulated (Classen et aI., 1998; Thickbroom et aI., 1998). Map shape is given in descriptive terms. Maps are usually ellipsoid with the long axis parallel to the direction of the inducing current (Wilson et aI., 1996). MEP map reorganization is demonstrated best by changes in map location while changes in map extent are confounded by corticospinal excitability (Ridding and Rothwell, 1997). MEP mapping has confirmed a rough somatotopical order of motor representations in human motor cortex with the face, hand, upper arm, neck/trunk and leg MEP located along a lateral-to-medial axis (Wassermann et al., 1992; Metman et aI., 1993; Thompson et aI., 1997; Classen et aI., 1998; Krings et al., 1998). Maps for different muscles of the hand and arm overlap, but hand representations are more

103

lateral than arm representations (Wassermann et al., 1992). MEP mapping was referred to the underlying anatomy (Levy et al., 1991; Wang et aI., 1994; Krings et aI., 1997, 1998; Singh et aI., 1997), and multi-modal approaches combined the results of MEP mapping with functional activation studies using positron emission tomography (Wassermann et aI., 1996c; Classen et aI., 1998) and functional magnetic resonance imaging (Krings et aI., 1997; Bastings et aI., 1998; Terao et aI., 1998; Boroojerdi et aI., 1999). These studies generally show that the MEP maps project onto the precentral gyrus and largely overlap with the functional activation areas.

8.3.3.2. Applications Testing motor cortex reorganization associated with cerebral stroke (e.g. (Hallett et aI., 1998; Liepert et aI., 1998; Byrnes et aI., 1999; Traversa et aI., 1999), dystonia (Byrnes et aI., 1998), facial palsy (Rijntjes et aI., 1997), hemi-facial spasm (Liepert et aI., 1999a), spinal cord injury (Levy et aI., 1990; Topka et aI., 1991; Streletz et aI., 1995), limb amputation (Cohen et aI., 1991a; Kew et al., 1994; Ridding and Rothwell, 1997; Karl et aI., 2001), motor practice and motor learning (Pascual-Leone et aI., 1994a; Pascual-Leone et al., 1995a, b; Liepert et aI., 1999b). MEP mapping may be used complementary to other techniques in pre-surgical identification of motor cortex. 8.3.4. Cortical silent period 8.3.4.1. Techniques and principles The cortical silent period (CSP) is defined as interruption of tonic voluntary EMG activity in a target muscle contralateral to the stimulated motor cortex (Cantello et al., 1992; Wilson et aI., 1993a). CSP duration is usually determined from the TMS stimulus artifact, or alternatively from MEP onset, to the resumption of sustained voluntary EMG activity. Automatic determination of CSP duration is possible by statistical procedures that compare the poststimulus EMG with the pre-stimulus EMG (Nilsson et aI., 1997; Garvey et aI., 2001). A CSP can be obtained in any muscle but is longest in the intrinsic hand muscles where it can exceed 250 ms (Cantello et al., 1992). CSP duration correlates approximately linearly with stimulus intensity (Cantello et al., 1992; Haug et aI., 1992; Inghilleri et aI., 1993; Triggs et aI., 1993; Wilson et al., 1993a). Ideally,

104

stimulus intensity should be related to CSP threshold, not to MEP threshold, as MEP and CSP may be differentially affected by disease (Chistyakov et al., 2001). The level of muscle contraction either does not affect CSP duration (Haug et al., 1992; Inghilleri et al., 1993; Roick et al., 1993) or leads to slight shortening (Cantello et al., 1992; Wilson et al., 1993a; Mathis et al., 1998). CSP duration is altered by specifics of the motor instruction, e.g. to prepare for contraction, or to hold isotonic vs. isometric contraction (Mathis et al., 1998; Hoshiyama and Kakigi, 1999; Mathis et al., 1999). Between-subject variability is high, while within-subject variability and side-to-side differences between homologous muscles are low (Cicinelli et al., 1997; Fritz et al., 1997). The late part of the CSP originates mainly or exclusively at the cortical level while the early part is mainly due to inhibition of the spinal alphamotoneuron (Fuhr et al., 1991; Cantello et al., 1992; Inghilleri et al., 1993; Ziemann et al., 1993; Davey et al., 1994; Chen et al., 1999). The contributing mechanisms of the late part of the CSP are not exactly known but most likely GABA-B receptor mediated cortical inhibition plays a crucial role. Pharmacological activation of GABA-B receptors leads to significant CSP lengthening (Siebner et al., 1998a; Werhahn et al., 1999). 8.3.4.2. Applications Epilepsy (e.g. Classen et al., 1995; Cincotta et al., 1998; Inghilleri et al., 1998; Macdonell et al., 2001), cerebral stroke (e.g. Schnitzler and Benecke, 1994; VonGiesen et al., 1994; Classen et al., 1997; Ahonen et al., 1998), Parkinson's disease (for review, Cantello et al., 2002), Huntington's disease (Priori et al., 1994; Modugno et al., 2001), dystonia (Filipovic et al., 1997; Rona et al., 1998; Amadio et al., 2000), tic disorder (Ziemann et al., 1997; Moll et al., 2001), congenital mirror movement (Cincotta et al., 1996), ALS (Prout and Eisen, 1994; Siciliano et al., 1999), stiff person syndrome (Sandbrink et al., 2000), cerebellar ataxia (Oechsner and Zangemeister, 1999), tetanus (Warren et al., 1999), migraine (Aurora et al., 1999; Werhahn et al., 2000). 8.3.5. Paired-pulse excitability at long interstimulus intervals (20-200 ms) 8.3.5.1. Techniques and principles Paired-pulse excitability at long inter-stimulus intervals (20-200 ms) refers to the modulating

V.ZIEMANN

effects of a supra-threshold conditioning pulse on the amplitude of the test MEP elicited by the subsequent supra-threshold test pulse. Both pulses are given through the same stimulating coil, usually at the same stimulus intensity of 110-150% of MT. The target muscle can either be at rest or voluntarily activated. The effect of the conditioning pulse is usually facilitatory at intervals of 20-40 ms, but inhibitory at intervals >50 ms (Claus et al., 1992; Valls-Sole et al., 1992). This paired-pulse facilitation and paired-pulse inhibition occur mainly through mechanisms at the level of the motor cortex (Claus and Brunholzl, 1994; Kaneko et al., 1996c; Nakamura et al., 1997; Chen et al., 1999). Paired-pulse inhibition at long inter-stimulus intervals underlies mechanisms different from those responsible for the paired-pulse inhibition at short inter-stimulus intervals (1-5 ms, see below) (Sanger et al., 2001), and does not test the same long-lasting inhibitory mechanisms as the CSP (Berardelli et al., 1996). 8.3.5.2. Applications Epilepsy (Brodtmann et al., 1999; Valzania et al., 1999; Manganotti et al., 2000), Parkinson's disease (for review, Cantello et al., 2002), Huntington's disease (Tegenthoff et al., 1996; Priori et al., 2000), dystonia (Chen et al., 1997c; Rona et al., 1998), cerebellar ataxia (Wessel et al., 1996). 8.3.6. Paired-pulse inhibition and facilitation at short inter-intervals (1-30 ms) 8.3.6.1. Techniques and principles Paired-pulse inhibition (PPI) and facilitation (PPF) at short inter-stimulus intervals (1-30 ms) refer to the modulating effects of a sub-threshold conditioning pulse on the amplitude of the test MEP elicited through the same stimulating coil by the subsequent supra-threshold test pulse (Fig. 3). The intensity of the conditioning pulse is usually set to 80% of resting MT or 90% of active MT, and the intensity of the test pulse to produce a test MEP of about 1 mV in peak-to-peak amplitude (Kujirai et al., 1993a; Ziemann et al., 1996b). PPI occurs at inter-stimulus intervals of 1-5 ms, PPF at intervals of 7-20 ms (Kujirai et al., 1993a; Ziemann et al., 1996b; Ziemann, 1999). Both originate through mechanisms at the level of the motor cortex (Di Kujirai et al., 1993a; Ziemann et al., 1996b; Nakamura et al., 1997; Lazzaro et al., 1998c). PPF is

105

TRANSCRANIAL MAGNETIC STIMULATION

ShOCkal Conditioning

.~ 200 III

-

150

0

100 ........ - -

"0 ~

lSI

=3

c: 0

ms , ~.

lSI

>=

0

.

.'

Cl

:"

!!c: 50 1

2 ms

1 mV

~ ~

Q)

a,

0

-~

_. f/

- - -- - -- _

.

+_1

Q)

',,:

--

f

+

"'-./t-'" 0

5

10

15

Interstimulus interval (ms)

10 ms Fig. 3. Paired-pulse inhibition and facilitation at short interstimulus-intervals (PPI, PPF). MEPs by a supra-threshold magnetic cortical test stimulus in relaxed first dorsal interosseous muscle are inhibited by a prior, sub-threshold conditioning stimulus at short inter-stimulus intervals of 1-5 ms (PPI) and facilitated at longer intervals of 10--15 ms (PPF). The left panel shows examples of EMG data from one healthy subject. The first trace shows absence of any MEP to the conditioning stimulus alone. The lower two records have two superimposed traces, the MEP to the test stimulus given alone, and the MEP to the test stimulus when given 3 (middle traces) or 2 ms (lower traces) after the conditioning stimulus. The larger MEP (dotted line) is the response to the test stimulus alone. It is dramatically suppressed at these two inter-stimulus intervals. Each trace is the average of 10 trials. The right panel shows the averaged group data of 6 subjects (means ±SD). The conditioned MEP is given as a percentage of the test MEP (y-axis) and expressed as a function of the inter-stimulus interval (x-axis) (with kind permission, from Kujirai et al., 1993a).

physiologically distinct from PPI and not merely a rebound facilitation (Ziemann et aI., 1996b; Strafella and Paus, 2001). PPI and PPF are tested at muscle rest, as both measures are suppressed by even slight voluntary contraction (Ridding et aI., 1995). PPI and PPF are studied predominantly in hand muscles but can be obtained similarly in many other muscles (Chen et aI., 1998b). PPI but not PPF decreases with age (Peinemann et aI., 2001) and may be affected by personal trait, such as the level of neuroticism (Wassermann et aI., 2001) and by the menstrual cycle (Smith et aI., 1999). GABA-A receptor agonists, N-methyl-o-aspartate (NMDA) receptor blockers, dopamine receptor agonists and serotonin result in PPI increase and/or PPF decrease (for review, Ziemann and Hallett (2000». Dopamine (D2) receptor antagonists, muscarinic receptor antagonists, GABA-B auto-receptor activation and norepinephrine agonists decrease PPI and/or increase PPF (Ziemann and Hallett, 2000; Bor-

oojerdi et aI., 2001; Liepert et aI., 2001; Plewnia et aI., 2001). Recent findings indicate that the PPI consists of at least two distinct phases of inhibition with different physiological properties, one at interstimulus intervals of about I ms, and the other at intervals of around 2.5 ms (Fisher et aI., 2002). Futhermore, PPI may in fact be a net inhibition, consisting of strong inhibitory and weaker facilitatory effects (see l-wave facilitation below). In summary, it is currently believed that PPI and PPF test the integrity and excitability of inhibitory and excitatory neuronal circuits in the motor cortex which are under the control of various neurotransmitter systems and in tum control the excitability of corticospinal neurons. 8.3.6.2. Applications (for reviews, Ziemann et al., 1998a; Ziemann, 1999; Cantello et al., 2002) Epilepsy, cerebral stroke, Parkinson's disease, Huntington's disease and other dyskinetic syn-

106

dromes, dystonia, cerebellar ataxia, ALS, autosomaldominant spastic paraplegia, stiff-person syndrome, restless legs syndrome, migraine, limb amputees, Alzheimer's disease, tic and other neuro-psychiatric disorders. Most of these diseases show a decrease in PPI and/or an increase in PPF. Therefore, while suggestive of a good sensitivity for the detection of abnormalities of motor cortex excitability, PPI and PPF at present do not differentiate between even very different pathological conditions. This weakness may be overcome by refined paired-pulse stimulation techniques (Fisher et al., 2002). 8.3.7. I-wave facilitation 8.3.7.1. Techniques and principles I-wave facilitation refers to the facilitatory effects of a sub-threshold second pulse on the amplitude of a test MEP elicited by a supra-threshold first stimulus given through the same stimulation coil 0.5-6.0 IDS earlier (Ziemann et al., 1998b). Alternatively, two pulses close to MT can be used (Tokimura et al., 1996). I-wave facilitation occurs at discrete inter-stimulus intervals of 1.1-1.5 ms, 2.3-2.9 ms and 4.1-4.5 ms with much less effect at intermediary intervals (Tokimura et al., 1996; Ziemann et al., 1998b). I-wave facilitation originates through mechanisms at the level of the motor cortex (Tokimura et al., 1996; Ziemann et al., 1998b; Di Lazzaro et al., 1999c; Hanajima et al., 2002). The inter-peak latency between the three facilitatory MEP peaks is approximately 1.5 ms, comparable to the succession of I-waves (see above). I-wave facilitation is reduced by GABAergic drugs (Ziemann et al., 1998c; Wischer et al., 2001). In summary, the available evidence suggests that this paired pulse technique probes the excitability of motor cortical circuits that are responsible for the generation ofl-waves. 8.3.7.2. Applications Patients with multiple sclerosis may show a reduction of l-wave facilitation or even a complete disorganization of the MEP facilitatory peaks (Ho et ai., 1999).

8.4. Motor cortex connectivity As a general principle, motor cortex connectivity is assessed by testing the effects of a conditioning

V.ZIEMANN

stimulus on the amplitude of a test MEP elicited by stimulation of the motor cortex. Another possibility is to stimulate the motor cortex and assess the activation of distant target structures by functional imaging (PET, fMRI) or electrophysiological methods (EEG, SEP, MRP). 8.4.1. Connectivity between different motor representations within motor cortex 8.4.1.1. Techniques and principles Focal conditioning stimulation of the leg area of the motor cortex inhibits a test MEP elicited by a supra-threshold pulse given a few milliseconds (1-5 ms) later over the hand area, and vice versa (Kujirai et al., 1993a). This suggests that within motor cortex connectivity is largely inhibitory. 8.4.1.2. Applications Lateral spread into motor representations of the proximal arm can occur during high-frequency repetitive TMS of the hand area of the motor cortex suggesting stimulation-induced break-down of cortico-cortical inhibitory mechanisms (Pascual-Leone et al., 1994c). Similarly, propagation of epileptic activity, such as the Jacksonian march, or an overflow of movement associated with an intended focal voluntary movement, such as in dystonia, may originate from deficient cortico-cortical inhibition between different motor representations, although this has not yet been tested. 8.4.2. Connectivity ofpremotor cortex and SMA with motor cortex 8.4.2.1. Techniques and principles Conditioning focal stimulation 3-5 em anterior to the hand area of the motor cortex, or 6 em anterior to the vertex inhibits the test MEP elicited by suprathreshold stimulation over the hand area of the motor cortex (Civardi et al., 2001). This effect is maximal at sub-threshold intensity (90% of active MT) and at an inter-stimulus interval of 6 IDS. It is thought that these conditioning sites correspond to the premotor cortex and the pre-SMA or SMA proper, and that their connections to the hand area of the motor cortex are largely inhibitory (Civardi et al., 2001; Gerschlager et al., 2001).

TRANS CRANIAL MAGNETIC STIMULATION

8.4.2.2. Applications No data available yet. Potentially useful in neurological diseases with a presumed abnormal connectivity between the pre-motor cortex or SMA and the primary motor cortex, in particular movement disorders. 8.4.3. Inter-hemispheric connectivity between the two motor cortexes 8.4.3.1. Techniques and principles The hand areas of the two motor cortexes are connected, although sparsely, by callosal fibers (Gould et al., 1986; Rouiller et al., 1994). This transcallosal connection can be tested by the ipsilateral silent period (ISP) (Wassermann et al., 1991; Ferbert et al., 1992b; Meyer et al., 1995; Meyer et al., 1998) and inter-hemispheric inhibition and facilitation measured with a paired stimulation protocol (Ferbert et al., 1992b; Ugawa et al., 1993; Netz et al., 1995; Gerloff et al., 1998; Di Lazzaro et al., 1999a; Hanajima et al., 200la). The ISP refers to the interruption of voluntary tonic EMG activity caused by TMS of the motor cortex ipsilateral to the target muscle. In hand muscles, the ISP onset is 10-15 ms later than the onset latency of the contralateral MEP. This difference corresponds to the estimated conduction time through the corpus callosum (Cracco et al., 1989). ISP duration in hand muscles is about 30 ms (Wassermann et al., 1991; Ferbert et al., 1992b; Meyer et al., 1995; Meyer et al., 1998). Children up to the age of 6 years do not show an ISP (Heinen et al., 1998), suggesting maturation of inter-hemispheric connections between the two motor cortexes later in life. Paired stimulation applies a conditioning stimulus over one motor cortex followed by a test stimulus over the other motor cortex. Inhibition of the test MEP occurs at inter-stimulus intervals of around 10 ms, if the intensities of both stimuli are clearly above MT (Ferbert et al., 1992b; Netz et al., 1995). Interhemispheric facilitation results at inter-stimulus intervals of 4-5 or 8 ms, if the intensity of the conditioning stimulus is close to MT (Ugawa, 1993 #1067; Hanajima et al., 200la ). Very likely, these inter-hemispheric interactions are mediated by transcallosal fibers (Di Lazzaro et al., 1999a), but some data point toward the contribution of other pathways

107

at the subcortical or even spinal level (Gerloff et al., 1998).

8.4.3.2. Applications ISP: Surgical lesions or agenesis of the trunk of the corpus callosum (Meyer et al., 1995; Meyer et al., 1998), multiple sclerosis (Boroojerdi et al., 1998; Hoppner et al., 1999; Schmierer et al., 2000), dystonia (Niehaus et al., 2001b), hydrocephalus (Roricht et al., 1998), schizophrenia (Hoppner et al., 2001). Inter-hemispheric facilitation/inhibition in the paired TMS protocol: Cortical myoclonus (Brown et al., 1996; Hanajima et al., 2001b), cortical-subcortical cerebral stroke (Boroojerdi et al., 1996), congenital mirror movements (Mayston et al., 1999), professional musicians (Ridding et al., 2000), schizophrenia (Daskalakis et al., 2002). 8.4.4. Connectivity from cerebellum to contralateral motor cortex 8.4.4.1. Techniques and principles The cerebellar hemispheres can be activated with percutaneous electrical (Ugawa et al., 1991a) or magnetic stimulation (Saito et al., 1995; Ugawa et al., 1995b; Werhahn et al., 1996). This leads to, on average, 50% inhibition of a test MEP elicited from the motor cortex contralateral to cerebellar stimulation at inter-stimulus intervals of 5-7 ms (Ugawa et al., 1995b; Werhahn et al., 1996). It is thought that this inhibition results from activation of the cerebello-dentato-thalamo-cortical pathway. An inhibition starting at slightly longer inter-stimulus intervals of 7-8 ms is probably caused by activation of peripheral nerve afferents at the level of the brachial plexus (Werhahn et al., 1996). 8.4.4.2. Applications The inhibitory interaction between cerebellum and motor cortex is reduced or absent in patients with lesions along the cerebello-dentato-thalamocortical pathway (Di Lazzaro et al., 1994; Ugawa et al., 1994a; Ugawa et al., 1997; Matsunaga et al., 2001). 8.4.5. Connectivity from motor cortex to ipsilateral spinal alpha-motoneurons

8.4.5./. Techniques and principles Ipsilateral corticospinal projections withdraw in an activity-dependent process during the first years

108

of life (Miiller et al., 1997; Eyre et aI., 2001). In adults, ipsilateral MEP in hand muscles are elicited only in a fraction of subjects, and only if strong voluntary contraction of the target muscle and high stimulus intensity are used (Wassermann et aI., 1991; Wassermann et aI., 1994; Ziemann et aI., 1999). Compared to contralateral MEP, ipsilateral MEP are much smaller, delayed by 5-10 ms, and the optimal stimulation site is slightly more lateral and anterior (Wassermann et aI., 1991; Wassermann et aI., 1994; Ziemann et aI., 1999). The ipsilateral MEP is either mediated by a weak residual uncrossed corticospinal tract (Eyre et aI., 2001) or an oligosynaptic cortico-reticulospinal projection (Ziemann et aI., 1999). 8.4.5.2. Applications Congenital/persistent mirror movements (e.g. Farmer et aI., 1990; Cohen et aI., 1991c), cerebral palsy (e.g. Farmer et aI., 1991; Carr et aI., 1993; Cincotta et al., 2000), adult cerebral stroke: ipsilateral MEP from the affected motor cortex (Fries et aI., 1991;Alagona et aI., 2001), adult cerebral stroke: ipsilateral MEP from the unaffected motor cortex (Turton et al., 1996; Netz et aI., 1997; Caramia et aI., 2000; Trompetto et aI., 2000), corticobasal ganglionic degeneration (Valls-Sole et aI., 2001). 8.4.6. Connectivity from cutaneous and muscle afferents to motor cortex 8.4.6.1. Techniques and principles Cutaneous and proprioceptive afferent information from the body can influence motor cortex excitability at short latencies. In upper limb muscles, electrical stimulation of a mixed nerve below or at motor threshold (resulting primarily in activation of Ia fibers) and muscle stretch produce MEP facilitation in the stimulated or stretched muscle at inter-stimulus intervals around 20-30 ms, usually followed by MEP inhibition at longer intervals (Troni et aI., 1988; Day et aI., 1991; Deuschl et aI., 1991; Mariorenzi et aI., 1991; Rossini et aI., 1991; Palmer and Ashby, 1992b; Baldissera and Leocani, 1995). The short-latency MEP facilitation after mixed nerve stimulation may be preceded by a shortlatency and short-lasting MEP inhibition at inter-stimulus intervals of 19-21 ms (Tokimura et al., 2000). Blockade of muscarinic receptors leads to a reduction of this inhibition (Di Lazzaro et aI.,

V.ZIEMANN

2000). Non-painful conditioning stimulation of digital cutaneous nerves produce predominantly short-latency inhibition in muscle adjacent to the stimulated finger, consistent with the It inhibitory period but merging also into the E2 excitatory period of the cutaneous reflex (Troni et aI., 1988; Mariorenzi et aI., 1991; Maertens de Noordhout et aI., 1992b; Clouston et aI., 1995; Manganotti et aI., 1997; Classen et al., 2000; Kofler et aI., 2001; Tamburin et aI., 2001). Short-latency MEP inhibition after median nerve stimulation occurs in antagonistic muscles (wrist extensors) supplied by the radial nerve, suggesting that the classical reciprocal inhibition at the spinal cord level might be assisted by a similar reciprocal inhibition at the level of motor cortex (Bertolasi et al., 1998). MEP modulation by afferent input from the hand is somatotopically organized (Terao et aI., 1995; Kofler et aI., 2001; Tamburin et al., 2001). All reported modulating effects of cutaneous and proprioceptive inputs on MEP amplitude occur largely or exclusively through mechanisms at the level of motor cortex because MEP evoked by TES and spinal alpha-motoneuron excitability as tested with H-reflexes and F waves are significantly less affected. 8.4.6.2. Applications Patients with lesions of the central somatosensory pathways lack a short-latency MEP modulation after conditioning stimulation of peripheral nerves (Bertolasi et al., 1998; Terao et aI., 1999). In contrast, patients with certain forms of epilepsy, such as progressive myoclonic epilepsy (Reutens et aI., 1993b; Cantello et aI., 1997) and patients with Creutzfeldt-Jacob disease (Yokotaet aI., 1994) show markedly increased short-latency MEP facilitation, indicating enhanced motor cortex excitability timelocked to the afferent input. 8.4.7. Other inputs to motor cortex (photic, auditory, nociceptive) 8.4.7.1. Techniques and principles MEPs in hand muscles are inhibited 55-70 ms after the unexpected presentation of a flash light (Cantello et al., 2000), or 30-60 ms after an unexpected loud sound (>80 dB, >50 ms duration) (Furubayashi et al., 2000). The latter effect habituates rapidly, and therefore, may transmit through the same system as the startle response. Nociceptive

TRANSCRANIAL MAGNETIC STIMULATION

electrical stimulation of digital nerves results in MEP inhibition of hand muscles but MEP facilitation of the biceps muscle (Kofler et aI., 1998; Kofler et aI., 2001). This MEP modulation occurs irrespective of whether the test MEP is elicited by TMS or TES, and is therefore best explained by a spinal withdrawal reflex. However, if nociceptive stimulation of the hand is produced by a COz laser, this results in MEP inhibition of hand muscles and the biceps if MEP are elicited by TMS (Valeriani et aI., 2001). MEP elicited by TES remain unaffected, suggesting a global inhibition of motor cortex following nociceptive input. 8.4.7.2. Applications No data available yet. MEP modulation by photic input may be useful in photic cortical reflex myoclonus. MEP modulation by auditory input may be useful in startle disease. 8.4.8. Motor cortex output in distant target structures tested by PET, fMRI. EEG. SEP, MRP 8.4.8.1. Techniques and principles Positron emission tomography (PET) can be used to detect metabolic change in brain areas distant from the cortex stimulated by repetitive TMS (rTMS). rTMS of the human frontal eye field results in visual cortex and superior parietal and medial parieto-occipital cortex network activation (Paus et aI., 1997). rTMS of motor cortex results in variable results depending on rTMS frequency, intensity and number of stimuli, and on the PET method ('8FDG, HzI50 ) (Fox et al., 1997; Paus et aI., 1998; Siebner et aI., 1998, 2000a, b, 2001). Most studies show a network activation, including the stimulated sensory-motor cortex, SMA and motor cortex of the opposite hemisphere. The combination of rTMS with functional magnetic resonance imaging (fMRI) is technically more difficult to achieve because the TMS pulses have to be interleaved with MR image acquisition. The first available studies demonstrate local activation of the stimulated motor cortex, both in block and singletrial designs (Bohning et aI., 1998; Bohning et aI., 1999; Bohning et aI., 2000). The combination of TMS and electroencephalography (EEG) allows assessment of TMS induced changes in electrical brain activity with high temporal resolution. Transcallosal responses appear

109

8.8-12.2 ms after TMS of one motor cortex at EEG electrodes over homologous areas of the opposite hemisphere (Cracco et aI., 1989). High-resolution multi-channel EEG and inversion algorithms show that TMS of the sensori-motor cortex elicits an immediate response at the stimulated site that spreads to adjacent ipsilateral motor areas within 5-10 ms and to homologous areas in the opposite hemisphere within 20 ms (Ilmoniemi et aI., 1997; Ilmoniemi et aI., 1999; Komssi et aI., 2002). Highfrequency rTMS of frontal cortex results in an increase in directed EEG coherence between the stimulated cortex and other electrode sites, mainly within the same hemisphere (ling and Takigawa, 2000). The P25 component of the median nerve somatosensory evoked potentials (SEP) is increased when conditioned by TMS over the motor cortex contralateral to median nerve stimulation (Kujirai et al., 1993b; Seyal et aI., 1993; Schiirmann et aI., 2001). This effect is maximal when TMS precedes the median nerve stimulus by 30-70 ms (Seyal et aI., 1993), by 10 ms (Kujirai et aI., 1993b), or is given simultaneously (Schiirmann et aI., 2001). This modulation of cortical components of the SEP may underlie the TMS induced degradation of sensory stimulus detection (Cohen et al., 1991b). Low-frequency (l Hz) rTMS of motor cortex given at 110% MT over a period of 15 min leads to a significant reduction in movement related potential (Bereitschaftspotential) amplitude (Rossi et aI., 2000), suggesting that rTMS interferes with movement-related brain activity, probably through influence on cortical inhibitory networks.

8.5. Repetitive transcranial magnetic stimulation 8.5.1.1. Techniques and principles RTMS refers to repeated TMS delivered to a single scalp site (Wassermann, 1998) and requires specially designed magnetic stimulators. RTMS is divided into low-frequency (~1 Hz) and highfrequency stimulation (> 1 Hz). This division is based on the different physiological effects and degrees of risk. Low-frequency rTMS results in a long-lasting depression of the excitability of the stimulated or connected cortex (Chen et aI., 1997a; Boroojerdi et aI., 2000; Maeda et al., 2000; Muellbacher et aI., 2000; Enomoto et aI., 2001;

V.ZIEMANN

110

Gerschlager et al., 2001; Touge et al., 2001; Tsuji and Rothwell, 2002) and has a low risk for adverse effects (Wassermann, 1998). In contrast, highfrequency rTMS leads to an increase in the excitability of the stimulated cortex (Pascual-Leone et al., 1994c; Maeda et al., 2000; Peinemann et al., 2000; Wu et al., 2000) and is associated with a higher risk for adverse effects (Wassermann, 1998). The parameters of stimulation are: (1) frequency; (2) intensity; (3) train length; (4) inter-train interval; and (5) total number of pulses. High-intensity and highfrequency rTMS bears the risk for spread of stimulus effects and induction of EMG discharge beyond the duration of stimulation (Pascual-Leone et al., 1993; Pascual-Leone et al., 1994c). This may result from a breakdown of cortico-cortical inhibition, and the generation of local epileptic activity. Accidental seizures were induced in altogether six healthy subjects by this form of high-intensity and -frequency rTMS (Wassermann, 1998). As a consequence, a table of the maximum safe duration of an rTMS train at a given combination of frequency and intensity was published, based on the NINDS experience (Wassermann, 1998). This table has two shortcomings as it does not include frequencies below 1 Hz and intensities below resting MT. Knowledge about the safety of the inter-train interval is limited but it was noted that two of the accidental seizures were induced at particularly short inter-train intervals ~ 1 s (Wassermann et al., I 996a; Chen et al., 1997d). Several safety studies did not find significant acute or short-term adverse effects toward motor, neuropychological, vegetative or neuro-hormonal function (Pascual-Leone et al., 1993; Wassermann et al., 1996b; Foerster et al., 1997; Jahanshahi et al., 1997; Niehaus et al., 1998; Evers et al., 2001; Niehaus et al., 2001a). It is currently unknown whether there exists an increased risk for any long-term adverse effects, in particular in those subjects who have received a large number of stimuli.

8.5.1.2. Applications Generally, rTMS is applied for two reasons, investigation of cortex function or therapy. Investigation of cortex function relies on the idea that rTMS can temporarily inactivate the stimulated cortex or neuronal network in the sense of a transient and fully reversible 'lesion', thus interfering with sensori-motor or cognitive tasks (for reviews,

Pascual-Leone et al., 1999; Walsh and Cowey, 2000). Therefore, in conjunction with PET and fMRI, rTMS can be used to determine the functional significance of metabolic activation that were demonstrated with functional neuro-imaging during sensori-motor or cognitive tasks (Cohen et al., 1997; Rossi et al., 2001). One application of particular relevance is interference of rTMS with language and speech that potentially might be used in the future as a non-invasive means for pre-surgical determination of language laterality (Pascual-Leone et al., 1991; Jennum et al., 1994; Michelucci et al., 1994; Epstein et al., 1996, 1999; Stewart et al., 2001). The most extensively investigated application of rTMS is as a therapeutic tool in major depression (for recent critical review, Lisanby and Sackheim, 2000). Other fields of potential therapeutic application are central pain following thalamic or brainstern stroke (Lefaucheur et al., 2001a, b), epilepsy (Tergau et al., 1999; Menkes and Gruenthal, 2000), Parkinson's disease (Pascual-Leone et al., 1994b; Siebner et al., 1999a; Siebner et al., 2000b), or writer's cramp (Siebner et al., 1999b). Although this appears to be an extremely important clinical avenue of rTMS, it should be noted that all of the quoted reports have not yet been rigorously replicated, and therefore have to be considered experimental work. Blinded and sham-controlled trials on large populations of patients are needed (Wassermann and Lisanby, 2001). Furthermore, recipes need to be developed for which parameters of rTMS to use for which therapeutic application. Finally, therapeutic rTMS sometimes makes things worse rather than better (Boylan et al., 2001), as may happen with any therapeutic application.

References Ahonen, JP, Jehkonen, M, Dastidar, P, Molnar, G and Hakkinen, V (1998) Cortical silent period evoked by transcranial magnetic stimulation in ischemic stroke. Electroencephalogr. Clin. Neurophysiol., 109: 224229. Alagona, G, Delvaux, V, Gerard, P, De Pasqua, V, Pennisi, G, Delwaide, PI, Nicoletti, F and Maertens de Noordhout, A (2001) Ipsilateral motor responses to focal transcranial magnetic stimulation in healthy subjects and acute-stroke patients. Stroke, 32: 13041309. Amadio S, Panizza M, Pisano F, Madema L, MiscioC, Nilsson, J, Volonte, MA, Comi, G and Galardi, G

TRANS CRANIAL MAGNETIC STIMULATION

(2000) Transcranial magnetic stimulation and silent period in spasmodic torticollis. Am. J. Phys. Med. Rehabil., 79: 361-368. Amassian, VE, Stewart, M, Quirk, GJ and Rosenthal, JL (1987) Physiological basis of motor effects of a transient stimulus to cerebral cortex. Neurosurgery, 20: 74-93. Ashby, P and Zilm, D (1982) Relationship between EPSP shape and cross-correlation profile explored by computer simulation for studies on human motoneurons. Exp. Brain Res., 47: 33-40. Aurora, SK, al-Sayeed, F and Welch, KM (1999) The cortical silent period is shortened in migraine with aura. Cephalalgia, 19: 708-712. Awiszus, F and Feistner, H (1994) Correlations between size parameters and the amplitude of the excitatory postsynaptic potential evoked by magnetic brain stimulation in human hand muscle motoneurons, Exp. Brain Res., 98: 128-134. Awiszus, F and Feistner, H (1995) Comparison of single motor unit responses to transcranial magnetic and peroneal nerve stimulation in the tibialis anterior muscle of patients with amyotrophic lateral sclerosis. Electroencephalogr. Clin. Neurophysiol., 97: 90-95. Awiszus, F, Feistner, H, Urbach, D and Bostock, H (1999) Characterisation of paired-pulse transcranial magnetic stimulation conditions yielding inhibition or I-wave facilitation using a threshold-hunting paradigm. Exp. Brain Res., 129: 317-324. Baker, SN, Olivier, E and Lemon, RN (1995) Task-related variation in corticospinal output evoked by transcranial magnetic stimulation in the macaque monkey. J. Physiol. (Lond.), 488: 795-801. Baldissera, F and Leocani, L (1995) Afferent excitation of human motor cortex as revealed by enhancement of direct cortico-spinal actions on motoneurons. Electroencephalogr. Clin. Neurophysiol., 97: 394-401. Barker, AT (1999) The history and basic principles of magnetic nerve stimulation. Electroencephalogr. CUn. Neurophysiol., Supp!. 51: 3-21. Barker, AT, Freeston, IL, Jalinous, R, Merton, PA and Morton, HB (1985a) Magnetic stimulation of the human brain (abstract). J. Physiol. (Lond.), 369: 3P. Barker, AT, Jalinous, Rand Freeston, IL (l985b) Noninvasive magnetic stimulation of human motor cortex (letter). Lancet, I: 1106-1107. Bastings, EP, Gage, HD, Greenberg, JP, Hammond, G, Hernandez, L, Santago, P, Hamilton, CA, Moody, DM, Singh, KD, Ricci, PE, Pons, TP and Good, DC (1998) Co-registration of cortical magnetic stimulation and functional magnetic resonance imaging. Neuroreport, 9: 1941-1946. Benecke, R and Meyer, BU (1991) Magnetic stimulation of corticonuclear systems and of cranial nerves in man:

III

physiological basis and clinical application. Electroencephalogr. CUn. Neurophysiol., Supp!. 43: 333-343. Berardelli, A, Rona, S, Inghilleri, M and Manfredi, M (1996) Cortical inhibition in Parkinson's disease. A study with paired magnetic stimulation. Brain, 119: 71-77. Bertolasi, L, Priori, A, Tinazzi, M, Bertasi, V and Rothwell, JC (1998) Inhibitory action of forearm flexor muscle afferents on corticospinal outputs to antagonist muscles in humans. J. Physiol. (Lond.), 511: 947-956. Bohning, DE, Shastri, A, Nahas, Z, Lorberbaum, JP, Andersen, SW, Dannels, WR, Haxthausen, ED, Vincent, DJ and George, MS (1998) Echoplanar BOLD fMRI of brain activation induced by concurrent transcranial magnetic stimulation. Invest. Radiol., 33: 336-340. Bohning, DE, Shastri, A, McConnell, KA, Nahas, Z, Lorberbaum, JP, Roberts, DR, Teneback, C, Vincent, DJ and George, MS (1999) A combined TMS/fMRI study of intensity-dependent TMS over motor cortex. Bioi. Psychiatr., 45: 385-394. Bohning, DE, Shastri, A, Wassermann, EM, Ziemann, U, Lorberbaum, JP, Nahas, Z, Lomarev, MP and George, MS (2000) BOLD-f MRI response to single-pulse transcranial magnetic stimulation (TMS). J. Magn. Reson. Imaging, II: 569-574. Boniface, SJ, Mills, KR and Schubert, M (1991) Responses of single spinal motoneurons to magnetic brain stimulation in healthy subjects and patients with multiple sclerosis. Brain, 114: 643-662. Boroojerdi, B, Diefenbach, K and Ferbert, A (1996) Transcallosal inhibition in cortical and subcortical cerebral vascular lesions. J. Neurol. Sci., 144: 160170. Boroojerdi, B, Hungs, M, Mull, M, Topper, Rand Noth, J (1998) Interhemispheric inhibition in patients with multiple sclerosis. Electroencephalogr. Clin. Neurophysiol., 109: 230-237. Boroojerdi, B, Foltys, H, Krings, T, Spetzger, U, Thron, A and Topper, R (1999) Localization of the motor hand area using transcranial magnetic stimulation and functional magnetic resonance imaging. Clin. Neurophysiol., 110: 699-704. Boroojerdi, B, Prager, A, Muellbacher, W and Cohen, LG (2000) Reduction of human visual cortex excitability using I-Hz transcranial magnetic stimulation. Neurology, 54: 1529-1531. Boroojerdi, B, Battaglia, F, Muellbacher, W and Cohen, LG (2001) Mechanisms influencing stimulus-response properties of the human corticospinal system. CUn. Neurophysiol., 112: 931-937. Boylan, LS, Pullman, SL, Lisanby, SH, Spicknall, KE and Sackeim, HA (2001) Repetitive transcranial magnetic stimulation to SMA worsens complex movements in

112 Parkinson's disease. Clin. Neurophysiol., 112: 259-

264. Brasil-Neto, JP, McShane, LM, Fuhr, P, Hallett, M and Cohen, LG (1992) Topographic mapping of the human motor cortex with magnetic stimulation: factors affecting accuracy and reproducibility. Eleetroeneephalogr. Clin. Neurophysiol., 85: 9-16. Britton, TC, Meyer, BU, Herdmann, J and Benecke, R (1990) Clinical use of the magnetic stimulator in the investigation of peripheral conduction time. Muse. Nerve, 13: 396-406. Brodtmann, A, Macdonell, RAL, Gilligan, AK, Curatolo, J and Berkovic, SF (1999) Cortical excitability and recovery curve analysis in generalized epilepsy. Neurology, 53: 1347-1349. Brouwer, B and Ashby, P (1990) Corticospinal projections to upper and lower limb spinal motoneurons in man. Electroencephalogr. Clin. Neurophysiol., 76: 509-519. Brown, P, Ridding, MC, Werhahn, KJ, Rothwell, JC and Marsden, CD (1996) Abnormalities of the balance between inhibition and excitation in the motor cortex of patients with cortical myoclonus. Brain, 119: 309317. Buhler, R, Magistris, MR, Truffert, A, Hess, CW and Resler, KM (2001) The triple stimulation technique to study central motor conduction to the lower limbs. Clin. Neurophysiol., 112: 938-949. Byrnes, ML, Thickbroom, GW, Wilson, SA, Sacco, P, Shipman, JM, Stell, R and Mastaglia, FL (1998) The corticomotor representation of upper limb muscles in .writer's cramp and changes following botulinum toxin injection. Brain, 121: 977-988. Byrnes, ML, Thickbroom, GW, Phillips, BA, Wilson, SA and Mastaglia, FL (1999) Physiological studies of the corticomotor projection to the hand after subcortical stroke. Clin. Neurophysiol., 110: 487-498. Cantello, R, Gianelli, M, Civardi, C and Mutani, R (1992) Magnetic brain stimulation: the silent period after the motor evoked potential. Neurology, 42: 1951-1959. Cantello, R, Gianelli, M, Civardi, C and Mutani, R (1997) Focal subcortical reflex myoclonus. A clinical and neurophysiological study. Arch. Neurol., 54: 187-196. Cantello, R, Civardi, C, Cavalli, A, Varrasi, C and Vicentini, R (2000) Effects of a photic input on the human cortico-motoneuron connection. Clin. Neurophysiol., 111: 1981-1989. Cantello, R, Tarletti, Rand Civardi, C (2002) Transcranial magnetic stimulation and Parkinson's disease. Brain Res. Rev., 38: 309-327. Cararnia, MD, Desiato, MT, Cicinelli, P, lani, C and Rossini, PM (1993) Latency jump of "relaxed" vs. "contracted" motor evoked potentials as a marker of cortico-spinal maturation. Electroencephalogr. Clin. Neurophysiol., 89: 61-66.

V.ZIEMANN

Cararnia, MD, Palmieri, MG, Giacomini, P, Iani, C, Dally, Land Silvestrini, M (2000) Ipsilateral activation of the unaffected motor cortex in patients with hemiparetic stroke. CUn. Neurophysiol., 111: 1990-1996. Carr, U, Harrison, LM, Evans, AL and Stephens, JA (1993) Patterns of central motor reorganization in hemiplegic cerebral palsy. Brain, 116: 1223-1247. Carroll, TJ, Riek, S and Carson, RG (2001) Reliability of the input-output properties of the cortico-spinal pathway obtained from transcranial magnetic and electrical stimulation. J. Neurosci. Meth., 112: 193-202. Catano, A, Houa, M, Caroyer, JM, Ducame, H and Noel, P (1996) Magnetic transcranial stimulation in acute stroke: early excitation threshold and functional prognosis. Electroeneephalogr. Clin. Neurophysiol., 101: 233-239. Chen, R, Classen, J, Gerloff, C, Celnik, P, Wassermann, EM, Hallett, M and Cohen, LG (1997a) Depression of motor cortex excitability by low-frequency transcranial magnetic stimulation. Neurology, 48: 1398-1403. Chen, R, Sarnii, A, Canos, M, Wassermann, EM and Hallett, M (1997b) Effects of phenytoin on cortical excitability in humans. Neurology, 49: 881-883. Chen, R, Wassermann, EM, Canos, M and Hallett, M (1997c) Impaired inhibition in writer's cramp during voluntary muscle activation. Neurology, 49: 10541059. Chen, RC, Gerloff, C, Classen, J, Wassermann, EM, Hallett, M and Cohen, LG (1997d) Safety of different inter-train intervals for repetitive transcranial magnetic stimulation and recommendations for safe ranges of stimulation parameters. Electroencephalogr. Clin. Neurophysiol., 105: 415-421. Chen, R, Corwell, B, Yaseen, Z, Hallett, M and Cohen, LG (1998a) Mechanisms of cortical reorganization in lower-limb amputees. J. Neurosci., 18: 3443-3450. Chen, R, Tam, A, Butefisch, C, Ziemann, D, Rothwell, JC and Cohen, LG (1998b) Intracortical inhibition and facilitation in different representations of the human motor cortex. J. Neurophysiology, 80: 2870-2881. Chen, R, Lozano, AM and Ashby, P (1999) Mechanism of the silent period following transcranial magnetic stimulation. Exp. Brain Res., 128: 539-542. Chistyakov, AV, Soustiel, JF, Hafner, H, Trubnik, M, Levy, G and Feinsod, M (2001) Excitatory and inhibitory corticospinal responses to transcranial magnetic stimulation in patients with minor to moderate head injury. J. Neurol. Neurosurg. Psychiatry, 70: 580-587. Chokroverty, S, Picone, MA and Chokroverty, M (1991) Percutaneous magnetic coil stimulation of human cervical vertebral column: site of stimulation and clinical application. Electroencephalogr. Clin. Neurophysiol., 81: 359-365.

TRANS CRANIAL MAGNETIC STIMULATION

Chokroverty, S, Flynn, D, Picone, MA, Chokroverty, M and Belsh, J (1993) Magnetic coil stimulation of the human lumbosacral vertebral column: site of stimulation and clinical application. Electroencephalogr. Clin. Neurophysiol., 89: 54-60. Cicinelli, P, Traversa, R, Bassi, A, Scivoletto, G and Rossini, PM (1997) Interhemispheric differences of hand muscle representation in human motor cortex. Muse. Nerve, 20: 535-542. Cincotta, M, Lori, S, Gangemi, PF, Barontini, F and Ragazzoni, A (1996) Hand motor cortex activation in a patient with congenital mirror movements: a study of the silent period following focal transcranial magnetic stimulation. Electroencephalogr. Clin. Neurophysiol., 101: 240-246. Cincotta, M, Borgheresi, A, Lori, S, Fabbri, M and Zaccara, G (1998) Interictal inhibitory mechanisms in patients with cryptogenic motor cortex epilepsy: a study of the silent period following transcranial magnetic stimulation. Electroencephalogr. Clin. Neurophysiol., 107: 1-7. Cincotta, M, Borgheresi, A, Liotta, P, Montigiani, A, Marin, E, Zaccara, G and Ziemann, D (2000) Reorganization of the motor cortex in a patient with congenital hemiparesis and mirror movements. Neurology, 55: 129-131. Civardi, C, Cantello, R, Asselman, P and Rothwell, JC (200l) Transcranial magnetic stimulation can be used to test connections to primary motor areas from frontal and medial cortex in humans. Neuroimage, 14: 1444-1453. Classen, J, Witte, OW, Schlaug, G, Seitz, RJ, Holthausen, H and Benecke, R (1995) Epileptic seizures triggered directly by focal transcranial magnetic stimulation. Electroencephalogr. Clin. Neurophysiol., 94: 19-25. Classen, J, Schnitzler, A, Binkofski, F, Werhahn, KJ, Kim, Y-S, Kessler, K and Benecke, R (1997) The motor syndrome associated with exaggerated inhibition within the primary motor cortex of patients with hemiparetic stroke. Brain, 120: 605-619. Classen, J, Knorr, D, Werhahn, KJ, Schlaug, G, Kunesch, E, Cohen, LG, Seitz, RJ and Benecke, R (1998) Multimodal output mapping of human central motor representation on different spatial scales. J. Physiol. (Lond.), 512: 163-179. Classen, J, Steinfelder, B, Liepert, J, Stefan, K, Celnik, P, Cohen, LG, Hess, A, Kunesch, E, Chen, R, Benecke, R and Hallett, M (2000) Cutaneomotor integration in humans is somatotopically organized at various levels of the nervous system and is task dependent. Exp. Brain Res., 130: 48-59. Claus, D (1990) Central motor conduction: method and normal results. Muse. Nerve, 13: 1125-1132.

113 Claus, D and Brunholzl, C (1994) Facilitation and disfacilitation of muscle responses after repetitive transcranial cortical stimulation and electrical peripheral nerve stimulation. Electroencephalogr. Clin. Neurophysiol., 93: 417--420. Claus, D, Weis, M, Jahnke, D, P1ewe, A and Brunholzl, C (1992) Corticospinal conduction studied with magnetic double stimulation in the intact human. J. Neurol. Sci., 111: 18D-188. Clouston, PD, Kiers, L, Menkes, D, Sander, H, Chiappa, K and Cros, D (1995) Modulation of motor activity by cutaneous input: inhibition of the magnetic motor evoked potential by digital electrical stimulation. Electroencephalogr. Clin. Neurophysiol., 97: 114-125. Cohen, LG, Bandinelli, S, Findley, TW and Hallett, M (1991a) Motor reorganization after upper limb amputation in man. A study with focal magnetic stimulation. Brain, 114: 615-627. Cohen, LG, Bandinelli, S, Sato, S, Kufta, C and Hallett, M (1991b) Attenuation in detection of somatosensory stimuli by transcranial magnetic stimulation. Electroencephalogr. Clin. Neurophysiol., 81: 366-376. Cohen, LG, Meer, J, Tarkka, I, Biemer, S, Leiderman, DB, Dubinsky, RM, Sanes, IN, Jabbari, B, Branscum, B and Hallett, M (1991c) Congenital mirror movements. Abnormal organization of motor pathways in two patients. Brain, 114: 381--403. Cohen, LG, Celnik, P, Pascual-Leone, A, Corwell, B, Faiz, L, Dambrosia, J, Honda, M, Sadato, N, Gerloff, C, Catala, MD and Hallett, M (1997) Functional relevance of cross-modal plasticity in the blind. Nature, 389: 18D-183. Corthout, E, Barker, AT and Cowey, A (2001) Transcranial magnetic stimulation. Which part of the current waveform causes the stimulation? Exp. Brain Res., 141: 128-132. Cowan, JM, Day, BL, Marsden, C and Rothwell, JC (1986) The effect of percutaneous motor cortex stimulation on H-reflexes in muscles of the arm and leg in intact man. J. Physiol. (Lond.), 377: 333-347. Cracco, RQ, Amassian, VE, Maccabee, PJ and Cracco, JB (1989) Comparison of human transcallosal responses evoked by magnetic coil and electrical stimulation. Electroencephalogr. Clin. Neurophysiol., 74: 417--424. Daska1akis, ZJ, Christensen, BK, Chen, R, Fitzgerald, PB, Zipursky, RB and Kapur, S (2002) Evidence for impaired cortical inhibition in schizophrenia using transcranial magnetic stimulation. Arch. Gen. Psychiatry, 59: 347-354. Davey, NJ, Romaiguere, P, Maskill, DW and Ellaway, PH (1994) Suppression of voluntary motor activity revealed using trans cranial magnetic stimulation of the motor cortex in man. J. Physiol. (Lond.), 477: 223-235.

114 Day, BL, Dressler, D, Maertens de Noordhout, A, Marsden, CD, Nakashima, K, Rothwell, JC and Thompson, PD (1989) Electric and magnetic stimulation of human motor cortex: surface EMG and single motor unit responses. J. Physiol. (Lond.), 412: 449-473. Day, BL, Riescher, H, Struppler, A, Rothwell, JC and Marsden, CD (1991) Changes in the response to magnetic and electrical stimulation of the motor cortex following muscle stretch in man. J. Physiol. (Lond.), 433: 41-57. Deuschl, G, Michels, R, Berardelli, A, Schenck, E, Inghilleri, M and Lucking, CH (1991) Effects of electric and magnetic transcranial stimulation on long latency reflexes. Exp. Brain Res., 83: 403-410. Devanne, H, Lavoie, BA and Capaday, C (1997) Inputoutput properties and gain changes in the human corticospinal pathway. Exp. Brain Res., 114: 329-338. Di Lazzaro, V, Molinari, M, Restuccia, D, Leggio, MG, Nardone, R, Fogli, D and Tonali, P (1994) Cerebrocerebellar interactions in man: neurophysiological studies in patients with focal cerebellar lesions. Electroencephalogr. Clin. Neurophysiol., 93: 27-34. Di Lazzaro, V,Oliviero, A, Profice, P, Saturno, E, Pilato, F, Insola, A, Mazzone, P, Tonali, P and Rothwell, JC (1998a) Comparison of descending volleys evoked by transcranial magnetic and electric stimulation in conscious humans. Electroencephalogr. Clin. Neurophysiol., 109: 397-401. Di Lazzaro, V, Restuccia, D, Oliviero, A, Profice, P, Ferrara, L, Insola, A, Mazzone, P, Tonali, P and Rothwell, JC (1998b) Effects of voluntary contraction on descending volleys evoked by transcranial magnetic stimulation in conscious humans. J. Physiol., 508: 625-633. Di Lazzaro, V, Restuccia, D, Oliviero, A, Profice, P, Ferrara, L, Insola, A, Mazzone, P, Tonali, P and Rothwell, JC (1998c) Magnetic transcranial stimulation at intensities below active motor threshold activates intracortical inhibitory circuits. Exp. Brain Res., 119: 265-268. Di Lazzaro, V,Oliviero, A, Profice, P, Insola, A, Mazzone, P, Tonali, P and Rothwell, JC (1999a) Direct demonstration of interhemispheric inhibition of the human motor cortex produced by transcranial magnetic stimulation. Exp. Brain Res., 124: 520-524. Di Lazzaro, V, Oliviero, A, Profice, P, Insola, A, Mazzone, P, Tonali, P and Rothwell, JC (1999b) Direct recordings of descending volleys after transcranial magnetic and electric motor cortex stimulation in conscious humans. Electroencephalogr. Clin. Neurophysiol., Supp!. 51: 120-126.

U.ZIEMANN

Di Lazzaro, V, Rothwell, JC, Oliviero, A, Profice, P, Insola, A, Mazzone, P and Tonali, P (1999c) Intracortical origin of the short latency facilitation produced by pairs of threshold magnetic stimuli applied to human motor cortex. Exp. Brain Res., 129: 494-499. Di Lazzaro, V, Oliviero, A, Profice, P, Pennisi, MA, Di Giovanni, S, Zito, G, Tonali, P and Rothwell, JC (2000) Muscarinic receptor blockade has differential effects on the excitability of intracortical circuits in human motor cortex. Exp. Brain Res., 135: 455-461. Ellaway, PH, Davey, NJ, Maskill, DW, Rawlinson, SR, Lewis, HS and Anissirnova, NP (1998) Variability in the amplitude of skeletal muscle responses to magnetic stimulation of the motor cortex in man. Electroencephalogr. Clin. Neurophysiol., 109: 104-113. Enomoto, H, Ugawa, Y, Hanajima, R, Yuasa, K, Mochizuki, H, Terao, Y, Shiio, Y, Furubayashi, T, Iwata, NK and Kanazawa, I (2001) Decreased sensory cortical excitability after 1 Hz rTMS over the ipsilateral primary motor cortex. Clin. Neurophysiol., 112: 21542158. Epstein, CM, Schwartzberg, DG, Davey, KR and Sudderth, DB (1990) Localizing the site of magnetic brain stimulation in humans. Neurology, 40: 666-670. Epstein, CM, Fernandez-Beer, E, Weissman, JD and Matsuura, S (1991) Cervical magnetic stimulation: the role of the neural foramen. Neurology, 41: 677-680. Epstein, CM, Lah, Jl, Meador, K, Weissman, JD, Gaitan, LE and Dihenia, B (1996) Optimum stimulus parameters for lateralized suppression of speech with magnetic brain stimulation. Neurology, 47: 1590-1593. Epstein, CM, Meador, KJ, Loring, DW, Wright, RJ, Weissman, JD, Sheppard, S, Lab, n, Puhalovich, F, Gaitan, L and Davey, KR (1999) Localization and characterization of speech arrest during transcranial magnetic stimulation. Clin. Neurophysiol., 110: 10731079. Evans, BA, Daube, JR and Litchy, WJ (1990) A comparison of magnetic and electrical stimulation of spinal nerves. Muse. Nerve, 13: 414-420. Evers, S, Hengst, K and Pecuch, PW (2001) The impact of repetitive transcranial magnetic stimulation on pituitary hormone levels and cortisol in healthy subjects. J. Affect. Disord., 66: 83-88. Eyre, JA and Miller, S (1992) Assessment of motor pathways. In: JA Eyre (Ed.), The Neurophysiological Examination of the Newborn Infant. Oxford: Mac Keith Press, pp. 124-140. Eyre, JA, Taylor, JP, Villagra, F, Smith, M and Miller, S (2001) Evidence of activity-dependent withdrawal of corticospinal projections during human development. Neurology, 57: 1543-1554.

TRANS CRANIAL MAGNETIC STIMULATION

Fanner, SF, Ingram, DA and Stephens, JA (1990) Mirror movements studied in a patient with Klippel-Feil syndrome. J. Physiol. (Lond.), 428: 467-484. Fanner, SF, Harrison, LM, Ingram, DA and Stephens, JA (1991) Plasticity of central motor pathways in children with hemiplegic cerebral palsy. Neurology, 41: 1505-1510. Ferbert, A, Caramia, D, Priori, A, Bertolasi, Land Rothwell, JC (l992a) Cortical projection to erector spinae muscles in man as assessed by focal transcranial magnetic stimulation. Electroencephalogr. Clin. Neurophysiol., 85: 382-387. Ferbert, A, Priori, A, Rothwell, JC, Day, BL, Colebatch, JG and Marsden, CD (l992b) Interhemispheric inhibition of the human motor cortex. J. Physiol. (Lond.), 453: 525-546. Fetz, EE and Gustafsson, B (1983) Relation between shapes of post-synaptic potentials and changes in firing probability of cat motoneurons. J. Physiol. (Lond.), 341: 387-410. Filipovic, SR, Ljubisavljevic, M, Svetel, M, Milanovic, S, Kacar, A and Kostic, VS (1997) Impairment of cortical inhibition in writer's cramp as revealed by changes in electromyographic silent period after transcranial magnetic stimulation. Neurosci. Lett., 222: 167-170. Fisher, RJ, Nakamura, Y, Bestmann, S, Rothwell, JC and Bostock, H (2002) Two phases of intracortical inhibition revealed by transcranial magnetic threshold tracking. Exp. Brain Res., 143: 240-248. Foerster, A, Schmitz, JM, Nouri, S and Claus, D (1997) Safety of rapid-rate transcranial magnetic stimulation: heart rate and blood pressure changes. Electroencephalogr. Clin. Neurophysiol., 104: 207-212. Fox, P, Ingham, R, George, MS, Mayberg, H, Ingham, J, Roby, J, Martin, C and Jerabekrn, P (1997) Imaging human intra-cerebral connectivity by PET during TMS. Neuroreport, 8: 2787-2791. Fries, W, Danek, A and Witt, TN (1991) Motor responses after transcranial electrical stimulation of cerebral hemispheres with a degenerated pyramidal tract. Ann. Neurol., 29: 646-650. Fritz, C, Braune, HJ, Pylatiuk, C and Pohl, M (1997) Silent period following transcranial magnetic stimulation: a study of intra- and inter-examiner reliability. Electroencephalogr. Clin. Neurophysiol., 105: 235240. Fuhr, P, Agostino, R and Hallett, M (1991) Spinal motor neuron excitability during the silent period after cortical stimulation. Electroencephalogr. Clin. Neurophysiol., 81: 257-262. Furubayashi, T, Ugawa, Y, Terao, Y, Hanajima, R, Sakai, K, Machii, K, Mochizuki, H, Shiio, Y, Uesugi, H, Enomoto, Hand Kanazawa, I (2000) The human hand

115 motor area is transiently suppressed by an unexpected auditory stimulus. Clin. Neurophysiol., Ill: 178-183. Garvey, MA, Ziemann, U, Becker, DA, Barker, CA and Bartko, 11 (2001) New graphical method to measure silent periods evoked by transcranial magnetic stimulation. Clin. Neurophysiol., 112: 1451-1460. Gerloff, C, Cohen, LG, Floeter, MK, Chen, R, Corwell, B and Hallett, M (1998) Inhibitory influence of the ipsilateral motor cortex on responses evoked by transcranial stimulation of the human cortex and pyramidal tract. J. Physiol., 510: 249-259. Gerschlager, W, Siebner, HR and Rothwell, JC (2001) Decreased corticospinal excitability after subthreshold I Hz rTMS over lateral premotor cortex. Neurology, 57: 449-455. Gould, HJd, Cusick, CG, Pons, TP and Kaas, JH (1986) The relationship of corpus callosum connections to electrical stimulation maps of motor, supplementary motor, and the frontal eye fields in owl monkeys. J. Comp. Neurol., 247: 297-325. Greenberg, BD, Ziemann, U, Cora-Locatelli, G, Hannon, A, Murphy, DL, Keel, JC and Wassermann, EM (2000) Altered cortical excitability in obsessive-compulsive disorder. Neurology, 54: 142-147. Gugino, LD, Romero, JR, Aglio, L, Titone, D, Ramirez, M, Pascual-Leone, A, Grimson, E, Weisenfeld, N, Kikinis, R and Shenton, ME (200 I) Transcranial magnetic stimulation coregistered with MRI: a comparison of a guided vs. blind stimulation technique and its effect on evoked compound muscle action potentials. Clin. Neurophysiol., 112: 1781-1792. Hallett, M, Wassermann, EW, Cohen, LG, Chmielowska, J and Gerloff, C (1998) Cortical mechanisms of recovery of function after stroke. Neurorehabilitation, 10: 131-142. Hallett, M, Chen, R, Ziemann, U and Cohen, LG (1999) Reorganization in motor cortex in amputees and in normal volunteers after ischemic limb deafferentation. Electroencephalogr. Clin. Neurophysiol., Suppl. 51: 183-187. Hamdy, S, Aziz, Q, Rothwell, JC, Singh, KD, Barlow, J, Hughes, DG, Tallis, RC and Thompson, DG (1996) The cortical topography of human swallowing musculature in health and disease. Nature Med., 2: 1217-1224. Hanajima, R, Ugawa, Y, Machii, K, Mochizuki, H, Terao, Y, Enomoto, H, Furubayashi, T, Shiio, Y, Uesugi, H and Kanazawa, I (200Ia) Interhemispheric facilitation of the hand motor area in humans. J. Physiol., 531: 849-859. Hanajima, R, Ugawa, Y, Okabe, S, Yuasa, K, Shiio, Y, Iwata, NK and Kanazawa, I (200 Ib) Interhemispheric interaction between the hand motor areas in patients

116 with cortical myoclonus. Clin. Neurophysiol., 112: 623-626. Hanajima, R, Ugawa, Y, Terao, Y, Enomoto, H, Shiio, Y, Mochizuki, H, Furubayashi, T, Uesugi, H, Iwata, NK and Kanazawa, I (2002) Mechanisms of intracortical 1wave facilitation elicited with paired-pulse magnetic stimulation in humans. J. Physiol., 538: 253-261. Haug, BA, Schonle, PW, Knobloch, C and Kohne, M (1992) Silent period measurement revives as a valuable diagnostic tool with transcranial magnetic stimulation. Electroencephalogr. Clin. Neurophysiol., 85: 158-160. Heinen, F, Glocker, FX, Fietzek, U, Meyer, B-U, LUcking, C-H and Korinthenberg, R (1998) Absence of transcallosal inhibition following focal magnetic stimulation in preschool children. Ann. Neurol., 43: 608-612. Hess, CW, Mills, KR and Murray, NM (1986) Magnetic stimulation of the human brain: facilitation of motor responses by voluntary contraction of ipsilateral and contralateral muscles with additional observations on an amputee. Neurosci. Lett., 71: 235-240. Hess, CW, Mills, KR and Murray, NM (l987a) Responses in small hand muscles from magnetic stimulation of the human brain. J. Physiol. (Lond.), 388: 397-419. Hess, CW, Mills, KR, Murray, NM and Schriefer, TN (l987b) Magnetic brain stimulation: central motor conduction studies in multiple sclerosis. Ann. Neurol., 22: 744-752. Ho, K-H, Lee, M, Nithi, K, Palace, J and Mills, K (1999) Changes in motor evoked potentials to short-interval paired transcranial magnetic stimuli in multiple sclerosis. Clin. Neurophysiol., 110: 712-719. Homberg, V, Stephan, KM and Netz, J (1991) Transcranial stimulation of motor cortex in upper motor neuron syndrome: its relation to the motor deficit. Electroencephalogr. Clin. Neurophysiol., 81: 377-388. Hoppner, J, Kunesch, E, Buchmann, J, Hess, A, Grossmann, A and Benecke, R (1999) Demyelination and axonal degeneration in corpus callosum assessed by analysis of transcallosally mediated inhibition in multiple sclerosis. Clin. Neurophysiol., 110: 748-756. Hoppner, J, Kunesch, E, Grossmann, A, Tolzin, CJ, Schulz, M, Schlafke, D and Ernst, K (2001) Dysfunction of transcallosally mediated motor inhibition and callosal morphology in patients with schizophrenia. Acta Psychiatr. Scand., 104: 227-235. Hoshiyama, M and Kakigi, R (1999) Shortening of the cortical silent period following transcranial magnetic brain stimulation during an experimental paradigm for generating contingent negative variation (CNV). Clin. Neurophysiol., 110: 1394-1398. Ikoma, K, Samii, A, Mercuri, B, Wassermann, EM and Hallett, M (1996) Abnormal cortical motor excitability in dystonia. Neurology, 46: 1371-1376.

V.ZIEMANN

llmoniemi, RJ, Virtanen, J, Ruohonen, J, Karhu, J, Aronen, HJ, Natanen, Rand Katila, T (1997) Neuronal responses to magnetic stimulation reveal cortical reactivity and connectivity. NeuroReport, 8: 3537-3540. Ilmoniemi, RJ, Ruohonen, J, Virtanen, J, Aronen, HJ and Karhu, J (1999) EEG responses evoked by transcranial magnetic stimulation. Electroencephalogr. Clin. Neurophysiol., Suppl. 51: 22-29. Inghilleri, M. Berardelli, A, Cruccu, G and Manfredi, M (1993) Silent period evoked by transcranial stimulation of the human cortex and cervicomedullary junction. J. Physiol. (Lond.), 466: 521-534. Inghilleri, M, Mattia, D, Berardelli, A and Manfredi, M (1998) Asymmetry of cortical excitability revealed by transcranial stimulation in a patient with focal motor epilepsy and cortical myoclonus. Electroencephalogr. Clin. Neurophysiol., 109: 70-72. Jahanshahi, M, Ridding, MC, Limousin, P, Profice, P, Fogel, W, Dressler, D, Fuller, R, Brown, RG, Brown, P and Rothwell, JC (1997) Rapid rate transcranial magnetic stimulation - a safety study. Electroencephalogr. Clin. Neurophysiol., 105: 422-429. Jennum, P, Friberg, L, Fuglsang-Frederiksen, A and Dam, M (1994) Speech localization using repetitive transcranial magnetic stimulation. Neurology, 44: 269273. Jing, H and Takigawa, M (2000) Observation of EEG coherence after repetitive transcranial magnetic stimulation. Clin. Neurophysiol., Ill: 1620-1631. Kaelin-Lang, A and Cohen, LG (2000) Enhancing the quality of studies using transcranial magnetic and electrical stimulation with a new computer-controlled system. J. Neurosci. Methods, 102: 81-89. Kammer, T, Beck, S, Thielscher, A, Laubis-Herrmann, U and Topka, H (2001) Motor thresholds in humans: a transcranial magnetic stimulation study comparing different pulse waveforms, current directions and stimulator types. Clin. Neurophysiol., 112: 250-258. Kaneko, K, Kawai, S, Fuchigami, Y, Morita, Hand Ofuji, A (l996a) The effect of current direction induced by transcranial magnetic stimulation on the corticospinal excitability in human brain. Electroencephalogr. Clin. Neurophysiol., 101: 478-482. Kaneko, K, Kawai, S, Fuchigarni, Y, Shiraishi, G and Ito, T (l996b) Effect of stimulus intensity and voluntary contraction on corticospinal potentials following transcranial magnetic stimulation. J. Neurol. Sci., 139: 131-136. Kaneko, K, Kawai, S, Fuchigami, Y, Shiraishi, G and Ito, T (1996c) Intracortical facilitation of the muscle response after transcranial magnetic double stimulation. Muse. Nerve, 19: 1043-1045.

TRANSCRANIAL MAGNETIC STIMULATION

Karl, A, Birbaumer, N, Lutzenberger, W, Cohen, LG and Flor, H (2001) Reorganization of motor and somatosensory cortex in upper extremity amputees with phantom limb pain. J. Neurosci., 21: 3609-3618. Kew, JJ, Ridding, MC, Rothwell, JC, Passingham, RE, Leigh, PN, Sooriakumaran, S, Frackowiak, RS and Brooks, DJ (1994) Reorganization of cortical blood flow and transcranial magnetic stimulation maps in human subjects after upper limb amputation. J. Neurophysiol., 72: 2517-2524. Kiers, L, Cros, D, Chiappa, KH and Fang, J (1993) Variability of motor potentials evoked by transcranial magnetic stimulation. Electroencephalogr. Clin. Neurophysiol., 89: 415-423. Kiers, L, Clouston, P, Chiappa, KH and Cros, D (1995) Assessment of cortical motor output: compound muscle action potential vs. twitch force recording. Electroencephalogr. cu« Neurophysiol., 97: 131-139. Kischka, U, Fajfr, R, Fellenberg, T and Hess, CW (1993) Facilitation of motor evoked potentials from magnetic brain stimulation in man: a comparative study of different target muscles. J. Clin. Neurophysiol., 10: 505-512. Kleine, BU, Praamstra, P, Stegeman, DF and Zwarts, MJ (2001) Impaired motor cortical inhibition in Parkinson's disease: motor unit responses to transcranial magnetic stimulation. Exp. Brain Res., 138: 477-483. Kloten, H, Meyer, BU, Britton, TC and Benecke, R (1992) Normal values and age-related changes in magnetoelectric evoked compound muscle potentials. EEG EMG Z Elektroenzephalogr Elektromyogr Verwandte Geb, 23: 29-36. Kofler, M, Glocker, FX, Leis, AA, Seifert, C, Wissel, J, Kronenberg, MF and Fuhr, P (1998) Modulation of upper extremity motoneuron excitability following noxious finger tip stimulation in man: a study with transcranial magnetic stimulation. Neurosci. Lett., 246: 97-100. Kofler, M, Fuhr, P, Leis, AA, Glocker, FX, Kronenberg, MF, Wissel, J and Stetkarova, I (2001) Modulation of upper extremity motor evoked potentials by cutaneous afferents in humans. Clin. Neurophysiol., 112: 10531063. Komssi, S, Aronen, HJ, Huttunen, J, Kesaniemi, M, Soinne, L, Nikouline, VV, Ollikainen, M, Roine, RO, Karhu, J, Savolainen, S and Ilmoniemi, RJ (2002) Ipsiand contralateral EEG reactions to transcranial magnetic stimulation. Clin. Neurophysiol., 113: 175-184. Kraus, KH, Gugino, LD, Levy, WJ, Cadwell, J and Roth, BJ (1993) The use of a cap-shaped coil for transcranial magnetic stimulation of the motor cortex. J. Clin. Neurophysiol., 10: 353-362. Krings, T, Buchbinder, BR, Butler, WE, Chiappa, KH, Jiang, HJ, Cosgrove, GR and Rosen, BR (1997)

II7 Functional magnetic resonance imaging and transcranial magnetic stimulation: Complementary approaches in the evaluation of cortical motor function. Neurology, 48: 1406-1416. Krings, T, Naujokat, C and Graf v. Keyserlingk, D (1998) Representation of cortical motor function and revealed by stereotactic transcranial magnetic stimulation. Electroencephalogr. Clin. Neurophysiol., 109: 85-93. Kujirai, T, Caramia, MD, Rothwell, JC, Day, BL, Thompson, PD, Ferbert, A, Wroe, S, Asselman, P and Marsden, CD (l993a) Corticocortical inhibition in human motor cortex. J. Physiol. (Lond.}, 471: 501519. Kujirai, T, Sato, M, Rothwell, JC and Cohen, LG (1993b) The effect of transcranial magnetic stimulation on median nerve somatosensory evoked potentials. Electroencephalogr. cu« Neurophysiol., 89: 227-234. Lefaucheur, JP, Drouot, X, Keravel, Y and Nguyen, JP (200la) Pain relief induced by repetitive transcranial magnetic stimulation of precentral cortex. NeuroReport, 12: 2963-2965. Lefaucheur, JP, Drouot, X and Nguyen, JP (200Ib) Interventional neurophysiology for pain control: duration of pain relief following repetitive transcranial magnetic stimulation of the motor cortex. Neurophysiol. Clin., 31: 247-252. Lemon, RN, Johansson, RS and Westling, G (1995) Corticospinal control during reach, grasp, and precision lift in man. J. Neurosci., IS: 6145-6156. Levy, WJ Jr., Amassian, VE, Traad, M amd Cadwell, J (1990) Focal magnetic coil stimulation reveals motor cortical system reorganized in humans after traumatic quadriplegia. Brain Res., 510: 130-134. Levy, WJ, Amassian, VE, Schmid, UD and Jungreis, C (1991) Mapping of motor cortex gyral sites noninvasively by transcranial magnetic stimulation in normal subjects and patients. Electroencephalogr. Clin. Neurophysiol., Suppl. 43: 51-75. Liepert, J, Miltner, WHR, Bauder, H, Sommer, M, Dettmers, C and Weiller C (1998) Motor cortex plasticity during constraint-induced movement therapy in stroke patients. Neurosci. Lett., 250: 5-8. Liepert, J, Oreja-Guevara, C, Cohen, LG, Tegenthoff, M, Hallett, M and Malin, J-P (l999a) Plasticity of cortical hand muscle representation in patients with hemifacial spasm. Neurosci. Lett., 272: 33-36. Liepert, J, Terborg, C and Weiller, C (1999b) Motor plasticity induced by synchronized thumb and foot movements. Exp. Brain Res., 125: 435-439. Liepert, J, Schardt, Sand Weiller, C (2001) Orally administered atropine enhances motor cortex excitability: a transcranial magnetic stimulation study in human subjects. Neurosci. Lett., 300: 149-152.

118 Linden, D and Berlit, P (1994) Magnetic motor evoked potentials (MEP) in diseases of the spinal cord. Acta Neurol. Scand., 90: 348-353. Lisanby, SH and Sackheim, HA (2000) TMS in major depression. In: MS George and RH Belmaker (Eds.), Transcranial Magnetic Stimulation in Neuropsychiatry. Washington, D.C.: American Psychiatric Press, pp. 185-200. Maccabee, PJ, Amassian, VE, Eberle, LP, Rudell, AP, Cracco, RQ, Lai, KS and Somasundarum, M (1991) Measurement of the electric field induced into inhomogeneous volume conductors by magnetic coils: application to human spinal neurogeometry. Electroencephalogr. Clin. Neurophysiol., 81: 224--237. Maccabee, PJ, Lipitz, ME, Desudchit, T, Golub, RW,Nitti, VW, Bania, JP, Willer, JA, Cracco, RQ, Cadwell, J, Hotson, GC, Eberle, LP and Amassian, VE (1996) A new method using neuromagnetic stimulation to measure conduction time within the cauda equina. Electroencephalogr. Clin. Neurophysiol., 10I: 153-166. Maccabee, PJ, Nagarajan, SS, Amassian, VE, Durand, DM, Szabo, AZ, Ahad, AB, Cracco, RQ, Lai, KS and Eberle, LP (1998) Influence of pulse sequence, polarity and amplitude on magnetic stimulation of human and porcine peripheral nerve. J. Physiol. (Lond.), 513: 571-585. Macdonell, RA, Shapiro, BE, Chiappa, KH, Helmers, SL, Cros, D, Day, BJ and Shahani, BT (1991) Hemispheric threshold differences for motor evoked potentials produced by magnetic coil stimulation. Neurology, 41: 1441-1444. Macdonell, RA, King, MA, Newton, MR, Curatolo, JM, Reutens, DC and Berkovic, SF (2001) Prolonged cortical silent period after transcranial magnetic stimulation in generalized epilepsy. Neurology, 57: 706-708. Maeda, F, Keenan, JP,Tormos, 1M, Topka, Hand PascualLeone, A (2000) Modulation of corticospinal excitability by repetitive transcranial magnetic stimulation. Clin. Neurophysiol., 111: 800-805. Maertens de Noordhout, A, Pepin, JL, Gerard, P and Delwaide, PJ (l992a) Facilitation of responses to motor cortex stimulation: effects of isometric voluntary contraction. Ann. Neurol., 32: 365-370. Maertens de Noordhout, A, Rothwell, JC, Day, BL, Dressler, D, Nakashima, K, Thompson, PD and Marsden, CD (1992b) Effect of digital nerve stimuli on responses to electrical or magnetic stimulation of the human brain. J. Physiol. (Lond.), 447: 535-548. Maertens de Noordhout, A, Rapisarda, G, Bogacz, D, P Gr, De Pasqua, V, Pennisi, G and Delwaide, PJ (1999) Corticomotoneuronal synaptic connections in normal

V.ZIEMANN

man: An electrophysiological study. Brain, 122: 1327-1340. Magistris, MR, Rosier, KM, Truffert, A and Myers, JP (1998) Transcranial stimulation excites virtually all motor neurons supplying the target muscle. A demonstration and a method improving the study of motor evoked potentials. Brain, 121: 437-450. Magistris, MR, Rosler, KM, Truffert, A, Landis, T and Hess, CW (1999) A clinical study of motor evoked potentials using a triple stimulation technique. Brain, 122: 265-279. Manganotti, P, Zanette, G, Bonato, C, Tinazzi, M, Polo, A and Fiaschi, A (1997) Crossed and direct effects of digital nerves stimulation on motor evoked potential: a study with magnetic brain stimulation. Electroencephalogr. Clin. Neurophysiol., 105: 280-289. Manganotti, P, Bongiovanni, LG, Zanette, G and Fiaschi, A (2000) Early and late intracortical inhibition in juvenile myoclonic epilepsy. Epilepsia, 41: 11291138. Mariorenzi, R, Zarola, F, Caramia, MD, Paradiso, C and Rossini, PM (1991) Non-invasive evaluation of central motor tract excitability changes following peripheral nerve stimulation in healthy humans. Electroencephalogr. Clin. Neurophysiol., 81: 90-10 1. Mathis, J, de Quervain, D and Hess, CW (1998) Dependence of the transcranially induced silent period on the 'instruction set' and the individual reaction time. Electroencephalogr. Clin. Neurophysiol., 109: 426435. Mathis, J, de Quervain, D and Hess, CW (1999) Taskdependent effects on motor-evoked potentials and on the following silent period. J. Clin. Neurophysiol., 16: 556-565. Matsunaga, K, Uozumi, T, Hashimoto, T and Tsuji, S (2001) Cerebellar stimulation in acute cerebellar ataxia. Clin. Neurophysiol., 112: 619-622. Mavroudakis, N, Caroyer, 1M, Brunko, E and Zegers de Beyl, D (1994) Effects of diphenylhydantoin on motor potentials evoked with magnetic stimulation. Electroencephalogr. Clin. Neurophysiol., 93: 428-433. Mayston, MJ, Harrison, LM and Stephens, JA (1999) A neurophysiological study of mirror movements in adults and children. Ann. Neurol., 45: 583-594. McKay, WB, Stokic, DS and Dimitrijevic, MR (1997) Assessment of corticospinal function in spinal cord injury using transcranial motor cortex stimulation: a review. J. Neurotrauma, 14: 539-548. Menkes, DL and Gruenthal, M (2000) Slow-frequency repetitive transcranial magnetic stimulation in a patient with focal cortical dysplasia. Epilepsia, 41: 240-242. Merton, PA and Morton, HB (1980) Stimulation of the cerebral cortex in the intact human subject. Nature, 285: 227.

TRANSCRANIAL MAGNETIC STIMULATION

Merton, PA, Hill, DK, Morton, HB and Marsden, CD (1982) Scope of a technique for electrical stimulation of human brain, spinal cord, and muscle. Lancet, 2: 597-600. Metman, LV, Bellevich, JS, Jones, SM, Barber, MD and Streletz, U (1993) Topographic mapping of human motor cortex with transcranial magnetic stimulation: Homunculus revisited. Brain Topogr., 6: 13-19. Meyer, BU, Roricht, S, Grafin Von Einsiedel, H, Kruggel, F and Weindl, A (1995) Inhibitory and excitatory interhemispheric transfers between motor cortical areas in normal humans and patients with abnormalities of the corpus callosum. Brain, 118: 429-440. Meyer, B-U, Roricht, S and Woiciechowsky, C (1998) Topography of fibers in the human corpus callosum mediating interhemispheric inhibition between the motor cortices. Ann. Neurol., 43: 360-369. Michelucci, R, Valzania, F, Passarelli, D, Santangelo, M, Rizzi, R, Buzzi, AM, Tempestini, A and Tassinari, CA (1994) Rapid-rate transcranial magnetic stimulation and hemispheric language dominance: usefulness and safety in epilepsy. Neurology, 44: 1697-1700. Mills, KR (1991) Magnetic brain stimulation: a tool to explore the action of the motor cortex on single human spinal motoneurons. Trends Neurosci., 14: 401-405. Mills, KR and Kimiskidis, V (1996) Motor cortex excitability during ballistic forearm and finger movements. Muse. Nerve, 19: 468-473. Mills, KR and Nithi, KA (1997a) Corticomotor threshold is reduced in early sporadic amyotrophic lateral sclerosis. Muse. Nerve, 20: 1137-1141. Mills, KR and Nithi, KA (1997b) Corticomotor threshold to magnetic stimulation: normal values and repeatability. Muse. Nerve, 20: 570-576. Modugno, N, Curra, A, Giovannelli, M, Priori, A, Squitieri, F, Ruggieri, S, Manfredi, M and Berardelli, A (2001) The prolonged cortical silent period in patients with Huntington's disease. Clin. Neurophysiol., 112: 1470-1474. Moll, GH, Heinrich, H, Trott, GE, Wirth, S, Bock, Nand Rothenberger, A (2001) Children with comorbid attention-deficit-hyperactivity disorder and tic disorder: evidence for additive inhibitory deficits within the motor system. Ann. Neurol., 49: 393-396. Mortifee, P, Stewart, H, Schulzer, M and Eisen, A (1994) Reliability of transcranial magnetic stimulation for mapping the human motor cortex. Electroencephalogr. Clin. Neurophysiol., 93: 131-137. Muellbacher, W, Ziemann, U, Boroojerdi, B and Hallett, M (2000) Effects of low-frequency transcranial magnetic stimulation on motor excitability and basic motor behavior. Clin. Neurophysiol., 111: 1002-1007. MUller, K, Homberg, V and Lenard, HG (1991) Magnetic stimulation of motor cortex and nerve roots in children.

119 Maturation of cortico-motoneuronal projections. Electroencephalogr. Clin. Neurophysiol., 81: 63-70. MUller, K, Kass-Iliyya, F and Reitz, M (1997) Ontogeny of ipsilateral corticospinal projections: a developmental study with transcranial magnetic stimulation. Ann. Neurol., 42: 705-711. Nakamura, H, Kitagawa, H, Kawaguchi, Y and Tsuji, H (1996) Direct and indirect activation of human corticospinal neurons by transcranial magnetic and electrical stimulation. Neurosci. Lett., 210: 45-48. Nakamura, H, Kitagawa, H, Kawaguchi, Y and Tsuji, H (1997) Intracortical facilitation and inhibition after transcranial magnetic stimulation in conscious humans. J. Physiol. (Lond.), 498: 817-823. Netz, J, Ziemann, U and Hamberg, V (1995) Hemispheric asymmetry of transcallosal inhibition in man. Exp. Brain Res., 104: 527-533. Netz, J, Lammers, T and Hamberg, V (1997) Reorganization of motor output in the non-affected hemisphere after stroke. Brain, 120: 1579-1586. Niehaus, L, Guldin, B and Meyer, B (2001a) Influence of transcranial magnetic stimulation on pupil size. J. Neurol. Sci., 182: 123-128. Niehaus, L, Meyer, BU and Roricht, S (1998) Magnetic stimulation over different brain regions: no differential effects on the elicited sympathetic skin responses. Electroencephalogr. Clin. Neurophysiol., 109: 94-99. Niehaus, L, VonAlt-Stutterheim, K, Roricht, S and Meyer, BU (2001b) Abnormal postexcitatory and interhemispheric motor cortex inhibition in writer's cramp. J. Neurol., 248: 51-56. Nielsen, JF (1996) Improvement of amplitude variability of motor evoked potentials in multiple sclerosis patients and in healthy subjects. Electroencephalogr. Clin. Neurophysiol., 101: 404-411. Nilsson, J, Panizza, M and Arieti, P (1997) Computeraided determination of the silent period. J. Clin. Neurophysiol., 14: 136-143. Oechsner, M and Zangemeister, WH (1999) Prolonged postexcitatory inhibition after transcranial magnetic stimulation of the motor cortex in patients with cerebellar ataxia. J. Neurol. Sci., 168: 107-111. Ofuji, A, Kaneko, K, Taguchi, T, Fuchigarni, Y, Morita, H and Kawai, S (1998) New method to measure central motor conduction time using transcranial magnetic stimulation and T-response. J. Neurol. Sci., 160: 26-32. Palmer, E and Ashby, P (1992a) Corticospinal projections to upper limb motoneurons in humans. J. Physiol. (Lond.), 448: 397-412. Palmer, E and Ashby, P (1992b) Evidence that a long latency stretch reflex in humans is transcortical. J. Physiol. (Lond.), 449: 429-440.

120 Pascual-Leone, A, Gates, JR and Dhuna, A (1991) Induction of speech arrest and counting errors with rapid-rate transcranial magnetic stimulation (see comments). Neurology, 41: 697-702. Pascual-Leone, A, Houser, CM, Reese, K, Shotland, LI, Grafman, J, Sato, S, Valls-Sole, J, Brasil-Neto, JP, Wassermann, EM and Cohen, LG et al. (1993) Safety of rapid-rate transcranial magnetic stimulation in normal volunteers. Electroencephalogr. Clin. Neurophysiol.,89: 120-130.

Pascual-Leone, A, Grafman, J and Hallett, M (1994a) Modulation of cortical motor output maps during development of implicit and explicit knowledge. Science, 263: 1287-1289. Pascual-Leone, A, Valls-Sole, J, Brasi1-Neto, JP, Cammarota, A, Grafman, J and Hallett, M (1994b) Akinesia in Parkinson's disease. II. Effects of subthreshold repetitive transcranial motor cortex stimulation. Neurology, 44: 892-898. Pascual-Leone, A, Valls-Sole, J, Wassermann, EM and Hallett, M (1994c) Responses to rapid-rate transcrania1 magnetic stimulation of the human motor cortex. Brain, 117: 847-858. Pascual-Leone, A, Nguyet, D, Cohen, LG, Brasil-Neto, JP, Cammarota, A and Hallett, M (1995a) Modulation of muscle responses evoked by transcranial magnetic stimulation during the acquisition of new fine motor skills. J. Neurophysiol., 74: 1037-1045. Pascual-Leone, A, Wassermann, EM, Sadato, Nand Hallett, M (1995b) The role of reading activity on the modulation of motor cortical outputs to the reading hand in Braille readers. Ann. Neurol., 38: 910-915. Pascual-Leone, A, Bartres-Faz, D and Keenan, JP (1999) Transcranial magnetic stimulation: studying the brainbehaviour relationship by induction of 'virtual lesions' . Phil. Trans. R. Soc. Lond. B, 354: 1229-1238. Patton, HD and Amassian, VB (1954) Single- and multiple-unit analysis of cortical stage of pyramidal tract activation. J. Neurophysiol., 17: 345-363. Paus, T, Jech, R, Thompson, CJ, Comeau, R, Peters, T and Evans, AC (1997) Transcranial magnetic stimulation during positron emission tomography: a new method for studying connectivity of the human cerebral cortex. J. Neurosci., 17: 3178-3184. Paus, T, Jech, R, Thompson, CJ, Comeau, R, Peters, T and Evans, AC (1998) Dose-dependent reduction of cerebral blood flow during rapid-rate transcranial magnetic stimulation of the human sensorimotor cortex. J. Neurophysiol., 79: 1102-1107. Peinemann, A, Lehner, C, Mentschel, C, Munchau, A, Conrad, Band Siebner, HR (2000) Subthreshold 5-Hz repetitive transcranial magnetic stimulation of the human primary motor cortex reduces intracortical paired-pulse inhibition. Neurosci. Lett., 296: 21-24.

V.ZIEMANN

Peinemann, A, Lehner, C, Conrad, Band Siebner, HR (2001) Age-related decrease in paired-pulse intracortical inhibition in the human primary motor cortex. Neurosci Lett., 313: 33-36. Pepin, JL, Bogacz, D, de Pasqua, V and Delwaide, PJ (1999) Motor cortex inhibition is not impaired in patients with Alzheimer's disease: evidence from paired transcranial magnetic stimulation. 1. Neurol. Sci., 170: 119-123. Pereon, Y, Genet, R and Guiheneuc, P (1995) Facilitation of motor evoked potentials: timing of Jendrassik maneuver effects. Muse. Nerve, 18: 1427-1432. Pierrot-Deseilligny, E (1996) Transmission of the cortical command for human voluntary movement through cervical propriospinal premotoneurons. Prog. Neurobiol., 48: 489-517. Plewnia, C, Bartels, M, Cohen, L and Gerloff, C (2001) Noradrenergic modulation of human cortex excitability by the presynaptic alpha(2)-antagonist yohimbine. Neurosci. Lett., 307: 41-44. Priori, A, Berardelli, A, Inghilleri, M, Polidori, L and Manfredi, M (1994) Electromyographic silent period after transcranial brain stimulation in Huntington's disease. Mov. Disord., 9: 178-182. Priori, A, Polidori, L, Rona, S, Manfredi, M and Berardelli, A (2000) Spinal and cortical inhibition in Huntington's chorea. Mov. Disord., 15: 938-946. Prout, AJ and Eisen, AA (1994) The cortical silent period and amyotrophic lateral sclerosis. Muse. Nerve, 17: 217-223. Ravnborg, M, Liguori, R, Christiansen, P, Larsson, H and Sorensen, PS (1992) The diagnostic reliability of magnetically evoked motor potentials in multiple sclerosis. Neurology, 42: 1296-1301. Reutens, DC, Berkovic, SF, Macdonell, RA and Bladin, PF (1993a) Magnetic stimulation of the brain in generalized epilepsy: reversal of cortical hyperexcitability by anticonvulsants. Ann. Neurol., 34: 351-355. Reutens, DC, Puce, A and Berkovic, SF (1993b) Cortical hyperexcitability in progressive myoclonus epilepsy: a study with transcranial magnetic stimulation. Neurology,43: 186-192.

Ridding, MC and Rothwell, JC (1997) Stimuluslresponse curves as a method of measuring motor cortical excitability in man. Electroencephalogr. Clin. Neurophysiol., 105: 340-344. Ridding, MC, Taylor, JL and Rothwell, JC (1995) The effect of voluntary contraction on cortico-cortical inhibition in human motor cortex. 1. Physiol. (Lond.), 487: 541-548. Ridding, MC, Brouwer, B and Nordstrom, MA (2000) Reduced interhemispheric inhibition in musicians. Exp. Brain Res., 133: 249-253.

TRANS CRANIAL MAGNETIC STIMULATION

Rijntjes, M, Tegenthoff, M, Liepert, J, Leonhardt, G, Kotterba, S, Muller, S, Kiebel, S, Malin, JP, Diener, HC and Weiller, C (1997) Cortical reorganization in patients with facial palsy. Ann. Neurol., 41: 621-630. Robinson, LR, Jantra, P and MacLean, IC (1988) Central motor conduction times using transcranial stimulation and F wave latencies. Muse. Nerve, II: 174-180. Roick, H, Von Giesen, HJ and Benecke, R (1993) On the origin of the postexcitatory inhibition seen after transcranial magnetic brain stimulation in awake human subjects. Exp. Brain Res., 94: 489-498. Rona, S, Berardelli, A, Vacca, L, Inghilleri, M and Manfredi, M (1998) Alterations of motor cortical inhibition in patients with dystonia. Mov. Disord., 13: 118-124. Roricht, S, Meyer, B-U, Woichiechowsky, C and Lehmann, R (1998) Callosal and corticospinal tract function in patients with hydrocephalus: a morphometric and transcranial magnetic stimulation study. J. Neurol., 245: 28Q....288. Rosler, KM, Truffert, A, Hess, CW and Magistris, MR (2000) Quantification of upper motor neuron loss in amyotrophic lateral sclerosis. Clin. Neurophysiol., III: 2208-2218. Rossi, S, Pasqualetti, P, Rossini, PM, Feige, B, Ulivelli, M, Glocker, FX, Battistini, N, Liicking, CH and KristevaFeige, R (2000) Effects of repetitive transcranial magnetic stimulation on movement-related cortical activity in humans. Cereb. Cortex, 10: 802-808. Rossi, S, Cappa, SF, Babiloni, C, Pasqualetti, P, Miniussi, C, Carducci, F, Babiloni, F and Rossini, PM (200l) Prefontal cortex in long-term memory: an "interference" approach using magnetic stimulation. Nat. Neurosci., 4: 948-952. Rossini, PM, Barker, AT, Berardelli, A, Caramia, MD, Caruso, G, Cracco, RQ, Dimitrijevic, MR, Hallett, M, Katayama, Y and LUcking, CH et al. (1994) Noninvasive electrical and magnetic stimulation of the brain, spinal cord and roots: basic principles and procedures for routine clinical application. Report of an IFCN committee. Electroencephalogr. Clin. Neurophysiol., 91: 79-92. Rossini, PM, Marciani, MG, Caramia, M, Roma, V and Zarola, F (1985) Nervous propagation along 'central' motor pathways in intact man: characteristics of motor responses to 'bifocal' and 'unifocal' spine and scalp non-invasive stimulation. Electroencephalogr. Clin. Neurophysiol., 61: 272-286. Rossini, PM, Caramia, M and Zarola, F (1987) Central motor tract propagation in man: studies with noninvasive, unifocal, scalp stimulation. Brain Res., 415: 211-225.

121 Rossini, PM, Paradiso, C, Zarola, F, Bernardi, G, Caramia, MD, Margari, L and Ferrari, E (1991) Brain excitability and long latency muscular arm responses: non-invasive evaluation in healthy and parkinsonian subjects. Electroencephalogr. Clin. Neurophysiol., 81: 454-465. Rossini, PM, Berardelli, A, Deuschl, G, Hallett, M, Maertens de Noordhout, AM, Paulus, W, and Pauri, F (1999) Applications of magnetic cortical stimulation. Electroencephalogr. Clin. Neurophysiol., Suppl. 52: 171-185. Roth, BJ, Saypol, JM, Hallett, M and Cohen, LG (1991) A theoretical calculation of the electric field induced in the cortex during magnetic stimulation. Electroencephalogr. Clin. Neurophysiol., 81: 47-56. Rothwell, JC, Thompson, PD, Day, BL, Dick, JP, Kachi, T, Cowan, JM and Marsden, CD (1987) Motor cortex stimulation in intact man. 1. General characteristics of EMG responses in different muscles. Brain, lID: 1173-1190. Rothwell, JC, Thompson, PD, Day, BL, Boyd, S and Marsden, CD (1991) Stimulation of the human motor cortex through the scalp. Exp. Physiol., 76: 159-200. Rothwell, J, Burke, D, Hicks, R, Stephen, J, Woodforth, I and Crawford, M (1994) Transcranial electrical stimulation of the motor cortex in man: further evidence for the site of activation. J. Physiol. (Lond.), 481: 243-250. Rothwell, JC, Hallett, M, Berardelli, A, Eisen, A, Rossini, P and Paulus, W (1999) Magnetic stimulation: motor evoked potentials. The International Federation of Clinical Neurophysiology. Electroencephalogr. Clin. Neurophysiol., Suppl. 52: 97-103. Rouiller, EM, Babalian, A, Kazennikov, 0, Moret, V, Yu, XH and Wiesendanger, M (1994) Transcallosal connections of the distal forelimb representations of the primary and supplementary motor cortical areas in macaque monkeys. Exp. Brain Res., 102: 227-243. Rudiak, D and Marg, E (1994) Finding the depth of magnetic brain stimulation: are-evaluation. Electroencephalogr. Clin. Neurophysiol., 93: 358-371. Saito, Y, Yokota, T and Yuasa, T (1995) Suppression of motor cortical excitability by magnetic stimulation of the cerebellum. Brain Res., 694: 20Q....206. Sakai, K, Ugawa, Y, Terao, Y, Hanajima, R, Furabayashi, T and Kanazawa, I (1997) Preferential activation of different I waves by transcranial magnetic stimulation with a figure-of-eight shaped coil. Exper. Brain Res., 113: 24-32. Sarnii, A, Luciano, CA, Dambrosia, 1M and Hallett, M (1998) Central motor conduction time: reproducibility and discomfort of different methods. Mus. Nerve, 21: 1445-1450.

122 Sandbrink, F, Syed, NA, Fujii, MD, Dalakas, MC and Floeter, MK (2000) Motor cortex excitability in stiffperson syndrome. Brain, 123: 2231-2239. Sanger, TD, Garg, RR and Chen, R (2001) Interactions between two different inhibitory systems in the human motor cortex. J. Physiol., 530(2): 307-317. Schmid, UD, Walker, G, Schmid-Sigron, J and Hess, CW (1991) Transcutaneous magnetic and electrical stimulation over the cervical spine: excitation of plexus roots - rather than spinal roots. Electroencephalogr. Clin. Neurophysiol., Suppl. 43: 369-384. Schmierer, K, Niehaus, L, Roricht, S and Meyer, BU (2000) Conduction deficits of callosal fibers in early multiple sclerosis. J. Neurol. Neurosurg. Psychiatry, 68: 633-638. Schnitzler, A and Benecke, R (1994) The silent period after transcranial magnetic stimulation is of exclusive cortical origin: evidence from isolated cortical ischemic lesions in man. Neurosci. Lett., 180: 41-45. Schurmann, M, Nikouline, VV, Soljanlahti, S, Ollikainen, M, Basar, E and Ilmoniemi, RJ (2001) EEG responses to combined somatosensory and transcranial magnetic stimulation. Glin. Neurophysiol., 112: 19-24. Seyal, M, Browne, JK, Masuoka, LK and Gabor, AJ (1993) Enhancement of the amplitude of somatosensory evoked potentials following magnetic pulse stimulation of the human brain. Electroencephalogr. cu« Neurophysiol., 88: 20-27. Siciliano, G, Manca, ML, Sagliocco, L, Pastorini, E, Pellegrinetti, A, Sartucci, F, Sabatini, A and Murri, L (1999) Cortical silent period in patients with amyotrophic lateral sclerosis. J. Neurol. Sci., 169: 93-97. Siebner, HR, Dressnandt, J, Auer, C and Conrad, B (1998a) Continuous intrathecal baclofen infusions induced a marked increase of the transcranially evoked silent period in a patient with generalized dystonia. Muse. Nerve, 21: 1209-1212. Siebner, HR, Willoch, F, Peller, M, Auer, C, Boecker, H, Conrad, B and Bartenstein, P (1998b) Imaging brain activation induced by long trains of repetitive transcranial magnetic stimulation. NeuroReport, 9: 943948. Siebner, HR, Mentschel, C, Auer, C and Conrad, B (1999a) Repetitive transcrania1 magnetic stimulation has a beneficial effect on bradykinesia in Parkinson's disease. NeuroReport, 10: 589-594. Siebner, HR, Tormos, 1M, Ceballos-Baumann, AO, Auer, C, Catala, MD, Conrad, B and Pascual-Leone, A (1999b) Low-frequency repetitive transcranial magnetic stimulation of the motor cortex in writer's cramp. Neurology, 52: 529-537. Siebner, HR, Peller, M, Willoch, F, Minoshima, S, Boecker, H, Auer, C, Drzezga, A, Conrad, Band Bartenstein, P (2000a) Lasting cortical activation after

V.ZIEMANN

repetitive TMS of the motor cortex: a glucose metabolic study. Neurology, 54: 956-963. Siebner, HR, Rossmeier, C, Mentschel, C, Peinemann, A and Conrad, B (2000b) Short-term motor improvement after sub-threshold 5-Hz repetitive transcranial magnetic stimulation of the primary motor hand area in Parkinson's disease. J. Neurol. Sci., 178: 91-94. Siebner, HR, Takano, B, Peinemann, A, Schwaiger, M, Conrad, B and Drzezga, A (2001) Continuous transcranial magnetic stimulation during positron emission tomography: a suitable tool for imaging regional excitability of the human cortex. Neuroimage, 14: 883-890. Singh, KD, Hamdy, S, Aziz, Q and Thompson, DG (1997) Topographic mapping of trans-cranial magnetic stimulation data on surface rendered MR images of the brain. Electroencephalogr. Clin. Neurophysiol., 105: 345-351. Smith, MJ, Keel, JC, Greenberg, BD, Adams, LF, Schmidt, PJ, Rubinow, DA and Wassermann, EM (1999) Menstrual cycle effects on cortical excitability. Neurology, 53: 2069-2072. Smith, HC, Davey, NJ, Savic, G, Maskill, DW, Ellaway, PH, Jamous, MA and Frankel, HL (2000) Modulation of single motor unit discharges using magnetic stimulation of the motor cortex in incomplete spinal cord injury. J. Neurol. Neurosurg. Psychiatry, 68: 516-520. Sohn, YR, Kaelin-Lang, A, Jung, HY and Hallett, M (200 1) Effect of levetiracetam on human corticospinal excitability. Neurology, 57: 858-863. Stedman, A, Davey, NJ and Ellaway, PH (1998) Facilitation of human first dorsal interosseous muscle responses to transcranial magnetic stimulation during voluntary contraction of the contralateral homonymous muscle. Muse. Nerve, 21: 1033-1039. Stewart, L, Walsh, VV, Frith, U and Rothwell, JC (2001) TMS produces two dissociable types of speech disruption. Neuroimage, 13: 472-478. Strafella, AP and Paus, T (2001) Cerebral blood-flow changes induced by paired-pulse transcranial magnetic stimulation of the primary motor cortex. J. Neurophysiol., 85: 2624-2629. Streletz, LJ, Be1evich, JK, Jones, SM, Bhushan, A, Shah, SH and Herbison, GJ (1995) Transcranial magnetic stimulation: cortical motor maps in acute spinal cord injury. Brain Topogr., 7: 245-250. Tamburin, S, Manganotti, P, Zanette, G and Fiaschi, A (2001) Cutaneomotor integration in human hand motor areas: somatotopic effect and interaction of afferents. Exp. Brain Res., 141: 232-241. Tegenthoff, M, Vorgerd, M, Juskowiak, F, Roos, V and Malin, JP (1996) Postexcitatory inhibition after transcranial magnetic single and double brain stimulation

TRANS CRANIAL MAGNETIC STIMULATION

in Huntington's disease. Electroencephalogr. Clin. Neurophysiol., 101: 298-303. Terao, Y, Ugawa, Y, Uesaka, Y, Hanajima, R, GembaShimizu, K, Ohki, Y and Kanazawa, I (1995) Input-output organization in the hand area of the human motor cortex. Electroencephalogr. Clin. Neurophysiol., 97: 375-381. Terao, Y, Ugawa, Y, Sakai, K, Miyauchi, S, Fukuda, H, Sasaki, Y, Takino, R, Hanajima, R, Furubayashi, T, Putz, B and Kanazawa, I (1998) Localizing the site of magnetic brain stimulation by functional MRI. Exp. Brain Res., 121: 145-152. Terao, Y, Ugawa, Y, Hanajima, R, Furubayashi, T, Machii, K, Enomoto, H, Shiio, Y, Mochizuki, H, Uesugi, H, Uesaka, Y and Kanazawa, I (1999) Air-puff-induced facilitation of motor cortical excitability studied in patients with discrete brain lesions. Brain, 122: 2259-2277. Terao,Y,Ugawa, Y,Hanajima, R, Machii, K, Furubayashi, T, Mochizuki, H, Enomoto, H, Shiio, Y, Uesugi, H, Iwata, NK and Kanazawa, I (2001) A single motor unit recording technique for studying the differential activation of corticospinal volleys by transcranial magnetic stimulation. Brain Res. Brain Res. Protoc., 7: 61-67. Tergau, F, Naumann, U, Paulus, W and Steinhoff, BJ (1999) Low-frequency repetitive transcranial magnetic stimulation improves intractable epilepsy (letter). Lancet, 353, 2209. Thickbroom, GW, Sammut, R and Mastaglia, FL (1998) Magnetic stimulation mapping of motor cortex: factors contributing to map area. Electroencephalogr. Clin. Neurophysiol., 109: 79-84. Thompson, ML, Thickbroom, GW and Mastaglia, FL (1997) Corticomotor representation of the sternocleidomastoid muscle. Brain, 120: 245-255. Tokimura, H, Ridding, MC, Tokimura, Y, Amassian, VE and Rothwell, JC (1996) Short latency facilitation between pairs of threshold magnetic stimuli applied to human motor cortex. Electroencephalogr. Clin. Neurophysiol., 101: 263-272. Tokimura, H, Di Lazzaro, V, Tokimura, Y, Oliviero, A, Profice, P, Insola, A, Mazzone, P, Tonali, P and Rothwell, JC (2000) Short latency inhibition of human hand motor cortex by somatosensory input from the hand. J. Physiol., 523: 503-513. Topka, H, Cohen, LG, Cole, RA and Hallett, M (1991) Reorganization of corticospinal pathways following spinal cord injury. Neurology, 41: 1276-1283. Touge, T, Gerschlager, W, Brown, P and Rothwell, JC (2001) Are the after-effects of low-frequency rTMS on motor cortex excitability due to changes in the efficacy of cortical synapses? Clin. Neurophysiol., 112: 21382145.

123 Traversa, R, Cicinelli, P, Filippi, M, Oliveri, M, Palmieri, MG, Pasqualetti, P and Rossini, PM (1999) A method to monitor motor cortical excitability in human stroke through motor evoked potentials. Brain Res. Protocols, 4: 44-48. Triggs, WJ, Cros, D, Macdonell, RA, Chiappa, KH, Fang, J and Day, BJ (1993) Cortical and spinal motor excitability during the transcranial magnetic stimulation silent period in humans. Brain Res., 628: 39-48. Trompetto, C, Assini, A, Buccolieri, A, Marchese, R and Abbruzzese, G (2000) Motor recovery following stroke: a transcranial magnetic stimulation study. Clin. Neurophysiol., 111: 1860-1867. Troni, W, Cantello, R, De Mattei, M and Bergamini, L (1988) Muscle responses elicited by cortical stimulation in the human hand: differential conditioning by activation of the proprioceptive and exteroceptive fibers of the median nerve. In: PM Rossini and CD Marsden (Eds.), Non-invasive Stimulation of Brain and Spinal Cord: Fundamentals and Clinical Applications. New York:Alan R Liss Inc., pp. 73-83. Truffert, A, Rosier, KM and Magistris, MR (2000) Amyotrophic lateral sclerosis vs. cervical spondylotic myelopathy: a study using transcranial magnetic stimulation with recordings from the trapezius and limb muscles. Clin. Neurophysiol., 111: 1031-1038. Tsuji, T and Rothwell, JC (2002) Long lasting effects of rTMS and associated peripheral sensory input on MEPs, SEPs and transcortical reflex excitability in humans. J. Physiol., 540: 367-"376. Turton, A, Wroe, S, Trepte, N, Fraser, C and Lemon, RN (1996) Contralateral and ipsilateral EMG responses to transcranial magnetic stimulation during recovery of arm and hand function after stroke. Electroencephalogr. Clin. Neurophysiol., 101: 316-328. Ugawa, Y (1999) Stimulation at the foramen magnum level. Electroencephalogr. Clin. Neurophysiol., Suppl. 51: 65-75. Ugawa, Y, Rothwell, JC, Day, BL, Thompson, PD and Marsden, CD (1989) Magnetic stimulation over the spinal enlargements. J. Neurol. Neurosurg. Psychiatry, 52: 1025-1032. Ugawa, Y, Day, BL, Rothwell, JC, Thompson, PD, Merton, PA and Marsden, CD (1991a) Modulation of motor cortical excitability by electrical stimulation over the cerebellum in man. J. Physiol. (Lond.), 441: 57-72. Ugawa, Y, Rothwell, JC, Day, BL, Thompson, PD and Marsden, CD (1991b) Percutaneous electrical stimulation of corticospinal pathways at the level of the pyramidal decussation in humans. Ann. Neurol., 29: 418-427.

124 Ugawa, Y, Hanajima, R and Kanazawa, I (1993) Interhemispheric facilitation of the hand area of the human motor cortex. Neurosci. Lett., 160: 153-155. Ugawa, Y,Genba-Shimizu, K, Rothwell, JC, Iwata, M and Kanazawa, I (1994a) Suppression of motor cortical excitability by electrical stimulation over the cerebellum in ataxia. Ann. Neurol., 36: 90-96. Ugawa, Y, Uesaka, Y, Terao, Y, Hanajima, Rand Kanazawa, I (1994b) Magnetic stimulation of corticospinal pathways at the foramen magnum level in humans. Ann. Neurol., 36: 618-624. Ugawa, Y, Genba-Shimizu, K and Kanazawa, I (1995a) Electrical stimulation of the human descending motor tracts at several levels. Can. J. Neurol. Sci., 22: 3~2. Ugawa, Y, Uesaka, Y, Terao, Y, Hanajima, R and Kanazawa, I (1995b) Magnetic stimulation over the cerebellum in humans. Ann. Neurol., 37: 703-713. Ugawa, Y, Terao, Y, Hanajima, R, Sakai, K, Furubayashi, T, Machii, K and Kanazawa, I (1997) Magnetic stimulation over the cerebellum in patients with ataxia. Electroencephalogr. Clin. Neurophysiol., 104: 453458. Valeriani, M, Restuccia, D, Di Lazzaro, V, Oliviero, A, Le Pera, D, Profice, P, Satumo, E and Tonali, P (2001) Inhibition of biceps brachii muscle motor area by painful heat stimulation of the skin. Exp. Brain Res., 139: 168-172. Valls-Sole, J, Pascual-Leone, A, Wassermann, EM and Hallett, M (1992) Human motor evoked responses to paired transcranial magnetic stimuli. Electroencephalogr. Clin. Neurophysiol., 85: 355-364. Valls-Soh~, J, Pascual-Leone, A, Brasil-Neto, JP, Cammarota, A, McShane, L and Hallett, M (1994) Abnormal facilitation of the response to transcranial magnetic stimulation in patients with Parkinson's disease. Neurology, 44: 735-741. Valls-Sole, J, Tolosa, E, Marti, MJ, Valldeoriola, F, Revilla, M, Pastor, P and Blesa, R (200I) Examination of motor output pathways in patients with corticobasal ganglionic degeneration using transcranial magnetic stimulation. Brain, 124: 1131-1137. Valzania, F, Strafella, AP, Tropeani, A, Rubboli, G, Nassetti, SA and Tassinari, CA (1999) Facilitation of rhythmic events in progressive myoclonus epilepsy: a transcranial magnetic stimulation study. Clin. Neurophysiol., 110: 152-157. Van der Kamp, W, Zwinderman, AH, Ferrari, MD and Van Dijk, JG (1996) Cortical excitability and response variability of transcranial magnetic stimulation. J. Clin. Neurophysiol., 13: 164-171. Von Giesen, HJ, Roick, H and Benecke, R (1994) Inhibitory actions of motor cortex following unilateral brain lesions as studied by magnetic brain stimulation. Exp. Brain Res., 99: 84-96.

V.ZIEMANN

Walsh, V and Cowey, A (2000) Transcranial magnetic stimulation and cognitive neurosciences. Nature Rev., I: 73-79. Wang, B, Toro, C, Zeffiro, TA and Hallett, M (1994) Head surface digitization and registration: a method for mapping positions on the head onto magnetic resonance images. Brain Topogr., 6: 185-192. Warren, JD, Kimber, TE and Thompson, PD (1999) The silent period after magnetic brain stimulation in generalized tetanus. Muse. Nerve, 22: 1590-1592. Wassermann, EM (1998) Risk and safety of repetitive transcranial magnetic stimulation: Report and recommendations from the international workshop on the safety of repetitive transcranial magnetic stimulation June 5-7, 1996. Electroencephalogr. Clin. Neurophysiol., 108: 1-16. Wassermannm, EM and Lisanby, SH (2001) Therapeutic application of repetitive transcranial magnetic stimulation: a review. Clin. Neurophysiol., 112: 1367-1377. Wassermann, EM, Fuhr, P, Cohen, LG and Hallett, M (1991) Effects of transcranial magnetic stimulation on ipsilateral muscles. Neurology, 41: 1795-1799. Wassermann, EM, McShane, LM, Hallett, M and Cohen, LG (1992) Noninvasive mapping of muscle representations in human motor cortex. Electroencephalogr. Clin. Neurophysiol., 85: 1-8. Wassermann, EM, Pascual-Leone, A and Hallett, M (1994) Cortical motor representation of the ipsilateral hand and arm. Exp. Brain Res., 100: 121-132. Wassermann, EM, Cohen, LG, Flitman, SS, Chen, Rand Hallett, M (1996a) Seizures in healthy people with repeated "safe" trains of transcranial magnetic stimuli (letter). Lancet, 347: 825-826. Wassermann, EM, Grafrnan, J, Berry, C, Hollnagel, C, Wild, K, Clark, K and Hallett, M (1996b) Use and safety of a new repetitive transcranial magnetic stimulator. Electroencephalogr. Clin. Neurophysiol., 101: 412-417. Wassermann, EM, Wang, B, Zeffiro, TA, Sadato, N, Pascual-Leone, A, Toro, C and Hallett, M (1996c) Locating the motor cortex on the MRI with transcranial magnetic stimulation and PET. Neuroimage, 3: 1-9. Wassermann, EM, Greenberg, BD, Nguyen, MB and Murphy, DL (2001) Motor cortex excitability correlates with an anxiety-related personality trait. Biol. Psychiatry, 50: 377-382. Weber, M, Eisen, A and Nakajima, M (2000) Corticomotoneuronal activity in ALS: changes in the peristimulus time histogram over time. Clin. Neurophysiol., Ill: 169-177. Werhahn, KJ, Fong, JK, Meyer, BU, Priori, A, Rothwell, JC, Day, BL and Thompson, PD (1994) The effect of magnetic coil orientation on the latency of surface EMG and single motor unit responses in the first dorsal

TRANS CRANIAL MAGNETIC STIMULATION

interosseous muscle. Electroencephalogr. CUn. Neurophysiol., 93: 138-146. Werhahn, KJ, Taylor, I, Ridding, M, Meyer, BD and Rothwell, IC (1996) Effect of transcranial magnetic stimulation over the cerebellum on the excitability of human motor cortex. Electroencephalogr. Clin. Neurophysiol., 101: 58-66. Werhahn, KJ, Kunesch, E, Noachtar, S, Benecke, R and Classen, I (1999) Differential effects on motorcortical inhibition induced by blockade of GABA uptake in humans. J. Physiol. (Lond.), 517: 591-597. Werhahn, KJ, Wiseman, K, Herzog, I, Forderreuther, S, Dichgans, M and Straube, A (2000) Motor cortex excitability in patients with migraine with aura and hemiplegic migraine. Cephalalgia, 20: 45-50. Wessel, K, Tegenthoff, M, Vorgerd, M, Otto, V, Nitschke, MF and Malin, IP (1996) Enhancement of inhibitory mechanisms in the motor cortex of patients with cerebellar degeneration: a study with transcranial magnetic brain stimulation. Electroencephalogr. CUn. Neurophysiol., 101: 273-280. Wilson, SA, Lockwood, RI, Thickbroom, GW and Mastaglia, FL (1993a) The muscle silent period following transcranial magnetic cortical stimulation. J. Neurol. Sci., 114: 216-222. Wilson, SA, Thickbroom, GW and Mastaglia, FL (1993b) Transcranial magnetic stimulation mapping of the motor cortex in normal subjects. The representation of two intrinsic hand muscles. 1. Neurol. Sci., 118: 134-144. Wilson, SA, Day, BL, Thickbroom, GW and Mastaglia, FL (1996) Spatial differences in the sites of direct and indirect activation of corticospinal neurons by magnetic stimulation. Electroencephalogr. CUn. Neurophysiol., 101: 255-261. Wischer, S, Paulus, W, Sommer, M and Tergau, F (2001) Piracetam affects facilitatory l-wave interaction in the human motor cortex. CUn. Neurophysiol., 112: 275279. Wu, T, Sommer, M, Tergau, F and Paulus, W (2000) Lasting influence of repetitive transcranial magnetic stimulation on intracortical excitability in human subjects. Neurosci. Lett., 287: 37-40. Yokota, T, Yoshino, A, Hirashima, F, Komori, T and Miyatake, T (1994) Increased central motor tract excitability in Creutzfeldt-Jakob disease. J. Neurol. Sci., 123: 33-37. Zidar, I, Trontelj, IV and Mihelin, M (1987) Percutaneous stimulation of human corticospinal tract: a single-fiber

125 EMG study of individual motor unit responses. Brain Res.,422: 196-199. Ziemann, D (1999) Intracortical inhibition and facilitation in the conventional paired TMS paradigm. Electroencephalogr. Clin. Neurophysiol., Suppl. 51: 127-136. Ziemann, D and Hallett, M (2000) Basic neurophysiological studies with TMS. In: MS George and RH Belmaker (Eds), Transcranial Magnetic Stimulation in Neuropsychiatry. Washington, D.C.: American Psychiatry Press, pp. 45-98. Ziemann, D, Netz, I, Szelenyi, A and Hamberg, V (1993) Spinal and supraspinal mechanisms contribute to the silent period in the contracting soleus muscle after transcranial magnetic stimulation of human motor cortex. Neurosci. Lett., 156: 167-171. Ziemann, D, Lonnecker, S, Steinhoff, BI and Paulus, W (1996a) Effects of antiepileptic drugs on motor cortex excitability in humans: a transcranial magnetic stimulation study. Ann. Neurol., 40: 367-378. Ziemann, D, Rothwell, IC and Ridding, MC (l996b) Interaction between intracortical inhibition and facilitation in human motor cortex. J. Physiol. (Lond.), 496: 873-881. Ziemann, D, Paulus, Wand Rothenberger, A (1997) Decreased motor inhibition in Tourette disorder: Evidence from transcranial magnetic stimulation. Am. J. Psychiatry, 154: 1277-1284. Ziemann, D, Steinhoff, BI, Tergau, F and Paulus, W (1998a) Trancranial magnetic stimulation: its current role in epilepsy research. Epilepsy Res., 30: 11-30. Ziemann, D, Tergau, F, Wassermann, EM, Wischer, S, Hildebrandt, I and Paulus, W (1998b) Demonstration of facilitatory I-wave interaction in the human motor cortex by paired transcranial magnetic stimulation. J. Physiol. (Lond.), 511: 181-190. Ziemann, D, Tergau, F, Wischer, S, Hildebrandt, I and Paulus, W (l998c) Pharmacological control of facilitatory I-wave interaction in the human motor cortex. A paired transcranial magnetic stimulation study. Electroencephalogr. CUn. Neurophysiol., 109: 321-330. Ziemann, D, Ishii, K, Borgheresi, A, Yaseen, Z, Battaglia, F, Hallett, M, Cincotta, M and Wassermann, EM (1999) Dissociation of the pathways mediating ipsilateral and contralateral motor-evoked potentials in human hand and arm muscles. J. Physiol., 518: 895-906.

Movement Disorders Handbook of Clinical Neurophysiology, Vol. I M. Hallett (Ed.) © 2003 Elsevier B.V. All rights reserved

127 CHAPTER 9

Movement disorders surgery: microelectrode recording from deep brain nuclei W.D. Hutchison'v", J.O. Dostrovsky" and A.M. Lozano'" a

Department of Surgery, Division of Neurosurgery, Toronto Western Hospital, 399 Bathurst St., Toronto, ON M5T 2S8, Canada b Department of Physiology, Faculty of Medicine, University of Toronto, Toronto, ON M5S IA8, Canada

9.1. Introduction The development of imaging techniques in stereotactic brain surgery has greatly enhanced the capabilities of direct targeting of subcortical structures, but there is still a need for functional confirmation and optimization of the target. Microelectrode recordings provide the most definitive and accurate method of localization. The use of microelectrode recording for localizing subcortical targets for stereotactic brain surgery began in the 1950s with Albe-Fessard and Guiot (Guiot et aI., 1962) who used the technique to precisely delineate the motor and sensory thalamic nuclei. Microelectrode recording involves the measurement of electrical activity of brain cells with a high spatial and temporal resolution with thin probes that produce minimal mechanical disturbance of the neuropil. For this reason the technique has long remained a principal method for the analysis of the function of neurons and nuclei in the brain. Indeed, important insights into the pathophysiology of various movement disorders can be gained by investigation of the individual properties and population characteristics of neurons in the globus pallidus, thalamus and subthalamic nucleus. Microelectrode techniques are also continuing to evolve to examine simultaneous recordings from neuron pairs (Hurtado et aI., 1999; Levy et aI., 2000) and neuronal assemblies in the same and different nuclei (Nicolelis et aI., 1998).

* Correspondence to: Dr. W.D. Hutchison, Div. of Neurosurgery, Toronto Western Hospital West Wing 4-433, 399 Bathurst St., East Wing 6-528, Toronto, ON M5T 2S8, Canada. E-mail address: [email protected] or [email protected] Tel.: (416)-603-5800, ext. 2226; fax: (416)-603-5298.

Several reports of microelectrode recording techniques have been published with a focus on thalamus (Lenz et aI., 1988; Tasker et al., 1998), globus pallidus (Sterio et al., 1994; Lozano et aI., 1996; Hutchison 1998; Vitek et aI., 1998), and subthalamic nucleus (Hutchison et al., 1998a) and general articles on the techniques of extracellular recording in laboratory animals and humans (Millar 1992; Dostrovsky 1999; Lalley et al., 1999). The purpose ofthe present chapter is to review the current techniques used in our operating room and briefly outline the major neurophysiological landmarks that need to be identified for target determination in each case.

9.2. Microelectrode assembly Some detail is required in a discussion of microelectrodes, since most problems with recording are due to faulty or damaged electrodes rather than the electronic instrumentation used to amplify, filter and display the signals. In particular the fine tip of the electrode is susceptible to damage by mechanical or electrical forces. Electrodes with thicker shanks and blunter tips are less susceptible to bending and tip damage but produce more tissue damage and yield poorer multi-unit recordings. Completed electrodes with extensions allowing them to be used in a stereotactic guide tube are commercially available, but some may prefer to assemble their own. The main advantages of local manufacture are substantial cost savings and avoidance of the common problem of bending or curling of the fine tip during transport which may not be visible unless inspected under a microscope or loupes. Shank skewing may be less problematic as far as recording of good spikes is concerned but the electrode may track obliquely to the desired course. We fabricate our own electrodes from commercially available

128

components. Microelectrode tips (Microprobe, Potomac, MD) are mounted on Kapton (polyimide, MicroML, College Point, NY, 23 gauge) insulated stainless steel tubing (Small Parts, Miami Lakes, FL) to extend the length for use in the stereotactic guide tube. Tungsten has the desirable properties of high tensile strength and stiffness as thin wire. Platinum is relatively malleable so unsuited to thin electrode construction on its own but when present in an alloy with iridium becomes suitably stiff as thin wire. Platinum itself is desirable as an electrode material since it reacts with free chloride in the tissue and forms a Pt-PtCI or platinum black tip interface which is non-polarizable and irreversible. This means it can convert excess hydrogen ions into hydrogen and hydroxyl ions instead of reducing the metal and physically eroding the tip during microstimulation. However, most medical electrical stimulators use biphasic pulses, so there is no net charge transfer to the patient. Tungsten microelectrodes can be plated with platinum to confer these desirable properties (see below). Tip exposures in the range of 15-25 urn are the most useful and give mostly single and occasionally multi-unit recordings. Larger tip sizes record from many more neurons making single unit discrimination more difficult whereas smaller tip sizes may record cells only in the immediate vicinity of the tip. The Parylene-C insulation on the portion of the electrode to be inserted into the extender tube is removed by mechanical stripping with fine emery paper or burning off by passing through a flame and the shank is crimped and inserted into 25 gauge stainless steel tubing so it is orthogonal with the extender tube. The 22 gauge Kapton tubing insulation fits over the extender and epoxy glue is used to make a continuous seal between the two insulators. The insulation can be tested by inserting just the electrode tip and then the rest of the shank into saline while observing the impedance reading, which should remain constant if there is no breech. Another method is to apply 3-10 V DC to the electrode in saline to watch electrolytic bubble formation, which should only occur at the tip. The thin coating of Parylene-C on the electrode tip is particularly sensitive to scratching so caution needs to be exercised during handling. Electrodes are plated using platinum and gold solutions of the free metal cations and cathodal current to attract those ions (Millar, 1992). The purpose of the prior gold plating is to aid in

W.D. HUTCHISON ET AL.

formation of a good bonding of the platinum on the tungsten. This is best carried out under microscopic control to monitor the amount of plating deposited, at least in the initial stages. The advantage of plating is to reduce the impedance and increase the signalto-noise but an added advantage is that it will protect the tungsten tip from oxidizing which occurs after prolonged storage (months). If electrodes are not plated they can be conditioned as described above in the bubble test to remove the tungsten oxides. Electrodes are best protected from damage by backloading them into 19 gauge shield tubing and securing them with adhesive tape. The shield tubing should be short enough and wide enough that the ethylene oxide will penetrate and sterilize the whole length of the electrode. Sometimes the tip can become curled during use and it will be noticeable that only vague small units are picked up in the background or there is background injury discharge before good single units can be seen. As a general rule, if there are no units recorded after about 5 mm of tracking down with the electrode, then probably the electrode is bad, since most targets are in cell dense areas, and even white matter regions show the occasional unit. With most commercially available recording systems, the impedance of the electrode can be measured in situ during a brief pause in the recording. Unplated 15-25 urn tip tungsten electrodes have impedances about 0.8-1.2 Mil. and the platinum-plated electrodes our group uses are typically 0.2-0.6 MO. Low values indicate a break in the insulation, and high values indicate that the tip may have been eroded by repeated high intensity stimulation. 9.3. Extracellular recording of spikes The term 'single unit' refers to extracellularly recorded potentials arising from action potentials generated by a single neuron. Typically, if the potential arises from the somatodendritic region of the cell the waveform is usually biphasic and of 1-2 ms duration with an after-potential, and this is referred to variously as a cell, spike or neuron. If the waveform of the potential is mono-phasic and short duration < I ms with no after-potential this usually means that its source is an axonal action potential. Multi-unit recording refers to a combination of potentials from many cells recorded simultaneously. While this allows several units to be sampled

RECORDING IN MOVEMENT DISORDERS SURGERY

simultaneously, which may expedite the identification of movement-related activity, the firing rate of individual cells cannot be determined with any accuracy. Electrodes with smaller tip sizes will record mostly single units and the occasional axon especially in white matter regions such as the internal capsule. With larger tipped electrodes, more potentials are recorded at a further distance from the tip, so that in addition to the multi-units, the background noise in the recording is larger when in cell-dense regions.

9.4. Amplification and filtering of signals The relatively small bioelectric potentials (- 100 J..L V) must be amplified several thousand times (usually -5,000-50,000) for passing the signal to audio monitors and driving oscilloscopes and data display systems of computers. Typically, differential amplification is employed, so that what is amplified is the difference in signal intensity on the positive and negative leads so any large electromagnetic induction producing an interference will be common to the two leads and subtracted from the signal of interest - usually termed common mode rejection. Filtering of the signal is used to remove unwanted frequencies in the very low «200 Hz) range to provide a stable DC baseline for spike discrimination. Most often noise occurs from interference from AC mains (60 Hz in North America, 50 Hz in some other countries) due to additional equipment in the operating room that is in the vicinity of the electrode leads. This can be due to electric patient beds, monitoring equipment, overhead fluorescent lights or projection equipment. With most recording systems commercially available, sufficient shielding for dealing with this stray capacitance has been incorporated into the low noise features of the amplifier design, so again, the noise is often due to a faulty electrode with a very high impedance or a poor ground connection. Very high frequencies (> 10 kHz) need to be removed to allow individual spike waveforms to be discriminated. One source of high frequency noise in the operating room is due to cautery equipment which still produces interference with the noise reduction features of new amplifiers. Some recording systems use digital filtering of signals, the theory of which is complex and beyond the scope of this review, but one advantage is the original signal is retained and

129

filtering can be changed by changing the software settings if other aspects are desired such as low frequencies of field potential oscillations.

9.5. Commercially available recording systems Several companies provide electronic equipment for recording extracellular action potentials during deep brain recording. The Axon Instruments Guideline System 3000 has been recently reviewed (Starr, 1999) and further information is available on their website (http://www.axon.com).This intraoperative system includes a "Clinical Micropositioner" that is mounted on either a Crossman-Roberts-Wells (CRW) or Leksell frame and allows movement in the X (anteroposterior) and Y (mediolateral) positions, without further adjustment of the frame settings. The recording system includes an ultra low-noise amplifier, combined with an electrical stimulation unit and a touch sensitive monitor which displays the spike waveforms as they occur on a millisecond time base as well as a more conventional oscilloscope-like tracing on a time base of 1-2 s. The level crossings viewed on the spike waveform window are displayed in a frequency histogram with 0.1-1 s bins "spike ratemeter" and the firing rate in Hz is displayed and refreshed every second. A useful feature of the Axon system is a hand-held module with controls for setting the intensity for stimulation through the tip of the electrode as well as changing the intensity of stimulation and measuring the impedance of the electrode with a 1 kHz sine wave. The stimulation parameters can be modified with pulse widths settings from 0.05 to 1 ms and the frequency range is from I to 300 Hz, and intensities range from 0-100 J..LA. A zero setting may be used as a control "null-stimulation" procedure to verify patient reports of non-motor effects such as phosphenes or tingling. Conventional parameters for microstimulation are 1-100 J..LA, a 1 s train of pulses at 300 Hz and pulse widths of 200 J..LS. These parameters are useful in determining visual and motor responses for the identification of optic tract and internal capsule respectively during pallidal procedures, internal capsule and medial lemniscus during STN and thalamic procedures as well as tremor arrest or reduction. In cases of mild tremor reduction often a longer train length of 5 s or 10 s can be carried out while visually inspecting the tremor to clarify any longer-lasting effects on tremor. In practice, stimula-

130

tion above 100 tJ-A is not recommended since the high impedance of the electrode (l Mil) dictates high voltages in excess of safety requirements (> 100 V) (and most stimulators will not be able to deliver voltages greater than this). Train durations can be given manually for stimulation as long as the switch is depressed or fixed durations from 0.5 to 20 s. The recording amplifier is blocked briefly while the stimulation pulses are delivered to the electrode so the leads, once attached to the sterile electrode shaft or connector of the drive holding it, do not need to be switched. Radionics (www.radionics.com) has a Neuroplan system including an Accudrive and Taha-Burchiel Recording Electrode Kit for use with the CRW frame system. The electrode drive in this system has a cable instead of hydraulics, which obviates possible fluid leakage or suction of air into the hydraulic drive upon retractions. The entire drive is lightweight and can be autoclaved. Further features allow integrated use with DBS implantation and the system is primarily designed for use with the CRW stereotactic frame. The system is fairly compact with all operations in desktop style computer system. Further details are available on their website. AlphaOmega (http://www.alphaomega-eng.com) makes the Neurotrek system which has the capability to record up to 5 channels of data from the 5 electrode configuration used by Benabid's group. The system also includes a stepper-motor microdrive with computer based settings, that can be also controlled via a hand-held remote and depth data logging features. It has microstimulation, macrostimulation and impedance measuring capabilities on any of the 5 possible electrodes and will also record EMG data on up to 7 additional channels. The high powered and high capacity computer provides online template matching algorithms and tagging features to identify specific traces as the trajectory progresses. The template matching is supported by alphaSort (Matlab) and provides comparison of spike shapes as well as spike clusters. Temporal features are included in a post-processing software package that should meet most high level research needs with plots such as interval time histograms and cross-correlogram displays of the various channels. Additional software will plot and print out the results of specific trajectories along with recognized or "tagged" traces of sample recordings and print out a color hard-copy of the full stereotactic map. Other independent

W.D. HUTCHISON ET AL.

workers have developed shareware programs (Interspike) that have processing features specifically designed for detection of spikes trains with temporal properties (bursting and pauses) found in pallidal structures (Favre et al., 1999). At the time of writing, Medtronic (www.medtronic.com) is also in the process of developing a Lead Point recording system for use with the CRW or else frame system. 9.6. Surgical technique Standard stereotactic technique is used to calculate the target from the AC and PC co-ordinates on MRI. More detailed descriptions of the surgical technique are available (Burchiel et al., 1997) as are complete monographs on the subject (Krauss, 1996; Germano, 1998; Lozano, 2000) and it will only be briefly reviewed here for the sake of completeness. A Leksell model G stereotactic frame is applied to the head with screw-pins inserted under local anesthetic, and a fiducial box containing channels filled with radio-opaque copper sulphate is placed over the frame to obtain three dimensional reference points (X - mediolateral; Y - anterior/posterior, Z inferior/superior) for the various MRI sections. Various techniques can be used to target the structure of interest, including direct calculation of the coordinates from the visualized structure on MRI scans, or indirect targeting by localization of wellvisualized landmarks such as the anterior and posterior commissures (AC, PC) and use of a standard stereotactic atlas (Schaltenbrand and Wahren, 1977) to infer the target location. In the former method, it has been noted that some significant distortion of images occurs with direct targeting. In the latter method, co-ordinates of AC and PC are obtained from the MRI console software (GE-Signa 1.5 Tesla magnet), and these are entered into a computer program that modifies the standard sagittal maps to obtain a customized stereotactic map for each patient. 9.7. Recordings in electrode trajectories targeting globus pallidus Several groups have described their methods for localizing the globus pallidus and surrounding structures (Hutchison et al., 1994a; Sterio et al., 1994; Lozano et al., 1996; Taha et al., 1996; Burchiel

RECORDING IN MOVEMENT DISORDERS SURGERY

et aI., 1997; Vitek et aI., 1998) and each uses slightly different techniques and places varying time and emphasis on the recording session. What follows is a description of our method. For pallidal procedures the target is generally considered to be in the ventral and lateral portion of GPi (Laitinen, 1990; Laitinen et aI., 1992), 20 mm lateral to the midline, 3-6 mm inferior to the AC-PC line and 2-3 mm anterior to the mid-commissural point. Recording starts about 15 mm from the target in an anterosuperior position, which corresponds to cells of GPe if initial targeting is correct. The objectives in localizing the target in globus pallidus are to identify: (1) characteristic cell types in GPe and GPi, and movement related activity of these cells; (2) the white matter lamina between these structures as well as border cells in the region; (3) optic tract ventral to the GPi; and (4) internal capsule posterior to GPi. The firing rates and patterns of basal ganglia neurons recorded in non-human primates are remarkably similar to those found in humans, so the terminology originally used in these studies is adopted here for the description of the cell types shown in Fig. 1. GPe cells have been described as occurring in two major types based on differences in firing rate and pattern as recorded in normal monkeys (DeLong, 1971; Filion et aI., 1988; Filion and Tremblay, 1991). These are the slow frequency discharge with pauses (SFD-P), and the low-frequency discharge with bursts (LFD-B). LFD-B neurons are not very common in GPe but are thought to be a characteristic feature of the region. The spontaneous ongoing activity is only about 5-10 Hz and the short bursts can reach about 300-500 Hz and occur at irregular time intervals. The known projection of GPe to thalamic reticular formation (Parent and Hazrati, 1995) and the high similarity in burst firing pattern between these cells suggests that LFB neurons are potential basal ganglia output neurons. SFD-P neurons have a higher spontaneous firing rate around 20-50 Hz which is sporadically interrupted by pauses in firing of duration about 150-300 ms (see Fig. 1). There are also cells in GPe with higher firing rates 50-70 Hz and these may be termed HFDP to follow the convention. Active and passive movements of limbs may modulate the firing rate of cells in GPe. Between GPe and GPi is a white matter lamina that is detectable by the absence of recorded units and relative quiet in background noise on the

131

recordings. Border cells are frequently encountered at its margins with a wider spike (due to a longer after-hyperpolarization) than pallidal cells, which imparts a regular firing pattern due to the longer

Fig. 1. Examples of well-isolated single units found in typical electrode trajectories penetrating the segments of globus pallidus during microelectrode exploration. All traces are 2 s in duration except bottom trace which is 3 s. Top trace shows the low firing rate of a striatal neuron, typically 1-5 Hz. Next two traces down are typical cells of the external segment of globus pallidus (GPe). Low frequency discharge with bursts (LFD-B); slow frequency discharge with pauses (SFD-P). At the margins of the pallidal segments and the medullary laminae are found border cells (Bor). Lowest two traces show high frequency discharge neurons (HFD), and a tremor cell (TC) with the accelerometer output from the back of the contralateral hand indicated below. GPi,i - internal segment of the globus pallidus internus.

132

relative refractory period (see Fig. 1). These electrophysiological characteristics indicate that the cells may be cholinergic and project to the cortical mantle (Carpenter, 1991). A sample of 17 border cells was found to have a mean firing rate of 35 Hz (Hutchison et aI., 1994a). After advancement of the electrode tip into GPi there is often an increase in background noise and multiunit recordings. Cells in GPi have relatively large amplitude spikes, irregular discharge patterns and the highest firing rates of all pallidal cells. The firing rates of GPi neurons are in the range of 60-80 Hz and the modal intervals for GPi neurons in the range of 5-7 ms. Some GPi neurons may show excitatory or inhibitory responses to passive and voluntary movements of limbs and orofacial structures. In some preliminary analyses of the data, there was no clear topographical organization of the body within GPi, as depicted in early studies by Hassler et al. (1979). In patients that have tremor, cells can be found with periodic oscillations in firing rate at the same frequency as the tremor (Hutchison et al., 1994b; Hutchison et al., 1997). These tremor cells show periods of coherent oscillation with limb tremor and appear to be located in ventral and lateral portions of the GPi (Hutchison et al., 1998b). Although each of the segments of GP has identified " signature" cell types that aid localization, there should be no misconception that a region can be immediately identified based on the firing rates and patterns of a few cells. Each pallidal segment has a range of firing rates and patterns and the population means of firing rates or pattern indices may show significant differences between segments. Normally a region is identified only after the track is completed and many well-isolated spikes have been recorded so that a comparison of all the firing rates and patterns can be integrated with the surmized position of the track based on the stereotactic anatomical map (Fig. 2). Below the GPi the cellular activity becomes sparse and background noise decreases. Upon entering the optic tract there may be some discernible increase in high frequency noise again due to axonal activity in the optic nerve. Stimulation at this site normally evokes phosphenes in the contralateral visual field at low current intensities (1-10 fLA). Patients frequently report white or yellow flashes of light, stars, sparkles or lightning-like patterns (see "Vi" in Fig. 2). This can be in a wedge-shaped

w.o. HUTCHISON ET AL. portion of the visual field which can sometimes be observed to move more ventral as one stimulates more dorsal in the optic tract, consistent with the rotation of the fibers in the tract at this level. In rare cases when patients do not report stimulation-evoked visual sensation, it is worthwhile to also carry out strobe-light evoked potentials by opening up the high pass filter and recording the slow-wave average (about 30-40 ms at this site) (see "VEP" in Fig. 2). If all of these features have been identified on the first trajectory through the pallidum, then the second trajectory is placed 3 rom posterior to attempt to identify the internal capsule. Recording again begins at 15 mm above the target and cells are usually again found at the top of the tract. In posteriorly located trajectories, cells may only be found at the initial part of the track indicating the posterior aspect of the GPi and the tip of the electrode will pass into the internal capsule. In the capsule, the recordings are usually quiet but the occasional unit or fiber is encountered. Passing electric current through the tip of the electrode, termed here microstimulation (up to 100 fLA, 0.2 ms pulse width, 300 Hz, 1 s train) will produce tetanic contraction of the contralateral body part or reduction or arrest of tremor (see "M" in Fig. 2). In a typical case, two or three electrode trajectories are required to complete a physiologic map and this includes unequivocal identification of the optic tract and internal capsule. If the targeting based on MRI co-ordinates is accurate, the physiological data should show a reasonably good spatial relation to the corresponding anatomical map. Macrostimulation using a large tipped (2-3 mm) electrode is used by some surgeons to locate the optic tract and internal capsule since the larger tip will allow a greater spread of electrical current. If no motor or visual effects are seen/reported at about 2 V, then the target can be considered safe from producing permanent adverse side effects.

9.8. Recording in electrode trajectories targeting the subthalamic nucleus (STN) In order to identify the STN with microelectrode recording the major features and landmarks to identify are: (1) anterior thalamus with bursting cells; (2) superior and inferior borders of STN; (3) dorsal border of SNr, and if possible; (4) the anterior and posterior borders of STN. Microelectrode

133

RECORDING IN MOVEMENT DISORDERS SURGERY

3

1

2\ #2048

GPe mel

Target

Fig. 2. Completed functional map of globus pallidus showing the location of identified neuronal types as described in Fig. 1. A total of three separate passes through the pallidum were made with the electrode in the order shown at the top of each track. Data from the mapping session is plotted on top of the customized sagittal section from the Schaltenbrand and Wahren stereotactic atlas 20 rom from the midline. Abbreviations of cell types as in Fig. 1; AC, PC - anterior and posterior commissures; mel - mid-commissural line; MEA - movement-evoked activity, "Target" refers to the tentative location chosen from the results of MRI scans, which was at the base of the pallidum but in this case would be too close to the optic tract (OT) indicated by the visual responses of the patient to microstimulation (Vi) and internal capsule (lC) indicated by motor or tetanic responses to microstimulation (M).

recording methods for the identification of STN have been described in detail already (Hutchison et al., 1998a). The target is in the center of the nucleus at about 10-12 rom lateral to the midline, I rom posterior to the mid-commissural point and 5 rom below the AC-PC line. Recording in microelectrode tracks targeting the STN starts 10 -15 rom above target. Depending on the trajectory angle in the sagittal plane, and given that the MRI-determined tentative target is reasonably accurate, recording will usually start in

thalamic reticular nucleus or in the anterior part of the ventral tier subnuclei of the thalamus (see Fig. 3). In this region and particularly in the thalamic reticular formation there are cells with spontaneous burst discharges, which have been reported previously (Raeva et al., 1991; Raeva and Lukashev, 1993) and can be considered characteristic for the region. The identification of the ventral portion of the thalamus is a useful landmark in determining the relative anterior and posterior position of the trajectory. Also the distance between the ventral border of

134

W.D. HUTCHISON ET AL.

Fig. 3. Right panel shows typical examples of spike recordings during microelectrode exploration of subthalamic nucleus. Left panel shows location of the electrode track on the customized Schaltenbrand and Wahren map at 12 mm from the midline. Hpth - hypothalamus, Rt - thalamic reticular nucleus, Voa - ventralis oralis anterior, Vop - ventralis oraIis posterior, Vim - ventralis intermedius, ZI - zona incerta, H2 - H2 fields of Forel, STN - subthalamic nucleus, SNr substantia nigra pars reticulata, other abbreviations as in Fig. I.

thalamus and the dorsal border of STN may give some indication of laterality since the anteromedial portion of the STN is pitched more ventral than the dorsolateral pole. Below the thalamus the electrode tip passes into a white matter region which is "quieter" in background electrical noise, corresponding to the thalamic and lenticular fasciculi with the zona incerta in-between. The zona incerta may have sparse cells that are bursting since it is continuous with the thalamic reticular nucleus at its lateral border. Others have reported cells in zona incerta with properties similar to subthalamic nucleus neurons i.e. responses to eye movements (Ma, 1996) as well as reaching movements (Crutcher et aI., 1980) so this should be borne in mind when determining the dorsal border of the subthalamic nucleus. The entry into the subthalamic nucleus is apparent when the level of background noise in the recordings increase and high amplitude spikes with firing rates of 25-45 Hz are found (25 and 75 percentiles). A

large number of STN neurons show clear modulation in firing rate with active and passive movements of the limbs and one sample revealed 80% of the neurons had excitatory responses. Ipsilateral as well as contralateral movements could elicit responses. In patients with tremor at rest, tremor cells have also been identified in the human STN with oscillations in firing rate both at the frequency of tremor and also at high frequency around 15-25 Hz (Levy et aI., 2000). This higher frequency component imparts a "chatter" or "flutter"-like sound to the background noise on the audio monitor that appears to be characteristic for the STN and has proven a useful feature for localization, at least in those patients with tremor. STN neurons may only be recorded over a short distance if the electrode does not track through the center of the nucleus, and trajectories to be selected as sites of implantation should have a 4-5 mm segment populated by cells with STN-like properties. A tracing of a typical STN neuron is shown in Fig. 3.

RECORDING IN MOVEMENT DISORDERS SURGERY

Below the STN is the substantia nigra, which is divided into the dopaminergic pars compacta (SNpc) and the basal ganglia output portion called the pars reticulata (SNpr). The division between the two structures is not clear and one expects that there is degeneration of the SNpc in these PD cases. Based on monkey recordings where histological confirmation is possible, there is known to be differences in the firing rates of the two groups. Putative dopaminergic pars compacta cells have very low firing rates 1-5 Hz and have inflections on the initial phase of the action potential (Schultz, 1986), but are expected to be rarely encountered in PD patients. In contrast, the characteristic features of SNpr or reticulata neurons are a high (60-90 Hz) and regular firing rate but there may be another group that have lower rates around 20-30 Hz, possibly reflecting functional differences (DeLong et aI., 1983, 1985). The different firing rates in SNpr may depend on the portion of the reticulata that is explored, since motor regions are more laterally located in this nucleus (about 15 mm from the midline). In the human, SNpr neurons frequently display cardiac-induced fluctuations in spike amplitude that can make spike discrimination problematic. Microstimulation (up to 100 j.LA, 300 Hz, 1 s train, 0.3 ms p.w.) during STN cases is not as useful as with pallidal or thalamic target localization, but stimulation-induced tremor arrest or reduction from within STN has been observed. Paresthesias have been encountered in more ventral and posterior positions and may be due to current spread to medial lemniscus or pre-Iemniscal radiation. 9.9. Recording and microstimulation in the motor and sensory thalamus Movement disorders surgery may also involve targets in the ventrointermediate (V.i.m.) nucleus of the ventral tier thalamus where input from proprioceptive primary afferents (joint capsules, Golgi tendon organs and muscle spindles) as well as cerebellar afferents converge. The thalamic target in V.i.m. is 14.5 mm lateral to the midline, 2-4 mm above the AC-PC line and 6-7 mm behind the midcommissural point (see Fig. 4). Frequently the ventrocaudal nucleus (V,c.) located more posteriorly is targeted for the first trajectory since the somatotopy of Vc is an important landmark for orientation

135

and also to avoid inadvertent damage to this structure that might lead to permanent sensory loss. Thalamic subnuclei cannot be visualized directly with MRI placing increased emphasis on functional identification of the various ventral tier nuclei with microelectrodes. The objectives in the thalamic procedures are to identify: (1) kinesthetic zone; (2) deep tactile zone; (3) cutaneous tactile zone; and (4) ventral border of the tactile region. Recordings at the top of trajectories pass thorough the "motor thalamus", comprised of the ventralis oralis anterior and posterior subnuclei (V.o.a, V.o.p.) and V.i.m. and have cells somewhat similar to those already described in thalamic areas slightly more medial. Bursting cells are frequently encountered as well as non-bursting cells. Movement-related activity (kinesthetic responses) may be present and cells may be found that respond to deep pressure or deep tactile stimuli. In patients with tremor, tremor cells may be recorded in this region (* in Fig. 4), and microstimulation will produce tremor arrest (TA in Fig. 4) or reduction in tremor. Passing the electrode further ventral, the spontaneous activity will usually increase with entry into the thalamic tactile relay of Vc. Cells in this region will respond to light stroke with a brush or cotton swab. The well-known somatotopic organization within Vc may help to guide the laterality of the procedure and the best place for the lesion is just anterior to the face and thumb region of the tactile core. Laterality in Vim, therefore, is largely determined with reference to Vc somatotopy, leg indicating a more lateral location than face and hand. Microstimulation within Vc produces focal paresthesia that shows a somatotopic organization roughly corresponding to that obtained by recording the cellular responses to touch. Microstimulation in ventral Vc and below Vc in V.c.p.c (parvocellularis) may elicit painful sensation. Ventral to Vc, microstimulation may also elicit hemibody paresthesias due to lemniscal fiber stimulation and activation of large numbers of neurons in the tactile core ofVc. Usually several trajectories are made to define the anterior border of Vc and map a large enough segment of the motor thalamus to find the focus and extent of sites of effective tremor suppression from microstimulation. The target lies 2-3 mm anterior to the Vc/Vim border in regions occupied by cells with kinesthetic responses and tremor reduction or arrest from microstimulation (see Fig. 4).

136

W.D. HUTCHISON ET AL.

Fig. 4. Functional localization of thalamic nuclei for movement disorders surgery. Findings from 3 microelectrode trajectories are overlaid on a Schaltenbrand and Wahren sagittal map 14.5 mm from the midline. The AC-PC line is shown intersecting Pc. Thalamic nuclei are labeled as follows Vo.a., ventralis oralis anterior, V.o.p., ventralis oralis posterior, V.i.m., ventralis intermedius, v.c., ventralis caudalis.

9.10. Summary Microelectrode recording is useful to accurately delineate deep brain structures and sub-nuclei for various stereotactic targets. In addition, it yields much information on the cellular pathophysiology of movement disorders and the rational development of surgical therapy for the treatment of movement disorders.

Acknowledgements The support of both Parkinson Society of Canada and the Canadian Institute for Health ResearchINIH is gratefully acknowledged.

References Burchiel, KJ, Taha, JM and Favre, J (1997) Posteroventral pallidotomy for Parkinson's disease patients. In: 55

Rengachary and RH Wilkins (Eds.), AANS Publications Committee. Park Ridge, Illinois, pp. 13-26. Carpenter, MB (1991) Corpus striatum and related nuclei. In: TS Satterfield (Ed.). Williams and Wilkins, Baltimore, pp. 325-360. Crutcher, MD, Branch, MR, DeLong, MR and Georgopoulos, AP (1980) Activity of zona incerta neurons in the behaving primate. Soc. Neurosci. Abstr., 6: 676. DeLong, MR (1971) Activity of pallidal neurons during movement. J. Neurophysiol., 34: 414-427. DeLong, MR, Crutcher, MD and Georgopoulos, AP (1983) Relations between movement and single cell discharge in the substantia nigra of the behaving monkey. J. Neuroscience, 3: 1599-606. DeLong, MR, Crutcher, MD and Georgopoulos, AP (1985) Primate globus pallidus and subthalamic nucleus: functional organization. J. Neurophysiol., 53: 530-543. Dostrovsky, JO (1999) Invasive techniques in humans: microelectrode recordings and microstimulation. In: U Windhorst and H Johansson (Eds.), Modem Techniques in Neuroscience Research. Springer Verlag, Berlin, pp.1199-1209.

137

RECORDING IN MOVEMENT DISORDERS SURGERY

Favre, J, Taha, JM, Baumann, T and Burchiel, KJ (1999) Computer analysis of the tonic, phasic, and kinesthetic activity of pallidal discharges in Parkinson patients. Surg. Neurol., 51: 665-672. Filion, M and Tremblay, L (1991) Abnormal spontaneous activity of globus pallidus neurons in monkeys with MPTP-induced parkinsonism. Brain Res., 547: 142151. Filion, M, Tremblay, L and Bedard, PJ (1988) Abnormal influences of passive limb movement on the activity of globus pallidus neurons in parkinsonian monkeys. Brain Res., 444: 165-176. Germano, 1M (1998) Neurosurgical treatment of movement disorders. AANS Publications Committee, Park Ridge, Illinois, 275 pp. Guiot, G, Hardy, J and Albe-Fessard, D (1962) Delimitation precise des structures sous-corticales et identification de noyaux thalamiques chez l'homme par l' electrophysiologie stereotaxique. Neurochirurgia (Stutt.), 51: 1-18. Hassler, R, Mundinger, F and Riechert, T (1979) Stereotaxis in Parkinson syndrome. Springer Verlag, Berlin, Hurtado, JM, Gray, CM, Tamas, LB and Sigvardt, KA (1999) Dynamics of tremor-related oscillations in the human globus pallidus: A single case study. Proc. Natl. Acad. Sci. USA, 96: 1674-1679. Hutchison, WD (1998) Microelectrode techniques and findings of globus pallidus. In: JK Krauss, RG Grossman and J Jankovic (Eds.). Lippincott-Raven, Philadelphia, pp. 135-152. Hutchison, WD, Lozano, CA, Davis, KD, Saint-Cyr, JA, Lang, AE and Dostrovsky, JO (1994a) Differential neuronal activity in segments of globus pallidus in Parkinson's disease patients. Neuroreport, 5: 15331537. Hutchison, WD, Lozano, AM, Kiss, ZHT, Davis, KD, Lang, AE, Tasker, RR and Dostrovsky, 10 (1994b) Tremor-related activity (TRA) in globus pallidus of Parkinson's disease (PD) patients. Soc. Neurosci. Abstr., 20: 783. Hutchison, WD, Lozano, AM, Tasker, RR, Lang, AE and Dostrovsky, JO (1997) Identification and characterisation of neurons with tremor-frequency activity in human globus pallidus. Exp. Brain Res., 113: 557563. Hutchison, WD, Allan, RJ, Opitz, H, Levy, R, Dostrovsky, JO, Lang, AE and Lozano, AM (1998a) Neurophysiological identification of the subthalamic nucleus in surgery for Parkinson's disease. Ann. Neurol., 44: 622-628. Hutchison, WD, Benko, R, Dostrovsky, JO, Lang, AE, and Lozano, AM (1998b) Coherent relation of rest tremor

and pallidal tremor cells in Parkinson's disease patients. Mov. Dis., 13 (Suppl. 2): 204. Krauss, JK, Grossman, RG and Jankovic, J (1998) Pallidal surgery for the treatment of Parkinson's disease and movement disorders. Lippincott-Raven, Philadelphia, 324 pp. Laitinen, LV (1990) Ventroposteromedial pallidotomy in Parkinsons disease. Stereotact. Funet. Neurosurg., 54+55. Laitinen, LV, Bergenheim, AT and Hariz, MI (1992) Leksell's posteroventral pallidotomy in the treatment of Parkinson's disease. J. Neurosurg., 76: 53-61. Lalley, PM, Moschovakis, AK and Windhorst, U (1999) Electrical activity of individual neurons in situ: extraand intracellular recording. In: U Windhorst and H Johansson (Eds.), Modem Techniques in Neuroscience Research. Springer, Berlin, pp. 127-172. Lenz, FA, Dostrovsky, JO, Kwan, HC, Tasker, RR, Yamashiro, K and Murphy, JT (1988) Methods for microstimulation and recording of single neurons and evoked potentials in the human central nervous system. J. Neurosurg., 68: 630-634. Levy, R, Hutchison, WD, Lozano, AM and Dostrovsky, JO (2000) High-frequency synchronization of neuronal activity in the subthalamic nucleus of parkinsonian patients with limb tremor. J. Neurosci., 20: 77667775. Lozano, AM (2000) Movement Disorder Surgery. Karger, Basel, 404 pp. Lozano, AM, Hutchison, WD, Kiss, ZHT, Davis, KD and Dostrovsky, JO (1996) Methods for microelectrodeguided posteroventral pallidotomy. J. Neurosurg., 84: 194-202. Ma, T (1996) Saccade-related omnivectoral pause neurons in the primate zona incerta. NeuroReport, 7: 27132716. Millar, J (1992) Extracellular single and multiple unit recording with microelectrodes. In: JA Stamford (Ed.), IRL Press at Oxford. New York, pp. 1-27. Nicolelis, MA, Stambaugh, CR, Brisben, A and Laubach, M (1998) Methods for simultaneous multi site neural ensemble recordings in behaving primates. In: MA Nicolelis (Ed.). CRC Press LLC, Boca Raton FLA, pp. 121-156. Parent, A and Hazrati, LN (1995) Functional anatomy of the basal ganglia. I. The cortico-basal ganglia-thalamocortical loop. Brain Res. Brain Res. Rev., 20: 91-127. Raeva, SN and Lukashev, A (1993) Unit activity in human thalamic reticularis neurons. II. Activity evoked by significant and non-significant verbal or sensory stimuli. Electroencephalogr. Clin. Neurophysiol., 86:

lID-I 22. Raeva, SN, Lukashev, A and Lashin, A (1991) Unit activity in human thalamic reticular nucleusI Spon-

138 taneous activity. Electroencephalogr. Clin. Neurophysiol.,79: 133-140. Schaltenbrand, G and Wahren, W (1977) Atlas for Stereotaxy of the Human Brain. Georg Thieme, Stuttgart, 69 plates. Schultz, W (1986) Responses of midbrain dopamine neurons to behavioural trigger stimuli in the monkey. J. Neurophysiol., 56: 1439-1461. Starr, P (1999) Instrumentation, technique, and technology. Axon Guideline System 3000. Neurosurgery, 44: 1354-1356. Sterio, D, Beric, A, Dogali, M, Fazzini, E, Alfaro, G and Devinsky, 0 (1994) Neurophysiological properties of pallidal neurons in Parkinson's disease. Ann. Neurol., 35: 586-591.

WO. HUTCHISON ET AL.

Taha, JM, Favre, J, Baumann, TK and Burchiel, KJ (1996) Characteristics and somatotopic organization of kinesthetic cells in the globus pallidus of patients with Parkinson's disease. J. Neurosurg., 85: 1005-1012. Tasker, RR, Davis, KD, Hutchison, WD and Dostrovsky, 10 (1998) Subcortical and thalamic mapping in functional neurosurgery. In: PL Gildenberg and RR Tasker (Eds.). McGraw-Hill, New York, pp. 945-94-31. Vitek, JL, Bakay, RA, Hashimoto, T, Kaneoke, Y, Mewes, K, Zhang, JY, Rye, D, Starr, P, Baron, M, Turner, Rand DeLong, MR (1998) Microelectrode-guided pallidotomy: technical approach and its application in medically intractable Parkinson's disease. J. Neurosurg., 88: 1027-1043.

Movement Disorders Handbook of Clinical Neurophysiology, Vol. 1 M. Hallett (Ed.) © 2003 Elsevier B.V. All rights reserved

139 CHAPTER 10

Polysomnography and related procedures s. Chokroverty* Department of Neurology/Cronin 466, St. Vincents Hospital, New York, NY 10011, USA

Scientific progress in the laboratory evaluation of sleep and its disorders has been rather slow but great advances have been made in the last century. The driving forces in this understanding have been the discovery of the human electroencephalogram (EEG) by Berger (1929) and rapid eye movements (REMs) during sleep by Aserinsky and Kleitman (1953). Polysomnography (PSG) has come to be viewed as the single most important laboratory technique for assessment of sleep and its disorders as well as for diagnosis and differential diagnosis of abnormal movements during sleep at night. PSG refers to recordings of multiple physiological characteristics during sleep whereas polygraphy denotes recordings of similar characteristics during any time of the day. The first polygraphic study to record motor activities during sleep was probably reported by Oswald in 1959 under the title of "sudden bodily jerks on falling asleep". In this chapter I briefly outline PSG recording techniques, indications for PSG, simultaneous video-PSG, and pertinent PSG findings in selected sleep disorders, computerized PSG, recording artifacts and related laboratory procedures for assessment of patients with movement disorders with or without complex behavior during sleep including multiple sleep latency test (MSLT), maintenance of wakefulness test (MWT) and actigraphy.

determine the patient's perception of quality of sleep and the actual test results. In order to assess chronic daytime sleepiness, the patient is asked to fill out the Epworth Sleepiness Scale (ESS) (Johns, 1991), which contains questions relating to the likelihood of dozing off in situations such as riding as a passenger in a car and watching television (Table 1). PSG allows assessment of sleep stages and wakefulness, respiration, cardiocirculatory functions and body movements (Keenan, 1999). EEG, electrooculogram (EOG) and electromyogram (EMG) of the chin muscle are recorded to study and score sleep staging (Rechtschaffen and Kales, 1968). Respiratory recording includes measurement of airflow and respiratory effort (Parisi, 1999; Kryger, 2000). PSG Table I Epworth sleepiness scale. Eight situations

Scores*

1. Sitting and reading 2. Watching television 3. Sitting in a public place (e.g., a theater or a meeting) 4. Sitting in a car as a passenger for an hour without a break 5. Lying down to rest in the afternoon

10.1. Techniques of PSG recording

PSG records multiple simultaneous physiological characteristics during sleep at night (Keenan, 1999). A pre- and post-study sleep questionnaire helps * Correspondence to: Prof. S. Chokroverty, Dept. of Neurology/Cronin 466, St. Vincents Hospital, 153 W l lth Street, New York, NY 10011, USA. E-mail address:[email protected] Fax: + 1 (212) 604-1555.

6. Sitting and talking to someone 7. Sitting quietly after a lunch without alcohol 8. In a car, while stopped for a few minutes in traffic

* Scale to determine total scores: 0= would never doze; I = slight chance of dozing; 2=moderate chance of dozing; 3=high chance of dozing. Source: Adapted from M.W. Johns. A new method for measuring daytime sleepiness: the Epworth Sleepiness Scale. Sleep 1991; 14: 540-545.

140

S. CHOKROVERTY

also records electrocardiogram (EKG), finger oxymetry, limb muscle activity, particularly EMG of the tibialis anterior muscles bilaterally, snoring and body positions. Special techniques not used routinely include measurements of intraesophageal pressure, esophageal pH and measurements of penile tumescence for assessing patients with erectile dysfunction. Equipment for recoding PSG contains AC and DC amplifiers. The AC amplifiers are used to record physiological characteristics showing high frequencies such as EEG, EOG, EMG, and EKG. A DC amplifier is typically used to record potentials with slow frequency such as for recording the output from the oxymeter, pH meter or CPAP titration pressure changes and recording of intraesophageal pressure. AC or DC amplifiers may be used to record respiratory flow and effort. Sensitivity and filter settings vary according to the physiological characteristics recorded (Table 2). The standard speed for recording traditional PSG is 10 mm1s, so that each monitor screen or page is a 30-s epoch. In patients with suspected nocturnal seizures, however, a 30 mm/s recording speed (a lO-s epoch) is used for easy identification of epileptiform activity. Analog recording using paper is currently being replaced in most of the laboratories by digital system recordings. It is important to have facility for simultaneous video recording to monitor the behavior during sleep. It is advantageous to use two cameras for sleep screen viewing covering the entire body. A low light level camera should be used to obtain good quality video in the dark and an infrared light source should be available after turning the laboratory lights

off. The monitoring station should have a remote control, which can zoom, tilt or pan the camera for adequate viewing. The camera should be mounted on the wall across from the head end of the bed. An intercom from a microphone near the patient should be available.

10.2. Technique of recording of multiple physiological characteristics 10.2.1. Electroencephalography Most laboratories using international 10-20 electrode placement system recorded from at least 4 channels (C3-A2, C4-Al, Ol-Al and 02-A2) to clearly document the onset of sleep. Some laboratories use 8-10 channels to cover the parasagittal and the temporal areas of the brain to record possible focal or diffuse EEG abnormalities as well as for epileptiform activities. Both bipolar montage and referential montage connecting the electrodes between an active and a relatively inactive site (e.g. AI, A2, Cz, pz) are recommended. The importance of multiple channel EEG recordings is to document focal or diffuse slow waves and particularly epileptiform discharges. Many patients are referred to the PSG laboratory for a possible diagnosis of nocturnal seizures. The standard recommended EEG recordings of 1-2 channels or even 4 channels of recordings will miss most of the epileptiform discharges during all night recording. Therefore, in patients suspected of nocturnal seizures, polysomnographic study should include multiple channels of EEG covering temporal and parasagittal regions and simultaneous video record-

Table 2 Filter and sensitivity settings for polysomnographic studies. Time constant (s)

Lowfrequency filter (Hz)

Sensitivity

70 or 35

0.4

0.3

5-7 u.V/mm

70 or 35

0.4

0.3

5-7 u.V/mm

90

0.04

5.0

2-3 fLV/mm

15

0.12

1.0

1 mV/cm to start; adjust

0.1

5-7 fLV/mm; adjust

Characteristics

Highfrequency filter (Hz)

Electroencephalogram Electro-oculogram Electromyogram Electrocardiogram Airflow and effort

15

141

POLYSOMNOGRAPHY AND RELATED PROCEDURES

ing (video-PSG study) for correlation of the electrical activity with the actual behavior of the patients. In computerized (digital) PSG recordings that are now performed in most laboratories it is easy to change the recording speed from the standard 10 mm/s of the usual sleep recording to 30 mm/s of the standard EEG recording for proper identification of the epileptiform discharges. Special seizure montages with full complement of standard electrodes and special electrode placements (e.g. T1 and T2 electrodes) should be used. The value of videoPSG for the diagnosis of seizure disorders and parasomnias has been clearly documented by Aldrich and Jahnke (Aldrich and Jahnke, 1991). 10.2.2. Electro-oculography (EDG)

EOG records corneoretina1 (relative positivity at the cornea and a relative negativity at the retina) potential difference (Walczak and Chokroverty, 1999). A typical electrode placement is one em superior and lateral to the outer canthus of one eye with a second electrode placed 1 em inferior and lateral to the outer canthus of the opposite eye. Both these electrodes are then connected to a single reference electrode, either the same ear or the mastoid process of the temporal bone. Therefore, right outer canthus (ROC) and left outer canthus (LaC) electrodes are referred to either A1 or A2. In this arrangement, conjugate eye movements produce out-of-phase deflections in the two channels whereas the EEG slow activities contaminating the eye electrodes are in-phase. Both conjugate horizontal and vertical eye movements are detected by this placement scheme. The sensitivity and filter settings for EOG are similar to those used for EEG (see Table 2).

10.2.3. EMG recordings during standard PSG

EMG activity is an important physiological characteristic that needs to be recorded for sleep staging as well as for diagnosis and classification of a variety of sleep disorders. In a standard PSG recording, EMGs are recorded from mentalis or submental and right and left tibialis anterior muscles. Mental or submental EMG activity is monitored to record axial muscle tone, which is significantly decreased during REM sleep, and, therefore, an important physiological characteristic for identifying REM sleep. Additional electrodes over the masseter muscles

may be needed in patients with bruxism (tooth grinding) to document bursts of EMG activities associated with bruxism. For recording from tibialis anterior muscles, surface electrodes are used and the distance between the two electrodes is 2-2.5 em. Bilateral tibialis anterior EMG is important to record in patients suspected of restless legs syndrome (RLS) because the periodic limb movements in sleep (PLMS), which are noted in 80% of such patients, may alternate between the two legs. Ideally, the recording should also include one or two EMG channels from the upper limbs in patients with RLS as occasionally PLMS are noted in the upper limbs. For patients with suspected REM behavior disorder, multiple EMGs from all four limbs are essential as there is often a dissociation of the activities between upper and lower limb muscles in such patients. If the upper limb EMGs are not included in patients suspected of REM behavior disorder, REM sleep without atonia may be missed in some cases. In patients presenting with abnormal movements such as dystonic, choreoathetoid or ballismic movements as noted in patients with nocturnal paroxysmal dystonia (a type of frontal lobe seizure disorder), multiple muscle surface EMG recordings in addition to video-PSG recordings may be obtained to document such activities. Other EMG recordings include intercostal and diaphragmatic EMG to record respiratory muscle activities. EMG shows progressively decreasing tone from wakefulness through stages I-IV of NREM sleep. In REM sleep the EMG is markedly diminished or absent. In REM behavior disorder, a characteristic finding is absence of muscle atonia during REM sleep in the EMG recording and the presence of phasic muscle bursts repeatedly during REM sleep. 10.2.4. Electrocardiography

A single channel of EKG is sufficient during PSG recording by placing one electrode over the sternum and the other electrode at a lateral chest location. Gold cup surface electrodes are used to record the EKG. Table 2 lists the filter settings and sensitivities for such recording. 10.2.5. Respiratory monitoring technique

PSG recording must routinely include methods to monitor airflow and respiratory effort adequately to

142

correctly classify and diagnose sleep disordered breathing (SDB). Respiratory effort can be measured by mercury-filled or piezoelectric strain gauges, inductive plethysmography, impedance pneumography, respiratory magnetometers and respiratory muscle EMG (Kryger, 20(0). Most commonly Piezoelectric strain gauges and inductive plethysmography are used to monitor respiratory effort. Airflow can be measured by thermistors, thermocouples or a nasal cannula-pressure transducer recording nasal pressure (Kryger, 2000). The best way to detect arterial O2 content (Pa02) is by invasive method using an arterial cannula. This is not viable from the practical standpoint and in any case intermittent sampling of blood through the cannula may not reflect the severity of hypoxemia during a particular disordered breathing event. Therefore, noninvasive method by finger pulse oximetry is routinely used to monitor arterial oxygen saturation (Sa02) or arterial oxyhemoglobin saturation, which reflects the percentage of hemoglobin that is oxygenated (Kryger, 2000). 10.2.6. Body position monitoring

Body position is monitored by placing sensors over one shoulder and using a DC channel. Snoring and apneas are generally worse in the supine position and therefore, CPAP titration must include observing patients in the supine position for evaluating optimal pressure for titration. 10.2.7. Snoring

This can be monitored by placing a miniature microphone on the patient's neck. There is no general accepted standardized technique to record quantitatively the intensity of snoring. 10.2.8. Beginning and ending the PSG

All important information including "lights out" and "lights on" as well as any unusual behavior and motor events should be clearly documented by the technologist before and during the recording. This must include the patient's name, age, date of study, the identification number, purpose of the recording (before referring the patient to the laboratory for PSG study, sleep clinicians should have performed complete history and physical examination formulat-

S. CHOKROVERTY

ing a provisional diagnosis) and the name of the technician. When awakened in the morning (either spontaneously or at a set time), the patient is asked to fill out a post-study questionnaire which includes estimation of time to fall asleep, total sleep time, number of awakenings and the quality of sleep

10.3. Sleep stage scoring technique The gold standard for scoring a particular sleep stage is still that recommended in the manual by Rechtschaffen and Kales (R-K) in 1968 (Rechtschaffen and Kales, 1968) following recommendations by an ad hoc committee. This was originally devised for sleep scoring in normal adults. There are, however, serious limitations using the R-K manual for scoring sleep stages in pathological states and many investigators recommend a computerized rather than the manual scoring technique. However, computerized techniques have many pitfalls and have not been accepted as a gold standard yet. In the future, with more sophisticated development of computer technology, a computer scoring technique most likely will supercede the manual scoring technique. Sleep stage scoring is based on three physiologic criteria: EEG, EOG and EMG. For R-K scoring an EEG recorded at C3/A2 or C4/Al should be used, especially for the purpose of amplitude criteria. The recommendation is for an epoch by epoch scoring and the most commonly used epoch is 30 s. 10.3.1. Scoring ofperiodic limb movements in sleep

Periodic limb movements in sleep are involuntary movements periodically recurring during sleep and are counted from the right and left tibialis anterior EMG recordings. The scoring criteria for PLMS can be summarized as follows (Atlas Task Force of the American Sleep Disorders Association, 1993): the movements must occur as part of 4 consecutive movements; the duration of each EMG burst should be 0.5 to 5 s; the interval between bursts should be 4-90 s; the amplitude of the EMG bursts, although variable, should be more than 25% of the EMG bursts recorded during the pre-sleep calibration recording. PLMS (Fig. 1) mayor may not be associated with arousals, and they should be scored separately. To score PLMS associated with arousal

POLYSOMNOGRAPHY AND RELATED PROCEDURES

143

Fig. 1. A portion of polysomnographic fragment showing periodic limb movements in sleep (PLMS). Left (LT) and right (RT) tibialis anterior (TIB) electromyography (EMG). The top 4 channels show EEG (international nomenclature). LOC: Left oculogram; ROC: Right oculogram. CHIN: Chin EMG. EKG: Electrocardiogram. ABD: Abdomen. S.02: Oxygen saturation in percent by finger oxymetry.

the arousal must occur within 3 s of onset of PLMS. PLMS is expressed as an index consisting of number of PLMS per hour of sleep. To be of pathologic significance PLMS index should be 5 or more. Leg movements may be noted occurring periodically associated with resumption of breathing following recurrent episodes of apneas or hypopneas. These respiratory related leg movements should not be counted as PLMS. PLMS are generally seen during NREM sleep but they can occur rarely during REM sleep. In patients with restless legs syndrome, however, periodic limb movements may occur during wakefulness when they are termed period limb movements in wakefulness (PLMW). 10.3.2. Indications for PSG and video-PSG

In addition to the standard indications for PSG as published in the guidelines (Indications for Poly-

somnography Task Force, 1997) by the American Academy of Sleep Medicine (e.g. suspected SDB, patients with excessive daytime sleepiness (EDS), CPAP titration in patients with obstructive sleep apnea syndrome (OSAS), prior to surgical procedures or dental appliances in patients with OSAS, suspected narcolepsy-cataplexy syndrome and atypical or violent parasomnias), video-PSG combining PSG with multiple EEG channels and simultaneous video recording is very useful in patients with abnormal movements and behavior during sleep at night. If these motor activities during sleep occur frequently, the changes of capturing these events in the video-PSG are much better than if these had been occurring infrequently. The video recording can include multiplex analog signal captured on a tape but currently many commercially available systems include digital video directly synchronizing and time-locking the abnormal behavior to the PSG

S. CHOKROVERTY

144 Table 3

Table 3

Classification of abnormal motor activities during sleep.

Continued.

Motor parasomnias

Drug-induced nocturnal dyskinesias

1. Sleep-wake transition disorders a. Rhythmic movement disorder b. Sleep talking (somniloquy) c. Nocturnal leg cramps d. Propriospinal myoclonus at the transition from wakefulness to drowsiness 2. NREM sleep parasomnias a. Confusional arousals b. Sleepwalking c. Sleep terror 3. REM sleep parasomnias a. Nightmares b. REM sleep behavior disorder 4. Diffuse parasomnias (no stage preference) a. Bruxism b. Neonatal sleep myoclonus

1. Levodopa-induced myoclonus in Parkinson's disease 2. Medication-induced (e.g. tricyclic antidepressants, levodopa, lithium) periodic limb movements in sleep (PLMS)

Restless legs syndrome (RLS) - Periodic limb movements A. Nocturnal jerks and body movements in obstructive sleep apnea syndrome B. Excessive fragmentary myoclonus seen in a variety of sleep disorders C. Sleep-related panic attacks D. Dissociative disorders E. Fatal familial insomnia F. Post-traumatic stress disorder G. Narcolepsy-cataplexy-sleep paralysis syndrome

Nocturnal seizures 1. True nocturnal seizures a. Tonic seizure b. Benign rolandic seizure c. Autosomal dominant nocturnal frontal lobe seizure d. Nocturnal frontal lobe epilepsy (Nocturnal paroxysmal dystonia) e. Paroxysmal arousals and awakenings f. Episodic nocturnal wanderings g. Electrical status epilepticus in sleep 2. True nocturnal and diurnal seizures (diffuse seizures) a. Generalized tonic-clonic seizure b. Myoclonic seizure c. Infantile spasms (West's syndrome) d. Partial complex seizure e. Frontal lobe seizure f. Epilepsia partialis continua 3. Pseudoseizure (psychogenic non-epileptic seizure)

Involuntary movement disorder 1. Always persisting during sleep: a. Palatal myoclonus or palatal tremor 2. Frequently persisting during sleep a. Spinal and propriospinal myoclonus b. Tics in Tourette's syndrome c. Hemifacial spasms d. Hyperekplexia or exaggerated startle syndrome 3. Sometimes persisting during sleep a. Tremor b. Chorea c. Dystonia d. Hemiballisms

signals. Depending on the availability of the channels and the electrode inputs in the equipment multiple channels of EEGs (e.g. for suspected nocturnal seizure disorder) and EMGs to include additional muscles (e.g. to record from forearm flexor and extension muscles, masseter and other muscles for patients with suspected rapid eye movement behavior disorder (RBD) and bruxism) are recommended. Video-PSG may help characterize the movements, differentiate one jerk from another, identify a specific entity and most importantly differentiate abnormal motor activities from nocturnal seizures. Video-PSG may aid in the diagnosis of other co-existing sleep disorder, e.g. OSAS, RBD, narcolepsy. Video-PSG thus helps us classify abnormal motor activities during sleep into several identifiable entities (Table 3), e.g. motor parasomnias, noctural seizures, involuntary diurnal movements persisting during sleep, PLMS, excessive fragmentary myoclonus seen in a variety of sleep disorders, dissociative disorders, nocturnal jerks and body movements seen in patients with OSAS. Many parasomnias (defined as abnormal movements and behavior introducing into sleep without necessarily disrupting sleep architecture) may be mistaken for nocturnal seizures. For example, confusional arousals, sleep walking, sleep terror, sleep talking, bruxism, rhythmic movement disorder, RBD, nightmares and dissociative

145

POLYSOMNOGRAPHY AND RELATED PROCEDURES

Table 4 Indications for video PSG. • Unusual and complex arousal disorders • Complex behaviors suspicious of RBD but not absolutely certain based on the history • Behavior and motor events at night suggesting possible nocturnal seizure disorder • EDS in patients with epilepsy to determine if excessive sleepiness is due to repeated nocturnal seizures, an undesirable side effect of antiepiieptic medications or due to an associated sleep disorder (e.g. sleep apnea) • Suspected psychogenic dissociative disorder • Other motor parasomnias (e.g. rhythmic movement disorder, bruxism) which may be mistaken for nocturnal seizures • Involuntary diurnal movement disorder persisting during sleep • Coexisting second sleep disorder (e.g. narcolepsy and RBD, OSAS and sleep walking, narcolepsy and sleep apnea) • For medicolegal purpose when the patient presents with violent behavior during sleep, video-PSG studies are mandatory to evaluate such patients for correct diagnosis of parasomnias or seizure disorders

disorders may be mistaken for seizures. RBD and nightmares occur during REM sleep. These conditions can be diagnosed and differentiated from one another based on characteristic clinical features combined with EEG and video-PSG findings. Table 4 lists indications for video-PSG. There is some controversy regarding the diagnosis of periodic limb movement disorder (PLMD) causing sleep fragmentation, arousals and excessive daytime sleepiness. Periodic limb movements in sleep (PLMS) have been noted in a number of sleep disorders as well as in normal individuals, particularly in patients over 60-65. Although the specificity of PLMS is not defined, at least 80% of patients with RLS show PLMS on PSG recordings. Therefore to document PLMS in RLS, PSG may be indicated. However, the diagnosis of RLS is a clinical one and has been based on international study group criteria (Allen et aI., 2003). PLMS as well as a sleep disturbance are not part of the essential criteria for the diagnosis of RLS. PSG indications, therefore, for RLS-PLMS remain somewhat dubious and contentious. Some investigators do not believe in the existence of PLMD as a

separate sleep disorder causing sleep dysfunction and EDS (Mahowald, 2002). The indications for pure PLMS or PLMD currently remain undetermined and further investigations including outcome studies are needed to document that PLMS or PLMD may cause sleep disturbance and EDS. Usually if one spends sufficient time in history taking and examining the patient it is possible to make a clinical diagnosis. However, even in clinically obvious cases it is important to confirm the clinical diagnosis by laboratory tests before instituting therapy because inappropriate or incorrect treatment may cause adverse side effects without necessarily helping the patient. On many occasions, however, the spells are atypical, unusual and often violent requiring video-PSG confirmation of the events. Correlation of the events with the time of the night and a particular sleep staging is important for correct diagnosis. For example, arousal disorder occurs during slow wave sleep in the first third of the night and RBD occurs during REM sleep usually in the last half to last third of the night. Rhythmic movement disorder (head banging, head rolling, body rocking) occurs during sleep stage transition from any stage of sleep whereas during psychogenic dissociated disorder EEG shows a wakeful pattern. The distinct disadvantage of video-PSG is additional expense and technologist time to place additional electrodes for extended EEG montage for suspected nocturnal seizure disorder and extended EMG montage to record multiple additional muscles for suspected RBD and bruxism. 10.4. Characteristic PSG findings in nocturnal movement disorders

There are no distinctive PSG patterns noted in most nocturnal movement disorders except characteristic epileptiform EEG patterns in nocturnal seizures. PSG findings are diagnostically nonspecific but video-PSG findings may help in the diagnosis, differential diagnosis, and in understanding pathophysiology. 1004.1. PSG findings in RLS-PLMS

These patients may show delayed sleep onset, fragmented sleep with repeated arousals and PLMS, which are noted in at least 80% of RLS patients. PLMS index (number of PLMS per hour of sleep)

146

below 5 is considered normal. Based on the PLMS index the severity of PLMS may be classified into mild (index of 5-25), moderate (index of more than 25-50) and severe (index of more than 50). 10.4.2. PSG findings in parasomnias and dissociative disorders

In NREM parasomnias (e.g. sleep walking, sleep terror and confusional arousals), spells arise out of slow wave sleep and are not stereotyped but are prolonged lasting for minutes and may be up to 10 min in contrast to patients with seizure disorders who may show stereotyped behavior lasting for a few seconds to a minute or two. Furthermore, EEG shows no evolving pattern in the arousal disorders, thus, differentiating from ictal EEG with rhythmically evolving pattern of slow waves, sharp waves or

S. CHOKROVERTY

spikes. High voltage delta waves, non-reactive alpha, stage I sleep and movement artifacts in the EEG are the other features found in patients with arousal disorders. In patients with REM behavior disorder (RBD) the behavior occurs out of REM sleep and, therefore, the EEG shows the characteristic REM sleep pattern of desynchronized EEG containing a mixture of alpha, beta and theta activities often associated with characteristic "saw-tooth" waves and rapid eye movements. The EMG shows absence of muscle atonia, phasic muscle bursts and the video may document excessive limb movements, which may be rhythmic or arrhythmic. Figure 2 is a representative sample from a patient with RBD. Psychogenic dissociative disorders may present with a variety of abnormal movements resembling those noted in frontal lobe seizures. However, in

Fig. 2. A fragment of polysomnographic tracing from a patient REM sleep behavior disorder. Note sustained muscle tone and phasic EMG bursts in the electromyograms from chin, left (L) and right (R) arms, left and right tibialis anterior muscles during REM sleep. EDGs (top two channels) show rapid eye movements and EEG (channels 3-6 from the top) is in stage 1 with a mixture of theta, alpha and beta rhythms. (Reproduced with permission from Drs. Carlos Schenck and Mark Mahowald.)

147

POLYSOMNOGRAPHY AND RELATED PROCEDURES

dissociative disorders the movements are nonstereotyped and the EEG shows wakeful patterns both during and after the spells. Rhythmic movement disorder can arise during sleep-wake state transition or rarely during sleep stage transition. This condition may mimic the behavior pattern of partial complex seizure but EEG shows no ictal pattern, thus, differentiating rhythmic movement disorder from patients with partial complex seizure. Orofacial movements in patients with bruxism sometimes resemble orofacial automatisms of partial complex seizure. However, EEG in bruxism shows no ictal pattern in contrast to patients with partial complex seizure. EMG in bruxism shows characteristic rhythmic bursts in the masseter muscles and these myeogenic artifacts are also reflected in the EEG electrodes.

10.4.3. PSG findings in diurnal movement disorders There has been a growing awareness amongst both movement disorder and sleep specialists about an interaction between sleep and the movement disorders. There is increasing understanding about the effects of diurnal movements on sleep and the effect of sleep on a variety of diurnal movement disorders, and how sleep-wake states modulate the daytime and night-time abnormal movements. A case in point is the recent controversy about "sleep attacks" in Parkinson's disease (PD) patients on newer dopamine agonists. Whether these agents or the disease itself are responsible for excessive sleepiness or unpredictable "sleep attacks" remain controversial but this controversy has re-emphasized the presence of sleep dysfunction which may be seen in up to 70% to 90% of PD patients. Overnight PSG findings in PD include decreased slow wave and REM sleep, reduced sleep spindles, decreased sleep efficiency, disruption of NREM-REM sleep cycling, rapid blinking at sleep onset, sleep fragmentation, and REM-onset blepharospasm. In those presenting with RBD the EMG shows absence of muscle atonia and increased phasic EMG bursts during REM sleep. In addition, some PD patients may document SDB and PLMS. Parkinsonian tremor may persist during NREM stages I and II, is absent in slow wave and REM sleep but may reappear during sleep stage transition.

PSG findings in progressive supranuclear palsy include increased sleep latency, repeated arousals and awakenings, decreased NREM stage I and REM sleep, decreased sleep spindles, in some patients, reduced REM latency and occasionally sleep apnea. In Tourette's syndrome, PSG recordings shows increased body movements and motor tics during all stages of sleep. An increased number of awakenings and mild reduction of REM sleep may also be seen. There is an increased prevalence of sleep walking and sleep terror in these patients. PSG findings in Huntington's chorea include sleep fragmentation with progressive deterioration as the disease progresses. Other findings include decreased sleep efficiency in 48%-80% of cases and a mild reduction of REM sleep. Persistence of the involuntary movements in NREM stages I and II and reemergence during REM sleep may also be seen. Sleep spindles are increased in amplified and density. PSG findings in torsion dystonia consist of prolonged sleep latency, repeated awakenings, reduced sleep efficiency and decreased REM sleep. Dystonic movements decrease during NREM stages I and II, and are absent during slow wave and REM sleep.

10.4.4. Computerized PSG The advantages of computerized PSG include easy data acquisition, display and storage (Hirshkowitz and Moore, 1994; Hasan, 1996; Kubicki and Hermann, 1996; Hirshkowitz and Moore, 1999; Penzel and Conradt, 2000; Agarwal and Gotman, 2002). It is easy to review the record on-line (i.e, the ability to look back at early tracing during progression of recording). The other advantages include the ability to manipulate large quantity of data for review and storage for permanent record keeping. It is also easy to review using a variety of filter settings, sensitivities, monitor speeds and reformatted montages (i.e, new montages may be created retrospectively from the electrode derivations used during actual recording). In a particular segment with a question of potential epileptiform event (e.g. spikes, sharp waves, spike and waves and sharp and slow waves) the standard PSG speed of 10 mm/s can be quickly changed to the usual EEG speed of 30 mm/s for recognizing the evolving pattern of activation for a correct diagnosis. These capabilities help

148

identify an abnormal EEG pattern and distinguish artifacts from true cerebral events. Computerized PSG makes it easy to document all events, to edit and report. The computerized PSG can store the raw data on relatively inexpensive CD-ROM or other suitable media making it easy to keep database and access raw data. European data format (EDF) is the most common format for exchange of digital PSGs between different laboratories in different countries. Simultaneous video monitoring during PSG recording is essential to obtain patient's behavior and motor manifestations, particularly in patients with parasomnias and nocturnal seizures. Recently introduced digital video incorporated into the computerized PSG system is an important advance over the traditional video tape recording. Digital video recordings, however, use a lot of space on the hard disk and one way to handle this problem is to save only small segments of digital video. The latest digital versatile disk (DVD) may solve the storage problems in the future. In order to overcome the limitations and fallacies of the R-K system and to reduce the time for scoring, automatic computer-assisted scoring techniques have been proposed (Hasan, 1996; Kubicki and Hermann, 1996; Hirshkowitz and Moore, 1999; Penzel and Conradt, 2000) and are commercially available. Some of the later methods are still evolving but none of the techniques have received wide popularity because of serious limitations in obtaining an acceptable scoring and because of lack of standardization, validation and precision in the methods. Some of the problems in computer-assisted scoring include artifact recognition, differentiating stage I NREM sleep from REM sleep, discriminating different sleep stages, inability to differentiate eye movements from high amplitude delta waves and failure to detect upper airway resistance. Attempts have also been made to develop computer methods to identify obstructive, central and mixed apneas and hypopneas as well as arousals and PLMS but the methods remain ambiguous, imprecise, and variable resulting in lack of universal acceptance yet. Computerized scoring has no real gold standard to compare the data. Another disadvantage of computer scoring is that there is no standardized procedure for scoring of various physiological characteristics. Furthermore, for comparison between visual and computerized scoring sampling remains a problem.

S. CHOKROVERTY

R-K manual scoring still remains today the gold standard in clinical practice.

10.5. Artifacts during PSG recording Artifacts refer to extraneous electrical activities not recorded from the regions of interest (e.g. the brain, muscles, eyes and heart). These extraneous electrical activities may obscure the biological signals of interest and, therefore, recognition and correction of these artifacts is an important task for the polysomnographic technologist. The artifacts can be divided into three categories: physiological, environmental and instrumental (Keenan, 1999; Walczak and Chokroverty, 1999). 10.5.1. Physiological artifacts

These include myogenic potentials, artifacts resulting from movements of the head, eyes, tongue, mouth and other body parts, sweating, pulse and EKG artifacts as well as rhythmic tremorogenic artifacts. 10.5.2. Environmental sources of electrical signals

These may simulate electrocerebral activity or may obscure the EEG activities and include 60 Hz (or 50 Hz), artifacts resulting from the telephone or the pager systems. Electrostatic artifacts result from movements of the subjects in the environment. Most important is keeping the impedance of recording electrodes below 5 K. 10.5.3. Instrumental artifacts

These arise from faulty electrodes, electrode wires, switches and the polygraph machine itself. A very common artifact is electrode "pops" which are transient sharp waves or slow waves limited to one electrode. These artifacts result from faulty electrode placement or insufficient electrode gel causing abrupt changes in impedance. The electrode should be reset and gel applied. If this persists then electrodes need to be changed. Other sources of artifacts are the electrode wires, cables and the switches. In the PSG machine random fluctuations of charges result in some instrumental noise artifacts. If the sensitivity is greater than 2 microvolts per mm, which are not generally used in PSG recordings, then these instrument artifacts may

149

POLYSOMNOGRAPHY AND RELATED PROCEDURES

interfere with the recording. Loose contacts in switches or wires may also cause sudden changes in voltage or loss of signal. 10.6. Pitfalls of PSG

Polysomnography is the single most important laboratory test for assessment of sleep disorders, particularly in patients presenting with excessive daytime somnolence and those suspected of nocturnal seizures, parasomnias or other abnormal motor activities. However, PSG has considerable limitations (Hinmanen and Hasan, 2000; Hirshkowitz, 2000). There is no standardized uniform protocol used in all sleep laboratories and this may make the comparison of the data from one laboratory to another somewhat misleading. The most serious limitation is that the ovemight in-laboratory PSG is labor intensive, time consuming and expensive. A single night's PSG may miss the diagnosis of mild OSAS, PLMS, parasomnias or nocturnal seizures. PSG data and patient's clinical findings may not be concordant. PSG data may be confounded by the first night effects (e.g. increased wakefulness and stage I NREM sleep and decreased slow wave and REM sleep). 10.7. Multiple sleep latency test

The most common indication for referring a patient for multiple sleep latency test (MSLT) is excessive daytime sleepiness (EDS), although sleep onset and sleep maintenance insomnia is the most common complaint in the general population. The initial step in assessment of the patient with EDS is a detailed sleep history and other history and physical examination. For assessment of persistent sleepiness the Epworth Sleepiness Scale (ESS) (Johns, 1991) is often used to assess a general level of sleepiness. This is a subjective propensity to sleepiness assessed by the patient under eight situations on a scale of 0-3, with three indicating a situation when chances of dozing off are highest. The maximum score is 24 and a score of 10 suggests the presence of EDS. This test has been weakly correlated with MSLT scores. The ESS and MSLT, however, test different types of sleepiness. MSLT tests the propensity to sleepiness objectively, and ESS the general feeling of sleepiness or subjective propensity to sleepiness. The Stanford Sleepiness

Scale (SSS) (Keenan, 1999) is a 7 point scale to measure subjective sleepiness but it does not measure persistent sleepiness. Visual Analog Scale (Keenan, 1999) is the other scale used to assess alertness and wellbeing in which subjects indicate their feelings of alertness at an arbitrary point on a line of 0-100 mm scale with 100 being the maximum sleepiness and 0 being the most alertness. 10.7.1. Technique of MSLT

The MSLT has been standardized and includes several general and specific procedures (Carskadon et al., 1986; Cherbin, 2003). The test is preceded by an overnight polysomnographic study and is scheduled about 2-3 hours after the conclusion of the overnight PSG study. The actual test consists of 4-5 opportunities for napping at 2 hour intervals and each recording session lasts for a maximum of 20 min. Between tests subjects must remain awake. The measurements include average sleep onset latency and the presence of sleep onset rapid eye movements (SOREMs). If no sleep occurs then the test is concluded 20 min after lights out. Fifteen minutes after the first 30-s epoch of any stage of sleep the test is terminated. If the finding is indefinite then it is better to continue the test than to end it prematurely. Mean sleep latency is calculated as the average of the latencies to sleep onset for each of the 4-5 naps. Mean sleep latency of less than 5 min is consistent with pathologic sleepiness. A mean sleep latency of 10-15 min is considered normal; and a mean sleep latency of up to 5-10 min is consistent with mild sleepiness. The occurrence of REM sleep within 15 min of sleep onset is defined as SOREMs. 10.7.2. Indications for MSLT

Narcolepsy is the single most important indication for performing the MSLT (Thorpy, 1992). A mean sleep latency of less than 5 min combined with SOREMs in 2 or more of the 4-5 recordings during MSLT is strongly suggestive of narcolepsy, although REM sleep dysregulation and circadian rhythm sleep disorders may also lead to such findings. In patients with upper airway obstructive sleep apnea syndrome (OSAS) the MSLT is indicated to assess the degree of severity of daytime sleepiness. Sometimes the patients underestimate the presence

150

of sleepiness and often deny daytime symptoms. In such patients it is important to assess the daytime sleepiness so that the appropriate advice regarding the driving and avoidance of dangerous situations during the daytime can be given to the patient to prevent disastrous consequences. Patients with excessive daytime sleepiness with no easily ascertainable cause should also have an MSLT to document objectively EDS and findings suggesting REM sleep dysregulation. In idiopathic hypersomnia the MSLT findings will be consistent with pathologic sleepiness without SOREMs.

10.7.3. Reliability, validity and limitation of the MSLT The sensitivity and specificity of the MSLT in detecting sleepiness have not been clearly determined (Cherbin, 2003). The test-retest reliability of the MSLT, however, has been documented in both normal subjects and patients with narcolepsy. In subjects with sleepiness caused by circadian rhythm sleep disorders, sleep deprivation and ingestion of hypnotics and alcohol pathologic sleepiness has been validated by MSLT. However, there is poor correlation between the MSLT and ESS. The patient's psychological and behavioral state also interferes with the MSLT results. MSLT objectively measures tendency to sleep rather than the likelihood of falling asleep. If the patient suffers from severe anxiety or psychological disturbances causing behavioral stimulation, MSLT may not show sleepiness even in a patient complaining of EDS.

10.8. Maintenance of wakefulness test The MWT is a variant of the MSLT to measure a patient's ability to stay awake (Doghramji et al., 1997). Sleep latency is defined as in MSLT from lights out to the first epoch of any stage of sleep. It has generally been accepted that if the mean sleep latency is less than 11 min there is impairment of wake tendency. The MWT is useful in differentiating groups with normal daytime alertness from those with EDS. The MWT is more sensitive than MSLT in assessing the effects of treatments (e.g. CPAP titration in OSAS and the stimulant treatment for narcolepsy). It is less useful and less sensitive than the MSLT as a diagnostic test for narcolepsy. The MSLT and the MWT do have separate functions: the MSLT

S. CHOKROVERTY

unmasks physiologic sleepiness, which depends on both circadian and homeostatic factors whereas the MWT is a reflection of the individual's capability to resist sleep and is influenced by physiologic sleepiness.

10.8.1. Actigraphy Monitoring of body movements and other activities can be performed continuously for days, weeks or even months by using an actigraph, also known as actometer or actimeter (Standards of Practice Committee of the American Sleep Disorders Association, 1995). This can be worn on the wrist or alternatively on the ankle for recording arm, leg and body movements. The actigraph uses piezoelectric sensors, which function as accelerometers to record acceleration or deceleration of movements rather than the actual movement. The principle of analysis is based on the fact that increased movements are seen during wakefulness in contrast to markedly decreased movements or no movements during sleep. Several actigraph models are in developing stage to carefully regulate the sampling frequencies and duration, filters, sensitivities and the dynamic range in order to detect and quantify PLMS but no generally accepted standardized technique of quantifying and identifying PLMS discriminating from other movements (e.g. those resulting from parasomnias, nocturnal seizures and other dyskinesias) is currently available. Currently the role of actigraphy in detecting, quantifying and differientiating abnormal motor activities remains controversial but there is immense potential in the future for such applications with the development of sophisticated models and techniques.

10.9. Conclusion Patients presenting with abnormal movements during sleep constitute a group of the most challenging sleep disorders. Many such patients remain undiagnosed or misdiagnosed for years and are often subjected to inappropriate treatment. We must make an effort to correctly diagnose and classify such disorders. In this chapter, I briefly summarized an important laboratory procedure (e.g. PSG) which might be helpful in assessment of patients presenting with abnormal motor activities during sleep. I must, however, emphasize that any laboratory procedure

POLYSOMNOGRAPHY AND RELATED PROCEDURES

must act as facilitator and be subservient to the clinical approach to such patients. References Agarwal, Rand Gotman, J (2002) Digital tools in polysomnography. J. Clin. Neurophysiol., 19: 136143. Aldrich, M and Jahnke, B (1991) Diagnostic value of video-EEG polysomnography. Neurology, 41: 1060. Allen, RP, Piechietti, D and Hening, WA et al., and the International Restless Legs Syndrome Study Group (2003) Restless Legs Syndrome: diagnostic criteria, special considerations and epidemiology. Sleep Med., 4(2): 101-119. Aserinsky, E and Kleitman, N (1953) Regularly occurring periods of eye motility and concomitant phenomena during sleep. Science, 118: 273. Atlas Task Force of the American Sleep Disorders Association (1993) Recording and scoring leg movements. Sleep, 16: 748. Berger, H (1929) Uber das Elektroencephalogramm des Menschen. Arch. Psychiatr. Nervenkr., 87: 527-570. Carskadon, MA, Dement, WC and Mitler, M et al. (1986) Guidelines for the Multiple Sleep Latency Test (MSLT): a standard measure of sleepiness. Sleep, 9: 519. Cherbin, R (2003) Assessment of sleepiness. In: S Chokroverty, WA Hening and AS Walters (Eds.), Sleep and Movement Disorders. Butterworth-Heinemann; Boston. Doghramji, K, Mitler, M and Sangal, RB et al. (1997) A normative study of the maintenance of wakefulness test (MWT). Electroencephalogr. Clin. Neurophysiol., 103: 554. Hasan, J (1996) Past and future of computer-assisted sleep analysis and drowsiness assessment. J. Clin. Neurophysiol., 13: 295-313. Hirshkowitz, M (2000) Standing on the shoulders of giants: The standardized sleep manual after 30 years. Commentary. Sleep Med. Rev., 4: 169-179. Hirshkowitz, M and Moore, CA (1994) Issues in computerized polysomnography. Sleep, 17: 105. Hirshkowitz, M and Moore, CA (1999) Computerized and portable sleep recording. In: S Chokroverty (Ed.), Sleep Disorders Medicine. Butterworth-Heinemann: Boston, pp. 237-244.

151 Hinmanen, S-L and Hasan, J (2000) Limitations of Rechtschaffen and Kales. Sleep Med. Rev., 4: 149-167. Indications for Polysomnography Task Force, American Sleep Disorders Association Standards of Practice Committee (1997) Practice parameters for the indications for polysomnography and related procedures. Sleep, 20: 406-422. Johns, MW (1991) A new method for measuring daytime sleepiness: the Epworth Sleepiness Scale. Sleep, 14: 540. Keenan, SA (1999) Polysomnographic technique: An overview: In: S Chokroverty (Ed.), Sleep Disorders Medicine. Butterworth-Heinemann: Boston, pp. 151174. Kryger, MH (2000) Monitoring respiratory and cardiac function. In: MH Kryger, T Roth and WC Dement (Eds.), Principles and Practice of Sleep Medicine. WB Saunders Company: Philadelphia, pp. 1217-1230. Kubicki, S and Hermann, WM (1996) The future of computer-assisted investigation of the polysomnogram: Sleep microstructure. J. Clin. Neurophysiol., 13: 285294. Mahowald, M (2002) Hope for the PLMS quagmire: Editorial. Sleep Med., 3: 463-464. Oswald, I (1959) Sudden body jerks on falling asleep. Brain, 82: 92. Parisi, RA (1999) Respiration and respiratory function: Technique of recording and evaluation. In: S Chokroverty (Ed.). Butterworth-Heinemann: Boston, pp. 215221. Penzel, T and Conradt, R (2000) Computer based sleep recording and analysis. Sleep Med. Rev., 4: 131-J48. Rechtschaffen, A and Kales, A (1968) A Manual of Standardized Terminology, Techniques and Scoring System for Sleep Stages of Human Subjects. Washington, D.C.: U.S. Govemment Prinitng Office. Standards of Practice Committee of the American Sleep Disorders Association (1995) Practice parameters for the use of actigraphy in the clinical assessment of sleep disorders. Sleep, 18: 285-228. Thorpy, M (1992) The clinical use of the multiple sleep latency test. Sleep, 15: 268-276. Walczak, T and Chokroverty, S (1999) Electroencephalography, electromyography, and electro-oculography: General principles and basic technology. In: S Chokroverty (Ed.), Sleep Disorders Medicine. Butterworth-Heinemann: Boston, pp. 175-203.

Movement Disorders Handbook of Clinical Neurophysiology, Vol. 1 M. Hallett (Ed.) © 2003 Elsevier B.Y. All rights reserved

153

CHAPTER II

Microneurography and motor disorders David Burke?", Simon C. Gandevia'' and Vaughan G. Macefield" "College of Health Sciences, The University of Sydney and h Spinal Injuries Research Centre, Prince of Wales Medical Research Institute, University of New South Wales, Sydney, Australia

The technique of microneurography was developed by Vallbo and Hagbarth (1968) and Hagbarth and Vallbo (1968), and has since been used to shed light on motor control mechanisms, cutaneous tactile sensations, pain and disturbances to sympathetic efferent function. This chapter addresses the technique and some of the contributions made using it to understanding the role of the 'Y efferent system in the control of movement and motor disorders.

11.1. Technique In microneurography, the experimenter inserts a sterilized microelectrode manually through the skin into an underlying nerve trunk and then guides the electrode tip into the desired nerve fascicle. The electrodes are usually monopolar tungsten electrodes with a shaft diameter of - 200 urn, tapered to a tip of 1-5 p.m and insulated to the tip (Fig. 1). To obtain good single unit recordings from large myelinated axons, the electrode impedance measured at 1 kHz is usually of the order of 100-300 kil. A special concentric needle electrode has also been used by some investigators (Hallin and Wiesenfeld, 1981). Different authorities use or eschew stimulation through the microelectrode to guide insertion, but all rely on auditory feedback when close to or within a fascicle. Figure 1 illustrates the recording technique, but there is a size distortion that belies the fact that the microelectrode is much larger (shaft 200 urn) than the largest axons «20 urn), Nevertheless, manipulating the position of recording tip within the fascicle is relatively easy for experienced experi-

* Correspondence to: David Burke, MD, DSc, College of Health Sciences, Medical Foundation Building - K25, University of Sydney, NSW 2006, Australia. E-mail address:[email protected]

Tel.: lnt +61.2.9036.3091; fax: lnt +61.2.9036.3092.

menters, and it is possible to focus on different types of activity - multi-unit activity, single unit activity and the activity in unmyelinated axons, afferent or efferent. Axons with background activity are preferentially detected, merely because their discharge can be heard and the recording can be focused on this activity. Single unit recordings are those in which the activity of a single unit stands out from background activity and noise, with a sufficiently large spike that it can be heard and seen reliably, and are usually from the largest axons because the amplitude of the action potential is a function of the square of axon diameter.

11.2. Fusimotor involvement in control of reflex function, muscle tone and voluntary movement Traditionally, muscle spindle afferents have held pride of place among muscle afferents, largely because their discharge can be directly modulated by 'Y efferent (fusimotor) drive, an unusual property for a sensory receptor. Recordings from muscle spindle endings have been used as measures of 'Y efferent activity. However, this practice is safe only if all other influences on spindle discharge are measured and controlled, and this is rarely possible in human experiments. Accordingly, there are conflicting data and conclusions in the literature and, inevitably, this review reflects the experience and biases of the authors.

11.2.1. Effects of immediate history on spindle discharge It is now well documented that the discharge of muscle spindle endings is affected by previous stretch (Edin and Vallbo, 1988) and by previous fusimotor activation (Ribot-Ciscar et al., 1991; Proske et al., 1993; Wilson et al., 1995), such that, for example, the discharge of muscle spindle endings

154

D. BURKE ET AL.

Fig. 1. Unitary recording from a human muscle spindle. For the recording on the right, the tungsten rnicroelectrode was inserted percutaneously into a motor fascicle of the ulnar nerve at the wrist and the target muscle identified by the responses to intraneural electrical stimulation and the responses to passive and active movements of the digits. The recording was made from a spontaneously-active (presumed secondary) spindle ending in the 4th dorsal interosseous. The spindle ending increased its discharge during extension (right panel) and passive abduction (not shown) at the 4th metacarpophalangeal joint, the responses to stretch and shortening being essentially static. A sketch of the technique is on the left. The rnicroelectrode is introduced manually and, when in situ, it is supported without rigid fixation at one end by its connecting lead and at the other by the skin and subcutaneous tissue. Its position is adjusted within the nerve until the tip penetrates the desired nerve fascicle. Minor adjustments are made to bring the desired neural activity into focus. Note that the rnicroelectrode has a shaft diameter of - 200 urn and that the largest axons have a diameter of - 20 urn,

may remain elevated for long after a voluntary contraction, as in Fig. 2 (Macefield et al., 1991; Wilson et al., 1995). This is not due to on-going fusimotor drive but to the persistence of actinmyosin bonds formed in intrafusal fibers by 'Y efferent -activity that accompanied the contraction but ceased with it. The resulting distortions of spindle responsiveness could account for some of the discrepancies between different studies. The effects of intrafusal thixotropy can be quite prominent, sufficient to produce changes in reflex function and distortions of proprioceptive sensations which depend on perception of muscle spindle discharge (e.g. Wise et al., 1998). 11.2.2. 'Y drive to resting muscle

In human subjects who are at rest, there is a very low, possibly negligible level of fusimotor drive, particularly in static 'Y efferents (Vallbo et al., 1979; Burke, 1981; Gandevia and Burke, 1992), such that

muscle spindle discharge and the response to stretch do not change appreciably following complete nerve block (Burke et al., 1976, 1981a). There may be some activity in dynamic 'Y efferents, and this may be altered by the reflex action of cutaneous afferents (Aniss et al., 1990; Gandevia et al., 1994) and, possibly, by reinforcement maneuvers (see below, Ribot-Ciscar et al., 2000). In normal subjects who are at rest, muscle tone and the tendon jerk are therefore not dependent on the level of fusimotor drive, and hypotonia and hyporeflexia cannot be due to the withdrawal of background v activity (for reviews, see Burke, 1983, 1988). 11.2.3. Reflex reinforcement

The potentiation of spinal reflexes by reinforcement maneuvers (such as the Jendrassik maneuver) is largely due to effects on reflex transmission within

155

MICRONEUROGRAPHY AND MOTOR DISORDERS

Fig. 2. Activation of a muscle spindle ending in tibialis anterior (TA) during a 60 s contraction. Panel A shows the first 20 s of the contraction, and panel B the last 10 s, the traces being the nerve recording, ankle dorsiflexion force and integrated EMG of TA. The axon had no background discharge at rest but was activated during the contraction, discharging at - 12 Hz. On cessation of the contraction, there was a high-frequency burst of impulses from the afferent as the spindle was stretched, but the discharge then continued despite the complete subsidence of EMG and force. Panel C shows the "twitch test" used to identify the afferent as of spindle origin (upper trace: discharge of the afferent in five supramaximal trials; second trace: corresponding twitch contractions). The lowest trace shows superimposed action potentials of the afferent. From Macefield et al. (1991), with permission.

the spinal cord, not to activation of the 'Y efferent system. It was believed that selective activation of dynamic 'Y efferents would potentiate the muscle spindle response to percussion sufficiently to enhance the tendon jerk (Paillard, 1955), but some authorities question whether dynamic 'Y activation could do so (Morgan et al., 1984; Wood et al., 1994), and there is evidence that it does not (Gregory et al., 2001). Some studies using microneurography have demonstrated that effective reinforcement maneuvers produce no enhancement of the background discharge or the response to stretch of muscle spindle afferents in EMG-silent muscles at constant length (Vallbo and Hagbarth, 1966; Hagbarth et al., 1975c; Burke, 1981; Burke et al., 1981b; Ribot et al., 1986), or their ease of activation in voluntary contractions (Burke et aI., 1980a). Others have reported that reinforcement maneuvers produce an increased background spindle discharge (Burg et al., 1974; Szumski et al., 1974; Ribot-Ciscar et aI., 2000), or an increased discharge of presumed dynamic 'Y efferent axons though, parodoxically, without an increase in group Ia activity (Ribot et al.,

1986). However, these maneuvers may also increase the discharge of at least some (X motoneurons (Ribot et al., 1986), raising questions about the selectivity claimed for the 'Y activation (Hagbarth et aI., 1975c; see also Ribot-Ciscar et al., 2000). If reinforcement increased the background spindle discharge, this should depress reflex transmission from the active afferents (through the mechanism known as "homosynaptic" depression). This would depress rather than enhance the tendon jerk and the H-reflex (Hultbom et al., 1996; Wood et aI., 1996), unless the spindle response to percussion could be enhanced sufficiently to overcome the "homosynaptic" depression due to increased background spindle activity. The available evidence suggests that it cannot (Gregory et al., 2001). If reflex reinforcement occurs within spinal cord circuitry, as is suggested by Fig. 3, one would expect the H-reflex to be potentiated, as indeed it is (Landau and Clare, 1964; Bussel et al., 1978; Burke et al., 1981b; Dowman and Wolpaw, 1988; Zehr and Stein, 1999; Gregory et al., 2001). Potentiation of the H-reflex is difficult to explain on a fusimotor mechanism. Similarly, when subjects are warned of the need to contract a muscle (anticipation), train on a task or mentally rehearse a movement, there is no evidence for selective activation of 'Y motoneurons although, in each instance, spinal reflex excitability is increased (Burke et al., 1980b; Gandevia and Burke, 1985; Gandevia et al., 1997). In addition, fusimotor drive does not contribute to the reflex enhancement accompanying a motor adaptation task (Al-Falahe et al., 1990). These studies indicate that there are effective mechanisms for controlling reflex "gain" independent of the fusimotor system. 11.2.4. Voluntary contractions

During a nerve block that affects (X motor axons preferentially, the effort to contract the paralyzed muscle increases the discharge of spindle endings, presumably because it activates 'Y efferents directed to the paralyzed muscle (Burke et aI., 1979a). This finding provided support for the view that, when normal subjects contract a muscle voluntarily (or unintentionally), static 'Y motoneurons innervating the contracting muscle (but not its inactive synergists) are activated (Vallbo and Hagbarth, 1966; Vallbo, 1971, 1974; Vallbo et al., 1979). When the contraction is isometric, the 'Y activation is usually

156

D. BURKE ET AL.

8

A

Jt3

. '0 •

26

'2

::J

e

°

-e 1°0 f! :t::

-ea:I

13

" ,. • I· •

•

0

0

E C>

w ~

-l

• l

.:••

0.2

l

..,,1

,

.

0

Tap (arbitrary units)

.

" P < 0.01

•

,,'' .......•... • !'e / 'I

Q)

30

.:··

/1.

0:::

15

"· ,/

~

Q)

z

~

0

Q)

->

0.6 0.4

~8 0

13

•..•..

26

Nerve (arbitrary units)

Fig. 3. Effects of the Jendrassik maneuver. Panel A shows the relationship between the intensity of tendon percussion to the Achilles tendon (in arbitrary units) and the amplitude of the integrated multi-unit afferent response from the nerve fascicle innervating soleus (in arbitrary units) for one subject. Circles: subject relaxed. Triangles: subject performing the Jendrassik maneuver. Filled symbols: percussion that produced a tendon jerk. Open symbols: no reflex response. The "threshold" for the tendon jerk was equivalent to 15 units of afferent activity (or 12 units of percussion) when relaxed but 9 units of afferent activity (4-5 units of percussion) when performing the Jendrassik maneuver. Panel B shows for the same data, using the same symbols, the relationship between the afferent volley and the reflex EMG. During the Jendrassik maneuver (triangles), the same afferent volley produced a significantly greater reflex response than at rest (circles). From Burke et aI. (l98Ib), with permission.

sufficient to enhance the background discharge of spindle endings (Fig. 2), increase their discharge variability, increase their static response to stretch, and diminish the pause in discharge that occurs on muscle shortening (Vallbo, 1971, 1973, 1974; Burke et al., 1979b). There is evidence that voluntary effort also activates dynamic 'Y efferents (Kakuda and Nagaoka, 1998), and that f3 (skeletofusimotor) efferents can be activated by both voluntary effort (Aniss et al., 1988; Kakuda et al., 1998) and transcranial stimulation of the motor cortex (Rothwell et aI., 1990). When voluntary contractions produce muscle shortening, the enhanced fusimotor drive can be sufficient to maintain or even increase spindle discharge (Vallbo, 1973), but this occurs only if the movement is slow or the muscle is contracting against a load (Burke et al., 1978a, b; Hulliger et al., 1985). The increase in spindle discharge usually occurs after the onset of EMG activity in the contracting muscle, at some 20-50 ms when the contractions are rapid and phasic (Vallbo, 1971; Hagbarth et al., 1975a). However, while there has been clear evidence of a-'Y co-activation in all

voluntary acts so far tested, the balance between the a and 'Y drives can be varied (Burke et al., 1980a;

Vallbo and Hulliger, 1981; Wessberg and Vallbo, 1995). This would be expected given that different descending pathways have quantitatively different effects on a and 'Y motoneurons, and that many peripheral afferent inputs have different reflex effects on a and 'Y motoneurons (Aniss et al., 1990; Gandevia et al., 1994). Nevertheless, the evidence for disproportionate activation of 'Y motoneurons during motor learning and precision finger movements is, at best, quite modest (Vallbo and AI-Falahe, 1990; Wessberg and Vallbo, 1995; Kakuda et al., 1996). In fatiguing submaximal isometric contractions the enhancement of muscle spindle discharge is maximal initially and then decreases by about onethird (Macefield et al., 1991), a finding that implies that feedback support to the contraction is maximal initially but subsequently wanes. A further implication is that, contrary to classical views (Merton, 1953), the 'Y efferent system is not mobilized to compensate for fatigue, at least under isometric conditions. When a motor nerve is blocked distal to

157

MICRONEUROGRAPHY AND MOTOR DISORDERS

the recording site, recordings can be made from the axons of ex motoneurons deprived of feedback support from endings in the now-paralyzed muscle (Gandevia et aI., 1990, 1993; Macefield et aI., 1993). The discharge rates of motor axons reach only about two-thirds of those of normally intact motor units, a finding that suggests significant feedback support to the contracting motoneuron pool. It is more difficult to maintain motor unit firing in the absence of this feedback, but subjects can still recruit and de-recruit motoneurons and modulate motoneuron firing rates, given only knowledge of the effort that they are sending to the muscle. 11.4. Cutaneous atTerents and motor control

The traditional view that cutaneous afferents play little role in motor control is just as fallacious as the view that muscle afferents play no role in sensation. In the control of hand function, cutaneous afferents are at least as important as muscle spindle afferents and, for some motor acts, they are arguably more important sources of afferent feedback, capable of reflexly modifying the command for movement at multiple levels - segmental, suprasegmental and cerebral. There are extensive data on the response patterns of different cutaneous mechanoreceptors to tactile stimuli, to skin stretch and to passive movement of the hand (for review see Macefield, 1998), yet there has been relatively little work on the roles of tactile afferents in motor control. Most of this work has dealt with the sensorimotor specializations of the hand which, given its high density of tactile afferents (particularly in the finger pads), is perhaps not surprising. At the very least, receptors in the skin provide facilitation of the motoneuron pool: electrical or mechanical stimulation of cutaneous afferents causes short-latency (spinal) and long-latency increases in the EMG of muscles acting on the digits (Darton et aI., 1982; Evans et aI., 1986; Macefield et aI., 1996b) and, in the absence of muscle afferent feedback (anesthetic block ofthe ulnar nerve), tactile afferents traveling in the median nerve have been shown to increase the size of a volitionaIly generated motor output (Gandevia et aI., 1990). Moreover, the input from a single cutaneous afferent is sufficiently strong that it can modulate the ongoing EMG of muscles acting on the receptor-bearing digit, at

spinal latencies (McNulty et aI., 1999); conversely, the synaptic input of a single muscle spindle afferent is weak, at least for muscle spindles in the leg (Gandevia et aI., 1986). However, more than this simple facilitation of motoneurons, tactile afferents are actively engaged in fine motor control of the hand. Because of their location at the skin/object interface, cutaneous mechanoreceptors are ideally placed to monitor the loads applied by or to the finger pads during manipulation of a gripped object. Indeed, anesthesia of the digits seriously compromises the capacity to perform a precision grip and to adjust the grip force automatically as a function of the load and surface conditions (Johansson et aI., 1992c; Hager-Ross and Johansson, 1996). Tactile afferents also provide information on the frictional conditions of the object, information that is incorporated automatically into grading the grip force required to hold an object (Cole and Johansson, 1993). Unpredictable pulling forces applied to an object held between finger and thumb evoke automatic increases in grip force that serve to prevent escape of the object from the grasp (Cole and Abbs, 1988; Johansson et al., 1992a, b, c; Macefield et aI., 1996a; Macefield and Johansson, 1996), and rnicroneurographic studies have shown that tactile afferents in the glabrous skin of the digits are the only receptors capable of triggering these increases in grip force (Macefield et aI., 1996a); muscle and joint afferents respond only during the resultant increases in grip force (Macefield and Johansson, 1996). 11.5. Studies in patients

11.5.1. Spasticity There are relatively few reports of muscle spindle activity in spastic patients (Hagbarth et aI., 1973, 1975b; Szumski et aI., 1974; Wilson et aI., 1999).All are from hemiplegic patients; there are no published studies from patients with spinal cord injury. The studies of Szumski et al. (1974) and Hagbarth et al. (1975b) involved discharge patterns during clonus (described below with Parkinsonian tremor). Hagbarth et aI. (1973) reported that the responses to controlled stretch of 9 endings in triceps surae of hemiplegic patients were not greater than those of 12 spindle endings in normal subjects. Wilson et al.

D. BURKEET AL.

158

Instantaneous frequency

Fig. 4. Muscle spindle afferent innervating extensor carpi radialis (ECR) during deliberate isometric wrist extension in a patient suffering from hemiplegic spasticity (extensor strength 68% of that of the contralateral side). The afferent did not maintain a background discharge when truly at rest. During voluntary efforts to contract the muscle ("Extension"), the spindle ending was activated together with the EMG of ECR. Between the deliberate contractions, the ending was activated unintentionally during inadvertent contractions, but again with EMG. The ending in ECR was not activated during an unintentional contraction of the forearm flexors that occurred when the patient was told to relax ECR. The insert (top right) shows action potential morphology (multiple superimposed discharges). From Wilson et a1. (1999), with permission.

(1999) documented the properties of 26 endings in the forearm extensors. Background spindle discharge rates were the same as in healthy volunteers, reinforcement and other maneuvers did not enhance spindle discharge selectively, cutaneo-fusimotor reflexes could not be demonstrated, and spindles in the paretic muscles were no more difficult to activate in voluntary attempts to contract the weak muscles than in normal subjects (Fig. 4). These studies lead to the conclusion that there is no primary defect of fusirnotor function in hemiplegic spasticity - the hyper-reflexia is not due to enhanced 'Y efferent drive, and the loss of dexterity is not due to inability to activate 'Y motoneurons appropriately for the degree of a activation. Whether the same holds true for patients with spinal spasticity remains to be demonstrated. It is not unreasonable to speculate that cutaneo-fusimotor reflexes could be disinhibited in spinal patients, much as are cutaneomuscular reflexes, and it is conceivable that "normal" peripheral afferent inputs from skin, joints,

etc. may produce a background discharge in 'Y efferents. 11.5.2. Parkinson's disease The only published data from parkinsonian patients are those of Hagbarth et a1. (1975b) on parkinsonian tremor and of Wallin et al. (1973) on multi-unit muscle afferent responses to muscle stretch, together with a re-analysis of some of these data (Burke et al., 1977). In parkinsonian tremor muscle spindle endings tended to discharge twice, during the shortening phase of the test muscle (with the EMG of that muscle) and during the lengthening phase (which would subject muscle spindles to stretch). There was a similar biphasic pattern in healthy subjects who made rapid alternating movements to mimic tremor (Hagbarth et al., 1975a) whereas, in the reflex-sustained movements of clonus in spastic patients, spindle discharge occurred only during the stretching phase of the oscillating

159

MICRONEUROGRAPHY AND MOTOR DISORDERS

movement (Szumski et al., 1973; Hagbarth et al., 1975b). In multi-unit recordings, it was noted that spontaneous or evoked fluctuations in rigidity involved parallel fluctuations in afferent activity and EMG and that there was more background afferent activity in rigid muscles than in normal subjects who were relaxed. The shortening reaction is a phenomenon characteristic of Parkinson's disease and other basal ganglia disturbances, but may also occur in normal human subjects. There is one recording of a shortening reaction in a normal subject: passive dorsiflexion produced an involuntary contraction of tibialis anterior, and this was associated with a muscle spindle discharge, much as was recorded when the subject made a voluntary dorsiflexion movement (Burke et aI., 1978b). The above data suggest that there is no primary defect of fusimotor function in parkinsonian rigidity. Parkinsonian rigidity seems to behave more as if there were a defect of supraspinal drives onto relatively normal spinal mechanisms, those drives affecting a and 'Y motoneurons much as would volitional drives. However, these conclusions are based on qualitative impressions rather than quantitative recordings, and more extensive studies are required before the views can be accepted as definitive. 11.5.3. Dystonia There have been no published microneurographic studies in dystonia, but there is literature suggesting that different forms of dystonia are due to or associated with a disturbance of sensory processing, affecting particularly the input from muscle spindle endings (Kaji et al., 1995; Grunewald et aI., 1997; Yoshida et al., 1998). In addition, some of the beneficial effects of botulinum toxin in dystonic syndromes may be due to effects on the neuromuscular junction of 'Y efferents on intrafusal fibers. 11.6. Clinical value

With microneurography, the findings depend on the site of the microelectrode within a fascicle. A recording from a single afferent axon is rarely representative of the population response, whether the afferent is of cutaneous or muscle origin. Multiunit recordings can provide a representative picture

of large-fiber activity within the fascicle, but are difficult to quantify. Accordingly, when one considers the data for an individual patient, microneurography currently has no place as a diagnostic procedure, even if insights into pathophysiology come when the data from a number of patients are pooled. Evoked compound action potentials to electrical stimulation have been recorded using microneurography, and the full range of afferent axons from large-myelinated to unmyelinated can be discriminated. However, the diagnostic value of such recordings is debatable, and similar data can be obtained with near-nerve needle electrodes. The clinical value of microneurography comes from the unique insights that it can provide into pathophysiology. References Al-Falahe, NA, Nagaoka, M and Vallbo, AB (1990) Lack of fusimotor modulation in a motor adaptation task. Acta Physiol. Scand., 140: 23-30. Aniss, AM, Gandevia, SC and Burke, D (1988) Reflex changes in muscle spindle discharge during a voluntary contraction. J. Neurophysiol., 59: 908-921. Aniss, AM, Diener, H-C, Hore, J, Burke, D and Gandevia, SC (1990) Reflex activation of muscle spindles in human pretibial muscles during standing. J. Neurophysiol., 64: 671-679. Burg, D, Szumski, AJ, Struppler, A and Velho, F (1974) Assessment of fusimotor contribution to reflex reinforcement in humans. J. Neurol. Neurosurg. Psychiatry, 37: 1012-1021. Burke, D (1981) The activity of human muscle spindle endings in normal motor behavior. In: R Porter (Ed.), International Review of Physiology, Vol. 25, Neurophysiology IV. University Park Press, Baltimore, pp.91-126. Burke, D (1983) Critical examination of the case for or against fusimotor involvement in disorders of muscle tone. In: JE Desmedt (Ed.), Motor Control Mechanisms in Health and Disease, Advances in Neurology, Vol. 39. Raven Press, New York, pp. 133-150. Burke, D (1988) Spasticity as an adaptation to pyramidal tract injury. In: SG Waxman (Ed.) Functional Recovery in Neurological Disease, Advances in Neurology, Vol. 47. Raven Press, New York, pp. 401-423. Burke, D, Hagbarth, K-E, Lofstedt, L and Wallin, BG (1976) The responses of human muscle spindle endings to vibration during isometric contraction. J. Physiol. (Lond.), 261: 695-711.

160 Burke, D, Hagbarth, K-E and Wallin, BG (1977) Reflex mechanisms in Parkinsonian rigidity. Scand. J. Rehab. Med., 9: 15-23. Burke, D, Hagbarth, K-E and Lofstedt, L (1978a) Muscle spindle responses in man to changes in load during accurate position maintenance. J. Physiol. (Lond.), 276: 159-164. Burke, D, Hagbarth, K-E and Lofstedt, L (1978b) Muscle spindle activity in man during shortening and lengthening contractions. J. Physiol. (Lond.), 277: 131-142. Burke, D, Hagbarth, K-E and Skuse, NF (1979a) Voluntary activation of spindle endings in human muscles temporarily paralysed by nerve pressure. J. Physiol. (Lond.), 287: 329-336. Burke, D, Skuse, NF and Stuart, DG (1979b) The regularity of muscle spindle discharge in man. J. Physiol. (Lond.), 291: 277-290. Burke, D, McKeon, B and Westerman, RA (1980a) Induced changes in the thresholds for voluntary activation of human spindle endings. J. Physiol. (Lond.),302: 171-181. Burke, D, McKeon, B, Skuse, NF and Westerman, RA (1980b) Anticipation and fusimotor activity in preparation for a voluntary contraction. J. Physiol. (Lond.), 306: 337-348. Burke, D, McKeon, B and Skuse, NF (198Ia) The irrelevance of fusimotor activity to the Achilles tendon jerk of relaxed humans. Ann. Neurol., 10: 547-550. Burke, D, McKeon, B and Skuse, NF (1981 b) Dependence of the Achilles tendon reflex on the excitability of spinal reflex pathways. Ann. Neurol., 10: 551-556. Bussel, B, Morin, C and Pierrot-Deseilligny, E (1978) Mechanism of monosynaptic reflex reinforcement during Jendrassik maneuver in man. J. Neurol. Neurosurg. Psychiatry, 41: 40-44. Cole, KJ and Abbs, JH (1988) Grip force adjustments evoked by load force perturbations of a grasped object. J. Neurophysiol., 60: 1513-1522. Cole, KJ and Johansson, RS (1993) Friction at the digitobject interface scales the sensorimotor transformation for grip responses to pulling loads. Exp. Brain Res., 95: 523-532. Darton, K, Lippold, OC, Shahani, M and Shahani, U (1985) Long-latency spinal reflexes in humans. J. Neurophysiol., 53: 1604-1618. Edin, BB and Vallbo, AB (1988) Stretch sensitization of human muscle spindles. J. Physiol. (Lond.), 400: 101-111. Evans, AL, Harrison, LM and Stephens, JA (1989) Taskdependent changes in cutaneous reflexes recorded from various muscles controlling finger movement in man. J. Physiol. (Lond.), 418: 1-12.

D. BURKE ET AL.

Gandevia, SC and Burke, D (1985) Effect of training on voluntary activation of human fusimotor neurons. J. Neurophysiol., 54: 1422-1429. Gandevia, SC and Burke, D (1992) Does the nervous system depend on kinesthetic information to control natural limb movements? Behav. Brain Sci., 15: 614632. Gandevia, SC, Burke, D and McKeon, B (1986) Coupling between human muscle spindle endings and motor units assessed using spike-triggered averaging. Neurosci. Lett., 71: 181-186. Gandevia, SC, Macefield, G, Burke, D and McKenzie, DK (1990) Voluntary activation of human motor axons in the absence of muscle afferent feedback. The control of the deafferented hand. J. Physiol. (Lond.), 113: 15631581. Gandevia, SC, Macefield, VG, Bigland-Ritchie, B, Gorman, RB and Burke, D (1993) Motoneuronal output and gradation of effort in attempts to contract acutely paralysed leg muscles in man. J. Physiol. (Lond.), 471: 411-427. Gandevia, SC, Wilson, L, Cordo, PJ and Burke, D (1994) Fusimotor reflexes in relaxed forearm muscles produced by cutaneous afferents from the human hand. J. Physiol. (Lond.), 479: 499-508. Gandevia, SC, Wilson, LR, Inglis, JT and Burke, D (1997) Mental rehearsal of motor tasks recruits alpha-motoneurons but fails to recruit human fusimotor neurons selectively. J. Physiol. (Lond.), 505: 259-266. Gregory, JE, Wood, SA and Proske, U (2001) An investigation into mechanisms of reflex reinforcement by the Jendrassik maneuver. Exp. Brain Res., 138: 366-374. Grunewald, RA, Yoneda, Y, Shipman, JM and Sagar, HJ (1997) Idiopathic focal dystonia: a disorder of muscle spindle afferent processing? Brain, 120: 2179-2185. Hagbarth, K-E and Vallbo, AB (1968) Discharge characteristics of human muscle afferents during muscle stretch and contraction. Exp. Neurol., 22: 674-694. Hagbarth, K-E, Wallin, G and Lofstedt, L (1973) Muscle spindle responses to stretch in normal and spastic subjects. Scand. J. Rehab. Med., 5: 156-159. Hagbarth, K-E, Wallin, G and Lofstedt, L (1975a) Muscle spindle activity in man during voluntary fast alternating movements. J. Neurol. Neurosurg. Psychiatry, 38: 625635. Hagbarth, K-E, Wallin, G, Lofstedt, L and Aquiionius, SM (1975b) Muscle spindle activity in alternating tremor of Parkinsonism and in clonus. J. Neurol. Neurosurg. Psychiatry, 38: 636-641. Hagbarth, K-E, Wallin, G, Burke, D and Lofstedt, L (1975c) Effects of the Jendrassik maneuver on muscle spindle activity in man. J. Neurol. Neurosurg. Psychiatry, 38: 1143-1153.

MICRONEUROGRAPHY AND MOTOR DISORDERS

Hager-Ross, C and Johansson, RS (1996) Non-digital afferent input in reactive control of fingertip forces during precision grip. Exp. Brain Res., 110: 131-141. Hallin, RG and Wiesenfeld, Z (1981) A standardized electrode for percutaneous recording of A and C fiber units in conscious man. Acta Physiol. Scand., 113: 561-563 Hulliger, M, Nordh, E and Vallbo, AB (1985) Discharge in muscle spindle afferents related to direction of slow precision movements in man. J. Physiol. (Lond.), 362: 437-453. Hultborn, H, Illert, M, Nielsen, J, Paul, A, Ballegaard, M and Wiese, H (1996) On the mechanism of the postactivation depression of the H-reflex in human subjects. Exp. Brain Res., 108: 450-462. Johansson, RS, Riso, R, Hager, C and Backstrom, L (1992a) Somatosensory control of precision grip during unpredictable pulling loads. I. Changes in load force amplitude. Exp. Brain Res., 89: 181-191. Johansson, RS, Hager, C and Riso, R (1992b) Somatosensory control of precision grip during unpredictable pulling loads. II. Changes in load force rate. Exp. Brain Res., 89: 192-203. Johansson, RS, Hager, C and Backstrom, L (1992c) Somatosensory control of precision grip during unpredictable pulling loads. III. Impairments during digital anesthesia. Exp. Brain Res., 89: 204-213. Kaji, R, Rothwell, JC, Katayama, M, Ikeda, T, Kubori, T, Kohara, N, Mezaki, T, Shibasaki, H and Kimura, J (1995) Tonic vibration reflex and muscle afferent block in writer's cramp. Ann. Neurol., 38: 155-162. Kakuda, N, Vallbo, AB and Wessberg, J (1996) Fusimotor and skeletomotor activities are increased with precision finger movement in man. J. Physiol. (Lond.), 492: 921-929. Kakuda, N, Miwa, T and Nagaoka, M (1998) Coupling between single muscle spindle afferent and EMG in human wrist extensor muscles: physiological evidence of skeletofusimotor (beta) innervation. Electroencephalogr. Clin. Neurophysiol., 109: 360-363. Kakuda, N and Nagaoka, M (1998) Dynamic response of human muscle spindle afferents to stretch during voluntary contraction. J. Physiol. (Lond.), 513: 621628. Landau, WM and Clare, MH (1964) Fusimotor function. Part IV. Reinforcement of the H-reflex in normal subjects. Arch. Neurol., 10: 117-122. Macefield, VG (1998) The signalling of touch, finger movements and manipulation forces by mechanoreceptors in human skin. In: JW Morley (Ed.), Neural Aspects of Tactile Sensation. Elsevier, Amsterdam, pp 89-130. Macefield, VG, Rothwell, JC and Day, BL (1996b) The contribution of transcortical pathways to long-latency

161 stretch and tactile reflexes in human hand muscles. Exp. Brain Res., 108: 172-184. Macefield, G, Hagbarth, K-E, Gorman, R, Gandevia, SC and Burke, D (1991) Decline in spindle support to alpha-motoneurons during sustained voluntary efforts. J. Physiol. (Lond.), 440: 497-512. Macefield, VG, Gandevia, SC, Bigland-Ritchie, B, Gorman, RB and Burke, D (1993) The firing rates of human motoneurons voluntarily activated in the absence of muscle afferent feedback. J. Physiol. (Lond.), 471: 429-443. Macefield, VG, Hager-Ross, C and Johansson, RS (1996a) Control of grip force during restraint of an object held between finger and thumb: responses of cutaneous afferents from the digits. Exp. Brain Res., 108: 155171. Macefield, VG and Johansson, RS (1996) Control of grip force during restraint of an object held between finger and thumb: responses of muscle and joint afferents from the digits. Exp. Brain Res., 108: 172-184. Merton, PA (1953) Speculations on the servo control of movement. In: JL Malcolm and JAB Gray (Eds.), The Spinal Cord. Ciba Foundation Symposium, Churchill, London, pp. 84-91. McNulty, PA, Tnrker, KS and Macefield, VG (1999) Evidence for strong synaptic coupling between single tactile afferents and motoneurons supplying the human hand. J. Physiol. (Lond.), 518: 883-893. Morgan, DL, Prochazka, A and Proske, U (1984) Can fusimotor activity potentiate the responses of muscle spindles to a tendon tap? Neurosci. Lett., 50: 209-215. Paillard, J (1955) Reflexes et Regulations d'Origine Proprioceptive Chez l'Homme. Arnette, Paris. Proske, U, Morgan, DL and Gregory, IE (1993) Thixotropy in skeletal muscle and in muscle spindles: a review. Prog. Neurobiol., 41: 705-721. Ribot, E, Roll, JP and Vedel, JP (1986) Efferent discharges recorded from single skeletomotor and fusimotor fibers in man. J. Physiol. (Lond.), 375: 2251-2268. Ribot-Ciscar, E, Tardy-Gervet, MF, Vedel, JP and Roll, JP (1991) Post-contraction changes in human muscle spindle resting discharge and stretch sensitivity. Exp. Brain Res., 86: 673-678. Ribot-Ciscar, E, Rossi-Durand, C and Roll, JP (2000) Increased muscle spindle sensitivity to movement during reinforcement maneuvers in relaxed human subjects. J. Physiol. (Lond.), 523: 271-282. Rothwell, JC, Gandevia, SC and Burke, D (1990) Activation of fusimotor neurons by motor cortical stimulation in human subjects. J. Physiol. (Lond.), 431: 743-756. Szumski, AJ, Burg, D, Struppler, A and Velho, F (1974) Activity of muscle spindles during muscle twitch and

162 clonus in normal and spastic human subjects. Electroencephalogr. Clin. Neurophysiol., 37: 589-597. Vallbo, AB (1971) Muscle spindle response at the onset of isometric voluntary contractions in man. Time difference between fusimotor and skeletornotor effects. J. Physiol. (Lond.), 218: 405-431. Vallbo, AB (1973) Muscle spindle afferent discharge from resting and contracting muscles in nonnal human subjects. In: JE Desmedt (Ed.), New Developments in Electromyography and Clinical Neurophysiology, Vol. 3. Karger, Basel, pp. 251-262. Vallbo, AB (1974) Human muscle spindle discharge during isometric voluntary contractions. Amplitude relations between spindle frequency and torque. Acta Physiol. Scand., 90: 319-336. Vallbo, AB and Hagbarth, K-E (1968) Activity from skin mechanoreceptors recorded percutaneously in awake human subjects. Exp. Neurol., 21: 270-289. Vallbo, AB and Hulliger, M (1981) Independence of skeletomotor and fusimotor activity in man? Brain Res., 223: 176-180. Vallbo, AB and Al-Falahe, NA (1990) Human muscle spindle response in a motor learning task. J. Physiol. (Lond.), 421: 553-568. Vallbo, AB, Hagbarth, K-E, Torebjork, HE and Wallin, BG (1979) Somatosensory, proprioceptive, and sympathetic activity in human peripheral nerves. Physiol. Rev., 59: 919-957. Wallin, BG, Hongell, A and Hagbarth, K-E (1973) Recordings from muscle afferents in Parkinsonian

D. BURKE ET AL.

rigidity. In: JE Desmedt (Ed.), New Developments in Electromyography and Clinical Neurophysiology, Vol. 3. Karger, Basel, pp. 263-272. Wessberg, J and Vallbo, AB (1995) Human muscle spindle afferent activity in relation to visual control in precision finger movements. J. Physiol. (Lond.), 482: 225-233. Wilson, LR, Gandevia, SC and Burke, D (1995) Increased resting discharge of human spindle afferents following voluntary contractions. J. Physiol. (Lond.), 488: 833840. Wilson, LR, Gandevia, SC, Inglis, IT, Gracies, J-M and Burke, D (1999) Muscle spindle activity in the affected upper limb after a unilateral stroke. Brain, 122: 20792088. Wise, AK, Gregory, JE and Proske, U (1998) Detection of movements of the human forearm during and after cocontraction of muscles acting at the elbow joint. J. Physiol. (Lond.), 508: 325-330. Wood, SA, Morgan, DL, Gregory, JE and Proske, U (1994) Fusimotor activity and the tendon jerk in the anaesthetized cat. Exp. Brain Res., 98: 101-109. Wood, SA, Gregory, JE and Proske, U (1996) The influence of muscle spindle discharge on the human Hreflex and the monosynaptic reflex in the cat. J. Physiol. (Lond.), 497: 279-290. Yoshida, K, Kaji, R, Kubori, T, Kohara, N, Iizuka, T and Kimura, J (1998) Muscle afferent block for the treatment of oromandibular dystonia. Movement Dis., 13: 699-705.

Movement Disorders Handbook of Clinical Neurophysiology, Vol. 1 M. Hallett (Ed.) © 200] Elsevier B.V. All rights reserved

163 CHAPTER 12

Imaging Scott T. Grafton * Center for Cognitive Neuroscience, Dartmouth College, Hanover; NH 03755, USA

12.1. Introduction Brain imaging plays an important role in the evaluation of patients with movement disorders. Anatomic imaging is essential for ruling out structurallesions in subcortical nuclei and cortex, and for identifying regional atrophic changes. Imaging of brain metabolism and neurotransmitter function is an important adjunct to the clinical examination in patients with atypical akinetic-rigid syndromes that might not be secondary to idiopathic Parkinson's disease. Functional activation studies provide unique insight into normal motor control as well as the pathophysiologic basis of abnormal motor control. These imaging methods encompass techniques based on conventional x-rays, magnetic resonance and radionuclide tomography. In this chapter these techniques are reviewed and related to clinical applications, basic research and assessment of pharmacological and surgical therapy for movement disorders.

12.2. Structural imaging Although conventional x-rays of the skull are no longer used diagnostically in movement disorders, they are important historically for lesion localization and premorbid clinical-radiological correlation. In a classic 1917 study of injured soldiers, Holmes used conventional x-rays to relate the location of bullets lodged in the cerebellar hemispheres to cardinal signs of cerebellar damage including unilateral ataxia, hypotonia and dysiadochokinesia (Holmes,

* Correspondence to: Dr. Scott T. Grafton, M.D., Director, Dartmouth Brain Imaging Center, Center for Cognitive Neuroscience, 6162 Moore Hall, Dartmouth College, Hanover, NH 03755, USA. E-mail address: [email protected] Tel.: + I (603) 646-0038; fax: + I (603) 646-1181.

1917). With the development of computer assisted tomographic imaging (CT) in the 1970s it became possible to identify supratentorial structural lesions that could cause secondary movement disorders. This early imaging work revealed that the most common structural lesion leading to parkinsonian symptoms was a large cortical or glial tumor with deformation of the basal ganglia. It is extremely rare for tumors located directly within the basal ganglia to cause parkinsonism (Waters, 1993). Other lesions, occasionally associated with parkinsonism are listed in Table 1. The advent of CT also brought attention to the incidental finding of basal ganglia calcification, i.e. Fahr's disease. The incidence of basal ganglia calcification in a general adult population is approximately 0.7%. Of these persons, less than 7% have any motor symptoms (Murphy, 1979; Brannan et aI., 1980). However, if the patient presents with hypoparathyroidism there is a 70% chance of basal ganglia calcification. This increases to almost 100% for patients with pseudohypoparathyroidism. The likelihood of motor symptoms also increases (Muenter and Whisnant, 1968; Sachs et aI., 1982; Illum and Dupont, 1985). With CT it also became possible to identify white matter changes consistent with subTable I Structura11esions associated with an akinesis or rigidity. Cortical tumors Glioma Meningioma Other Subdural hematoma Striatal abscess Midbrain tuberculoma Ventriculomegally Posterior fossa cyst Normal pressure hyodrocephalus Vascular parkinsonism

164

cortical infarction and associated arteriosclerotic parkinsonism (Critchley, 1929), i.e. subcortical arteriosclerotic encephalopathy (Binswanger's disease) (Thompson and Marsden, 1987; Bennett et al., 1990). CT was the first method to generate reliable volumetric measurements of brain anatomy in vivo. Striatal atrophy in advanced Huntington's disease was readily measured and it became possible to correlate clinical severity with tissue loss in the head of the caudate nucleus (Grafton et aI., 1992). With the introduction of magnetic resonance imaging in the early 1980s, image resolution and tissue contrast improved dramatically. The primary use of anatomic MR imaging in movement disorders is to exclude vascular disease or neoplasm causing symptoms that could mimic a neurodegenerative disease (Waters, 1993). Infratentoriallesions such as cerebellar atrophy in the hereditary ataxias can also be screened reliably. MRI changes in the basal ganglia can be seen in a variety of systemic diseases, as listed in Table 2. Most of these can be readily diagnosed clinically. Structural imaging with MRI allows for unprecedented accuracy in volumetric measurements of complete nuclei, such as the putamen or caudate. Large databases of normal and pathologic brain anatomy are currently being generated for probabilistic assessment of structure, form and volume (Mazziotta et al., 2001; Toga and Thompson, 2001). These measures can be correlated with clinical progression in Huntington's disease and possibly used to detect presymptomatic gene-positive persons at risk for the disease (Aylward et aI., 2000). Using special acquisition parameters, it may be possible to identify subtle changes in other neurodegenerative disorders including Parkinson's disease (Hu et aI., 2001).

12.3. Functional imaging 12.3.1. Radionuclide imaging The advent of single photon emission tomographic (SPECT) imaging provided early measurements of brain cerebral blood flow. With this method patients are injected with a radioactive agent that binds to cerebral tissue in proportion to local cerebral blood flow, a receptor or some other biologic marker (Podreka et aI., 1987). Injections and images are acquired with the subject at rest. Gamma-ray energy is detected with a set of

S.T. GRAFTON

Table 2 Diseases with MRI signal changes in basal ganglia. Hypointensity Wilson's diease Leigh's disease CO intoxication Anoxia Hallervorden-Spatz disease Cyanide poisoning Methanol intoxication GM2-gangliosidosis Hemolytic uremic disease Hyperintensity Wilson's disease Creutzfeldt-Jakob disease Manganese toxicity Hepatic encephalopathy AIDS Normal aging Calcified basal ganglia Hypo- and pseudohypoparathyroidism Fahrs syndrome CO intoxication Birth anoxia Tuberous sclerosis Mitochondrial encephalopathies Radiation and methotrexate therapy AIDS Congenital folate deficiency, dihydropteridine reductase deficiency Japanese B encephalitis, herpes simplex encephalitis Down syndrome Cockayne's syndrome MRI, magnetic resonance imaging; CO, carbon monoxide; AIDS, acquired immune deficiency syndrome.

collimated detectors rotating slowly around the head. Images are of low resolution (> 1.5 em) and nonuniform. Deep brain structures such as basal ganglia are of low image intensity due to attenuation of the radioactive emitter by overlying tissue. SPECT studies using blood flow tracers provided early evidence for changes in basal ganglia with Huntington's disease and temporo-parietal hypoperfusion in Alzheimer's disease. More recently, the cocaine analog 213-carbomethoxy-313-4-iodophenyl-tropane (beta-CIT) labeled with 1231 and related compounds have played an essential role in the assessment of the presynaptic striatal dopamine transporter uptake site (Brucke et al., 1993).

IMAGING

Development of positron emission tomography (PET) imaging in the early 1980s resolved many of the technical limitations of SPECT (better resolution, no attenuation artifacts) (Phelps et al., 1975). The range of biologic radiotracers that could be created with cyclotron produced radioisotopes was greatly expanded. Dominating these new compounds was "F-fluorodeoxyglucose (FDG) (Reivich et aI., 1979). The tracer is trapped within cells in proportion to glucose transport and utilization. Imaging of regional radioactivity within the brain provided a direct, simple assessment of relative glucose metabolism. Glucose metabolism is strongly correlated with local neuronal activity (Jueptner and Weiller, 1995). In particular, lesions and physiologic studies in rodents and non-human primates have established that regional metabolism reflects both excitatory and inhibitory neuronal activity and this activity is predominantly a reflection of pre-synaptic function (Nudo and Masterton, 1986). Under pathologic conditions glucose metabolism is altered when there is a change of neuronal density. Importantly, this measure was observed to be highly sensitive to underlying pathologic conditions and more reliable than other imaging methods such as blood flow imaging with SPECT or PET agents. Early clinical studies identified marked metabolic changes in temporoparietal cortex in both early and advanced Alzheimer's disease and mesial temporal hypometabolism in complex partial epilepsies. Thus, one might hope to identify subtle alterations of function circuits in movement disorders using metabolic rather than structural imaging. However, glucose metabolism also shows large changes in association with normal neuronal activation (Sokoloff, 1977). Thus, the behavioral state of the human or animal during the 20-30 min uptake period of FDG after intravenous injection will have a strong impact on the regional metabolism measured by PET imaging. In disorders with involuntary movement neural systems associated with movement production could have increased metabolic activity (Colebatch et aI., 1990; Brooks et aI., 1992b). This can potentially blur the distinction between metabolic abnormalities due to a disease (trait) with those due to a symptom (state).

12.3.1.1. Hypokinetic movement disorders A variety of cortical and subcortical metabolic changes are observed in the hypokinetic movement

165

disorders, i.e, disorders where there is a reduction of volitional movement. In Parkinson's disease the characteristic finding on PET imaging is elevated glucose metabolism in the striatum and mild to moderate reductions of cortical metabolism (Kuhl et aI., 1984; Eidelberg et aI., 1994). The hyperactivity in striatum is consistent with autoradiographic studies of non-human primates with parkinsonian symptoms secondary to the neurotoxin N-methyl-4-phenyl-l,2,3,6-tetrahydropyridine (MPTP) (Crossman et aI., 1985). Using this pattern of striatal hypermetabolism of PD as a benchmark, it was apparent that atypical parkinsonian syndromes, including multiple systems atrophy, striatonigral degeneration and olivopontocerebellar atrophy had different metabolic signatures as listed in Table 3 (Rosenthal et aI., 1988; De VoIder et aI., 1989; Fulham et aI., 1991; Otsuka et aI., 1991; Eidelberg et aI., 1993; Gilman et al., 1994; Otsuka et aI., 1994). An important generality is that all of the atypical syndromes are likely demonstrate striatal hypometabolism with variable involvement of cortical or cerebellar hypometabolism. Large clinical series have not yet been performed to establish the sensitivity and specificity of PET imaging. Nevertheless, the available evidence from smaller studies supports the utility of PET glucose metabolic imaging as an adjunct for diagnosing patients with clinically atypical akinetic-rigid movement disorders. Approximately 15% of PD patients will develop a significant dementia. With dementia there is a reduction of temporal-parietal cortical metabolism in the same areas as seen in Alzheimer's disease (Kuhl et aI., 1985). Whether this dementia and metabolic finding represents PD+AD, a special form of PD, or diffuse Lewy body disease with dementia is unknown. The neuropharmacology of movement disorders can be evaluated with PET radioisotopes that reflect presynaptic doparninergic function C8F-DOPA), post-synaptic DIID2 dopamine receptor binding (Spiperone, Raclopride) and non-specific opiod receptor binding (Garnett et aI., 1983). In Parkinson's disease there is an approximately 30% loss of F-DOPA uptake in striatum compared to normal subjects at symptom onset, progressing to a 60% reduction with advanced disease (Garnett et aI., 1984; Leenders et al., 1984; Leenders et aI., 1986; Martin et aI., 1987). There is a greater loss of FDOPA in the putamen than the caudate, whereas in

166

S.T. GRAFTON

Table 3 Imaging in hypokinetic movement disorders.

Metabolism

PD

PD-Dementia

Atypical PD

PSP

CBGD

Inc striatum

Inc striatum

Dec striatum

Dec frontal

Mild dec frontal

Mild dec frontal

Dec frontal

Dec striatum, cerebellum, thalamus

Dec thalamus, parietal, temporal Asymmetric!

Dec Dec cerebellar (ataxic) temperoparietal Presynaptic dopamine Postsynaptic D2

Dec putamen

Dec putamen

Dec putament

Dec putamen

Dec putamen

Mild dec caudate

Mild dec caudate

Dec caudate

Dec caudate

Dec caudate

Dec striatum

Dec striatum

Dec striatum

Dec striatum

Dec striatum

Nl-mild inc striatum (untreated) Nl-mild dec striatum (treated)

Opioid receptors Normal PMRS

Normal striatum?

Normal striatum?

Dec NANcreatine Dec NANcreatine Dec NANcreatine

Pl): Parkinson's disease; PSP: Progressive supranuclear palsy; CBGO: Corticobasal ganglionic degeneration; Atypical PO includes striatonigral degeneration, olivopontocerebellar degeneration, and multiple systems atrophy; PMRS: Proton magnetic resonance spectroscopy.

the atypical parkinsonian syndromes both caudate and putamen are typically involved (Table 3) (Brooks et aI., 1990a, b; Laihinen et al., 1995; Brucke et aI., 1997). The reliability of using these findings for radiologic diagnosis in an individual patient depends on the experience of the imaging center performing F-DOPA imaging. Individual subject diagnosis requires the study of a large normative population with low measurement variability that patient data can be compared to. Post-synaptic dopamine receptors are normal or mildly increased in untreated early Parkinson's disease, suggestive for receptor upregulation (Rinne et al., 1990a, b). With long standing treatment with L-DOPA the postsynaptic binding is normal or reduced, consistent with mild receptor down regulation (Brooks et aI., 1992a; Turjanski et al., 1997). 12.3.1.2. Hyperkinetic movement disorders The prototypic hyperkinetic movement disorder is Huntington's disease (HD), in which the loss of

medium aspiny neurons in the striatum is accompanied by profound hypometabolism and reductions of dopaminergic, opioid and GABA associated benzodiazepine binding. All of these changes can be observed in vivo with PET imaging (Myers et al., 1988; Kuwert et aI., 1990). Reductions of metabolism likely precede clinical onset and then parallel disease progression (Mazziotta et aI., 1985b; Young et al., 1986; Mazziotta et al., 1987; Young et aI., 1987; Grafton et al., 1990; Grafton et aI., 1992). The development of a direct genetic test for Huntington's disease obviates the use of functional brain imaging as a diagnostic aid for this disease (Gusella et aI., 1983, 1993). It is interesting to note that different causes of chorea can have opposing changes of striatal metabolism as listed in Table 4. There is a common pattern of striatal hypometabolism in HD, benign familial chorea and neuroacanthocytosis (Suchowersky et aI., 1986; Hosokawa et aI., 1987; Dubinsky et aI., 1989). Hypermetabolism is observed in Sydenham's chorea, lupus and tardive

167

IMAGING

NL

PO

F-OOPA

FECNT

Fig. 1. Functional neurochemistry of the basal ganglia. Integrity of pre-synaptic dopamine synthesis can be assessed with fluoro-dopa (F-DOPA). In Parkinson's disease (PD) there is a marked reduction of uptake and decarboxylation of this compound compared to normal controls (NL), particularly in the putamen. Integrity of presynaptic dopamine terminals can also be assessed by labeling the dopamine transporter protein with compounds such as 213-carbomethoxy-313-(4-chlorophenyl)8-(2-C HF)fluoroethyl)nortropane (FECNT). This proteinis normally involved in reuptake of synaptic dopamine and is a marker of dopamine terminal density. Note the marked reduction in Parkinson's disease. Images provided by Mark Goodman and Margaret Davisof Emory University, AtlantaGA. dyskinesia (Guttman et al., 1987; Weind1 et al., 1993; Pahl et al., 1995). No changes in post-synaptic dopamine receptor function have been observed in TD, suggesting the clinical symptoms may be a result of GABA related disinhibition of motor circuits rather than upregulation of the dopaminergic pathways (Blin et al., 1989; Andersson et al., 1990). The other important set of hyperkinetic movement disorders are the dystonias. The etiology of focal, segmental, hemi- or generalized dystonia, irrespective of the distribution of symptoms is remarkably diverse. MRI has been useful in identifying focal lesions within the spine, brainstem, striatum, thalamus and white matter resulting in acquired dystonia (Grafton et al., 1988; Gille et al., 1996; Kostic et al., 1996; Lehericy et al., 1996; Karsidag et al., 1998; Kurita et al., 1998). This diversity of lesion location makes it difficult to generate a unifying pathophysio-

logic model that predicts the occurrence of dystonic movements. Functional imaging is an important alternative approach for characterizing the pathophysiology of dystonia. By definition, there is forceful and prolonged simultaneous co-contraction of agonist and antagonist muscles which distort the affected extremities into stereotypic postures (Oppenheim, 1911). Thus, imaging studies examining neural substrates of the dystonias can potentially be complicated by movement related activation. Using fluorodeoxyglucose (FDG), brain glucose metabolism has been measured in both focal and generalized dystonia (Stoessl et al., 1986; Martin et al., 1988; Karbe et al., 1992; Hirato et al., 1993; Eidelberg et al., 1995; Galardi et al., 1996; Dethy et al., 1998; Mazziotta et al., 1998). Experimental strategies to avoid movement-related activation include scanning subjects in their sleep or scanning presymptomatic subjects who test positive for the dystonia gene DYTl (Eidelberg et al., 1998; Mazziotta et al., 1998). The main finding in DYTl patients was an increased covariance of metabolism within the lentiform nucleus, cerebellum and supplementary motor area, suggesting disregulated control between cortical and subcortical motor areas.

12.3.2. Proton magnetic resonance spectroscopy Given appropriate technical modifications, conventional MRI scanners can be used to perform proton magnetic resonance spectroscopy (PMRS) of brain metabolites. The most commonly detected signals are related to N-acetylaspartate (NAA) a relative marker of neuronal density, choline-containing compounds (Cho) and creatine-phosphocreatine (Cr). Absolute quantification is difficult and most studies investigate altered ratios of these metabolites with each other. Comparative studies of PD, MSA, PSP and CBGD have been performed (Federico et al., 1999; Abe et al., 2000). Single volume assays, localized to the lentiform nucleus as well as frontal cortex assays, usually demonstrate reductions of the NAA/Cho and NANCr peak ratio in all of the atypical parkinsonian syndrome patients compared to controls. Reductions of NAA/Cho or NAA/Cr are less dramatic and inconsistently observed in the frontal lobe or striatum of PD, in part due to measurement error secondary to inorganic paramagnetic substances within the basal ganglia (Clarke and Lowry, 2000). When a reduction is observed it

168

S.T. GRAFTON

Table 4 Imaging in hyperkinetic movement disorders. Huntington's disease

Neuroacanthocystosis

Benign familial chorea

DRPLA

SLE

Sydenham's chorea

Tardive dyskinesia

Metabolism

Dec dorsal striatum Dec frontal (advanced)

Dec dorsal striatum

Dec dorsal striatum

Dec dorsal striatum

Inc striatum

Inc striatum

Inc striatum

Postsynaptic D2

Dec striatum

Dec striatum

()pioidreceptors

Dec striatum

Central benzodiazepine

Dec striatum

Normal

Normal

DRPLA: Dentatorubropallidoluysian atrophy SLE: Systemic lupus erythematosis PMRS: Proton magnetic resonance spectroscopy

can correlate with disease severity (Abe et al., 2000). Recent studies show reductions of NANCr ratios in both motor cortex and temporo-parietal cortex compared to healthy controls, suggesting alterations of thalamocortical projection areas in PD (Lucetti et al., 2001). Animal models of Parkinson's disease reveal an increase of striatal glutamate activity. However, several proton magnetic resonance spectroscopy studies of striatal glutamate + glutamine relative to Cr have been normal in PD patients who are dyskinetic, non-dyskinetic and there has been no change with acute dopaminergic treatment by apomorphine (Clarke et al., 1997; Taylor-Robinson et al., 1999). This suggests the changes observed in animal models are currently too subtle to be detected byPMRS.

such as PET as well as magnetic resonance imaging (Mazziotta et al., 1985a; Belliveau et al., 1991).

12.3.3. Functional brain mapping

12.3.3.1. PET CBF The PET blood flow method requires injections of radioactive water or inhalation of radioactive CO 2 (which is converted to water in the lungs by carbonic anhydrase). The amount of radioactivity appearing in the brain is proportional to local blood flow. The temporal resolution is limited to the time it takes to acquire sufficient radioactive counts, typically on the order of 45-90 s. Spatial resolution is nominally 5 mm and more realistically 10-15 mm after image processing. Only 10-15 scans are acquired per subject due to limits on human exposure to radioactivity. Subject motion leads to image blurring, rather than signal dropout, thus the technique can be useful in patients with abnormal movements.

Over a century ago Sherrington and Roy noticed the relationship of brain blood flow and regional activity (Roy and Sherrington, 1890). It is a remarkable fact that increases of neuronal activity, down to the columnar level of spatial resolution will lead to corresponding changes of local blood flow across a slightly larger volume of tissue and with a delay of approximately 4 s (Malonek and Grinvald, 1996; Logothetis et al., 2001). This change of blood flow can be measured with radionuclide techniques

12.3.3.2. FMR1 BOLD imaging The most commonly used functional magnetic resonance imaging technique is the blood oxygen level dependent method (BOLD) (Ogawa et al., 1990). The method detects change in the contrast of T2* weighted images by varying levels of oxygen saturation. As blood flow to an area increases, so does the delivery of oxygenated blood. The method is enhanced with MRI gradients that are capable of

IMAGING

rapid acquisition using echo planar imaging (EPI) techniques (Cohen and Weisskoff, 1991). A typical commercial 1.5 Tesla scanner is capable of acquiring 10-12 slices per second with EPI imaging. Signal detection is improved with surface coils, stronger magnetic fields and acquisition at lower sampling densities (64 x 64 matrix). The method is very sensitive to head movement (signal dropout rather than signal blurring), artifacts from motion in the magnetic field (from eye or limb movements) and susceptibility artifacts maximal at air tissue interfaces such as near the sinuses. Run to run and across session variance in tMRI can be significant and create challenges for across session experimental designs (Aguirre et aI., 1998; Glover, 1999; Waldvogel et al., 2000). The tight confines of an MRI scanner have also set limits on the types of movements and behavior that can be examined in this restrictive environment. Nevertheless, tMRI has replaced PET as the most commonly used method for investigating functional anatomy in normal subjects. 12.3.3.3. Functional imaging of normal motor control Nearly two decades of experiments have mapped the functional anatomy of normal human motor behavior while subjects performed a broad range of motor tasks during brain imaging. The scope of this work is beyond the capacity of this chapter. Core observations include: (l) the delineation of the somatotopic organization of motor cortex, SMA and premotor areas (Colebatch et aI., 1991; Grafton et aI., 1991; Walter et aI., 1992; Grafton et aI., 1993; Sanes et aI., 1995); (2) the identification of premotor and parietal areas for movement selection, preparation, and on-line control (Deiber et aI., 1991, 1996; Honda et aI., 1998b; Desmurget et aI., 1999); (3) the involvement of cerebellum in movement timing and coordinated motor control (Jueptner et aI., 1996; Jueptner and Weiller, 1998; Wolpert et aI., 1998; Miall et aI., 2001); (4) the involvement of motor cortex and SMA in procedural and sequential learning (Jenkins et aI., 1994; Grafton et aI., 1995a; Karni et aI., 1995; Sadato et aI., 1996; Doyon et aI., 1997; Hazeltine et aI., 1997; Boecker et aI., 1998; Honda et aI., 1998a; Toni et aI., 1998; Grafton et aI., 2001); (5) modulation of activity in motor cortex and cerebellum as a function of force and velocity (Dettmers et aI., 1995, 1996a, b; Turner et aI., 1998).

169

These experiments form a critical background for interpreting changes of functional circuits in patients with movement disorders. 12.3.3.4. Functional brain mapping of movement disorders Functional brain imaging has been used most intensively to understanding the pathophysiologic basis of Parkinson's disease. This work forms an essential benchmark for interpreting future investigations of the functional topography of other movement disorders. The goal in PD imaging research has been to determine how altered basal ganglia (BG) information processing due to dopamine deficiency leads to altered control of movements at both the cortical and subcortical levels. A key advance was developing reliable methods that could detect movement-related activity throughout cortical and subcortical circuits. For example, PET and tMRI studies of simple movement can detect activation in almost all of the nuclei of the cortico-subcortical motor circuit (Bucher et aI., 1995; Winstein et aI., 1997; Turner et aI., 1998). A related goal asks if patterns of activity observed by imaging studies correspond to specific parkinsonian signs such as bradykinesia or akinesia. Most published imaging studies of PD have focused on the hypothesis that SMA underactivity is a cause of akinesia. In this model BG dysfunction culminates in an inadequate recruitment of SMA neurons resulting in impaired movement initiation. In principal, this is a reasonable approach as the SMA is one of the main cortical receiving areas of the BG motor circuit (Schell and Strick, 1984) and the SMA has been linked to a variety of motor behaviors that are impaired in PD, including, most notably, the selection and generation of internallyguided movements. Thus, tasks that require repeated internal selection and initiation of discrete movements should provide a good substrate for testing the association between parkinsonian akinesia and SMA activity. As predicted, PD patients show a smallerthan-normal increase in CBF in the SMA during movement tasks that require selection and execution of unidirectional ballistic joystick movements (Playford et al., 1992). In a critical follow-up experiment, a more carefully designed movement task was used to compare internally and externally generated movements in normal subjects and PD patients (Jahanshahi et aI., 1995). Subjects were trained to

170

make simple index finger extensions every 3 s by self initiation or external triggering, yoked to the same rate. The tasks required minimal working memory or other cognitive demands. PD patients had a smaller-than-normal activation of SMA for self-initiated movements. It is noteworthy that no differences in brain activity between normal subjects and PD patients were found in this study when they performed similar movements under an externally triggered condition. When PD patients performing the internal generation task are treated with dopamine agonists (apomorphine) there is a "normalization" of the movement-related activation of SMA accompanied by a reduction in reaction times (Jahanshahi et al., 1995). A similar effect of dopamine replacement therapy was observed by Rascol et aI. in PD patients performing a sequential movement task which requires frequent initiation of self-generated discrete finger-to-thumb movements (Rasco1 et al., 1992). They showed with single photon emission tomography (SPECT), that the SMA is under-activated in PD patients during this task (i.e. that that SMA had a smaller-than-normal task-related increase in CBF) and that the SMA defect normalized with apomorphine therapy. These results provide additional evidence that SMA activation is modulated by the BG motor circuit and that dopamine replacement therapy can ameliorate the inadequate thalamocortical facilitation of the SMA. Dopamine replacement therapy, by releasing thalamocortical facilitation, restores normal SMA activation patterns and movement initiation improves. Alternative models are emerging from imaging experiments to understand the symptoms of PD. One of these models is task specific compensation. Imaging studies have detected patterns of CBF in PD patients that may reflect adaptive changes, some of which may be closely linked to the particular motor task being performed. Using SPECT, Rascol et al. (1997) found that untreated PD patients demonstrated an abnormally high activation of the cerebellum ipsilateral to the moving arm when they performed sequential finger-to-thumb movements. Coincident with the cerebellar overactivation was a smaller-than-normal activation of the SMA, as predicted by the akinesia model. The increased activity in cerebellum was not seen in a separate group of PD subjects who were studied when on their normal dopamine replacement therapy. Cer-

S.T. GRAFrON

ebellar overactivation in untreated PD patients may be part of a compensatory recruitment of alternate motor circuits in the parkinsonian brain (including the visually driven cortico-ponto-cerebellar loop (Glickstein and Stein, 1991» in an attempt to overcome impaired function of the mesial frontal cortical circuits. Other studies also provide evidence of abnormal increased cerebral activity (CBF) in PD patients and indicate, additionally, that the specific patterns of under- and over-activation hinge on what behavioral task is used. Using PET, Samuel et al. found a bilateral task-related increase in CBF in dorsolateral premotor and inferior parietal cortices in untreated PD subjects performing a sequential finger tapping task (Samuel et al., 1997a). These areas were not activated in normal subjects performing the same task. Samuel et al. also found a task-related underactivation of mesial frontal and prefrontal areas in the PD subjects. These observations have been confirmed and extended recently by Catalan et aI. (Catalan et aI., 1999) in a PET study of PD and normal subjects performing either sequential finger movements of increasing complexity or an internal generation task (similar to the internal generation task first used by Playford et al. (1992). During sequential finger movements, they found a relative overactivation (i.e. a greater task-related increase in CBF than observed in normals) of bilateral parietal cortices, lateral premotor areas, and precuneus. Interestingly, Catalan et aI. observed that mesial frontal areas (anterior SMA/cingulate cortex) were activated during motor sequence performance in both PD and normal subjects, but that CBF increased progressively with more complex sequences only in the PD subjects. In contrast, when the same PD subjects performed the internal generation task, no parietal or premotor overactivations were observed and the mesial frontal areas, including SMA, were under-active, as previous studies predicted. Although some of the results described thus far can be interpreted within the model for parkinsonian akinesia, other results call for a revised or expanded model. The contrasting results for sequential movement and internal generation tasks in the Catalan et aI. study, for instance, indicate that the specific differences in brain activity between PD and normal subjects depend critically on the nature of the behavioral task being performed. The use of tasks that accentuate different facets of parkinsonian

171

IMAGING

motor impairment may expand our understanding of the functional substrates of parkinsonian symptoms other than akinesia. 12.4. Imaging therapy in movement disorders 12.4.1. Ablative surgical therapy

The current model of PO pathophysiology provides a clear rationale for surgical treatment of PO by stereotaxic ablation of the posteroventral GPi (pallidotomy). Both in PO patients and in primate models of PO, pallidotomy can reduce significantly the cardinal symptoms of PO while producing no overt side-effects (Laitinen et aI., 1992; Dogali et aI., 1995; Baron et al., 1996). The presumed mechanism of action for pallidotomy is an elimination of excessive pallidothalamic inhibition and a subsequent recovery of function in the previously under-excited frontal cortical areas. The efficacy of pallidotomy as a treatment for PO points clearly to the conclusion that most of the symptoms of PO arise from the impaired function of cortical motor areas secondary to excessive inhibitory outflow from the pallidum and not, as might be assumed, from impaired BG function per se (Wichmann and DeLong, 1996). Functional imaging studies of pallidotomy have provided results consistent with the akinesia model of PO pathophysiology (Ceballos-Baumann et aI., 1994; Grafton et al., 1994, 1995b; Samuel et aI., 1997b). A consistent finding across studies has been that following pallidotomy, there is a movement

related increase of activity in the SMA compared to rest conditions. 12.4.2. Deep brain stimulation

A relative drawback of surgical pallidotomy is the potential morbidity (acute and chronic) resulting from a permanent brain lesion. The introduction of high frequency deep brain stimulation (DBS) is an important alternative to ablation because the electrode can be introduced without producing significant brain damage and, by adjusting stimulation sites and parameters, the optimal response can be obtained. Reports of clinical response to DBS are promising (Siegfried and Lippitz, 1994; Limousin et aI., 1997; Krack et aI., 1998; DBS study group, 2001). The stimulating electrode can be positioned at several nodes of the subcortical motor circuit, including the GPi, subthalamic nucleus (STN) and the motor thalamus. Evidence to date in unblinded, non-randomized trials suggest similar maximal benefit for placement in the STN and pallidum, although patients with STN stimulators may require lower amounts of supplemental L-DOPA therapy (DBS study group, 2001). The mechanism by which DBS achieves therapeutic results remains speculative. PET has been used to examine the effects of therapeutic DBS on CBF. In the first report, Limousin et aI. explored the effects on cerebral blood flow of DBS in GPi and STN (Limousin et al., 1997). Clinically effective levels of stimulation in STN led to a greater task-related increase in CBF in

Fig. 2. Functional adaptation in Parkinson's disease. PET blood flow imaging was used to assess motor system activity during visually guided tracking at different velocities. Areas in white represent sites where PD patients show a greater increase of activity as movements become faster relative to controls. These sites include bilateral premotor cortex, motor cortex, globus pallidus and cerebellum. In PD these areas are recruited to a greater degree than normal subjects to achieve the same level of performance. Images provided by Robert Turner, DC San Francisco, California.

172

the SMA and dorsolateral prefrontal cortex compared to ineffective stimulation. Of concern, however, clinically effective stimulation of the GPi produced no significant changes in CBF. In a second study, Davis et. al., examined the effect of GPi DBS on brain activity during a "rest" condition (Davis et al., 1997). Clinically beneficial stimulation in the GPi was associated with a CBF increase in mesial frontal cortex anterior to the SMA. This result suggests that DBS altered the inhibitory GPi output in a manner analogous to ablation and thereby disinhibited the frontal thalamocortical circuit. The authors proposed that the increased CBF in the mesial cortical areas, although observed under a "resting" condition, could be responsible for a reduction of akinesia. In a more recent study, patients were examined while they performed simple paced sequential reaching movements. Concurrent regional cerebral blood flow recordings revealed a significant enhancement of motor activation responses in the left sensorimotor cortex and bilateral supplementary motor area. Significant correlations were evident between the improvement in motor performance and the regional cerebral blood flow changes mediated by stimulation (Fukuda et al., 2001). The combined results of these different imaging studies can be taken as further evidence that surgical therapeutic interventions for PD lead to increased cerebral activity in areas that are targets from pallido-thalamic connections. 12.4.3. Fetal transplantation

Functional imaging of dopaminergic function is extremely useful for assessing the in vivo viability and growth of transplants of fetal substantia nigra tissue in patients with advanced Parkinson's disease (Lindvall et al., 1989; Lindvall et al., 1994). Fluorodopa imaging can be used as an independent measure of tissue viability (Freed et al., 1990; Lindvall et al., 1990). In a recent large randomized trial there was significant evidence for increased fluorodopa uptake in the patients treated by transplantation therapy suggesting dopamine producing fiber outgrowth of transplanted tissue (Freed et al., 2001). An interesting observation emerging from the randomized clinical trials of PD using fetal transplantation has been a potent placebo effect in the patients receiving sham surgery. A functional imaging study helps to explain this puzzling response. PD

S.T. GRAFTON

patients who were told they were to get a new medical therapy for their disease were scanned and the availability of post-synaptic dopamine receptors was assessed with PET (de la Fuente-Fernandez et al., 2001). Patients given a placebo showed reduced receptor availability, suggesting they were releasing endogenous dopamine in the setting of increased reward expectancy (a new therapy). This measurable increase of endogenous dopamine could also improve parkinsonian symptoms. This finding is consistent with recent studies in non-human primates establishing the importance of the BG for facilitating reward expectancy and learning (Schultz, 2001). Acknowledgments

Supported NS33504.

by

PHS

Grants

NS37470

and

References Abe, K, Terakawa, H, Takanashi, M, Watanabe, Y, Tanaka, H, Fujita, N, Hirabuki, N and Yanagihara, T (2000) Proton magnetic resonance spectroscopy of patients with parkinsonism. Brain Res. Bull., 52: 589-595. Aguirre, GK, Zarahn, E and D'Esposito, M (1998) The variability of human, BOLD hemodynamic responses. Neuroimage, 8: 360-369. Andersson, U, Eckemas, SA, Hartvig, P, Ulin, J, Langstrom, B and Haggstrom, JE (1990) Striatal binding of llC-NMSP studied with positron emission tomography in patients with persistent tardive dyskinesia: no evidence for altered dopamine D2 receptor binding. J. Neural Transm. Gen. Sect., 79: 215-226. Aylward, EH, Codori, AM, Rosenblatt, A, Sherr, M, Brandt, J, Stine, OC, Barta, PE, Pearls on, GD and Ross, CA (2000) Rate of caudate atrophy in presymptomatic and symptomatic stages of Huntington's disease. Mov. Disord., 15: 552-560. Baron, MS, Vitek, JL, Bakay, RAE, Green, J, Kaneoke, Y, Hashimoto, T,Tumer, RS, Woodard, JL, Cole, SA, McDonald, WM and DeLong, MR (1996) Treatment of advanced Parkinson's disease by posterior GPi pallidotomy: l-year results of a pilot study. Ann. Neurol., 40: 355-366. Belliveau, JW, Kennedy, DN, McKinstry, RC, Buchbinder, BR et al. (1991) Functional mapping of the human visual cortex by magnetic resonance imaging. Science, 254(5032): 716-718. Bennett, DA, Wilson, RS, Gilley, DW and Fox, JH (1990) Clinical diagnosis of Binswanger's disease. J. Neurol. Neurosurg. Psychiatry, 53: 961-965.

173

IMAGING Blin, J, Baron, JC, Cambon, H, Bonnet, AM, Dubois, B, Loc'h, C, Maziere, B and Agid, Y (1989) Striatal dopamine D2 receptors in tardive dyskinesia: PET study. J. Neurol. Neurosurg. Psychiatry, 52: 12481252. Boecker, H, Dagher, A, Ceballos-Baumann, AO, Passingham, RE, Samuel, M, Friston, KJ, Poline, J, Dettmers, C, Conrad, B and Brooks, DJ (1998) Role of the human rostral supplementary motor area and the basal ganglia in motor sequence control: investigations with H 2 15 0 PET (published erratum appears in J. Neurophysiol. 1998 Jun; 79(6): 3301). J. Neurophysiol., 79: 107Q....1080. Brannan, TS, Burger, AA and Chaudhary, MY (1980) Bilateral basal ganglia calcification visualized on CT scan. J. Neurol. Neurosurg. Psychiatry, 40: 403-406. Brooks, DJ, Ibanez, V, Sawle, GV, Quinn, N, Lees, AJ, Mathias, CJ, Bannister, R, Marsden, CD and Frakowiak, RSJ (1990a) Differing patterns of striatal 18F_DOPA uptake in Parkinson's disease, multiple systems atrophy, and progressive supranuclear palsy. Ann. Neurol., 28: 547-555. Brooks, DJ, Salmon, EP, Mathias, CJ, Quinn, N, Leenders, KL, Bannister, R, Marsden, CD and Frackowiak, RS (1990b) The relationship between locomotor disability, autonomic dysfunction, and the integrity of the striatal dopaminergic system in patients with multiple system atrophy, pure autonomic failure, and Parkinson's disease, studied with PET. Brain, 113: 1539-1552. Brooks, DJ, Ibanez, V, Sawle, GV, Playford, ED, Quinn, N, Mathias, CJ, Lees, AJ, Marsden, CD, Bannister, R and Frackowiak, RSJ (1992a) Striatal D2 receptor status in patients with Parkinson's disease, striatonigral degeneration, and progressive supranuclear palsy, measured with 11C-raclopride and positron emission tomography. Ann. Neurol., 31: 184-192. Brooks, DJ, Jenkins, IH, Bain, P, Colebatch, JG, Thompson, PD, Findley, LJ and Marsden, CD (1992b) Comparison of the abnormal patterns of cerebral activation associated with neuropathic and essential tremor. Neurology, 42 (Suppl. 3): 423. Brucke, T, Kornhuber, J, Angelberger, P, Asenbaum, S, Frassine, Hand Podreka, I (1993) SPECT imaging of dopamine and serotonin transporters with 23I)betaCIT. Binding kinetics in the human brain. J. Neural Transm. Gen. Sect., 94: 137-146. Brucke, T, Asenbaum, S, Pirker, W, Djamshidian, S, Wenger, S, Wober, C, Muller, C and Podreka, I (1997) Measurement of the dopaminergic degeneration in Parkinson's disease with 23I) beta-CIT and SPECT. Correlation with clinical findings and comparison with multiple system atrophy and progressive supranuclear palsy. 1. Neural Transm. Suppl., 50: 9-24.

e

e

Bucher, SF, Seelos, KC, Stehling, M, Oertel, WH, Paulus, W and Reiser, M (1995) High-resolution activation mapping of basal ganglia with functional magnetic resonance imaging. Neurology, 45: 18Q....182. Catalan, MJ, Ishii, K, Honda, M, Samii, A and Hallett, M (1999) A PET study of sequential finger movements of varying length in patients with Parkinson's disease. Brain, 122: 483-495. Ceballos-Baumann, AO, Obeso, JA, Vitek, JL, Delong, MR, Bakay, R, Linazosoro, G and Brooks, DJ (1994) Restoration of thalamocortical activity after posteroventral pallidotomy in Parkinson's disease. Lancet, 344: 814. Clarke, CE and Lowry, M (2000) Basal ganglia metabolite concentrations in idiopathic Parkinson's disease and multiple system atrophy measured by proton magnetic resonance spectroscopy. Eur. J. Neurol., 7: 661-665. Clarke, CE, Lowry, M and Horsman, A (1997) Unchanged basal ganglia N-acetylaspartate and glutamate in idiopathic Parkinson's disease measured by proton magnetic resonance spectroscopy. Mov. Disord., 12: 297-301. Cohen, MS and Weisskoff, RM (1991) Ultra-fast imaging. Magn. Reson. Imag., 9: 1-37. Colebatch, JG, Findley, LJ, Frackowiak, RSJ, Marsden, CD and Brooks, DJ (1990) Preliminary report: activation of the cerebellum in essential tremor. Lancet, 336: 1028-1030. Colebatch, JG, Deiber, M-P, Passingham, RE, Friston, KJ and Frackowiak, RSJ (1991) Regional cerebral blood flow during voluntary arm and hand movements in human subjects. J. Neurophysiol., 65: 1392-1401. Critchley, M (1929) Arteriosclerotic parkinsonism. Brain, 52: 23-83. Crossman, AR, Mitchell, IJ and Sambrook, MA (1985) Regional brain uptake of 2-deoxyglucose in N-methyl4-phenyl-l,2,3,6- tetrahydropyridine (MPTP)-induced parkinsonism in the macaque monkey. Neuropharmacology, 24: 587-591. Davis, KD, Taub, E, Houle, S, Lang, AE, Dostrovsky, JO, Tasker, RR and Lozano, AM (1997) Globus pallidus stimulation activates the cortical motor system during alleviation of parkinsonian symptoms. Nat. Med., 3: 671-674. DBS Study Group (2001) Deep-brain stimulation of the subthalamic nucleus or the pars interna of the globus pallidus in Parkinson's disease. N. Engl. J. Med., 345: 956-963. De la Fuente-Fernandez, R, Ruth, TJ, Sossi, V, Schulzer, M, Calne, DB and Stoessl, AJ (2001) Expectation and dopamine release: mechanism of the placebo effect in Parkinson's disease. Science, 293: 1164-1166. De Voider, AG, Francart, J, Laterre, C, Dooms, G, Bol, A, Michel, C and Goffinet, AM (1989) Decreased glucose

174 utilization in the striatum and frontal lobe in probable striatonigral degeneration. Ann. Neurol., 26: 239-247. Deiber, MP, Passingham, RE, Colebatch, JG, Friston, KJ, Nixon, PD and Frackowiak, RS (1991) Cortical areas and the selection of movement: a study with positron emission tomography. Exp. Brain Res., 84: 393-402. Deiber, MP, Ibanez, V, Sadato, N and Hallett, M (1996) Cerebral structures participating in motor preparation in humans: a positron emission tomography study. J. Neurophysiol., 75: 233-247. Desmurget, M, Epstein, CM, Turner, RS, Prablanc, C, Alexander, GE and Grafton, ST (1999) Role of the posterior parietal cortex in updating reaching movements to a visual target. Nat. Neurosci., 2: 563-567. Dethy, S, Van Blercom, N, Damhaut, P, Wikler, D, Hildebrand, J and Goldman, S (1998) Asymmetry of basal ganglia glucose metabolism and DOPA responsiveness in parkinsonism. Mov. Disord., 13: 275-280. Dettmers, C, Fink, GR, Lemon, RN, Stephan, KM, Passingham, RE, Silbersweig, D, Holmes, A, Ridding, MC, Brooks, DJ and Frackowiak, RS (1995) Relation between cerebral activity and force in the motor areas of the human brain. J. Neurophysiol., 74: 802-815. Dettrners, C, Connelly, A, Stephan, KM, Turner, R, Friston, KJ, Frackowiak, RS and Gadian, DG (1996a) Quantitative comparison of functional magnetic resonance imaging with positron emission tomography using a force-related paradigm. Neuroimage, 4: 201-209. Dettmers, C, Lemon, RN, Stephan, KM, Fink, GR and Frackowiak, RSJ (1996b) Cerebral activation during the exertion of sustained static force in man. NeuroReport, 7: 2103-2110. Dogali, M, Fazzini, E, Kolodny, E, Eidelberg, D, Sterio, D, Devinsky, 0 and Betic, A (1995) Stereotactic ventral pallidotomy for Parkinson's disease. Neurology, 45: 753-761. Doyon, J, Gaudreau, D, Laforce, R Jr., Castonguay, M, Bedard, PJ, Bedard, F and Bouchard, JP (1997) Role of the striatum, cerebellum, and frontal lobes in the learning of a visuomotor sequence. Brain Cogn., 34: 218-245. Dubinsky, RM, Hallett, M, Levey, R and Di Chiro, G (1989) Regional brain glucose metabolism in neuroacanthocytosis. Neurology, 39: 1253-1255. Eidelberg, D, Takikawa, S, Moeller, JR, Dhawan, V, Redington, K, Chaly, T, Robeson, W, Dahl, JR, Margouleff, D, Fazzini, E, Przedborski, S and Fahn, S (1993) Striatal hypometabolism distinguishes striatonigral degeneration from Parkinson's disease. Ann. Neurol., 33: 518-527. Eidelberg, D, Moeller, JR, Dhawan, V, Spetsieris, P, Takikawa, S, Ishikawa, T, Chaly, T, Robeson, W, Margouleff, D and Przedborski, S et al. (1994) The

S.T. GRAFI'ON

metabolic topography of parkinsonism. J. Cereb. Blood Flow Metab., 14: 783-801. Eidelberg, D, Moeller, JR, Ishikawa, T, Dhawan, V, Spetsieris, P, Przedborski, S and Fahn, S (1995) The metabolic topography of idiopathic torsion dystonia. Brain, 118: 1473-1484. Eidelberg, D, Moeller, JR, Antonini, A, Kazumata, K, Nakamura, T, Dhawan, V, Spetsieris, P, Del.eon, D, Bressman, SB and Fahn, S (1998) Functional brain networks in DYTl dystonia. Ann. Neurol., 44: 303312. Federico, F, Simone, IL, Lucivero, V, Mezzapesa, DM, de Mari, M, Lamberti, P, Petruzzellis, M and Ferrari, E (1999) Usefulness of proton magnetic resonance spectroscopy in differentiating parkinsonian syndromes. Ital. J. Neurol. Sci., 20: 223-229. Freed, CR, Breeze, RE, Rosenberg, NL, Schneck, SA, Wells, TH, Barrett, IN, Grafton, ST, Huang, SC, Eidelberg, D and Rottenberg, DA (1990) Transplantation of human fetal dopamine cells for Parkinson's disease. Arch. Neurol., 47: 505-512. Freed, CR, Greene, PE, Breeze, RE, Tsai, WY, DuMouchel, W, Kao, R, Dillon, S, Winfield, H, Culver, S, Trojanowski, JQ, Eidelberg, D and Fahn, S (2001) Transplantation of embryonic dopamine neurons for severe Parkinson's disease. N. Engl. J. Med., 344: 710-719. Fukuda, M, Mentis, M, Ghilardi, MF, Dhawan, V, Antonini, A, Hammerstad, J, Lozano, AM, Lang, A, Lyons, K, Koller, W, Ghez, C and Eidelberg, D (2001) Functional correlates of pallidal stimulation for Parkinson's disease. Ann. Neurol., 49: 155-164. Fulham, MJ, Dubinsky, RM, Polinsky, RJ, Brooks, RA, Brown, RT, Curras, MT, Baser, S, Hallett, M and Di Chiro, G (1991) Computed tomography, magnetic resonance imaging and positron emission tomography with C8F)fluorodeoxyglucose in multiple system atrophy and pure autonomic failure. Clin. Auton. Res., I: 27-36. Galardi, G, Perani, D, Grassi, F, Bressi, S, Amadio, S, Antoni, M, Comi, GC, Canal, N and Fazio, F (1996) Basal ganglia and thalamo-cortical hypermetabolism in patients with spasmodic torticollis. Acta Neurol. Scand., 94: 172-176. Garnett, ES, Firnau, G and Nahmias, C (1983) Dopamine visualized in the basal ganglia of living man. Nature, 305: 137-138. Garnett. ES, Nahmias. C and Firnau, G (1984) Central dopaminergic pathways in hemiparkinsonism examined by positron emission tomography. Can. J. Neurol. Sci., 11: 174-179. Gille, M, Van den Bergh, P, Ghariani, S, Guettat, L, Delbecq, J and Depre, A (1996) Delayed-onset hemidystonia and chorea following contralateral infarction

IMAGING of the posterolateral thalamus. A case report. Acta Neurol. Belg., 96: 307-311. Gilman, S, Koeppe, RA, Junek, L, Kluin, KJ, Lohman, M and St Laurent, RT (1994) Patterns of cerebral glucose metabolism detected with positron emission tomography differ in multiple system atrophy and olivopontocerebellar atrophy. Ann. Neurol., 36: 166-175. Glickstein, M and Stein, M (1991) Paradoxical movement in Parkinson's disease. Trends Neurosci., 14: 480-483. Glover, GH (1999) Deconvolution of impulse response in event-related BOLD fMRI. Neuroimage, 9: 416-429. Grafton, ST, Bahls, FH and Bell, KR (1988) Acquired focal dystonia following recovery from central pontine myelinolysis (letter). J. Neurol. Neurosurg. Psychiatry, 51: 1354-1355. Grafton, ST, Mazziotta, JC, Pahl, JJ, St. George-Hyslop, P, Haines, JL, Gusella, J, Hoffman, IM, Baxter, LR and Phelps, ME (1990) A comparison of neurological, metabolic, structural, and genetic evaluations in persons at risk for Huntington's disease. Ann. Neurol., 28: 614-621. Grafton, ST, Woods, RP, Mazziotta, JC and Phelps, ME (1991) Somatotopic mapping of the primary motor cortex in humans: activation studies with cerebral blood flow and positron emission tomography. J. Neurophysiol., 66: 735-743. Grafton, ST, Mazziotta, JC, Pahl, JJ, St. George-Hyslop, P, Haines, JL, Gusella, J, Hoffman, JM, Baxter, LR and Phelps, ME (1992) Serial changes of cerebral glucose metabolism and caudate size in persons at risk for Huntington's disease. Arch. Neurol., 49: 1161-1167. Grafton, ST, Woods, RP and Mazziotta, JC (1993) Withinarm somatotopy in human motor areas determined by positron emission tomography imaging of cerebral blood flow. Exp. Brain Res., 95: 172-176. Grafton, ST, Couldwell, W, Sutton, J, Lew, MF, Mallory, L and Waters, C (1994) Pallidotomy increases movement related activity in motor cortical areas in Parkinson's disease: A PET activation study. Ann. Neurol., 36: 315. Grafton, ST, Hazeltine, E and Ivry, R (1995a) Functional anatomy of sequence learning in normal humans. 1. Cogn. Neurosci., 7: 497-510. Grafton, ST, Waters, C, Sutton, J, Lew, MF and Couldwell, W (l995b) Pallidotomy increases activity of motor association cortex in Parkinson's disease: a positron emission tomographic study. Ann. Neurol., 37: 776783. Grafton, ST, Salidis, J and Willingham, DB (2001) Motor learning of compatible and incompatible visuomotor maps. J. Cogn. Neurosci., 13: 217-231. Gusella, JF, Wexler, NS and Conneally, PH et al. (1983) A polymorphic DNA marker genetically linked to Huntington's disease. Nature, 306: 234-238.

175 Gusella, JF, MacDonald, ME, Ambrose, CM and Duyao, MP (1993) Molecular genetics of Huntington's disease. Arch. Neurol., 50: 1157-1163. Guttman, M, Lang, AE, Garnett, ES, Nahmias, C, Firnau, G, Tyndel, FJ and Gordon, AS (1987) Regional cerebral glucose metabolism in SLE chorea: further evidence that striatal hypometabolism is not a correlate of chorea. Mov. Disord., 2: 201-210. Hazeltine, E, Grafton, ST and Ivry, R (1997) Attention and stimulus characteristics determine the locus of motorsequence encoding. A PET study. Brain, 120: 123140. Hirato, M, Horikoshi, S, Kawashima, Y, Satake, K, Shibazaki, T and Ohye, C (1993) Different cerebral metabolism between parkinsonian rigidity and hyperkinesia (DID, chorea, dystonia). A PET study. Adv. Neurol., 60: 511-514. Holmes, G (1917) The symptoms of acute cerebellar injuries due to gunshot injuries. Brain, 40: Honda, M, Deiber, MP, Ibanez, V, Pascual-Leone, A, Zhuang, P and Hallett, M (l998a) Dynamic cortical involvement in implicit and explicit motor sequence learning. A PET study. Brain, 121: 2159-2173. Honda, M, Wise, SP, Weeks, RA, Deiber, MP and Hallett, M (l998b) Cortical areas with enhanced activation during object-centerd spatial information processing. A PET study. Brain, 121: 2145-2158. Hosokawa, S, Ichiya, Y, Kuwabara, Y, Ayabe, Z, Mitsuo, K, Goto, I and Kato, M (1987) Positron emission tomography in cases of chorea with different underlying diseases. J. Neurol. Neurosurg. Psychiatry, 50: 1284-1287. Hu, MT, White, SJ, Herlihy, AH, Chaudhuri, KR, Hajnal, N and Brooks, DJ (2001) A comparison of 18F_DOPA PET and inversion recovery MRI in the diagnosis of Parkinson's disease. Neurology, 56: 1195-1200. Illum, F and Dupont, E (1985) Prevalences ofCT-detected calcification in the basal ganglia in idiopathic hypoparathyroidism and pseudohypoparathyroidism. Neuroradiology, 27: 32-37. Jahanshahi, M, Jenkins, IH, Brown, RG, Marsden, CD, Passingham, RE and Brooks, DJ (1995) Self-initiated vs. externally triggered movements. I. An investigation using measurement of regional cerebral blood flow with PET and movement-related potentials in normal and Parkinson's disease subjects (see comments). Brain, 118: 913-933. Jenkins, IH, Brooks, DJ, Nixon, PD, Frackowiak, RS and Passingham, RE (1994) Motor sequence learning: a study with positron emission tomography. J. Neurosci., 14: 3775-3790. Jueptner, M and Weiller, C (1995) Review: does measurement of regional cerebral blood flow reflect synaptic

176 activity? - implications for PET and tMRI (In Process Citation). Neuroimage, 2: 148-156. Jueptner, M, Flerich, L, WeiHer, C, Mueller, SP and Diener, H-C (1996) The human cerebellum and temporal information processing - results from a PET experiment. NeuroReport, 7: 2761-2765. Jueptner, M and Weiller, C (1998) A review of differences between basal ganglia and cerebellar control of movements as revealed by functional imaging studies. Brain, 121: 1437-1449. Karbe, H, Holthoff, VA, Rudolf, J, Herholz, K and Heiss, WD (1992) Positron emission tomography demonstrates frontal cortex and basal ganglia hypometabolism in dystonia. Neurology, 42: 1540-1544. Karni, A, Meyer, G, Jezzard, P, Adams, MM, Turner, R and Ungerleider, LG (1995) Functional MRI evidence for adult motor cortex plasticity during motor skill learning. Nature, 377: 155-157. Karsidag, S, Ozer, F, Sen, A and Arpaci, B (1998) Lesion localization in developing poststroke hand dystonia. Eur. Neurol., 40: 99-104. Kostic, VS, Stojanovic-Svetel, M and Kacar, A (1996) Symptomatic dystonias associated with structural brain lesions: report of 16 cases. Can. J. Neurol. Sci., 23: 53-56. Krack, P, Pollak, P, Limousin, P, Hoffmann, D, Benazzouz, A, Le Bas, JF, Koudsie, A and Benabid, AL (1998) Opposite motor effects of pallidal stimulation in Parkinson's disease. Ann. Neurol., 43: 180-192. Kuhl, DE, Metter, EJ and Reige, WH (1984) Patterns of local cerebral glucose utilization determined in Parkinson's disease by the CSF)fiuorodeoxyglucose method. Ann. Neurol., 15: 419-424. Kuhl, DE, Metter, EJ, Benson, DF et al. (1985) Similarities of cerebral glucose metabolism in Alzheimer's and Parkinsonian dementia. J. Cereb. Blood Flow Metab., 5: S169-S170. Kurita, H, Sasaki, T, Suzuki, I and Kirino, T (1998) Basal ganglia arteriovenous malformation presenting as "writer's cramp". Childs Nerv. Syst., 14: 285-287. Kuwert,T, Lange, HW, Langen, K-J, Herzog, H, Aulich, A and Feinendegen, LE (1990) Cortical and subcortical glucose consumption measured by PET in patients with Huntington's disease. Brain, 113: 1405-1423. Laihinen, AO, Rinne, JO, Nagren, KA, Lehikoinen, PK, Oikonen, VJ, Ruotsalainen, UH, Ruottinen, HM and Rinne, UK (1995) PET studies on brain monoamine transporters with carbon-11-beta-CIT in Parkinson's disease. J. Nucl. Med., 36: 1263-1267. Laitinen, LV, Bergenheim, AT and Hariz, MI (1992) Leksell's posterventral pallidotomy in the treatment of Parkinson's disease. J. Neurosurg., 76: 53-61. Leenders, KL, Herold, S, Brooks, DJ, Palmer, AJ, Turton, D, Fimau, G, Garnett, ES, Nahmias, C and Veall, N

S.T. GRAFTON

(1984) Pre-synaptic and post-synaptic dopaminergic system in human brain. Lancet, 2: 110-111. Leenders, KL, Palmer, AJ and Quinn, N et al. (1986) Brain dopamine metabolism in patients with Parkinson's disease measured with positron emission tomography. J. Neurol. Neurosurg. Psychiatry, 49: 853-860. Lehericy, S, Vidailhet, M, Dormont, D, Pierot, L, Chiras, J, Mazetti, P, Marsault, C and Agid, Y (1996) Striatopallidal and thalamic dystonia. A magnetic resonance imaging anatomoclinical study. Arch. Neurol., 53: 241-250. Limousin, P, Greene, J, Pollak, P, Rothwell, J, Benabid, ALand Frackowiak, RSJ (1997) Changes in cerebral activity pattern due to subthalamic nucleus or internal pallidum stimulation in Parkinson's disease. Ann. Neurol., 42: 283-291. Lindvall, 0, Rehncrona, S, Brundin, P, Gustavii, B, Astedt, B, Widner, H, Lindholm, T, Bjorklund, A, Leenders, KL, Rothwell, JC, Frackowiak, R, Marsden, D, Johnels, B, Steg, G, Freedman, R, Hoffer, BJ, Seiger, A, Bygdeman, M, Strombert, I and Olson, L (1989) Human fetal dopamine neurons grafted into the striatum in two patients with severe Parkinson's disease: a detailed account of methodology and a 6-month follow-up. Arch. Neurol., 46: 615-631. Lindvall, 0, Brundin, P, Widner, H, Rehncrona, S, Gustavii, B, Frackowiak, R, Leenders, KL, Sawle, G, Rothwell, JC, Marsden, CD and Bjorklund, A (1990) Grafts of fetal dopamine neurons survive and improve motor function in Parkinson's disease. Science, 247: 574-577. Lindvall, 0, Sawle, G, Widner, H, Rothwell, JC, Bjorklund, A, Brooks, D, Brundin, P, frackowiak, R, Marsden, CD, Odin, P and Rehncrona, S (1994) Evidence for long-term survival and function of dopaminergic grafts in progressive Parkinson's disease. Ann. Neurol., 35: 172-180. Logothetis, NK, Pauls, J, Augath, M, Trinath, T and Oeltermann, A (2001) Neurophysiological investigation of the basis of the tMRI signal. Nature, 412: 150-157. Lucetti, C, Del Dotto, P, Gambaccini, G, Bernardini, S, Bianchi, MC, Tosetti, M and Bonuccelli, U (2001) Proton magnetic resonance spectroscopy (1H-MRS) of motor cortex and basal ganglia in de novo Parkinson's disease patients. Neurol. Sci., 22: 69-70. Malonek, D and Grinvald, A (1996) Interactions between electrical activity and cortical microcirculation revealed by imaging spectroscopy: implications for functional brain mapping. Science, 272: 551-554. Martin, WR, Stoessl, AJ, Adam, MJ, Ammann, W, Bergstrom, M, Harrop, R, Laihinen, A, Rogers, JG, Ruth, TJ and Sayre, CI et al. (1987) Positron emission

IMAGING

tomography in Parkinson's disease: glucose and DOPA metabolism. Adv. Neurol., 45: 95-98. Martin, WR, Stoessl, AJ, Palmer, M, Adam, MJ, Ruth, TJ, Grierson, JR, Pate, BD and Caine, DB (1988) PET scanning in dystonia. Adv. Neurol., 50: 223-229. Mazziotta, JC, Huang, S-C, Phelps, ME, Carson, RE, MacDonald, NS and Mahoney, K (1985a) A noninvasive positron computed tomography technique using oxygen-15-labeled water for the evaluation of neurobehavioral task batteries. J. Cereb. Blood Flow Metab., 5: 70-78. Mazziotta, JC, Wapenski, J, Phelps, ME et al. (l985b) Cerebral glucose utilization and blood flow in Huntington's disease: symptomatic and at risk subjects. J. Cereb. Blood Flow Metab., 5: S25. Mazziotta, JC, Phelps, ME, Pahl, JJ, Huang, SC, Baxter, LR, Riege, WH, Hoffman, JM, Kuhl, DE, Lanto, AB, Wapenski, JA and Markham, CH (1987) Reduced cerebral glucose metabolism in asymptomatic subjects at risk for Huntington's disease. N. Engl. J. Med., 316: 357-362. Mazziotta, JC, Hutchinson, M, Fife, TD and Woods, R (1998) Advanced neuroimaging methods in the study of movement disorders: dystonia and blepharospasm. Adv. Neurol., 78: 153-160. Mazziotta, J, Toga, A, Evans, A, Fox, P, Lancaster, J, Zilles, K, Woods, R, Paus, T, Simpson, G, Pike, B, Holmes, C, Collins, L, Thompson, P, MacDonald, D, Iacoboni, M, Schormann, T, Amunts, K, PalomeroGallagher, N, Geyer, S, Parsons, L, Narr, K, Kabani, N, Le Goualher, G, Feidler, J, Smith, K, Boomsma, D, Pol, HR, Cannon, T, Kawashima, Rand Mazoyer, B (2001) A four-dimensional probabilistic atlas of the human brain. J. Am. Med. Inform. Assoc., 8: 401-430. Miall, RC, Reckess, GZ and Imarnizu, H (2001) The cerebellum coordinates eye and hand tracking movements. Nat. Neurosci., 4: 638-644. Muenter, MD and Whisnant, JP (1968) Basal ganglia calcification, hypoparathyroidism, and extrapyramidal motor manifestations. Neurology, 18: 1075-1083. Murphy, MJ (1979) Clinical correlations of CT scandetected calcifications of the basal ganglia. Ann. Neurol., 6: 507-511. Myers, R, Vonsattel, J, Steven, T, Cupples, LA, Richardson, EP, Martin, JB and Bird, ED (1988) Clinical and neuropathologic assessement of severity in Huntington's disease. Neurology, 38: 341-347. Nudo, RJ and Masterton, RB (1986) Stimulation-induced C4C)rdeoxyglucose labeling of synaptic activity in the central auditory system. 1. Compo Neurol., 245: 553565. Ogawa, S, Lee, TM, Nayak, AS and Glynn, P (1990) Oxygenation-sensitive contrast in magnetic resonance

177 image of rodent brain at high magnetic fields. Magn. Reson. Med., 14: 68-78. Oppenheim, H (1911) Uber eine eigenartige Krampfkrankheit des kinlichen und jugendlichen Alters (Dysbasia lordotic a progressiva, Dystonia musculorum deformans). Neurol Zbl., 30: 1090-lI07. Otsuka, M, Ichiya, Y, Hosokawa, S, Kuwabara, Y, Tahara, T, Fukumura, T, Kato, M, Masuda, K and Goto, I (1991) Striatal blood flow, glucose metabolism and lXF_ DOPA uptake: difference in Parkinson's disease and atypical Parkinsonism. J. Neurol. Neurosurg. Psychiatry, 54: 898-904. Otsuka, M, Ichiya, Y, Kuwabara, Y, Hosokawa, S, Akashi, Y, Yoshida, T, Fukumura, T, Masuda, K, Goto, I and Kato, M (1994) Striatal IXF_DOPA uptake and brain glucose metabolism by PET in patients with syndrome of progressive ataxia. J. Neurol. Sci., 124: 198-203. Pahl, JJ, Mazziotta, JC, Bartzokis, G, Cummings, J, Altschuler, L, Mintz, J, Marder, SR and Phelps, ME (1995) Positron-emission tomography in tardive dyskinesia. J. Neuropsychiatry Clin. Neurosci., 7: 457-465. Phelps, ME, Hoffman, EJ and Mullani, NA et al. (1975) Application of annihilaton coincidence detection to trans axial reconstruction tomography. 1. Nucl. Med., 16: 210-224. Playford, ED, Jenkins, IH, Passingham, RE, Nutt, J, Frackowiak, RSJ and Brooks, OJ (1992) Impaired mesial frontal and putamen activation in Parkinson's disease: a positron emission tomography study. Ann. Neurol., 32: 151-161. Podreka, I, Suess, E, Goldenberg, G et al. (1987) Initial experience with technitium-99m HM-PAO brain SPECT. J. Nucl. Med., 28: 1657-1666. Rascol, 0, Sabatini, U, Chollet, F, Celsis, P, Montastruc, JL, Marc-Vergnes, J-P and Rascol, A (1992) Supplementary and primary sensory motor area activity in Parkinson's disease. Regional cerebral blood flow changes during finger movements and effects of apomorphine. Arch. Neurol., 49: 144-148. Rascol, 0, Sabatini, U, Fabre, N, Brefel, C, Loubinoux, I, Celsis, P, Senard, JM, Montastruc, JL and Chollet, F (1997) The ipsilateral cerebellar hemisphere is overactive during hand movements in akinetic parkinsonian patients. Brain, 120: 103-110. Reivich, M, Kuhl, D, Wolf, A, Greenberg, J, Phelps, ME, Ido, T, Casella, V, Fowler, J, Hoffman, E, Alavi, A, Som, P and Sokoloff, L (1979) The C8F)fluorodeoxyglucose method for the measurement of local cerebral glucose utilization in man. Circ. Res., 44: 127-137. Rinne, JO, Laihinen, A, Nagren, K, Bergman, J, Solin, 0, Haaparanta, M, Ruotsalainen, U and Rinne, UK (1990a) PET demonstrates different behaviour of striatal dopamine D-I and D-2 receptors in early Parkinson's disease. J. Neurosci. Res., 27: 494-499.

178 Rinne, UK, Laihinen, A, Rinne, JO, Nagren, K, Bergman, J and Ruotsalainen, U (1990b) Positron emission tomography demonstrates dopamine D2 receptor supersensitivity in the striatum of patients with early Parkinson's disease. Mov. Disord., 5: 55-59. Rosenthal, G, Gilman, S, Koeppe, RA, Kluin, KJ, Markel, DS, Junek, Land Gebarski, SS (1988) Motor dysfunction in olivopontocerebellar atrophy is related to cerebral metabolic rate studied with positron emission tomography. Ann. Neurol., 24: 414-419. Roy, CS and Sherrington, CS (1890) On the regulation of the blood supply of the brain. J. Physiol. (Lond.), 11: 85-108. Sachs, C, Sjoberg, HE and Ericson, K (1982) Basal ganglia calcifications on CT: relation to hypoparathyroidism. Neurology, 32: 779-782. Sadato, N, Campbell, G, Ibanez, V, Deiber, M and Hallett, M (1996) Complexity affects regional cerebral blood flow change during sequential finger movements. J. Neurosci., 16: 2691-2700. Samuel, M, Ceballos-Baumann, AO, Blin, J, Uema, T, Boecker, H, Passingham, RE and Brooks, DJ (1997a) Evidence for lateral premotor and parietal overactivity in Parkinson's disease during sequential and bimanual movements. A PET study. Brain, 120: 963-976. Samuel, M, Ceballos-Baumann, AO, Turjanski, N, Boecker, H, Gorospe, A, Linazasoro, G, Holmes, AP, DeLong, MR, Vitek, JL, Thomas, DG, Quinn, NP, Obeso, JA and Brooks, DJ (1997b) Pallidotomy in Parkinson's disease increases supplementary motor area and prefrontal activation during performance of volitional movements an Hz(15)0 PET study. Brain, 120: 1301-1313. Sanes, IN, Donoghue, JP,Thangariaj, V, Edelman, RR and Warach, S (1995) Shared neural substrates controlling hand movements in human motor cortex. Science, 268: 1775-1777. Schell, GR and Strick, PL (1984) The origin of thalamic inputs to the arcuate premotor and supplementary motor areas. J. Neurosci., 4: 539-560. Schultz, W (2001) Reward signaling by dopamine neurons. Neuroscientist, 7: 293-302. Siegfried, J and Lippitz, B (1994) Bilateral chronic electrostimulation of ventroposterolateral pallidum: a new therapeutic approach for alleviating all parkinsonian symptoms. Neurosurgery, 35: 1126-1129; discussion 1129-1130. Sokoloff, L (1977) Relation between physiological function and energy metabolism in the central nervous system. 1 Neurochem., 29: 13-26. Stoessl, AI, Martin, WR, Clark, C, Adam, MJ, Ammann, W, Beckman, JH, Bergstrom, M, Harrop, R, Rogers, JG and Ruth, TJ et al. (1986) PET studies of cerebral

S.T. GRAFTON

glucose metabolism in idiopathic torticollis. Neurology, 36: 653-657. Suchowersky, 0, Hayden, MR, Martin, WR, Stoessl, AJ, Hildebrand, AM and Pate, BD (1986) Cerebral metabolism of glucose in benign hereditary chorea. Mov. Disord., 1: 33-44. Taylor-Robinson, SD, Turjanski, N, Bhattacharya, S, Seery, JP, Sargentoni, J, Brooks, DJ, Bryant, DJ and Cox, IJ (1999) A proton magnetic resonance spectroscopy study of the striatum and cerebral cortex in Parkinson's disease. Metab. Brain Dis., 14: 45-55. Thompson, PD and Marsden, CD (1987) Gait disorder of subcortical arteriosclerotic encephalopathy: Binswanger's disease. Mov. Disord., 2: 1-8. Toga, AW and Thompson, PM (2001) Maps of the brain. Anat. Rec., 265: 37-53. Toni, I, Krams, M, Turner, R and Passingham, RE (1998) The time course of changes during motor sequence learning: a whole-brain fMRI study. Neuroimage, 8: 50-61. Turjanski, N, Lees, AI and Brooks, DJ (1997) In vivo studies on striatal dopamine D I and D2 site binding in L-DOPA-treated Parkinson's disease patients with and without dyskinesias. Neurology, 49: 717-723. Turner, RS, Grafton, ST, Votaw, JR, Delong, MR and Hoffman, JM (1998) Motor subcircuits mediating the control of movement velocity: A PET study (In Process Citation). 1 Neurophysiol., 80: 2162-2176. Waldvogel, D, Van Gelderen, P, Immisch, I, Pfeiffer, C and Hallett, M (2000) The variability of serial fMRI data: correlation between a visual and a motor task. NeuroReport, 11: 3843-3847. Walter, H, Kristeva, R, Knorr, U, Schlaug, G, Huang, Y, Steinmetz, H, Nebeling, B, Herzog, H and Seitz, RJ (1992) Individual somatotopy of primary sensorimotor cortex revealed by intermodal matching of MEG, PET, and MRI. Brain Topogr., 5: 183-187. Waters, CH (1993) Structural lesions and parkinsonism. In: MB Stem and WC Koller (Eds.), Parkinsonian Syndromes. Marcel Dekker, New York, pp. 483-501. Weindl, A, Kuwert, T, Leenders, KL, Poremba, M, Grafin Von Einsiedel, H, Antonini, A, Herzog, H, Scholz, D, Feinendegen, LE and Conrad, B (1993) Increased striatal glucose consumption in Sydenham's chorea. Mov. Disord., 8: 437-444. Wichmann, T and DeLong, MR (1996) Functional and pathophysiological models of the basal ganglia. Curr. Opin. Neurobiol., 6: 751-758. Winstein, CJ, Grafton, ST and Pohl, PS (1997) Motor task difficulty and brain activity: investigation of goaldirected reciprocal aiming using positron emission tomography. J. Neurophysiol., 77: 1581-1594.

IMAGING

Wolpert, DM, Miall, RC and Kawato, M (1998) Internal models in the cerebellum. Trends Cogn. Sci., 2: 338-347. Young, AB, Penney, JB, Starosta-Rubinstein, S, Markel, DS, Berent, S, Giordani, B, Ehrenkaufer, R, Jewett, D and Hichwa, R (1986) PET scan investigations of Huntington's disease: Cerebral metabolic correlates of

179 neurological features and functional decline. Ann. Neural., 20: 296-303. Young, AB, Penney, JB, Starosta-Rubinstein, S, Markel, D, Berent, S, Rothley, J, Betley, A and Hichwa, R (1987) Normal caudate glucose metabolism in persons at risk for Huntington's disease. Arch. Neural., 44: 254-257.

Movement Disorders Handbook of Clinical Neurophysiology, Vol. I M. Hallett (Ed.) © 2003 Elsevier B.V. All rights reserved

181 CHAPTER 13

Accelerometry Rodger J. Elble* Department of Neurology, Southern Illinois University School of Medicine, P.O. Box 19643, Springfield, IL 62794-9643, USA

Accelerometers are small lightweight motion transducers that are capable of measuring accelerations less than 0.02 G (G=9.807 m/s', the static acceleration of gravity). The use of accelerometers in human motion analysis is the focus of this chapter. The Internet is a rich source of additional information pertaining to basic technology and manufacturers, and Table 1 contains an incomplete list of useful sites.

13.1. Common types of accelerometers Several types of accelerometers are now commercially available, but piezoresistive, piezoelectric and capacitance accelerometers are employed most commonly in human applications. These accelerometers are based on Newton's law of mass acceleration (Force=massxacceleration) and Hooke's law of spring action (Force=spring constant x change in length of a spring). Therefore, for a known mass attached to an elastic material, one can relate acceleration to the extent that the elastic material is stretched or compressed. Piezoresistive accelerometers consist of a small mass attached to a semiconductor beam that behaves like a spring. Deflection of the beam is measured with strain gauges that are connected in a Wheatstone bridge. The output voltage of the Wheatstone bridge is proportional to acceleration. Piezoelectric accelerometers contain a mass that is attached to a piezoelectric crystal, which behaves as a spring. Deformation of the crystal produces a small voltage (- millivolts) that is proportional to acceleration. Capacitance accelerometers contain a variable

* Correspondence to: Dr. R.I. Elb1e, Department of Neurology, Southern lllinois University School of Medicine, P.O. Box 19643, Springfield, IL 62794-9643, USA. E-mail address: [email protected] Tel.: 217-524-7881 (ext. 3002); fax: 217-524-1903.

capacitor, in which the gap between the capacitor plates changes in proportion to acceleration.

13.2. Technical specifications and considerations In selecting an accelerometer, one must consider the required size, weight, durability, frequency range, linear amplitude range, sensitivity, transverse sensitivity, and resolution. These specifications for accelerometers are provided on the Internet web sites of most manufacturers (Table 1). Miniature accelerometers are now so small and lightweight that many are suitable for most human applications. Triaxial accelerometers with an approximate weight and volume of 10-20 g and 1-2 em' are common. Accelerometers ofthis size are necessary when recording from small body parts, such as the finger and when multiple accelerometers are used. Durability is specified in terms of the maximum acceleration that the device can experience without damage. Accelerometers are frequently constructed with mechanical stops that prevent damage by excessive sudden acceleration (i.e. shock), as may occur if the accelerometer is dropped or struck against a hard object. Sufficiently durable accelerometers can withstand shock accelerations of at least ±2000 G. Shock accelerations of this magnitude may be achieved when an accelerometer is dropped onto a hard surface, so these devices must be handled with care. The frequency range (i.e. frequency response) of most accelerometers is flat from approximately 0 Hz to 500 Hz or greater. The upper limit of the frequency range should be more than 4 times the highest frequency of movement that will be encountered. Most accelerometers easily satisfy this requirement for human applications, in which the frequency content of motion is 0-30 Hz. Accelerometers usually contain some form of damping to

182

R.J. ELBLE

Table I Accelerometers and gyroscopes for human motion analysis. Manufacturer

Internet web site

Entran Devices, Inc.

www.entran.com

Piezoelectric

Kistler Instrument Corp.

www.kistler.com

Piezoresistive

Endevco U.K., Ltd.

www.endevco.co.uk

Accelerometers Type

Piezoresistive

Piezoelectric Capacitance Capacitance

Analog Devices

www.analog.com

Capacitance

Silicon Designs, Inc.

www.silicondesigns.com

Piezoelectric

Cambridge Neurotechnology

www.camntech.co.uk

Piezoelectric

1M Systems

www.imsystems.net

Motus Bioengineering, Inc.

www.motusbioengineering.com

Activity monitors (accelerometers)

Gyroscopes

Gyroscope

prevent resonant oscillation at the accelerometer's natural frequency, which resides just beyond the upper limit of the frequency range. The frequency response of piezoresistive and capacitance accelerometers extends down to 0 Hz, making them sensitive to the static acceleration of gravity and providing a useful means of static calibration. The low-frequency limit of piezoelectric devices is 0.1 Hz or higher. The sensitivity (mV/G) of an accelerometer is inversely proportional to its amplitude range. Therefore, accelerometers with excessively large ranges should be avoided. The amplitude range should have a linear sensitivity ± 1%. An amplitude range of ± lOG is suitable for most human applications, and a typical sensitivity for this range is 5-10 mV/G (Verplaetse, 1996). An amplitude range of ±20 G may be needed for accelerometers mounted on the feet or ankles during walking and running (Bouten et al., 1997). The transverse axis (cross) sensitivity is the degree to which an accelerometer erroneously detects Decelerations perpendicular to the axis of sensitivity. The transverse sensitivity should be 3% or less. The resolution of an accelerometer is the lowest acceleration that can be measured. Resolution is

determined mainly by the level of transducer noise and is 0.02 G or less for most accelerometers used in human applications. Intrinsic sources of accelerometer error and noise are: (1) electronic device noise (e.g. due to fluctuations in the power source); (2) transverse axis sensitivity; and (3) thermal drift in de response (piezoresistive devices). Ambient sources of noise are: (1) electrical interference (e.g. 60 Hz noise); (2) ambient vibrations (e.g. as when riding a car or bike); (3) inadvertent bumping or jarring with another physical object (e.g. striking the accelerometer against a table or door frame); and (4) loose or faulty attachment of the accelerometer to a body part, resulting in extraneous mechanical resonance (Bouten et al., 1997). These sources of noise have a cumulative effect when acceleration is numerically integrated over long periods of time to obtain velocity and position estimates, and this problem is a major impediment to the use of accelerometers in motion analysis of complex movements such as walking and reaching.

13.3. Signal conditioning Most accelerometers require a power source, and the power source should be stable and free of noise

ACCELEROMETRY

in order to avoid power fluctuations that cause measurement error. The output of the accelerometer is filtered and amplified before being sampled into a computer with an analog-digital converter. Highpass filters and AC-coupled amplifiers do not eliminate the sensitivity of accelerometers to gravity but are useful when the DC component of acceleration is of no interest. Low-pass filters are needed to attenuate noise at frequencies greater than the maximum frequency of biological interest. The sampling frequency of the analog-digital converter should be greater than twice the cutoff frequency of the low-pass filter, in order to avoid aliasing. 13.4. Measuring motion in 3-dimensional space

The output of accelerometers must be integrated once to obtain velocity and twice to obtain position. For a sinusoidal displacement of amplitude A (onehalf peak-to-peak amplitude), the velocity and acceleration are the first and second derivatives of displacement: displacement=A sin (wt) velocity=Aw cos(wt) accelerationw-Ass' sin(wt)

where w is the frequency of oscillation in radians per second (l cycle/s= 1 Hz=2'lT radians/s) and t is time. Thus, for two oscillations (e.g. tremors) of identical displacement amplitude but different frequencies, the oscillation with the higher frequency will have a larger velocity and acceleration. A corollary to this rule is that high-frequency low-amplitude noise will be more evident in accelerometric measurements than in position measurements, as obtained with a photogrammetric motion analysis system. Another corollary is that for a sinusoidal movement like tremor, displacement and velocity can be estimated with accelerometry by dividing the measured acceleration by w2 and to, respectively. Motion of a limb or other body part rarely occurs in a single direction. In general, a body part may exhibit translational motion in any of three orthogonal (X, Y, Z) directions, and it may rotate about any of these axes. An accelerometer will record translational and rotational inertial accelerations to the extent that these acceleration vectors are in line with the accelerometer's axis of sensitivity. In addition, gravitational acceleration is recorded to the extent

183

that the axis of sensitivity is in the vertical direction of earth's gravity. Gravitational acceleration may contribute significantly to the total acceleration detected by the axis of an accelerometer and thereby limits the ability of accelerometry to reflect translational and rotational motion (i.e. inertial acceleration) of a body part. The task of separating the gravitational and inertial components of acceleration is impossible unless multiple accelerometers are used, and even with multiple accelerometers, measurement error may preclude the accurate separation of gravitational and inertial accelerations. Consider the situation in Fig. 1 in which a rodshaped body part (e.g. the index finger) rotates vertically about a fixed axis (e.g. the metacarpophalangeal joint), such that there is no translational motion. A biaxial accelerometer is attached to the body part at distance R from the axis of rotation. When the body part is perfectly horizontal, gravity is parallel to the t-axis and perpendicular to the r-axis, and the effect of gravity is reflected only in the t-axis of the accelerometer. However, the influence of gravity on the t-axis decreases as the body part rotates up and down, and the influence of gravity on the r-axis increases. The gravitational components in the t and r axes are G cos and G sin <1>. Therefore, for fluctuations in of ± 10 degrees (20 degrees peak to peak), the gravitational component of t-axis acceleration will fluctuate between 1.0 G and 0.985 G, and the gravitational component of r-axis acceleration will be ±0.174 G. Thus, gravitational fluctuations in the r-axis of the accelerometer are substantially greater than in the t-axis, and the influence of gravity on the t-axis is nearly constant (i.e. varies less than 0.015 G). The differences in t-axis and r-axis gravitational fluctuations are significant in situations such as the one depicted in Fig. 1 because the inertial acceleration in the t-axis is much greater than the inertial acceleration in the r-axis. For example, a 6-Hz tremor producing vertical accelerometer movements of ± 1.22 em (0.0244 m peak-to-peak) produces joint rotations of ± sin:' ( 1.2217) = ±0.175 radians = ± 10 degrees, assuming the accelerometer is mounted 7 ern from the joint (i.e, R=0.07 m in Fig. 1). For sinusoidal joint rotations of this magnitude, the t-axis inertial acceleration is R times the angular acceleration, which equals ±0.07· 0.175(2'lT6)2/ 9.807=±1.77 G. These fluctuations in inertial acceleration are much larger than the 0.015 G

184

fluctuations in t-axis gravitational acceleration, so the output of the t-axis of the accelerometer will reflect the fluctuations in inertial acceleration plus the nearly constant effect of gravity, which is easily removed with a highpass filter or by numerically subtracting 1.0 G. By contrast, the r-axis inertial acceleration would fluctuate between zero and R times the angular velocity squared, which equals 0.07(0.175· 2'lT6)2/9.807 =0.311 G. These fluctuations in inertial acceleration are comparable to the ±0.174 G fluctuations r-axis gravitational acceleration, so the output of the r-axis of the accelerometer will reflect significant contributions from inertial acceleration and gravity. The gravitational component will tend to obscure the inertial component, which is the measure of body motion.

RJ. ELBLE

Angular acceleration (a=d 24>/dr) cannot be computed from a, and a, in Fig. 1 because 4> is not known, and hence the contributions of gravity to a, and a, cannot be computed. However, angular acceleration can be computed if a second accelerometer is mounted between the joint and the first accelerometer. Both accelerometers will have the same gravitational influence if they are mounted with the same orientation (i.e. their t-axes are parallel; Fig. 2), and the equations for angular acceleration are R,a=a1t+G cos(4)(t)) and R2a= a 2t+G cos(4)(t)), where the subscripts 1 and 2 refer to accelerometers 1 and 2. Subtracting these two equations gives the following equation for angular acceleration: a=(a't - a2t)/(R, - R2). This approach can be extended to 3-dimensional space using at least six uniaxial accelerometers that are strategically positioned on the body segment (Padgaonkar et al., 1975). The computed angular acceleration can be integrated to obtain angular velocity and rotation (4)), and having computed 4>, the r-axis and t-axis gravitational accelerations could be computed at any time t. Note that this approach assumes the precise strategic alignment of the pairs of accelerometers, and malalignment will produce error in computing a. This error and any noise in a, will have a cumulative effect when a is integrated to obtain 4>. Consequently, accelerometers are not suitable for measuring absolute translational position and rotation in space over extended periods of time because

a, = Rd 2c'P/dt2-Gcosc'P(t) For c'P(t) = c'P o sin(2JZji),

at = -R[ c'P o(2JZff sin(2JZ"fi)]-Gcos[c'Po sin(2JZ"fi)]

a, =-R(dc'P/dt)2-Gsin(c'P(t» For c'P(t) = c'P o sin(2JZji),

a, = -R[c'P o2JZfcos(2JZ"fi)f - Gsin(c'Po sin(2JZji» =-O.5R(c'Po2JZf)2[cos(4trft) + 1]-Gsin(c'Po sin(2trft» Fig. 1. Schematic diagram of a rigid body rotating about a fixed axis. A biaxial accelerometer (shaded box) is attached to the rigid body at a distance R from the axis of rotation. The X-Y reference axes are fixed to the axis of rotation. The t and r axes are the accelerometer axes of sensitivity. The equations for acceleration in the t and r axes (a, and a,) are given for a sinusoidally varying angle of rotation <1>, with amplitude c'P o and frequency f (Hz). G is the acceleration of gravity.

Fig. 2. Same as Fig. I but for two biaxial accelerometers, mounted in parallel for the measurement of angular acceleration a=d 2c'P/df

185

ACCELEROMETRY

accelerometer error and noise accumulate in proportion to f. Commercially available electromagnetic and photogrammetric motion analysis systems are better suited for measuring position and rotation (Ladin et aI., 1989). Accelerometers are best suited for measuring relative motions (e.g. tremor and other involuntary movements) and short-duration changes in position and rotation. In Fig. 1, the frequency of oscillation in the inertial component of a,. is twice the frequency of sinusoidal oscillation in because of the trigonometric relationship cos 2wt= 2 cos' wt - 1. Similarly, the oscillation in the t-axis gravitational component occurs at 2w because cos(. Thus, rhythmic oscillation in will produce t-axis and r-axis acceleration signals with a mixture of oscillations at frequencies wand 2w. The 2w oscillations in the contribution of gravity to the t-axis acceleration will be negligible for angular rotations of ± 10 degrees or less, but these 2w oscillations in gravitational acceleration will assume an increasing proportion of the t-axis acceleration for larger angular rotations. For the r-axis acceleration, the 2w oscillations in centrifugal acceleration will be comparable to the w oscillations in gravitational acceleration for angular rotations of approximately ± 10 degrees, and these 2w oscillations assume an increasing proportion of the r-axis acceleration as the angular rotations exceed ± 10 degrees. Therefore, 2w oscillations in taxis and r-axis accelerations become more apparent as the amplitude of oscillation increases, and this mixture of harmonic oscillations could be misinterpreted as neurogenic when they are, in fact, purely mechanical phenomena. The resultant acceleration (=y'(a~+a~+a~)) of a single triaxial accelerometer is not a complete measure of motion in space, and the error will increase in proportion to the amount of rotational motion. At least six one-dimensional linear accelerometers must be strategically mounted on a body part to record the translational and rotational movement about three orthogonal axes (X, Y, Z), and in most instances, one triaxial and three biaxial accelerometers are needed (Padgaonkar et aI., 1975). Alternatively, accelerometers can be used in combination with gyroscopic transducers to completely record the translational and rotational motion of a body part. Gyroscopic transducers measure angular velocity, from which angular rotation and accelera-

tion can be computed by numerical integration and differentiation, respectively (Verplaetse, 1996). Gyroscopic sensors are not sensitive to gravity. Their main disadvantages are their relatively large weight and bulk.

13.5. Applications Resolving complex translational and rotational motion of a single body segment (e.g. the thigh) with 6 or more uniaxial accelerometers is a difficult task, and capturing complex motion of multiple body segments, as in walking and throwing, requires too many accelerometers and complicated mathematical algorithms to be practical. Consequently, accelerometers are used far less than photogrammetric systems in the motion analysis of walking and other complex movements involving multiple body segments. Accelerometers are occasionally used in combination with photogrammetric methods when there is a need to measure as accurately as possible the acceleration of a particular site on the body. Accelerometers are also used in isolation when only a rough or representative measure of body motion is needed during a particular motor act (Bussmann et aI., 2000; Sekine et aI., 2000). Even then, the user must remain mindful of the limitations of accelerometry discussed in Section 13.4 of this chapter. Accelerometers have been used extensively in the study of tremor and are ideally suited for this purpose. Tremor is easily distinguished from most other movements because tremor is a rhythmic movement at a particular frequency. This is particularly true of pathologic tremors, which produce a narrow (i.e. finely tuned) Fourier spectral peak at the tremor frequency. These finely tuned spectra are typical of tremor and are not produced by other forms of voluntary or involuntary movement, except for rhythmic voluntary movements, such as chewing and cursive writing (Hoff et aI., 2001). Most voluntary and involuntary movements are not rhythmic, and their frequency content is primarily 0-3 Hz (Manson et aI., 2000; Hoff et aI., 2001). Very abrupt transient movements (e.g. myoclonus and chorea) produce Fourier spectra with broader frequency content, ranging from a to more than 30 Hz. Accelerometry and Fourier analysis cannot distinguish dyskinesias from voluntary movements but are useful in the quantification of dyskinesias when patients are relatively sedentary

186

RJ. ELBLE

Coherence Spectrum

EMG

Rectification Low-pass Filtering

Phase Spectrum

Fig. 3. Flow diagram summarizing the steps in performing auto- and cross-spectral analyses of two signals such as accelerometry and EMG. Filtering and rectification can be performed with analog devices before digitization with an analog-to-digital (AID) converter, as shown in this figure, or these procedures could be performed with computer software after digitization of the acceleration and EMG signals. Rectification of the EMG is performed before low-pass filtering to produce the envelope of amplitude fluctuations in EMG over time. The cut-off frequency of the low-pass filter is typically set to 40 Hz or less, depending upon the frequency range of biological interest. The fluctuations in rectified-filtered EMG are proportional to muscle force (Journee et al., 1983). Signals such as acceleration, EEG and neuronal spike trains do not require rectification (Elble and Randall, 1976).

(e.g. seated in a chair) (Manson et al., 2000; Hoff et al., 2001). The optimum sites for mounting the accelerometer(s) have not been determined. Most commercially available activity monitors contain one or more piezoresistive or piezoelectric accelerometers (Table 1). Digitized data are stored within these devices and later downloaded to a personal computer for analysis. They are often worn on the wrist but may be worn elsewhere. Acceleration can be recorded continuously, or accelerations above a specified threshold can be summed over time to give a measure of activity. These "actigraph" devices have been used extensively in sleep analysis and assessment of daytime activity levels. Polysomnography is a more accurate method of sleep assessment because actigraphs tend to overestimate sleep when people lie awake and motionless in bed and underestimate sleep when there is increased motor activity during sleep (Jean-Louis et al., 2000; Verbeek et al., 2001). Nevertheless, actigraphy has been used successfully in the quantification of periodic leg movements during sleep because the pseudoperiodicity of these movements distinguishes them from other motor activity when the patient is in bed (Kazenwadel et al., 1995).

13.6. Frequency (spectral) analysis of accelerometry and EMG 13.6.1. Spectral analysis

Tremor consists of one or more oscillations that are superimposed on a background of broad-frequency activity, which is often regarded as noise.

Consequently, tremor and other rhythmic signals (e.g. EEG) are ideally suited for analysis in the frequency domain, using spectral techniques. A detailed description of spectral analysis is beyond the scope of this chapter. There is an extensive literature on this subject, and the principles and mathematical methods have been reviewed in the context of tremor (Elble and Koller, 1990; Timmer et al., 1996). Spectral methods have been applied extensively to the analysis of tremor, EMG, EEG and neuronal spike trains recorded from muscle (Elble and Randall, 1976; Halliday et al., 1999), peripheral nerve, and central nervous system (Elble et al., 1984; Hua et al., 1998; Zirh et al., 1998; Lemstra et al., 1999). Here, an overview of these methods is provided, and the steps in performing auto- and cross-spectral analyses of two signals such as EMG and accelerometry are summarized in Fig. 3. Spectral analysis of a signal is a mathematical process that decomposes the digitized signal into its frequency components. The most popular method of spectral analysis is Fourier transformation, which produces series of sine and cosine waves that in sum provide an approximation of the recorded signal. The frequencies of the sine and cosine waves reflect the frequency content of the signal, and the amplitudes of the sine and cosine waves reflect the amplitude of the signal at each frequency. The sum of the squared sine and cosine amplitudes at each frequency plotted vs. frequency is called an autospectrum (power spectrum). For a signal x(t) that is digitized to produce a continuous time series x( i) (i= 1,2,3, ... ,N), the Fourier transform X(w p ) is

187

ACCELEROMETRY

given by the following three equations, where p=O, 1,2, ... , N/2): X(wp)=Ax(w p) - jBx(wp)

2

Ai wp ) = ( N) ~N

Blwp) =

2

(

N ~

N

)

27Tpi

IV

xU) cos ( )

27Tpi

IV

xU) sin ( )

The autospectrum (power spectrum) is A;( wp ) + B .~( wp ) .

Gl wp ) =

The coefficient of the sine and cosine components varies among authors and programmers. Therefore, the user of any program should "calibrate" the program by computing the autospectrum of an artificial sinusoidal signal of known amplitude and frequency. The coefficient can then be adjusted so that the autospectrum yields the mean squared, baseline-to-peak squared, or peak-to-peak squared amplitude vs. frequency. A continuous time series is the realization of a continuous signal (e.g. accelerometry, rectifiedfiltered EMG, EEG) observed and sampled in time, at repeated intervals of Llt. The sampling frequency 1/Llt must be greater than twice the highest frequency contained in the signal. Violation of this requirement is called undersampling and will cause aliasing. Aliasing is a phenomenon in which frequencies greater than the Nyquist folding frequency, /N= 1/(2Llt), appear to be frequencies less than/N0 In other words, for any frequency / in the range 0~f5fN' the frequencies (2nfN±f), n= 1,2, ... are indistinguishable from f Thus, if the sampling frequency is 70 samples per second lfN=35 Hz), 60-Hz and 80-Hz noise in digitized recording will be mistaken for lO-Hz oscillations. The truncated ends of a data segment produce artificial side lobes or peaks in the autospectrum, surrounding the true or biological spectral peak. To reduce this spectral leakage, the ends of a data segment can be tapered to zero by multiplying the segment by an appropriate "window" of the same length. Several data windows are available, each with unique merits and limitations (Harris, 1978). The Hamming and Hanning cosine windows are commonly used in tremor and EEG analyses. These windows suppress the side lobes (leakage) but have

the adverse effect of broadening the true spectral peak. This is a problem only when one is trying to resolve two closely-spaced frequency components. Prior to windowing, the mean and any linear trend must be removed from the data segment. Neuronal spike trains are examples of a stochastic point process. A stochastic point process is a series of N identical events, occurring at times t; where i= 1,2, ... N. Only the times of occurrence of the action potentials are of interest. Therefore, the spike train is represented by a series of impulses (Dirac delta functions) at i, For a neuronal spike train y(t) of total duration T, the Fourier transform Y(wp ) is given by the following three equations, where p=O, 1,2, ... ,N12 and W p =27Tp/T: Y(wp)=A/w p) - jB,{wp)

The autospectrum (power spectrum) is G/wp ) = A;(wp)+B;(wp) (Bartlett, 1963; Lewis, 1970). For a

totally aperiodic spike train with Poisson distributed interspike intervals, the autospectrum is statistically flat, with a value of 2 at all frequencies (Bartlett, 1963; Lewis, 1970). Here again, the coefficient of the sine and cosine FFT components varies among authors and programmers, and the user of any program is advised to "calibrate" the program by computing the autospectrum of an artificial spike train with various values of N and T. This approach to computing the auto spectrum of a spike train entails two special problems: the fast Fourier transform algorithm cannot be used, and there is spectral leakage from the non-zero mean of the point process (French and Holden, 1971). Consequently, alternative approaches in which the point process is converted to a continuous time series by low-pass filtering have been developed (French and Holden, 1971; Lago and Jones, 1982). G(wp ) is merely a statistical estimate of the true spectral power at each frequency wp- G(w p ) is proportional to a chi square random variable with 2 degrees of freedom. To reduce the variance of the spectral estimates, the time series is divided into L

188

non-overlapping segments, and the spectra of these segments are averaged. This reduces the variance of the spectral estimates by a factor of L, and G( wp ) becomes a chi square random variable with 2L degrees of freedom. This method of averaging is limited by the length of the time series and by the required frequency resolution of the final spectrum. For the entire time series, resolution of the spectrum is liT or lI(Nt1t), where Ilt is the sampling interval used in digitization. For a spectrum that is the average of L segments of the total time series, the frequency resolution is reduced by a factor of L. This can be a problem when dealing with short time series. An alternative method of reducing the variance of spectral estimates is to smooth the spectrum with a smoothing window, such as a moving average. The width of the window must be large enough to produce adequate statistical reliability but not so large that the spectral peaks are excessively attenuated. Timmer and colleagues described an approach in which the width of the smoothing window is adjusted to a preliminary estimate of the bandwidth of the spectral peak (Timmer et al., 1996; Lauk et al., 1999). Regardless of the approach used, there is obviously an unavoidable element of subjectivity in spectral estimation. The problem of leakage is increased when short time series are analyzed. The shorter the data segment, the greater the effect of the truncated ends of the segment. Therefore, an autoregressive (AR) method of spectral analysis has been developed for short data segments. This method first fits the time series to an autoregressive equation of order k, and the spectrum is computed from this equation. This approach is described and contrasted with the fast Fourier transform (FFT) in three recent papers (Cappello et al., 1997; Muthuswamy and Thakor, 1998; Spyers-Ashby et al., 1998). The FFT and AR methods are used with the assumption that the signal is statistically stationary or very slowly varying (Muthuswamy and Thakor, 1998). A stationary signal has first, second and higher order statistical properties (i.e. mean, variance) that do not change with time. The FFT and AR methods of spectral analysis are performed easily with Matlab and the Matlab signal processing toolbox (The MathWorks, Inc., Natick, MA). The FFT method is mathematically efficient and is familiar to most clinical and basic neuro-

RJ. ELBLE

physiologists. It is well suited for the typical tremor, EEG or EMG recording of a minute or so. The AR method is better suited for short recordings of several seconds, as may occur in the recording of intention tremor during a reaching task (Morgan et al., 1975).

13.6.2. Cross-spectral, coherence and phase analyses Given the auto spectra Gx(wp) and Giwp) of two signals x(t) and y(t), the cross-spectrum Gxy(wp) between the two signals is Gxiwp)=X (wp)Y*(wp)=Axy(wp)+jB,i wp)

where Y*(wp) is the complex conjugate of Y(wp). The cross-spectrum is commonly normalized by the product of the two auto spectra to produce the coherence spectrum K~(wp).

K~( w

p)

Gxiwp)G ~(wp) Giwp)Giwp)

The coherence spectrum is the squared linear correlation between two signals (time series) at frequency w p (Benignus, 1969). The coherence spectrum is bounded by values ranging from zero (no linear relationship) to 1 (perfect linear relationship). Note that low coherence (i.e. < 1.0) can be caused by a lack of linear correlation between two signals or by a nonlinear relationship between two signals. The confidence limit for coherence is 1(l - a)I/(L-I), where a is the confidence level (e.g. 0.95) and L is the number of non-overlapping segments for which the auto- and cross-spectra are computed and averaged in the computation of coherence. Coherence values less than this limit are not statistically different from zero. Amjad and coworkers (Amjad et al., 1997) described a method for computing the coherence of data from two or more experimental trials. The phase spectrum Pxy(wp) between x(t) and y(t) is

and the 95% confidence interval for a phase estimate is

189

ACCELEROMETRY

[l( 1

±1.96 - - 2- - 1 )]112 . L K xy(Wp ) The coherence spectrum must be calculated with smoothed auto- and cross-spectral estimates, and the phase spectrum must be calculated with smoothed cross-spectral estimates.

an analog-to-digital converter at a sampling rate of at least twice the highest frequency content of the signal, to avoid frequency aliasing. Acknowledgments

This work was supported by the Spastic Paralysis Research Foundation of Kiwanis International, Illinois-Eastern Iowa District.

13.7. Summary References

Accelerometry is ideal for measuring low-amplitude rapid movements. However, accelerometers are sensitive to earth's gravity, so their rotation in space produces a time-varying gravitational signal that tends to obscure the inertial acceleration of the body part to which the accelerometer is attached. The magnitude of fluctuations in gravitational acceleration relative to changes in inertial acceleration depends upon the orientation of the accelerometer's axis of sensitivity to earth's gravity and to the direction(s) of body motion. This limitation of accelerometry must be considered when using an accelerometer to quantify body motion, which is typically translational and rotational. The task of separating the gravitational and inertial components of acceleration is impossible unless multiple accelerometers are used, and even with multiple accelerometers, measurement error may preclude the accurate separation of gravitational and inertial accelerations. A three-dimensional accelerometer is not sufficient to record three-dimensional movement in space if there is rotation of the body part. At least six one-dimensional linear accelerometers must be strategically mounted on a body part to record the translational and rotational movement about three orthogonal axes (X, Y, Z), and in most instances, one triaxial and three biaxial accelerometers are needed. Spectral analysis is best suited for detecting rhythmic activity in signals such as tremor, EEG, EMG and neuronal spike trains. Coherence and phase analyses are useful for quantifying the linear relationship between rhythmic activity in two signals (e.g. accelerometry and rectified-filtered EMG), recorded simultaneously. These methods can also be applied to neuronal spike trains, which are treated as stochastic point processes. A continuous signal (e.g. accelerometry, EEG, EMG) must be digitized with

Amjad, AM, Halliday, DM, Rosenberg, JR and Conway, BA (1997) An extended difference of coherence test for comparing and combining several independent coherence estimates: theory and application to the study of motor units and physiologic tremor. J. Neurosci. Methods, 73: 69-79. Bartlett, MS (1963) The spectral analysis of point processes. J. Roy. Stat. Soc. B, 25: 264-281. Benignus, VA (1969) Estimation of the coherence spectrum and its confidence intervals using the fast Fourier transform. IEEE Trans. Audio and Electroacoustics, 17: 145-150. Bouten, CV, Koekkoek, KT, Verduin, M, Kodde, R and Janssen, JD (1997) A triaxial accelerometer and portable data processing unit for the assessment of daily physical activity. IEEE Trans. Biomed. Eng., 44: 136-147. Bussmann, JB, Damen, Land Stam, HJ (2000) Analysis and decomposition of signals obtained by thigh-fixed uni-axial accelerometry during normal walking. Med. Bioi. Eng. Comput., 38: 632-8. Cappello, A, Leardini, A, Benedetti, MG, Liguori, R and Bertani, A (1997) Application of stereophotogrammetry to total body three-dimensional analysis of human tremor. IEEE Trans. Rehabil. Eng., 5: 388-393. Elble, RJ and Koller, WC (1990) Tremor. The Johns Hopkins University Press, Baltimore. Elble, RJ and Randall, JE (1976) Motor-unit activity responsible for 8- to 12-Hz component of human physiological finger tremor. J. Neurophysiol., 39: 370383. Elble, RJ, Schieber, MH and Thach, WT, Jr. (1984) Activity of muscle spindles, motor cortex and cerebellar nuclei during action tremor. Brain Res., 323: 330-4. French, AS and Holden, AV (1971) Alias-free sampling of neuronal spike trains. Kybemetic, 8: 165-171. Halliday, DM, Conway, BA, Farmer, SF and Rosenberg, JR (1999) Load-independent contributions from motor-

190 unit synchronization to human physiological tremor. J. Neurophysiol., 82: 664-675. Harris, FJ (1978) On the use of windows for harmonic analysis with the discrete Fourier transform. Proc. IEEE, 66: 51-83. Hoff, Jl, Van den Plas, AA, Wagemans, EA and Van Hilten, JJ (2001) Accelerometric assessment of levodopa-induced dyskinesias in Parkinson's disease. Mov. Disord., 16: 58-61. Hua, SE, Lenz, FA, Zirh, TA, Reich, SG and Dougherty, PM (1998) Thalamic neuronal activity correlated with essential tremor. J. Neurol. Neurosurg. Psychiatry, 64: 273-276. Jean-Louis, G, Mendlowicz, MV, Gillin, JC, Rapaport, MH, Kelsoe, JR, Zizi, F, Landolt, H and Von Gizycki, H (2000) Sleep estimation from wrist activity in patients with major depression. Physiol. Behav., 70: 49-53. Joumee, HL, Van Manen, J and Van der Meer, JJ (1983) Demodulation of e.m.g.s of pathological tremours. Development and testing of a demodulator for clinical use. Med. BioI. Eng. Comput., 21: 172-175. Kazenwadel, J, Pollmacher, T, Trenkwalder, C, Oertel, WH, Kohnen, R, Kunzel, M and Kruger, HP (1995) New actigraphic assessment method for periodic leg movements (PLM). Sleep, 18: 689-697. Ladin, Z, Flowers, WC and Messner, W (1989) A quantitative comparison of a position measurement system and accelerometry. J. Biomechanics, 22: 295308. Lago, PJA and Jones, NB (1982) Note on the spectral analysis of neural spike trains. Med. BioI. Eng. Comput., 20, 44-48. Lauk, M, Timmer, J, LUcking, CH, Honerkamp, J and Deuschl, G (1999) A software for recording and analysis of human tremor. Comput. Meth. Prog. Biomed., 60: 65-77. Lemstra, AW, Verhagen Metman, L, Lee, Jl, Dougherty, PM and Lenz, FA (1999) Tremor-frequency (3-6 Hz) activity in the sensorimotor arm representation of the

R.J. ELBLE internal segment of the globus pallidus in patients with Parkinson's disease. Neurosci. Lett., 267, 129-132. Lewis, PAW (1970) Remarks on the theory, computation and application of the spectral analysis of series of events. J. Sound Vihr., 12: 353-375. Manson, AJ, Brown, P, O'Sullivan, JD, Asselman, P, Buckwell, D and Lees, AJ (2000) An ambulatory dyskinesia monitor. J. Neurol. Neurosurg. Psychiatry, 68: 196-201. Morgan, MH, Hewer, RL and Cooper, R (1975) Intention tremor - a method of measurement. J. Neurol. Neurosurg. Psychiatry, 38: 253-258. Muthuswamy, J and Thakor, NY (1998) Spectral analysis methods for neurological signals. J. Neurosci. Methods, 83: 1-14. Padgaonkar, AJ, Krieger, KW and King, AI (1975) Measurement of angular acceleration of a rigid body using linear accelerometers. J. Appl. Mech. (Trans. ASME), 42: 552-556. Sekine, M, Tamura, T, Togawa, T and Fukui, Y (2000) Classification of waist-acceleration signals in a continuous walking record. Med. Eng. Phys., 22: 285291. Spyers-Ashby, lM, Bain, PG and Roberts, SJ (1998) A comparison of fast fourier transform (FFT) and autoregressive (AR) spectral estimation techniques for the analysis of tremor data. J. Neurosci. Methods, 83: 35-43. Timmer, J, Lauk, M and Deuschl, G (1996) Quantitative analysis of tremor time series. Electroencephalogr. Clin. Neurophysiol., 101: 461-468. Verbeek, I, Klip, EC and Declerck, AC (2001) The use of actigraphy revised: the value for clinical practice in insomnia. Percept. Mot. Skills, 92: 852-6. Verplaetse, C (1996) Inertial proprioceptive devices: selfmotion-sensing toys and tools. IBM Syst. J., 35: 639-650. Zirh, TA, Lenz, FA, Reich, SG and Dougherty, PM (1998) Patterns of bursting occurring in thalamic cells during parkinsonian tremor. Neuroscience, 83: 107-121.

Movement Disorders Handbook of Clinical Neurophysiology, Vol. I M. Hallett (Ed.) © 2003 Elsevier S.v. All rights reserved

191 CHAPTER 14

Kinesiology Christian Dohle* and Hans-Joachim Freund Department ofNeurology, University Hospital Dusseldorf, D-40225 Dusseldorf, Germany

14.1. Overview In motor control studies, recordings of the kinematics and dynamics of movement can be employed at a high level of sophistication. These techniques have been applied for research, for the study of pathophysiology and for diagnostic purposes. In clinical settings, their use is still restricted because they are expensive, time consuming and require skilled technical personnel. Another reason is that better insights into movement kinematics are often not critical for differential diagnosis or therapeutic decisions. The situation is going to change as rehabilitative strategies and procedures become more specific. The previously used qualitative or semiquantitative scores are increasingly complemented by electrophysiological and kinematic recordings. Kinematic techniques will therefore be required for both the study of normal motor behavior and for restorative neurology. Neuroimaging techniques are providing new insights into brain function. They produce high resolution, spatial information about brain structure and functional activation states during a broad range of cognitive and sensorimotor behaviors. In contrast to the highly sophisticated imaging techniques the assessment of behavior is often poorly controlled. For motor control, this is partly due to the restricted mechanical conditions for measurement in the scanners or to interference with the magnetic devices. This situation is just about to change as more recording techniques are available for application in neurofunctional studies. This article gives an outline of the most commonly used kinematic recording techniques and their

* Correspondence to: Dr. C. Dohle, Neurologische Klinik, Universitatsklinikum Dusseldorf, MoorenstraBe 5, D-40225 Dusseldorf, Germany. E-mail address:[email protected] Tel.: +49-211-81 18678; fax: +49-211-81 18485.

methodical backgrounds. The major emphasis is put on the analysis of upper limb movements, as gait analysis is dealt with in a separate chapter.

14.2. Technical principles Depending on the purpose of the movement study, different movement recording systems can be used. One-dimensional systems, such as goniometers or lever constructions can record only one degree of freedom - usually unidirectional limb movements. In contrast, three-dimensional recording systems allow the recording of the entire movement of a limb in three-dimensional space. Both systems have specific advantages. One-dimensional techniques are simpler and cheaper and often serve the desired purpose. Three-dimensional movement recording systems are usually marker-based: the position of (active or passive) markers that are fixed to certain body parts is recorded by detectors. Their position in three-dimensional space has to be calibrated before the measurement. In principle, two different types of systems can be distinguished: (a) Active systems with markers emitting a signal (usually infrared light) in well-defined time intervals which is recorded by CCD cameras (e.g. Optotrak, Selspot (now out of business». (b) Passive systems with markers consisting of colorful or light-reflecting material. Their position is recorded by video- or infrared-based camera systems (e.g. Elite, MacReflex, Vicon). Usually, the recording procedure is based on infrared light. Active systems make use of infrared emitting diodes (LEOs) that are recorded by sensitive cameras. In passive systems the camera is coupled with an array of infrared emitting diodes, and the light is reflected by the markers. The most prominent problem of both systems is noise due to reflections from extraneous shiny surfaces. Other systems can detect markers directly from video recordings with-

192

out any additional technical setup. In principle, these systems are very versatile as they do not require any extensive technical set-up. On the other hand, they require considerable manual post-processing and assignment of the markers at the beginning of each recording session. These systems are also much more viable to optical interferences. A completely different technique is based on ultrasound (ZEBRIS). The markers consist of small ultra-sound emitting sources and the signals are recorded by an array of three microphones fixed on a rack at well-defined distances to each other. Although these markers make use of small piezo elements to generate ultrasound, they barely emit electromagnetic waves, so that they can also be used in environments that are sensitive to electromagnetic 'noise' such as magnetoencephalography or magnetic resonance tomography. On the other hand, the system has only a limited range and is very liable to acoustic noise. Common to all these systems is the limitation that one camera can only detect positions in two dimensions. Thus, at least two cameras are necessary in order to obtain a three-dimensional position. Other systems are available that make use of different technologies. Gained from developments for the market of virtual reality techniques, electromagnetic tracking devices are partly employed for movement analysis (mainly Polhemus, Flock of Birds). In these systems, one emitter generates a static electromagnetic field of a certain dimension. The sensor consists of a coil that indicates position and orientation of the moving part relative to the emitter. In principle, these systems are rather cheap, very accurate and can provide both position and orientation of a marker in real-time. On the other hand, they are very liable to any type of metal within the range of the emitter as well as electromagnetic 'noise', e.g. of nearby located large computer monitors. All systems have specific advantages and disadvantages. Active systems are usually more expensive. For measurements, it is necessary to 'wire' the subjects, i.e. to connect the marker and the control unit. This ensures that every marker is always uniquely defined, even if it is occluded during certain measurement epochs. Passive measurement systems are usually cheaper and the measurement is easier to realize, because they do not require direct connections of the markers

C. DOHLE AND H.-J. FREUND

and the control unit. The major drawback of these systems is the ambiguity of marker identification. Consequently, the analysis of every trial has to be preceded by a manual assignment of markers that are then tracked further by the analysis software. Depending on the type of movement, marker trajectories can cross - at least in the field of view of one of the cameras. In this case, the marker assignment has to be corrected manually. In summary, passive systems are cheaper and easier in the initial acquisition of data. This advantage, however, has to be counterbalanced against a much higher demand in manual post-processing. All measurement systems of this type deliver time series of X-, Y: and z-coordinates as a result, that are sampled at certain frequencies. The further analysis of these data is poorly standardized, so that not many commercial programs are available. Usually, different laboratories make use of different, custom-made software packages. A good (and still valid) overview of the general principles of human movement analysis can be found in Winter (1990). Since this review is mostly restricted to applications in neurophysiology and sports medicine, we will mainly concentrate on the analysis of patients. Such recordings offer two advantages: They can quantify certain aspects of motor behavior and disclose qualitative features that cannot be detected otherwise.

14.3. Coordinate systems As outlined above, the majority of three-dimensional movement measurement systems provide (Cartesian) X-, y- and z-coordinates in time. The point of origin and the orientation of these coordinates is defined by a calibration procedure that has to be performed prior to the measurement. The calibration should be adjusted to match the coordinate axes to the main axes of the movement task to be recorded. Generally, the x- and y-axis define the floor or table plane with the z-axis pointing upwards. Although it is clear that the brain does not use cartesian coordinates for its computations, it is less clear what type of coordinates are used instead. In their pioneering studies on goal-directed movements, Soechting and Flanders (Soechting and Terzuolo, 1988; Soechting and Flanders, 1989a, b; Flanders et al., 1992) analyzed the variability of upper-limb pointing movements directed at targets in

193

KINESIOLOGY

three-dimensional space. From their analysis, they proposed that arm movements are organized in a polar system based at the respective shoulder, having two angles (yaw and elevation) and one length (distance) as coordinate values. This proposal was theoretically based and inferred solely from kinematic measurements. Recent cell recordings in the awake monkey, however, demonstrated different cell populations responding specifically to the variation of one of the three values (Lacquaniti et al., 1995). On the other hand, a study of sensorimotor transformation errors in patients with lesions of the inferior parietal lobule showed dissociations of distance and direction errors, supporting the idea of different central processing modules for these two variables (Darling et al., 2001). 14.4. Levels of observation When subjects are required to write their signature either on a sheet of paper with a fine pen or on a blackboard with a piece of chalk attached to a broomstick, the individual features of their signature remain basically preserved - even when the writing is performed with the foot or a pen between their teeth. This phenomenon is called motor equivalence and neatly demonstrates that the brain is capable to transfer a given movement blueprint into a distinct kinematic pattern, irrespective of the biomechanical requirements. This is accomplished by converting the movement scheme into joint torques and muscle activation patterns specific for the limb that is actually performing the movement (inverse dynamics). For the analysis of human movements, this implies that movement analysis can be performed at different levels of observation: (1) Examination of (one-dimensional) limb angles.

This is particularly useful in the consideration of intra- and inter-limb coupling. When trajectories are recorded in using threedimensional measuring systems, more information is available about the position of the entire limb in space. Thus, additional levels of observation can be analyzed. (2) Examination of trajectories in three-dimensional space (kinematics). (3) Examination of multi-joint angles (limb configuration)

(4) Examination of joint torques (dynamics) Obviously, these considerations mainly apply to whole limb movements, i.e. the analysis of arm movements or gait. As gait analysis is covered in a separate chapter of this volume, the following analysis will focus on upper limb movements. Especially, we will demonstrate how the different levels of observation have been used to determine features of normal and pathological movements in patients with neurological disorders. 14.5. Analysis of patient populations The optimum strategy for the different applications depends whether the kinematic analysis should either elaborate previously defined patient's profiles or to allow the identification of different subpopulations of patients based on the kinematic data. In the first case, the mean values of pre-assigned kinematic parameters are determined for certain, apparently homogenous patient groups. These data can then be referenced either to those of normal subjects, or - if the type of disturbance only affects one side of the body - to the 'unaffected' body side. This procedure can only be applied if the patient groups are well defined and homogenous, such as hemiparkinsonism or hemiparesis. Conversely, kinematic data can be used to identify subgroups of patients with similar kinematic abnormalities and then to examine a possible correlation with distinct lesion localizations. This approach is more demanding, because the identification of the respective kinematic measures that are suitable to differentiate 'affected' from 'not affected' patients requires extensive prior research. 14.6. Temporal kinematics The kinematic analysis of recorded movements is exclusively based upon the trajectory of the recorded body segment, e.g. the hand or the fingertip. As the movement recording is usually performed at sampling frequencies around 100 Hz, it produces a large number of data points. In order to compare the trajectories of one person under different experimental conditions or trajectories of different patient populations it is necessary to decrease the dimensions of the trajectory, i.e. to parameterize the movement. One way to parameterize movements is the determination of fixed temporal 'landmarks'. This approach was extensively used in the analysis of

194

grasping movements by Jeannerod and his coworkers (Jeannerod, 1981, 1984, 1988). He focused on hand velocity as a measure for the (proximal) reach component of the arm and grip aperture as a measure for the (distal) grasp component of the fingers. This work demonstrated the independence of these two components, by studying the times of maximum values of the two measures as representatives of the two components of the prehension movement (for review, see Paulignan and Jeannerod, 1996). The independence of these two components was the focus of many subsequent studies in normal subjects. In a typical paradigm in normal subjects, a determinant of one of the two components was varied (e.g. object size) and its effect on the second component (e.g. time of maximum hand velocity) was studied. Further support for the independent organization of these two components came from patients with focal cerebral lesions, especially of the posterior parietal cortex. Single case reports confirmed the role of the posterior parietal cortex for these types of sensorimotor transformations (Jeannerod, 1986a, b; Pause et al., 1989; Jakobson et al., 1991). In a first systematic study in patients with lesions of the parietal cortex, presenting the clinical picture of optic ataxia, Perenin and Vighetto showed deficits in the transport component of both arms in the left visual field following right parietal lesions ('field effect'). On the other hand, the transport component of patients with optic ataxia due to left parietal lesions was not only found to be impaired in the right visual field, but additionally also with all movements of the right arm in the left visual field and with central fixation ('hand effect'). The grasping component was tested for by adjustment of hand orientation in the frontal plane, following the paradigm of Haaxma and co-workers (Haaxma and Kuypers, 1974). Deficits in hand orientation showed the same distribution as deficits in the transport component, so that in this condition no dissociation between the two components could be demonstrated. It should be noted, however, that hand orientation is a kind of intermediate task that has to be accomplished in a synergy of proximal and distal joint angles (Dohle et al., 1995; Desmurget et aI., 1996). A kinematic study of prehension movements in patients with parietal lesions (Binkofski et al., 1998b) demonstrated an impairment of the grasp task but not of the reach component, supporting the

C. DOHLE AND H.-1. FREUND

concept of two independent visuomotor channels. This finding is of special interest as it corresponds to experimental inactivation studies in monkeys where the transport and the grasp component could be impaired independently (Gallese et aI., 1994). In an attempt to identify measures for disturbances of apraxic arm movements, Poizner and co-workers applied the approach of comparing temporal landmarks on repetitive movements. In the analysis of the times of maximum hand velocity and minimum hand path curvature during each stroke of a cyclical arm movement, the decoupling of these two measures was proposed to indicate a specific deficit in apraxic movements (Fig. 1, Poizner et al., 1990). Such interpretations have to be met with caution by two reasons. First, the underlying mathematics is not clear. Based on basic differential geometry (Morasso, 1983; Bartsch, 1986), the absolute value of hand path curvature c(t) at a given point of the three-dimensional trajectory r(t) (with its temporal derivatives velocity r'(t)=v(t) and acceleration r'(t)=a(t)) is calculated as: c(t)

=1(r' (r) x r'(t))/I r' (t) 131 =1(v(t) x a(t))/I v(t) 131

In other words, the value of curvature is calculated as a derivative of hand velocity. This fact on its own makes it highly unlikely that the two measures can be varied by the central commands independently. Second, Poizner's analysis was based mainly on the consideration of maximums and minimums of the time course of velocity and curvature. In normal subjects' movements, no significant time delay was observed between the timing of these extreme values. In apraxic patients, however, their occurrence did not correlate at all. Inspection of the raw data clearly shows that the movements were fragmented and consisted of several movement segments. It is clear, that this type of noisy data produce more extreme values that do not match each other. This leaves the reader wondering what is the phenomenom of decoupling of tangential velocity and hand path curvature. 14.7. Temporal aspects of movement organization

Quantitative measurements can provide insights in new principles of organization. An example are recordings from hand and finger movements that

195

KINESIOLOGY

Fig. I. Velocity - curvature coupling during carving movements in an apraxic patient (right) in comparison to a normal subject (left). Upper panels: tangential velocity (mis, broken lines) and radius of curvature (m, continuous lines). Lower panels: time correspondences between velocity minima and radius of curvature minima (from Poizner et aI., 1990).

provided evidence for a clear cut functional dichotomy of manipulative hand and finger movements. On a qualitative basis, movements of the hand as a whole used for writing, shading and hammering had been previously classified as extrinsic and movements of the fingers against each other or the palm as intrinsic hand movements. Quantitatively, it has been demonstrated that these two types of hand movements are performed at distinctly different preferred frequencies. Extrinsic hand movement are usually performed at 4-7 Hz (type I) and intrinsic hand movements at 1-1.5 Hz (type II; Kunesch et aI., 1989). Surprisingly, there is no overlap between the two ranges (Fig. 2). Since the difference in their characteristic frequencies cannot be explained by mechanical reasons it has been concluded (Kunesch et aI., 1989) that the grouping into distinct frequency ranges represents different types of sensorimotor interactions. The slower type I movements require a continuous sensory guidance for their execution and must be carried out slowly in order to meet the time requirements for the detailed sensory analysis. The faster type II movements are rapid automated subroutines that are largely predictive and require only some crude sensory monitoring.

The observed grouping at preferred frequencies fits well into the more general scheme that most of our repetitive motor behaviors such as speaking, eyemovements during reading, chewing, walking, swimming or dancing is performed at characteristic frequencies. Central pattern generators in the brainstem play an important role for their generation. Therefore, the assessment of the temporal aspects of motor organization is only possible on the basis of quantitative movement analysis.

14.8. Spatial kinematics Apart from determining the timing of certain landmarks during the movement, the spatial position of some body part at the time is of interest as well. For prehension movements, this approach was first proposed by Haggard and co-workers (Haggard et aI., 1994). In their study, it could be demonstrated that the aperture during a prehension movement was not only scaled on the basis of the temporal pattern of the prehension movement, but also depending on the spatial position of the hand in the entire trajectory. In a patient with cerebellar damage, the strategic trial-to-trial coordination seemed to be

196

Fig. 2. Distribution of the frequencies at which finger movements are performed during four tactile exploratory tasks (left) and of hand movements during execution of three normal skills (right) (from Kunesch et aI., 1989).

preserved in spite of gross abnormalities of the arm trajectory (Haggard et al., 1994). Another spatial feature of arm trajectories is preservation of the movement plane, which was presented by Poizner and colleagues as a typical feature of movements in apraxic patients (Poizner et al., 1990). In their study, it could be shown that the plane of motion in cyclical arm movements shows a high inter-trial variability when compared to those of normal subjects. These findings convincingly characterize the deranged movements. However, similar deficits were later shown to occur in deafferented patients (Sainburg et al., 1993), raising the question of specificity of this finding. 14.9. Limb and body configuration

The analysis methods presented so far were only focused on the hand and the finger. However, the three-dimensional recording systems also offer the possibility to catch the movement of the entire limb,

C. DOHLE AND H.-I. FREUND

namely the upper arm and forearm. By adding two further markers at the shoulder and at the elbow, three further limb angles can be calculated: upper arm elevation, upper arm yaw and elbow angle (Soechting and Lacquaniti, 1981). This method does not take into account forearm pronation or supination, which can simply be regarded as a radial segment rotating around an ulnar segment (Reich and Daunicht, 2000). Its calculation out of recorded kinematic data requires a further marker on the forearm defining the transversal axis of the hand joint in space. Soechting and Lacquaniti's pilot studies in normal subjects demonstrated that upper arm elevation (usually referred to as shoulder elevation) and elbow angle are tightly coupled during a pointing (or grasping) task (Soechting and Lacquaniti, 1981). Subsequently, this relationship was found to be disturbed in a variety of neurological disorders, such as deafferentation (Sainburg et al., 1993; Ghez and Sainburg, 1995) (Fig. 3), apraxia (Poizner et al., 1995), in patients with parietal lesions with and without apraxia (Binkofski et al., 1998a), cerebellar ataxia (Bastian et al., 1996; Massaquoi and Hallett, 1996), hemiparesis (Levin, 1996; Beer et al., 2000; Cirstea and Levin, 2000) or Parkinson's disease (Seidler et al., 2001). The authors of these studies noted the similarities of their findings, but it remains unclear if the loss of interjoint coupling is a specific pathological phenomenon, or rather some epiphenomenon that is to be found in different types of deranged arm movements. 14.10. Dynamics

When both information about the joint angles and estimates about the inertial mass of the body segments (e.g. Winter, 1990) are available, inverse dynamics equations can be applied to calculate the joint torques that are necessary to produce the observed movements. Different authors have used different terminologies (Soechting and Lacquaniti, 1981; Hollerbach and Flash, 1982; Bastian et al., 1996), but in general, one has to distinguish (according to the terminology used by Bastian and co-workers): • Net torque: sum of all the torques acting at a joint.

197

KINESIOLOGY

Control 2

Fig. 3. Coupling of elbow and shoulder motions in two patients with deafferentation (middle and bottom) in comparison to a normal subject (top). The elbow flexion! extension angle has been plotted against the upper arm elevation (top) and yaw (bottom) angles for individual cycles of a four cycle movement (from Sainburg et al., 1993).

According to the formulas presented in Soechting and Lacquaniti (1981) and Bastian et al. (1996), this consists from the sum of:

• muscle torques: sum of all muscle and passive tissue forces that are necessary to generate the observed movement; • gravitational torque: torque produced by the force of gravity acting on the limb segments of the arm; • interaction torques: motion-dependent torques that occur at any joint due to movements that are linked. In normal subjects, it could be demonstrated that phasic EMG activity occurs in muscles solely acting at adjacent joints, even if those joints are mechanically immobilized (Gribble and Ostry, 1999). This was interpreted as experimental support for the theoretical proposal that central control signals to muscles are adjusted, in a predictive manner, to compensate for interaction torques. In only a very limited number of studies have these analysis techniques been applied to patient movements. In studies on patients with deafferentation (Sainburg et al., 1993, 1995), cerebellar ataxia (Bastian et al., 1996) and Parkinson's disease (Seidler et al., 2001) these interaction torques were found to be inadequately adjusted in order to generate smooth, straight movements. It remains unclear, however, if this deficit can be pinpointed to the abnormal control of interaction torques as result of a deficient processing of somatosensory information or whether this disturbance is the pure description of a lack of joint coordination at a different level of observation. Further studies are necessary to clarify this point. In a later study, Bastian and co-workers (Bastian et al., 2000) tried to exclude that this inaccurate adjustment of interaction torques in cerebellar patients was solely due to a general inability to generate sufficient levels of phasic torque. For this purpose, they studied practically identical upper-arm pointing movements in cerebellar patients, either with their shoulder free or their shoulder fixed (Bastian et al., 2000). Cerebellar patients changed the pattern of muscle activity fundamentally when changing between these two conditions, producing inadequately scaled interaction torques, due to deficits in coordination of multi-joint movements. 14.11. Manipulative hand movements

Apart from whole-limb movements of either arm, manipulative movements, such as those during exploration of small objects, can be recorded and

198

C. DOHLE AND H.-I. FREUND

analyzed quantitatively. In this case, movements of the fingertips of the thumb and the forefinger have to be recorded. Although movement paths are restricted to a narrow workspace, the same trajectory in space is never repeated (Kunesch et al., 1989). The temporal profile, however, shows a regular, periodic pattern (Fig. 4). Spectral analysis typically exhibits a single peak in the frequency spectrum between 1 and 2 Hz, showing high variability across, but not within different subjects (Kunesch et al., 1989). This is typically impaired in subjects with lesions of the posterior parietal cortex, showing deficits in the recognition of objects by haptic exploration (astereognosis). In these patients, the space of exploration is grossly enlarged and the mean frequency and the regularity of the movement is reduced (Kunesch et al., 1995; Binkofski et al., 2001). A more detailed analysis of subgroups of 3D-trajectories

these patients show that astereognosis is correlated with grossly enlarged space of exploration, but not with slower or more irregular movements per se (Binkofski et al., 2001).

14.12. Bilateral synergies The last paragraphs lead to another question: To which extent can different limb segments be controlled independently, especially between both sides of the body? The most basic model of coupled limb movements is that of rhythmic flexion and extension movements of the index fingers or hands of both arms, that are moving either in-phase or anti-phase. This paradigm was extensively studied by Kelso and co-workers who noted that in-phase movements are much more stable than anti-phase movements (Kelso, 1981, Movement<:haraeteflslics of thumb andindex finger

Movement characteristics of lhumb and indexfinger

Frequency spectrum

OIlhumb movemems

Frequencyspectrum 01 thumb movements

Fig. 4. Exploratory finger movements in two patients with focal cerebral lesions (middle and bottom) in comparison to a normal subject (top). Left panels: 3-D reconstruction of thumb (T) and index finger (FF) movement trajectories. Middle panels: scan paths of thumb and index finger. Right panels: frequency distribution of the thumb movements (from Binkofski et aI.,

2001).

199

KINESIOLOGY

1984; Kelso and Tuller, 1984). Furthermore, if subjects perform externally paced anti-phase movements with increasing movement frequency, a sudden transition to in-phase movements occurs. These observations led to the Haken-Kelso-Bunz model, explaining the superior stability of the inphase pattern by a theoretical model of coupled oscillators (Haken et al., 1985). Although this model has been extensively studied and refined since then (Kay et al., 1987; Forrester and Whitall, 2000; Beek et al., 2002), it is only recently that the perceptual component of this phenomenon has come into the focus of attention. For example, it could be shown that the same stability of the in-phase pattern when compared to the anti-phase holds true when two people have to

synchronize their movements (Bingham et al., 1999). Finally, in a recent study Mechsner and coworkers studied bimanual finger movements when one hand was either pronated or supinated (Mechsner et al., 2001). Surprisingly, in this setup, the symmetrical movements of the fingers were also more stable than parallel movements when one of the two hands was pronated and the other supinated. In other words, the superior stability of the in-phase pattern only depended on the actual movement direction of the fingers, but not on the finger muscles involved. In a second experiment, the subjects had to synchronize two circular-moving handles. In one handle, however, the ratio of transmission was altered in such a way that three rotations of the hand resulted in four rotations of the handle (Mechsner et

B

elbow

1-------4

_1oomm

18

C

,. \/V\NVVVWVV 1= 2z

v\[\MMM!VV I~

wrist

1-------4

D

1= 1z

18

ISOmm

E

F

I~

I:n

I:n 1-------4

1.

~10mm

Fig. 5. Bilateral synchronous movements of shoulder (top), elbow (middle) and wrist (bottom) in a patient with right sided hemispherectomy. Left panels: movement cycles (right side: upper trace). Right panels: bilateral correlation (from Miiller et aI., 1991).

200 al., 2001). In this set-up, stability of in-phase circular movements was achieved flawlessly, further supporting the idea that it is the perception of in-phase pattern that is controlled rather than the implementation in the subjects' biomechanical system. All these experiments were performed with movements of distal parts of the arm (hand or finger movements) and synchrony was usually assessed by comparison of the inter-tap-times of each hand. In this analysis procedure, the actual movement trajectory is not considered, but only the final 'goal' (Wiesendanger et al., 1994). Analysis of phase diagrams comparing right- and left sided movements and calculating correlation coefficients between the entire trajectory of either side might help to understand the mechanisms of bilateral coordination in more detail. In a study on bilateral prehension movements using this approach, Dohle and coworkers could demonstrate different coupling mechanisms for the reach and the grasp component respectively (Dohle et al., 2000). Whereas the grasp (thus distal) component shows only little coordination during the movement, suggesting largely independent movement organization mechanisms of either side, the reach (thus proximal) component seems to be 'hard-wired' connected between both sides. This is compatible with a study in a patient with right-sided hemispherectomy, showing preserved coordination between proximal arm movements of either side, but largely independent, fractionated movements of more distal parts of the body (Muller et al., 1991) (Fig. 5).

14.14. Summary In this chapter, we have presented an overview about the different techniques and methods in kinesiology, especially with respect to their application in the movement analysis of neurological patients. A number of different disturbance patterns provide useful information about the characteristic changes in a particular pathological condition. Some of them are specific, while others reflect more general abnormalities.

References Bartsch, H-I (1986) Mathematische Formeln, 21. Auflage. Leipzig:VEB Fachbuchverlag. Bastian, AJ, Martin, TA, Keating, JG and Thach, WT (1996) Cerebellar Ataxia: Abnormal control of inter-

C. DOHLEAND H.-J. FREUND

action torques across multiple joints. J. Neurophysiol., 76: 492-509. Bastian, AJ, Zackowski, KM and Thach, WT (2000) Cerebellar Ataxia: Torque deficiency or torque mismatch between joints? J. Neurophysiol. 83: 30193030. Beek, PI, Peper,CE and Daffertshofer, A (2002) Modeling rhythmic interlimb coordination: beyond the Hakenkelso-Bunz model. Brain Cogn., 48: 149-165. Beer, RF, Dewald,IPA and Rymer, WZ (2000) Deficits in the coordination of multijoint arm movements in patients with hemiparesis: evidence for disturbed control of limb dynamics. Exp. Brain Res., 131: 305319. Bingham, GP, Schmidt, RC and Zaal, FT (1999) Visual perception of the relative phasing of human limb movements. Percept. Psychophys., 61: 246--258. Binkofski, F, Dohle, C, Hefter, H, Schmitt, M, Kuhlen, T, Seitz, R and Freund, H-I (1998a) Deficits in ThreeDimensional Limb Coordination in Parietal Patients With and WithoutApraxia. In: M Fetter,T Haslwanter, H Misslisch and D Tweed (Eds.), 3-D Kinematic Principles of Eye, Head and Limb Movements in Health and Disease. Amsterdam: Harwood Publishers.

Binkofski,F, Dohle, C, Posse, S, Stephan, KM, Hefter, H, Seitz, RJ and Freund, HI (1998b) Human anterior intraparietal area subserves prehension: a combined lesion and functional MRI activation study. Neurology, 50: 1253-1259. Binkofski,F, Kunesch, E, Classen, J, Seitz, RJ and Freund, HI (2001) Tactile apraxia: unimodal apractic disorder of tactile object exploration associated with parietal lobe lesions. Brain, 124: 132-144. Cirstea, MC and Levin, MF (2000) Compensatory strategies for reaching in stroke. Brain, 123: 940--953. Darling, WG, Rizzo, M and Butler, AJ (2001) Disordered sensorimotor transformations for reaching following posterior cortical lesions. Neuropsychologia, 39: 237254. Desmurget, M, Prablanc, C, Arzi, M, Rossetti, Y, Paulignan, Y and Urquizar, C (1996) Integrated control of hand transport and orientation during prehension movements. Exp. Brain Res., 110: 265-278. Dohle, C, Hefter, H, Meermagen, S, Nies, A and Freund, H-J (1995) Disturbances in orienting finger-thumb opposition space in patients with parietal lesions. In: Parietal Lobe Contributions to Orientation in 3-D space. Ttibingen.

Dohle, C, Ostermann, G, Hefter, H and Freund, H-I (2000) Different coupling for the reach and the grasp components in bimanual prehension movements. NeuroReport, 11: 3787-3791.

KINESIOLOGY

Flanders, M, Helms Tillery, SI and Soechting, JF (1992) Early Stages in a sensorimotor transformation. Behav. Brain Sci., 15: 309-362. Forrester, Land Whitall, J (2000) Bimanual finger tapping: effects of frequency and auditory information on timing consistency and coordination. J. Mot. Behav., 32: 176-191. Gallese, V, Murata, A, Kaseda, M, Niki, N and Sakata, H (1994) Deficit of hand preshaping after muscimol injection in monkey parietal cortex. NeuroReport, 5: 1525-1529. Ghez, C and Sainburg, R (1995) Proprioceptive control of interjoint coordination. Can. 1. Physiol. Pharmacol., 73: 273-284. Gribble, PL and Ostry, OJ (1999) Compensation for interaction torques during single- and multijoint limb movements. J. Neurophysiol., 82: 2310-2326. Haaxma, Rand Kuypers, H (1974) Role of occipitofrontal cortico-cortical connections in visual guidance of reatively independent hand and finger movements in rhesus monkey. Brain Res., 71: 361-366. Haggard, P, Jenner, J and Wing, A (1994) Coordination of aimed movements in a case of unilateral cerebellar damage. Neuropsychologia, 32: 827-846. Haken, H, Kelso, JA and Bunz, H (1985) A theoretical model of phase transitions in human hand movements. BioI. Cybem., 51: 347-356. Hollerbach, JM and Flash, T (1982) Dynamic interactions between limb segments during planar movements. Bioi. Cybem., 44: 67-77. Jakobson, LS, Archibald, YM, Carey, DP and Goodale, MA (1991) A kinematic analysis of reaching and grasping movements in a patient recovering from optic ataxia. Neuropsychologia, 29: 803-809. Jeannerod, M (1981) Intersegmental coordination during reaching at natural visual objects. In: J Long and A Baddeley (Eds.), Attention and Performance IX. Hillsdale, NJ: Lawrence, pp. 153-169. Jeannerod, M (1984) The timing of natural prehension movements. 1. Mot. Behav., 16: 235-254. Jeannerod, M (1986a) The formation of finger grip during prehension. A cortically mediated visuomotor pattern. Behav. Brain Res., 19: 99-116. Jeannerod, M (1986b) Mechanisms of visuomotor coordination: a study in normal and brain-damaged subjects. Neuropsychologia, 24: 41-78. Jeannerod, M (1988) The neural and behavioural organization of goal-directed movements. Oxford: Oxford University Press. Kay, BA, Kelso, JA, Saltzman, EL and Schoner, G (1987) Space-time behavior of single and bimanual rhythmical movements: data and limit cycle model. J. Exp. Psychol. Hum. Percept. Perform., 13: 178-192.

201 Kelso, JAS (1981) On the oscillatory basis of movement. Bull. Psychon. Soc., 18. Kelso, JAS (1984) Phase transitions and critical behavior in human bimanual coordination. Am. J. Physiol., 246: R 1000-1 004. Kelso, JAS and Tuller, B (1984) A dynamical basis for action systems. In: MS Gazzaniga (Ed.), Handbook of Cognitive Neuroscience. New York: Plenum, pp. 321356. Kunesch, E, Binkofski, F and Freund, HJ (1989) Invariant temporal characteristics of manipulative hand movements. Exp. Brain Res., 78: 539-546. Kunesch, E, Binkoski, F, Steinmetz, H and Freund, H-J (1995) The pattern of motor deficits in relation to the site of stroke lesion. Eur. Neurol., 35: 20-26. Lacquaniti, F, Guigon, E, Bianchi, L, Ferraina, S and Caminiti, R (1995) Representing spatial information for limb movement: role of area 5 in the monkey. Cereb. Cortex, 5: 391-409. Levin, MF (1996) Interjoint coordination during pointing movements is disrupted in spastic hemiparesis. Brain, 119: 281-293. Massaquoi, S and Hallett, M (1996) Kinematics of initiating a two-joint arm movement in patients with cerebellar ataxia. Can. J. Neurol. Sci., 23: 3-14. Mechsner, F, Kerzel, D, Knoblich, G and Prinz, W (2001) Perceptual basis of bimanual coordination. Nature, 414: 69-73. Morasso, P (1983) Three dimensional arm trajectories. Bioi. Cybem., 48: 187-194. Muller, F, Kunesch, E, Binkofski, F and Freund, H-J (1991) Residual sensorimotor functions in a patient after right-sided hemispherectomy. Neuropsychologia, 29: 125-145. Paulignan, Y and Jeannerod, M (1996) Prehension Movements. The Visuomotor Channels Hypothesis Revisited. In: AM Wing, P Haggard and R Flanagan (Eds.), Hand and Brain. San Diego: Academic Press, pp. 265-282. Pause, M, Kunesch, E, Binkofski, F and Freund, HJ (1989) Sensorimotor disturbances in patients with lesions of the parietal cortex. Brain, 112: 1599-1625. Poizner, H, Mack, L, Verfaellie, M, Rothi, LJG and Heilman, KM (1990) Three-dimensional computergraphic analysis of apraxia. Brain, 113: 85-101. Poizner, H, Clark, MA, Merians, AS, Macauley, B, Rothi, LJG and Heilman, KM (1995) Joint coordination deficits in limb apraxia. Brain, 118: 227-242. Reich, J and Daunicht, WJ (2000) A rigid body model of the forearm. J. Biomech., 33: 1159-1168. Sainburg, RL, Poizner, H and Ghez, C (1993) Loss of proprioception produces deficits in interjoint coordination.1. Neurophysiol., 70: 2136-2147.

202 Sainburg, RL, Ghilardi, MF, Poizner, H and Ghez, C (1995) Control of limb dynamics in normal subjects and patients without proprioception. J. Neurophysiol., 73: 820-835. Seidler, RD, Alberts, JL and Stelmach, GE (2001) Multijoint movement control in Parkinson's disease. Exp. Brain Res., 140: 335-344. Soechting, JF and Lacquaniti, F (1981) Invariant characteristics of a pointing movement in man. J. Neurosci., 1: 710-720. Soechting, JF and Terzuolo, CA (1988) Sensorimotor transformations underlying the organization of arm

C. DOHLE AND H.-J. FREUND

movements in three-dimensional space. Can. J. Physiol. Pharmacol., 66: 502-507. Soechting, JF and Flanders, M (1989a) Sensorimotor representations for pointing to targets in threedimensional space. J. Neurophysiol., 62: 582-594. Soechting, JF and Flanders, M (1989b) Errors in pointing are due to approximations in sensorimotor transformation. J. Neurophysiol., 62: 595-608. Wiesendanger, M, Kaluzny, P, Kazennikov, 0, Palmeri, A and Perrig, S (1994) Temporal coordination in bimanual actions. Can. J. Physiol. Pharmacol., 72: 591-594. Winter, DA (1990) Biomechanics and motor control of human movement. New York: Wiley.

Movement Disorders Handbook of Clinical Neurophysiology, Vol. 1 M. Hallett (Ed.) © 2003 Elsevier B.v. All rights reserved

203 CHAPTER 15

Reaction time as an index of motor preparation/programming and speed of response initiation Marj an Jahanshahi * Sobell Research Department ofMotor Neuroscience and Movement Disorders, Institute ofNeurology, Queen Square, London WCIN 3BG, UK

15.1. Introduction

15.2. Historical background of the study of RTs

Measuring the accuracy and speed of responding have been two key methods for quantifying a subject's performance under a variety of experimental conditions. The speed of responding to a stimulus, the so-called 'reaction time' RT has been widely used to measure the speed of information processing in a diversity of tasks and paradigms such as covert orienting of attention (Posner, 1980) or memory scanning (Sternberg, 1966). Reaction times have also been employed as an index of movement preparation and programming. The main focus of this chapter will be on the latter use of RT, that is as an index of motor preparation and movement initiation. Following a brief historical overview, I will describe the common types of RT tasks used to study motor preparation and programming, and then outline the factors known to influence RT. I will then consider the functional anatomy of RT tasks based on evidence from imaging studies. The important issue of parallel vs. serial information processing will be a good starting point for discussion of the advantages and disadvantages of RT measurement and the value of complementary use of other techniques such as recording of event-related potentials. Finally, I will consider the application of RT measurement to a range of movement disorders.

The measurement of the time of mental processes has a long history that goes back to the work of Donders and Sternberg. Donders (1968/1969) made a distinction between A, B and C type RT tasks, respectively requiring responses to a single stimulus (type A), a particular response to each of several stimuli (type B) and a response to only one of several stimuli (type C). These respectively correspond to what is now known as simple, choice and 'go/no go' RT tasks (see below for description). With himself as the experimental subject and using vowel sounds as stimuli, Donders reported average RTs which were fastest for A task, intermediate for C task and slowest for B task. He noted that the difference between the C and A tasks, was approximately half (39 ms) of the difference between the B and A tasks (75 ms). The former difference was considered to reflect 'the time required for conception of a certain sound' or in modem parlance the time for stimulus discritnination necessary in the 'go/no go' RT task (type C) but not the simple RT task (type A). Donders considered the difference between the B and A tasks to reflect the time required "for that same conception combined with the corresponding expression of the will" (p. 424). In the language of modem neuroscience, the 'expression of the will' refers to the process of response selection, or (choosing the particular response associated with the particular stimulus presented) involved in choice RT (type B) but not simple RT (type A). Donders' logic for the 'substraction method' illustrated above, was based on his idea that a number of successive processes or stages intervene between stimulus presentation and response production and that each process or stage only begins when the preceding stage has ended. As discussed in a

* Correspondence to: M. Jahanshahi, Sobell Research Department of Motor Neuroscience and Movement Disorders, Institute of Neurology, Queen Square, London WCIN 3BG, UK. E-mail address:[email protected] Tel.: 44 20 7837 3611 (ext. 3055); fax: 44 20 7278 9836.

204

later section, the latter assumption has been questioned. In addition, Donders' 'substraction method' has been criticized since it is based on the assumption of 'pure insertion', that is, it suggests that it is possible to add or remove one of the processing stages without altering other stages of processing that intervene between the stimulus and response. In practice, it is difficult to devise experimental tasks that only differ in a key stage of processing. To overcome the problems associated with Donders' substraction method, Sternberg (1969) proposed an 'additive factor model' for stage analysis of RTs. According to this model, if two experimentally manipulated variables or factors have an additive effect on RT such that the combined effect of the two factors is equal to the sum of the effect of each factor assessed independently, then it is assumed that each factor is affecting a different stage of processing. In contrast, if the two factors interact, that is the combined effect of the two factors is either greater or smaller than the effect of each factor assessed independently, then it is assumed that the two factors are influencing at least one common stage of processing. The additive factors model, similar to Donders' subtraction method, is based on the assumption of serial and independent stages of processing between a stimulus and a response, and has been widely criticized (e.g. Sanders, 1980). Nevertheless, even its critics (Sanders, 1980) have considered the Sternberg model a valuable heuristic tool for studying reaction processes.

15.3. Measurement of RT and MT RT refers to the time interval between the presentation of a stimulus and the initiation of a response, usually involving some form of movement. Movement time (MT) is the time taken to execute a movement from its moment of initiation. A variety of mechanical and electronic techniques have been used to measure RTs and MTs. Whatever method is used for quantification of RT and MT, it should allow accuracy in the millisecond range. A distinction has been made between premotor vs. motor RTs. Premotor RTs are measured as the time between the presentation of a stimulus and the start of electromyographic (EMG) activity in the muscles acting as the prime movers in initiating and executing the response. Motor RTs refer to the interval

M. JAHANSHAHI

between the onset of EMG activity and start of the behavioral response such as lifting the finger from a microswitch. Premotor and motor RTs respectively reflect central and peripheral processes. Tomberg et al. (1991) compared different methods for recording RTs such as onset of EMG responses, pressing microswitches or photoelectric detection of finger extension or flexion. They concluded that RTs based on EMG onset or photoelectric detection of finger extension provided the most reliable RT recordings. For reliable RTs, measurements have to be obtained across a number of trials. This allows development of an attentional or response 'set' and ensures that the values are not unduly affected by lapses in attention that may occur on some trials. To ensure that subjects have understood task instructions and to familiarize them with the task requirements, it is also customary to provide a number of practice trials before the actual measurement of RTs. The optimal number of trials for reliable measurement of RTs will vary depending on the precise nature of the RT task, but while too few trials render the measurements unreliable because of inadequate sampling, too many trials may be affected by fatigue, given the need to maintain focused attention throughout the task. A much replicated finding in the RT and MT literature is the 'speed-accuracy trade-off' (for review see Meyer et aI., 1988). This refers to the finding that with faster RTs and MTs, subjects make more errors. Thus, when speed of responding improves this seems to be at the expense of poorer accuracy of performance. Trial by trial analysis of response times has also shown that following an error RTs are slower on the next trial and then gradually become faster (Rabbitt, 1966). Keyobservations relating to the speed-accuracy trade-off were made by Woodworth (1938) and Fitts (1954). In Fitts' paradigm, subjects were required to move a metal-tipped stylus between two metallic target plates as quickly and accurately as possible. The width of the target plates (0.25, 0.5, 1 and 2 inches) and the distance between them (2, 4, 8, and 16 inches) were experimentally manipulated. Fitts defined an 'index of difficulty' as a function of the distance between (D) and the width of the target plates (W). This index of difficulty, log, (2DIW) is known as Fitts' law. Fitts' results showed that MTs are a linear function of the index of difficulty. With the distance between the targets kept constant, when

REACTION TIME AS AN INDEX OF MOTOR PREPARATIONIPROGRAMMING

the target width is reduced, subjects have to slow down their movements as otherwise they will make more errors. Therefore, these results showed that an inverse relationship exists between the spatial accuracy and speed of rapid aimed movements (Fitts, 1954). While in Fitts' paradigm, spatial accuracy of the aimed movements was of interest, in other RT paradigms, the precise nature of the errors depend on the type of task used. Anticipations or premature responses are errors in timing of release of the response: the response is made before presentation of the stimulus. Such anticipatory errors usually occur in RT tasks where the subject builds up high levels of preparedness or readiness to respond such as in simple or primed choice RT tasks. Incorrect response selection is an error encountered in choice RT tasks, particularly those with a large number of stimulus(S)-response(R) alternatives or with low SR compatibility such as with poorly learned or arbitrary S-R associations. Long RTs which are two or more standard deviations outside the mean RTs for the individual subject are another type of error that can occur. Errors of omission (failure to respond) or commission (making a response when none required) are other types of errors that can be respectively encountered on 'go' and 'no go' trials of go/no go RTs. In most studies, the mean and standard deviation of RTs are reported. However, typically, RT distributions have a positively skewed shape. The presence of a skewed distribution means that the mean and standard deviation of the sample's RT, the two parameters most frequently used to quantify RT data, do not convey precise information about the performance of the sample. Use of median RTs and providing information about the fastest and the slowest responses may also be of value. Heathcote et al. (1991) recommend that "distributional analysis should not be treated as a supplementary technique. Rather, we contend that RT measures should always be analyzed using a distributional analysis." 15.4. Motor programs Complex movements such as those involved in playing the piano occur too quickly to be influenced by peripheral feedback. For performance of even the simplest movements, the motor system has to generate and process a great deal of information very

205

quickly. One way of achieving this is for movements to be prepared in detail or 'programmed' before their execution. This central programming model of motor control dates back to the pioneering observations of Lashley (1917). The basic component of this model is a motor program, defined as "a set of muscle commands that are structured before a movement sequence begins and that allows the entire sequence to be carried out" (Keele, 1968). An important piece of evidence in support of the central programming model was the observation, first reported by Henry and Rogers (1960), that the time to initiate rapid responses increased as a function of the number of components in the response. Henry and Rogers (1960) suggested that this 'sequence length effect' (SLE) was due to programming of the response, with more complex responses requiring longer programs and longer compilation times. A similar finding was reported by Sternberg et al. (1978) who found that the RT to initiate the first response in a string of typewritten or spoken responses became longer with the increase in the number of elements in a string. They proposed that the SLE reflected the longer time required for search of a motor buffer and subprogram retrieval in the case of more complex responses. The infinite number and permutations of movements that are possible raise the problem of storage of the motor programs that would be required. This storage problem is overcome by the concept of 'generalized motor programs' (Schmidt, 1975). These are abstract representations which are modified to meet current demands by setting the appropriate motor parameters. Such generalized motor programs ensure storage 'economy' while allowing infinite variability and novelty in motor performance. A second line of evidence supporting the central programming model is the existence of invariant motor patterns. For example, a person's signature retains a number of constant features whether produced using a pen on paper or a brush on a canvas or with a piece of chalk on a board. This invariance and constancy of performance has been considered to derive from the use of the same generalized motor program in the different situations. While feedforward control and lack of reliance on feedback are the key components of the central programming model, nevertheless it is clear that sensory feedback is also used by the motor system.

206 This is, for example, evident in the case of aimed movements which deteriorate in the absence of visual feedback (Keele and Posner, 1968). In fact, when describing aimed movements, Woodworth (1938) considered these to consist of two phases, an 'initial impulse' at the start of movement and a terminal phase involving 'current control', respectively corresponding to centrally programmed and feedback-dependent components.

15.5. Some common types of RT tasks A variety of RT tasks or paradigms have been used to measure specific aspects of perceptual and cognitive processing and learning. These RT paradigms are too numerous to consider individually here. Instead, given the focus of this volume on Movement Disorders, I will consider some of the most common types of RT task that have been used to study speed of processing in these disorders. These include Simple, Choice (uncued and precued), and Go/no go RT tasks, which have been employed to investigate the processes of motor preparation and programming and response initiation and inhibition. If programming of the movement occurs before the onset of a stimulus, this so-called preprogramming of a response, confers a speed advantage to RT over the conditions in which the subject engages in motor programming after the onset of the stimulus. 15.5.1. Simple, uncued and precued choice RTs In its classic form, Simple RT (SRT) is defined as the time taken to initiate a single invariant response to a single invariant stimulus. As the precise nature of the response is known in advance, it can be preprogrammed, i.e. the response can be selected and fully specified before stimulus presentation (Klapp et al., 1979). Furthermore, as the nature of the stimulus is also invariant, registering the occurrence of the stimulus is sufficient for initiation of the preprogrammed response (Fig. la). However, preprogramming of the response before onset of the stimulus, which is essential for conferring its speed advantage to SRT, is optional. As noted by Klapp (1977) "Such advance programming would not be obligatory when the SRT paradigm is used since subjects could merely ignore the advance information and postpone programming until the onset of the 'go' signal" (pp. 234-235).

M. JAHANSHAHI

A Choice RT (CRT) task involves more than one stimulus each with its associated response. CRT is the time taken to initiate one of a set of possible responses to one of a set of stimuli. On any particular trial, as the stimulus and its associated response varies, the response cannot be preprogrammed. On each trial, the subject has to identify the particular stimulus presented, refer to the stimulus-response (S-R) codes in order to select the appropriate response, and then to program, initiate and execute that response (Fig. lb). So, in CRT conditions programming of a response does occur, but unlike SRT, it is performed after stimulus presentation, rather than before. In CRT, additional processes intervene between presentation of the stimulus and execution of the response. The relative prolongation of CRT compared to SRT, or the so-called "central delay", is considered to reflect central information processing operations such as stimulus evaluation and identification, S-R decoding, response selection, and motor preparation/programming which occur following the onset of the stimulus. In situations where the stimuli and responses are highly compatible or have a "natural" or a priori relationship such as in some cases of tactile CRT, reference to S-R mapping rules or matching lists (Hasbroucq et al., 1990) for response selection may not be necessary. In such conditions, preparation of the response can proceed following identification of the stimulus. Besides the "classical" SRT and CRT paradigms depicted in Figs la and b, other RT tasks have also been used in the study of motor programming. One of the most frequently employed tasks has been the precued CRT paradigm originally devised by Rosenbaum (1980). In this task (Fig. lc), a precue presented at some fixed or variable interval prior to the imperative stimulus provides the subject with partial or complete advance information about the nature of the required response. This advance information can be potentially used by the subject to partially or fully preprogram the response. Precued CRTs are faster than uncued CRTs, since the stages of processing that in a standard CRT (Fig. 1b) would normally occur after presentation of the imperative stimulus can take place before it (Fig. lc). SRT and precued CRT tasks are similar in-so-far as in both preprogramming of the response prior to the imperative stimulus is possible. However, given that the response to be preprogrammed remains constant in

Response Preprogramming

I~I Response

+

Response InitiatIon

~

Stimulus RegistratIon

+

ImperatIVe Stimulus

Simpl. AT

I

I

I

I

I

I

b.

Response

~

Response Initiallon

~

Response Programming

+

R.sponN Selection

~

S-R Mapping·

Stimulus identifICation

•

Cholc» RT

Imperative Stimulus

Uncu~

1

I

I

I

Precue

Precued Choice RT

I

~

I

I

:I

:

I I I

I!

! !

:I

*

Response

Response InitIation

~

StImulus Registration

+

Imperative Stimulus

Response Preprogramming

~

Response Selection

+

I

S·R Mappmg •

~

Stimulus Identification

t

I:

I

I

.J

I

II

,

I I I

II

I I

,..----- f -----,

c.

'1:1

~

-.l

~

~ o ~

~

(3

::c 2:i

~

G;

~

"f1

~

.

Informatlon

ot

~ 0

Use

o

Z

Z

;I>

CIl

;I>

~m

~

~

Aavance

Fig. I. A schematic presentation of the stages of processing involving simple (SRT), uncued choice (CRT) and preeued CRT tasks. From Jahanshahi et at. (1992) with permission.

I

(

8.

208

SRT but varies from trial to trial in precued CRT, the exact nature of the attentional processing and extent of volitional control involved in the preprogramming of the response is likely to differ in the two tasks. For example, in SRT, because of the constant nature of the response across trials, development of a motor set is possible, which is not the case in precued CRT where a degree of S-R uncertainty remains since a different set of stimuli requiring different responses are presented across trials. The use of SRT and CRT paradigms in the study of motor programming is based on the fundamental fact that the time to initiate a response in a SRT task is less than the time taken in a CRT task. The faster speed of response initiation in SRT tasks results from two sources. First, in SRT tasks but not in CRT, the subject has the opportunity to prepare or preprogram the response in advance of the go signal. Second, after stimulus onset, additional processing stages are involved in CRT that are not necessary in SRT, partly as a result of preprogramming in the latter case. The extra stages increase the time between presentation of the stimulus to initiation of the response. Given that in precued CRT, preprogramming of the response on the basis of the information provided by the precue is also possible, the speed advantage of precued CRT relative to uncued CRT has been used as another index of motor programming. Frith and Done (1986) have distinguished three routes to action. They refer to a 'fast' route used in SRT conditions, a 'direct' route used in CRT tasks with high S-R compatibility, and a 'slow' route underlying responses in CRT conditions where S-R compatibility is low. They further proposed that the degree of volitional control necessary is different in the three routes. Volitional control is minimally involved in the 'direct' route (compatible CRT) in which response preparation and initiation is entirely ruled by the onset of the stimulus. In contrast, the stimulus plays a minimal role in the 'fast' route (SRT), where preprogramming of the response depends on an act of will. In the 'slow' route, volitional control is necessary to some extent for decoding of the arbitrary S-R codes. On the basis of the results of a study examining the degree of interference produced by a secondary task on the performance of SRT and CRT tasks, Frith and Done (1986) proposed that focused attention is not needed for the direct route whereas it is required for the fast

M. JAHANSHAHI

route. In summary, in Frith and Done's terminology, in SRT the response is preprograrnmed through an act of will, a process which requires focused attention, whereas in CRT, programming of a response is triggered by the onset of the stimulus and requires less volitional control and makes fewer demands on attentional resources. 15.5.2. Tasks to study the sequence length effect

Another way in which RT tasks have been used to examine motor programming has been by experimental investigation of the sequence length effect (SLE). To study the SLE, sequential finger tapping is commonly incorporated in RT paradigms, although more complex hand movements have also been studied by some (e.g. Haadland et al., 1987). In SRT, the same movement of a fixed sequence length is repeated throughout a block of trials. Therefore, the subject knows the number of required components in advance and can preprogram the movement sequence prior to the go stimulus. In an uncued CRT task, the sequence length randomly varies across trials in a block. The subject knows which sequence length to perform only after the go signal appears. In a variation of the precued CRT, on each trial a precue provides information about the number of components in a sequence before the go stimulus. This allows subjects to preprogram the movement during the precue-go signal interval of each trial. In such tasks, the effect of sequence length on measures of movement execution such as movement time (Hulstijn and Van Galen, 1983), inter-tap-intervals (Rafal et al., 1987; Stelmach et al., 1987) and mean duration (Sternberg et al., 1978) have also been examined. Since the initial observations and experiments of Henry and Rogers (1960) and Sternberg et al. (1978) showing that the RT to initiate the first movement in a sequence increased with the complexity and the number of components in the sequence, the study of SLE using finger tapping tasks has yielded inconsistent results in relation to the linearity of the SLE (Garcia-Colera and Semjen, 1987; Rafal et al., 1987; Stelmach et al., 1987; Balfour et al., 1991; Franks and Donkelaar, 1991). This inconsistency of the SLE may reflect methodological differences across studies. The precise type of movement used (tapping with single finger or 2 or 3 fingers), the nature of the RT task (SRT, uncued or precued CRT), the length of the sequence (3 to 8

REACTION TIME AS AN INDEX OF MOTOR PREPARATIONIPROGRAMMING

components), and the amount of practice are some of the variables that vary across studies.

15.5.3. Go/no go RTs Go/no go tasks allow measurement of the participants' ability to discriminate between 'go' and 'no go' stimuli and to differentially initiate and execute a response on 'go' trials and inhibit or withhold responses on 'no go' trials. A two-stimulus (51-S2) paradigm is commonly used to study go/no go RTs. Variations of the SI-S2 paradigm have been employed in the literature depending on whether S1 or S2 are the informative stimulus providing go/no go information. In the most basic form of the go/no go RT task, S1 is a warning signal which is followed after a preparatory interval by S2 which indicates that a response should be initiated (e.g. green circle or 1000 Hz tone) or withheld (e.g. red circle or 2000 Hz tone) on that trial. In Donders' original formulation, the go/no go RT or type C task required the subject to make a response to only one of several stimuli. Such go/no go RTs involved preparation of a single response which was then either initiated on go and withheld on no go trials. Therefore, compared to a simple RT with an invariant stimulus across trials, in a go/go RT task the main requirement was for the subject to discriminate go from no go stimuli and then release the preprograrnmed response. More

209

recently 'hybrid' choice go/no go RT tasks have been devised which also require more than one type of response to be prepared; In these hybrid tasks, go/ no go and movement parameter information (e.g. go to left or right button) are respectively provided by SI and S2 (Hackley and Miller, 1995) or both sets of information are communicated by different features of S2, and S1 simply acts as a warning signal (Osman et al., 1992). In go no RT tasks, a key factor affecting the strategies adopted by subjects and the latency of their RTs on go trials and their success in withholding responding on no go trials is the percentage of go to no go trials in a block. A higher percentage of no go trials increases action uncertainty. Conversely, a higher percentage of go trials increases the likelihood and extent of advance preparation by subjects which will in tum increase the probability of errors of commission on no go trials. RTs get faster when the percentage of go trials increase from 25% to 50% to 75% (Van der Molen et al., 1989).

15.6. Factors that influence RT Some of the most widely investigated factors that have been shown to influence RTs are listed in Table 1. These factors can be broadly divided into two classes: subject/operator-related and procedural.

Table I Some of the most widely investigated factors shown to influence RTs. Subject/operator-related factors

Age Handedness Arousal/attentional state of subject: attentional and motor 'set', alcohol, drug, sleep deprivation effects Instructions to subjects: emphasizing speed or accuracy Practice/fatigue effects Motivational influences Procedural factors Nature of stimuli, modality of presentation, discriminability, probability Nature of responses Extent of S-R compatibility Number of S-R alternatives Presence/absence and nature of advance information (warning signal, temporal cue, precue) Timing of presentation of stimuli: interval between warning signal/precue and go stimulus, duration of the responsestimulus interval, inter-stimulus interval

210

Detailed consideration of the influence of these factors on RT is beyond the scope of the present chapter, and I will only briefly consider the ways in which RTs are altered by some of these factors. 15.6.1. Subject/operator-related factors

The speed of processing and reacting to external stimuli changes with age. The function relating RTs to age is U-shaped. During infancy RTs are slow but with maturation of the central nervous system and particularly the motor system, RTs become faster up to the age of about 30 and from then on gradually become slower (Wilkinson and Allison, 1989). RTs are generally faster with the dominant than the nondominant hand, and the latency of responses are effector-dependent with RTs with the hand, for example, being faster than those with the foot (Boff and Lincoln, 1988). The arousal/attentional state of the subject has a major impact on RTs which is the reason why alcohol and drugs that alter the arousaV attentional state of the individual such as benzodiazepines and barbiturates affect RTs (Maylor and Rabbitt, 1973; Borland and Nicholson, 1975). For similar reasons, factors such as sleep deprivation or fatigue also prolong RTs and reduce accuracy (Boff and Lincoln, 1988). Practice has been shown to generally improve RTs and confers a differentially greater advantage on CRTs than SRTs (Mowbray and Rhoades, 1959). Practice has been shown to interact with procedural factors such as the number of S-R alternatives or S-R compatibility, such that practice results in a greater speeding of RTs under conditions with a larger number of S-R alternatives and higher S-R incompatibility (Techner and Krebs, 1974). The motivational state of the subject plays a fundamental role, with RTs being faster when success, gain or reward is anticipated (e.g. Shankweiler, 1959). Response selection has been shown to be sensitive to reward expectancy even at the neuronal level. For example, in the monkey, in a cued saccade task, evidence from single cell recordings has shown that not only target selection behavior but also the extent of neuronal activity in the lateral intra-parietal area are determined by the expected gain and outcome probability (Platt and Glimcher, 1999). The specific instructions to subjects, whether emphasizing speed or accuracy of performance, is another factor that can influence RTs (Fitts, 1966).

M. JAHANSHAHI

15.6.2. Procedural factors

The precise nature of the stimuli and responses and how stimuli and responses are related are key determinants of the speed and accuracy of responding (Keele, 1986). Simple, Choice and go/no go RTs are influenced by a set of common factors, for example, all variables that influence the subject's performance as well as a number of procedural variables such as stimulus discriminability. However, there are also factors that specifically prolong or speed up only SRTs, CRTs or go/no go RTs. Under optimal conditions, RTs to tactile stimuli are fastest and those to visual stimuli slowest, with reactions to auditory stimuli being intermediate (Postman and Egan, 1949). Stimuli that are difficult to discriminate or of low intensity are associated with longer RTs (Woodworth, 1938; Postman and Egan, 1949). In CRT, two factors that can influence response selection are response uncertainty (the number of SR alternatives) and S-R compatibility (Sanders, 1980). The interaction of these two factors in CRT tasks (e.g. Broadbent and Gregory, 1965) suggests that they affect a common stage of processing. A reliable finding in RT studies is that an increase in the number of S-R alternatives results in an increase in RT. This association is represented by the HickHyman law: RT=a+b log, N, where a and b are constants and N is the number of alternatives (Hick, 1952; Hyman, 1953). In the process of response selection, the subject must perform a sensory-motor transformation to retrieve the appropriate response representation from the response set based on the stimulus presented on each trial (Hasbroucq et aI., 1990). In conditions where the relationship between the stimuli and responses are not natural or highly compatible, decoding of S-R relations becomes a significant factor in response selection. The nature of the relationship between the stimuli and responses and the degree of S-R compatibility determines the amount and type of decoding necessary for response selection. RTs are faster when stimuli and responses are highly compatible (Fitts and Seeger, 1953; Fitts and Deiginger, 1954). The nature of the S-R pairing may be spatial (e.g. left stimulus indicating a left response) or symbolic (e.g. the number 12 indicating a response to the top position of a circular clock-like array of buttons). In conditions of low S-R compatibility, the S-R relationship may be based on some

REACTION TIME AS AN INDEX OF MOTOR PREPARATIONIPROGRAMMING

arbitrary association (e.g. red stimulus indicates left response), or on some transformational rule (e.g. a stimulus on the left indicates a right response). The S-R compatibilty effect has been discussed in terms of the strategies that subjects use in performing the S-R transformation. For S-R decoding in CRT tasks with low S-R compatibility, Hasbroucq et al. (1990) have distinguished between the subjects' use of strategies involving rule implementation (when a common concept links the stimuli and responses) and memory list scanning (when stimuli and responses cannot be represented in terms of a common concept). When independently investigated, higher S-R compatibility results in faster RTs and a greater number of S-R alternatives prolong RTs. When manipulated concurrently, S-R compatibility interacts with the number of S-R alternatives and high S-R compatibility can reduce the effect of having a larger number of S-R alternatives on RTs (Boff and Lincoln, 1988). Related to S-R compatibility, is the so-called 'Simon' effect (Simon, 1969). This is the prolongation of RTs due to a taskirrelevant feature. For example, right and left hand RTs to arrows pointing to the left or right are prolonged when the task relevant information (direction of arrow) is incongruent with the task irrelevant information such as the spatial location of arrow presentation relative to a central fixation point (e.g. left pointing arrow presented to the right of central fixation cross). The presence/absence, the precise nature of the advance information provided by preparatory stimuli (warning signal or precue) and their timing relative to the go stimulus also influence RTs. In general, RTs are faster with than without a preparatory stimulus. A preparatory stimulus can act as a warning signal and/or temporal cue and/or a movement parameter precue. A warning signal informs the subject that an imperative stimulus is about to be presented. It allows the subject to increase his/her readiness to respond. A temporal cue is a preparatory signal which allows the subject to prepare to respond at a particular time. For example, a warning signal presented at a fixed interval prior to an imperative stimulus across trials can act as a temporal cue as it will allow the subject to develop an anticipatory set or temporal expectancy for the occurrence of the imperative stimulus. A movement parameter cue provides information about the precise character-

211

istics of the response, such that it allows the subject to select and prepare a specific response in advance of the imperative stimulus. The interval between the preparatory and imperative stimuli, the so-called S1S2 interval also affects RTs. RTs initially become faster with longer intervals of up to 200 ms (which is the optimal foreperiod resulting in the fastest RTs) and then gradually level out and then become slower with S1-S2 intervals in the 4-8 s range (Posner et al., 1973; Posner and Snyder, 1975; Van der Molen et al., 1989). Thus RTs have a U-shaped relation to the duration of the warning signal interval in the 0 (unwarned) to 800 ms range. In contrast, the function relating percent of errors to the warning signal interval has an inverted U shape, with the highest percent of errors observed for foreperiods of 100 ms (Ponser et al., 1973). RTs are prolonged when intervals between the response and presentation of the next stimulus are short and become faster when the R-S intervals are increased from 20 to 200 ms (Rabbitt, 1980). Another timing-related factor of importance is the inter-stimulus interval. The socalled 'psychological refractory period' refers to the finding that if the interval between two stimuli both of which require responses is shorter than the time required to produce a response to the first stimulus, the RT to the second stimulus is delayed (Boff and Lincoln, 1988). The probability/frequency of presentation of a stimulus across trials also determines the latency of responses to it. RTs are faster to more frequent stimuli (Fitts et al., 1963; Hawkins and Underhill, 1971). Evidence suggests that it takes less time to perceive a highly probable stimulus than a less probable stimulus (Miller and Pachella, 1973). So faster perceptual processing times could partly explain the faster RTs with higher probability stimuli. For disjunctive RTs, the percentage of go and no go stimuli are altered across blocks in some studies, for example, from 75, to 50 to 25%. RTs to the go stimuli are faster in blocks with higher go to no go probabilities (Low and Miller, 1999). In these circumstances, it is also possible that the changes in go probability alters the usefulness of motor preparation across trials (Low and Miller, 1999). Related to such probability effects is the 'repetition effect' . This refers to the finding that RTs are faster on trials when the same stimulus is repeated as on the preceding trial (Berthelson, 1961).

212

15.7. Functional anatomy of RT tasks A hierarchical model of action representation has been proposed by Requin (1992), with each level of the hierarchy considered to be associated with activity in specific brain areas. At the most abstract level, are 'action goals', non-motoric and symbolic representations of what the goal is. It is suggested that the representation of action goals is mediated by the frontal and parietal association cortices. At the next level, are motor programs which are also nonmotoric representations involved in setting the best parameters for movement based on past experience and current environmental constraints and associated with activity in the lateral and medial premotor cortices. Finally, there are the basic motor units involved in production of the motor output, namely the primary motor cortex and cortico-spinal pathways. Animal studies, investigations of patients with focal brain lesions, functional imaging, and studies using transcranial magnetic stimulation (TMS) have provided evidence for the involvement of these brain regions in the preparation, initiation and execution of movements. Animal studies have established that motor preparatory signals can be recorded from the motor, lateral and medial premotor, posterior parietal, and prefrontal cortices as well as from the striatum and cerebellum (Fuster and Alexander, 1971; Evarts and Tanji, 1974; Tanji et al., 1980; Alexander, 1987; Crarnmond and Kalaska, 1989; Muschiake and Strick, 1995; Wise et al., 1997). In one of the first studies of this kind, Evarts and Tanji (1974) used a delayed CRT task, in which monkeys were trained to pull a handle when a red light was presented and to push the handle when a green light was presented. Neuronal activity in the motor cortex started with presentation of the stimulus and continued until movement execution. It has been unclear whether delay related neuronal activity in the prefrontal cortex reflects motor preparation or the role of this area in working memory. There is now evidence from animal lesion (Quintana and Fuster, 1993) and imaging (D'Espositio et al., 2000) studies that the prefrontal cortex is involved in both delayed response tasks that involve working memory or motor preparation. Patients with unilateral lesions of the prefrontal cortex (Alivisatos and Milner, 1989) or the medial frontal lobe including the SMA (Verfaellie and

M. JAHANSHAHI

Heilman, 1987) do not benefit from advance movement parameter information provided by precues, suggesting a failure of motor preparation. Functional imaging with PET or fMRI has confirmed the role played by the prefrontal, premotor and parietal cortices in response selection and preparation. Compared to rest, a SRT involving performance of a simple movement (right index finger extension) on presentation of a tone with irregular inter-stimulus intervals of 2-7 s across trials, significantly activated the sensorimotor cortex, caudal SMA, parietal area 40, insular cortex and the putamen (Jenkins et al., 2000). The same movement, when performed with regular inter-stimulus intervals, which allows stimulus anticipation and motor preparation, also activated the lateral premotor cortex, the rostral SMA and the anterior cingulate (Jahanshahi et al., 1995), suggesting that the activation of these additional areas related to the motor preparation possible with regular but not irregularly timed SRTs. Naito et al. (2000) also found that relative to a rest condition, visual, auditory and somatosensory SRTs were associated with activation of the sensorimotor cortex, dorsal premotor cortex, SMA and anterior cingulate, and that there were no modality-specific activations in motor areas. They also found that greater activation in the anterior cingulate was associated with faster RTs. The lack of differential effects of stimulus modality on activation of motor areas has been confirmed by some (Weeks et al., 2001) but not others (Sugiura et al., 2001). Nevertheless, all three studies (Naito et al., 2000; Suguira et al., 2001; Weeks et al., 2001) underline the activation of sensorimotor, dorsal and medial premotor areas with RT tasks. When response selection is required, the prefrontal cortex is also activated. For example, compared to a 'fixed' control condition, during which subjects moved a joystick in a predetermined direction, performance of joystick movements in one of four possible directions which necessitates response selection on each trial was found to significantly activate the dorsolateral prefrontal cortex (BA 46, 9), premotor cortex, SMA and the superior parietal cortex (Deiber et al., 1991). CRTs require response selection on every trial which is not the case in SRT, where a pre-selected response is initiated on presentation of the stimulus. Comparison of CRTs and SRTs has also revealed greater activation of the inferior and middle frontal areas, dorsal premotor cortex and intraparietal sulcus

REACTION TIME AS AN INDEX OF MOTOR PREPARATIONIPROGRAMMING

associated with response selection (Shulter et al., 2001). Use of precued RT tasks providing full or partial information about movement parameters or the time of presentation of the imperative stimulus with PET have been found to be associated with greater activation of the left parietal cortex compared to CRTs providing no advance information (Deiber et al., 1991; Coull and Nobre, 1998), thus indicating the importance of this area in motor preparation. This has been confirmed in an event-related fMRI study (Toni et al., 1999) which revealed that with a precued CRT task sustained activity during the precue-imperative stimulus was observed in the dorsal premotor cortex and the posterior parietal cortex. Compared to control conditions, RT tasks with low S-R compatibility are associated with greater activation of the superior parietal cortices (Deiber et al., 1991; Iacoboni et al., 1996) and the dorsolateral prefrontal cortex (BA 46,9) (Deiber et al., 1991). The Simon effect on RTs is also associated with greater activation of the parietal cortices relative to the Stroop interference task (Petersen et al., 2002). Functional imaging has been useful in identifying the brain regions activated during performance of a variety of RT tasks. However, from these data it is difficult to establish the areas that are essential for preparation, initiation or execution of a response. Such information which previously was only available from studies of patients with focal brain lesions or animal lesion studies, can now be also obtained by using TMS to create a temporary 'virtua11esion' in specific areas under the coil and examine whether or not and at what point performance is disrupted. In an SRT task, magnetic stimulation of the motor cortex in the interval between the onset of the imperative stimulus and execution of the response, was shown to delay the motor response for up to 150 ms (Day et al., 1989). The pattern of agonist/antagonist EMG activation was unaffected, which was interpreted as indicating that the motor program for the movement was held unaltered in a motor buffer and was released after the inhibitory process had subsided. In a precued RT task, Hasbroucq et al. (1997) used TMS to demonstrate that corticospinal excitability was reduced prior to and then increased following presentation of the response signal. Using TMS with go/no go RTs, it has also been demonstrated that while MEPs in the agonist muscle increased 100-200 after the go signal they decreased after the

213

no go signal; respectively reflecting the activation and inhibition of the corticocospina1 pathway (Hoshiyama et al., 1996, 1997; Leocani et al., 2000). Shulter et al. (1999) delivered TMS over three sites at different times after the presentation of the visual cue. They found that TMS over the anterior premotor cortex slowed movements when applied 140 ms after the visual signal; whereas TMS over the posterior motor cortex slowed movements when applied 180 ms after the cue; and movements were delayed with TMS over the sensorimotor cortex when this was delivered 220 ms after the cue. The results were considered to reflect the change of activity from signal to movement-related processing from the premotor to motor cortex. Shulter et al. (1999) also used a precued RT task with the precue and the 'go' signal respectively informing subjects which movement to make and when to initiate it. TMS over the anterior premotor cortex or the sensorimotor cortex during the interval between the precue and the 'go' signals both slowed the movement, indicating the involvement of both areas in motor 'set'. Using repetitive TMS, Praamstra et al. (1999) found that stimulation over the premotor cortex increased interference effects due to the Simon effect. Evidence for the role of the dorsolateral prefrontal cortex in response selection as distinct from working memory was provided by Hadland et al. (2001) who showed that repetitive TMS disrupted performance in a task requiring response selection but without any working memory demands. Several lines of evidence suggest that performance of go/no go RTs depends on the integrity of the prefrontal cortex. A female patient with a meningioma in the medial aspects of the prefrontal cortex bilaterally had impaired go/no go RTs, with a large number of errors of commission on go trials. Following surgical excision of the tumor, the go/no go performance of the patient became normal (Leimkuhler and Mesulam, 1985). In a group study of patients with frontal or posterior regions, Godefroy et ai. (1996) found that responses to irrelevant stimuli on bimodal go/no go RT tasks were more frequent with lesions of the dorsolateral prefrontal cortex and the caudate nucleus. In a study by Ikeda et al. (1996), field potentials associated with a warned choice go/no go RT task were recorded from subdural electrodes over the prefrontal, SMA and primary motor cortex in 5 patients with epilepsy. Both go and no go trials elicited a slowly rising

214

potential between SI and S2, with part of the potential at least corresponding to the late CNV. This was seen not only at electrodes over the SMA, but also over orbitofrontal and mesial prefrontal areas. This potential was considered to be related to decision making upon S2 in both go and no go conditions. There is now also mounting evidence from imaging studies (e.g. Humberstone et aI., 1997; Konisihi et aI., 1999, Rubia et aI., 2000) for the involvement of various prefrontal areas in go/no go RTs. For example, using fMRI, Humberstone et al. (1997) reported that the SMA proper was activated only on the go trials associated with motor execution, while the pre-SMA was activated in both go and the go go trials, representing the process of decision making about whether to initiate or inhibit a response. 15.8. Parallel vs. serial information processing

While the above section has been concerned with where in the brain the processing required for performance of RT tasks takes place, there is also the issue of the time course of information processing that precedes the production of a response to a stimulus. In much of the foregoing account and in Fig. 1, a discrete serial or stage model of RTs has been adopted. According to such a model, the stages of processing intervening between presentation of a stimulus and production of an appropriate response occur one after another in a serial fashion. In the "classical" versions of stage models (Donders, 1969; Sternberg, 1969), processing in anyone stage does not begin until processing in preceding stage has been completed. In contrast, alternative conceptualizations of RT, the so-called "parallel" processing models, for example, the "continuous flow model" (Grice et aI., 1977, 1982; Eriksen and Schultz, 1979; McClelland, 1979), maintains that the processes necessary for production of an appropriate response can be activated with presentation of the stimulus and operate simultaneously and in parallel rather than sequentially. A third type of model, the "asynchronous discrete" model of Miller (1982), combines features of the discrete serial and parallel models. It suggests that transfer occurs between the processing elements only when an element has completely processed a "grain" of information. While information represented by a single grain is transferred discretely, various processing elements

M. JAHANSHAHI

can work in parallel when more than one grain of information is involved. A number of studies have addressed the question of whether processes that intervene between presentation of a stimulus and production of a response occur in series or in parallel. Recording of eventrelated potentials, particularly the lateralized readiness potential (Coles et aI., 1986; Smid et aI., 1987) has proven of value in investigation of the 'serial vs. parallel' processing issue, since it has supplied an index of covert processing to supplement the measure of overt responses provided by RTs. Two paradigms have been commonly used, the conflict paradigm or a modified 'hybrid' go/no go RT task. In the conflict paradigm, the subject has to respond with either the left or right hand to target stimuli presented in the center of a VDU. On compatible trials, the target stimulus is flanked by stimuli indicating the same response as the target, for example, a central arrow pointing to the right, flanked by other arrows also pointing to the right. In contrast, on incompatible trials, the target and flanking stimuli indicate different responses, for example, a central arrow pointing to the right, flanked by arrows pointing to the left. The critical issue in relation to the 'serial vs. parallel' processing question is what happens on the incompatible trials when the target and flanking stimuli provide conflicting information. If partial information transmission occurs, then information about the incompatible flankers may also reach the response system and result in preparation of the incorrect response in parallel with preparation of the correct response. In the hybrid go/no go RT task, one feature of the stimulus provides information about hand selection while another feature indicates whether to initiate or withhold a response on that trial. In this task, the critical result is what happens on the no go trials, and presence of an initial LRP on no go trials would provide evidence in favor of partial transmission of information. Recording of the LRP has shown that the incorrect response is in fact covertly prepared on incompatible trials in the conflict paradigm (Gratton et al., 1988; Smid et aI., 1991) and that partial information about the responding hand does result in preparation of a response on no go as well as go trials (Miller and Hackley, 1992; Osman et aI., 1992; Smid et aI., 1992). This evidence favors partial transmission of information as suggested by the continuous flow (Grice et al., 1977, 1982; Eriksen

REACTION TIME AS AN INDEX OF MOTOR PREPARATIONIPROGRAMMING

and Schultz, 1979; McClelland, 1979) and asynchronous discrete (Miller, 1982) models rather than the serial models (Donders, 1969; Sternberg, 1969) of RT. 15.9. RT and MT in movement disorders In the study of movement disorders, the measurement of RTs and MTs to quantify the deficits in movement initiation and execution has a long history. In Parkinson's disease, the first "formal" study of RT was by Wilson (1925), who obtained an average RT of 360 ms for patients compared to the mean RT of 240 ms for normal controls. A report in 1959 by King provided no quantitative data but made a number of observations on the RT of four patients with Parkinson's disease compared to sexand age-matched normal controls. The first quantitative SRT and CRT data in Parkinson's disease were provided by Talland (1963) who assessed SRT and CRT in 25 patients and 25 age- and sex-matched normals. The eight most severely affected patients were noted to be significantly slower under all conditions than the controls and the rest of the patient sample. The more recent work on RTs in Parkinson's disease and other movement disorders has been partly driven by attempts to test the hypothesis that the basal ganglia are normally responsible for "the automatic execution of learned motor plans" (Marsden, 1982), a corollary of which is the proposal that motor symptoms such as bradykinesia and akinesia are problems with programming of movements. Below, a brief overview of the RT literature in movement disorders will be provided, starting with Parkinson's disease which has been most extensively studied with RT paradigms. 15.9.1. RT in Parkinson's disease

Several approaches have been used to study motor programming with RT paradigms in Parkinson's disease. As SRT provides the possibility for preprogramming of motor responses, the first approach has been through a comparison of SRT in patients with Parkinson's disease and normals. The second approach has involved a comparison of SRT and CRT in patients with Parkinson's disease relative to normals. A selective impairment of SRT and normality of CRT in Parkinson's disease has been considered to indicate deficits in preprogramming of

215

movements. The third approach has examined whether advance information about the required response can be used by patients with Parkinson's disease to preprogram movements and hence speed up their RT. The fourth approach has investigated the sequence length effect in Parkinson's disease to determine whether the time to initiate the first response in a sequence increases linearly as a function of the length of the sequence or not. As a result, there is now a sizeable literature on RTs in Parkinson's disease. For each of these four RT approaches to study motor programming in Parkinson's disease, the results of the existing studies are inconsistent. With regard to the first approach, while most studies have found slower SRTs in Parkinson's disease relative to normals (e.g. Bloxham et al., 1984; Sheridan et aI., 1987; Pullman et aI., 1988; Goodrich et aI., 1989; Daum and Quinn, 1991; Jahanshahi et aI., 1992a; Brown et al., 1993; Harrison et aI., 1995; Kutukcu et aI., 1999; Henderson et al, 2001), some have not found such a deficit (Talland, 1963; Wiesendanger et aI., 1969; Girotti et aI., 1986; Heitanen and Teravainen, 1986; Kerr et aI., 1988). Nevertheless, of all the findings relating to RTs in Parkinson's disease, the most consistent is a deficit in SRT (Jahanshahi et al., 1992a; Gauntlett-Gilbert and Brown, 1998). However, in SRT preprogramming of the response before onset of the stimulus is optional. Therefore, in isolation the results of these studies do not allow an unequivocal attribution of the prolongation of SRT in Parkinson's disease to deficits in motor programming, which could be due to slowness of the other two stages of processing that precede execution of the response (Fig. la), i.e. stimulus registration or response initiation. In fact, Klapp (1977) recommended that "research on response programming should be based on choice rather than simple reaction time and, furthermore, that effects appearing in simple reaction time may not necessarily be given a programming interpretation." In relation to the second approach comparing SRT and CRT in Parkinson's disease and normals, a selective slowing of SRT in patients relative to normals has only been found in some (Evarts et aI., 1981; Bloxham et aI., 1984; Sheridan et al., 1987; Pullman et al., 1988; Goodrich et aI., 1989; Henderson et aI., 2001) but not other studies (Talland, 1963; Wiesendangeret al., 1969; Girotti et aI., 1986; Heitanen and Teravainen, 1986; Stelmach et aI.,

216

M. JAHANSHAHI

General slowing

SRT

No slowing

CRT

CRT

Selective slowing of SRT

Selective slowing of CRT

SRT

SRT

CRT

SRT

CRT

Fig. 2. The four patterns of SRT vs. CRT found in Parkinson's disease (filled symbols) relative to normal controls (white symbols) in the literature.

1986; Mayeux et al., 1987; Lichter et al., 1988; Reid et al., 1989; Pullman et al., 1990; Jahanshahi et aI., 1992a; Brown et al., 1993; Kutukcu et al., 1999). In fact, as discussed elsewhere (Jahanshahi et al., 1993b), when comparing Parkinson's disease patients and controls, the literature suggests that the relationship between SRT and CRT can be described by one of 4 possible patterns. These 4 patterns are shown in Fig. 2a-d. These patterns which can be respectively labeled "no deficit" (Talland 1963; Wiesendanger et al., 1969; Girotti et aI., 1986; Heitanen and Teravainen, 1986; Kerr et aI., 1988), "general slowing" (Stelmach et aI., 1986; Mayeux et aI., 1987; Dubois et aI., 1988; Rafal et aI., 1989; Yanagisawa et aI., 1989; Pullman et al 1990; Daum and Quinn, 1991; Jahanshahi et aI., 1992a), "selective slowing of SRT" (Evarts et aI., 1981; Bloxham et al., 1984; Sheridan et aI., 1987; Pullman et aI.,

1988; Goodrich et al., 1989) and "selective slowing of CRT" (Wiesendanger et al., 1969; Lichter et aI., 1988; Reid et al., 1989; Brown, Jahanshahi and Marsden; 1993; Kutukcu et aI., 1999) are each supported by evidence from a number of studies. The third approach has examined whether advance information about the required response provided by a precue can be used by patients with Parkinson's disease to preprogram movements and hence speed up their RTs. The studies which have used this approach have shown that the use of advance movement parameter or temporal information provided by precues is qualitatively normal in Parkinson's disease (e.g. Stelmach et al., 1986; Jahanshahi et aI., 1992a; Labutta et aI., 1994), but that patients only take full advantage of such advance information with longer intervals between precues and go stimuli (Jahanshahi et al., 1992a).

REACTION TIME AS AN INDEX OF MOTOR PREPARATIONIPROGRAMMING

From a quantitative meta-analysis of the results of the studies which have examined use of advance information in Parkinson's disease, Gauntlett-Gilbert and Brown (1998) concluded that there is no evidence of motor programming deficits from these results. The fourth approach based on the sequence length effect has produced inconsistent results with some evidence suggesting deficits in motor programming in Parkinson's disease (e.g. Harrington et aI., 1991) and others finding no such deficits (e.g. Rafal et al., 1987; Stelmach et aI., 1987). From the existing evidence, the most parsimonious conclusion is that patients with Parkinson's disease have a deficit in response initiation, which is a stage of processing common to both SRT and CRT (J ahanshahi et al., 1992a). Regardless of the approach that has been used to study RTs in Parkinson's disease, it is clear that the literature is inconsistent about the existence and the nature of deficits in RTs in this disorder. One possible reason for this inconsistency may be the heterogeneous nature of Parkinson's disease, whereby the presentation may vary from a pure motor disorder to one accompanied by depression, cognitive (frontal) dysfunction or even dementia. Berry et al. (1999) examined the influence of 'frontal lobe impairment' as one factor that could influence presence or absence of motor slowness in Parkinson's disease. They divided their sample into two groups: those who were impaired on a test of frontal lobe function, the Wisconsin Card Sorting test, and those who were not. Only the 'frontally impaired' subgroup had significantly slower RTs on visual search tasks than controls and the 'non-frontal' subgroup had RTs that were similar to those of normals. This suggests that RT deficits may only be present for those patients with Parkinson's disease, for whom the pathophysiology affects the frontal lobes. This issue of heterogeneity is clearly worth further investigation with regards to other features of the disorder. 15.9.2. MT in Parkinson's disease

In addition to slowness in initiating movements, patients with Parkinson's disease have problems in execution of movement. This problem is evident for execution of all forms of movement from simple ballistic movements aimed at targets, to simultaneous or sequential and continuous movements (Brown and Jahanshahi, 1996).

217

Performance of fast ballistic movements aimed at a target depend on a characteristic pattern of preprogrammed agonist/antagonist EMG activity. In patients with Parkinson's disease the size of the first agonist burst is reduced (Hallett et al., 1977; Hallett and Khoshbin, 1980), and although patients can modulate the size of the first agonist burst as a function of amplitude of movement (Berardelli et aI., 1986); movements are hypometric, requiring additional bursts of agonist activity to reach the target. Besides the step-wise manner in which targets are achieved during aimed ballistic movements, patients with Parkinson's disease are shown to be dependent on visual feedback to a greater extent than normals during performance of fast movements (Cooke et al., 1977). Execution of pointing movements by patients with Parkinson's disease are particularly slow and hypometric when performed without vision of the moving hand (Klockgether and Dichgans, 1994). In fact, patients are described as executing ballistic movements in a 'closed loop' fashion, requiring online feedback control, instead of in an 'open-loop', fast and pre-programmed way (Flowers, 1976). For aimed movements, movement time is a function of difficulty defined by factors such as target size or distance (Fitts, 1954). With distance kept constant, when the target width is reduced, even normal subjects have to slow down their movements or rely on visual guidance to ensure end-point accuracy, otherwise they will make more errors. In two experiments where participants were required to make rapid or slow movements to targets at different distances or widths, Sheridan et al. (1987, 1990), found that like normals, patients with Parkinson's disease can produce fast ballistic movements but these had disproportionately lower end-point accuracy than in normals. In real-life situations, for example, when reaching for a cup of tea, underreaching has less drastic consequences than over-reaching. While the latter may result in spilling the contents of the cup, under-reaching could be corrected by one or more additional movements. Thus, in Parkinson's disease, the fact that movements are hypometric and dependent on visual feedback may represent a strategic adaptation to a 'noisy' motor control system which has a basic deficit in force generation (Sheridan et aI., 1987; Brown and Jahanshahi, 1996). More complex movements requiring bimanual coordination and simultaneous and sequential move-

218

ments are also impaired in Parkinson's disease. Early studies showed that simultaneous performance of movements such as pressing a tally counter and picking up beads with a tweezer (Talland and Schwab, 1964) are impaired in Parkinson's disease. Relative to controls, patients with Parkinson's disease show greater decline in bimanual index finger tapping, bimanual peg insertion and simultaneous tapping with one hand and peg insertion with the other hand compared to when performing these tasks unimanually (Brown et al., 1993). In the study of Benecke et al. (1986), two movements, a rapid forearm flexion and a 30N isometric contraction of the hand were performed either alone or simultaneously. When executed alone, patients with Parkinson's disease were slow in performing each movement relative to controls. When the two movements were performed simultaneously, the patients were even more dramatically slower than normals. Similar problems with performance of simultaneous eye and hand movements (Warabi et al., 1988) or reach and grasp movements (Castiello et al., 1993) have been reported in Parkinson's disease. The patients' problems with simultaneous movements are less pronounced if the tasks performed with the two hands are discrete and involve a common timing element (Stelmach and Worringham, 1988). In another study, Benecke et al. (1987) required subjects to perform a set of movements sequentially. Patients with Parkinson's disease showed an increased inter-onset-latency between sequential movements. Marked hesitations between movement segments in Parkinson's disease relative to controls has also been documented by others (e.g. Agostino et al., 1992). In Parkinson's disease, reduction of visual guidance impairs performance of sequential movements further (Georgiou et al., 1993). These problems with sequential movements have been interpreted as reflecting difficulty in rapid switching between motor programs, particularly when this has to be internally driven rather than externally triggered. The ability to make continuous and predictive movements required for accurately tracking a moving target have also been investigated in Parkinson's disease. The results are inconsistent, with some studies suggesting that patients with Parkinson's disease show impaired learning of manual pursuit tracking (e.g. Frith et al., 1986) and others finding no such deficits in non-demented and medicated

M. JAHANSHAHI

patients (e.g. Bloxham et al., 1984; Soliveri et al., 1997). Such predictive movements require development of an 'internal representation' incorporating the spatial and temporal characteristics of the moving target. Day et al. (1984) found that when subjects were explicitly told that the target track was repetitive, patients with Parkinson's disease similar to the controls showed predictive tracking and were able to reduce their phase lag. Nevertheless, patients are more dependent on visual feedback, and tracking which is quite normal with visual feedback, deteriorates when visual feedback is not available (Cooke et al., 1978; Flowers, 1978).

15.9.3. Effect of Levodopa on RT and MT in Parkinson's disease In idiopathic Parkinson's disease, the motor symptoms improve with levodopa. This clinical benefit of levodopa would lead one to expect that RTs and MTs would also be significantly faster following such medication. This is not borne out by the evidence which has shown that levodopa medication does not consistently improve RTs or MTs across studies. Several methodological approaches have been used to study the effect of medication on RTs and MTs in PD (Gauntlett-Gilbert and Brown, 1998), including: (i) assessing de novo patients before and after starting therapy or comparing de novo and medicated patients (e.g. Velasco and Velasco, 1973; Jordan et al., 1992; Van Hilten et al., 1998); (ii) examining the effect of withdrawal of medication on performance (e.g. Bloxham et al., 1987; Starkstein et al., 1989; Jahanshahi et al., 1992b; Zappia et al., 1994); (iii) investigating the effect of spontaneous on-off fluctuations of motor symptoms on RT and MT (e.g. Rafal et al., 1984; Girotti et al., 1986); and (iv) measuring the effects of intravenous injections of levodopa (Pullman et al., 1988, 1990; Van Hilten et al., 1997). Regardless of the approach used, the majority of studies have shown that while patients with Parkinson's disease have faster RTs and MTs on compared to off medication, the benefits of levodopa on RT are often small and non-significant (Velasco and Velasco, 1973; Heilman et al., 1976; Girotti et al., 1986; Bloxham et al., 1987; Pullman et al., 1990; Jahanshahi et al., 1992b; Jordan et al., 1992; Starkstein et al., 1992; Van Hilten et al., 1997, 1998). Other studies have found SRT (Trevainen and Calne,

REACTION TIME AS AN INDEX OF MOTOR PREPARATIONIPROGRAMMING

1980; Rafal et aI., 1984; Viallet et aI., 1987; Montgomery and Nuessen, 1990; Harrison et aI., 1995) or CRT (Pullman et aI., 1988; Brown et al., 1993; Zappia et aI., 1994; Harrison et aI., 1995) to be significantly faster following levodopa medication. This inconsistent effect of levodopa on RT is also mirrored in the association of RT with ratings of severity of various symptoms. While some studies (e.g. Yanagisawa et aI., 1989) have found positive associations between RT and ratings of bradykinesia and rigidity, others (e.g. Dubois et aI., 1988) have found no such associations or have found RT to be more closely related to rigidity and tremor than to bradykinesia (e.g. Lichter et aI., 1988). Some investigators (Zappia et aI., 1994) have proposed that MT is of greater value in the assessment of Parkinson's disease than RT, since levodopa has a greater effect on MT which is also more strongly related to the severity of symptoms, particularly bradykinesia than RT. Similar suggestions have been made in favor of tapping rate, due to its greater responsiveness to dopaminergic manipulation than MT or RT (Van Hilten et aI., 1997). However, as noted by Harrison et aI. (1995) RT should not be considered as a laboratory test for akinesia or bradykinesia, but rather as a task involving a specific set of cognitive and motor processes. A number of explanations have been proposed for the inconsistent effect of levodopa on RT. First, it is possible that in Parkinson's disease RT deficits result from deficiency of a neurotransmitter other than dopamine. In fact, CSF levels of a noradrenaline metabolite significantly correlate with SRT and CRT in Parkinson's disease (Stem et al., 1984). Second, Jahanshahi et al. (1992b) suggested that it is possible that some form of threshold effect is operational such that the clinical symptoms of Parkinson's disease, RT and MT deficits appear at different levels of dopamine depletion, with clinical symptoms appearing first, followed by slowing of MT and then RT slowing. This proposal that the effects of levodopa may be threshold related, with differentially greater effects with increasing complexity of movement, is supported by the results of studies showing greater improvement of complex than simple movements with levodopa (e.g. Benecke et aI., 1987). Third, as discussed in more detail below and elsewhere (Jahanshahi et aI., 1993a, b), it is possible that RT slowness is a non-specific feature of brain damage and impaired not only in Parkinson's disease but also

219

other disorders involving striato-frontal dysfunction as well as other forms of neurological illness. 15.9.4. RT and MT in other movement disorders

Although less extensively studied, RT and MT have also been assessed in other movement disorders such as Huntington's disease, dystonia and other parkinsonian syndromes such as progressive Supranuclear Palsy (PSP), Multiple System Atrophy (MSA) and Cortico-basal degeneration (CBD). Slowness of RT and MT has been documented in patients with Huntington's disease. Girotti et al. (1988) found that the CRTs of 20 non-demented patients with Huntington's disease were significantly longer than those of 44 hospital controls and 67 patients with Parkinson's disease. Halsband et aI. (1990) compared the performance of 14 patients with Huntington's disease, nine Parkinson's disease sufferers and 20 normal controls on two tasks. On the first task requiring precise aiming movements, the patients with Huntington's and Parkinson's disease showed similar increases in MT as a function of task difficulty relative to normals. On the second writing task, the patients with Huntington's disease showed disproportionate increases in writing time when letter size was increased. Viallet et aI. (1987b) reported that the SRTs of six patients with PSP were significantly longer than those of 13 Parkinson's disease patients. Similarly, in the study of Dubois et aI. (1988) 10 patients with PSP had significantly longer RTs on a SRT and three go-no go CRT tasks compared to 33 patients with Parkinson's disease and 20 normal controls. In addition, they found that while 'decision time' (RT difference between go/no go CRT and SRT) and 'analysis time' (RT difference between two go/no go CRTs with increased complexity of stimulus discrimination) were significantly longer for patients with PSP than controls, these measures did not differ significantly for the Parkinson's disease patients and controls. More recently, Pirtosek et aI. (2001) compared RTs and cognitive event-related potentials in patients with Parkinson's disease without dementia, PSP, MSA and CBD and matched groups of younger and older controls. Only the patients with PSP and CBD showed ERP abnormalities in the auditory oddball P300 paradigm. In the auditory and visual selective attention tasks, all patient groups showed ERP abnormalities to varying degrees, with

220 the changes in PSP being most prevalent and those in Parkinson's disease and MSA being minimal. The patients with CBD had the longest RTs which were significantly different from those of controls and patients with Parkinson's disease in all tasks. The patients with PSP had also significantly longer RTs than controls across all tasks. The RTs of the patients with MSA were longer than those with Parkinson's disease, and significantly so in the auditory selective attention task. Inzelberg et aI. (1995) found no differences in RTs between dystonia patients and normals and concluded that motor preparation is normal in this disorder. The latter conclusion was supported by the results of Jahanshahi et aI. (2001) who examined SRT, uncued and precued CRTs in 12 patients with dystonia and 12 age-matched controls and suggested that while response initiation and execution are significantly slower in patients with dystonia than normals, movement preparation is not quantitatively or qualitatively different. Curra et aI. (2000) used kinematic analysis and compared self-initiated and externally-triggered sequences of arm movements in patients with generalized or focal dystonia relative to age-matched controls. From the results it was concluded that "patients with dystonia have a general impairment of sequential movements" and that dystonia impairs internal cueing more than external cueing of movements. In summary, there is evidence that deficits in RT are present in other forms of striatal pathology and not just in Parkinson's disease and that the slowing of RT may be even more prominent in other movement disorders than in idiopathic Parkinson's disease. Most of the existing literature has addressed the question of "normality" of RTs in Parkinson's disease, by comparing the performance of the patients with that of matched normals on various RT tasks. With regard to interpreting the significance of the RT deficits found in Parkinson's disease, a major issue is whether such impairments are "specific" to patients with this disorder. 15.10. RT as a measure sensitive to brain injury/ insult The first aspect of the "specificity" question is whether RT deficits are specific to Parkinson's disease or are also present in other disorders involving striatal pathology. Given that a role in

M. JAHANSHAHI

motor preparation has been attributed to the lateral cerebellum (Allen and Tasakuhara, 1978) and the SMA (Evarts and Wise, 1984) as well as the basal ganglia (Marsden, 1982), the second aspect of the "specificity" question is whether the impairment of RT is specific to disorders of the basal ganglia or is also a feature of other movement disorders involving damage to the cerebellum or SMA. Finally, given that RTs measure a host of cognitive and motor processes that precede the production of a response to a stimulus, it is also possible that brain injury of any form affects RTs. Relatively few comparative studies of RTIMT in Parkinson's disease relative to other disorders involving striatal pathology such as Huntington's disease (Girotti et al., 1988; Halsband et al., 1990; Jahanshahi et al., 1993a) or Progressive Supranuclear Palsy (PSP) (Viallet et al., 1987b; Dubois et al., 1988) or extra-striate pathology (e.g. cerebellar disease; Nakamura and Taniguchi; 1980; Jahanshahi et al., 1993a) have been carried out. In one such comparative study, Nakamura and Taniguchi (1980) studied premotor SRT following a warning signal in 13 patients with Parkinson's disease, 10 with cerebellar degeneration and 14 controls. The RTs of the patients with Parkinson's disease were not different from those of controls. The patients with cerebellar disease, however, had RTs which were significantly slower than those of the Parkinson's disease sufferers and the controls. Verfaellie and Heilman (1987) examined preprogramming of motor responses in two patients with unilateral lesions of the medial frontal lobe. The patient with a left-sided lesion used the advance information about the required response to speed up RT, whereas the case with a right-sided lesion did not. On the basis of the results a role in motor preparation was attributed to the medial frontal lobe which includes the SMA, and it was further suggested that the right hemisphere may have a dominant role in mediating these processes. In another comparative study, Jahanshahi et aI. (1993a), the SRT and CRTs of eight patients with Parkinson's disease tested after withdrawal of dopaminergic medication, eight non-demented patients with Huntington's disease and nine patients with cerebellar disease were compared. The SRTs of the patients with Huntington's disease were slower than those of the other two groups, and the patients with cerebellar disease had the slowest CRTs. However, none of the group differences were

REACTION TIME AS AN INDEX OF MOTOR PREPARATIONIPROGRAMMING

significant. The CRTs of all patients were speeded up by presentation of a warning signal prior to the imperative stimulus. In a precued CRT task, advance movement parameter information was used to speed up RTs relative to an uncued condition by patients in all three groups. Thus, behavioral slowness as indexed by RT was observed in all three patient groups and there were no major differences between the groups with regard to the beneficial effects of a warning signal or in terms of the use of advance information. Montgomery et aI. (2000) compared the RTs of patients with Parkinson's disease or essential tremor. Both groups of patients had longer RTs and slower movement velocities than the normal controls. From the results of these studies, it is evident that RT deficits are not limited to patients with Parkinson's disease, as they are present in patients with other disorders involving the basal ganglia such as those with Huntington's disease, dystonia, PSP or MSA as well as in patients with movement disorders in which the basal ganglia are spared such as those with cerebellar disease or SMA lesions. Furthermore, from the available evidence it appears that prolongation of SRT and/or CRT is a feature of other neurological illnesses. Significant prolongation of SRT and/or CRT relative to normals has been reported for patients with epilepsy (Bruhn and Parsons, 1977), head injury (Miller, 1976; Van Zomeron and Deelman, 1978; MacFlynn et aI., 1984) or Alzheimer's disease (Ferris et aI., 1976; Pirozzolo et aI., 1981; Gordon and Carson, 1990). The impairment of RT with brain damage of any type was illustrated by the results of Elsass and Hartelius (1985) who found that 485 patients with various forms of cerebral dysfunction had prolonged RTs compared to 60 hospitalized controls and that the RT deficit was not influenced by etiology or chronicity of the disorder, although RTs were more impaired by progressive than non-progressive disease. As noted by Milner (1986) it appears that brain damage of any kind results in slower reactions, which is also observed in psychiatric disorders such as depression and schizophrenia (King, 1954; Fuller and Jahanshahi, 1999), as well as in normal ageing (Hicks and Birren, 1970). What would be of value would be the demonstration that specific types of RT such as SRT or CRT are selectively impaired or that the various stages of processing in these tasks are differentially or selec-

221

tively affected by different types of brain damage (Milner, 1986). The demonstration of such differential effects has been attempted in some investigations (e.g. Ferris et aI., 1976; Miller, 1976; Van Zomeren and Deelman, 1978; Pirozzolo et al., 1981). The general picture that emerges from this evidence is that RT is a measure sensitive to injury! insult to various cortical and subcortical regions of the brain, which has been particularly of value in quantifying the motor deficits of patients with movement disorders.

15.11. Conclusions Regardless of whether a serial or parallel model of information processing is adopted, RT paradigms provide behavioral data which reflect the total time taken for the various processes which intervene between presentation of a stimulus and execution of a response. On the basis of such RT data it is not possible to differentiate the time taken up by each of the processes intervening between the stimulus and response, such as those depicted in Figure 1. In fact, in 1938, Woodworth stated that "Since we cannot break up the reaction time into successive acts and obtain the time of each act, of what use is the reaction time?" (p. 83). Nevertheless, the major advantage of RTs is that they are non-invasive techniques for measuring the speed of processing. As noted by Jahanshahi et aI. (1993b), examination of the processes involved in the preparation and execution of movement can be greatly enhanced through the use of other techniques to supplement RT data. Three techniques that are particularly relevant for investigating the brain structures involved in motor preparation and movement initiation and execution and the time course of their activation are recording of event-related potentials (ERPs) both cognitive and motor, TMS, and functional imaging using PET or fMRI. Adoption of a multidimensional approach to the study of motor preparation, initiation and execution through the simultaneous use of different techniques, for example, RTs and ERPs or TMS and imaging; would overcome some of the shortcomings of the individual procedures when used in isolation.

References Agostino, R, Berardelli, A, Formica, A, Stocchi, F, Accomero, N and Manfredi, M (1994) Analysis of

222 repetitive and nonrepetititve sequential arm movements in patients with Parkinson's disease. Mov. Disord., 9: 311-314. Alexander, GE (1987) Selective neuronal discharge in monkey putamen reflects intended direction of planned limb movements. Exp. Brain Res., 67: 623-634. Alexander, GE, DeLong, MR and Strick, PL (1986) Parallel organization of functionally segregated circuits linking basal ganglia and cortex. Ann. Rev. Neurosci., 9: 337-381. Alivisatos, B and Milner, B (1989) Effects of frontal or temporal lobectomy on the use of advance information in a choice reaction time task. Neuropsychologia, 27: 495-503. Allen, GI and Tsukahara, N (1974) Cerebrocerebellar communication systems. Physiolog. Rev., 54: 9571006. Balfour, CA, Clark, CR and Geffen, GM (1991) Interhemispheric differences in the planning and execution of sequences of skilled finger movements. Neuropsychalgia, 29: 889-905. Barbeau, A (1966) The problem of measurement of akinesia. J. Neurosurg., 24: 331-338. Benecke, R, Rothwell, JC, Dick, JP, Day, BL and Marsden, CD (1986) Performance of simultaneous movements in patients with Parkinson's disease. Brain, 109: 739-757. Benecke, R, Rothwell, JC, Dick, JP, Day, BL and Marsden, CD (1987) Disturbance of sequential movements in patients with Parkinson's disease. Brain, 1l0: 361-379. Beradelli, A, Benecke, R, Manfredi, M, Marsden, CD. Academic Press: London, pp. 181-190. Berardelli, A, Dick, JP, Rothwell, JC, Day, BL and Marsden CD (1986) Scaling of the size of the first agonist EMG burst during rapid wrist movements in patients with Parkinson's disease. J. Neurol. Neurosurg. Psychiatry, 49: 1273-1279. Bertelson, P (1961) Sequential redundancy and speed in a serial two-choice responding task. Quart. J. Exper. Psychology, 13: 90-102. Berry, EL, Nicholson, RI, Foster, JK, Behrmann, M and Sagar, HJ (1999) Slowing of reaction time in Parkinson's disease: the involvement of the frontal lobes. Neuropsychologia, 37: 787-795. Bloxham, CA, Mindell, TA and Frith, CD (1984) Initiation and execution of predictable and unpredictable movements in Parkinson's disease. Brain, 107: 371-384. Bloxham, CA, Dick, OJ and Moore, M (1987) Reaction times and attention in Parkinson's disease. J. Neurol., Neurosurg. Psychiatry, 50: 1178-1183. Boff, KR and Lincoln, JE (1988) Engineering data compendium. Human Perception and Performance. AAMRL, Wright-Patterson, AFB, Ohio.

M. JAHANSHAHI

Borland, RG and Nicholson, AN (1975) Comparison of the residual effects of two benzodiazepines (nitrazepam and flurazepam hydrochloride) and pentobarbitone sodium on human performance. Br. J. Clin. Pharmacol., 2: 9-17. Boschert, J and Deecke, L (1986) Cerebral potentials precding voluntary toe, knee and hip movements and their vectors in human precentral gyrus. Brain Res., 376: 175-179. Broadbent, DE and Gregory, M (1965) On the interaction of S-R compatibility with other variables affecting reaction time. Br. J. Psychology, 56: 61-67. Brown, RG and Jahanshahi, M (1996) Cognitive-motor function in Parkinson's disease. Eur. J. Neurology, 36 (Suppl. 1): 24-31. Brown, RG, Jahanshahi, M and Marsden, CD (1993) The execution of bimanual movements in patients with neurological disease. J. Neurol. Neurosurg. Psychiatry, 56: 295-297. Brown, RG, Jahanshahi, M and Marsden, CD (1993) Response choice in Parkinson's disease. Brain, 116: 869-885. Brown, V, Zeffiro, TA, Bowman, EM, Robinson, DL and Hallett, M (1991) Task-specific movement initiation impairments in Parkinson's disease. Neurology, 41 (Suppl. 1): 191. Bruhn, P and Parsons, OA (1977) Reaction time variability in epileptic and brain-damaged patients. Cortex, 13: 373-384. Brumlik, J and Boshes, B (1966) The mechanism of bradykinesia in parkinsonism. Neurology, 16: 337344. Castiello, U and Bennett, KMB (1997) The bilateral reach-to-grasp movement of Parkinson's disease sugjects. Brain, 120: 593-604. Coles, MGH, Gratton, G, Bashore, TR, Eriksen, CW and Donchin, E (1985) A psychophysiological investigation of the continuous flow model of human information processing. J. Exper. Psychology: Hum. Percept. Perform., ll: 529-552. Coles, MGH and Gratton, G (1986) Cognitive psychophysiology and the study of states and processes. In: GRJ Hockey, AWK Gaillard and MGH Coles (Eds.), Energetics and Human Information Processing. Dordrecht, The Netherlands: Martinus Nijhof, pp. 409424. Cooke, JD, Brown, JD and Brooks, VB (1978) Increased dependence on visual information for movement control in patients with Parkinson's disease. Can. J. Neurolog. sa; 5: 413-415. Coull, IT and Nobre, AC (1998) Where and when to pay attention: the neural systems for directing attention to spatial locations and to time intervals as revealed by both P and fMRI. J. Neuroscience, 18: 7426-7435.

REACTION TIME AS AN INDEX OF MOTOR PREPARATION/PROGRAMMING

Crammond, DJ and Kalaska, JF (1989) Neuronal activity in primate pariental cortex area 5 varies with intended movement direction during an instructed-delay period. Exper. Brain Res., 76: 458-462. Curra, A, Berardelli, A, Agostino, R, Giovannelli, M, Koch, G and Manfredi, M (2000) Movement cueing and motor execution in patients with dystonia: a kinematic study. Mov. Disord., 15: 103-112. Daum, I and Quinn, N (1991) Reaction times and visuospatial processing in Parkinson's disease. J. Clin. Exper. Neuropsychol., 13: 972-982. Day, BL, Dick, JP and Marsden, CD (1984) Patients with Parkinson's disease can employ a predicitive motor strategy. J. Neurol. Neurosurg. Psychiatry, 47: 12991306. Day, BL, Rothwell, IC, Thompson, PD, Maertens de Noordhout, A, Nakashima, K, Shannon, K and Marsden, CD (1989) Delay in the execution of voluntary movement by electrical or magnetic brain stimulation in intact man. Evidence for the storage of motor programs in the brain. Brain, 112: 649-663. Deiber, MP, Ibanez, V, Sadato, N and Hallett, M (1996) Cerebral structures participating in motor preparation in humans: a positron emission tomography study. J. Neurophysiol., 75: 233-247. De Jong, R, Wierda, M, Mulder, G and Mulder, LIM (1988) Use of partial stimulus in response processing. J. Exper. Psychol.: Hum. Percept. Perform., 14: 682692. D'Esposito, M, Ballard, D, Zarahn, E and Aguirre, GK (1999) The role of prefrontal cortex in sensory memory and motor preparation: An event-related fMRI study. Neurolmage, II: 4(}()-408. Dick, JPR, Cantello, R, Buruma, 0, Gioux, M, Benecke, R, Day, BL, Rothwell, IC, Thompson, PD and Marsden, CD (1987) The Bereitshaftspotential, LDOPA and Parkinson's disease. EEG and Clin. Neurophysiol., 66: 233-244. Dick, IPR, Rothwell, JC, Day, BL, Cantello, R, Buruma, 0, Gioux, M, Benecke, R, Beradelli, A, Thompson, PD and Marsden, CD (1989) The Bereitshaftspotential is abnormal in Parkinson's disease. Brain, 112: 233-244. Deiber, MP, Passingham, RE, Colebatch, JG, Friston, KJ, Nixon, PD and Frackowiak, RS (1991) Cortical areas and the selection of movement: a study with positron emission tomography. Exper. Brain Res., 84: 393-402. Donders, FC (1969) On the speed of mental processes. In: W.G. Koster (Ed.), Attention and Performance II. Amsterdam: North-Holland, pp. 412-431. Dubinsky, RM, Hallett, M and Bromfield, E (1989) Regional brain metabolic changes with simple hand movements in humans. Neurology, 39 (Suppl. I): 165.

223

Dubois, B, Pillon, B, Legault, F, Agid, Y and Lherrnite, F (1988) Slowing of cognitive processing in progressive supranuclear palsy. Arch. Neurol., 45: 1194-1199. Dubois, B, Danze, F, Pillon, B, Cusiman, G, Lherrnite, F and Agid, Y (1987) Cohinergic-dependent cognitive deficits in Parkinson's disease. Ann. Neurology, 22: 26-30. Dunne, MP and Hartley, LR(1985) The effect of scopolamine upon verbal memory: evidence for an attentional hypothesis. Acta Psychologica, 58: 205-217. Elsass, P and Hartelius, H (1985) Reaction time and brain disease: relations to location, etiology and progression of cerebral dysfunction. Acta Neurologia Scandanavia, 17: ll-19. Eriksen, CW and Schultz, DW (1979) Information processing in visual search: A continuous flow conception and experimental results. Percept. Psychophys., 25: 249-263. Evarts, EV and Tanji, J (1974) Gating of motor cortex reflexes by prior instruction. Brain Res., 71: 479-494. Evarts, EV and Wise, SP (1984) Basal ganglia outputs and motor control. In: D Evered and M O'Connor (Eds.), Functions of the Basal Ganglia. CIBA Foundation Symposium, Vol. 107, pp. 83-96. Evarts, EV, Teravainen, H and Calne, DB (1981) Reaction time in Parkinson's disease. Brain, 104: 167-186. Ferris, S, Crook, T, Sathananthan, G and Gershon, S (1976) Reaction time as a diagnostic measure in senility. J. Amer. Geriat. Soc., 24: 529-533. Fitts, PM (1954) The information capacity of the human motor system in controlling the amplitude of movement. J. Exp. Psychol., 47: 381-391. Fitts, PM (1966) Cognitive aspects of information processing. III. Set for speed vs. accuracy. J. Exp. Psychol., 71: 849-857. Fitts, PM and Seeger, CM (1953) S-R compatibility: spatial characteristics of stimulus and response codes. J. Exper. Psychol., 46: 199-210. Fitts. PM and Deininger. RL (1954) S-R compatibility: correspondence among paired elements within stimulus and response codes. J. Exper. Psychol., 48: 483-492. Fitts, PM, Peterson, JR and Wolpe, G (1963) Cognitive aspects of information processing. II. Adjustments to stimulus redundancy. J. Exp. Psychol., 65: 423-432. Flowers, K (1975) Ballistic and corrective movements in an aiming task: intention tremor and parkinsonian movement disorders compared. Neurology, 25: 413421. Franks, 1M and Van Donkelaar, P (1990) The effects of demanding temporal accuracy on the programming of simple tapping sequences. Acta Psychologica, 74: 1-14.

224 Frith, CD, Bloxham, CA and Carpenter, KN (1986) Impairments in the leaming and performance of a new manual skill in patients with Parkinson's disease. J. Neurol. Neurosurg. Psychiat., 49: 661-668. Frith, CD and Done, DJ (1986) Routes to action in reaction time tasks. Psycholog. Res., 48: 169-177. Fuller, R and Jahanshahi, M (1999) Impairment of willed actions and use of advance information for movement preparation in schizophrenia. J. Neurol. Neurosurg. Psychiatry, 66: 502-509. Fuster, JM and Alexander, GE (1971) Neuron activity related to short-term memory. Science, 173: 652-654. Garcia-Colera, A and Semjen, A (1987) The organization of rapid movement sequences as a function of sequence length. Acta Psychologica, 66: 237-250. Gauntlett-Gilbert, J and Brown, VJ (1998) Reaction time deficits and parkinson's disease. Neurosci. Biobehav. Rev., 22: 865-881. Georgiou, N, Iansek, R, Bradshaw, JL, Philips, JG, Mattingley, JB and Bradshaw, JA (1993) An evaluation of the role of internal cues in the pathogenesis of parkinsonian hypokinesia. Brain, 116: 1575-1587. Girotti, F, Carella, F, Grassi, MP, Soliveri, P, Marano, R and Caraceni, T (1986) Motor and cognitive performances of parkinsonian patients in the on and off phases of the disease. J. Neurol., Neurosurg. Psychiatry, 49: 657-660. Girotti, F, Marano, R, Soliveri, P, Geminiani, G and Scigliano, G (1988) Relationship between motor and cognitive disorders in Huntington's disease. J. Neurol., 236: 454-457. Godefroy, 0 and Rousseaux, M (1996) Divided and focused attention in patients with lesion of the prefrontal cortex. Brain Cognition, 30: 155-174. Godefroy, 0, Lhullier, C and Rousseaux, M (1996) Nonspatial attention disorders in patients with frontal or posterior brain damage. Brain, 119: 191-202. Goodrich, S, Henderson, L and Kennard, C (1989) On the existence of an attention-demanding process peculiar to simple reaction time: Converging evidence from Parkinson's disease. Cogn. Neuropsychol., 6: 309331. Gordon, B and Carson, K (1990) The basis for choice reaction time slowing in Alzheimer's disease. Brain Cogn., 13: 148-166. Granjon, MJ, Requin, J, Dump, H and Reynard, G (1973) Effects of a timing signal on reaction time with 'nonaging' foreperiod. J. Exper. Psychol., 101: 139-145. Gratton, G, Coles, MGH Sirevaag, EJ, Eriksen, CW and Donchin, E (1988) Pre- and post-stimulus activation of response channels: A psychophysiological analysis. J. Exper. Psychol.: Hum. Percept. Perform., 14: 331344.

M. JAHANSHAHI

Gratton, G, Bosco, CM, Kramer, AF, Coles, MGH, Wickens, CD and Donchin, E (1990) Event-related potentials as indices of information extraction and response priming. EEG Clin. Neurophysiol., 75: 419-432. Grice, GR, Nullmey, R and Spiker, VA (1977) Application of variable criterion theory to choice reaction time. Percept. Psychophys., 22, 431-449. Grice GR, Nullmeyer R, Spiker VA (1982) Human reaction times: Toward a general theory. J. Exper. Psycho/.: General, Ill: 135-153. Hackley, SA and Miller, J (1995) Response complexity and precue interval effects on the lateralized readiness potential. Psychophysiology, 32: 230-241. Hadland, KA, Rushworth, MFS, Passingham, RE, Jahanshahi, M and Rothwell, JC (2001) Interference with performance of a response selection task that has no working memory component: An rTMS comparison of dorsolateral prefrontal cortex and medial frontal cortex. J. Cogn. Neurosci., 13: 1097-1108. Hale, DJ (1967) Sequential effects in a two-choice serial reaction task. Quart. J. Exper. Psychol., 19: 133-141. Hallett, M, Shahani, BT and Young, RR (1977) Analysis of stereotyped voluntary movements at the elbow in patients with Parkinson's disease. J. Neurol. Neurosurg. Psychiat., 40, 1129-1135. Hallett, M and Khoshbin, S (1980) A physiological mechanism of bradykinesia. Brain, 103: 301-314. Halsband, U, Homberg, V and Lange, HJ (1990) Slowing of different types of voluntary movement in extrapyramidal disease: Fitt's law and idiographic writing. In: Motor Disturbances II. Harrington, DL and Haadland, KY (1987) Programming sequences of hand postures. J. Motor Behav., 19: 77-95. Hasbroucq, T, Guiard, Y and Ottomani, L (1990) Principles of response determination: the list-rule model of SR compatibility. Bull. Psychonom. Soc., 28: 327330. Hasbroucq, T, Kaneko, H Akamatsu, M and Possama, C-A (1997) Preparatory inhibition of cortico-spinal excitability: a transcranial magnetic stimulation study in man. Cogn. Brain Res., 5: 185-192. Harrington, DL and Haaland, KY (1991) Sequencing in Parkinson's disease. Abnormalities in programming and controlling movement. Brain, 114 (Pt. lA): 99115. Harrington, DL, Haaland, KY, Yeo, RA and Marder, E (1990) Procedural memory in Parkinson's disease: impaired motor but not visuoperceptual learning. J. Clin. Exp. Neuropsychol., 12(2): 323-39. Harrison, J, Henderson, L and Kennard, C (1995) Abnormal refractoriness in patients with Parkinson's

REACTION TIME AS AN INDEX OF MOTOR PREPARATION/PROGRAMMING

disease after brief withdrawal of levodopa treatment. J. Neurol. Neurosurg. Psychiat., 59(5): 499-506. Hawkins, HL and Underhill, JR (1971) S-R compatibility

and the relative frequency effect in choice reaction time. J. Exp. Psychol., 96: 334-337. Heathcott, A, Popiel, SJ and Mewnhort, DJK (1991) Analysis of response time distributions: an example using the stroop task. Psycholog. Bull., 109: 340-347. Heilman, KM, Bowers, D, Watson, RT and Greer, M (1976) Reaction time in Parkinson's disease. Arch. Neurol., 33: 139-140. Henderson, L, Harrison, J and Kennard, C (2001) Selectively impaired reaction time in Parkinson's disease: persistent absence of simple reaction advantage in a patient with frontal complications. Neurocase, 4: 319-330. Henry, FM and Rogers, DE (1960) Increased response latency for complicated movements and a "memory drum" theory of neuromotor control. Res. Quart. Exer. Sport, 31: 448-458. Hick, WE (1952) On the rate of gain of information. Quart. J. Exper. Psycho I., 4: 11-26. Hicks, LH and Birren, JE (1970) Ageing, brain damage, and psychomotor slowing. Psychol. Bull., 74: 27-396. Hietanen, M and Teravainen, H (1986) Cognitive performance in early Parkinson's disease. Acta Neurol. Scand., 73: 151-159. Hohle, RH (1965) Inferred components of reaction times as functions of foreperiod duration. J. Exper. Psychol., 69: 382-386. Hoshiyama, M, Koyama, S, Kitamura, Y, Shimojo, M, Watanabe, S and Kakigi, R (1996) Effects of judgement process on motor evoked potentials in GolNo-go hand movement task. Neurosci. Res., 24: 427-430. Hoshiyama, M, Kakigi, R, Koyama, S, Takeshima, Y, Watanabe, Sand Shimojo, M (1997) Temporal changes of pyramidal tract activities after decision of movement: a study using transcranial magnetic stimulation of the motor cortex in humans. EEG Clin. Neurophysiol., 105: 255-261. Hulstijn, Wand Van Galen, GP (1983) Programming in handwriting: Reaction time and movement time as a function of sequence length. Acta Psychologica, 54: 23-49. Humberstone, M, Guy, V, Sawle, DM, Clare, S, Hykin, J, Coxon, R, Bowtell, R, Macdonald, IA and Morris, PG (1997) Functional magnetic resonance imaging of single motor events reveals human presupplementary motor area. Ann. Neurol., 42: 632-637. Hyman, R (1953) Stimulus information as a determinant of reaction time. J. Exper. Psychol., 45: 188-196. Iacoboni, M, Woods, RP and Mazziotta, IC (1996) Brainbehavior relationships: evidence from practice effects

225

in spatial stimulus-response compatibility. J. Neurophysiol., 76: 321-331. Ikeda, A, Luders, HO, Collura, TF, Burgess, RC, Morris, HH, Hamano, T and Shibasaki, H (1996) Subdural potentials at orbitofrontal and mesial prefrontal areas accompanying anticipation and descision making in humans: a comparison with beretischaftspotential. Electroencephalogr. Clin. Neurophysiol., 98: 206-212. Inzelberg, R, Flash, T, Schechtman, E and Korczyn, AD (1995) Kinematic properties of upper limb trajectories in idiopathic torsion dystonia. J. Neurol. Neurosurg. Psychiat., 58: 312-319. Jahanshahi, M, Brown, RG and Marsden, CD (l992a) Simple and choice reaction time and the use of advance information for motor preparation in Parkinson's disease. Brain, 115: 539-564. Jahanshahi, M, Brown, RG and Marsden, CD (l992b) The effect of withdrawal of dopaminergic medication on simple and choice reaction time and the use of advance information in Parkinson's disease. J. Neurol. Neurosurg. Psychiat., 55: Il 69-11 76. Jahanshahi, M, Brown, RG and Marsden, CD (1993a) A comparative study of simple and choice reaction time in Parkinson's, Huntington's and cerebellar disease. J. Neurol. Neurosurg. Psychiat., 56: 1169-1177. Jahanshahi, M, Brown, RG and Marsden, CD (l993b) Motor Slowness in Parkinson's disease. In: GE Stelmach and V Homberg (Eds.), Sensorimotor Impairment in the Elderly: Are They Reversible? Kluwer Academic: The Netherlands, pp. 265-291. Jahanshahi, M, Jenkins, HI, Brown, RG, Marsden, CD, Passingham, RE and Brooks, CD (1995) Self-initiated vs. externally-triggered movements: I. An investigation using regional cerebral blood flow and movementrelated potentials in normals and in patients with Parkinson's disease. Brain, 118: 913-933. Jahanshahi, M, Rowe, J and Fuller, R (2001) Impairment of movement initiation and execution but not preparation in idiopathic dystonia. Exp. Brain Res., 140: 460-468. Jenkins, HI, Jahanshahi, M, Jueptner, M, Passingham, RE and Brooks, CD (2000) Self-initiated vs. externallytriggered movements: II. The effects of stimulus predictability studied with PET. Brain, 123: 12161228. Jordan, N, Sagar, HJ and Cooper, JA (1992) Cognitive components of reaction time in Parkinson's disease. J. Neurol. Neurosurg. Psychiatry, 55: 658-664. Keele, SW (1986) Motor control. In KR Boff, L Kaufman and JP Thomas (Eds.), Handbook of Perception and Human Perception, Volume II, Cognitive Processes and Performance (Chapter 30). Wiley, New York.

226

M. JAHANSHAHI

Keel, SW and Posner, ML (1968) Processing of visual feedback in rapid movements. 1. Exper. Psychol., 77: 155-158. Kerr, R, Dall, JLC, Grimes, JD and Teaffe, M (1988) The response to predictable and unpredictable movements by individuals with Parkinson's disease. J. Hum. Mov. Studies., 14: 85-92. King, HE (1954) Psychomotor aspects of mental disease. Harvard University Press: Cambridge. King, HE (1959) Defective psychomotor movement in Parkinson's disease: exploratory observations. Percept. Motor Skills, 9: 326. Klapp, ST (1976) Short-term memory as a response preparation state. Mem. Cogn., 4: 721-729. Klapp, ST (1977) Reaction time analysis of programmed control. Exer. Sport Sci., 5: 231-253. Klapp, ST, Abbott, J, Coffman, K, Greim, D, Sneider, R and Young, F (1979) Simple and choice reaction time methods in the study of motor programming. J. Motor Behav., II: 91-101. Klockgether, T and Dichgans, J (1994) Visual control of arm movement in Parkinson's disease. Mov. Disord., 9: 48-56. Konishi, S, Nakajima, K, Uchida, I, Sekihara, K and Miyashita, Y (1998) No-go dominant brain activity in human inferior prefrontal cortex revealed by functional magnetic resonance imaging. Eur. J. Neurosci., 10: 1209-1213. Kornhuber, HH and Deecke, L (1965) Himpotentialanderungen bei Wilkburbewegugen und passiven Bewegugungen den Menschen: Beritschaftspotential und reafferente Potentiale. Pfiugers Archiv., 284:

1-17. Kutukcu, Y, Marks, WJ, Jr., Goodin, D-S and Aminoff, MJ (1999) Simple and choice reaction time in Parkinson's disease. Brain Res., 815(2): 367-72. Labutta, RJ, Miles, RB, Sanes, IN and Hallett, M (1994) Motor program memory storage in Parkinson's disease patients tested with a delayed response task. Mov. Disord., 9: 218-222. Leimkuhler, ME and Mesulam, MM (1985) Reversible go-no go deficits in a case of frontal lobe tumor. Ann. Neurol., 18: 617-619. Leocani, L, Cohen, L-G, Wassermann, E-M, Ikoma, K and Hallett, M (2000) Human corticospinal excitability evaluated with transcranial magnetic stimulation during different reaction time paradigms. Brain, 123: 1161-1173. Lichter, DG, Corbett, AJ, Fitzgibbon, GM, Davidson, OR, Hope, A, Goddard, GV, Sharples, KJ and Pollock, M (1988) Cognitive and motor dysfunction in Parkinson's disease. Clinical.. performance, and computed tomographic correlations. Arch. Neurol., 45: 854-860.

Low, KA and Miller, J (l999)The usefulness of partial information: Effects of probability in the choice/no go task. Psychophysiology, 36: 288-297. MacFlynn, G, Montgomery, EA, Fenton, GW and Rutherford, W (1984) Measurement of reaction time following minor head injury. J. Neurol., Neurosurg. Psychiatry, 47: 1326-1331. Marsden, CD (1982) The mysterious motor function of the basal ganglia: The Robert Wartenberg Lecture. Neurology, 32: 514-539. Martin, J (1967) The Basal Ganglia and Posture. London: Pitman Medical, pp. 33-35. Mayeux, R, Stem, Y, Sano, M, Cote, L and Williams, JBW (1987) Clinical and biochemical correlates of bradyphrenia in Parkinson's disease. Neurology, 37: 1130-1134. Maylor, EA and Rabbitt, PM (1993) Alcohol, reaction time and memory: a meta-analysis. Br. J. Psycho I., 84: 301-317. McCarthy, G and Donchin, E (1981) A metric for thought: A comparison of P300 latency and reaction time. Science, 211. McClelland, JL (1979) On the time relations of mental processes: An examination of systems of processes in cascade. Psycholog. Rev., 86: 287-330. Miller, E (1976) Simple and choice reaction time following severe head injury. Cortex, 6: 121-127. Miller, E, Berrios, GE and Plitynska, B (1987) The adverse effect of benzhexol on memory in Parkinson's disease. Acta Neurologia Scandanavia, 76: 278-282. Miller, J (1982) Discrete vs. continuous stage models of human information processing: In search of partial output. J. Exper. Psycho/.: Human Percept. Perform., 8: 273-296. Miller, J and Pachella, RG (1973) Locus of the stimulus probability effect. J. Exp. Psychol., 101: 227-231. Milner, AD (1986) Chronometric analysis in neuropsychology. Neuropsychologia, 24: 115-128. Montgomery, EB and Nuessen, J (1990) The movement speed/accuracy operator in Parkinson's disease. Neurology, 40: 269-272. Montgomery, EB, Baker, KB, Lyons, K and Koller, WC (2000) Motor initiation and execution in essential tremor and Parkinson's disease. Mov. Disord., 15: 511-515. Mowbray, G and Rhoades, M (1959) On the reduction of choice reaction times with practice. Quart. J. Exper. Psychol., 43: 169-194. Mushiake, H and Strick, PL (1995) Cerebellar and pallidal activity during instructed delay periods. Soc. Neurosci. Abstr., 21: 411. Naito, E, Kinomura, S, Geyer, S, Kawashima, R, Roland, PE and Zilles, K (2000) Fast reaction to different sensory modalities activates common fields in the

REACTION TIME AS AN INDEX OF MOTOR PREPARATIONIPROGRAMMING

motor areas, but the anterior cingulated cortex is involved in the speed of reaction. 1. Nurophysiol., 83: 1701-1709. Nakamura, R and Taniguchi, R (1980) Dependence of reaction times on movement patterns in patients with Parkinson's disease and those with cerebellar degeneration. Tohoku 1. Exper. Med., 132: 153-158. Osman A, Bashore TR, Coles MGH, Donchin E, Meyer DE (1992) On the transmission of partial information: inferences from movement-related brain potentials. J. Exper. Psychol.: Hum. Percept. Perform., 18: 217232. Peterson, BS, Kane, MJ, Alexander, GM, Lacadie, C, Skudlarski, P, Leung, HC, May, J and Gore, JC (2002) An event-related functional MRI study comparing interference effects in the Simon and Stroop task. Cogn. Brain Res., 13: 427-440. Pirozzolo, FJ, Christensen, KJ, Ogle, KM, Hansch, EC and Thompson, WG (1981) Simple and choice reaction time in dementia: Clinical implications. Neurobiol. Ageing, 2: 113-117. Pirtosek, Z, Jahanshahi, M, Barrett, G and Lees, A (2001) Attention and cognition in bradykinetic-rigid syndromes: an event-related potential study. Ann. Neurol., 50: 567-573. Playford, ED, Jenkins, IH, Passingham, RE, Nutt, J, Frackowiak, RSJ and Brooks, DK (1992) Impaired mesial frontal and putamen activation in Parkinson's disease: A PET study. Ann. Neurol., in press. Posner, MI (1980) Orienting of attention. Quart. J. Exp. Psychol., 32: 3-25 .. Posner, MI and Snyder, CRR (1975) Facilitation and inhibition in the processing of signals. In: PMA Rabbitt and S Domic (Eds.), Attention and Performance V. Academic Press: London, pp. 669-682. Posner, MI, Klein, R, Simmers, J and Buggies, S (1973) On the selection of signals. Mem. Cogn., 1: 2-12. Postman, L and Egan, J (1949) Experimental Psychology: An Introduction. Harper, New York. Praamstra, P, Kleine, BU and Schnitzler, A (1999) Magnetic stimulation of the dorsal premotor cortex modulates the Simo effect. NeuroReport, 10: 36713674. Pullman, SL, Watts, RL, Juncos, JL, Chase, TN and Sanes, IN (1988) Dopaminergic effects on simple and choice reaction time performance in Parkinson's disease. Neurology, 38: 249-254. Pullman, SL, Watts, RL, Juncos, JL and Sanes, IN (1990) Movement amplitude choice reaction time performance in Parkinson's disease may be independent of dopaminergic status. 1. Neurol. Neurosurg. Psychiatry., 53: 279-283.

227

Quintana, J and Fuster, JM (1993) Spatial and temporal factors in the role of prefrontal and pariental cortex in visuomotor integration. Cereb. Cortex, 3: 122-132. Rabbitt, PMA (1966) Errors and error correction in choice-response tasks. J. Exp. Psychol., 71: 264-272. Rabbitt, PMA (1980) The effects of R-S interval duration on serial choice reaction time: preparation time or response monitoring time? Rafal, RD, Posner, MI, Walker, JA and Friedrich, FJ (1984) Cognition and the basal ganglia. Separating mental and motor components of performance in Parkinson's disease. Brain, 107: 1083-1094. Rafal, RD, Inhoff, AW, Friedman, JH and Bernstein, E (1987) Programming and execution of sequential movements in Parkinsons's disease. J. Neurol. Neurosurg. Psychiat. 50: 1267-1273. Rafal, RD, Friedman, J and Lannon, MC (1989) Preparation of manual movements in hemiparkinsonism. J. Neurol., Neurosurg. Psychiatry, 52: 399-402. Ratcliffe, R (1979) Group reaction time distributions statistics. Psycho log. Bull., 86: 446-461. Reid, WGJ, Broe, GA, Hely, A, Morris, JGL, Williamson, PM, O'Sullivan, DJ, Rail, D, Genge, S and Moss, NG (1989) The neuropsychology of de novo patients with idiopathic Parkinson's disease: the effects of age of onset. Int. J. Neurosci., 48: 205-217. Requin, J (1992) From action representation to movement control. In: GE Stelmach and J Requin (Eds.), Tutorials in Motor Behavior II, North-Holland Publishing Company, pp. 159-179. Rosenbaum, DA (1980) Human movement initiation: specification of arm, direction, and extent. 1. Exper. Psycho/.: General, 109: 444-474. Rubia, K, Russell, T, Overmeyer, S, Brammer, MJ, Bullmore, ET, Sharma, T, Simmons, A, Williams, SCR, Giampietro, V, Andrew, CM and Taylor, E (2001) Mapping motor inhibitions: Conjunctive brain activations across different versions of Go/No-Go and stop tasks. Neurolmage, 13: 250-261. Sanders, AF (1980) Stage analysis of reaction processes. In: GE Stelmach and J Requin (Eds.), Tutorials in Motor Behaviour. Amsterdam: North-Holland Publishing Company, pp. 331-354. Sano, M, Rosen, W, Stem, Y, Rosen, J and Mayeux, R (1995) Simple reaction time as a measure of global attention in Alzheimer's disease. J. Int. Neuropsychol. Soc., I: 56-61. Shankweiler, D (1959) Effects of success and failure instructions on reaction time in patients with brain damage. J. Compar. Psychol., 52: 546-549. Schmid, HGOM, Mulder, G and Mulder, LJM (1987) The continuous flow model revisited: Perceptual and central motor aspects. In: R Johnson, JW Rohrbaugh and R Parasuraman (Eds.), Current Trends in Event-related

228

Potential Research (EEG Supp!. 40). Amsterdam: Elsevier, pp. 270-278. Schmidt, RA (1975) A schema theory of discrete motor skillieaming. Psycholog. Rev., 82: 225-260. Schluter, ND, Krams, M, Rushworth, MFS and Passingham, RE (2001) Cerebral dominance for action in the brain: the selection of actions. Neuropsychologia, 39: 105-113. Sheridan, MR, Flowers, KA and Hurrell, J (1987) Programming and execution of movement in Parkinson's disease. Brain, llO: 1247-1271. Sheridan, MR and Flowers, KA (1990) Movement variability and bradykinesia in Parkinson's disease. Brain, 113: ll49-ll61. Simon, JR (1969) Reactions towards the source of stimulation. J. Exp. Psychol., 81: 174-176. Smid, HGOM, Mulder, G and Mulder, UM (1987) The continuous flow model revisited: Perceptual and central motor aspects. In: R Johnson, Jr., JW Rohrbaugh and R Parasuraman (Eds.), Current Trends in EventRelated Potential Research (EEG Supp!. 40). Amsterdam: Elsevier, pp. 270-278. Smid, HGOM, Lamain, W, Hogeboom, MM, Mulder, G and Mulder, UM (1991) Psychophysiological evidence for continuous information transmission between visual search and response processes. J. Exper. Psychol.: Hum. Percept. Perform., 17: 696-714. Smid, HGOM, Mulder, G, Mulder, UM and Brands, OJ (1992) A psychophysiological study of the use of partial information in stimulus-response translation. J. Exper. Psychol.: Hum. Percept. Perform., 18: llOlll19.. Soliveri, P, Brown, RG, Jahanshahi, M, Caraceni, R and Marsden, CD (1997) Learning manual pursuit tracking skills in patients with Parkinson's disease. Brain, 120: 1325-1337. Starkstein, S, Esteguy, M, Berthier, ML, Garcia, H and Leiguarda, R (1989) Evoked potentials, reaction time and cognitive performance in on and off phases of Parkinson's disease. J. Neurol. Neurosurg. Psychiatry, 52: 338-340. Stelmach, GE and Worringham, CJ (1988) The control of bimanual aiming movements in Parkinson's disease. J. Neurol. Neurosurg. Psychiatry, 51: 223-231. Stelmach, GE, Worringham, CJ and Strand, EA (1986) Movement preparation in Parkinson's disease. The use of advance information. Brain, 109: 1179-ll94. Stelmach, GE, Worringham, CJ and Strand, EA (1987) The programming and execution of movement sequences in Parkinson's disease. Int. J. Neurosci., 36: 55-65. Stem, Y, Mayeux, R and Cote, L (1984) Reaction time and vigilance in Parkinson's disease. Possible role of

M. JAHANSHAHI

altered norepinephrine metabolism. Arch. Neurol., 41: 1086-1089. Sternberg, S (1966) High-speed scanning in human memory. Science, 153: 652-654. Sternberg, S (1969) The discovery of processing stages: Extensions of Donder's method. In: WG Koster (Ed.), Attention and Performance II. Amsterdam: North Holland. pp. 276-315. Sternberg, SS, Monsell, RL and Knoll Wright, CE (1978) The latency and duration of rapid movement sequences: comparisons of speech and typewriting. In: G.E. Stelmach (Ed.), Information Processing in Motor Control and Learning. New York: Academic Press. Suci, OJ, Davidoff, MD and Surwillo, WW (1960) Reaction time as a function of stimulus information and age. J. Exper. Pscyhol., 60: 242-244. Sugiura, M, Kawashima, R, Takahashi, T, Xiao, R, Tsukiura, T, Sato, K, Kawano, K, Iijima, T and Fukuda, H (2001) Different distribution of the activated areas in the dorsal premotor cortex during visual and auditory Neurolmage, 14: ll68raction-time tasks. 1174. Syndulko, K, Gilden, ER and Hansch, EC (1981) Decreased verbal memory associated with anticholinergic treatment in Parkinson's disease patients. Int. J. Neurosci., 14: 61. Talland, GA (1963) Manual skills in Parkinson's disease. Geriatrics, 18: 613-620. Talland, GA and Schwab, RS (1964) Performance with Multiple sets in Parkinson's disease. Neuropsychologia, 2: 45-53. Tanji, J, Taniguichi, K and Saga, T (1980) Supplementary motor area: neuronal response to motor instructions. J. Neurophysiol., 43: 60-68. Teasdale, N and Stelmach, GE (1988) Movement disorders: The importance of the movement context. J. Mot. Behav., 20: 186-191. Techner, WH and Krebs, MJ (1974) Laws of visual choice reaction time. Psycholog. Rev., 81: 75-98. Teravainen, H and CaIne, D (1980) Quantitative assessment of Parkinsonian deficits. In: UK Rinne, M Klinger and G Stamm (Eds.), Parkinson's Disease Current Progress, Problems and Management. North Holland Biomedical Press: Elsevier, pp. 145-164.. Tomberg, C, Levarlet-Joye, H and Desmedt, JE (1991) Reaction times recording methods: reliability and EMG analysis of patterns of motor commands. Electroencephalogr. Clin. Neurophysiol., 81: 269-278. Toni, I, Schluter, ND, Josephs, 0, Friston, K and Passingham, RE (1999) Signal-, Set- and movementrelated activity in the human brain: An event-related fMRI study. Cereb. Cortex, 9: 35--49.

REACTION TIME AS AN INDEX OF MOTOR PREPARATIONIPROGRAMMING

Van Donkelaar, P and Franks, 1M (1991) Preprogramming vs. on-line control in simple movement sequences. Acta Psychol. (Arnst.), 77: 1-19. Van Hilten, JJ, Wagemans, EA, Ghafoerkhan, SF and Van Laar, T (1997) Movement characteristics in Parkinson's disease: determination of dopaminergic responsiveness and threshold. Clin. Neuropharmacol., 20(5): 402-408. Van Hilten, n. Van Eerd, AA, Wagemans, EA, Middelkoop, HA and Roos, RA (1998) Bradykinesia and hypokinesia in Parkinson's disease: what's in a name? J. Neural Transm., 105(2-3): 229-237. Van der Molen, MW, Boomsma, 01, Jennings, JR and Nieuwboer, RT (1989) Does the heart know what the eye sees? A cardiac/pupillometric analysis of motor preparation and response execution. Psychophysiology, 26(1): 70-80. Van Zomeron, AH and Deelman, BG (1978) Long-term recovery of visual reaction time after closed head injury. J. Neurol. Neruosurg. Psychiatry, 41: 452-457. Velasco, F and Velasco, M (1973) A quantitative evaluation of the effects of L-DOPA on Parkinson's disease. Neuropharmachology, 12: 80-99. Verfaellie, M and Heilman, KM (1987) Response preparation and response inhibition after lesions of the medial frontal lobe. Arch. Neurol., 44: 1265-1271. Viallet, F, Masson, R and Khalil, R (l987a) Performance of a bimanual load-lifting task by Parkinsonian patient. J. Neurol., Neurosurg. Psychiatry, 50: 1274-1283. Viallet, F, Trouche, E, Beaubaton, D, Legallet, E and Khalil, R (I 987b) Visual feedback and motor performance in human and animal basal ganglia dysfunction. In: JS Shneider and TI Libby (Eds.), Basal Ganglia and Behavior. Toronto: Huber, pp. 71-82. Warabi, T, Yanagisawa, N and Shindo, R (1988) Changes in strategy of aiming tasks in Parkinson's disease. Brain, Ill: 497-505. Weeks, RA, Honda, M, Catalan, MJ and Hallett, M (2001) Comparison of auditory, somatosensory, and visually

229

instructed and internally generated finger movements: A PET study. Neurolmage, 14: 219-230. Wiesendanger, M, Schneider, P and Villoz, IP (1969) Electromyographic analysis of a rapid volitional movement. Am. J. Phys. Med., 48: 17-24.. Wilkinson, RT and Allison, S (1989) Age and simple reaction time. Decade differences for 5,325 subjects. J. Gerontol.., 44: 29-35. Wilson, SAK (1925) Disorders of motility and of muscle tone with special reference to corpus striatum. Lecture I. Lancet, ii: 1-10.. Wise, SP, Boussaoud, D, Johnson, PB and Caminiti, R (1997) The premotor and parietal cortex: Corticocortical connectivity and combinatorial computations. Annu. Rev. Neurosci., 20: 25-42. Woodworth, RS (1938) Experimental psychology. Holt, Rinehart & Winston, New York. Yamanaka, K, Kimura, T, Miyazaki, M, Kawashima, N, Nozaki, D, Nakazawa, K and Yamamoto, Y (2002) Human cortical activities during Go/NoGo tasks with opposite motor control paradigms. Exper. Brain Res., 142: 301-307. Yangagisawa, N, Fujimot, S and Tamaru, F (1989) Bradykinesia in Parkinson's disease: Disorders of onset and execution of fast movement. Eur. Neurol., 29 (Supp!. 1): 19-28. Yokochi, F, Nakamura, Rand Narabayashi, H (1985) Reaction time of patients with Parkinson's disease with reference to assymetry of neurological signs. J. Neurol., Neurosurg. Psychiatry, 48: 702-705. Zahn, TP and Rosenthal, D (1966) Simple reaction time as a function of the relative frequency of the preparatory interval. J. Exper. Psychol., 72: 15-19. Zappia, M, Montesanti, R, Colao, R and Quattrone, A, (1994) Usefulness of movement time in the assessment of Parkinson's disease. J. Neurol., 241: 543-550.

Movement Disorders Handbook of Clinical Neurophysiology, Vol. 1 M. Hallett (Ed.) © 2003 Elsevier B.Y. All rights reserved

231 CHAPTER 16

Spinal reflexes Mary Kay Floeter* Electromyography Section, National Institutes of Neurologic Disorders and Stroke. National Institutes of Health, Bethesda, MD 20892-1404. USA

16.1. Introduction Reflexes are stereotyped motor actions evoked by specific sensory stimuli. The clinical usefulness of reflexes is well established (Kimura et al., 1994). Reflexes are important in the study of movement disorders for several reasons. First, reflexes play a normal role in adjusting movements in response to changes in sensory feedback. Muscle, joint and skin afferents will each be activated in a quite distinct way during a movement, signaling displacement of individual joints, pressure on particular regions of the skin, or stretch of certain muscles. These multisensory signals feed back to the motor system and reinforce the expected movement. Should perturbations occur as the movement proceeds, the sensory signal relayed back to the nervous system will evoke a reflex response that changes the motor output. Abnormal reflex responses are a characteristic of many movement disorders. Reflexes are also important in the study of movement disorders because they can provide a non-invasive means for assessing the central nervous system and exploring changes in the strength of the connections of spinal intemeurons. Simple reflexes, such as the stretch reflex, can be used as a tool to assess the excitability of motor neurons in patients with movement disorders. Spinal reflexes, defined as those whose circuitry lies within the spinal cord, can be categorized as either excitatory or inhibitory. Excitatory spinal reflexes produce movement or activation of motor neurons. Inhibitory reflexes are those in which the

* Correspondence to: Dr. Mary Kay Floeter, EMG Section, NINDS, NIH, 10 Center Drive, MSC 1404, Bethesda, MD 20892-1404, USA. E-mail address: [email protected] Tel.: + 1-(301)-496-7428; fax: + 1-(301)-402-8796.

sensory stimulus either terminates ongoing motor activity or suppresses a motor response. Inhibitory reflexes can produce either postsynaptic inhibition of motor neurons, which will reduce the motor neuron's excitability to all inputs, or pre-motor inhibition, which reduces the response to either a selected set of afferents (presynaptic inhibition) or to the intemeurons that relay inputs to motor neurons. Most spinal reflexes involve more than one muscle. Generally speaking, reflexes will affect muscles that have synergistic actions in the same manner. But reflexes can be complex, and in some, such as the flexor reflex, one group of muscles may be excited (the flexors) whereas another group of muscles (the extensors) is inhibited.

16.2. Stretch reflexes: T-waves and H-reflexes The stretch reflex, assessed clinically by tapping on the muscle tendon of a relaxed, slightly stretched muscle, is the simplest spinal reflex. The tap causes a brief stretch which activates the muscle spindle: the spindle's primary (la) afferents respond with a burst of spikes that may be repeated as the muscle oscillates until stopping (Burke et al., 1983). Ia afferents synapse monosynaptically on homonymous motor neurons and, less potently, on heteronymous motor neurons such as those that innervate synergist muscles. Ia afferents have later effects on motor neurons through non-monosynaptic pathways. The latency of a stretch reflex can be measured using a tendon hammer with a contact-activated switch to trigger a sweep of surface EMG activity from the muscle being stretched. The response, called the Twave has relatively constant latency upon repeated tapping, but the amplitude will vary, reflecting changes in the number of motor neurons responding with each tap. The latency of the T-wave is

232

M.K. FLOETER

proportional to the proximal-distal location of the muscle. T-waves can be elicited from any superficial muscle in which a tendon reflex can be obtained (Fig. 1). T-waves are useful for assessment of proximal sensory function and for estimating peripheral conduction time, but are less useful in studies of motor control than the H-reflex. The H-reflex is an electrically elicited homologue of the T-wave that shares the same efferent pathway but has a slightly different afferent origin (Magladery and McDougal, 1951; Burke et al., 1984). H-reflexes are evoked by electrical stimulation of the muscle nerve with long stimulus pulses (0.5 ms-l ms duration). Unlike mechanical taps, electrical stimulation is not selective for the activation of muscle spindles, although it selectively activates larger diameter sensory axons at a lower intensity than the motor axons in many nerves because of differences in their strength-

duration properties (Panizza et al., 1994; Kiernan et al., 1996; Mogyoros et al., 1996; Panizza et al., 1998). An electrical stimulus activates the Ia fibers, as well as other large diameter sensory fibers and some motor axons, in a synchronous volley. Because the nerve is stimulated proximal to the spindle, the H-reflex is independent of spindle properties, such as thixotropy, as well as the level of activity in the gamma motor neurons. The latency of the H-reflex in a given muscle is shorter than its T-wave latency because electrical stimulation is given at a more proximal site (Fig. 1B vs. l C). T-waves and Hreflexes amplitudes both vary with stimulus intensity and with moment-to-moment changes in the excitability of the motor neuron pool. H-reflexes can be easily elicited in only a few muscles at rest, most notably the calf muscles and the forearm flexor muscles. H-reflexes can be obtained in many other muscles, however, during slight muscle contraction.

A +

'22.5 rns

B

c

~

36.0 ms

••

.+

31.4 illS Fig. 1. Stretch reflexes - T-waves and H-reflexes. (A) T-wave elicited from the quadriceps muscle. The patellar tendon tap was delivered at the beginning of the trace. (B) T-wave elicited from the soleus muscle in the same subject. The latency is slightly longer. The Achilles tendon tap was delivered at the beginning of the trace. (C) H-reflex elicited from the soleus muscle in the same subject, stimulating the tibial nerve in the popliteal fossa.

SPINAL REFLEXES

16.3. H-reflexes: methodological considerations

16.3.1. Recruitment curves The amplitude of the H-reflex has a If-shaped relationship to the stimulus intensity. A plot of the Hreflex amplitude and the amplitude of the compound muscle action potential (M-wave) against the stimulus intensity is called a recruitment curve. As stimulus intensity increases, the H-reflex amplitude increases to a maximum amplitude, then declines with further increases in intensity (Fig. 2). The Hreflex amplitude increases on the ascending limb of the recruitment curve because progressively more Ia fibers are activated. The H-reflex amplitude declines when the stimulus intensity begins to activate progressively more motor axons, as demonstrated by the growing amplitude of the M-wave in the same trace. There are several proposed mechanisms for the decline in H-reflex amplitude. Collision between the antidromically and orthodromically propagated action potentials within the motor axon will prevent reflex volleys from reaching the muscle, and is one of the proposed mechanisms. Another mechanism is that the antidromic invasion of the motor neuron cell

233

body produces a short period of refractoriness, that prevents Ia impulses from activating the motor neuron. There has been some debate whether recurrent inhibition, activated by antidromic impulses traveling in motor axon collaterals, contributes significantly to the decline of the H-reflex amplitude. The participation of recurrent inhibition depends on the timing - whether the recurrent inhibitory synaptic potentials, relayed through Renshaw intemeurons (Eccles et al., 1954), arrive at the motor neuron before the arrival of the Ia volley. In most circumstances, the slightly faster conduction velocities of sensory afferents (Macefield et al., 1989) coupled with the additional synaptic delay in the recurrent circuit, will tend to produce maximal recurrent inhibition after the peak of the Ia volley has arrived at the motor neuron. Construction of the H-reflex recruitment curve provides several experimental parameters useful for describing H-reflexes. The ratio of the maximum Hreflex amplitude to the maximum M-wave amplitude (Hma/Mma,) provides a rough measure of the number of motor neurons in the pool that are excitable to peripheral stimuli (Koelman et al., 1993; Hilgevoord et al., 1994). Hma/Mma, ratios are altered in periph-

Fig. 2. H-reflex recruitment curve. The amplitude of the H-reflex initially increases as the stimulus intensity is increased until a maximum amplitude (Hma, ) is reached, shown in waveforms on the left and in the ascending limb of the recruitment curve (shading) in the plot on the right. The amplitude of the H-reflex declines as the stimulus intensity increases further and the M-wave amplitude increases.

234

eral neuropathies as well as in central disorders, particularly those producing spasticity. Expression of the amplitude of an H-reflex as a percentage of the maximal M-wave normalizes for individual differences in M-wave size, thus allowing comparison between different persons. 16.3.2. Techniques for evoking H-reflexes

Soleus: The H-reflex is easily obtained in the soleus muscle and gastrocnemius muscles at rest. Surface EMG recordings from the soleus muscle can be made by placing the active recording electrode a few em distal to the gastrocnemii muscles with a reference electrode on the Achilles tendon or 3-4 em distal to the active electrode (Fig. 3A). The tibial nerve is stimulated in the popliteal fossa either using monopolar stimulation, with a remote anode on the patella, or using bipolar stimulation with the cathode facing proximal. After eliciting an If-reflex, the

M.K. FLOETER

amplitude of a subsequent H-reflex is initially depressed, taking as long as 10 s to recover fully. A stimulus rate no faster than every 3-5 s is recommended for routine clinical evaluation, and slower rates may be preferable for experimental paradigms. The criteria for recognizing an H-reflex includes: appropriate latency, elicited by stimulus intensities below those needed to evoke an M-wave, and enhancement by voluntary contraction. Forearm flexors: In many, but not all, normal subjects an H-reflex can be elicited from the median innervated forearm muscles at rest (Fig. 3B). The active recording electrode is placed over the belly of the flexor carpi radialis (FeR), and the reference electrode is positioned either 3 em distally or on the ulnar styloid. Although the longer inter-electrode distance produces a higher amplitude waveform, the larger pick-up area often contains volume conducted responses from muscles outside the forearm that may interfere with observation of the H-reflex in

Fig. 3. H-reflexes. (A) Locations for stimulating the tibial nerve in the popliteal fossa and recording from the soleus muscle to elicit an H-reflex. (B) Locations for stimulating the median nerve at the elbow and recording from the forearm (FeR) muscles to elicit an H-reflex

SPINALREFLEXES

some experimental paradigms. To elicit the H-reflex, the median nerve is stimulated at the elbow, usually using bipolar stimulation with the cathode facing proximal, or less frequently, using monopolar stimulation with a remote anode near the elbow. Like the soleus H-reflex, the FCR H-reflex increases in amplitude with increasing stimulus intensity. In some individuals the FCR M-wave duration exceeds the latency of the H-reflex, such that the maximal Hreflex response may be obscured by the M-wave. When this occurs it may also be difficult to verify that the H-reflex amplitude declines with increasing stimulus intensities. The identification of an H-reflex is supported by an appropriate latency (15-20 ms), appearance at stimulus intensities below those needed to evoke an M-wave, and enhancement by voluntary contraction (Kimura et al., 1994). Other muscles: Techniques for measuring Hreflexes in resting quadriceps (Sabbahi and Khalil, 1990; Valls-Sole et al., 1998) and biceps (Miller et al., 1995) muscles have been described. For muscles in which H-reflexes are not readily elicited at rest, voluntary contraction can be used to visualize the reflex (Burke et al., 1989). The muscle nerve is stimulated using low intensity, long duration (0.5 to 1 ms) stimuli while the subject maintains a slight, steady contraction. If the H-reflex is small, it may not be detected unless the EMG signal is rectified and averaged. Identification of the resulting waveform as an H-reflex requires that it is reproducibly time-locked to the stimulus, that it has an appropriate latency for the muscle (similar to the F-wave obtained by supramaximal nerve stimulation), that it increases in amplitude with increasing stimulus intensity and disappears with stimulation supramaximal for the M-wave. 16.3.3. H-reflexes as a probe for motor neuron excitability

The H-reflex is commonly used to assess how stimuli affect the excitability of motor neurons using a conditioning-test paradigm (Pierrot-Deseilligny and Mazevet, 2000). Conditioning-test paradigms compare the amplitudes of H-reflexes evoked by a fixed intensity "test" stimulus with those evoked when the test stimulus is paired with a "conditioning" stimulus. Changes in the H-reflex amplitude during conditioning stimulus trials are used to infer changes in the excitability of the motor neuron pool.

235

Because there is normally trial-to-trial variability in H-reflex amplitude, it is important to average a number of trials with each type of stimulus. There are several important methodological considerations when the H-reflex is used to test motor neuron excitability. (1) The intensity of the "test" stimulus must remain constant throughout the experiment. When the paradigm involves testing H-reflexes only in relaxed muscles, it may be sufficient to firmly fasten the stimulator to the skin and use constant current stimulation. However, if the H-reflex is being elicited during movement, minor displacements of the stimulator with respect to the underlying nerve will change the stimulation intensity delivered. To monitor for stimulator movement, an accepted practice is to use an intensity that evokes a small M-wave (Capaday, 1997). The M-wave is generated by motor axons that are activated directly by the stimulus, and these respond in an all-or-none fashion. A constant amplitude of the M-wave indicates that the stimulus intensity remains constant. Trials with changes in the M-wave amplitude should be discarded. (2) The size of the test H-reflex influences the magnitude of the conditioning effect (Crone et al., 1990). Small H-reflexes, consisting of a few early-recruited motor units, may show disproportionately large effects since many inputs to motor neurons have more potent effects on the first-recruited motor units. On the other hand, large H-reflexes which approach the maximal Hreflex amplitude will be relatively insensitive to weak inhibitory inputs and may exhibit occlusion with excitatory conditioning inputs. An optimal test H-reflex size allows facilatatory and inhibitory actions to be observed. One common practice is to use a test H-reflex that is 50% Hmax• Another practice is to use a test H-reflex that is 15-25% of the maximal M-wave amplitude. The latter practice allows comparison across subjects and ensures that a similar portion of the motor neuron pool will be assessed. (3) Fluctuations in the excitability of the motor neuron pool are unavoidable, but the paradigm should be designed to ensure that they are distributed across conditioning and test trials. To avoid systematic effects of long term trends in

236

excitability, conditioning and test trials should be interleaved, rather than delivered as separate blocks of trials. Randomization or psuedorandom balanced designs are preferable to performing a strict alternation between conditioning and test trials, to avoid systematic effects of long-lasting changes induced by the conditioning stimulation. (4) If the H-reflex is used to assess motor neuron excitability during movement, the level of muscle contraction must be controlled (Capaday, 1997). The amplitude of the H-reflex increases approximately linearly with the background rectified EMG activity, although the slope of that increase depends on the amplitude of the Hreflex (Capaday and Stein, 1986). Muscle contraction can be a significant problem when studying patients with movement disorders. In patients with rigidity or stiffness, complete muscle relaxation may not be possible. A compromise is to assess the H-reflex at the same level of voluntary background contraction in control subjects as involuntary contraction in patients. The strategy of matching levels of muscle contraction, however, is not practical when H-reflexes are used to assess motor neuron excitability in different phases of a complex movement - walking for example - in which EMG activity varies throughout the task. A further source of H-reflex variability is the movement-induced central modulation of the H-reflex by presynaptic mechanisms. The investigator must devise control conditions for each task that address the possible confounding influences of the movement itself. (5) The H-reflex is affected by its recent history of activation (Crone and Nielsen, 1989a). The recovery cycle of the soleus H-reflex has been explored using paired stimulation pulses (Koelman et aI., 1993). When a second stimulus is given at a delay of 30-75 ms after an H-reflex occurs, the amplitude of the second H-reflex is depressed. At a slightly longer delay of around 200 ms, the H-reflex is potentiated. At later intervals, the amplitude is again depressed, with recovery occurring over several seconds, taking as long as 10 s for complete recovery. The depression of the H-reflex has several causes. The depression at short delays has been attributed to post-spike after hyperpolarization, and to

M.K. FLOETER

postsynaptic inhibition through recurrent inhibition and Ib-activated inhibition. Presynaptic inhibition, mediated by GABAergic interneurons, persists up to several hundred ms (Hultborn et aI., 1996) but cannot explain the very long phase of recovery lasting for several seconds. The prolonged recovery is best explained by an activity-dependent process intrinsic to the presynaptic terminal (Capek and Esplin, 1977; Hultborn et aI., 1996; Kohn et al., 1997) that has been termed homosynaptic depression or postactivation depression (Hultborn and Nielson, 1998). Unlike classic presynaptic inhibition, post-activation depression is not mediated through interneurons, and is likely to be related to presynaptic autoreceptors or mechanisms controlling transmitter turnover and release. Most studies of post-activation depression have focused on the soleus If-reflex, but differences in H-reflex recovery between upper and lower limb muscles have been reported (Rossi-Durand et aI., 1999). 16.4. Methods using single motor units 16.4.1. Peristimulus time histograms

The reflex effects of a conditioning stimulus on a motor neuron pool can be assessed by sampling its effects on single motor units. Normally a motor unit fires at a fairly regular rate during steady contraction. Variations produced by a conditioning stimulus can be quantified by constructing a peristimulus time histogram (PSTH) of motor unit firing (Fig. 4). Excitatory stimuli produce time-locked peaks above baseline in the PSTH, whereas inhibitory stimuli produce troughs. An advantage of single motor unit analysis is that it can be applied to any voluntary muscle, unlike an H-reflex. Furthermore, additional analysis of the timing and synchrony of the peaks can be used to infer whether the connection is likely to be monosynaptic. The disadvantages of the technique include the requirement that a patient can produce a steady contraction for several minutes, and that a number of single units may need to be sampled to understand effects across the motor neuron pool, since the method preferentially samples early recruited units. In addition, the tonic contraction required to observe the motor unit could alter the excitability of the reflex pathway being studied.

SPINAL REFLEXES

237

Fig. 4. Peristimulus time histogram of firing of a tibialis anterior motor unit with stimulation of the peroneal nerve below motor threshold. The peak of firing in the PSTH at a latency of 45 ms after the stimulus corresponds to the reflex response to stimulation of Ia afferents. Bin size 1 ms. y-axis indicates the counts per bin.

16.4.2. Unitary H-reflex

An alternative method to assessing variations in firing patterns is a technique described by Shindo et al. (1994), in which the threshold for single motor unit firing in response to stimulation of the muscle nerve is assessed. Like the H-reflex, firing of the motor unit is elicited using electrical stimulation of the muscle nerve, but the stimulus intensity is continually adjusted to produce a 50% probability of motor unit firing, using a computer controller. When a conditioning stimulus produces excitatory effects, the current required to maintain firing declines, whereas inhibitory stimuli increase the current required. This method can be used at rest or during contraction. 16.5. Vibration-induced inhibition of H-reflexes

Muscle or tendon vibration is a handy way to deliver small, repeated stretches to activate primary muscle spindle afferents, which can follow at

frequencies of 20-200 cycles per second (Burke et aI., 1976). When muscle vibration is applied continuously for several seconds or longer, the excitability of motor neurons innervating the vibrated muscle increases, as demonstrated either by the appearance of a tonic vibration reflex or by an increase in the probability of F-wave firing. In contrast, the H-reflex amplitude declines after a short train of vibration, a dissociation that has been called the "vibration paradox" (Desmedt, 1983). The vibration-induced increase in motor neuron excitability has been attributed to the induction of plateau potentials, self sustaining states of membrane depolarization, by the barrage of spindle volleys (Kiehn and Eken, 1997; Gorassini et aI., 1998). Although it is clear that the vibration-induced inhibition of Hreflexes has a presynaptic origin, the relative contributions of different presynaptic mechanisms has not been well delineated (Katz, 1999). Repeated activation of the Ia afferent leads to post-activation depression (Crone and Nielsen, 1989a; Hultbom and

238 Nielson, 1998), but the vibration probably also activates classic GABAergic mediated presynaptic inhibition by the spread of vibration to antagonist muscles. With prolonged vibration, depletion of transmitter or mobilization of synaptic vesicles may be a limiting consideration. Nevertheless, impairment of vibration-induced inhibition of If-reflexes is one of the most consistent physiological findings in patients with spasticity (Ashby et al., 1980; Pierrot-Deseilligny, 1990; Calancie et al., 1993; Delwaide and Pennisi, 1994). Several methods for quantifying vibration-induced inhibition of H-reflexes have been described (Bour et al., 1991; Koelman et al., 1993). One method is to measure the H-reflex recruitment curve, interleaving trials with and without a preceding vibration stimulus at each intensity. Vibration-induced inhibition of the H-reflex amplitude is dependent on the size of the control reflex, with smaller H-reflexes showing the greatest inhibition. For this reason, it is best assessed over a range of intensities on the ascending limb of the recruitment curve. The amplitude of the vibrated H-reflex will be smaller than the control Hreflex elicited at the same intensity, whereas the amplitude of the M-wave remains unchanged.

16.6. Flexorand cutaneomuscular reflexes A flexor reflex is a limb response that occurs upon activation of several different types of afferents cutaneous, joint, and some muscle afferents - often in combination. The movement generally produces a mass movement of the limb that tends to move it away from the stimulus, and thus flexor reflexes are often viewed as protective reflexes. The afferents that produce flexor reflexes are commonly grouped as "flexor reflex afferents" or FRAs. Animal studies have shown that FRAs activate multiple circuits, such that the final movement represents the combination of a number of alternative reflex pathways whose state of excitability varies with respect to ongoing movement (Lundberg et al., 1987). Subtle differences occur in the flexor reflex with differences in the afferents stimulated and the site of stimulation. Most electrophysiological studies of flexor reflexes in people have used electrical stimulation of the skin or skin nerves to evoke the movement. The discussion below will describe several reflexes evoked by cutaneous stimulation which can be viewed as belonging to the family of flexor reflexes.

M.K. FLOETER

16.6.1. Cutaneomuscular reflexes

Stimulation of the skin produces a mixture of excitatory and inhibitory reflex actions. The combined actions can be observed by assessing the effect of skin stimulation upon tonic EMG activity. The response, termed the cutaneomuscular reflex, is the summation of the action of several reflex circuits, each with different latencies, time course and sign. The pattern and timing of excitatory and inhibitory peaks is specific to the muscle and the stimulation site (Meinck et al., 1983). Electrical stimulation of one of the fingers during a tonic contraction of the first dorsal interosseus muscle, for example, produces a sequence of changes as shown in Fig. 5, an average of 250 traces. The earliest excitatory peak, E1, has a peak latency of about 40 ms. E1 is followed by transient suppression of EMG, and this inhibitory period, 11, has a peak latency of about 50 ms. E1 and 11 are spinal reflexes. A second excitatory peak, E2, occurs at a latency of 60-66 ms. The magnitude of E2 is reduced in patients with dorsal column lesions and is enhanced by cortical stimulation (De Noordhout et al., 1992), suggesting a supraspinal contribution to its generation. The cutaneomuscular reflex as described above is elicited using electrical stimulation of moderate intensity, 2-4 times the perceptual sensory threshold (Fuhr and Friedli, 1987), below the threshold for producing pain. A short train of stimuli is typically more effective than a single shock (Meinck et al., 1983). The amplitude, but not the latency, of the peaks varies with stimulus strength. Muscle contraction strength also influences the magnitude of the excitatory and inhibitory peaks. Moderate contractions, e.g. 20% maximum force, are commonly used. Strong contractions can lead to occlusion of excitatory peaks. The surface EMG signal is rectified and averaged, triggering each sweep from the electrical stimulus in such a way as to include a baseline period prior to the stimuli. After averaging, the rectified signal peaks are identified using a predetermined criterion, for example an amplitude change of 50% from the mean baseline EMG. 16.6.2. Cutaneous silent period

Shocks ofrelative1y high intensity, in the range of ten to fifteen times higher than the perceptual threshold (Shefner and Logigian, 1993), produce a

239

SPINAL REFLEXES

El

E2

• •

12.0mA 500 uV N: 251

20ms

t

11 Fig. 5. Cutaneomuscular reflex in the FDI, stimulating the index finger at an intensity of four times perceptual threshold. Average of 251 traces.

strong inhibition of tonic EMG activity, termed the cutaneous silent period (Leis et al., 1995; Inghilleri et al., 1997; Manconi et al., 1998). The cutaneous silent period is very pronounced in the intrinsic hand muscles with stimulation of the digits, but can also be observed in some arm and leg muscles (Uncini et al., 1991) (Fig. 6B). In hand muscles it has a latency approximately 70-80 ms, and a duration of approximately 50 ms. It exhibits little habituation, even with intervals as short as 100 ms between pulses (Inghilleri et al., 1997). Leis (1998) has suggested that the cutaneous silent period is a protective reflex that allows the grasp to open before withdrawing the hand. Several lines of evidence suggest that small diameter afferents are primarily responsible for evoking the cutaneous silent period, including estimates of afferent conduction velocity (Inghilleri et al., 1997), relative sparing in large fiber neuropathy (Leis et al., 1992), and reduction in small fiber neuropathy (Syed et al., 2000). Cutaneous inhibition consistent with the cutaneous silent period has been demonstrated in patients after spinal injury, indicating a spinal localization of the reflex circuitry (Logigian et al., 1999). Although there has been debate whether the inhibition is presynaptic or postsynaptic (Leis et al., 1995), studies using magnetic stimulation favor a postsynaptic inhibition of motor neurons, mediated through spinal interneurons (Kaneko et al., 1998; Manconi et al., 1998).

16.6.3. Flexor reflexes Flexor reflexes of the leg occur when a moderately strong stimulation is applied to the sole of the foot. The predominant movement is flexion of the hip, knee, and ankle that produces a withdrawal of the foot. However, on closer examination, relaxation of the extensor muscles is observed to occur simultaneously (Sherrington, 1910; Hagbarth, 1960). With electrical stimulation of the posterior tibial nerve at the ankle, the flexor reflex is seen to be widely distributed across the leg flexor muscles, and the EMG signal reveals two separate components (Shahani and Young, 1971). The EMG response in the tibialis anterior muscle consists of two relatively short latency bursts (Fig. 7). The first component (Rl) has a latency < 100 ms, and the second component (R2) has a latency of about 150 ms. The Rl component has a slightly lower threshold than the R2 component. The flexor reflex is present in normal subjects, but the R2 component, which corresponds to the flexor movement itself, is enhanced in patients with pyramidal tract lesions (Shahani and Young, 1971; Fisher et al., 1979; Roby-Brami et al., 1989). Different subgroups of interneurons are thought to produce the early and late components of the flexor reflex, which can be differentially inhibited by contralateral stimulation (Bussel et al., 1989). The flexor reflex can be elicited in muscles at rest or during muscle contraction.

240

Inhibitory effects are observed in antagonist extensor muscles when the flexor reflex is elicited while extensors are active. 16.7. Mixed nerve silent periods

Supramaximal stimulation of a mixed nerve during ongoing contraction of the muscle it innervates produces an interrupted inhibition of the EMG called the mixed nerve silent period (Shahani and Young, 1973). The first segment of the silent period follows the M-wave and lasts until the F-wave, approximately 30 ms in the intrinsic hand muscles. This period of silence is caused by collision between the (volitional) orthodromic motor action potentials and the antidromic action potentials evoked by nerve stimulation. The silence which follows the F-wave has a more complex origin. Originally it was attributed to the pause in the Ia afferent firing that occurs when the spindle is unloaded following a

M.K. FLOETER

muscle twitch (Shahani and Young, 1973). However, the demonstration that this portion of the silent period also occurs in heteronymous muscles and shifts its latency in proportion to the site of afferent stimulation (Leis et al., 1991) focused attention on contributions from inhibitory spinal reflex circuits activated by nerve stimulation. Recurrent inhibition activated by the antidromic volleys in motor axons contributes to the second period of inhibition (Shefner et al., 1993) as do inhibitory reflexes produced by activation of Ia and Ib (Golgi tendon organ) afferents. Contributions from cutaneous inhibition occur after 50 ms. The second period of silence in the hand muscles is often interrupted by the long latency response around 60-70 ms. When this response is absent there is no clear distinction between a second and third segment of the silent period. The end of the mixed nerve silent period has been attributed to cutaneous inputs, since it corresponds fairly well to the end of the cutaneous silent period (Fig. 6A).

Fig. 6. Silent periods in the Abductor pollicis brevis muscle. (A) Mixed nerve silent period, stimulating the median nerve at the wrist with supramaximal stimulation (arrow) and recording from the APB during moderate contraction. Four traces are superimposed, and there is a 20 ms period of baseline EMG prior to the stimulus. The first portion of the silent period lies between the M-wave and the F-wave. The second (SP2) and third (SP3) portions are indicated (see text). (B) Cutaneous silent period (CSP) in the same subject, stimulating the index finger at 13 times perceptual threshold (arrow) and recording from the APB muscle during moderate contraction. Four traces are superimposed.

241

SPINAL REFLEXES

j~l

Vastus Lateralis

~O.5mv SOms

j~~~i\~:::f' A

Rl

R2

Fig. 7. Flexor reflex elicited in the leg by stimulation of the sole of the foot with a train of 4 pulses at 300 Hz at an intensity 1.75 times twitch threshold (arrowhead). Single traces, rectified EMG. The lower two traces show a two-component response (Rl and R2) in the tibialis anterior and hamstring, two flexor muscles. In this subject, activation of the vastus lateralis muscle, an extensor muscle (top trace) occurred at a slight delay with a timing appropriate for alternation.

16.8. Reciprocal inhibition

Contraction of a muscle is normally accompanied by inhibition of its antagonist muscle, a phenomenon called reciprocal inhibition. The effectiveness of reciprocal inhibition in curbing co-contraction depends on its strength and the duration of its action. Because a number of movement disorders exhibit abnormal co-contraction, the modulation of circuits producing reciprocal inhibition has been extensively studied in people. Spinal and supraspinal mechanisms both contribute to reciprocal inhibition. Descending systems exert many of their effects through the interneurons in spinal reflex circuits that effect reciprocal inhibition. Separate spinal reflex circuits produce postsynaptic inhibition of antagonist motor neurons and presynaptic inhibition of la afferents from antagonist muscles. The shortest latency reflex for reciprocal inhibition is disynaptic, mediated through the glycinergic la inhibitory interneuron (laIN) which has been well characterized in animals (Hultborn et al., 1976a, b). The laiN produces a brief inhibitory postsynaptic potential in motor neurons, transiently reducing the response to all excitatory inputs. lalNs are activated reflexively by la afferents from the antagonist muscle, and also by descending corticospinal, rubro-

spinal and vestibulospinal inputs and by various segmental inputs (Hultborn et al., 1976b). The pattern of connections to lalNs follows the principle that inputs which excite a motor neuron pool will also excite the lalNs that inhibit the antagonist motor neuron pool. The convergence of inputs upon the laiN allows the strength of laiN reciprocal inhibition to be dynamically modulated (Crone and Nielsen, 1994). For example, reciprocal inhibition can be suppressed when the planned movement requires cocontraction and can be facilitated in movements with alternating activation of antagonists, such as in stepping. Descending modulation of laIN reciprocal inhibition can be impaired in movement disorders even when the reflex reciprocal inhibition is normal (Meunier et al., 2000). A final point is that the strength of reciprocal inhibition at rest varies considerably for different pairs of antagonistic muscles, and that normative ranges must be established for the muscles under study. At a slightly longer delay, the reciprocal inhibition produced by lalNs is reinforced by presynaptic inhibition of the la afferents from the antagonist muscle. Presynaptic inhibition reduces the likelihood that passive lengthening of the antagonist muscle during the movement will induce a counteracting stretch reflex, but permits the motor neuron to

242

remain responsive to other inputs. Presynaptic inhibition is generated through polysynaptic circuits, with a GABAergic interneuron serving as the final interneuron (Rudomin, 1990). Like the laIN, the interneurons in the reflex circuit producing presynaptic inhibition are subject to modulation through supraspinal and segmental inputs. Several techniques have been described for measuring the strength of reciprocal inhibition in people. One method uses the H-reflex in a conditioning-test paradigm in which the conditioning stimulus consists of stimulating the antagonist nerve at different delays. In the arm, stimulation of the radial nerve in the spiral groove with long duration stimuli (0.5-1 ms) at an intensity below the motor threshold produces two periods of inhibition of the FCR Hreflex at short latencies (Day et aI., 1984) (Fig. 8) and a longer latency third period of inhibition. The

M.K. FLOETER

first period occurs when the radial and median nerve stimulation are delivered nearly simultaneously. The reduction in the H-reflex represents postsynaptic inhibition, thought to be mediated by the laIN. During the second period of inhibition, occurring at delays from 7-50 ms, motor neurons remain excitable to corticospinal activation by TMS, supporting a presynaptic origin of inhibition (Berardelli et al., 1987). The late phase of inhibition, from 50-500 ms is not well understood, and is likely to involve supraspinal mechanisms. Reversal of the third period of inhibition has been reported in dystonia and Parkinson disease (Panizza et aI., 1990; Tsai et al., 1997). Conditioning-test paradigms also have been used to demonstrate reciprocal inhibition between ankle flexor and extensor muscles, measuring the effects of common peroneal nerve stimulation near the fibular

Fig. 8. Reciprocal inhibition between forearm flexor and extensor muscles. (A) FCR H-reflexes in response to median nerve stimulation alone (test condition). Gray box highlights the FCR H-reflex, several traces superimposed. (B) A radial nerve conditioning stimulus reduced the amplitude of the test FCR H-reflexes when both nerves were simultaneously stimulated. (C) Plot of the ratio of the conditioned/test FCR H-reflex amplitude (mean±SD) in 10 normal subjects with different conditioning-test interstimulus intervals. Two periods of inhibition occur, as described in text.

243

SPINAL REFLEXES

head on the amplitude of the soleus H-reflex (Crone and Nielsen, 1994). At the shortest delay, 2-4 ms, a small inhibitory effect is observed in the resting soleus muscle although there is marked variability in the magnitude of inhibition among individuals. This phase of inhibition is enhanced at the onset of antagonist contraction, reflecting the activation of supraspinal inputs that converge upon the underlying interneurons (Crone and Nielsen, 1989b; Nielsen et al., 1992; Shindo et al., 1995). A later period of inhibition, from delays of 7-50 ms was first demonstrated by assessing changes in tonic soleus EMG following peroneal nerve stimulation at intensities near motor threshold (Mizuno et al., 1971). This period of inhibition, termed D1, is maximal at about a 20 ms delay, and is thought to be mediated through presynaptic inhibition. An even later period of depression, D2, beginning about 130 ms after the stimulus, is caused by passive stretch of the calf muscles and can be reduced by fixation of the foot. Reciprocal inhibition in the opposite direction from ankle extensor afferents upon flexor motor neurons - has also been demonstrated in healthy subjects. Stimulation of the tibial nerve produces a sharp trough in the PSTH of tibialis anterior motor unit firing at a latency of about 35 ms (Ashby and Wiens, 1989), an effect that can be presumed to represent disynaptic inhibition from IaINs by its timing. Patients with spasticity and weakness of the tibialis anterior were found to have increased reciprocal inhibition from ankle extensors with this method (Ashby and Wiens, 1989); the same finding was reported in spastic patients who had H-reflexes in their tibialis anterior muscles using a conditioning-test paradigm (Yanagisawa et al., 1976). In contrast, when reciprocal inhibition was assessed in the same patients in the opposite direction -- i.e. inhibition of soleus H-reflexes by peroneal nerve stimulation -- reciprocal inhibition was markedly reduced or, in some patients, peroneal stimulation even produced facilitation (Yanagisawa et al., 1976; Crone et al., 1994). Both changes were in a direction to cause increased excitability of ankle extensor reflexes and inhibition of ankle dorsiflexors. This example highlights the point that reciprocal inhibition is not a unitary process modulated in the same way for every muscle pair. When assessing reciprocal inhibition, or other spinal reflexes, it is important to consider the actions produced by the muscles

under study, and the normal function of the reflex during movement.

References Ashby, P and Wiens, M (1989) Reciprocal inhibition following lesions of the spinal cord in man. 1. Physiol., 414: 144-157.

Ashby, P, Verrier, M and Carleton, S (1980) Vibratory inhibition of the monosynaptic reflex. In: JE Desmedt (Ed.), Spinal and Supraspinal Mechanisms of Voluntary Motor Control and Locomotion. Basel: Karger, pp. 256-262. Berardelli, A, Day, B, Marsden, C and Rothwell, J (1987) Evidence favouring presynaptic inhibition between antagonist muscle afferents in the human forearm. J. Physiol., 391: 71-83.

Bour, LJ, Visser, BWOD, Koelman, JHTM, Baruggen, GJV and Speelman, JD (1991) Soleus H-reflex tests in spasticity and dystonia: a computerized analysis. J. Electromyogr. Kinesiol., I: 9-19.

Burke, D, Hagbarth, KE, Lofstedt, L and Wallin, BG ( 1976) The responses of humanmusclespindle endings to vibration of non-contracting muscles. J. Physiol., 261: 673-693.

Burke, D, Gandevia, SC and McKeon, B (1983) The afferent volleys responsible for spinal proprioceptive reflexes in man. J. Physiol., 339: 535-552. Burke, D, Gandevia, SC and McKeon, B (1984) Monosynaptic and oligosynaptic contributions to human ankle jerk and H-reflex. J. Neurophysiol., 52: 435448.

Burke, D, Adams, RW and Skuse, NF (1989) The effects of voluntary contraction on the H-reflex of humanlimb muscles. Brain, 112: 417-433. Bussel, B, Roby-Brami, A, Yakovleff, A and Bennis, N (1989) Late flexion reflex in paraplegic patients. Evidence for a spinal stepping generator. Brain Res. Bull., 22: 53-56.

Calancie, B, Broton, JG, Klose, J, Traad, M, Difini, J and Ayyar, DR (1993) Evidence that alterations in presynaptic inhibition contribute to segmental hypoand hyperexcitability after spinal cord injury in man. Electroencephalog. and Clin. Neurophysiol., 89: 177186. Capaday, C (1997) Neurophysiological methods for stud-

ies of the motor system in freely moving human subjects. J. Neurosci. Methods, 74: 201-218. Capaday, C and Stein, RB (1986) Amplitude modulation of the soleus H-reflex in the humanduring walking and standing. J. Neurosci., 6: 1308-1313. Capek, R and Esplin, B (1977) Homosynaptic depression and transmitter turnover in spinal monosynaptic pathways. J. Neurophysiol., 40: 95-105.

244 Crone, C and Nielsen, J (1989a) Methodological implications of the post-activation depression of the soleus H-reflex in man. Exp. Brain Res., 78: 28-32. Crone, C and Nielsen, J (1989b) Spinal mechanisms in man contributing to reciprocal inhibition during voluntary dorsiflexion of the foot. J. Physiol., 416: 255-272. Crone, C and Nielsen, J (1994) Central control of disynaptic reciprocal inhibition in humans. Acta. Physiol. Scand., 152: 351-363. Crone, C, Hu1tborn, H, Mazieres, L, Morin, C, Nielsen, J and Pierrot-Deseilligny, E (1990) Sensitivity of monosynaptic test reflexes to facilitation and inhibition as a function of the test reflex size: a study in man and cat. Exp. Brain Res., 81: 35-45. Crone, C, Nielsen, J, Petersen, N, Ballegaard, M and Hultborn, H (1994) Disynaptic reciprocal inhibition of ankle extensors in spastic patients. Brain, 117: 1161-1168. Day, BL, Marsden, CD, Obeso, JA and Rothwell, JC (1984) Reciprocal inhibition between the muscles of the human forearm. J. Physiol., 349: 519-534. Delwaide, PJ and Pennisi, G (1994) Tizanidine and electrophysiologic analysis of spinal control mechanisms in humans with spasticity. Neurology; 44 (Suppl. 9): S21-S28. De Noordhout, AM, Rothwell, JC, Day, BL, Dressler, D, Nakashima, K, Thompson, PD and Marsden, CD (1992) Effect of digital nerve stimuli on responses to electrical or magnetic stimulation of the human brain. J. Physiol., 447: 535-548. Desmedt, JE (1983) Mechanisms of vibration-induced inhibition or potentiation: tonic vibration reflex and vibration paradox in man. Adv. Neurol., 39: 671-683. Eccles, JC, Patt, P and Koketsu, K (1954) Cholinergic and inhibitory synapses in a pathway from motor-axon collaterals to motoneurons. 1. Physiol., 126: 524-562. Fisher, MA, Shahani, BT and Young, RR (1979) Electrophysiologic analysis of the motor system after stroke: the flexor reflex. Arch. Phys. Med. Rehabil., 60: 7-11. Fuhr, P and Friedli, WG (1987) Electrocutaneous reflexes in upper limbs - reliability and normal values in adults. Eur. Neurol., 27: 231-238. Gorassini, MA, Bennett, DJ and Yang, JF (1998) Selfsustained firing of human motor units. Neurosci. Lett., 247: 13-16. Hagbarth, K-E (1960) Spinal withdrawal reflexes in the human lower limbs. J. Neurol. Neurosurg. Psychiatry, 23: 222-227. Hilgevoord,AA, Koelman, JH, Bour, LJ and Ongerboer de Visser, BW (1994) Normalization of soleus H-reflex recruitment curves in controls and a population of spastic patients. Electroencephalogr. Clin. Neurophysiol., 93: 202-208.

M.K. FLOETER

Hultborn, H, Illert, M and Santini, M (1976a) Convergence on interneurons mediating the reciprocal Ia inhibition of motoneurons 1. Disynaptic Ia inhibition of Ia inhibitory interneurons. Acta. Physiol. Scand., 96: 193-201. Hultborn, H, mert, M and Santini, M (1976b) Convergence on interneurons mediating the reciprocal Ia inhibition of motoneurons III. Effects from supraspinal pathways. Acta. Physiol. Scand., 96: 368-391. Hultborn, H, mert, M, Nielsen, J, Paul, A, Ballengard, M and Wiese, H (1996) On the mechanism of the post-activation depression of the H-reflex in human subjects. Exp. Brain Res., 108: 450-462. Hultborn, H and Nielson, J (1998) Modulation of transmitter release from Ia afferents by their preceding activity - a "postactivation depression." In: P Rudomin, R Romo and L Mendell (Eds.), Presynaptic Inhibition and Neural Control. New York: Oxford University Press, pp. 178-191. Inghilleri, M, Cruccu, G, Argenta, M, Polidori, Land Manfredi, M (1997) Silent period in upper limb muscles after noxious cutaneous stimulation in man. Electroencephalogr. Clin. Neurophysiol., 105: 109115. Kaneko, K, Kawai, S, Taguchi, T, Fuchigami, Y, Yonemura, H and Fujimoto, H (1998) Cortical motor neuron excitability during cutaneous silent period. Electroencephalogr. Clin. Neurophysiol.; 109: 364-368. Katz, R (1999) Presynaptic inhibition in humans: a comparison between normal and spastic patients. J. Physiol. Paris, 93: 379-385. Kiehn, a and Eken, T (1997) Prolonged firing in motor units: evidence of plateau potentials in human motoneurons? 1. Neurophysiol., 78: 3061-3068. Kiernan, MC, Mogyoros, I and Burke, D (1996) Differences in the recovery of excitability in sensory and motor axons of human median nerve. Brain, 119: 1099-1105. Kimura, J, Daube, J, Burke, D, Hallett, M, Cruccu, G, Ongerboer de Visser, BW, Yanagisawa, N, Shimamura, M and Rothwell, J (1994) Human reflexes and late responses. Report of an IFCN committee. Electroencephalogr. Clin. Neurophysiol., 90: 393-403. Koelman, JH, Bour, LJ, Hilgevoord, AA, Van Bruggen, OJ and Ongerboer de Visser, BW (1993) Soleus H-reflex tests and clinical signs of the upper motor neuron syndrome. J. Neurol. Neurosurg. Psychiatry, 56: 776781. Kohn, AF, Floeter, MK and Hallett, M (1997) Presynaptic inhibition compared with homosynaptic inhibition as an explanation for low frequency depression. Exp. Brain Res., 116: 375-380. Leis, AA, Ross, MA, Emori, T, Matsue, Y and Saito, T (1991) The silent period produced by electrical stim-

SPINAL REFLEXES

ulation of mixed peripheral nerves. Muscle Nerve, 14: 1202-1208. Leis, AA, Kofler, M and Ross, MA (1992) The silent period in pure sensory neuronopathy. Muscle Nerve, 15: 1345-1348. Leis, AA, Stetkarova, I, Beric, A and Stokic, DS (1995) Spinal motor neuron excitability during the cutaneous silent period. Muscle Nerve, 18: 1464-1470. Logigian, EL, Plotkin, GM and Shefner, JM (1999) The cutaneous silent period is mediated by a spinal inhibitory reflex. Muscle Nerve, 22: 467-472. Lundberg, A, Malmgrem, K and Schomburg, ED (1987) Reflex pathways from group II muscle afferents. 3. Secondary spindle afferents and the FRA: a new hypothesis. Exp. Brain Res., 65: 294-306. Macefield, G, Gandevia, SC and Burke, D (1989) Conduction velocities of muscle and cutaneous afferents in the upper and lower limbs of human subjects. Brain, 112: 1519-1532. Magladery, JW and McDougal, DB (1951) Electrophysiological studies of nerve and reflex activity in normal man. Bull. Johns Hopkins Hosp., 86: 265-290. Manconi, FM, Syed, NA and Floeter, MK (1998) Mechanisms underlying spinal motor neuron excitability during the cutaneous silent period in humans. Muscle Nerve, 21: 1256-1264. Meinck, HM, Benecke, R, Kuster, S and Conrad, B (1983) Cutaneomuscular (Flexor) reflex organization in normal man and in patients with motor disorders. In: JE Desmedt (Ed.), Motor Control Mechanisms in Health and Disease. New York: Raven Press, pp. 787-796. Meunier, S, Pol, S, Houeto, JL and Vidailhet, M (2000) Abnormal reciprocal inhibition between antagonist muscles in Parkinson's disease. Brain, 123: 10171026. Miller, TA, Mogyoros, I and Burke, D (1995) Homonymous and heteronymous monosynaptic reflexes in biceps brachii. Muscle Nerve, 18: 585-592. Mizuno, Y, Tanaka, Rand Yanagisawa, N (1971) Reciprocal group I inhibition on triceps surae motoneurons in man. J. Neurophysiol., 34: 1010-1017. Mogyoros, I, Kiernan, MC and Burke, D (1996) Strengthduration properties of human peripheral nerve. Brain, 119: 439-447. Nielsen, J, Kagarnihara, Y, Crone, C and Hultborn, H (1992) Central facilitation of Ia inhibition during tonic ankle dorsiflexion revealed after blockade of peripheral feedback. Exp. Brain Res., 88: 651-656. Panizza, M, Lelli, S, Nilsson, J and Hallett, M (1990) Hreflex recovery curve and reciprocal inhibition of H-reflex in different kinds of dystonia. Neurology, 40: 824-828. Panizza, M, Nilsson, J, Roth, BJ, Rothwell, J and Hallett, M (1994) The time constants of motor and sensory

245 peripheral nerve fibers measured with the method of latent addition. Electroencephalogr. Clin. Neurophysiol., 93: 147-154. Panizza, M, Nilsson, J, Roth, BJ, Grill, SE, Demirci, M and Hallett, M (1998) Differences between the time constant of sensory and motor peripheral nerve fibers: further studies and considerations. Muscle Nerve, 21: 48-54. Pierrot-Deseilligny, E (1990) Electrophysiological assessment of the spinal mechanisms underlying spasticity. In: PM Rossini and F Mauguieres (Eds.), New Trends and Advanced Techniques in Clinical Neurophysiology. Amsterdam: Elsevier, pp. 364-373. Pierrot-Deseilligny, E and Mazevet, D (2000) The monosynaptic reflex: a tool to investigate motor control in humans. Interest and limits. Neurophysiol. CUn., 30: 67-80. Roby-Brami, A, Ghenassia, J and Bussel, B (1989) Electrophysiologic study of the Babinski sing in paraplegic patients. J. Neurol. Neurosurg. Psychiatry, 52: 1390-1397. Rossi-Durand, C, Jones, KE, Adams, Sand Bawa, P (1999) Comparison of the depression of H-reflexes following previous activation in upper and lower limb muscles in human subjects. Exp. Brain Res., 126: 117-127. Rudomin, P (1990) Presynaptic inhibition of muscle and tendon organ afferents in the mammalian spinal cord. Trends Neurosci., 13: 499-505. Sabbahi, M and Khalil, M (1990) Segmental H-reflex studies in upper and lower limbs of healthy subjects. Arch. Phys. Med. Rehabil., 71: 216-222. Shahani, B and Young, R (1971) Human flexor reflexes. J. Neurol. Neurosurg. Psychiatry, 34: 616-627. Shahani, BT and Young, RR (1973) Studies of the normal human silent period. In: JE Desmedt (Ed.), New Developments in Electromyography and Clinical Neurophysiology. Basel: Karger, pp. 589-602. Shefner, JM and Logigian, EL (1993) Relationship between stimulus strength and the cutaneous silent period. Muscle Nerve, 16: 278-282. Shefner, JM, Berman, SA and Young, RR (1993) The effect of nicotine on recurrent inhibition in the spinal cord. Neurology, 43: 2647-2651. Sherrington, CS (1910) Flexion-reflex of the limb, crossed extension-reflex, and reflex stepping and standing. J. Physiol., 40: 28-121. Shindo, M, Yanagawa, S, Morita, H and Yanagisawa, N (1994) Conditioning effect in single human motoneurons: a new method using the unitary H-reflex. J. Physiol., 481: 469-477. Shindo, M, Yanagawa, S, Morita, Hand Yanagisawa, N (1995) Increase in reciprocal Ia inhibition during

246 antagonist contraction in the human leg: a study of motor units and the H-reflex. J. Physiol., 489: 275289. Syed, NA, Sandbrink, F, Luciano, CA, Altarescu, G, Weibel, T, Schiffmann, R and Floeter, MK (2000) Cutaneous silent periods in patients with Fabry disease. Muscle Nerve, 23: 1179-1186. Tsai, CH, Chen, RS and Lu, CS (1997) Reciprocal inhibition in Parkinson's disease. Acta. Neurol. Scand., 95: 13-18.

MK FLOETER

Uncini, A, Kujirai, T, Gluck, B and Pullman, S (1991) Silent period induced by cutaneous stimulation. EEG Clin. Neurophysiol., 81: 344-352. Valls-Sote, J, Hallett, M and Brasil-Neto, J (1998) Modulation of vastus medialis motoneuronal excitability by sciatic nerve afferents. Muscle Nerve, 21: 936-939. Yanagisawa, N, Tanaka, R and Ito, Z (1976) Reciprocal inhibition in spastic hemiplegia of man. Brain, 99: 555-574.

Movement Disorders Handbook of Clinical Neurophysiology, Vol. I

M. Hallett (Ed.) © 2003 Elsevier B.Y.All rights reserved

247 CHAPTER 17

Cranial nerve reflexes: anatomical pathways, recording techniques and normative data M. Aramidehv'<, G. Cruccu" and B.W. Ongerboer de Visser b b

a Department ofNeurology/Clinical Neurophysiology, Medical Center Alkmaar, Alkmaar, The Netherlands Department ,of Neurology/Clinical Neurophysiology Unit, Academic Medical Center, Amsterdam, The Netherlands Department of Neurological Sciences, University of Rome, "La Sapienza ", Rome, Italy

17.1. Introduction Recording of the brainstem reflexes enables us to assess the functional integrity of these reflexes within the brainstem and their afferent and efferent connections. Many structural abnormalities, such as tumors or infarctions at the level of the brainstem may abolish reflexes. Recording of different brainstem reflexes may enable us to exclude structural lesions or, on the other hand, to localize more accurately the lesion within the brainstem. In the latter case a more tailored MR imaging examination of this region may be sufficient. Recording of the brainstem reflexes also plays an important role in the pre-selection of patients for additional MR imaging examination and in objectively assessing the functional abnormality of the cranial nerves by elucidating subtle changes in one or more variables of the reflex features in terms of amplitude or latency of the responses (Majoie et al., 1999). The studies of different reflex responses also provide relevant information about the physiology and pathophysiology of segmental and suprasegmental control mechanisms upon the brain stem reflexes and help to differentiate between segmental or suprasegmental origin of an abnormality. In this chapter, we describe the recording techniques, physiology, central pathways and normative data of different brainstem reflexes. As it is impossible to discuss all different aspects of the brainstem reflexes and functions in a chapter of a text book, only those neurophysiological tests that we believe

* Correspondence to : Dr. Majid Aramideh MD PhD Department of Neurology/Clinical Neur;physiology: Medical Center Alkmaar, P.O. Box 501, 1800 AM Alkmaar, The Netherlands. E-mail address:[email protected]

to be relevant in the assessment of patients with movement disorders are selected. 17.2. Orbicularis oculi reflexes 17.2.1. Blink reflex responses

The British physician Overend first elicited the blink reflex by tapping one side of the forehead (Overend, 1896). Kugelberg (1952) analyzed the blink reflex electromyographically by electrically stimulating the supraorbital nerve. When interpreting the findings, one should realize that supratentorial lesions may change corneal and blink reflex features. Reflex changes produced by lesions at different anatomical levels must therefore be analyzed with caution. In hemispheric disorders, blink reflex and corneal reflex responses may be abnormal when the affected side of the face is stimulated. In patients with extrapyramidal disorders, such as Parkinson's disease, parkinsonism, and in patients with idiopathic dystonic disorders, such as blepharospasm, Meige's syndrome and spasmodic torticollis, the orbicularis oculi reflexes should be normal. This indicates that afferent, efferent and central pathways involved in the generation of these reflexes are intact in these patients. However, recovery curves of these reflexes may be affected (vide infra). 17.2.1.1. Technique of recording During recording of the electrical blink reflex the subject lies supine on a bed with the eyes open. The supraorbital nerve is stimulated transcutaneously. The cathode is placed over the supraorbital notch on one side; the anode is placed about 2 em higher and rotated laterally at an oblique angle to avoid spread of current to the contralateral supraorbital nerve

248

Fig. 1. Electrode placements for recording responses to supraorbital nerve and corneal stimulations. GI: active electrode; G2: reference electrode.

(Fig. 1). If the second trigeminal division is to be examined, the infraorbital nerve is stimulated by placing the cathode over the nerve as it exits through the infraorbital foramen at the inferior rim of the orbit. The anode is then placed about 2 em below the cathode. Reflex responses from the inferior portion of both orbicularis oculi muscles are recorded simultaneously by surface or needle electrodes. The active surface electrode is placed on the mid-third of the inferior orbital rim and the reference electrode on the lateral surface of the nose or the temple. A ground electrode is taped under the chin or placed around the upper arm. The optimal time to stimulate is between spontaneous blinks. To minimize habituation, stimuli should be delivered at intervals of 7 s or longer, while the subject is kept alert (Boelhouwer and Brunia, 1977). Low-intensity shocks are used initially and the intensity is gradually increased until maximum and nearly stable responses are obtained with repeated trials. 17.2.1.2. Physiology and normative data The common afferent limb of the reflex components is mediated via ophthalmic division of the trigeminal nerve. The facial nerve is the common efferent limb. The electrical stimulation of the supraorbital nerve elicits two responses (Fig. 2); the first or early reflex, RI, is a brief unilateral response that occurs at about 10 ms latency in the orbicularis oculi muscle ipsilateral to the side of stimulation. The second or late response, R2, has a latency of about 30 ms. The second response is longer in duration and occurs bilaterally. RI is not visible clinically. R2 responses of the blink reflex correlate with closure of the eyelids. A contralateral RI response may sometimes occur in healthy subjects.

M. ARAMIDEH ET AL.

RI response is delayed, if it exceeds 13.0 ms and R2 is delayed if it exceeds 41 ms. (Kimura and Powers, 1969; Ongerboer de Visser and Goor, 1974). A difference between the two sides greater than 1.5 ms for RI and 8.0 ms for R2 is also considered abnormal. In addition, the difference between the ipsilateral and contralateral R2 should not exceed 5 ms (Kimura and Powers, 1969) or 8 ms (Ongerboer de Visser and Goor, 1974). Amplitudes vary considerably from one subject to the next. The normal values (mean±SD) are 0.38±0.23 mV for RI, 0.53±0.24 mV for ipsilateral R2, and 0.49±0.24 mV for contralateral R2 (Kimura et aI., 1969). Stimulation of the infraorbital nerve always evokes an R2 response, but not necessarily an RI. When R1 is not present, it is difficult to evaluate R2, because of the wide range of latency times. An absent R2, however, is certainly abnormal. While recording blink reflexes it is important to keep the subject alert. In a state of diminished consciousness or sleep, R2 and to a lesser extent RI, shows a prolonged latency and decreased response amplitude. 17.2.1.3. Central pathways The central pathways, through which the blink reflex responses are mediated, are still incompletely understood. It seems that the impulses for the RI response are conducted through the pons and are relayed via an oligosynaptic arc, probably consisting of one or two intemeurons, located in the vicinity of the main sensory nucleus of the trigeminal nerve (Fig. 2) (Shahani and Young, 1972; Kimura, 1975; Ongerboer de Visser, 1983). For the R2 responses, it has been established that afferent impulses are conducted through the descending spinal tract of the trigeminal nerve in the pons and medulla oblongata before they reach the caudal spinal trigeminal nucleus (Kimura and Lyon, 1972; Ongerboer de Visser and Kuypers, 1978). From there, impulses are relayed via medullary pathway that ascends bilaterally to reach the facial nuclei in the pons. These trigemino-facial connections are thought to pass through the lateral tegmental field, that lies medial to the spinal trigeminal nucleus (Ongerboer de Visser and Kuypers, 1978 in the human; Holstege et aI., 1986 in the cat). The observations of Aramideh et al. (1997) established that the uncrossed, ascending trigeminofacial pathway originates at the level of the lower

CRANIAL NERVE REFLEXES

249

Fig. 2. (A) Normal early (R I) and bilateral late (R2) responses of the blink reflex. The responses are shown from the right (r) and left (I) orbicularis oculi muscles after stimulation of the right (r*) and (1*) supraorbital nerves. (B) Diagram showing the presumed location of the bulbar intemeurons subserving the two components of the blink reflex. (Vll e facial nucleus; Vle abducens nucleus; Vpre principal trigeminal nucleus; Vm etrigeminal motor nucleus; Med. Tegm. Field=medial tegmental field).

medulla oblongata and that the contralateral R2 response is established by way of an ascending trigemino-facial connection that crosses the midline at the level of the lower third of the medulla oblongata. The blink reflex is influenced by many suprasegmental structures, including the motor cortex, the postcentral area of the cortex and the basal ganglia (see Esteban, 1999, for a review). 17.2.2. Corneal reflex responses 17.2.2.1. Technique of recording During recording of the corneal reflex responses, the subject lies supine on a bed or sits in a reclining

chair. Responses are recorded simultaneously with surface electrodes positioned as for recording blink reflex responses (Fig. 1). The optimal time to stimulate the cornea is between spontaneous blinks. The cornea can be stimulated mechanically or electrically. With mechanical stimulation, orbicularis oculi responses are evoked by successive manual application of a small metal sphere, 2 rom in diameter, to the cornea (Ongerboer de Visser et al., 1977). The examiner holds the upper eyelid open with one finger. When the sphere touches the cornea, contact is made between the subject and an electronic trigger circuit; which delivers a pulse. When the corneal reflex is studied by electrical stimulation, the cornea is touched lightly with a thin saline-

250

soaked cotton thread connected to the cathode of a constant-current stimulator. The anode is placed on the earlobe or forearm (Accornero et al., 1980). Square pulses, 1 ms duration, 0.1 to 3 rnA are delivered manually and the oscilloscope is triggered by the stimulus. Electrical shocks excite the A-delta nerve fibers directly. 17.2.2.2. Physiology and normative data The corneal reflex is typically nociceptive and serves to protect the eye. The cornea is innervated by unmyelinated (C) and small myelinated (A-delta) fibers (Lele and Wedell, 1959). After penetrating the cornea, the myelinated axons lose myelin and both types ofaxons terminate in the stroma and epithelium as free nerve endings. The mechanical or electrical stimulation of the cornea gives rise to a bilateral contraction of the orbicularis oculi, leading to closure of the eyelids. Three pairs of latency times should be assessed from stimulus artifact to onset of the EMG response (Fig. 3). In contrast to the blink reflex, the corneal reflex does not evoke the early R1 response. When cornea is touched mechanically, the latency time of the direct (ipsilateral) response should not exceed the consensual (contralateral) response latency by more than 8 ms. The latencies of the direct responses, evoked by stimulation of both corneas separately, should never differ by more than 10 ms. This also applies to the consensual response latencies (Ongerboer de Visser et al., 1977). With an electrical stimulus, the difference between the direct and consensual responses never exceeds 5 ms. The difference between the direct responses should never exceed 8 ms. The reflex threshold in normal subjects rarely exceeds 0.5 rnA (Accornero et al., 1980). Mechanical and electrical stimuli elicit reflex responses with similar latency times. Absolute latency values range from 36 ms to 64 ms with mechanical stimulation and from 35 ms to 50 ms with electrical stimulation. This wide range of latencies narrows if control subjects are divided into age groups, to account for the influence of age. 17.2.2.3. Central pathway The reflex afferents are A-delta fibers (Cruccu et al., 1987) passing through the long ciliary nerves and the ophthalmic division of the trigeminal sensory root, to reach the pons (Dengler et al., 1982;

M. ARAMIDEH ET AL.

Ongerboer de Visser, 1983). The central circuit is grossly similar to that of the R2 responses of the blink reflex (Fig. 3). The corneal reflex differs from R2, however, because it is a purely nociceptive reflex. The corneal reflex is relayed through different and fewer interneurons than R2 (Ongerboer de Visser and Moffie, 1979) and it is far more resistant to suprasegmental influences (Cruccu et al., 1997b).

Fig. 3. (A) Normal corneal reflex. The responses are shown from the right (r) and left (I) orbicularis oculi muscles after mechanical stimulation of the right (r*) and (1*) cornea. (B) Diagram showing the presumed central pathways subserving the corneal reflex. (For key abbreviations, see legend to Fig. 2).

251

CRANIAL NERVE REFLEXES

Afferent impulses descend along the spinal trigeminal tract, reach below the obex at the level of the trigeminal subnucleus caudalis and ascend along a multisynaptic chain of intemeurons in the lateral tegmental field before impinging on the facial motoneurons. 17.2.3. Blink reflexes responses evoked by non-trigeminal inputs 17.2.3.1. Somatosensory blink reflex An electrical stimulus to peripheral nerves, specifically the median nerve at the wrist, may induce responses in the orbicularis oculi muscles (VallsSole et aI., 1994). The circuits involved in those responses are not fully comprehended. However, the somatosensory induced blink reflex can be part of a generalized activation of the startle circuit (Karpukhina et aI., 1986), or the expression of a release phenomenon (Miwa et aI., 1998). Electrical stimulation of the median nerve may also induce responses in lower facial muscles, as an electrophysiological equivalent of the palmomental reflex (Dehen et al., 1975). 17.2.3.2. Acoustic blink reflex Sound induces a response in the orbicularis oculi muscle, which can be limited to this muscle, or involve neck and extremity muscles as in the generalized auditory startle reaction. When the response is limited to the orbicularis oculi, it is known as the auditory blink reflex (ABR). Some authors consider the ABR as the least expression of the generalized startle response (Wilkins et aI., 1986). In practical terms, however, if two different responses are generated by sound stimulation in the orbicularis oculi muscles, they cannot be easily differentiated in single individuals in routine practice. The latency of the orbicularis oculi response to sound is usually between 40 and 60 ms. 17.2.3.3. Photic blink reflex Yates and Brown (1981a) thoroughly studied orbicularis oculi reflex responses evoked by light stimuli using a photic stimulator. In a control group they obtained the optimal response and the shortest latency (50.0 ms±4.5 ms), with the stimulator held at a distance of 200 mm in front of the eyes. Afferent optic fibers probably enter the brainstem in the pretectum and impulses are then conveyed to the

facial nuclei in the pons (Tavy et al., 1984). It seems that the cerebral cortex is not involved in the generation of the photic blink reflex, since experimental ablation of occipital cortex does not influence the reflex (Weiskrantz et aI., 1974). As mentioned earlier, similar to the orbicularis oculi reflexes, the above mentioned reflexes should be normal in patients with extrapyramidal disorders such as Parkinson's disease, Parkinsonism or dystonic disorders. 17.3. Levator palpebrae inhibitory reflex The levator palpebrae superioris (LP) muscle and the orbicularis oculi (00) muscle act antagonistically during various movements of the eyelids (see subsection recording of movements). The LP muscle is innervated by the third cranial nerve. The inhibitory reflex of the LP can be examined together with the 00 reflex (Aramideh et aI., 1998). The reflex is normal in patients with cranial dystonia such as blepharospasm and oromandibular dystonia (Aramideh et aI., 1998), and whether it is altered in patients with the so-called apraxia of eyelid opening, also known as involuntary levator palpebrae inhibition, should yet be examined. 17.3.1. Technique of recording

The EMG recordings from the LP and the 00 muscles can be obtained with the subject supine. To record from the LP, a bipolar needle electrode should be inserted through the skin in the middle portion of the upper eyelid and directed toward the LP, while the subject looks downward and keeps the eyelids gently closed. The subject is then asked to open the eyes. This maneuver results in an EMG activity of the LP, which can be verified on the monitor and by the sound signal. To record from the 00, a bipolar needle electrode is inserted into the upper or lower eyelids. The position of the needles can be adjusted by asking the subject to blink or to close the eyes gently. The supraorbital nerve stimulation is similar to that for the electrical blink reflex. The stimulation should be given in the absence of spontaneous blinking, while the subject keeps the eyes voluntarily open. 17.3.2. Physiology and normative data

Stimulation of the supraorbital nerve evokes two silent periods in the LP muscle (Fig. 4) (Aramideh et

252

M. ARAMIDEH ET AL.

Fig. 4. Left panel: superimposition of three traces of the right levator palpebrae muscle (upper traces) and the right orbicularis oculi muscle (lower traces) after stimulation of the ipsilateral right supraorbital nerve (R*). The ipsilateral stimulation causes an ipsilateral early (SPl = first silent period) and late (SP2=second silent period) in the levator palpebrae and an ipsilateral RI and R2 in the orbicularis oculi. Right panel: the same superimposition of three traces, but after stimulation of the contralateral left supraorbital nerve (L*). In contrast to Rl response, regardless of the stimulation site, the first silent period, SPI, could still be evoked.

aI., 1998). The first or early silent period (SPI), and the second or late period (SP2). In contrast to Rl response of the 00 muscle, the SPI occurs bilaterally regardless of the stimulation site. The latency of the SPI varied from 9 to 13 ms and is slightly shorter than the latency of the corresponding R1 response after the ipsilateral stimulation. The SPI has a duration of 12 to 15 ms. Previous experimental works, including tracing studies, have shown that a small percentage of the LP motoneurons have axon collaterals to both LP muscles (Vanderwerf et al., 1997). Whether these motoneurons are involved in the generation of the first silent period is unclear. The majority of LP motoneurons project to one LP muscle. The latency of the SP2 varies from 27 to 35 ms and is again slightly shorter than that of the corresponding R2 responses. The SP2 has a duration of 32 to 50 ms. There is a slight variability in the features of the LP inhibitory reflex, depending on the pre-stimulus

contraction level of the LP muscle and the stimulus intensity; when the pre-stimulus contraction of the levator is weak, SPI and SP2 may form a one large inhibitory period. At lower stimulus intensities, SP2 appears sooner than SPl, Rl and R2 responses, shows a longer latency and has a shorter duration. The SP2 latency decreases and its duration increases following supramaximal stimulation of the supraorbital nerve. Preliminary data indicate that the inhibition responses of the LP muscle are probably relayed through other central pathways than those involved in Rl and R2 responses of the 00 muscle.

17.4. Jaw reflexes 17.4.1. Jaw jerk

The jaw-jerk induced by a tap on the chin was first described by De Watteville (1886). It is also called jaw reflex, mandibular reflex or masseter reflex. Without EMG recording, the clinical value of the

253

CRANIAL NERVE REFLEXES

mandibular tendon jerk is generally confined to the distinction between normal and brisk reactions, because in healthy subjects the movement of the mandible is often undetectable. Furthermore, on clinical examination alone, a unilateral interruption of the reflex arc can never be disclosed. The jaw-jerk is unaltered in patients with extrapyramidal disorders. In patients with hemirnasticatory spasm, a rare condition characterized by painful involuntary contractions of the masseter or temporal muscles on one side, often associated with hemifacial atrophy, the jaw-jerk is absent on the affected side. A bilaterally absent reflex at advanced ages has no definite clinical significance because this may also occur in healthy subjects. 17.4.1.1. Technique of recording To elicit the jaw-jerk, the examiner puts one finger on the subject's chin and taps it with a reflex hammer provided with a microswitch that triggers with a microswitch the sweep of the oscilloscope (Goodwill and O'Tuama, 1969; Ongerboer de Visser and Goor, 1974 and 1976). Electromyographic responses are recorded simultaneously from the two sides by surface electrodes. The active electrode is placed on the masseter muscle belly, in the lower third of the

--G1

-02

/ /

/ I

II

stimulus Fig. 5. Electrode placement for recording the jaw-jerk and masseter inhibitory reflex with electrode placements for stimulating the mental nerve.

distance between the zygoma and the lower edge of the mandible, and the reference electrode is placed below the mandibular angle (Fig. 5). Reflex responses can also be picked up by a small diameter concentric needle electrode inserted into each masseter muscle. In this case, a ground electrode is taped onto the forehead, neck or upper arm. To ensure a constant latency time taps should be delivered at intervals of 5 s or more. 17.4.1.2. Physiology and normative data

The latency time, which provides the most useful parameter, should be evaluated in several trials or measured on the averaged signal (Ongerboer de Visser and Goor, 1976; Yates and Brown, 1981b). The mean latency in healthy subjects is 6.8 ms (SD 0.8 ms) and a range of 5 to 10 ms (Fig. 6) (Cruccu and Ongerboer de Visser, 1999). Comparison of the latency time of the jaw-jerk responses, recorded simultaneously on both sides in one subject is of great value. A difference of more than 0.5 ms or a consistent unilateral absence of the reflex is an abnormal finding. The masseter reflex is strongly influenced by dental occlusion and can be asymmetrical or even absent in patients with temporomandibular disorders (Cruccu et aI., 1997a). Therefore, whenever the jaw-jerk is the only abnormal trigeminal reflex, it should be tested not only in the postural position but also in intercuspal occlusion or during clenching. Changing the position of the mandible or the level of pre-innervation may strongly reduce or worsen the asymmetry in patients with dental problems, but not in patients with a lesion along the reflex arc. 17.4.1.3. Central pathway Whether the afferent fibers travel in the trigeminal motor root (McIntyre and Robinson, 1959; Pennisi et aI., 1991) or the trigeminal sensory root (Goor and Ongerboer de Visser, 1976; Ferguson, 1978; Ongerboer de Visser, 1982) is still controversial (Fig. 6). Unique among the primary sensory neurons, these afferents have their cell body within the central nervous system, in the trigeminal mesencephalic nucleus, rather than in the ganglion. Collaterals from the trigeminal mesencephalic nucleus descend to the midpons to activate monosynaptically jaw-closing motoneurons of the ipsilateral side only.

254

M. ARAMIDEH ET AL.

The MIR should generally be normal in patients with extrapyramidal disorders. In patients with hemimasticatory spasm, study of the MIR during the spasm, shows an efferent block, i.e. SPI and SP2 are completely absent in the affected muscle, regardless of the side of stimulation. The silent periods are absent probably because the motor potentials are ectopically generated along the nerve, and cannot be suppressed by the reflex inhibitory input on the motoneurons (Cruccu et al., 1991).

Fig. 6. (A) Normal jaw-jerk responses from the right (R) and left (L) masseter muscles. The lower traces are made by averaging, and arrows mark the latency times. (B) Diagram showing the presumed central pathways subserving the jaw-jerk responses. (Ncl Mes N V =mesencephalic nucleus of the trigeminal nerve; Ncl Mot N V=motor nucleus of the trigeminal nerve; Ncl Prine N V= principal sensory nucleus of the trigeminal nerve; Ncl Tract Spin N V =nucleus of the trigeminal spinal tract; Nill= oculomotor nerve; Ophth=ophthalmic trigeminal root; Max =maxillary trigeminal root; Mand=mandibular trigeminal root; Mot Root N V =trigeminal motor root; N Vle abducens nerve.

17.4.2. Masseter inhibitory reflex

The masseter inhibitory reflex (MIR), also called the cutaneous silent period or exteroceptive suppression reflex, was first described by Hoffman and Tonnies (1948), as the inhibitory component of the tongue-jaw reflex seen after electrical stimulation of the tongue. The EMG silent period (SP) refers to a transitory relative or absolute decrease in EMG activity evoked in the midst of an otherwise sustained contraction (Shahani and Young, 1973).

17.4.2.1. Technique of recording The masseter inhibitory reflex is recorded bilaterally with the same position for electrodes as described for the jaw-jerk (Fig. 5). Subject is sitting upright, and instructed to clench the teeth as hard as possible, for a period of 2-3 s, with the aid of auditory feedback. The reflex can only be measured properly if the subject is able to clench the teeth in intercuspal occlusion and to produce an interferential EMG pattern. It may be necessary to use a concentric needle electrode, instead of surface electrodes, particularly if the signal is contaminated by facial muscle activity. Single electrical shocks 0.2 ms in duration are delivered to the mentalis or infraorbital nerves, through surface electrodes placed over the homonymous foramina. A stimulus intensity of about 2-3 times of the reflex threshold (usually 20-50 rnA) yields the best results. It is always necessary to examine several trials, usually 8-16, allowing 10-30 s of rest between the contractions. Some authors measure the latency at the last EMG peak, some at the last crossing of the isoelectric line, and others at the beginning of the electric silence. Each of these methods is clinically satisfactory if the same criterion is maintained and intra-individual differences between right and left stimulations are examined. 17.4.2.2. Physiology and normative data Mechanical or electrical stimulation, applied anywhere within the mouth or on the facial skin of the maxillary and mandibular trigeminal divisions, evokes a reflex inhibition in the jaw-closing muscles. These reflexes probably play a role in the reflex control of mastication, by preventing intraoral damage that could occur with uncontrolled contraction of jaw closing muscles, and in jaw movements during speech.

255

CRANIAL NERVE REFLEXES

The MIR consists of two electrical silent periods, interrupting the voluntary EMG activity in the ipsilateral and contralateral masseter muscles (Shahani and Young, 1973; Godeaux and Desmedt, 1975; Ongerboer de Visser and Goor, 1976) (Fig. 7). The early silent period, SP1, has a latency of 10-15 ms. The late silent period, SP2, has a latency of 40-50 ms. A latency difference between the ipsilateral and contralateral responses greater than 2 ms for SP1 or 6 ms for SP2 is abnormal (Cruccu, 1987). In a few subjects, little or no EMG activity occurs between the two SPs, i.e. SPI and SP2 merge in a single long-lasting SP, even when the strength of contraction is maximum. In this case, the latency of the recurrence of EMG activity is taken as a measure of SP2. The latency difference between right and left stimulations following recording from one muscle should not exceed 8 ms. If full-wave rectification is available, 8-16 trials should be averaged. The latency and duration of SPs can be measured from the intersection of the rectified and averaged signal and a line indicating 80% of the background EMG level (Cruccu et aI., 1987). In 100 normal subjects, aged 15-80 years, the mean latency of SP1 was 11.8 ms (SD 0.8) and that of SP2 was 45 ms (SD 5.2). The duration of SP1 was 20 ms (SD 4) and the duration of SP2 was 40 ms (SD 15) (Cruccu and Ongerboer de Visser, 1999). Probably because electrical stimuli yield a mixed nociceptive and non-nociceptive input, whether the SP1 or the SP2, or both components, are nociceptive reflexes remains controversial (Miles and Turker, 1987; Ellrich et aI., 1997). Both inhibitory responses can nevertheless be elicited with innocuous mechanical stimuli, and indirect evidence supports the view that the afferents belong to the intermediately myelinated A beta group (Shahani, 1970, Cruccu et al., 1989). 17.4.2.3. Central pathway After stimulation of the mental or infraorbital nerve, impulses reach the pons via the sensory mandibular or maxillary root of the trigeminal nerve, respectively (Fig. 7) (Ongerboer de Visser and Goor, 1976). The SP1 response is probably mediated by one inhibitory interneuron, located close to the ipsilateral trigeminal motor nucleus. The inhibitory interneuron projects onto jaw-closing motoneurons bilaterally. The whole circuit lies in the mid-pons (Ongerboer de

A. stimulus

*-

Fig. 7. (A) Normal early (SPI = first silent period) and late (SP2=second silent period) phase of the masseter inhibitory reflex. The responses are shown from the right (upper trace) and left (lower trace) masseter muscles after stimulation of the right (r*) mental nerve. (B) Diagram presenting the presumed location of the bulbar interneurons subserving (I) the early (SP!) and (2) the late (SP2) phase of the masseter inhibitory reflex. (For key abbreviations, see legend to Fig. 25.6).

256

Visser et al., 1990). The afferents for SP2 descend in the spinal trigeminal tract and connect with a polysynaptic chain of excitatory interneurons, probably located in the lateral reticular formation, at the level of the pontomedullary junction. The last interneuron of the chain is inhibitory and gives rise to ipsilateral and contralateral collaterals that ascend medial to the right and left spinal trigeminal complexes, to reach the trigeminal motoneurons (Ongerboer de Visser et al., 1990). 17.5. Assessment of cranial nerves motor function

17.5.1. Cranial nerves conduction examination

Of the cranial nerves, only the facial nerve and the spinal accessory nerve are commonly studied by stimulating the nerve along its course and recording the direct motor response of the muscle, because stimulation of the other nerves, lying for most of their course in the depth of craniofacial structures is difficult. By means of magnetic transcranial stimulation, however, the motor cortex and the intracranial root of most cranial motor nerves can be excited painlessly, and the potential can be recorded from the target muscles. Measurement of the latency and amplitude of the muscle potential provides quantitative information on nerve function. 17.5.1.1. Facial nerve Electrical stimuli are delivered to the facial nerve through surface electrodes placed near the stylomastoid foramen, i.e. just below and anterior to the mastoid bone. Selective stimulation of the upper (zygomatic) or lower (mandibular) branches of the nerve can also be applied. Surface recording electrodes are placed over the orbicularis oculi, nasalis, orbicularis oris, or mentalis muscles, and the direct muscle potential (M-wave) can be recorded by surface electrodes. The reference electrode is placed either over the nasal bone or over the same muscle on the opposite side. The position of the ground electrode is not critical and it is often placed on the lip, chin, or wrist. The main problem with surface recordings from facial muscles is that signals from nearby muscles may be picked-up by volume conduction. This often alters the M wave, which makes the measurement of the onset latency and amplitude difficult. In order to obtain an M wave starting with a negative deflection

M. ARAMIDEH ET AL.

of maximum amplitude, it is sometimes necessary to move the active and reference electrodes, since the optimal position varies slightly between subjects. Some investigators prefer recording by concentric needle-electrode placed at the comer of the mouth or the lateral epicanthus of the eye. In needle recordings, contamination by nearby muscles is avoided because the signal is only picked up from the target muscle. On the other hand, only a proportion of motor units contribute to the signal. Measurement of latency and amplitude is, therefore, less accurate with needle electrode than it is with surface electrode. Normal latency values, as measured by surface electrodes, vary between 2.5 and 5 ms, with a mean of 3.2 ms (SD 0.7). The amplitude varies widely among muscles and individuals and is only examined by side-to-side comparison. 17.5.1.2. Spinal accessory nerve The spinal accessory nerve has two principal components; the larger and more important component is the spinal portion and the smaller branch is the accessory nerve, so called because of its function as accessory to the vagus nerve. The spinal accessory nerve can be stimulated percutaneously in the posterior triangle of the neck at the midportion of the posterior edge of the sternocleidomastoid muscle. Evoked potentials can be recorded over the trapezius muscle at the angle of neck and shoulder. Normal latencies vary between 1.8 and 3.0 ms in healthy subjects, aged 10-60 years (Cherington, 1968; Berry et al., 1991). 17.5.1.3. Hypoglossal nerve Hypoglossal nerve provides motor innervation to the extrinsic and intrinsic muscles of the tongue. The hypoglossal nucleus contains afferent fibers, which appear to be largely spindle afferents (Thomas and Mattias, 1992). Hypoglossal nerve conduction can be accomplished by placing surface electrodes on the top of the tongue and stimulating in the submandibular region, just medial to the angle of the mandible (Redmond and Di Benedetto, 1988). 17.5.2. Recording of movements

Electromyographic recording of spontaneous movements occurring normally in muscles of the

CRANIAL NERVE REFLEXES

257

with a search coil, Bour et al. (2000) also showed that during the electrical blink reflex an early ipsilateral response and late bilateral responses can be identified in the upper eyelid movement (Fig. 11). Based on the coordinated movements of the eyelids and eyes during different types of blinking, it may be suggested that within the brainstem there is a premotor neural structure acting as a generator to

Fig. 8. Schematic drawing of the three components of the orbicularis oculi muscle.

face is of some utility for understanding physiological and pathophysiological aspects of the brainstem and suprasegmental structures. Eyelid kinematics can be investigated using electromagnetic recordings (Evinger et al., 1984; Bour et al., 2000). Furthermore, needle EMG recordings simultaneously from the levator palpebrae and the orbicularis oculi muscles provide information on the normal reciprocal activity between these two muscles (Gordon, 1951; Bjork and Kugelberg, 1953; Aramideh et al., 1994b, 2001). Needle electrodes can be inserted into different portions of the orbicularis oculi (Fig. 8); the pretarsal portion is involved particularly in different types of blinks. The orbital portion is involved in forceful closure of the eyelids and the preseptal portion is more or less activated during both movements. A needle electrode can be inserted into the levator palpebrae as discussed earlier (see levator palpebrae inhibitory reflex). In physiological conditions, during a blink the levator relaxes, followed by the contraction of the orbicularis. As soon as the orbicularis returns to its quiescent state, the levator resumes its tonic activity and raises the eyelid to its previous position. The same reciprocal activity can be observed during voluntary closure and opening of the eyelids (Fig. 9) (Aramideh et al., 1994b, 1995b, 2001). By using double magnetic induction technique or the search coil technique, it is also possible to record the eye movements during different movements of the eyelids (Collewijn et al., 1985; Riggs et al., 1987; Bour et al., 2000). During all types of blinking, Aramideh et al. (l994a) and Bour et al. (2000) demonstrated a downward and nasalward displacement ofthe eyes (Fig. 10). By recording of the eyelid

Fig. 9. Electromyograms from the levator palpebrae superioris (LP) and the orbicularis oculi (00) muscles, showing the reciprocal inhibition between these two muscles in a healthy subject. When the subject is asked to close the eyes gently, LP activity ceases abruptly, followed by contraction of 00 (upper two traces at black arrow). Note the occurrence of dense bursts of action potentials with high amplitude preceding the return of LP activity on the order "open eyes", following the inhibition of 00 (lower third trace at open arrow). The lowest two traces show the total inhibition of the LP muscle activity and a brief contraction of the 00 during the act of spontaneous blinking (two times).

258

Horizonta' 00(°)

M. ARAMIDEH ET AL.

~Oj

A Horizontal 00 (0)

0.0

~j

Horizontal OS (0)

0.0

(0)

~Oj

Vergence

0.0

(0)

-10.0

Vertical 00(°)

Vertical

~Oj

Vertical 00 (0)

0.0

(0)

~l

Vertical 05(°)

0.0

0.0

~Oj 00

~Oj

0.0

-10.0

~Oj

Vertical Vergence

0.0

-10.0

~Oj

-10.0

-10.0

Vertical Vergence

00

-10.0

-10.0 OS (0)

~Oj -100

-10.

Vergence

0.0 -100

-10.0 Horizontal OS (0)

~Oj

B

(0)

15

~Oj

0.0

----v-.

-

•

-10.0

15

Fig. 10. The features of eye movement during spontaneous blinking (A) and gentle closing and opening of the eyelids (B), recorded in two healthy subjects. Positive changes denote right gaze, divergence, or upward gaze. For vertical vergence, positive changes denote that the right eye (00) is higher than the left eye (OS).

coordinate the incoming impulses to different subnuclei of the ocular motor system as well as the facial motoneurons. Swallowing is another natural spontaneous movement that occurs repeatedly in awake humans. Although muscles involved in swallowing are not readily accessible to electromyographic recordings, they can be studied with combined EMG and mechanical methods (Ertekin et aI., 1995). A simple approach to the functional assessment of superficial muscles involved in mastication and swallowing can be performed by surface electrodes placed on the masseter and orbicularis oris muscles. These are activated in a well-defined alternating pattern while chewing that becomes synchronous when swallowing (Fig. 12).

17.6. Recovery curves to paired stimuli Because of passive mechanisms (e.g. after-hyperpolarization potential) or the intervention of negative feedback circuits, the excitability of a reflex circuit is depressed after the passage of an earlier impulse. With the double-shock technique it is possible to draw the recovery curve (or excitability cycle) of a given reflex and thus obtain a measure of the excitability of the reflex circuit (Kimura, 1973). Stimulus intensity and position of the stimulating and recording electrodes are the same as those used in a standard reflex study. Electrical stimuli of equal intensity are delivered in pairs, at varying interstimuli time-intervals. The response to the first shock

259

CRANIAL NERVE REFLEXES

OS(20mA)

Time(ms)

OD(20mA)

Time(ms)

Fig. 11. Two components of eyelid movement during evoked blink reflex. From top to bottom: horizontal (Eh(OD», and vertical (Ev(OD» movements of the right eye, upper eyelid movement (lid(OD» and upper eyelid velocity (lidve,(OD» for the right eye and EMG of the right orbicularis oculi muscle (OOEMO(OD» and the left orbicularis oculi (OOEMO(OS». Several responses after electrical stimulation of the left (A) and right (B) supraorbital nerves with an 8-mA current are superimposed. Upward deflections relate to rightward eye movement for the horizontal component, upward eye movement for the vertical component and upward movement or upward velocity of the upper eyelid. Arrows in B indicate the start of the early component.

is called "conditioning" and the response to the second shock is called "test" response. The size (amplitude, duration or area) of the two responses is measured, and that of the test response is expressed as a percentage of the conditioning response. The recovery curve is drawn by plotting the size of the test response, as a percentage of the conditioning response, on the Y-axis and the time-interval on the

X-axis (see Chapter Aramideh, Valls-Sole, Ongerboer, Fig. 8) (Kimura 1973). The studies on the recovery curves of different cranial nerve reflexes have been performed in various movement disorders, particularly in patients with dystonic disorders, Parkinson's disease and parkinsonism disorders. The recording of the recovery curve enables us to examine the excitability of

260

M. ARAMIDEH ET AL.

Fig. 12. Normal alternating masseter and orbicularis oris muscle activation patterns during chewing, and synchronous activation during swallowing.

the interneuronal circuits and the integrity of the suprasegmental structures. Patients with blepharospasm have an abnormal brain stem interneuronal excitability enhancement, as investigated with the recovery curves of R2 response of the orbicularis oculi reflex, similar to that found in patients with parkinsonism (Berardelli et al., 1985; Tolosa et al., 1988; Cruccu et al., 1991; Aramideh et al., 1995a). In a group of 33 patients with involuntary eyelid closure, based on EMG patterns, Aramideh et al. (1995a) showed that recovery of R2 was enhanced in all patients with pure blepharospasm, but it was normal in all patients with pure involuntary levator palpebrae inhibition and in 75% of patients with a combination of both disorders. Nakashima et al. (1990), Pauletti et al. (1993) and Eekhof et al. (1996) found that the abnormalities of the recovery curve are also present in patients with cervical and generalized dystonia, even in those without blepharospasm. Normal recovery curves are usually seen in focal arm dystonia only (Tolosa et al., 1988; Nakashima et al., 1990). The recovery curves of the SP2 component of the masseter inhibitory reflex may be enhanced in dystonic patients, with or without facial dystonia (Cruccu et al., 1991). It is interesting that the

recovery of SP2 was also facilitated in dystonic patients not exhibiting jaw closing dystonia (Pauletti et al., 1993), as was the recovery of the R2 component of the blink reflex, which was facilitated in patients without blepharospasm (Nakashima et al., 1990). In Parkinson's disease, the enhanced excitability of brain stem interneurons leads to a rapid recovery of the responses (Kimura, 1973). Similar abnormalities have been described in the excitability recovery curve of the MIR (Cruccu et al., 1991). The blink reflex excitability recovery curve has also been studied in other parkinsonian disorders such as progressive supranuclear palsy and multiple system atrophy, with abnormalities that were similar to those found in patients with Parkinson's disease.

17.6.1. Recovery of orbicularis oculi reflexes A complete excitability cycle of the orbicularis oculi reflex can be examined with the time-intervals between 10 ms and 100 ms every 10 ms, and those between 100 ms and 1500 ms every 100 ms. However, it is not clinically indispensable to check all these time-intervals. The test R1 summates with the conditioning R1 or R2 at short interstimuli time-intervals, up to 60-70

261

CRANIAL NERVE REFLEXES

ms. The apparent facilitation may reach 250% at 30-40 ms intervals. With longer intervals, Rl is affected little by the conditioning shock. It may be very slightly reduced (80% at the 100 ms interval) and slowly recovers to 90-100% of the conditioning response at the intervals of 200-500 ms (Kimura, 1973). The test R2 is usually completely abolished at interstimuli time intervals shorter than 200, it then very slowly recovers, reaching about 40-50% at the 500 ms interval and 70-90% at the 1500 ms interval (see Chapter Aramideh, Valls-Sole, Ongerboer, Fig. 8) (Aramideh et al., 1995a; Eekhof et al., 1996). The recovery of the corneal reflex parallels but is more rapid than that of R2. The test corneal reflex already measures about 30% at the 200 ms interval, while R2 of the blink reflex is still abolished, and reaches 90-100% at the 1500 ms interval (Kimura, 1973; Cruccu et al., 1986). Since the same motoneurons are shared by the various orbicularis oculi reflexes, the difference in recovery times (progressively longer for R1, corneal reflex, and R2) is commonly attributed to the difference in the interneuronal net. The R2 of the blink reflex response is most susceptible to changes in excitability, and the R2 recovery curve has provided valuable information in research and clinical settings. It may also be valuable to measure the recovery index (Aramideh et al., 1995a; Eekhof et al., 1996).

80

E Q)

60

E

'5 ... 0

40

Q)

a: 0~

20 00

400 600 800 Stimulus Interval Recove of O. Oculi

1000

400 600 800 Stimulus Interval Recove of LP and 00

1000

100

-oS

80

c: Ql

..

'5 0

Ql

II:

60 40

~ 0

20 00 100

E Ql

200

80

-...

E 60

'5

17.6.2. Recovery of levator palpebrae inhibitory reflex

With the same needle position as for the recording of the levator palpebrae EMG activity and the inhibition reflex (see earlier), the recovery of the inhibition periods of the levator palpebrae muscle can be examined with the recovery curve of the orbicularis oculi responses (Fig. 13) (Aramideh et al., 1998). The durations of the two silent periods are measured, and that of the test response is expressed as a percentage of the conditioning response. The recovery curve is drawn by plotting the size of the test response, as a percentage of the conditioning response, on the Y-axis and the time interval on the X-axis. At interstimuli time-intervals higher than 500 ms, the percentage of inhibition recovery of the second

0

Ql

II:

40

~ 0

20 00

200

400 600 Stimulus Interval

800

1000

Fig. 13. The recovery curve of the second silent period (SP2) of the levator palpebrae (L. Palpebrae) and the recovery curve of the R2 response of the orbicularis oculi (0. Oculi). The lowest panel shows the recovery curves of both muscles together. At higher stimulus intervals, for instance 1 s and 0.5 s, the percentage of inhibition recovery of the SP2 was similar to that of the excitability recovery of the R2 response. At lower stimulus intervals, the recovery curve of inhibition responses were shifted upward, i.e. SP2 recovered faster than corresponding R2 response. For example, at 200 ms where R2 response did not show any recovery, the SP2 could still be recruited for about 20-30%. The SPI is not examined systematically.

262

M. ARAMIDEH ET AL.

silent period (SP2) was similar to that of the excitability recovery of the R2 response (Fig. 13). At lower intervals, the recovery curve of SP2 was shifted upward, i.e. SP2 recovered faster than corresponding R2 response. For example, at 200 ms where R2 response was completely abolished, the SP2 could still be recruited for about 20-30%. These findings are in agreement with our earlier observation (see earlier) and indicate that responses are relayed on different central pathways.

I I I I

~

17.6.3. Recovery of masseter inhibitory reflex

L!¥; I

o

100

me

250

c

100

20 100

250

500

Fig. 14. Recovery curvesof the masseterinhibitoryreflex. (A) Eight rectified and averaged signals, recorded in a healthy subject after stimulation of the mental nerve. The first (conditioning) shock is indicated by the dashed line. The second(test) shock is indicated by the arrows. After a secondshockwith an interstimulus interval of 100 ms, the test SP2 is almostabolished(AI) and partly recovers with an interval of 250 ms (A2). (B) The same as in (A) but in a patient with Parkinson's disease. The test SP2 is only slightlysuppressed at the interstimulus intervalof 100 ms (B1) and completely recovers at the interval of 250 ms (B2). (C) Recovery curve of the SP2 component of the masseter inhibitory reflex. X-axis: interstimulus interval (ms); Y-axis: area of the test response expressed as percentage of the conditioning response. The two curves are ± standard errors of the estimate in 20 healthy subjects. The two squares indicate the recovery value for the 100 ms and 250 ms intervals in (B). Note that the recovery of SP2 is enhanced in the patient with Parkinson's disease.

The recovery cycle of the masseter inhibitory reflex is studied by delivering paired stimuli at interstimulus time-intervals of 100 ms, 150 ms, 250 ms, and 500 ms, with the same low-rate stimulation and alternation of "clench" and "rest" phases described for the assessment of the reflex values. The recovery curve is drawn by plotting the timeintervals on the X-axis and the size (area or duration) of the test response as a percentage of the size of the conditioning response on the Y-axis (Fig. 14). The recovery of SP1 in normal subjects varies from about 85% at 100 ms interval to approximately 96% at 500 ms (Fig. 14). The recovery of SP2 varies from 24% at 100 ms to 79% at 500 ms. The function of correlation between time interval (milliseconds) and size of the test response (percent) is linear for SP1 (Y=0.02 X+85) and logarithmic in base 10 for SP2 (Y =75 log X - 120) (Cruccu et al., 1991). Recovery can also be evaluated however by simply measuring the duration of SP1 and SP2 in non-rectified recordings, and for clinical use, it may be sufficient to measure the recovery of SP2 at the 250 ms interval (Cruccu et al., 1997a; Cruccu and Ongerboer de Visser, 1999).

References Accornero, N, Berardelli, A, Bini, G, Cruccu, G and Manfredi, M (1980) Corneal reflex elicited by electrical stimulation of the human cornea. Neurology, 30: 782-785. Aramideh, M, Bour, LJ, Koelman, JHTM, Speelman, JD and Ongerboer de Visser, BW (1994a) Abnormal eye movements in blepharospasm and involuntary levator palpebrae inhibition: clinical and pathophysiological considerations. Brain, 117: 1457-1474. Aramideh, M, Ongerboer de Visser, BW, Devriese, PP, Bour, LJ and Speelman, JD (1994b) Electromyo-

CRANIAL NERVE REFLEXES

graphic features of levator palpebrae superioris and orbicularis oculi muscles in blepharospasm. Brain, 117: 27-38. Aramideh, M, Eekhof, JLA, Bour, U, Koelman, JHTM, Speelman, JD and Ongerboer de Visser, BW (1995a) Electromyography and blink reflex recovery in involuntary eyelid closure: A comparative study. J. Neurol. Neurosurg. Psychiatry, 58: 692-698. Aramideh, M, Ongerboer de Visser, BW, Koelman, JHTM and Speelman, JD (1995b) Motor persistence of orbicularis oculi muscle in eyelid opening disorders. Neurology, 45: 897-902. Aramideh, M, Ongerboer de Visser, BW, Koelman, JHTM, Majoie, CB and Holstege, G (1997) The late blink reflex abnormality due to lesions of the lateral tegmental field. Brain, 120: 1685-1692. Aramideh, M, Ongerboer de Visser, BW and Van der Werf, F (1998) The inhibitory response of levator palpebrae muscle to evoked blink reflexes. In: J Valls-Sole and E Tolosa (Eds.), Brainstem Reflexes and Functions. ENE Publicidad, Madrid, pp. 89-98. Aramideh, M, Koelman, JHTM, Speelman, JD and Ongerboer de Visser, BW (2001) Eyelid movement disorders and electromyography. Lancet, 357: 805806. Berry, H, MacDonald, EA and Mrazeh, AC (1991) Accessory nerve palsy: a review of 23 cases. Can. J. Neural. Sci., 18: 337-341. Bjork, A and Kugelberg, E (1953) The electrical activity of the muscles of the eyes and eyelids in various positions and during movement. Electroencephalogr. cu« Neurophysiol., 5 : 595-602. Boelhouwer, AJW and Brunia, CHM (1977) Blink reflexes and the state of arousal. J. Neural. Neurosurg. Psychiatry, 40: 58-63. Bour, U, Aramideh, M and Ongerboer de Visser, BW (2000) Neurophysiological aspects of eye and eyelid movements during blinking in man. J. Neurophysiol., 83: 166--176. Collewijn, H, Van Der Steen, J and Steinman, RM (1985) Human eye movements associated with blinks and prolonged eyelid closure. J. Neurophysiol., 54: 11-27. Cruccu, G and Ongerboer de Visser, BW (1999) The jaw reflexes. In: G Deuschl and A Eisen (Eds.), Recommendations for the Practice of Clinical Neurophysiology: Guidlines of the International Federation of Clinical Neurophysiology. Elsevier Science Publisher B.Y., pp. 243-247. Cruccu, G, Agostino, R, Berardelli, A and Manfredi, M (1986) Excitability of the corneal reflex in man. Neurosci. Lett., 63: 320-324. Cruccu, G, Inghilleri, M, Fraioli, B, Guidetti, B and Manfredi, M (1987) Neurophysiological assessment of

263 trigeminal function after surgery for trigeminal neuralgia. Neurology, 37: 631-638. Cruccu, G, Agostino, Rand Inghilleri, M (1989) The masseter inhibitory reflex is evoked by innocuous stimuli and mediated by A beta afferent fibers. Exp. Brain Res., 77: 447-450. Cruccu, G, Pauletti, G, Agostino, R, Berardelli, A and Manfredi, M (1991) Masseter inhibitory reflex in movement disorders. Huntington's chorea, Parkinson's disease, dystonia, and unilateral masticatory spasm. Electroencephalogr. Clin. Neurophysiol., 81: 24--30. Cruccu, G, Frisardi, G, Pauletti, G, Romaniello, A and Manfredi, M (1997a) Excitability of the central masticatory pathways in patients with painful temporomandibular disorders. Pain, 73: 447-454. Cruccu, G, Leandri, MG, Ferracuti, S and Manfredi, M (1997b) Corneal reflex responses to mechanical and electrical stimuli in coma and narcotic analgesia in humans. Neurosci. Lett., 222: 33-36. Dengler, R, Rechl, F and Struppler, A (1982) Recruitment of single units in human blink reflex. Neurosci. Lett., 34: 301-305. Dehen, H, Bathien, N and Cambier, J (1975) The palmomental reflex. An electrophysiological study. Eur. Neurol., 13: 395-404. De Watteville, A (1886) Note on the jaw-jerk, or masseteric tendon reaction, in health and disease. Brain, 8: 518-519. Eekhof, JLA, Aramideh, M, Bour, LJ, Hilgevoord, AAJ, Speelman, JD and Ongerboer de Visser, BW (1996) Blink reflex recovery curves in blepharospasm, torticollis spasmodica and hemifacial spasm. Muscle Nerve, 19: 10-15. Ellrich, J, Hopf, HC and Treede, RD (1997) Nociceptive masseter inhibitory reflexes evoked by laser radiant heat and electrical stimuli. Brain Res., 764: 214--220. Ertekin, C, Pehlivan, M, Aydogdu, I, Ertas, M, Uludag, B, Celebi, G, Colakoglu, Z, Sagduyu, A and Yuceyar, N (1995) An electrophysiological investigation of deglutition in man. Muscle Nerve, 18: 1177-1186. Esteban, A (1999) A neurophysiological approach to brainstem reflexes. Blink reflex. Neurophysiol. Clin., 29: 7-38. Evinger, C, Shaw, MD, Peck, CK, Manning, KA and Baker, R (1984) Blinking and associated eye movements in humans, guinea pigs and rabbits. J. Neurophysiol., 52: 323-339. Ferguson, IT (1978) Electrical study of jaw and orbicularis oculi reflexes after trigeminal nerve surgery. J. Neurol. Neurosurg. Psychiatry, 41: 819-823. Godeaux, E and Desmedt, JE (1975) Exteroceptive suppression and motor control of the masseter and temporalis muscles in normal man. Brain Res., 85: 447-458.

264 Goodwill, CJ and O'Tuama, L (1969) Electromyographic recording of the jaw reflex in multiple sclerosis. J. Neural. Neurosurg. Psychiatry, 32: 6--10. Goor, C and Ongerboer de Visser, BW (1976) Jaw and blink reflexes in trigeminal nerve lesions. Neurology, 26: 95-97. Gordon, G (1951) Observation upon the movement of the eyelids. Br. J. Ophthalmol., 35: 339-351. Hoffman, P and Tonnies, JF (1948) Nachweis des vollig konstanten verkommens des zungen-kiefer-reflexes bei menschen. Pfliigers Archiv., 250: 103-108. Holstege, G, Tan, J, Van Ham, 11 and Graveland, GA (1986) Anatomical observations on the afferent projections to the retractor bulb motoneuronal cell group and other pathways possibly related to the blink reflex in the cat. Brain Res., 374: 321-334. Karpukhina, MY, Gokin, AP and Limanskii, yP (1986) Activation of pontine and bulbar reticulo-spinal neurons in the cat by somatosensory stimuli of different modalities. Neurophysiology, 18: 329-336. Kimura, J (1973) Disorder of interneurons in parkinsonism. The orbicularis oculi reflex to paired stimuli. Brain, 96: 87-96. Kimura, J (1975) Electrical elicited blink reflex in diagnosis of multiple sclerosis. Review of 260 patients over a seven-year period. Brain, 98: 413-426. Kimura, J (1982) Conduction abnormalities of the facial and trigeminal nerves in polyneuropathy. Muscle Nerve, 5: 149-144. Kimura, J and Lyon, LW (1972) Orbicularis oculi reflex in Wallenberg syndrome: alteration of the late reflex by lesion of the spinal tract and nucleus of the trigeminal nerve. J. Neural. Neurosurg. Psychiatry, 35: 228-233. Kimura, J, Powers, 1M and Van Allen, MW (1969) Reflex response of orbicularis oculi muscles to supraorbital nerve stimulation. Study in normal subjects and in peripheral facial paresis. Arch. Neurol., 21: 193-199. Kugelberg, E (1952) Facial reflexes. Brain, 75: 385-396. Lele, PP and Wedell, G (1959) Sensory nerves of the cornea and cutaneous sensibility. Exp. Neurol., 1: 334-359. McIntyre, AK and Robinson, RG (1959) Pathway for the jaw jerk in man. Brain, 75: 385-396. Miles, TS and Turker, KS (1987) Decomposition of the human electromyogramme in an inhibitory reflex. Exp. Brain Res., 65: 337-342. Miwa, H, Nohara, C, Hotta, M, Shimo, Y and Amemiya, K (1998) Somatosensory-evoked blink reflex response: investigation of the pathophysiological mechanism. Brain, 121: 281-291. Nakashima, K, Rothwell, JC, Thompson, PD et al. (1990) The blink reflex in patients with idiopathic torsion dystonia. Arch. Neurol., 47: 413-416.

M. ARAMIDEH ET AL.

Ongerboer de Visser, BW (1982) Afferent limb of the human jaw reflex: electrophysiologic and anatomic study. Neurology, 32: 536--546. Ongerboer de Visser, BW (1983) Anatomical and functional organization of reflexes involving the trigeminal system in man: jaw reflex, blink reflex, corneal reflex and exteroceptive suppression. Adv. Neurol., 39: 729-738. Ongerboer de Visser, BW (1986) The recorded corneomandibular reflex. Electroencephalogr. CUn. Neurophysiol., 63: 25-31. Ongerboer de Visser, BW and Goor, C (1974) Electromyographic and reflex study in idiopathic and symptomatic trigeminal neuralgias: Latency of the jaw and blink reflexes. 1. Neurol. Neurosurg. Psychiatry, 37: 12251230. Ongerboer de Visser, BW and Goor, C (1976) Jaw reflexes and masseteric electromyograms in mesencephalic and pontine lesions. An electrophysiologic study. J. Neural. Neurosurg. Psychiatry, 39: 90-92 Ongerboer de Visser, BW and Kuypers, HGJM (1978) Late blink reflex changes in lateral medullary lesions. An electrophysiological and neuroanatomical study of Wallenberg's syndrome. Brain, 101: 285-294. Ongerboer de Visser, BW and Moffie, D (1979) Effects of brainstem and thalamic lesions on the corneal reflex. An electrophysiological and anatomical study. Brain, 102: 595-608. Ongerboer de Visser, BW, Mechelse, K and Megens, PHA (1977) Corneal reflex latency in trigeminal nerve lesions. Neurology, 27: 1164-1167. Ongerboer de Visser, BW, Cruccu, G, Manfredi, M and Koelman, JHTM (1990) Effects of brainstem lesions on the masseter inhibitory reflex. Functional mechanisms of reflex pathways. Brain, 113: 781-792. Overend, W (1986) Preliminary note on a new cranial reflex. Lancet, I: 619. Pauletti, G, Berardelli, A, Cruccu, G, Agostino, R and Manfredi, M (1993) Blink reflex and masseter inhibitory reflex in patients with dystonia. Mov. Disord., 8: 495-500. Pennisi, E, Cruccu, G, Manfredi, M and Palladini, G (1991) Histometiric study of myelinated fibers in the human trigeminal nerve. J. Neural. Sci., 105: 22-28. Redmond, MD and Di Benedetto, M (1988) Hypoglossal nerve conduction in normal subjects. Muscle Nerve, 11: 447-452. Riggs, LA, Kelly, JP, Manning, KA and Moore, RK (1987) Blink-related eye movements. lnv. Ophthalmol. Vis. Sci., 28: 334-342. Shahani, BT (1970) The human blink reflex. J. Neurol. Neurosurg. Psychiatry, 33: 792-800. Shahani, BT and Young, RR (1972) Human orbicularis oculi reflexes. Neurology, 22: 149-154.

CRANIAL NERVE REFLEXES

Tavy, DL, Van Woerkom, TCAM, Bots, GTAM and Endtz, U (1984) Persistence of the blink reflex to sudden illumination in a comatose patient. Arch. Neurol., 42: 323-324. Thomas, PK and Mattias, CJ (1992) Diseases of the ninth, tenth and eleventh and twelfth cranial nerves. In: PJ Dyck and PK Thomas (Eds.), Peripheral Neuropathy (3rd ed). Philadelphia, Saunders, pp. 869-883. Tolosa, E, Montserrat, L and Bayes, A (1988) Blink reflex studies in focal dystonias. Enhanced excitability of brain stem intemeurons in cranial dystonia and spasmodic torticollis. Mov. Disord., 3: 61-69. Valls-Solt\ J, Cammarota, A, Alvarez, R and Hallett, M (1994) Orbicularis oculi responses to stimulation of nerve afferents from upper and lower limbs in normal humans. Neurosci. Lett., 650: 313-316. Van der Werf, F, Aramideh, M, Ongerboer de Visser, BW, Baljet, B, Speelman, JD and Otto, JA (1997) A

265 retrograde double fluorescent tracing study of the levator palpebrae superioris muscle in the cynomolgous monkey. Exp. Brain Res., 113: 174-179. Weiskrantz, L, Wanington, EK, Sanders, MD and Marshall, J (1974) Visual capacity in the hemianoptic field following a restricted occipital ablation. Brain, 97: 700-728. Wilkins, DE, Hallett, M and Wess, MM (1986) Audiogenic startle reflex of man and its relationship to startle syndromes: a review. Brain, 109: 561-573. Yates, SK and Brown, WF (198Ia) Light-stimulus-evoked blink reflex: methods, normal values, reaction to other blink reflexes, and observation in multiple sclerosis. Neurology, 32: 272-281. Yates, SK and Brown, WF (198Ib) The human jaw-jerk: electrophysiologic methods to measure the latency, normal values, and changes in multiple sclerosis. Neurology, 31: 632-634.

Movement Disorders Handbook of Clinical Neurophysiology, Vol. I M. Hallett (Ed.) © 2003 Elsevier B. V. All rights reserved

267 CHAPTER 18

Startle and prepulse effects Josep Valls-Sole" Unitat d'EMG, Servei de Neurologia, Departament de Medicina, Hospital Clinic, /nstitut d'Investigacio Biomedica August Pi i Sunyer (ID/BAPS), Universitat de Barcelona, Villarroel /70, Barcelona 08036, Spain

18.1. Introduction The startle reaction (SR) is an involuntary, reflex, behavioral reaction, generated in the brainstem reticular formation by unexpected external stimuli. It serves the purpose of protecting the organism from the effects of a presumed danger. Usually, the SR is induced by loud acoustic stimuli (auditory SR), although visual (Stitt et al., 1976; McManis et aI., 2001), somatosensory (Gokin and Karpukhina, 1985; Matsumoto et aI., 1992), and vestibular stimuli (Bisdorff et aI., 1994) have also been employed. The most apparent component of the auditory SR is its short latency motor response (Wilkins et al., 1986; Brown et aI., 1991a), which consists of the activation of many muscles, in a pattern that follows a rostrocaudal progression, in accordance to the spread of the excitatory volley presumably initiated in the caudal brainstem (Fig. 1A). A startle gives rise also to changes in autonomic function (Gogan, 1970; Gautier and Cook, 1997), neuroendocrine indices (Campeau et aI., 1997), and electroencephalographic activity (Andermann and Andermann, 1986) that have been less thoroughly investigated. These longlatency effects may be part of the so-called 'orienting' reaction (Turpin, 1986). Being a behavioral response, the SR is modified according to the emotional state, and hence has been used as a probe for such emotions as fear, aversion, pleasure, etc. (Lang et aI., 1990; Grillon et aI., 1991). The excitability of the SR is modulated not only by inputs from many central nervous system structures (Ho et al., 1987; Liegois-Chauvel et aI., 1989), but also by preceding stimuli, called 'prepulse'

* Correspondence to: Dr. Josep Valls-Sole, Unitat d'EMG, Servei de Neurologia, Hospital Clinic, Villarroel, 170, Barcelona 08036, Spain. E-mail address:[email protected] Tel.: 34-93-2275413; fax: 34-93-2275783.

stimuli. A prepulse stimulus is typically a stimulus of an intensity so weak that is incapable of inducing a response by itself. However it causes inhibition of the response to an incoming startling stimulus (Fig. 1B). Prepulse inhibition (PI), known also as 'reflex modification' (Hoffman and Ison, 1980; Ison and Hoffman, 1983), 'stimulus anticipation' (Ison et aI., 1990), or 'sensorimotor gating' (Rossi and Scarpini, 1992; Swerdlow et aI., 1992), does not depend on learning or on previous experiences with the same kind of stimuli. The effect might be due to the attentional shift required to process the information carried by the afferent volley generated by the prepulse stimulus (Graham, 1975; Blumenthal and Gescheider, 1987). PI may also be regarded as a method to assess sensorimotor integration at a subcortical level and, as a consequence, the technique is being increasingly used in the study of physiological mechanisms of behavior, and of the pathophysiology underlying certain diseases, in experimentation animals and in humans. In the following paragraphs, we will deal with some physiological aspects of SR and PI, emphasizing the utility of both as neurophysiological tools to examine and quantify functional aspects of the brainstem that are not readily available using other means of evaluation.

18.2. Startle 18.2.1. Circuits of the SR Research in animals has provided basic information on the circuits involved in the motor component of the auditory SR (Davis et aI., 1982, Lingenhohl and Friauf, 1994; Yeomans and Frankland, 1996). The response persists in experimental animals after intercollicular decerebration (Davis and Gendelman, 1977), and, therefore, it should depend on structures of the lower brainstem. Davis and co-workers (Davis

1. VALLS-SOLE

268

A

~~-I/~-~~-'\M 1

- -~,.

~r Acoustic stimulus

,~

I~\_,/.~,_-

.,~_~

V·l

Finger

Ilexors

_~~_._.

Orbicularis oculi

HIli

illS

B

~

__.-+-

Finger

w~

flexors

... Biceps

~---~------._------

Sterno mastoid

___ ___ _

Electrical stimulus (3'" finger)

~

__ • _Olbicularis oculi

Acoustic stimulus

Fig. 1. Startle reaction (A) and prepulse inhibition (B) in a healthy volunteer. (A) An auditory stimulus of an intensity of 130 dB is given at the vertical line. (B) The same stimulus is given 100 ms after a weak electrical stimulus to the 3rd finger (prepulse). Note the absence of the startle reaction in B.

et al., 1982) reported that, in the rat, the SR could be generated by electrical stimulation of the nucleus reticularis pontis caudalis (nRPC), but not of the nucleus reticularis gigantocellularis (Davis et al.,

1982), and proposed a circuit consisting of the cochlear nucleus, the nucleus of the lateral lemniscus, the nRPC, and the brainstem and spinal motoneurons. Later, Lee et al. (1996) found that the cochlear nucleus and the nRPC, but not the nucleus of the lateral lemniscus, were required for the response, and Lingenhohl and Friauf (1994) described that the nRPC received direct inputs from the cochlear nuclei. The nRPC projects to the brainstem and spinal cord alpha motoneurons via the medial reticulospinal tract. In present days, the simplest circuit of the auditory SR is believed to involve the cochlear nuclei, the nRPC, and the alpha motoneurons (Davis, 1996; Yeomans and Frankland, 1996; Koch, 1999). The giant neurons of the nRPC are not modality specific (Wu et al., 1988), likely responding to inputs other than those generated by acoustic stimuli. It is conceivable that somatosensory, visual and vestibular inputs induce startle responses through activation of the same efferent circuit. There is no reason to suggest that the SR in humans uses other circuits. Instead, there are several lines of evidence in favor of the origin of the human SR at the lower brainstem: It is present in anencephalic newborns (Brown et al., 1991a); the pattern of muscle recruitment is consistent with a brainstem origin (Wilkins et al., 1986; Brown et al., 1991a), and symptomatic lesions causing enhanced or depressed SR involve the caudal brainstem reticular formation (Brown et al., 1991b; Vidailhet et al., 1992). In humans, the muscle activated with the shortest latency is the orbicularis oculi, followed by the sternocleidomastoid, masseter, and the muscles of the upper and lower limb (Table 1). Some authors believe that the earliest part of the orbicularis oculi response (beginning at about 40 ms) is actually an auditory blink reflex (ABR), separated from the SR, while the true startle reaction starts at a relatively longer latency, and overlaps with the ABR (Brown et al., 1991a). These authors suggest that the physiological mechanisms of the ABR differ from those of the auditory SR. This theory is supported by the anatomic description of a specific circuitry for the ABR, involving the inferior colliculus and the mesencephalic reticular formation (Hori et al., 1986). The possibility of physiological differences between the ABR and the auditory SR is an important point because of the large amount of

269

STARTLE AND PREPULSE EFFECTS

Table 1 Latencies in ms of the auditory startle reaction measured in 15 healthy volunteers, aged 31 to 56 y.o. Mean

SD

Orbicularis oculi

48.3

6.9

Sternocleidomastoid

53.9

10.4

Masseter

54.1

9.2

Wrist flexors

74.3

8.6

Wrist extensors

75.1

9.5

Tibialis anterior

123.5

19.4

literature published in which the SR was identified with just the responses recorded in the orbicularis oculi (Graham, 1975; Lang et aI., 1990; Boelhouwer et al., 1991; Swerdlow et al., 1995). In theory, the findings reported in studies exploring only the responses of the orbicularis oculi could be related to the ABR but not necessarily to the auditory SR. However, the idea that the ABR and the auditory SR have separate anatomic pathways and physiological mechanisms is not universally accepted. From the anatomical point of view, Koch (1999) reported that, in animals, the same nRPC neurons mediate facial and limb muscle responses, suggesting that the cranial nerve and spinal alpha motoneurons respond to inputs from the same source. From the physiological point of view, Valls-Sole et aI. (1999a) reported that responses to auditory stimuli limited to the orbicularis oculi (and, therefore, considered to be ABRs) were modulated by prepulse stimuli in the same way as those involving also the masseter and the sternocleidomastoid muscles (and, therefore, considered to be auditory SRs). An alternative view to the understanding of the SR as the result of single or multiple excitatory volleys spreading from the nRPC was given by Bisdorff et aI. (1994) on the basis of their analysis of the SR generated through vestibular inputs. These authors hypothesized that the differences between onset of EMG activity in various muscles are not simply governed by the respective distance from the brainstem nuclei but that the SR is the expression of a patterned response, built in the reticular formation with a temporo-spatial distribution. This pattern would be specifically modulated by sensory inputs, according to the examination conditions, and by the

biological weight of certain muscle components. A similar reasoning was put forward by Matsumoto et aI. (1992) in their explanation of the results obtained in patients with hyperekplexia. If this is the case, the different behavior of the blink component of the SR could be explained by it being relatively hard-wired in the reticular formation (Jenny and Saper, 1987), reflecting the predominant role of the mechanisms protecting the eye during potentially harmful stimuli. 18.2.2. Long latency motor and non-motor aspects of the SR

A startling stimulus generates not only the short latency SR, but other effects occurring at a longer latency (Gogan, 1970). The orienting reaction (OR) is an ill-defined behavioral response that depends on psychological factors and may contain effects related to the subject's curiosity, fear, annoyance, etc. The motor component of the OR can account for complex movements, and may involve defense reactions and the 'flight or fight' response (Turpin, 1986). In some of our subjects, we observed an OR consisting in the turning of the head towards the source of the loud noise, a relatively long lasting activity of cranial nerve innervated muscles (Fig. 2), the articulation of a few words of surprise, or expressions of emotion such as laughing or guttural noises. It is difficult to know at present whether there is any relationship between the long-latency motor phenomena induced by a startling stimulus in normal subjects and the tonic spasms observed in patients with hyperekplexia (Saenz-Lope et al., 1984; Brown et al., 1991b) or startle epilepsy (Anderman and Anderman, 1986; Brown et aI., 1991b). The EEG correlates of the normal human startle reflex are not clear. Most descriptions report that the unavoidable artifact related to blinking is followed by a brief desynchronization of the background EEG activity (Wilkins et aI., 1986). The same findings have been reported in patients with hyperekplexia, although some of them might actually have enhanced excitability of the motor cortex manifesting with enhanced amplitude of somatosensory evoked potentials (Markand et al., 1984). In startle epilepsy, ictal activity is delayed with respect to the changes related to the initial startle reflex (Bancaud et al., 1975). Autonomic changes induced by a startling stimulus can be part of the OR, and include the

270

J. VALLS-SOLl~,

Fig, 2. Two motor components of the startle reaction. The short latency component (shown as the bursts close to the stimulus, applied at the vertical line) is followed by a long latency reaction that may be part of the orienting reaction (underlined),

elicitation of the galvanic cutaneous sudomotor response, and a relatively long-lasting modulation of blood pressure and heart rate (Dimberg, 1990; Gautier and Cook, 1997; Holand et al., 1999; VallsSole et al., 2002). 18.2.3. Other features related to the external activation of the startle pathways

Apart from the generation of an overt short latency motor response, and other long latency changes, the auditory SR has other features of physiological interest such as the generation of the PI potential, the interindividual variability and habituation rate, the effects induced on voluntary motor reactions, and the modulation of segmental

reflexes. Table 2 summarizes the motor and nonmotor effects that can be induced by an auditory startling stimulus. 18.2.3.1. The P1 potential Any auditory stimulus generates auditory evoked potentials of short, medium and long latency.Among these responses, the one that has been thought to be generated in startle-related structures is the PI potential (Reese et al., 1995). This potential is recorded in the vertex at a latency of about 50 ms and is preceded by another potential known as the Pa potential (Buchwald et al., 1991). In contrast to the Pa potential, the PI potential amplitude decreases significantly as the stimulus rates exceed l/s. Also, intravenous injection of cholinergic antagonists

271

STARTLE AND PREPULSE EFFECTS

Table 2 Some effects of an auditory startling stimulus. Auditory blink reflex

Fox, 1979; Hori et al., 1986

Short latency startle reaction

Davis et al., 1982; Wilkins et al., 1986; Brown et aI., 1991

Reflex habituation and sensitization

Davis et aI., 1982; Rimpel et al., 1982

Reflex modification

Hoffmann and Ison, 1980; Ison et al., 1990

Orienting reaction

Gogan, 1980

Autonomic reactions

Turpin, 1986

PI potential

Reese et al., 1995

Acceleration of reaction time

Valls-Sole et aI., 1995; Valldeoriola et al., 1998

Audiospinal reflex

Rossignol and Melvill-Jones, 1976; Delwaide et aI., 1993

resulted in a decrease of the PI potential. These characteristics were considered as indicating that the PI potential is generated in the reticular ascending system, and likely in the pedunculopontine tegmental nucleus (PPTn) (Reese et aI., 1995). An evoked potential with characteristics similar to those of the PI potential was recorded in the rat, near the PPTn and the mesencephalic locomotor region, Ebert and Ostwald (1991). Ninomiya et aI. (1997) suggested that the PI potential is the result of overlapping of temporally coincident evoked potentials, and that part of the potential might relate to a motor response process. 18.2.3.2. 1nterindividual variability and habituation The magnitude of the startle reaction varies from subject to subject, making it difficult to draw normative values (Chokroverty et aI., 1992). However, one can reasonably expect that the first time a loud auditory stimulus is applied, it causes short latency responses seen as EMG bursts in at least several cranial and neck muscles. Apart from interindividual differences, groups of normal subjects have been identified to respond differently to a startling stimulus. Variables such as age, gender, smoking habits, and body mass might be important (Kumari et aI., 1996; Faraday et aI., 1999; Lehmann et al., 1999; Kofler et aI., 200la, b). The size of the responses can be potentiated or inhibited depending in part on the subject's posture (Brown et aI., 1991c). However, in any given posture, the short latency SR fades away quite fast with repeated stimuli in normal subjects (Wilkins et

aI., 1986; Brown et aI., 1991a; Chokroverty et al., 1992; Matsumoto et aI., 1992). Habituation may be indeed inconvenient for clinical applicability of the startle reaction, because the examiner may not have the opportunity to replicate the results obtained from the first stimulus. In a study of habituation rate of SR in different environmental and behavioral conditions, Valls-Sole et aI. (1997) found significantly reduced habituation rate when subjects were prepared to execute a ballistic movement in a reaction time task paradigm, in comparison to other conditions (Fig. 3). This observation suggests that preparation for execution of a voluntary movement causes an increase in the excitability of the circuits conveying the SR. At the same time, the voluntary

Fig. 3. Comparison of habituation rate in the responses recorded in the orbicularis oculi and the sternocleidomastoid muscles in two different conditions: Quiet environment and preparedness to perform a ballistic movement. Histograms represent the mean size of the responses of 15 healthy volunteers to startling stimuli delivered every 3Q-.45 s in each condition, in percentage of the size of the response to the first stimulus. Note the reduced habituation of the response recorded in the sternocleidomastoid in the condition of preparedness.

J. VALLS-SOLl~

272

reaction was speeded up but not otherwise modified by the startling stimulus (Valls-Sole et al., 1995, 1999b), suggesting that the circuits of the SR might have been used in the execution of the voluntary ballistic movement. 18.2.3.3. The effects of a startle on reaction time The execution of ballistic movements in a reaction time task paradigm involves activation of many descending motor tracts, but the relative role of each one of them in the accurate performance of fast movements is mostly unknown (Porter and Lemon, 1993). That the corticospinal tract is involved in the execution of fast ballistic movements in humans is assumed in part from the observations of an increase in the excitability of the motor pathway to transcranial electrical and magnetic stimulation, beginning at about 80 ms before onset of the EMG activity in the agonist muscle (Starr et aI., 1988; Pascual-Leone et aI., 1992). When testing such an effect, Pascual-Leone et aI. (1992) observed also that subthreshold transcranial magnetic stimulation (TMS) induced acceleration of the voluntary reaction. Reaction time shortening by subthreshold TMS was interpreted as the consequence of an externally induced acceleration in the process of motor cortex energization to reach the level of excitability appropriate for activation of the execution channel. However, alternative explanations were given such as the possibility to activate subcortical motor structures. We found out later that a startling stimulus, which is supposed to directly activate subcortical motor structures, induced even more acceleration of the reaction time than TMS (Valls-

Sole et al., 1995). Furthermore, the shortening effect involved the whole three-burst pattern constituting the motor program for ballistic movements (Hallett et aI., 1975). In such a study, we asked healthy volunteers to perform a ballistic wrist movement. They were trained to produce the triphasic agonistantagonist-agonist EMG pattern by vigorous activation of the wrist extensors and wrist flexors. In test trials, a startling stimulus was delivered together with the imperative signal. The mean results for all subjects as a group are summarized in Table 3, and individual mean data are shown in Fig. 4. In test trials, the latency of the first agonist EMG burst was significantly shorter than in control trials, but the configuration of the triphasic pattern did not change between control and test trials. These results indicate that the whole programmed three bursts pattern of the ballistic movement was moved to a shorter latency, as if it was triggered by the startle itself. Our interpretation is that the excitability of the pathways involved in the startle reaction was modulated according to the patterned three burst response during preparation for the execution of the ballistic movement. If this were not the case, one should have expected a significant change in the configuration of the three burst pattern. These observations suggest that the reticulospinal tract is actively involved and plays a most important role in the execution of ballistic movements (Valls-Sole et aI., 1999b). Startle-induced reaction time shortening offers an interesting question to the physiologist: Is there enough time for cortical sensory processing before execution of the 'voluntary' task? Preparation for the

Table 3 Acceleration of reaction time by a startle. Test

Control Movement of the hand

203.6

(74.5)

104.3

(17.6)*

Latency of 1st agonist burst

171.4 (50.9)

77.3

(10.7)*

72.9

(11.2)

35.6

(10.4)

Duration of 1st agonist burst Interval agonist-antagonist

76.6

(15.6)

36.2

(11.9)

Interval 1st-2nd agonist bursts

143.2 (34.6)

138.4 (24.8)

Duration of antagonist burst

121.5 (47.0)

109.3 (21.3)

Measures of wrist movement and the electromyographic three bursts pattern in control condition (control) and when the 'go' signal was accompanied by a startling stimulus (test). * Statistically significant differences with respect to control.

J

I

'go' signal

I

.'

I "'1'1

'~~~T . l' f1,. . ~W

MOV

- \VI'

lOOms

WF

~ ' .d ,~"' ......,"t";': ': .·~'l"WE '~'"

MOV

o Control trials A Test trials

~60

~gl-dur ~80

~100

~140

Agl-Ag2 . .

--¥r-

~~

Ag-Antag

B

40 I

o

100 J

80

120

100) 160

oo~

60 40 20 0

i

I

50

A

•

~A

A

A

A

0

o

0

0

0 0

o

0

,---.-

o

0 0 0

0

150

o

0

0

I

200

o

o

8

0

0

---.

Reaction time (ms)

•

0 0

100

I

0 0

o

0

0

A O

AA

A

A~AtA

~

A AAA4

Fig. 4. Acceleration of the reaction time, with no other changes in the three burst EMG pattern of a ballistic movement, when subjects receive an acoustic startling stimulus at the same time as the imperative visual 'go' signal, in a reaction time task paradigm. The graphs in A show recordings from a representative individual. The top graph corresponds to a trial with no startle, while the bottom graph corresponds to a trial with startle. The graphs in B show the mean results in 10 healthy subjects. The graphs plot the reaction time, in the horizontal axis, against three different measurements describing the intrinsic characteristics of the three burst pattern, in the vertical axis. Open circles correspond to trials with no startle, while filled triangles correspond to trials with startle.

startle

+

'go' signal

-+jr~~

~~

,

A

N -.J

w

til

i--l

o

~

!iJ

tIl

til

ce-

~

0

~

[;;

i--l

~

til

274

voluntary reaction to a sensory stimulus of a specific modality might block the input from other sensory modalities in their pathway to the cognitive centers, while the excitability of the motor pathway may be enhanced to prepare for the execution of a voluntary reaction (Brunia, 1993). In these conditions, the startling stimulus may trigger the whole reaction, before all sensory information processing has been completed. 18.2.3.4. Audiospinal reflex An auditory stimulus causes changes in the excitability of some reflex circuits. The effect on the H-reflex was first documented by Rossignol and Melvill Jones (1976), and later confirmed and further explored by Delwaide and coworkers, who applied the technique to the study of pathophysiological mechanisms of rigidity in parkinsonian patients (Delwaide et aI., 1993, 2000; Delwaide and Schepens, 1995). The audiospinal reflex consists of the facilitation of the H-reflex, beginning 50 to 60 ms after a sound stimulus, and lasting about 200 ms. It has been found to be of similar size and resolution in both tibialis anterior and soleus, which is taken as a proof that the audiospinal reflex is not a flexor reflex (Delwaide and Schepens, 1995). Similar long-latency effects on the soleus H-reflex have been found using TMS rather than auditory stimuli (Wolfe et aI., 1996; Goulart et aI., 2000). TMS induces two excitatory phases on the H-reflex, the first one being likely due to the generation of a motoneuronal EPSP from activation of the corticospinal tract (Goulart et aI., 2000). The second facilitatory phase peaks at about 70 ms, and is of approximately the same size as the first phase. Its origin is not clear, but, because of its similarity with the audiospinal reflex (Fig. 5), it has been suggested to arise from activation of the reticulospinal tract through converging sources of inputs. These would include somatosensory inputs to the trigeminal nerve from the TMS induced contraction of scalp muscles, auditory inputs from the acoustic stimulus generated by the magnetic coil, and descending inputs from activation of cortico-reticular connections (Goulart et aI., 2000).

18.3. Prepulse 18.3.1. Prepulseeffects on the startle reaction

The size of the auditory SR is reduced when the auditory stimulus is preceded by a stimulus of an

J. VALLS-SOLE

intensity so weak that is unable to elicit a response by itself. This effect, presently known as prepulse inhibition (PI), was first described as the phenomenon of 'reflex modification' more than 30 years ago by Hoffman and coworkers (see the reviews published by Graham, 1975; Hoffman and Ison, 1980; Ison and Hoffman, 1983). Table 4 shows a list of the different observations found in the literature with regard to the effects of a weak stimulus on the response to a subsequent strong stimulus. PI is a very robust phenomenon that occurs with stimuli of the same or different modality as the one eliciting the SR, is similar with stimulus onset and offset, and does not require learning, being present from the first time the prepulse stimulus is applied (Graham, 1975). Most studies of PI have been carried out in the domain of psychophysiology and behavioral psychology, where PI has been found to be abnormally reduced in psychosis prone and schizophrenic subjects (Braff and Geyer, 1990; Simons and Giardina, 1992; Bradley et aI., 1993). The cause of this abnormality can be the attentional deficit and information processing abnormalities exhibited by these patients, who, according to Braff and Geyer (1990), have 'normal sensory registration but defective acute directed attention'. The impairment of internal screening causes failure of information selection, with the consequence of a continuous bombardment of undifferentiated and involuntary data. Actually, normal subjects are probably able to integrate at a subcortical level inputs generated by environmental conditions of our daily life, such as visual, acoustic, somatosensory or others. In certain conditions, these punctiform environmental inputs may play the role of prepulse stimuli, causing inhibition of the reaction to unwanted interfering stimuli. Hoffman and Fleschler (1963) reported the observation of significantly reduced SRs in rats in an environment containing pulsed noise with respect to another in which there was a steady background noise. In our experience, the SRs in humans are largest when uncontrolled environmental punctiform stimuli are limited by steady background noise (Valls-Sole et aI., 1997). 18.3.2. Prepulse effects on the blink reflex and on other facial reflexes

The blink reflex is a very useful tool for the study of some integrative functions taking place in the brain stem. Most of the methods allowing for such a

Fig. 5. Modulation of the soleus If-reflex by auditory stimuli (A and C), corresponding to the audiospinal reflex, and by transcranial magnetic stimulation (B and D). A and B show the superimposition of 12 traces in which the H-reflex was elicited by a posterior tibial nerve stimulus delivered at the interval marked in the horizontal axis with respect to the conditioning stimulus, either the auditory stimulus (A) or the transcranial magnetic stimulus (B), given always at onset of the trace. C and D show the mean and one standard deviation of the soleus H-reflex amplitude in test trials with respect to control trials, marked with the letter 'C', in a group of 15 healthy subjects. The bars under C and D unerline the intervals at which the amplitude of the Il-reflex was statistically significantly larger than in the control condition. Note the similarities in the long latency facilitation by each of the two stimuli.

tv

VI

-:I

en

-3

~

tIl

en

~

~

s;? ~ ~

en

276

J. VALLS-SOLl~

Table 4 Some effects seen with a pair of stimuli. Effect

Reference

Reflex modification

Hoffmann and Ison, 1980; Ison et al., 1990

Gating of reflexes

Rossi and Scarpini, 1992

Prepulse effects

Graham, 1975; Boelhouwer et aI., 1991; Swerdlow et aI., 1995

Postpulse effects

Valls-Sole et aI., 1996

Classical conditioning

Gormezano, 1966

study make use of the collision between inputs of different sensory modalities. Orbicularis oculi responses to stimuli of a certain modality are modulated by inputs from other sensory modalities within a certain time period (Rimpel et al., 1982). When the conditioning stimulus is of suprathreshold intensity and induces a response in the orbicularis oculi muscle itself, the effects on the response to a stimulus of another sensory modality can be explained by suggesting refractoriness in a common polysensory interneuronal circuit (Rimpel et al., 1982). However, the same effect may also be obtained when using a prepulse stimulus (i.e. a stimulus of an intensity sub-threshold for elicitation of a response). The effects of prepulse stimuli were originally investigated on the blink reflex generated by auditory stimuli (Graham, 1975). However, similar observations have been reported with the blink reflex responses elicited by supraorbital nerve stimulation (Ison et al., 1990), and using other prepulse modalities (Rossi and Scarpini, 1992; Valls-Sole et al., 1994). Investigation of the prepulse effects through the electrically induced blink reflex brings the possibility to study separately the effects occurring on the Rl and on the R2. In most studies, a prepulse causes facilitation of the R1 at intervals between 40 and 100 ms, and inhibition of R2 (and of R2c) at intervals beyond 70 ms (Ison et al., 1990; Boelhouwer et al., 1991; Boulu et al., 1991; Rossi and Scarpini, 1992; Valls-Sole et al., 1994). The dissociation between the effects of a prepulse on the Rl and R2 responses suggest that the effects occur at a presynaptic level (Ison et al., 1990; Rossi and Scarpini, 1992). Prepulse effects can also be induced by activation of small myelinated fibers with laser stimuli (Valls-Sole et al., 2000a). In these instances,

the inhibitory effect occurs between 200 and 300 ms after the prepulse, the delay being related to the slow conduction of the afferent pathway. Prepulse stimuli are not only effective on the blink reflex but also on other facial reflexes. We investigated the effects of somatosensory prepulses on the masseteric silent period (Gomez-Wong and VallsSole, 1996), and saw that the second phase of the masseteric silent period to mentalis nerve stimulation (SP2) is significantly reduced with respect to control conditions when a weak electrical stimulus is applied to the 3rd finger 100 ms before. The percentage reduction of SP2 habituation was not different from the percentage reduction of the R2 during voluntary facilitation. These observations suggest that the effects of a prepulse take place at the trigeminal nerve afferents themselves, before synapsing with facial or trigeminal motoneurons. This is in keeping with the effects of inputs from peripheral nerves reported in experimental animals (Baldissera et al., 1967). 18.3.3. The physiological mechanisms of the PI

The circuit of prepulse inhibition has been delineated in animal studies (Koch et al., 1993; Swerdlow and Geyer, 1993; Inglis and Winn, 1995). Among the structures involved, the subthalamic and pallidal projections to the pedunculo-pontine tegmental nucleus (PPTn), and the cholinergic neurons that project from the PPTn to the nucleus reticularis pontis caudalis (nRPC), playa major role. Prepulse inhibition is preserved in decerebrate animals (Fox, 1979). Therefore, inputs generated by prepulse stimuli should access a group of neurons with capability of sensori-motor integration in the prepulse circuit by way of subcortical pathways. One possible structure that conforms with these require-

277

STARTLE AND PREPULSE EFFECTS

ments is the caudal pontine or pontomedullary reticular formation, and more specifically, the nRPC itself, which would activate the PPTn through reciprocal pathways (Swerdlow and Geyer, 1993) and initiate an inhibitory response. If this is the case, it should be expected that the prepulse causes subthreshold activation of the pontomedullary reticular formation before actually exerting its inhibitory action. The expected facilitation is actually found with short interval somatosensory prepulses on acoustically induced blink reflexes. To analyze better the differences between prepulse effects over the blink reflex and over the ABR, we studied the effects of auditory and somatosensory prepulses on the auditory and electrically induced blink reflexes (Valls-Sole et aI., 1999a). The modulatory effects of somatosensory prepulses on the ABR were facilitatory with interstimuli intervals of up to 50 ms, and inhibitory with interstimuli intervals of 60 ms or beyond. Regarding the electrically induced blink reflex, both somatosensory and auditory prepulses facilitated the R1 between 40 and 80 ms, and inhibited the R2 between 70 and 150 ms. We did not observe any facilitation of the electrically induced R2, which is in contrast with the facilitation observed in the auditory blink reflex by somatosensory stimuli. Therefore, although PI is indeed a cross-modality phenomenon, subtle differences can be seen between modalities. Some differences between auditory and somatosensory prepulses can be explained according to the differences in the relative length of the circuit for somatosensory and auditory inputs and, consequently, in the arrival time of the afferent volley to the brainstem centers (Fig. 6). Both stimuli cause a

Prepulse

Stimul:.~~~~~~= Responseeliciting stimulus

Second modality '--,------,."-" pathway Integrative polysensory system

Fig. 6. Diagram of the prepulse inhibition of a startle reaction. A prepulse stimulus reaches first the brainstem centers causing moderate facilitation of the efferent structures and, at the same time, engage in a prepulse circuit with inhibitory effects over the startle reaction.

window of facilitation followed by a long-lasting inhibition. Because of the different distance, afferents from auditory stimuli arrive earlier to the brainstem than those generated by somatosensory stimuli. Therefore, the window of facilitation involves only the Rl response, because of its shorter latency. On the contrary, since the ABR has a fairly short latency compared to the time at which the afferents from somatosensory prepulses will reach the brainstem, the whole ABR falls into the window of facilitation at short inter-stimulus intervals.

18.3.4. The relationship between PI and blink reflex excitability

In PI, the blink reflex is used as a probe. For this reason, it has been proposed that the results obtained with examining prepulse modulation would in part depend on the excitability of the blink reflex circuit (Schicattano et aI., 2000). However, although reduced PI goes together with enhanced blink reflex excitability recovery in patients with Parkinson's disease as well as in patients with cranial dystonia (Nakashima et aI., 1993; Gomez-Wong et al., 1998; Schicatano et aI., 2000), such coincidence does not occur in other groups of patients. One example is found in patients with Huntington's disease (HD). In these patients, several authors (Esteban and Gimenez-Roldan, 1975; Agostino et aI., 1988) have reported an abnormal decrease in blink reflex excitability, while others have reported reduced PI (Swerdlow et al., 1995). We investigated this aspect as part of a larger study of patients with HD, and confirmed that some patients might indeed have abnormally depressed blink reflex excitability together with an abnormally reduced PI (Valls-Sole, 2000). Thus, the results of prepulse studies could not be predicted from the observations made when studying blink reflex excitability in patients with HD. Why patients with Parkinson's disease and patients with Huntington's disease might have similarly reduced prepulse inhibition but opposite abnormalities in blink reflex excitability recovery is presently an unexplained observation. It is known that the reticular formation has reciprocal connections with nuclei of the basal ganglia. These connections use mainly two different pathways. One

278

might be implicated in the regulation of blink reflex excitability. and involves the internal globus pallidus (GPi) and the substantia nigra pars reticulata (SNr), which send inhibitory inputs to the superior collicuIus. The superior colliculus sends excitatory inputs to the nucleus raphe magnus, which inhibits the excitability of trigeminal neurons (Basso et al., 1996a, b). Since the GPi/SNr complex is hyperactive in Parkinson's disease and hypoactive in Huntington's disease, the effects on blink reflex should be opposite in the two diseases (Kimura, 1973; Esteban and Gimenez-Roldan, 1975). Another link between the basal ganglia and the brains tern reticular formation is established through the PPTn, which is reciprocally connected mainly to the GPi/SNr complex and the subthalamic nucleus (STn), and could have some role in the regulation of prepulse modulation of the startle reaction (Koch et al., 1993; Inglis and Winn, 1995). In Parkinson's disease, all output connections from the basal ganglia to the PPTn are hyperactive, the output of the STn being excitatory and those from the GPi and the SNr being inhibitory. According to current concepts on how the basal ganglia are interconnected, the hyperfunctioning STn will cause even more hyperactivity of the GPi, probably resulting in a net inhibitory output to the PPTn. In Huntington's disease, both the STn and the GPi/SNr are hypofunctioning. However, this is not exactly the opposite behavior with respect to PD. The reduced excitatory input from the STn to the GPi could cause this nucleus to be more dependent on the inputs from the direct pathway, which is also the one more likely implicated in dopamine-related influences on prepulse inhibition (Swerdlow et al., 1992). The decreased inhibition that the GPi receives from the direct pathway could oppose or even overcome the increased inhibition from the indirect pathway and the resulting output from the GPi could be negligible or even turn to more inhibition than normal. Whatever the case, the reduced excitatory output from the STn would cause reduced activation of the PPTn, resulting in an effect similar to the one occurring in PD patients. Therefore, even if the same prepulse effects are seen in PD and HD patients, the mechanisms inducing these effects could be different. They could involve excess inhibition from the GPi to the PPTn in PD, and reduced excitation from the STn to the PPTn in HD (Fig. 7).

J. VALLS-SOLl~

APD

B:HD

Fig. 7. Circuits of the basal ganglia suggested to control the activity in the pedunculopontine tegmental nucleus in relation to prepulse inhibition in Parkinson's disease (A) and in Huntington's disease (B).

18.3.5. Postpulse effects In a paired stimulation paradigm the effects are not limited to the response induced by the second stimulus. Repeated presentation of a trigeminal nerve stimulus generates excitability enhancement in neurons in or near the facial nerve nucleus that may contribute in part to the generation of the conditioned response (Matsumura and Woody, 1986).

STARTLE AND PREPULSE EFFECTS

279

Fig. 8. Postpulse effect in a series of 5 traces combining mechanical taps to the forearm and electrical stimuli to the supraorbital nerve. In all columns, the series of 5 stimuli begin at the bottom of the trace and go upwards. Recordings are always done in the O. oculi of both sides (the right side is the top trace of each pair). In A, supraorbital nerve electrical stimuli were given at the vertical line. The R1, R2 and R2c responses show some tendency to diminish their size with successive trials. In B, a tap is applied to the forearm at onset of the traces, and responses are present in the O. oculi to just the first and the second stimuli. In C, forearm taps and supraorbital nerve electrical stimuli are applied synchronously, separated by an interval of 100 ms. See that the R2 and R2c responses to the supraorbital nerve electrical stimulus are inhibited (prepulse inhibition), and that the O. oculi responses to the tap remain present up to the 5th trial, indicating reduced habituation (postpulse effect).

We investigated the effects of repeated presentations of paired low intensity (prepulse) auditory stimuli and supraorbital nerve stimuli of an intensity leading to the known blink reflex responses. Apart from the expected prepulse inhibition of the R2 response, we observed that the small responses induced occasionally by the low intensity auditory stimuli were larger and showed less habituation when given paired to the supraorbital nerve stimulus than when given on their own (Fig. 8). This effect, described as the 'postpulse' effect, might have physiological bases akin to those underlying pavlovian conditioning (Woodruff-Pak and Thompson, 1988; Solomon et aI., 1989) but differs from it in two important respects (Valls-Sole et al., 1996): (l) The postpulse effect occurs earlier than the acquisition of the conditioned response; and (2) While postpulse enhancement relates to the response time locked to the first stimulus of a pair, the conditioned response is time locked to the presentation of the second stimulus.

References Agostino, R, Berardelli, A, Cruccu, G, Pauletti, G, Stocchi, F and Manfredi, M (1988) Correlation between facial involuntary movements and abnormalities of blink and corneal reflexes in Huntington's chorea. Mov. Disord., 3: 281-289. Anderrnann, F and Anderrnann, E. Excessive reflex syndromes: startle disease, jumping and startle epilepsy. Adv. Neurol., 43: 321-338. Bancaud, J, Ta1airach, J, Lamarche, M, Bonis, A and Trottier, S (1975) Hypotheses neuro-physio-pathologiques sur l'epilepsie-sursaut chez l'homme. Revue Neurologique, 131: 559-571. Basso, MA and Evinger, C (1996) An explanation for reflex blink hyperexcitability in Parkinson's disease. II Nucleus raphe magnus. J. Neurosci., 16: 7318-7330. Basso, MA, Powers, AS and Evinger, C (1996) An explanation for reflex blink hyperexcitability in Parkinson's disease. I. Superior colliculus, J. Neurosci., 16: 7308-7317. Bisdorff, AR, Bronstein, AM and Gresty, MA (1994) Responses in neck and facial muscles to sudden free

280 fall and a startling auditory stimulus. Electroencephalogr. Clin. Neurophysiol., 93: 409-416. Blumenthal, TD and Gescheider, GA (1987) Modification of the acoustic startle reflex by a tactile prepulse: the effects of stimulus onset asynchrony and prepulse intensity. Psychophysiology, 24: 320-327. Boelhouwer, AJW, Teurlings, RJMA and Brunia, CHM (1991) The effect of an acoustic warning stimulus upon the electrically elicited blink reflex in humans. Psychophysiology, 28: 133-139. Boulu, Ph, Willer, JC and Cambier, J (1981) Analyse electrophysiologique du reflexe de clignement chez l'homme: interaction des afferences sensitives segmentaires et intersegmentaires, des afferences auditives et visuelles. Rev. Neurol., 137: 523-533. Bradley, MM, Cuthbert, BN and Lang, PJ (1993) Pictures as prepulse: attention and emotion in startle modification. Psychophysiology, 30: 541-545. Braff, DL and Geyer, MA (1990) Sensorimotor gating and schizophrenia. Arch. Gen. Psychiatry, 47: 181-188. Brown, P, Day, BL, Rothwell, JC, Thompson, PD and Marsden, CD (1991a) The effect of posture on the normal and pathological auditory startle reflex. J. Neurol. Neurosurg. Psychiatry, 54: 892-897. Brown, P, Rothwell, JC, Thompson, PD, Britton, TC, Day, BL and Marsden, CD (1991b) New observations on the normal auditory startle reflex in man. Brain, 11: 1891-1902. Brown, P,Thompson, PD, Rothwell, JC, Britton, TC, Day, BL and Marsden, CD (1991c) The hyperekplexias and their relationship to the normal startle reflex. Brain, 114: 1903-1928. Brunia, CHM. Waiting in readiness: gating in attention and motor preparation. Psychophysiology, 30: 327339. Buchwald, JS, Rubinstein, EH, Schwafel, J and Strandburg, RJ (1991) Midlatency auditory evoked responses. Differential effects of a cholinergic agonist and antagonist. Electroencephalogr. Clin. Neurophysiol., 80: 303-309. Chokroverty, S, Walczak, T and Hening, W (1992) Human startle reflex: technique and criteria for abnormal response. Electroencephalogr. Clin. Neurophysiol., 85: 236-242. Campeau, S, Falls, WA, Cullinan, WE, Helrnreich, DL, Davis, M and Watson, SJ (1997) Elicitation and reduction of fear: behavioral and neuroendocrine indices and brain induction of the immediate-early gene c-fos. Neuroscience, 78: 1087-1104. Davis, M (1996) Differential roles of the amygdala and bed nucleus of the stria terrninalis in conditioned fear and startle enhanced by corticotropin-releasing hormone. In: T Ono, BL McNaughton, S Molotchnikoff,

J. VALLS-SOLl~

ET Rolls and H Nishijo (Eds.), Perception, Memory and Emotion. Oxford: Elsevier Ltd., pp. 525-548. Davis, M and Gendelman, PM (1977) Plasticity of the acoustic startle response in the acutely decerebrate rat. J. Compo Physiol. Psychol., 91: 549-563. Davis, M, Gendelman, DS, Tischler, MD and Gendelman, PM (1982) A primary acoustic startle circuit: lesion and stimulation studies. J. Neurosci., 2: 791-805. Delwaide, PJ and Schepens, B (1995) Auditory startle (audio-spinal) reaction in normal man: EMG responses and H-reflex changes in antagonistic lower limb muscles. Electroencephalogr. Clin. Neurophysiol., 97: 416-423. Delwaide, P, Pepin, JL and Maertens de Noordhout, A (1993) The audiospinal reaction in parkinsonian patients reflects functional changes in reticular nuclei. Ann. Neurol.,33: 63-69 Delwaide, P, Pepin, JL, De Pasqua, V and Maertens de Noordhout, A (2000) Projections from the basal ganglia to tegmentum: a subcortical route for explaining the pathophysiology of Parkinson's disease signs? J. Neurol., 247 (Suppl. 2): 75-81. Dimberg, U (1990) Facial electromyographic reactions and autonomic activity to auditory stimuli. Biol. Psychol., 31: 137-147. Ebert, U and Ostwald, J (1991) The mesencephalic locomotor region is activated during the auditory startle response of the unrestrained rat. Brain Res., 565: 209-217. Esteban, A and Gimenez-Roldan, S (1975) Blink reflex in Huntington's chorea and Parkinson's disease. Acta. Neurol. Scand., 52: 145-157. Faraday, MM, O'Donoghue, VA and Grunberg, NE (1999) Effects of nicotine and stress on startle amplitude and sensory gating depend on rat strain and sex. Pharmacol. Biochem. Behav., 62: 273-284. Fox, JE (1979) Habituation and prestimulus inhibition of the auditory startle reflex in decerebrate rats. Physiol. Behav., 23: 291-297. Gautier, CH and Cook, EW (1997) Relationship between startle and cardiovascular reactivity. Psychophysiology, 34: 87-96. Gogan, P (1970) The startle and orienting reactions in man. A study of their characteristics and habituation. Brain Res., 18: 117-135. Gokin, AP and Karpukhina, MV (1985) Reticular structures in the cat brain involved in startle to somatic stimuli of various modalities. Neurophysiology, 17: 278-286. Gomez-Wong, E and Valls-Sole, J (1996) Effects of a prepulse stimulus on the masseteric inhibitory reflex in humans. Neurosci. Lett., 208: 183-186. Gomez-Wong, E, Marti, MJ, Tolosa, E and Valls-Sole, J (1998) Sensory modulation of the blink reflex in

STARTLE AND PREPULSE EFFECTS

patients with blepharospasm. Arch. Neurol., 55: 12331237. Gormezano, I (1966) Classical conditioning. In: JB Sidowski (Ed.), Experimental Methods and Instrumentation in Psychology. McGraw Hill. New York, pp: 385-420. Goulart, F, Valls-Sole, J and Alvarez, R (2000) Posturerelated changes of soleus H-reflex excitability. Muscle Nerve, 23: 925-932. Graham, FK (1975) The more or less startling effects of weak prestimulation. Psychophysiology, 12: 238-248. Grillon, C, Ameli, R, Wood, SW, Merikangas, K and Davis, M (1991) Fear-potentiated startle in humans: effects of anticipatory anxiety on the acoustic blink reflex. Psychophysiology, 28: 588-595. Hallett, M, Shahani, BW and Young, RR (1975) EMG analysis of stereotyped voluntary movements in man. 1. Neurol. Neurosurg. Psychiatry., 38: 1154-1162. Ho, KJ, Kileny, P, Paccioretti, D and McLean, DR (1987) Neurologic, audiologic, and electrophysiologic sequelae of bitemporal lobe lesions. Arch. Neurol., 44: 982-987 Hoffman, HS and FleshIer, M (1963) Startle reaction: modification by background acoustic stimulation. Science, 141: 928-930. Hoffman, HS and Ison, JR (1980) Reflex modification in the domain of startle: I. Some empirical findings and their implications for how the nervous system processes sensory input. Psychological Review, 87: 175-189. Holand, S, Girard, A, Laude, D, Meyer-Bisch, C and Elghozi, JL (1999) Effects of an auditory startle stimulus on blood pressure and heart rate in humans. J. Hypertens., 17: 1893-1897. Hori, A, Yasuhara, A, Naito, Hand Yasuhara, M (1986) Blink reflex elicited by auditory stimulation in the rabbit. J. Neurol. Sci., 76: 49-59. Inglis, WL and Winn, P (1995) The pedunculo-pontine tegmental nucleus: where the striatum meets the reticular formation. Prog. Neurobiol., 47: 1-29. Ison, JR and Hoffman, HS (1983) Reflex modification in the domain of startle. II. The anomalous history of a robust and ubiquitous phenomenon. Psychol. Bull., 94: 3-17. Ison, JR, Sanes, IN, Foss, JA and Pinckney, LA (1990) Facilitation and inhibition of the human startle blink reflexes by stimulus anticipation. Behavioral Neuroscience, 104: 418-429. Jenny, AB and Saper, CB (1987) Organization of the facial nucleus and the corticofacial projection in the monkey: a reconsideration of the upper motor neuron facial palsy. Neurology, 37: 930-939. Koch, M (1999) The neurobiology of startle. Progr. Neurobiol.,59: 107-128.

281 Koch, M, Kungel, M and Herbert, H (1993) Cholinergic neurons in the pedunculopontine tegmental nucleus are involved in the mediation of prepulse inhibition of the acoustic startle response in the rat. Exp. Brain Res., 97: 71-82. Kofler, M, MUller, J, Reggiani, L and Valls-Sole, J (2001) Influence of age on auditory startle responses. Neurosci. Lett., 307: 65-68. Kofler, M, MUller, J, Reggiani, L and Valls-Sole, J (2001) Influence on gender on auditory startle responses. Brain Res., 921: 206-210. Kumari, V, Checkley, SA and Gray, JA (1996) Effect of cigarette smoking on prepulse inhibition of the acoustic startle reflex in healthy male smokers. Psychopharmacology, 128: 54-60. Lang, PJ, Bradley, MM and Cuthbert, BN (1990) Emotion, attention and the startle reflex. Psychol. Rev., 97: 377-395. Lee, Y, Lopez, DE, Meloni, EJ and Davis, M (1996) A primary acoustic startle pathway: obligatory role of cochlear root neurons and the nucleus reticularis pontis caudalis. J. Neurosci., 16: 3775-3789. Lehmann, J, Pryce, CR and Feldon, J (1999) Sex differences in the acoustic startle response and prepulse inhibition in Wistar rats. Behav. Brain Res., 104: 113-117. Liegois-Chauvel, C, Morin, C, Musolino, A, Bancaud, J and Chauvel, P (1989) Evidence for contribution of the auditory cortex to audiospinal facilitation in man. Brain, 112: 375-391. Lingenhohl, K and Friauf, E (1994) Giant neurons in the rat reticular formation: a sensorimotor interface in the elementary acoustic startle circuit? J. Neurosci., 14: 1176-1194. Markand, ON, Garg, BP and Weaver, DD (1984) Familial startle disease (hyperekplexia): electrophysiological studies. Arch. Neurol., 41: 71-74. Matsumoto, J, Fuhr, P, Nigro, M and Hallett, M (1992) Physiological abnormalities in hereditary hyperekplexia. Ann. Neurol., 32: 41-50. Matsumura, M and Woody, CD (1986) Long term increases in excitability of facial motoneurons and other neurons in and near the facial nuclei after presentations of stimuli leading to acquisition of a Pavlovian conditioned facial movement. Neurosci. Res., 3: 568-589. McManis, MH, Bradley, MM, Berg, WK, Cuthbert, BN and Lang, PJ (2001) Emotional reactions in children: verbal, physiological, and behavioral responses to affective pictures. Psychophysiology, 8: 222-23 I. Nakashima, K, Shimoyama, R, Yokoyama, Y, Takahashi, K (1993) Auditory effects on the electrically elicited

282 blink reflex in patients with Parkinson's disease. Electroencephalogr. cu« Neurophysiol., 89: 108-112. Ninomiya, H, Onitsuka, T, Chen, CH and Kinukawa, N (1997) Possible overlapping potentials of the auditory P50 in humans: factor analysis of middle latency auditory evoked potentials. Electroencephalogr. CUn. Neurophysiol., 104: 23-30. J, Wassermann, EM, BrasilPascual-Leone, A, Valls-Soh~, Neto, J, Cohen, LG and Hallett, M (1992) Effects of focal transcranial magnetic stimulation on simple reaction time to acoustic, visual and somatosensory stimuli. Brain, US: 1045-1059. Porter, R and Lemon, R (1993) Corticospinal function and voluntary movement. Oxford Science Publications, Clarendon Press. Oxford, pp. 428. Reese, NB, Garcia-Rill, E and Skinner, RD (1995) The pedunculopontine nucleus - auditory input, arousal and pathophysiology. Progr. Neurobiol., 42: 105-133. Rimpel, J, Geyer, D and Hopf, HC (1982) Changes in the blink responses to combined trigeminal, acoustic, and visual repetitive stimulation, studied in the human subject. Electroencephalogr. CUn. Neurophysiol., 54: 552-560. Rossi, A and Scarpini, C (1992) Gating of trigernino-facial reflex from low-threshold trigeminal and extratrigeminal cutaneous fibers in humans. J. Neurol. Neurosurg. Psychiatry, 55: 774-780. Rossignol, S and Melvill Jones, G (1976) Audiospinal influence in man studied by the H-reflex and its possible role on rhythmic movements synchronized to sound. Electroencephalogr. CUn. Neurophysiol., 41: 83-92. Saenz-Lope, E, Herranz-Tanarro, FJ, Masdeu, JC and Chacon Pefia, JR (1984) Hyperekplexia: a syndrome of pathological startle responses. Ann. Neurol., IS: 3~1.

Schicatano, EJ, Peshori, KR, Gopalaswamy, R, Sahay, E and Evinger, C (2000) Reflex excitability regulates prepulse inhibition. J. Neurosci., 20: 4240-4247 Simons, RF and Giardina, BD (1992) Reflex modification in psychosis-prone young adults. Psychophysiology, 29: 8-16. Starr, A, Cararnia, M, Zarola, F and Rossini, PM (1988) Enhancement of motor cortex excitability in humans by non-invasive electrical stimulation appears prior to voluntary movement. Electroencephalogr. CUn. Neurophysiol., 70: 26-32. Stitt, CL, Hoffman, HS, Marsh, RR and Schwartz, GM (1976) Modification of the pigeon's visual startle reaction by the sensory environment. J. Comp. Physiol. Psychol., 90: 601-619. Solomon, PR, Pomerleau, D, Bennett, L, James, J and Morse, DL (1989) Acquisition of the classically

1. VALLS-SOLl~

conditioned eyeblink response in humans over the life span. Psychol. Aging, 4: 34-41. Swerdlow, NR and Geyer, MA (1993) Prepulse inhibition of acoustic startle in rats after lesions of the pedunculopontine tegmental nucleus. Behavioral Neuroscience, 107: 104-117. Swerdlow, NR, Caine, SB, Braff, DL and Geyer, MA (1992) The neural substrates of sensorimotor gating of the startle reflex: a review of recent findings and their implications. J. Psychopharmacol., 6: 176-190. Swerdlow, NR, Paulsen, J, Braff, DL, Butters, N, Geyer, MA and Swenson, MR (1995) Impaired prepulse inhibition of acoustic and tactile startle response in patients with Huntington's disease. J. Neurol. Neurosurg. Psychiatry, 58: 192-200. Turpin, G (1986) Effects of stimulus intensity on autonomic responding: the problem of differentiating orienting and defense reflexes. Psychophysiology, 23: 1-14. Valls-Soh~, J (2000) Blink reflex excitability recovery curve and auditory prepulse inhibition in patients with Huntington's chorea and Parkinson's disease. Muscle Nerve, 23: 1629 (abstract). Valls-Sole, J, Cammarota, A, Alvarez, R and Hallett, M (1994) Orbicularis oculi responses to stimulation of nerve afferents from upper and lower limbs in normal humans. Brain Res., 650, 313-316. Valls-Sole, J, Sole, A, Valldeoriola, F, Munoz, E, Gonzalez, LE and Tolosa, ES (1995) Reaction time and acoustic startle in normal human subjects. Neurosci. Lett., 195: 97-100. Valls-Sole, J, Hallett, M and Alvarez, R (1996) Postpulse effects on blink reflex responses. Electroencephalogr. CUn. Neurophysiol., 101: 504-510. Valls-Sole, J, Valldeoriola, F, Tolosa, E and Nobbe, F (1997) Habituation of the auditory startle reaction is reduced during preparation for execution of a motor task in normal human subjects. Brain Res., 751: 155-159. Valls-Sole, J, Valldeoriola, F, Molinuevo, JL, Cossu, G and Nobbe, F (1999a) Prepulse modulation of the startle reaction and the blink reflex in normal human subjects. Exp. Brain Res., 129: 49-56. Valls-Sole, J, Rothwell, JC, Goulart, F, Cossu, G and Munoz, JE (1999b) Patterned ballistic movements triggered by a startle in healthy humans. J. Physiol., 516: 931-938. Valls-Sole, J, Veciana, M, Serra, J, Cruccu, G and Romaniello, A (2000) Prepulse inhibition of the blink reflex by laser stimuli in normal humans. Neurosci. Lett., 286: 79-82. Valls-Sole, J, Veciana, M, Leon, L and Valldeoriola, F (2002) Effects of a startle on heart rate in patients with multiple system atrophy. Mov. Disord., 17: 546-549.

STARTLE AND PREPULSE EFFECTS

Veciana, M, Valls-Solt\ J, Valldeoriola, F, Munoz, JE and Tolosa, E (2000) Startle effects on the R-R interval in normal controls and patients with multisystem atrophy. Mov. Disord., IS: 78 (abstract). Vidailhet, M, Rothwell, JC, Thompson, PD, Lees, AJ and Marsden, CD (1992) The auditory startle response in the Steele-Richardson-Olszewsky syndrome and Parkinson's disease. Brain, ll5: ll81-ll92. Wilkins, DE, Hallett, M and Wess, MM (1986) Audiogenic startle reflex of man and its relationship to startle syndromes. Brain, 109: 561-573. Wolfe, DL, Hayes, KC, Potter, PJ and Delaney, GA (1996) Conditioning lower limb H-reflexes by transcranial

283 magnetic stimulation of motor cortex reveals preserved innervation in SCI patients. J. Neurotrauma, 13: 281291. Woodruff-Pak, DS and Thompson, RF (1988) Classical conditioning of the eyeblink in the delay pardigm in adults aged 18-83 years. Psychol. Aging, 3: 219-229. Wu, MF, Suzuki, SS and Siegel, 1M (1988) Anatomical distribution and response patterns of reticular neurons active in relation to acoustic startle. Brain Res., 457: 399-406. Yeomans, JS and Frankland, PW (1996) The acoustic startle reflex: neurons and connections. Brain Res. Rev., 21: 301-314.

Movement Disorders Handbook of Clinical Neurophysiology, Vol. 1 M. Hallett (Ed.) © 2003 Elsevier B.Y. All rights reserved

285 CHAPTER 19

Long-latency reflexes following stretch and nerve stimulation G. Deuschl* Neurologische Klinik, Christian-Albrechts-Universitiit Kiel, D-24/05 Kiel, Germany

19.1. Introduction Long-latency reflexes are a group of reflexes of which the stretch reflexes of hand and arm muscles and the electrically elicited hand muscle reflexes have been most extensively studied. Stretch reflexes need more sophisticated equipment and have been used mainly to assess physiological questions. Reflexes in hand muscles following nerve stimulation can be elicited and recorded with routine EMG equipment and are therefore more suited for diagnostic questions. They can be a useful tool to assess sensorimotor functions of the central nervous system. They are known to exhibit distinct patterns of abnormality in various diseases, including movement disorders or focal lesions within the spinal cord, brainstem and brain. If a reflex is to function normally not only must its afferent and efferent pathways be intact but all central processing must be normal. Reflex tests and tests using evoked potentials or transcranial stimulation therefore provide distinctly different information. Lesions along the afferent, central, or efferent pathways suppress the reflex. Central lesions involving suprasegmental areas that facilitate or inhibit the reflex circuit may suppress or enhance the response but may leave the conduction times unaffected. Stretch reflexes and electrically elicited hand-muscle reflexes are such examples (Cruccu and Deuschl, 2000). Reflexes in hand and arm muscles can be elicited with various kinds of stimuli: muscle stretch (Marsden et al., 1976; Noth et al., 1985), cutaneous stimulation with air-puffs (Deuschl et al., 1995), or

* Correspondence to: Prof. Dr. G. Deuschl, Department of Neurology, Christian-Albrechts-Universitat Kiel, Niemannsweg 147,0-24105 Kiel, Germany. E-mail address:[email protected] Tel.: 0431-597-2610; fax: 0431-597-2712.

electrical stimulation of mixed or pure cutaneous nerves (Caccia et al., 1973; Deuschl et al., 1985; Meinck et al., 1987; Deuschl and Lucking 1990). Arm and hand muscles exhibit a variety of reflex responses that are probably transmitted through distinct pathways. On account of these variant reflex pathways, attempts to generalize the findings obtained in one muscle to other muscles are both inappropriate and confusing (Marsden et al., 1981). The present chapter will focus on hand muscle reflexes that have been studied most extensively, the physiology of these reflexes is sufficiently understood and the clinical applications have proven reliable in several neurological diseases. The stretch reflexes will be mentioned whenever special results have been obtained.

19.2. The pattern and physiology of stretch reflexes and electrically elicited reflexes The reflex pattern depends on the mode of stimulation and the nomenclature for the different reflex components may vary from author to author and muscle to muscle (see Table 1). After muscle stretch the pattern consists of the so-called Ml, M2 and M3 responses (Marsden et al., 1981; Noth et al., 1985). The Ml-response is equivalent to the monosynaptic reflex response elicited by la-afferents. The long-latency response M2 of hand muscles is most likely secondary to the activation of low-threshold muscle and cutaneous afferents (Deuschl et al., 1985). Like the circuit for responses to cutaneous stimulation, the central circuit for M2 includes a transcortical loop. Recently a pure cutaneous loop has been proposed for M2 of finger muscles (Corden et al., 2000), but a long-latency reflex (LLR) mediated by la-afferents has been convincingly demonstrated earlier (Deuschl et al., 1985; Noth et al., 1985). In proximal arm muscles (m. biceps and m. brachioradialis) the short-latency Ml-response is also transmitted through the la-monosynaptic reflex

G.DEUSCHL

286 Table I

Nomenclature of long-latency, long loop and stretch reflexes in the literature and possible correspondence between the different reflexes. Long-latency excitatory (medium)

Long-latency excitatory (late)

Reference

B

M3

(Marsden et al., 1976)

M2

M3

(Noth et al., 1985)

C-reflex

C-reflex

C-reflex

(Conrad and Aschoff, 1977)

LLRI

LLRll

LLR III

(Deuschl and Liicking, 1990)

Radial superficial nerve stimulation

cLLRI

cLLRll

cLLR III

(Caccia et al., 1973; Deuschl et al., 1985)

Finger stimulation

E1

E2

E3

(Caccia et al., 1973; Jenner and Stephens, 1982)

Spinal monosynaptic

Long-latency excitatory (early)

Stretch reflexes of the thumb

MI

A

Stretch reflexes of the thumb

MI

Median nerve stimulation

HR

Median nerve stimulation

HR

Long-latency inhibitory

II

Stretch reflexes of the hand

M1

M2

M3

(Marsden et aI., 1976)

Stretch reflexes of the biceps/ triceps

M1

M2

M3

(Marsden et al., 1976)

Stretch reflexes of the triceps surae

M1

M2

M3

(Berardelli et al., 1982)

pathway. But the M2-response in arm muscles is probably a spinal response mediated by group II afferents (Thilmann et al., 1991; Fellows et al., 1993; Fellows et al., 1996). For the triceps surae and the tibial anterior muscle, the situation is less clear. But a group II-afferent spinal pathway seems to be more likely (Berardelli et al., 1982, 1983a, b) than a generation by repetitive discharges of Ia-afferents (Hagbarth et al., 1981; Bongiovanni and Hagbarth, 1990). This is also true for more complex ways of eliciting reflexes like sudden slips of the surface during standing (Dietz, 1993). The reflex pattern elicited by adequate cutaneous stimulation (air-puffs) to the tip of the finger has

been assessed only in hand muscles and is characterized by an early excitation (cLLR I), an early inhibition (11), and a late excitation (cLLR II) (Deuschl et al., 1995), comparable to the reflex pattern elicited by electric stimulation of finger nerves (Jenner and Stephens, 1982). The responses to electric stimulation have been characterized in several studies (Deuschl et al., 1985; Claus and Jakob, 1986; Claus et al., 1986; Eisen, 1987; Eisen et al., 1989; Eisen and Deuschl, 1994; Friedli and Deuschl, 1994). After mixed-nerve stimulation the first response is the M-wave due to direct excitation of the motor axons. It has a latency of about 3-10 ms depending on the site of stimula-

287

LONG-LATENCY REFLEXES FOLLOWING STRETCH AND NERVE STIMULATION

tion. The reflex pattern consists of the Hoffmann reflex (HR), followed by up to three long latency reflexes (LLR), termed LLR I, II, and III. We prefer to use the term long-latency instead of long-loop reflexes because the central pathways for LLR I and LLR III remain unclear. These reflexes have been assessed in thenar muscles after median nerve stimulation at the wrist (Deuschl and Lucking, 1990) and in the first dorsal interosseus after ulnar nerve stimulation at the wrist (Noth et al., 1985; Claus and Jakob, 1986; Claus et aI., 1986). The reflex pattern after stimulation of purely cutaneous nerves, such as the finger nerves (Caccia et aI., 1973; Jenner and Stephens, 1982) or the radial superficial nerve (Caccia et al., 1973; Deuschl et aI., 1988, 1989), consists of an early excitatory component (El, cLLR I), an inhibitory component (11), and a second excitatory component (E2, cLLR II) (Jenner and Stephens, 1982). Similarly cutaneous reflexes of leg muscles have been described (Gibbs et aI., 1999). The reflex pathway for the HR is well-known: afferent impulses are conducted along la-fibers and relayed monosynaptically to the homonymous motoneurons. The H-reflex is usually easily evoked in normal subjects; in elderly subjects it can occasionally be difficult to record without a sizable M-wave. The only other reflex response present in all normal subjects is the LLR II. The afferents for the LLR II are fast-conducting cutaneous fibers and la-fibers (Deuschl et al., 1985); afferent impulses are transmitted along the dorsal columns to the nucleus cuneatus and along the lemniscal pathway to the sensory cortex. It is assumed that the excitatory volley is then relayed to the motor cortex, which in tum activates the spinal motoneurons via the corticospinal tract (Fig. 1 top). Evidence in favor of this pathway comes from animal experiments (Bawa et aI., 1979; Bawa and McKenzie, 1981; Evarts and Fromm, 1981; Cheney, 1985) as well as from several studies in normal subjects (Bawa and McKenzie, 1981; Bawa et aI., 1983; Noth et aI., 1983, 1985; Deuschl et aI., 1985) and patients (Noth 1986; Deuschl et aI., 1988, 1989; Podivinsky et al., 1992; Michels et aI., 1993; Naumann and Reiners 1997). The pathways for the LLR I and the LLR III are much less clear. The LLR I is also considered a transcortical reflex, mainly because under pathologic conditions (patients with cortical myoclonus and giant SEPs) reflexes of similar latency (C-reflexes) can occur at the same latency as the LLR I

Primarysonsorv cortex

Thalamus

Net cuneatus

Spinal cord

s~

MU'iCle

H-ref1ex A

B

c

~

~-;-.-----'---

~

~II o

100

~

:

_I

~~

~

200 (msl

Fig. I. Reflex pathways of the hand muscle reflexes (top) and normal patterns of reflexes elicited by median nerve (left) and radial superficial nerve (right) stimulation in three normal subjects (A, B, C) (bottom). From Cruccu and Deuschl (2000).

(Shibasaki, 1995). The C-reflex could therefore be an enhanced LLR I. Because electrical stimulation of the median nerve in some patients elicits reflex myoclonic jerks at the same latency as the LLR III (Deuschl and Lucking, 1990; Deuschl, 1992), LLR III could in theory be a transcortical reflex, with a similar argument as for the LLR I. Current evidence, however, suggests that the LLR III has a more complex pathway, possibly including a transcerebellar loop (Claus et aI., 1986). Even for the hand muscles it is difficult to match the various responses evoked by stretch, cutaneous or electric nerve stimulations. Ml corresponds to the H-reflex and M2 to the LLR II/cLLR lIfE 2 responses (see below). But the other responses (Ell cLLR I/LLR I and LLR IIIIcLLR IIIIM3) of the hand muscles are more difficult to compare. Whether

288 they are equivalent across the different modalities of stimulation also remains unclear. For proximal arm and leg muscles the situation is even more complex and further studies are necessary to compare them with the hand muscle reflexes.

19.2.1. Methods Hand muscle reflexes are studied with routine EMG-equipment, using surface electrodes for stimulation and recording. Either the median nerve or the radial superficial nerve are stimulated at the wrist. The stimulus intensity for the electrical stimulus is kept at motor threshold for mixed nerves and at 2.5 x sensory threshold for sensory nerves (squarewave pulse, 200 p.s, 3 Hz repetition rate). EMG signals are recorded from the abductor pollicis brevis, filtered (5 Hz-3000 Hz), full-wave rectified and averaged (256-512 sweeps). Some commercial EMG-machines have specially implemented filters (integrators) for the rectified EMG. Because these devices sometimes alter the shape of the reflex pattern they should be turned off. It has been shown that rectified and non-rectified responses have different latencies (Kurusu and Kitamura, 1999; Sartucci et al., 1999). The latencies are measured at the onset. Reflexes preceded by a pronounced inhibitory component sometimes have a questionable onset. If the precise onset remains in doubt, the latency can be taken at the point where EMG responses and background activity cross. The amplitude depends on the number and size of subliminal fringe motoneurons and hence on the amount of background activation. But because the rectified background EMG and the reflex size are roughly linearly related (Berkefeld et al., 1986; Meinck et al., 1987; Deuschl and LUcking, 1990) the reflex size can be defined as a fraction of baseline EMG. The duration is simply defined as the difference between the onset-latency and endlatency of the reflex. Mixed reflexes can also be elicited by stimulating the median or ulnar nerve at the wrist and recording from the first dorsal interosseus muscle (Claus and Jakob, 1986; Claus et al., 1986). Cutaneous reflexes can also be elicited by stimulation of finger nerves, e.g. the second finger and recording from the first dorsal interosseus muscle (Jenner and Stephens, 1982). However, far more sweeps are necessary to get clear responses compared with radial superficial nerve stimulation.

G.DEUSCHL

We refer to the original publications for stretch reflexes of the hand muscles (Marsden et aI., 1976; Noth et al., 1985), wrist muscles (Marsden et al., 1976; Rothwell et al., 1980), cutaneous reflexes of handmuscles elicited with airpuffs (Deuschl et al., 1995) or stretch reflexes of the legs elicited with stretch (Berardelli et aI., 1982) or gait disturbances (Dietz, 1993).

19.2.2. Normal values and abnormal patterns of thenar reflexes Median nerve stimulation elicits the H-reflex (HR) at 25-34 ms in the thenar muscle followed by the LLR I in about 30%, LLR II in 100%, and LLR III in about 20% of normal subjects. The latency of these reflexes depends on the height of the subject. Adult ranges are 35-46 ms for LLR I, 45-58 ms for LLR II, and >68 ms for LLR III (Fig. 1). For normal values see Table 2. Radial superficial nerve stimulation elicits an early cLLR I (35-43 ms) in 35% and a cLLR II (43-59 ms) in 100% of normal subjects (Fig. 1). The cLLR I is never preceded by HR-like responses and is seldom followed by a discernible inhibitory period. The cLLR I corresponds to the E1 after finger nerve stimulation. Stimulation of the superficial radial nerve stimulation only occasionally elicits a cLLR III 00-82 ms). The latency of the hand-muscle reflexes depends critically on the time spent along the peripheral pathways and therefore on the individual subjects' arm-length and conduction velocity. Variance induced by these factors can be reduced in two ways. The first is to use the difference between the LLR and the median nerve HR latency as a measure to assess the central conduction time. The second is to express the LLR-Iatencies as a function of the HR-Iatency according to the following formulas (Deuschl et al., 1985): LLR II-latency =19.3 ms+1.07xHR-latency, LLR I-latency =12.6 ms+0.981 x HR-Iatency. Both approaches permit assessment of the reflexlatency even in patients with peripheral neuropathies. In some of these patients the peripheral conduction velocity is prolonged but the afferent volley remains sufficiently synchronized to elicit the full pattern of LLR.

LONG-LATENCY REFLEXES FOLLOWING STRETCH AND NERVE STIMULATION

289

Table 2 Normal values of the hand muscle reflexes after median nerve stimulation (HR and LLR) or radial superficial nerve stimulation (cLLR) in 102 normal subjects (age: 18-85 years). Mean

SD

SE

Count

Minimum

Maximum

Latency (ms) HR

28.9

2.4

0.2

102

24.1

35.4

LLRI

40.6

2.5

0.5

26

36.8

47.2

LLRII

50.3

3.2

0.3

102

43.1

59.3

LLRill

76.0

4.6

1.0

20

70.3

92.1

cLLRI

37.6

2.6

1.0

7

35.0

43.0

cLLR II

50.2

3.0

0.3

100

43.0

60.0

cLLRill

75.9

3.6

0.7

31

70.0

82.0

Amplitude (mV) HR

1.9

1.1

0.1

102

0.3

4.8

LLRI

0.4

0.1

0.0

25

0.2

0.8

LLRII

1.2

0.6

0.1

102

0.3

3.0

LLRill

0.8

0.5

0.1

20

0.3

2.3

cLLRI

0.4

0.2

0.1

7

0.1

0.6

cLLR II

1.2

0.5

0.0

100

0.3

2.5

cLLRill

0.8

0.5

0.1

31

0.3

2.4

Duration (ms) HR

10.4

2.5

0.2

102

6.0

17.0

LLR I

8.9

7.8

1.5

26

3.0

46.0

LLR II

22.3

5.4

0.5

101

12.0

40.0

LLRill

27.6

8.7

2.0

20

9.0

44.0

cLLRI

10.0

3.7

1.4

7

7.0

17.0

cLLR II

24.0

6.6

0.7

100

9.0

50.0

cLLRill

27.7

6.2

1.1

31

15.0

40.0

Hand-muscle reflexes vary little with age. In adults, the amplitude of the HR, LLR II, and cLLR II show a negative correlation with age. This decrease is small and clinically irrelevant. The maturation of the pattern of cutaneous reflexes has been studied in detail for the reflexes elicited by index finger stimulation (Issler and Stephens 1983; Evans et al., 1990). Up to the age of 8 years, the El can be larger than the E2. Later in life, the E2 predominates. Thenar reflexes in response to median nerve or radial superficial nerve stimulation can show five

distinct types of abnormality: an absent HR, an enhanced HR, an absent LLR II, a delayed LLR II, and an enhanced LLR I. (Table 3, Fig. 2). 19.3. Clinical applications 19.3.1. Intra-axial focal and multifocal lesions

The pathways of the hand-muscle long-latency reflexes cover the larger area of the cervical cord, brainstem, and brain along a narrow pathway. Focal lesions within this pathway lead to abnormal reflex

G.DEUSCHL

290 Table 3 Most typical abnormalities of hand-muscle reflexes and their diagnostic indications. Reflex abnormality

Involved pathway

Typical lesions/diseases

Absent HR

Peripheral reflex pathways

Peripheral neuropathy. radiculopathy

Enhanced HR

Suprasegmental descending projections

Various lesions along the corticospinal or bulbospinal tracts

Absent LLR II

Central reflex arc

Various lesions of the lemniscal pathways. cortex or cortico-spinal tracts; Huntington's disease

Delayed LLR II

Central reflex arc

Mostly demyelinating lesions in multiple sclerosis

Enhanced LLR I

Unknown

Myoclonus of various origin (cortical. subcortical); corticobasal degeneration; Parkinson's disease; essential tremor

patterns (Jenner and Stephens, 1982; Deuschl and Lucking, 1990; Rothwell, 1990). Especially reduced or absent LLR II are found in lesions affecting the afferent pathway to the motor cortex (dorsal columns, lemniscus medialis, thalamus, thalamocortical pathway). An enhanced HR strongly implies damage to the corticospinal tract. radial superficial nerve

·L"", ,~ D

o

100

Almost all the abnormal reflex patterns can be found in multiple sclerosis. The precise abnormality depends on the type of central lesion. The most common findings are an enlarged HR and absent LLR II. The most specific finding, found in up to 50% of patients with multiple sclerosis, is a delayed LLR II (Deuschi et al., 1988; Michels et al., 1993; lovichich, 1994). The LLR seems to yield abnormal findings more often than SEP-testing. It also sometimes provides useful diagnostic information not obtainable by SEPs and transcranial magnetic stimulation (Michels et al., 1993; lovichich, 1994). A similar delay of the LLR II has been found in two patients with adrenomyeloneuropathy (Liao et al., 2001).

19.3.2. Spasticity

~

200 0

'00

200 ems]

Fig. 2. Abnormal reflex patterns in various diseases. (A) shows a normal reflex pattern for comparison with HR and LLR II after median nerve stimulation and a cLLR II after radial superficial nerve stimulation. (B) shows the typical pattern for spasticity: an enlarged HR and an absent LLR II and cLLR II. (C) shows a delayed LLR II and cLLR II with an enlarged HR in Multiple Sclerosis. (D) shows a pattern with an enlarged LLR I and normal HR and normal cutaneous reflexes inessential tremor or Parkinson's disease. From Deuschl and Lucking (1990).

The typical pattern of hand reflexes in disorders associated with spasticity is the enhanced HR with relative amplitudes (multiples of baseline amplitude) of more than 4.5 and an absent or reduced LLR II. It is sometimes hard to decide if an LLR II is still present or not because the statistically defined lower limit for its amplitude is near zero (Table 1). We assume an LLR II to be present (and normal), if its latency is within the normal range (see Table 3) and if it has a discernible amplitude. When the HR is very large, the amplitude of the LLR II following stimulation of the median nerve can be indiscernible probably because the motoneurons are refractory. If the median-nerve LLR II is hard to discern then it is

LONG-LATENCY REFLEXES FOLLOWING STRETCH AND NERVE STIMULATION

especially helpful to test the cutaneous LLR after superficial radial nerve stimulation. 19.3.3. Huntington's disease In Huntington's disease the LLR II-amplitude is reduced or the LLR II (and cLLR II) is absent (Noth et aI., 1985; Deuschl et aI., 1989; Eisen et aI., 1989; Siedenberg et aI., 1999). In contrast to the findings in spasticity, the HR-amplitude is only slightly increased. Choreatic hyperkinesias can occur in a variety of conditions other than Huntington's disease, commonly labeled as symptomatic chorea (Durr et aI., 1995; Sinard and Hedreen, 1995). The available evidence indicates that LLR II is absent only in Huntington's disease, not in symptomatic chorea (Deuschl et aI., 1989; Eisen et aI., 1989). 19.3.4. Parkinson's disease and parkinsonian syndromes The M21M3 reflexes have been found to be increased in PD (Lee et aI., 1983; Lee, 1989; Johnson et aI., 1991). A recent study has shown that this component is decreasing after Pallidotomy (Hayashi et aI., 2001). The same reduction has been found earlier after thalamotomy (Struppler, 1983; Struppler et aI., 1984). Cutaneous reflexes have been studied and the cutaneous inhibitory period (11) after digital nerve stimulation in Parkinson's disease was found to be decreased which is probably related to Parkinsonian rigidity (Fuhr et al., 1992). Median nerve stimulation often elicits an enhanced LLR I. How this finding is related to the symptoms of Parkinson's disease remains unclear. No correlation has been found with rigidity, some correlation with the occurrence of an action tremor. Interestingly, the cutaneous reflexes do not always show this enhanced LLR I component. The enhanced LLR I is usually much larger in Parkinsonian syndromes with reflex myoclonus, e.g. corticobasal degeneration, than in idiopathic Parkinson's disease and is always accompanied by an increased cutaneous LLR I (radial superficial or finger nerve stimulation). In these syndromes the combination of an akinetic-rigid syndrome with apraxia or alien limb syndrome and reflex myoclonus is itself suggestive for corticobasal degeneration (Chen et aI., 1992; Feifel et aI., 1994; Thompson et aI., 1994; Thompson, 1995). Although reflex myo-

291

clonus and especially an enhanced LLR are almost always found in corticobasal degeneration, in the very early stages of disease both may still be normal (Lu et aI., 1998). An open question is whether the abnormal LLR I is found only in corticobasal degeneration or also in other non-idiopathic Parkinsonian syndromes. A recent study found also an increased LLR I in patients with MSA and myoclonus (Salazar et aI., 2000).

19.3.5. Dystonia Earlier investigations of LLR in dystonia showed an increased amplitude of the LLR following stretch of the wrist muscles (Rothwell, 1990). More recent investigations (Naumann and Reiners, 1997) of the electrically-elicited hand muscle reflexes have shown an increased incidence of abnormally enlarged LLR I and reduced amplitudes of the LLR II. Botulinum toxin injections lead to even smaller LLR II.

19.3.6. Cerebellar disease and Friedreich's disease Patients with Friedreich's disease whose HR and LLR-responses are preserved may have an LLR-II latency delay, compatible with delayed conduction within the dorsal columns (Alfonsi et aI., 1992). The enhanced LLR III-response seen in various cerebellar degenerations (Claus et aI., 1986; Alfonsi et aI., 1992) suggests a major role of the cerebellum in the generation of the LLR III.

19.3.7. Essential tremor About 40% of patients with essential tremor have an enhanced LLR I. Only the LLR I to median nerve stimulation is usually enhanced whereas the LLR I to radial superficial nerve stimulation is not (Deuschl et aI., 1987a). A statistical association has been found between reciprocal alternating activity in antagonistic muscles of the tremor bursts and enhanced LLR I amplitude but the pathophysiological significance of this finding remains unknown in essential tremor.

19.3.8. Myoclonus One of the most important clinical applications of hand-muscle reflex testing is the diagnosis and classification of myoclonus.

292

The various forms of myoclonus can be separated by means of clinical and electrophysiological criteria (Shibasaki, 2000). The electrophysiological criteria include LLR-testing, the measurement of the cortical SEP following median nerve stimulation and backaveraging of the ongoing EEG triggered by spontaneous myoclonic jerks. All forms of reflex myoclonus have abnormal LLR, no matter whether their origin is cortical or subcortical (Deuschl and LUcking, 1990). The most common abnormality is an enhanced LLR I. It can occur with or without a giant-SEP or a spike preceding the spontaneous myoclonic jerks by about 15-25 ms. This is mostly interpreted as indicating enhanced excitability of the primary sensory cortex when the SEP is enlarged or of the primary motor cortex when the backaveraging is positive. In some patients the LLR III is enhanced while the LLR I is not (Deuschl et al., 1987b). The LLR enhancement is often associated with an increased amplitude of the later SEP components (N2-P2-N3). Rarely, also the LLR II can be enlarged (Shibasaki, 1995). Some forms of myoclonus are notably difficult to separate from tremors. These cases have been labeled as cortical myoclonic tremor (Ikeda et al., 1990; Toro et al., 1993; Terada et al., 1997). Most of them have enhanced LLRs.

Acknowledgments GD was supported by the Deutsche Bundesministerium fur Bildung und Forschung (Kompetenznetz Parkinson).

References Alfonsi, E, Bono, G, Mertel, J, Sandrini, G, Bollani, E and Moglia, A (1992) Electrically elicited short and longlatency responses of intrinsic hand muscles in hereditary ataxias. Effects of isometric and ballistic isotonic voluntary contractions. Electromyogr. Clin. Neurophysiol., 32: 435---442. Bawa, P and McKenzie, DC (1981) Contribution of joint and cutaneous afferents to longer-latency reflexes in man. Brain Res., 2ll: 185-189. Bawa, P, Stein, RB and Tatton, WG (1979) Dynamics of a long-latency reflex pathway in the monkey. Biol. Cybernet., 34: 107-110. Bawa, P, Mekjavic, IE and Calancie, B (1983) Effect of core cooling on short and long latency reflex responses. Brain Res., 264: 320-322.

G.DEUSCHL

Berardelli, A, Hallett, M, Kaufman, C, Fine, E, Berenberg, Wand Simon, SR (1982) Stretch reflexes of triceps surae in normal man. J. Neurol. Neurosurg. Psychiatry, 45: 513-525. Berardelli, A, Sabra, AF and Hallett, M (1983a) Physiological mechanisms of rigidity in Parkinson's disease. J. Neurol. Neurosurg. Psychiatry, 46: 45-53. Berardelli, A, Sabra, AF, Hallett, M, Berenberg, Wand Simon, SR (1983b) Stretch reflexes of triceps surae in patients with upper motor neuron syndromes. J. Neurol. Neurosurg. Psychiatry, 46: 54-60. Berkefeld, J, Morich, J and Meinck, HM (1986) (Cutaneomuscular reflexes of the human hand. I. Methodologic principles). Eeg Emg Z Elektroenzephalogr Verwandte Geb, 17: 189-196. Bongiovanni, LG and Hagbarth, KE (1990) Tonic vibration reflexes elicited during fatigue from maximal voluntary contractions in man. J. Physiology, 423: 1-14. Caccia, MR, McComas, AJ, Upton, AR and B10gg, T (1973) Cutaneous reflexes in small muscles of the hand. J. Neurol. Neurosurg. Psychiatry, 36: 960-977. Chen, R, Ashby, P and Lang, AE (1992) Stimulussensitive myoclonus in akinetic-rigid syndromes. Brain. Cheney, PD (1985) Role of cerebral cortex in voluntary movements. A review. Phys. Ther., 65: 624-635. Claus, D and Jakob, S (1986) The relationship between long latency responses and height. J. Neurol., 233: 271-273. Claus, D, Schocklmann, HO and Dietrich, HJ (1986) Long latency muscle responses in cerebellar diseases. Europ. Arch. Psychiatry Neurol. Sci., 235: 355-360. Conrad, B and Aschoff, JC (1977) Effects of voluntary isometric and isotonic activity on late transcortical reflex components in normal subjects and hemiparetic patients. Electroencephalogr. Clin. Neurophysiol., 42: 107-ll6. Corden, DM, Lippold, OC, Buchanan, K and Norrington, C (2000) Long-latency component of the stretch reflex in human muscle is not mediated by intramuscular stretch receptors. J. Neurophysiol., 84: 184-188. Cruccu, G and Deuschl, G (2000) The clinical use of brainstem reflexes and hand-muscle reflexes. Clin. Neurophysiol., lll: 371-387. Deuschl, G (1992) Myoklonien. In: Neurologie in Praxis und Klinik, 2 (Vol. 4). Stuttgart: Thieme Verlag, pp. 61-4.73. Deuschl, G and LUcking, CH (1990) Physiology and clinical applications of hand muscle reflexes. Electroencephalogr. Clin. Neurophysiol., 41 (Suppl.): 84101. Deuschl, G, Schenck, E and Lucking, CH (1985) Longlatency responses in human thenar muscles mediated

LONG-LATENCY REFLEXES FOLLOWING STRETCH AND NERVE STIMULATION

by fast conducting muscle and cutaneous afferents. Neurosci. Lett., 55: 361-366. Deuschl, G, LUcking, CH and Schenck, E (1987a) Essential tremor: electrophysiological and pharmacological evidence for a subdivision. J. Neurol. Neurosurg. Psychiatry, 50: 1435-1441. Deuschl, G, Schenck, E, LUcking, C and Ebner, A (1987b) Cortical reflex myoclonus and its relation to normal long-latency reflexes. In: R Benecke, B Conrad and CD Marsden (Eds.), Motor Disturbances I. London: Academic press, pp. 305-319 Deuschl, G, Strahl, K, Schenck, E and LUcking, CH (1988) The diagnostic significance of long-latency reflexes in multiple sclerosis. Electroencephalogr. Clin. Neurophysiol., 70: 56-61. Deuschl, G, LUcking, CH and Schenck, E (1989) Hand muscle reflexes following electrical stimulation in choreatic movement disorders. 1. Neurol. Neurosurg. Psychiatry, 52: 755-762. Deuschl, G, Feifel, E, Guschlbauer, B and LUcking, CH (1995) Hand muscle reflexes following air puff stimulation. Exp. Brain Res., 105: 138-146. Dietz, V (1993) Gating of reflexes in ankle muscles during human stance and gait. Progr. Brain Res., 97: 181188. DUff, A, Dode, C, Hahn, V, Pecheux, C, Pillon, B, Feingold, J, Kaplan, JC, Agid, Y and Brice, A (1995) Diagnosis of "sporadic" Huntington's disease. J. Neurol. Sci., 129: 51-55. Eisen, A, Bohlega, S, Bloch, M and Hayden, M (1989) Silent periods, long-latency reflexes and cortical MEPs in Huntington's disease and at-risk relatives. Electroencephalogr. Clin. Neurophysiol., 74: 444-449. Evans, AL, Harrison, LM and Stephens, JA (1990) Maturation of the cutaneomuscular reflex recorded from the first dorsal interosseous muscle in man. J. Physiol., 428: 425-440. Evarts, EV and Fromm, C (1981) Transcortical reflexes and servo control of movement. Can. J. Physiol. Pharmacol., 59: 757-775. Feifel, E, Brenner, M, Teiwes, R, LUcking, CH and Deuschl, G (1994) Corticobasale degeneration. Die bedeutung klinischer kriterien zur diagnosestellung. Nervenarzt, 65: 653-659. Fellows, SJ, Ross, HF and Thilmann, AF (1993) The limitations of the tendon jerk as a marker of pathological stretch reflex activity in human spasticity. J. Neurol. Neurosurg. Psychiatry, 56: 531-537. Fellows, SJ, Topper, R, Schwarz, M, Thilmann, AF and Noth, J (1996) Stretch reflexes of the proximal arm in a patient with mirror movements: absence of bilateral long-latency components. Electroencephalogr. Clin. Neurophysiol., 101: 79-83.

293

Fuhr, P, Zeffiro, T and Hallett, M (1992) Cutaneous reflexes in Parkinson's disease. Muscle Nerve, 15: 733-739. Gibbs, J, Harrison, LM, Stephens, JA and Evans, AL (1999) Cutaneomuscular reflex responses recorded from the lower limb in children and adolescents with cerebral palsy. Dev. Med. Child Neurol., 41: 456-464. Hagbarth, KE, Hagglund, JB, Wallin, EU and Young, RR (1981) Grouped spindle and electromyographic responses to abrupt wrist extension movement in man. J. Physiol., 312: 81-96. Hayashi, R, Hashimoto, T, Tada, T and Ikeda, S (2001) Relation between changes in long-latency stretch reflexes and muscle stiffness in Parkinson's disease comparison before and after unilateral pallidotomy. Clin. Neurophysiol., 112: 1814-1821. Iovichich, A (1994) Long latency reflexes and somatosensory potentials in multiple sclerosis patients. Neurosci. Behav. Physiol., 24: 440-441. Issler, H and Stephens, JA (1983) The maturation of cutaneous reflexes studied in the upper limb in man. J. Physiol., 335: 643-654. Jenner, JR and Stephens, JA (1982) Cutaneous reflex responses and their central nervous pathways studied in man. J. Physiol., 333: 405-419. Johnson, MT, Kipnis, AN, Lee, MC, Loewenson, RB and Ebner, TJ (1991) Modulation of the stretch reflex during volitional sinusoidal tracking in Parkinson's disease. Brain, 114: 443-460. Kurusu, K and Kitamura, J (1999) Long-latency reflexes in contracted hand and foot muscles and their relations to somatosensory evoked potentials and transcranial magnetic stimulation of the motor cortex. Clin. Neurophysiol., 110: 2014-2019. Lee, RG (1989) Pathophysiology of rigidity and akinesia in Parkinson's disease. Eur. Neurol., 29: 13-18. Lee, RG, Murphy, JT and Tatton, WG (1983) Longlatency myotatic reflexes in man: mechanisms, functional significance, and changes in patients with Parkinson's disease or hemiplegia. Adv. Neurol., 39: 489-508. Liao, KK, Chen, JT, Lin, KP, Chen, CC, Kao, KP and Wu ZA (2001) Brain dysfunction explored by long latency reflex: a study of adrenomyeloneuropathy. Acta. Neurol. Scand., 104: 105-109. Marsden, CD, Merton, PA and Morton, HB (1976) Stretch reflex and servo action in a variety of human muscles. J. Physiol., 259: 531-560. Marsden, CD, Merton, PA, Morton, HB, Rothwell, JC and Traub, MM (1981) Reliability and efficacy of the longlatency stretch reflex in the human thumb. J. Physiol., 316: 47-60. Meinck, HM, Berkefeld, J and Conrad, B (1987) Cutaneo-muscular reflexes of the human hand. II.

294 Neurophysiologic aspects of reflex organization and coordination (German). Eeg Emg Zeitschrift fur Elektroenzephalographie Elektromyographie und Verwandte Gebiete, 18: 101-107. Michels, R, Wessel, K, Klohn, S and Kompf, D (1993) Long-latency reflexes, somatosensory evoked potentials and transcranial magnetic stimulation: relation of the three methods in multiple sclerosis. Electroencephalogr. Clin. Neurophysiol., 89: 235-241. Naumann, M and Reiners, K (1997) Long-latency reflexes of hand muscles in idiopathic focal dystonia and their modification by botulinum toxin. Brain, 409-416. Noth, J (1986) Long loop reflexes: concepts and consequences. Appl. Neurophysiol., 49: 262-268. Noth, J, Friedemann, HH, Podoll, K and Lange, HW (1983) Absence of long latency reflexes to imposed finger displacements in patients with Huntington's disease. Neurosci. Lett., 35: 97-100. Noth, J, Podoll, K and Friedemann, HH (1985) Long-loop reflexes in small hand muscles studied in normal subjects and in patients with Huntington's disease. Brain, 108: 65-80. Podivinsky, F, Jergelova, M and Koncek, V (1992) The role of short and long latency reflexes in the small muscles of the hand in the diagnosis of movement disorders - an electrophysiology study (Slovak). Bratislavske Lekarske Listy, 93: 291-294. Rothwell, JC (1990) Long latency reflexes of human arm muscles in health and disease.; Electroencephalogr. Clin. Neurophysiol., 41 (Suppl.): 251-263. Rothwell, JC, Traub, MM and Marsden, CD (1980) Influence of voluntary intend. on the human longlatency stretch reflex. Nature, 286: 496-498. Salazar, G, Valls-Solt~, J, Marti, MJ, Chang, H and Tolosa, ES (2000) Postural and action myoclonus in patients

G.DEUSCHL

with parkinsonian type multiple system atrophy. Mov. Disord., 15: 77-83. Sartucci, F, Bonfiglio, L, Logi, F, Pellegrinetti, A and Murri, L (1999) Changes in long-latency reflexes onset latencies across full-wave rectified and non-rectified recordings. Clin. Neurophysiol., 110: 1975-1977. Shibasaki, H (1995) Myoclonus. Curr. Opin. Neurol., 8: 331-334. Shibasaki, H (2000) Electrophysiological studies of myoclonus. Muscle Nerve, 23: 321-335. Siedenberg, R, Goodin, DS and Aminoff, MJ (1999) Changes of forearm EMG and cerebral evoked potentials following sudden muscle stretch in patients with Huntington's disease. Muscle Nerve, 22: 1557-1563. Sinard, JH and Hedreen, JC (1995) Neuronal loss from the subthalamic nuclei in a patient with progressive chorea. Mov. Disord., 10: 305-311. Struppler, A (1983) The control of isometric contractions in patients suffering from different types of hypotonia. Rev. Electroencephalogr. Neurophysiol. Clin., 13: 273-281. Struppler, A, Lehmann-Hom, F, Klein, W, LUcking, CH and Deuschl, G (1984) Effect of stereoencephalotomy on long-latency EMG responses and motor control of arm movements in Parkinson's syndrome. Adv. Neurol., 40: 437-445. Thilmann, AF, Schwarz, M, Topper, R, Fellows, SJ and Noth, J (1991) Different mechanisms underlie the longlatency stretch reflex response of active human muscle at different joints. J. Physiol., 444: 631-643. Thompson, PD (1995) Myoclonus in corticobasal degeneration. Clin. Neurosci., 3: 203-208. Thompson, PD, Day, BL, Rothwell, JC, Brown, P, Britton, TC and Marsden, CD (1994) The myoclonus in corticobasal degeneration. Evidence for two forms of cortical reflex myoclonus. Brain, 1197-1207.

Movement Disorders Handbook of Clinical Neurophysiology, Vol. 1

M. Hallett (Ed.)

© 2003 Elsevier B.Y. All rights reserved

295

CHAPTER 20

Posturography Bastiaan R. Bloem":", Jasper E. Visser" and John H.J. Allum" a

Department of Neurology, University Medical Center St. Radboud, Nijmegen, The Netherlands h Department ofORL, University Hospital. Basel. Switzerland

20.1. Introduction The term posturography literally means "description of posture". The various techniques that are bundled under this term actually have a much wider perspective. First of all, most techniques aim to describe not only posture - that is, the position of various body parts during quiet stance or while being seated - but also the active and passive regulation of balance. Secondly, many posturography techniques provide much more than mere descriptions of posture or balance under unperturbed conditions. Indeed, many posturography techniques actively manipulate posture or balance, thus hoping to learn more about how healthy persons respond to such interventions, and why the underlying regulatory processes fail in patients with postural instability. Finally, although this is not made explicit by the term posturography itself, an inherent component of posturography is an objective assessment using quantitative outcome measures. Thus, simply inspecting subjects with the naked clinical eye is not a posturography technique owing to its qualitative nature. 20.1.1. Why posturography?

Regulation of normal posture and balance is very complex. The central nervous system must be informed continuously about internally or externally generated postural perturbations. To achieve this, different sources of afferent information - visual, vestibular and proprioceptive - need to be gathered,

* Correspondence to: Dr. Bastiaan R. Bloem, MD, PhD, Department of Neurology, 326, University Medical Center St. Radboud, P.O. Box 9101, 6500 HB Nijmegen, The Netherlands. E-mail address:[email protected] Tel.: 31-24-3615202; fax: 31-24-3541122.

weighted and centrally integrated. Humans require knowledge about the position of the various body segments relative to each other, relative to gravity, and relative to the surroundings - body and earthfixed coordinates - in order to regulate their posture. They also need information about the specific environmental circumstances, such as the nature or stability of the supporting surface on which they stand. This information must be combined with the plans and intentions of the subject, and weighed against the expectations based upon prior experience or specific instructions. If balance is jeopardized, humans can choose from a wide repertoire of different balance reactions, ranging from purely passive postural responses to voluntarily initiated movements. The complex physiological mechanisms underlying these reactions can only be unraveled under carefully controlled experimental conditions, with calibrated control of the environment, standardized bodily perturbations and use of quantitative outcome measures (Table 1). Apart from its use in normal physiology, posturography is essential to disentangle the complex pathophysiology of balance disorders in patients. The "disease focused" eye of an experienced clinician may remain superior in detecting that something is wrong, but has relatively limited accuracy when deciding how balance regulation is affected. For most fallers, multiple factors can be Table 1 Advantages of posturography. • • • •

Standardized and reproducible postural perturbations Quantitative and objective outcome measures Simultaneous recording of multiple variables Ability to selectively manipulate elements of postural control

296 identified that might be causally involved, but it remains difficult to determine their relative importance during a clinical examination. To complicate matters further, clinicians observe the net result of primary pathological processes and secondary compensatory strategies; separating these two aspects often calls for careful experimental manipulation. Posturography can help with this separation, thereby providing a detailed insight into pathophysiological processes. This knowledge is required for optimal development of therapeutic and fall-prevention strategies. Another potential advantage of posturography is its use as a supplement or even surrogate to clinical tests of posture and balance. A drawback of most clinical balance tests is the inherent difficulty in standardizing performance, and the subjective scoring of the outcome. For these reasons, clinical tests of posture and balance have been of limited use for early detection of subjects prone to falls, and as objective outcome measure in intervention studies. Posturography offers interesting perspectives in these fields (Table 1). Important advantages include the delivery of highly standardized postural perturbations, and the ability to quantify balance performance in an objective manner. 20.1.2. Static vs. dynamic posturography

The literature contains descriptions of a wide range of different posturography techniques. Many are custom made and can be found only in selected laboratories, but a few are commercially available. It is common practice to allocate a particular technique to one of two major categories: static posturography and dynamic posturography (Furman et aI., 1993). Although these terms are commonly used, their definitions are surprisingly poor. At first sight, the term static posturography would suggest that only stationary body positions or spontaneous sway oscillations are studied in persons who are quietly standing on a fixed support surface with imbedded strain gauges to measure surface reaction forces. In contrast, the term dynamic posturography is usually equated with experimental settings where standing subjects are actively being perturbed by a moving support surface (Furman et aI., 1993). But how does one classify an experiment where subjects standing on a fixed support surface are being perturbed by external perturbations applied directly to the trunk?

B.R. BLOEM ET AL.

Or studies where a person's gait is perturbed by a dynamic posturography system hidden in the walkway? And what about studies where subjects are required to balance on an unstable support surface, thereby perturbing themselves? To our mind, static posturography refers to all those techniques that measure quiet standing, with or without an instrumented fixed support surface, and without any physical body perturbation. In contrast, dynamic posturography techniques employ physical perturbations of stance, using either an unstable or motorized support surface, or an external force applied to one or more body parts. A further distinction should be made between these stance techniques and those where a gait task forms the basis of the experiment. These definitions ignore whether or not balance and gait are being manipulated in any other way, for example by eye closure or specific prior instructions. After all, such manipulations can be used with either technique. These definitions also leave aside how posture or balance is measured, again because this is not unique to one particular technique. Note that the initial body position is not specified by these definitions. Although posturography has been used widely to study upright standing persons, interesting information can also be obtained from studies in jumping, seated and even recumbent subjects (Martin, 1965; Forssberg and Hirschfeld, 1994; Perennou et aI., 1997; Bisdorff et aI., 1999; McDonagh and Duncan, 2002). This discussion of definitions sets the stage for the remainder of this chapter. We will first describe the general principles of static posturography. This is followed by an overview of the different types of physical body perturbations, as these determine the difference between the various types of dynamic posturography. The next section deals with the recording equipment that can be used to quantify posture and balance. Use of these outcome measures will be illustrated for various static or dynamic posturography experiments. The subsequent section reviews several options to manipulate posture and balance. We conclude with paragraphs on the shortcomings and possible future developments of posturography. The emphasis of this chapter is on methodology, and outcomes of studies are only mentioned when they serve to illustrate the merits or shortcomings of a particular approach. Given the rapid pace at which new techniques appear, readers

POSTUROGRAPHY

should not expect a comprehensive summary of posturography variants, but rather an illustration of the rich spectrum of possibilities. 20.2. Static posturography

Static posturography aims to measure quiet standing while subjects maintain stance in a relatively unperturbed state (Overstall et al., 1977; Fernie et aI., 1982; Lord et al., 1991; Collins and De Luca, 1994; Riley et aI., 1997; Winter et al., 1998). Because no external perturbations are used, a relatively long trial duration is required to capture the entire range of spontaneously occurring balance responses. Depending on the type of experiment, trial duration may vary between several seconds up to 30 minutes. Subjects are often instructed to simply stand quietly on a fixed support surface, with eyes opened or closed. More recent approaches involved the use of different types of manipulations to make the balancing task more challenging; these manipulations are outlined in more detail in Section 20.5. Examples include changing the size or texture of the support surface and controlled variations in stance width - ranging from wide to tandem stance (Day et al., 1993; Baloh et aI., 1995; Gatev et al., 1999; Lord et aI., 1999; Rogers et aI., 2001). Somewhere in between static and dynamic posturography are the techniques where self-inflicted body perturbations are used - for example voluntary weight shifts while standing on a stable underground (Beckley et aI., 1995; Ondo et aI., 2000) - or where subjects actively balance on an unstable support surface (Begbie, 1967; Dietz et aI., 1980; Fitzpatrick et aI., 1992; Perennou et al., 1997). An important difference between the various static posturography studies is based on the choice of the outcome measures and the analytical approach of the obtained data, for example whether one assumes a closed or open loop control strategy of balance during quiet standing. Illustrations of actual static posturography experiments will be presented below in the section on recording equipment. 20.3. Dynamic posturography 20.3.1. Moving platforms

Dynamic posturography is best known for its use of moving platforms. In the 1920s and 1930s,

297

platforms were introduced that could be rotated relative to the horizontal plane by the manual force of an experimenter (Rademaker and Garcin, 1932). This rotating platform technique gained more widespread recognition by Martin's seminal studies in the 1960s on tilting reactions in man (Martin, 1965). Horizontal "translational" perturbations have also been used since the 1930s, for example by pulling a cord attached to a waggonette upon which subjects were standing (Jongkees and Groen, 1942). Other movable platforms consisted of an unstable support surface (Begbie, 1967), hence the subject's own continuous movements caused the perturbations (Fig. 1A). In the 1970s and 1980s, important new developments took place that were stimulated to a large extent by medical engineering scientists with an interest in balance control among patients (Allum and Budingen, 1979; Nashner et al., 1982; Diener et aI., 1983; Dietz et aI., 1984). One significant improvement was the introduction of powerful torque motors that assured delivery of rapid perturbations (Figs 1B and 1C). These perturbations could be highly standardized across subjects of different height and weight owing to development of servocontrolled motors. An additional new feature was the presence of a visual enclosure that could move along with, or independent from, the movements of the support surface (Fig. 1B). Also, posturography became increasingly computerized, allowing for automatic delivery of sophisticated test protocols with randomly mixed variations of perturbations (Nashner, 1982; Diener et aI., 1983). Such motorized and computerized platforms were among the first devices to be marketed commercially, leading to more widespread use of dynamic posturography in scientific and clinical settings. For many years, the movable platforms were capable of delivering postural perturbations in only one direction, usually the pitch (anterior-posterior) plane. However, in the 1990s physiologists began to develop a new interest in the control of roll (mediallateral) stability (Day et aI., 1993; Maki et aI., 1996). At the same time, epidemiological surveys highlighted the fact that falls in daily life could occur in any possible direction, and that lateral instability could be particularly relevant for understanding the genesis of hip fractures (Nevitt and Cummings, 1993; Greenspan et al., 1998; Lord et aI., ,1999). These new insights formed the basis for the de-

298

B.R. BLOEM ET AL.

Fig. 1. Several developments in the field of moving platform posturography. A: An early example of a movable platform. The support surface is attached to two crossed struts and is in equilibrium by itself; subjects standing on the platform render the system unstable - within predefined limits, as determined by physical stops. 1\\10 of the pivots were equipped with potentiometers, thus providing one of the earliest forceplates. Modified after Begbie (1967), with permission. B: Example of a movable platform - manufactured by NeuroCom International, Clackamas, Oregon - where a servo-controlled torque motor delivers highly reproducible postural perturbations (Nashner et al., 1982). This platform has only one axis, hence postural perturbations are restricted to a single direction. The platform movements consist of rotations, co-linear with the ankle joints, or horizontal translations. The visual enclosure can also move, along with or independent from the movements of the support surface. Redrawn after a figure provided courtesy of NeuroCom. C: Example of a dual-axis movable platform that allows for multidirectional postural perturbations: purely forward or backward, purely sideways, and combinations thereof (Carpenter et al., 1999; Allum et al., 2002). 0: Example of a multidirectional movable platform that is driven by gravity. The platform drops when the supporting magnets on three of the four sides of the platform are released, leaving one side attached. The perturbation size can be varied by changing the amount of slack in the suspending cables (controlled by a simple motor). Note the large support surface - 1 m 2 - that permits study of protective stepping responses. Modified after Comrnissaris et al. (2002), with permission.

velopment of multidirectional platforms that could deliver postural perturbations in either the pitch plane, the roll plane, or combinations thereof (Maki et al., 1996; Henry et al., 1998;Allum et al., 2002)see Fig. l C and D. Another interesting area was the

shift in attention from "feet-in-place" reactions where subjects are forced to deal with postural perturbations without moving the feet - to the analysis of compensatory stepping responses. This calls for a large supporting surface, such as the one

POSTUROGRAPHY

299

illustrated in Fig. lD. This particular platform, which is driven by the force of gravity, has another useful feature because it can deliver highly destabilizing postural perturbations with a very large rotation angle; an alternative would be to deliver very rapid perturbations by a torque motor. Hence, subjects can be tested around their limits of stability, and this might be necessary to understand the physiological mechanisms associated with actual falling incidents.

missaris et al., 2002; McDonagh and Duncan, 2002). Commonly used stimuli are toes-up rotational perturbations or forward translations, both types leading to a backward displacement of the center of mass (COM). Their counterparts are the toes-down rotations and the backward translations, which both induce a forward body displacement. There are several relevant differences between rotations and translations. First, rotational perturbations more consistently elicit stretch reflexes than horizontal translations because it is easier to drive the rotating platform faster. Second, rotations are subjectively more destabilizing than translations, and this may explain why postural responses habituate less rapidly following rotations. Third, the postural strategies used to restore balance are very different for both perturbation types. For example, hip and knee movements following backward translations are different from those following toe-up rotations, despite a similar amount of ankle joint rotation (Allum et al., 1989). Also, vertical head displacements are differently directed, and often more pronounced following rotational perturbations. A possible theoretical advantage to translations is the similarity to some daily life situations, such as standing in an accelerating bus or train. However, use of "unfamiliar" types of perturbations, such as rotations, may have different benefits to study falling mechanisms, because falls are almost by definition unexpected and unpractised events.

20.3.2. Types ofplatform movements

Movable platforms can deliver various types of postural perturbations (Fig. 2). As mentioned earlier, most platforms can move in only one direction, but more recent versions allow for perturbations in multiple directions (Henry et al., 1998; Rogers et al., 2001; Allum et al., 2002). The platform movements can be divided into: (a) rotations - "toes-up" or "toes-down" - usually around an axis co-linear with the ankle joints, (b) horizontal translations - forward or backward - or (c) purely vertical displacements. A specific variant involves the combined use of rotations and translations, in order to manipulate the amount of ankle rotation (Bloem et al., 2000). Most perturbation types are intended to induce a controlled mini-fall. The purely vertical displacements can be used to elicit startle-like responses or to study landing mechanisms (Bisdorff et al., 1999; Com-

A

B

Rotation

c

Translation

D

Translation & rotation

E

Vertical displacement

Fig. 2. Different types of platform movements. A: Baseline position. B: Rotational movements of the support surface, usually about the ankle axis as shown here. C: Horizontal translations of the support surface, shown here for the anteriorposterior direction, but medial-lateral or diagonal translations are used as well. D: Combinations of simultaneously occurring rotations and translations (explained in more detail in Fig. 15). E: Purely vertical displacements.

B.R. BLOEM ET AL.

300

It is possible to vary the magnitude, velocity and displacement profile of the platform movements. Most investigators use rapid and brief perturbations in order to study defensive postural reactions, in particular a "ramp" perturbation profile as this is optimal for evoking postural responses. Rotational perturbations usually have an amplitude of anywhere between 2 and 10 degrees, and the displacement velocity usually varies between 20 and 150 deg/s; the stimulus duration is always less than 200 ms. A rotation velocity of around 20 deg/s is approximately the lowest threshold to elicit stretch responses, at least in soleus muscles. Horizontal displacements usually have an amplitude of anywhere between 1 and 15 em, and the velocity of displacement usually varies between 5 and 100 cm/s. It must be realized that only the largest and most rapid platform movements are destabilizing for most healthy subjects, and even patients with moderate balance impairment can correct for most perturbations without much difficulty. This is an important drawback when it comes to studying the mechanisms associated with actual falls. For this reason, several investigators have begun to use more destabilizing stimuli, in order to study healthy subjects and unstable patients at the limits of their stability (Schieppati et aI., 1994; Hsiao and Robinovitch, 1998). An alternative to these rapid perturbations is the use of a platform that "oscillates" continuously back and forth at a much lower velocity; the oscillation frequency can be predictable or random (Diener et al., 1982; Maki et aI., 1990; Dietz et aI., 1993; Baloh et aI., 1995; Coma et al., 1999). Such predictable perturbations can be used to study stimulus anticipation and feed-forward postural control mechanisms.

20.3.3. Other types ofpostural perturbations Although dynamic posturography is often equated with moving platforms (Furman et aI., 1993), various other types of stimuli can be used to perturb posture (Table 2). One possibility is the use of external forces applied directly to one or more body segments, such as the trunk, shoulders or waist (Wolfson et aI., 1986; Cresswell et aI., 1994; Gilles et aI., 1999; Rietdyk et aI., 1999). A widely used example is the "postural stress test", where a weight is attached via a cable to a belt worn around the waist (Fig. 3A). Sudden release of this weight

Table 2 Types of physical perturbations, as used in dynamic posturography. External perturbations • Moving support surface o Translation, rotation or vertical displacement o Unidirectional vs. multidirectional o Abrupt vs. continuous (e.g. sinusoidal) o Predictable vs. non-predictable stimuli

• Stimuli applied to upper body parts o Hips o Trunk o Head Self-inflicted perturbations • Voluntary weight shifts • Anticipatory postural responses • Balancing on unstable support surface

induces a rapid postural perturbation which, depending on the direction of the pull, leads to a fall forwards, backwards or sideways. Subjects can compensate for such perturbations in different ways, for example flexion of the hips with forward flexion of the arms - when pulled backwards - or a stepping response (Wolfson et aI., 1986). A drawback here is the difficulty in standardizing the stimulus for subjects of different height and weight, but this can be partially overcome when servo-controlled torque motors deliver the pull. A promising new development involves the use of such motors to deliver perturbations in multiple directions (Matjacic, 2000). Alternatively, subjects can be placed in a forward or backward reclining position, while being supported by a firmly attached cable. Release of this cable then induces a fall forwards or backwards (Do et aI., 1988). Another variant uses an external push applied directly to the pelvis or the trunk (Fig. 3B). An advantage of all these upper body perturbation techniques is the possible relevance for daily life, where subjects are frequently being pushed or pulled aside, for example while walking in a busy crowd. Another advantage of upper body perturbations is the ability to study how afferent sensory information from the pelvis and trunk - "proximal triggers" helps to shape balance corrections. Another possibility is the use of self-inflicted postural perturbations. While such perturbations are "dynamic", the border between static and dynamic

POSTUROGRAPHY

301

Fig. 3. Other types of postural perturbations. A: The "postural stress test", where the stimulus is applied at waist level. Note the different types of normal balance reactions, and the absence of postural corrections on the far right. Redrawn from Wolfson et al. (1986), with permission. B: An example of physical stimuli applied directly to upper body segments, in this case the pelvis or shoulders (four different types of perturbations are possible). This set-up was used by Rietdyk et al. (1999).

posturography is hazy because the tests are usually performed while balancing on a static platform. One example is the analysis of voluntary weight shifts in an anterior-posterior or medial-lateral direction (Fig. 4A). These weight shifts can be guided externally, for example by a continuously moving visual stimulus (Beckley et al., 1995), or by fixed visual targets towards which the subject has to move (Ondo et al., 2000). Alternatively, subjects can be asked to shift their weight as far as possible. Such experiments provide valuable insights into the actual and perceived limits of stability under normal and pathological conditions. A second example of selfinflicted postural perturbations involves the use of voluntary upper body movements, such as weight lifting or raising the arms (Fig. 4B). This type of experiment is well suited for the study of anticipatory postural responses (Cordo and Nashner, 1982; Paulignan et al., 1989; Aruin and Latash, 1996; Benvenuti et al., 1997). A third option is to use an unstable rocking support surface, so that subjects

perturb themselves by their continuous shifts in the COM (Figs. lA and 4C). A final possibility formally beyond the range of posturography as we have defined it - is to study posture and balance in freely moving subjects, for example while walking or climbing stairs. Such gait assessment techniques are common to those of posturography and are described in Chapter 2.20. 20.4. Recording equipment 20.4.1. General principles

Inherent to all posturography measurement techniques is the ability to quantify posture or balance in an objective manner. This can be achieved in various different ways (Table 3). One widely applied option is the use of biomechanical outcome measures, which can be separated into kinematics - analysis of how body parts move - and kinetics - analysis of forces and joint moments. Another possibility is the

B.R. BLOEM ET AL.

302

Fig. 4. Various types of "self-inflicted" postural perturbations. A: Voluntary weight shifts, in this example guided by two visual targets and display of the person's own center of foot pressure. B: Lifting of a weight. C: Balancing on an unstable support surface.

use of electromyography (EMG) to describe muscle activation patterns. We will first describe these different techniques separately. This is followed by an illustration of the combined use of different modalities ("multimodal posturography").

20.4.2. Kinematics

The body's kinematics can be recorded in different ways. An early application in this area involved the use of cinematography (Martin, 1965), but this

Table 3 Recording equipment and outcome measures. Equipment

Outcome measures

Kinetics

Forceplates

Center of foot pressure Torques Shear forces and moments

Kinematics

Motion sensors • Linear or angular • Displacement, velocity or acceleration

Center of gravity Segment motion

Optical motion analysis

3-D spatial representation of body (parts) in time

Surface electrodes Needle electrodes Inserted wire electrodes

Background muscle activity Individual postural responses Postural synergies

Electromyography

POSTUROGRAPHY

technique was hampered by low sampling frequency. Another possibility is the use of lightweight position or motion sensors that can be attached to different body segments in order to record their position and, provided the sampling rate was fast enough, velocity or acceleration (Fig. SA). Ideally, such sensors should not interfere with the subject's natural movements. Linear or angular accelerometers may also be used to record acceleration directly. The information derived includes absolute changes in joint angles, and changes in the position of body segments relative to each other or with respect to an earth fixed joint. This approach has been used to quantify movements of the head, trunk, arms and upper or lower legs. Recordings from multiple body segments can also be used to arrive at an estimate of the COM (Peterka and Black, 1990). As such, kinematic studies can provide insight into postural strategies: fixed movement patterns of different body segments with respect to each other that are used to restore balance in response to specific postural perturbations. Examples of such postural strategies include the ankle strategy - where small support surface translations lead to movements that occur predominantly about the ankle joint - and the more complex "multi-link" strategy - where rapid translations or rotations of the support surface induce multisegmental movements, involving the ankles, knees and hips (Allum et aI., 1989; Allum and Honegger, 1992). One use of motion sensors is illustrated in Fig. SA, and more detailed in Fig. 6, which shows two lightweight angular velocity transducers that are mounted orthogonally in a portable box attached to a waist-worn belt (Allum et aI., 2001; Gill et aI., 2001; De Roon et aI., 2003). Depending on their orientation with respect to upright, these sensors can measure the angular velocity of the lower trunk in two different directions - usually the pitch and roll plane. This device can be used to quantify the kinematics of the trunk for both stance and gait tasks of posturography, for example during quiet stance, while climbing stairs or while undergoing a sudden pull at the shoulders from behind (Fig. 6C, D and E). Using mathematical integration or differentiation, the angular velocity signals can be converted into angular displacements or accelerations. Note that upper body movements in freely moving subjects have also been measured using linear accelerometers, but here one must

303

correct the sensor output for the subject's body inclination (Moe-Nilssen, 1998). A drawback of motion sensors is that these only provide information on the body segments on which they are mounted. A more attractive, but also more expensive and complex possibility involves the use of optical motion analysis systems for stance and gait tasks, which can visualize overall body movements (Fig. SB). Such systems use multiple - up to six - cameras that are aimed at the subject from different angles. These cameras monitor signals derived from markers that are generally attached to standard anatomical landmarks, such as the acromion or lateral malleolus. Other systems use passive reflective markers that reflect the light of a centrally positioned source. Further variants use diodes that actively emit infra-red light. Use of multiple markers - usually about 20 - assures that body segment motion can be viewed in multiple planes. Use of many cameras assures that each marker is "seen" throughout its full range of movement, and permits calculation of marker position in three-dimensional space. The resultant coordinates can subsequently be connected to generate a "stick figure" representation of the human body moving in three-dimensional space (Fig. SC). 20.4.3. Kinetics

Measurement of kinetics provides a net estimate of active and passive contributions to joint torques comprising the postural neuromuscular response. This offers a clear advantage over kinematic measures and EMG, because active muscle forces as well as the viscoelastic properties of stretched tissues can be estimated. In early experiments, only reactive forces at the ankle joints were assessed in a simple manner, for example by placing subjects with one foot on each of two scales and recording the difference in pressure from both feet. Nowadays researchers use forceplates equipped with multiple strain gauges that are embedded into a supporting platform (Fig. 7A). These strain gauges measure the postural reaction forces exerted by the feet onto the platform during standing. From this information, the center of foot pressure (COP) and the reactive torques can be calculated. Most forceplates record only the vertical reaction forces - from which the anterior-posterior and medial-lateral directed torques can be calculated - but some also measure the

Fig. 5. Outcome measures used in posturography; similar systems are also used for gait analysis. A: Different types of motion sensors. These sensors can be linear or angular. B: Optical motion analysis systems. Note the two different types of markers. C: Example of a three-dimensional "stick figure" representation of a healthy person while balancing on a tilting platform, as derived from an optical motion analysis system.

~

~

~

5

t:tl

~

t:tl

~

w

POSTUROGRAPHY

305

Fig. 6. Illustration of the use of motion sensors. A: Two orthogonally mounted angular velocity transducers are attached to a waist-worn belt (SwayStar, manufactured by Balance Int. Innovations, Switzerland). The recorded angular velocities are transmitted to a PC interface via a long cable (Gill et aI., 2001; Allum et aI., 2001; De Hoon et aI., 2003). This permits measurement of trunk sway in freely-moving subjects, for example while climbing stairs (A). or while responding to external perturbations. B: Angular velocities of the trunk in the roll and pitch plane are illustrated for a young healthy subject while standing quietly (C), while climbing stairs (D) and while receiving a sudden shoulder pull from behind (E).

306

horizontal (shear) forces and moments. The COP deviations are estimated from the torques, assuming the body can be modeled as an inverted pendulum, with control of equilibrium originating in the lower extremities. With a multi-joint model of posture, torques at each joint can be estimated from recordings of segment inertias and angular deviations of the segments (Rietdyk et al., 1999). We will illustrate the use of COP recordings for a static posturography experiment. The spontaneous sway oscillations of the human body while standing upright are difficult to detect with the naked eye, but cause measurable movements of the COP. These movements can be plotted as a function of time (Fig. 7B). Alternatively, displacements of the COP in one direction can be plotted as a function of displacement in another direction, creating so-called "spaghetti plots" within a horizontal plane (Fig. 7C). Various summary derivations can be obtained from the COP recordings (Table 4). Stabilogram-diffusion analysis is an interesting approach to analyze

B.R. BLOEM ET AL.

whether COP trajectories behave as random walks (Collins and De Luca, 1994; Mitchell et al., 1995). This provides physiological information about the underlying balance control mechanisms, and is sensitive to changes that occur with balance disorders such as Parkinson's disease. However, there are various other approaches to analyze the changes in COP, particularly in relation to changes in the COM and EMG (Fitzpatrick et al., 1992; Winter et al., 1998; Gatev et al., 1999). Note that COP recordings are often equated with "body sway", but this is only correct if the body moves as a rigid pendulum - which is often not the case - and only when sway angles and the inertial forces are small. Sway can refer to the movements of the body's COM in three-dimensional space. It can also be used to describe how certain parts of the body "sway", for example the trunk. During balance control, we try to keep our "sway" within certain safety limits. The vertical projections of reaction forces onto the platform are not always a good

Fig. 7. Analysis of the center of foot pressure (COP). A: Illustration of strain gauges imbedded into the support surface. As shown here, most forceplates have strain gauges in all four comers. This particular forceplate can record both the vertical and horizontal reaction forces - indicated by x, y and z. B: Plotting of the COP in the medial-lateral (x) and anteriorposterior (y) directions, as a function of time. C: Plotting of the COP in one direction as a function of displacement in another direction. Figures Band C were redrawn from Collins et al. (1994), with permission,

307

POSTUROGRAPHY

Table 4

20.4.4. Electromyography

Derivatives of static posturography measurements.

Another commonly used outcome measure is EMG. Inserted needle studies are anatomically more focal than surface EMG, but are rarely used because these might interfere with the subject's natural movements and because they present small health risks. Inserted wire electrodes hardly interfere with movement, but record only from a small portion of the muscle and produce comparable interindividual variability as surface electrodes (Horstmann et al., 1988). Surface EMG is therefore most used widely. Here, drawbacks to consider include the problem of crosstalk, the variable impedance - both across and within subjects, even during a particular experiment - and the need to normalize EMG responses for intra- and interindividual comparisons. No normalization procedure is generally accepted as the best, but one possibility is to relate EMG response amplitudes to an independent measure of maximal muscle contraction, achieved either voluntarily or by a supramaximal nerve shock (Bloem et al., 1996). Surface EMG can be used for various purposes. First, it can be used in a quantitative manner, for example to provide an estimate of background muscle activity (BGA) levels in "resting" muscles when standing quietly. This BGA depends on many variables, such as the specific posture of the person and the anticipated balance threats. For example, recordings during quiet stance suggest that BGA in the tibialis anterior muscle rarely exceeds 5% of maximal voluntary contraction (Panzer et al., 1995). Considerably higher activity can be noted in persons who actively assume a stooped "parkinsonian" posture (Bloem et al., 1999). Online visual or auditory feedback of BGA could be used as a feedback mechanism to assure a consistent posture over time within a given individual, or across different subjects. Levels of BGA are often abnormal - usually increased - in patients with balance disorders. This can be caused by the underlying disease process, e.g. rigidity in Parkinson's disease, a fear of falling, an abnormal stance, e.g. a stooped posture, or a combination of these factors (Bloem et al., 1993; Horak et al., 1996; Carpenter et al., 2001). During dynamic posturography experiments, surface EMG can be used to measure active balance reactions in individual muscles. These postural responses can be characterized in terms of onset

• Summary statistics (Furman et aI., 1993) o Anterior-posterior displacement of center of foot pressure (COP) o Medial-lateral displacement of COP o Displacement area o Displacement path o Power spectrum • Stabilogram-diffusion analysis (study of random walks) (Collins and De Luca, 1994) o Diffusion coefficients o Scaling exponents o Critical point coordinates • Stiffness measures (Fitzpatrick et aI., 1992; Winter et aI., 1998; Lauk et aI., 1999; Loram et aI., 2001) • Torques

description of sway. In fact, the COP is a reflection of the controlling elements, that is, of the passive and active forces that try to prevent sway from becoming excessive. COP recordings are therefore more dynamic than sway recordings, particularly when the subject's sway is relatively fast. These differences are illustrated in Fig. 8 for both static and dynamic conditions. Calculations of such timing differences error signals - between the COP and COM have been used to estimate stiffness (Winter et al., 1998; Loram et al., 2001). Measurements of COP can be used in an attempt to standardize each individual's baseline posture prior to every trial within a given experiment. For example, the COP can be sampled at the beginning of an experiment while the subject assumes a preferred stance position. During the experiment, subjects can be confronted with on-line visual feedback of their actual COP, relative to their own "preferred-stance" reference values. Using this visual feedback, subjects can be asked to maintain their COP within a predefined range. This helps to prevent subjects from gradually changing their initial position, either due to fatigue or due to stimulus anticipation (Bloem et al., 1999).

308

-

B.R. BLOEM ET AL.

6.2

A

Static posturography

6.0 E u

Q)

::i2

c

5.8

ItI

-

E 5.6 0 ~

Q)

u

c

5.4

ItI

+"" III

(5

5.2

Center of mass Center of pressure

5.0 0

4

1

2

3

B

4 5 Time (s)

6

7

8

Dynamic posturography

2

E

~

0

+""

cQ) E Q)

-2

a.

-4

U ItI III

....

_-_

..... ;

(5

-6

Center of mass Center of pressure

-8

o

0.5

1.0 Time (s)

1.5

2.0

Fig. 8. A: Recordings of the center of foot pressure (COP) and center of mass (COM) for a single subject under static conditions - quiet stance on a stable platform. Note that the COP moves more rapidly and with a greater amplitude than the COM. This figure was redrawn from Winter (1990), with permission. B: Similar recordings under dynamic conditions in a single upright standing subject, whose balance was perturbed by a sudden toe-up rotational movement of the support surface. Note again the difference in dynamics between the COP and COM.

309

POSTUROGRAPHY

latency, duration and amplitude - the latter can be calculated as the area under the curve, or as the peak amplitude. Such calculations are essential to evaluate the modifiability of postural responses according to functional demands, and to determine whether and how such postural responses are pathologically changed in patients with balance disorders. Onset latencies are usually defined by the moment when the response amplitude consistently exceeds a predetermined level of prestimulus BGA, typically by

Individually determined onset and offset of response

more than two standard deviations. More difficulty arises when the response amplitude and duration from this onset need to be determined. This problem is illustrated schematically in Fig. 9. Some responses, in particular the early occurring stretch reflexes, have a clear onset and offset. In this case, response amplitudes can easily be calculated, and comparisons between differently sized and differently timed responses are straightforward (Fig. 9A). However, for many other responses the precise offset

Fixed time windows

Fixed length of interval, with variable onset latency

Fig. 9. Calculation of EMG response amplitudes. A: Determining the exact onset and offset of each individual response provides accurate estimates of response duration (d), and distinguishes between similarly shaped responses with a different amplitude (a) - response types 1 and 2. However, this technique cannot distinguish between differently shaped responses with a similar area under the curve, such as response types 1 and 3. Additional problems arise when the response offset is unclear, for example due to blending with later occurring responses (response type 4). B: Calculation of net EMG activity over a fixed time window - irrespective of any identified response onsets - is functionally informative as it reveals how much balance correcting activity is present at a given stage of the postural reaction (Carpenter et al., 1999; Allum et al., 2002). Delayed onsets are thus "translated" into a low level of EMG activity - compare response types 1 and 2. A drawback is the lack of insight into actual reflex behavior, for example delayed onsets or changes in waveform. C: Calculation of response amplitudes over a fixed time window following response onset avoids contamination by later occurring responses (Hansen et al., 1988; Beckley et al., 199Ib). A disadvantage is the differential outcome for responses with a different waveform. Thus, amplitudes for response types I or 2, which rapidly reach their peak amplitude following onset, are relatively overestimated relative to response type 3. D: A final possibility is to determine the peak amplitude of each postural response and to calculate the response area over a fixed time interval on each side of this peak (Grey et al., 2002). This method gives a good estimate of the maximal response size. Problems are the difficulty in identifying the peak - certainly in single subject traces - and the need to have variable intervals for differently shaped responses.

310

and onset latency remain unclear due to blending with a subsequent other response, or with tonic bursts of voluntary muscle activity.This problem can be addressed in three different ways. First, response amplitudes can be expressed as the average magnitude of the EMG signal within a predetermined and fixed time window (Fig. 9B). Based upon prior knowledge of postural control in healthy subjects, such time windows can be rationally determined in order to "capture" the expected appearance of relevant postural activity (Carpenter et aI., 1999). Examples of such predetermined time intervals are 40-100 ms from stimulus onset - to capture stretch reflex activity - or 120-220 ms - to capture balance correcting responses. An advantage is the direct functional relevance to postural control: is there little or much activity in a particular muscle during important stabilizing periods? A disadvantage is the obscured insight into merely delayed responses. For example, lack of EMG activity within a given time window could be caused by either complete absence or a mere delay of the normal response; in the latter case, the delayed response would show up in a later window, giving rise to increased activity that could be falsely interpreted as "compensatory". The second approach accommodates this drawback by means of a fixed time window that starts at the measured onset of the postural response (Fig. 9C). For example, amplitudes can be calculated over a fixed 75 ms window following onset of the response (Hansen et al., 1988). Although this technique does not always encapsulate the entire response duration, it effectively avoids contamination with later response activity without blurring physiological insights. However, this approach still assumes that onset latencies can be defined correctly. A third approach is to avoid determining the onset of a response and seek its maximum amplitude (Grey. et aI., 2002). Once this is done, the area is calculated over a fixed interval, usually about 50 ms to either side of the maximum. Further interpretation of the response size depends on knowledge of BGA levels, because these show a strong positive correlation to the amplitude of postural responses evoked in stretched and, to a lesser extent, shortened muscles (Bloem et al.• 1993, 1999). It is therefore necessary to take prestimulus BGA into account if valid interpretations of changes in "pure" response amplitudes are to be made.

B.R. BLOEM ET AL.

Ideally, investigators should aim at keeping the amount of BGA constant within and across subjects, for example by supplying muscles with a constant pre-loading force - standing on a slightly tilted support surface - or by instructing subjects to relax completely. However, a disadvantage is that small differences in BGA, which are likely to persist, may still considerably change reflex amplitudes. Furthermore, this approach may fail in patients with neurological disorders due to their inability to relax or inability to maintain a constant pre-loading force throughout the period of examination. An alternative approach, which consists of an "after-the-fact" mathematical correction for differences in BGA, is therefore often needed. For this purpose, it is common practice in dynamic posturography experiments to record the level of prestimulus BGA, usually over a period of 50-100 ms prior to the stimulus. A common correction technique consists of subtracting this amount of BGA from the average response amplitude. This "difference method" is reasonably effective in correcting for BGA in most subjects, but cannot be used routinely since BGA continues to exert a strong influence on corrected reflex amplitudes in some individuals (Bloem et al., 1993). Covariance analysis is a more flexible correction technique that can be applied routinely to every subject, irrespective of the magnitude of the relation between BGA and response amplitudes (Fig. 10). Covariance analysis is a common statistical procedure designed to remove the influence of a confounding variable by estimating its relation to the dependent variable and providing post-hoc statistical control for this influence, leaving all other sources of variability intact. Once determined reliably, onset latencies and modifiability of response amplitudes can be used as criteria to distinguish between pure reflexes, automatic postural responses and voluntarily initiated movements. Pure reflexes are at one end of this spectrum. These are stereotyped responses to sudden muscle stretch, with an early and fixed onset latency, and a response amplitude that cannot be modified easily. Voluntary postural corrections are at the other end of the spectrum. They can be activated reactively following external postural perturbations, or in a feedforward manner, for example while making selfinitiated weight shifts. Their onset latencies are much later and more variable than those of stretch reflexes, and the magnitude depends on conscious

311

POSTUROGRAPHY

OJ

'"0 ::J

.t:

0..

E


OJ

'"0 ::J

.t:

0.. E
Background

activity (a.u.)

Fig. 10. Correction for background muscle activity (BOA), using covariance analysis (Bloem et al., 1993). A: There is a strong correlation between BOA levels and the actually observed reflex amplitudes (open circles) which can be estimated using linear regression analysis. If the regression perfectly predicted changes in reflex amplitudes, all observations should lie on top of the regression line (closed circles). This is not the case because of many other influences on reflex amplitudes, hence the observed amplitudes are scattered around the regression line. This variability can be quantified by calculating residuals, i.e. the difference between observed and predicted values. These residuals are a reflection of reflex amplitudes, completely independent from the effects of BOA. This is illustrated in B which shows that BOA has no influence on residual amplitudes. Covariance analysis automatically provides both the regression coefficient and the residuals. Abbreviation: a.u.=arbitrary units.

decision making. Most balance reactions reside somewhere between the two ends of this spectrum. Their onset latency is rather fixed and too early or invariable to be voluntary, yet factors such as prior instruction or expectation can markedly influence the response amplitude (Horak et al., 1989; Bloem et al., 1995) - see Section 20.6.3 for more details on response modification. For this reason, the term "automatic postural responses" is best used for those balance reactions that are neither pure stretch reflexes nor strictly voluntary movements. The terminology for responses within the category of automatic postural responses is inconsistent and sometimes confusing (Table 5). One approach emphasizes the timing relative to stimulus onset.

Here all automatic postural responses are labeled "long latency", this to distinguish them from the preceding "short latency" stretch reflexes. This proved helpful to describe balance reactions in lower leg muscles following translational perturbations of posture, because most long latency responses occurred in the same stretched muscle (Nashner, 1976). However, rotational platform movements evoke distinct automatic postural responses in antagonist muscles of the lower legs (Diener et al., 1983). This led to the introduction of the term "medium latency" for balance reactions that occurred in stretched muscles; the term "long latency" response was now reserved for the slightly later occurring response in the shortened antagonist

312

Table 5 Terminology of automatic postural responses. Named after timing relative to stimulus onset

• Short latency • Medium latency • Long latency Named after presumed postural function

• Simple stretch reflexes • Functional stretch reflex • Balance correcting responses (primary or secondary) Named after supposed underlying circuitry

• Monosynaptic spinal responses • Polysynaptic spinal responses • "Longloop" postural responses

muscle (Fig. 11). The problem with this approach is that medium latency responses, although of shorter duration than long-latency responses in antagonistic muscles, have similar onset latencies. Another approach tries to reflect the presumed function for postural control in its terminology. Thus, some investigators refer to automatic postural responses as "functional stretch reflexes" or "balance correcting responses", thus separating them from the functionally less effective stretch reflexes that occur shortly after the perturbation (Nashner, 1976; Carpenter et al., 1999). Depending on the onset latency relative to stimulus onset, balance-correcting responses can be subdivided into primary or secondary. Recognition of physiological differences among these functionally important responses formed the basis for the parsing algorithms described in Fig. 9B. A third possibility is to name postural responses after their supposed underlying circuitry. Commonly used terms are monosynaptic spinal responses - a synonym for the early stretch reflexes - and polysynaptic spinal responses - to describe the majority of . automatic postural responses in leg muscles. The older literature also contains descriptions of "long loop" postural responses, as a reflection of assumptions that neural pathways involving the brainstem or cortex might be involved in their generation. This term is best avoided for postural responses in the legs, trunk and upper arms, as there is no conclusive

B.R. BLOEM ET AL.

evidence to support the existence of such long loop pathways in humans. Examples of actual EMG responses during a dynamic posturography experiment are shown in Fig. 12. Most initial studies focused on lower leg responses (Fig. 12A), but later studies went on to analyze more proximal muscles of the upper leg, pelvis, trunk, neck and arms (Fig. 12B). These figures also illustrate how quantitative analyses of individual EMG responses can be used to detect differences in patients with balance disorders, and to study the pathophysiological mechanisms leading to postural instability and falls. For example, recordings from lower leg muscles of patients with a severe polyneuropathy usually show absence of stretch reflexes, as well as a delayed onset and diminished amplitude of balance correcting responses (Fig. 12A). In contrast, patients with Parkinson's disease typically have enlarged amplitudes of medium latency and balance correcting amplitudes in all muscles involved in postural control illustrated for the gluteus medius and paraspinals in Fig. 12B. Apart from providing quantitative measures of individual postural responses, surface EMG can also be used in a qualitative manner to describe the fixed and pre-programmed spatiotemporal activation patterns of automatic postural responses; these are termed postural synergies. For this purpose, it is best to record from multiple muscles spanning different body segments, including the limbs and trunk, because recording from only a few muscles can provide a distorted view of the synergy. An example of this is the frequently mentioned "distal-toproximal" strategy that is induced by horizontal translations of a supporting surface. Here postural responses seem to appear first in ankle muscles, while subsequent responses can be observed in muscles of respectively the thigh and trunk (Nashner and McCollum, 1985). However, later studies showed that responses appeared in neck and arm muscles even earlier than in the leg muscles (Keshner et al., 1988; Allum et al., 1993; McIlroy and Maki, 1995). Other types of postural perturbations - for example, rotational movements of a supporting surface or forces applied primarily to the trunk - can elicit different postural synergies, with prominent stretch and balance correcting responses appearing first in proximal muscles spanning the hips and trunk (Do et al., 1988; Allum et al., 2002).

Fig. II. Terminology of automatic postural responses, illustrated for the commonly used toe-up rotations which induce a rapid ankle dorsiflexion. Stretched soleus: I = short latency response or monosynaptic spinal stretch reflex; 2=medium latency response, polysynaptic spinal response or functional stretch reflex. Shortened tibialis anterior muscle: 3 = long latency response, balance correcting response or shortening reaction. These automatic postural responses are followed by more voluntary balance reactions (labeled 4) which, although shown here as clearly separate responses, usually blend with preceding balance reactions.

v.> ..... v.>

d!-<

~~

en

(3

314

B.R. BLOEM ET AL.

Fig. 12. Illustration of actual EMG responses, recorded with surface electrodes. A: An example of timing differences, shown for postural reactions to toe-up rotations of a supporting platform in a single patient with dorsal root ganglionopathy. Note the absence of stretch reflexes and clearly delayed balance corrections in the patient. Redrawn and modified from Bloem et a1. (2002), with permission. B: An example of amplitude differences in patients with Parkinson's disease, shown for postural reactions to a sudden tilt of a supporting dual-axis platform (simultaneously to the left and backwards - 2250 in the annotation illustrated in Fig. Ie). Note the increased response amplitudes in patients, as well as their high level of prestimulus background muscle activity. Redrawn and modified from Bloem et a1. (2001), with permission.

315

POSTUROGRAPHY

20.4.5. Multimodal posturography

A comprehensive view of postural control is obtained when multiple domains are measured simultaneously, including the kinetics, kinematics and surface EMG. An example is illustrated in Fig. 13 for a dynamic posturography experiment, where young healthy persons were exposed to sudden rotational movements of a supporting platform. The stimulus consisted of a combined movement of a dual-axis platform in both the pitch plane (toes-up) and roll plane (to the left). We will illustrate how the kinematics can help to interpret particular EMG findings. Surface EMG is illustrated for six different muscles, two with bilateral recordings. For example, the stretched soleus shows a clear stretch reflex with an onset at about 50 ms after the onset of first platform movement; this is followed by a prominent burst of activity in the shortened tibialis anterior muscle. In this case, tibialis anterior activity would theoretically help to regain balance control because the COM is forced backwards by the platform movement. Conversely, soleus activity would act to propel the body even further backwards (Dichgans and Diener, 1987). The net influence of these muscle responses on postural control cannot be retrieved from EMG alone. For this, one needs to examine the simultaneously recorded kinetics. The anterior-posterior ankle torque trace first shows a positive deflection shortly after stimulus onset, reflecting the passively increased pressure on the forefeet due to the upward moving platform that stretched triceps surae muscles. This will have been aggravated by the destabilizing plantar flexion force of the stretch reflex in soleus muscle. This is followed by a negative deflection of the anteriorposterior ankle torque, reflecting build-up of active stabilizing forces. Simultaneous inspection of the EMG and torque traces suggests that the tibialis anterior activity would be appropriately timed to generate such stabilizing activity, certainly when a 25 to 30 ms electromechanical coupling delay is taken into account (Allum and Mauritz, 1984). The overall "impact" of the postural perturbation, as well as the result of the stabilizing muscle forces, can be derived from various kinematic parameters. In this experiment, trunk angular velocity in the pitch and roll planes was collected using motion sensors mounted to a metal plate that hung at the level of the sternum (Fig. 13). The trunk initially

pitches forward and rolls towards the right - that is, opposite to the stimulus direction in both planes and is then gradually returned to its initial position, indicating successful restoration of upright balance. Interestingly, these kinematic parameters are nicely paralleled by corresponding muscle activation patterns in proximal muscles. For example, the rightward roll of the trunk is reflected by an early stretch reflex in the right paraspinals, as well as brief release of the left paraspinals. The subsequent bursts of balance correcting activity in both paraspinals help to keep the trunk from falling towards the direction of platform movement. Furthermore, prominent balance correcting activity can be seen in the left gluteus medius, with much less activity occurring in the right gluteus medius. This activity serves to stabilize the pelvis which rotates in the opposite direction to that of the trunk.

20.5. Manipulations 20.5.1. General aspects

A key element of posturography is the ability to selectively manipulate elements of postural control. Two elements are particularly suitable for experimental manipulation: afferent feedback signals stemming from proprioceptive, vestibular and visual sources - and cognitive or postural set. 20.5.2. Afferentfeedback signals

All afferent feedback systems can be manipulated experimentally. This can be achieved in various ways, during both static and dynamic posturography experiments. Broadly speaking, there are three different approaches: (a) reducing or entirely eliminating the feedback signal; (b) distorting the feedback signal, thereby providing false information to the central nervous system; or (c) providing subjects with more (correct) feedback than under normal circumstances. The essence lies in a selective interference with one particular feedback signal, without altering the others. However, this is not always straightforward, as we shall discuss below. 20.5.2.1. Visualfeedback Visual feedback can simply be removed by asking subjects to close their eyes or to provide them with blindfolds. A more elegant manner is to distort visual feedback by providing subjects with faulty or conflicting visual information. Early experiments in

316

B.R. BLOEM ET AL.

Fig. 13. Multimodal posturography, showing simultaneous recordings of EMG (multiple distal and proximal muscles), kinematics (ankle torque) and kinetics (trunk velocities), relative to the actual platform movement (a sudden tilt of a dualaxis platform to the left and backwards - 225 0 in the annotation illustrated in Fig. IC). Redrawn and modified from Allum et al. (2002), with permission.

317

POSTUROGRAPHY

this field indicated that a moving visual scene could be used to examine a subject's spatial orientation, in particular the perception of subjective and postural verticality (Dichgans et aI., 1972). This technique can also be used to search for causes of postural instability in patients with vestibular or neurological

disorders (Bronstein et aI., 1996). In the example shown in Fig. 14A, seated subjects faced a visual background that was either stationary or continuously rotating at different velocities to the left or right. The visual roll-motion stimuli provided disorienting visual information and created an illusion

Visual roll stimuli Fig. 14. An example of visual feedback manipulation. A: Subjects are asked to align a straight line with perceived verticality while facing a rotating visual scene. B: Perceived visual vertical, as a function of visual roll. Patients with defectivelabyrinthsare more sensitive to visual roll stimuli than healthy controls. STAT =static condition;CW =clockwise rotation at 15, 30 and 60 CCW= counterclockwise rotation at 15, 30 and 600/s. Redrawn and modified from Bronstein et al. (1996), with permission. 0/s;

318

of self-rotation. This could be quantified by asking subjects to set a straight indicator line - displayed in the center of the visual scene - in what they presumed to be a pure vertical position. Clockwise rotation of the visual background induced a placement of the indicator line to the right, counterclockwise rotation to the left (Fig. l4B). Patients with a bilateral vestibular deficit showed much larger deviations, suggesting a role for the vestibular system in processing visual biases for the perception of verticality. More recent studies have distorted visual feedback by immersing subjects into a virtual reality environment (Cobb and Nichols, 1998; Owen et al., 1998; Tossavainen et al., 2001). Visual feedback can be enhanced by providing extra "cues", for example stripes on the floor to reduce freezing during walking (Nieuwboer et al., 1997). 20.5.2.2. Proprioceptive feedback Proprioceptive feedback can be reduced through cooling, anesthesia or ischemia of the legs (Dietz et al., 1980; Diener et al., 1984; Schieppati and Nardone, 1997; Grey et al., 2001). These interventions all lead to greater excursions of the COP, suggesting an increase in body sway. Note that neither of these techniques is a purely selective proprioceptive manipulation. For example, limb ischemia or anesthesia reduces proprioceptive feedback, typically only for a short period of time, but may also affect muscle strength which in turn impairs balance control (Wolfson et al., 1995). A better way to eliminate proprioceptive feedback from the lower legs is use of an "ankle nulling" procedure. This can be achieved by fixating the ankle joints, thus preventing proprioceptive feedback originating from joint receptors or stretched lower leg muscles (Gurfinkel et al., 1979). Alternatively, ankle inputs can be nulled by matching the platform movements to the measured movements of the subject (Allum and Honegger, 1998; Bloem et al., 2000). For example, computer-controlled delivery of a simultaneous rearward translation and toes-down rotation of a supporting platform results in negligible ankle movement over the first 250 ms (Fig. 15A). Indeed, ankle motion remains restricted to less than 10 rotation over this period, which indicates that the "ankle nulling" procedure had been successful (Fig. 15B). This is an effective way of eliminating lower leg proprioceptive drive of automatic postural responses. Indeed, short latency

B.R. BLOEM ET AL.

stretch reflexes are virtually eliminated by this ankle nulling procedure (Fig. 15C). Proprioceptive feedback can be distorted by placing subjects on a thick foam mattress (Baloh et al., 1995; Gill et al., 2001; Rogers et al., 2001). The problem here is that standing on foam may lead to instability simply because the balance reactions are less effective on a compliant surface. Another possibility is applying high-frequency vibration directly over tendons or muscles. This causes massive stimulation of type Ia afferents and thus produces a false impression of muscle stretch. Commonly targeted muscles include the triceps surae, tibialis anterior, gluteus maximus and dorsal neck muscles (Gurfinkel et al., 1988; Pyykko et al., 1991; Lekhel et al., 1997). When vibration is delivered during upright stance with eyes shut, the perceived muscle stretch creates an illusory body tilt and altered perception of body position. Healthy subjects compensate for this kinesthetic illusion by actively moving in the opposite direction of presumed body inclination. Figure 17A illustrates this for blindfolded young subjects who underwent vibration of both Achilles tendons (frequency 70 Hz, amplitude 0.5 mm, duration 25 s). The subjects perceived this as stretch of the triceps surae caused by dynamic forward leaning, and actively compensated by leaning backwards. This is reflected by the COP recordings in the anterior-posterior direction, which show a rapid backward displacement that stabilized after about 3 s of vibration when presumably vestibular and residual proprioceptive signals indicated the body was "vertical" (Fig. 17B). Proprioceptive feedback from the lower legs can be increased using sudden support-surface movements of variable size or speed (Diener et al., 1988; Beckley et al., 1991a). An example is shown in Fig. 15A, where the movements of a supporting platform consist of a simultaneous rearward translation and toes-up rotation. This combined movement induces a greater ankle dorsiflexion than a simple translation (Fig. 15B), thereby leading to an enhanced proprioceptive drive of postural responses. This is confirmed by surface EMG recordings from the soleus muscle, which shows brisk and early stretch reflexes during enhanced ankle input trials (Fig. 15C). A problem with this type of manipulation is that variable support-surface movements not only manipulate lower leg proprioceptive input, but also induce different movements of other body segments.

0

Fig. 15. Manipulation of ankle proprioception. A: A rearward platform translation causes ankle dorsiflexion (angle a), but this can be cancelled out by a simultaneous toes-down rotation of the platform which - on its own - would cause ankle plantarflexion. The net result is a "nulled" ankle angle (angle 13). Conversely, the combination of rearward translation and toes-up rotation of the platform - both of which cause ankle dorsiflexion - enhances the overall change in ankle angle (angle 'Y), thus producing "enhanced" ankle proprioceptive drive. B: The net effect ofthese manipulations is reflected by the ankle angle trace, which remains constant within ± 1 over the first 250 ms during ankle nulling trials, but rapidly reaches a large dorsiflexion angle during enhanced ankle input trials. C: Surface EMG recorded from the soleus muscles is consistent with the kinematics, because stretch reflexes (solid arrow) are minimally present during ankle nulling trials, but are brisk and early during enhanced ankle input trials. Note that the subsequent balance correcting responses (open arrow) are hardly affected by these manipulations, suggesting that lower leg proprioception is not required to drive these postural responses. Traces in B and C represent the population average of 15 healthy subjects. Data modified after Bloem et al. (2000), with permission.

25

\0

w ......

-<

I

d

[Il

320

This is by itself enough reason to expect difference balance reactions. Another way of providing increased somatosensory feedback is passive or active touch (Jeka and Lackner, 1995; Riley et al., 1997; Rogers et al., 2001). In the experiment shown in Fig. 16, subjects were instructed to stand on a thick foam mattress during some trials, and on a normal floor during others; this was done with and without visual feedback. To provide subjects with additional somatosensory feedback, subjects were gently touched from behind at the level of either the knees or the shoulders. The results showed a reduction of postural sway with either tactile cue, and simultaneous touch at both levels gave the greatest sway reduction. Furthermore, the effects of touch were most pronounced when baseline sway was greatest, i.e. with foam support, absence of vision, or both.

20.5.2.3. Vestibular feedback An elegant way to reduce the influence of vestibular feedback, along with proximal proprioceptive feedback, was developed by Fitzpatrick et al. (1992). Their subjects were strapped at the head and waist to a rigid post. By rotating their ankle joints, they had to balance a load that was matched to the subjects' own load stiffness - as measured in a separate experiment while normally standing. In this way, an "equivalent body" swayed at the ankles without affecting other sensory inputs. Low intensity electrical (galvanic) stimulation of the labyrinth is a commonly used technique to distort vestibular feedback, presumably by stimulating distal vestibular nerve afferents, particularly those with an irregular discharge (Lund and Broberg, 1983; Goldberg et al., 1984). The central nervous system apparently interprets the galvanic input as roll-tilt of the support surface. Thus, with a positive stimulating current polarity, the body sways in the direction of the mastoid under stimulation. With a negative polarity, the body sways in the opposite direction. These effects depend on the spatial orientation of the head: body sway is in the frontal plane when the head is held in the normal position, and in the sagittal plane when the head is rotated sideways. Vestibular information processing can also be distorted by changing the orientation of the head, for example to a hyperextended position (Brandt et al., 1986). Vestibular feedback can be enhanced using isolated head perturbations (Horak et al., 2001).

B.R. BLOEM ET AL.

Interpretation of the resultant responses can be difficult because head perturbations also stimulate afferents from the neck and, sometimes, from lower body segments.

20.5.2.4. Weightlessness The effects of weightlessness have been studied by analyzing adaptation of postural control in astronauts during or after space flights (Clement et al., 1984; Speers et al., 1998). Because such unique experimental conditions are rarely available, Dietz et al. (1989) developed an interesting alternative by immersing a movable platform into a large bath. This underwater paradigm reduces the natural force of gravity and permits study of loading mechanisms by providing subjects with a range of different weights, but of course increases the amount of tactile feedback. Such experiments provided valuable new insights into the physiology of load receptors, and their role in the pathophysiology of balance disorders. 20.5.2.5. Combined manipulations It is possible to systematically manipulate multiple sources of afferent information during a given experiment. One example was presented earlier (Fig. 16). However, the most widely used technique is the so-called "sensory organization test" (SOT), which selectively disrupts the subject's somatosensory feedback, visual feedback, or both (Nashner et al., 1982). The SOT uses a movable platform that is equipped with a movable visual surround (Fig. lB). Subjects stand quietly for 20 s during six different conditions, while the platform continuously records displacements of the vertical reaction forces (Fig. 18). An interesting aspect is that afferent feedback is not only eliminated during some conditions - for example eye closure - but also distorted during other conditions with "sway-referencing". Here the platform, the visual surround, or both, move after a small delay in the same direction as, and proportional to, a filtered version of the anteriorposterior displacement of the on-line recorded COP. This creates a false impression of the visual environment or of ankle movements, and forces subjects to rely more upon vestibular sensations and proprioceptive input from joints other than from the ankle to remain standing. A problem is that the SOT assumes that posture behaves as an inverted pendu-

Fig. 16. Influence of external touch on postural sway, assessed using an optical motion analysis system. A: Somatosensory feedback was enhanced by providing an external tactile cue to the knees, the shoulders, or both. Additional manipulations included eye closure - yes or no - and alterations in support surface - firm floor or foam mattress. B: Touch of the shoulders and, to lesser extent, the knees reduces static sway. Combined touch was most effective in reducing sway, particularly when visual and lower leg proprioceptive feedback was reduced. Data redrawn and modified from Rogers et al. (2001), with permission.

w

.....

IV

-<

~

s

d

@

Fig. 17. Example of distorted proprioceptive feedback. A: Vibration applied bilaterally over the Achilles tendons (frequency 70 Hz, amplitude 0.5 mm), without visual feedback. B: Effects on the position of the COP (pooled data of 10 young healthy subjects). The thick horizontal bar indicates the period of vibration. Achilles tendon vibration rapidly induces an illusory forward tilt of the body, which is "compensated" by shifting the COP in a posterior direction. The effects immediately disappear following cessation of vibration.

~

~

~

ttl

~

ttl

IV IV

w

323

POSTUROGRAPHY

Fig. 18. The six different test conditions of the Sensory Organization Test (Nashner et al., 1982). Condition I: all sensory inputs available. Condition 2: absent visual feedback. Condition 3: distorted visual feedback because visual surround is "center of pressure referenced" (movements are proportional to the anterior-posterior displacement of the COP). Condition 4: distorted somatosensory feedback because supporting platform is "center of pressure referenced". Condition 5: same as condition 4, but now with eyes closed. Condition 6: distorted visual and somatosensory feedback because both visual surround and supporting platform are "center of pressure referenced". Redrawn after a figure provided courtesy of NeuroCom.

lum, but this is not always the case. For example, head movements appear to precede, rather than follow, the movements of the visual surround, suggesting a "top-down" control of posture with the head as the primary target for postural correction (Di Fabio et al., 1998).

20.5.3. Postural and cognitive set

A popular area of research focuses on "postural set": manipulating the functional demands of the ongoing postural task. This rather vaguely defined term covers a widespread array of conditions

324

B.R. BLOEMET AL.

Table 6 Manipulations of postural or cognitive set. Actual changes in environmental situation • Stimulus characteristics o Direction o Size o Velocity o Acceleration o Duration

• Nature of support surface o Size o Friction • Handrails o Present yes or no o Distance to subject • Touch o Active (e.g. holding on to frame) vs. passive' o Gentle vs. stabilizing force • Physical restraints o Ankle straps o Immobilization (e.g. corset, fixing the trunk to a post) o Supporting harness Changes in environmental perception' • Amount of predictability" o Based upon experience o From prior information - Correct vs. incorrect - Complete vs. partial • Fear of falling • Specific prior instructions (e.g. yield to the perturbation) Changes in initial body position • Sitting vs. standing" • Amount of preleaning'' Multiple tasking • Secondary cognitive task • Secondary motor task • Combinations of cognitive and motor tasks a Under these circumstances, an increased feeling of security (a form of "cognitive set") contributes, possibly together with the fact that additional somatosensory feedback is provided; b changes in postural responses which merely depend upon altered perceptions of the environmental context (and not upon actual changes in the environment) could be referred to as changes in "cognitive set"; C this includes both the functional attenuation of postural reflexes if perturbations become predictable through repetition, and the development of fixed, "default" responses if the nature of upcoming perturbations is unpredictable; d under these circumstances, postural reflexes are altered not only because the functional demands are changed (for example, stabilizing reflexes are not required while sitting), but also because the amount of preloading, and therefore the amount of BGA, is altered.

325

POSTUROGRAPHY

(Table 6). Changes in postural responses which merely depend upon altered expectations or perceptions of the environmental context - and not upon actual changes in the environment - are a form of postural set which could be referred to as changes in "cognitive set". The distinction between postural and cognitive set is somewhat arbitrary, because changes in postural set often lead to concurrent changes in cognitive set. For example, allowing subjects to hold on to a manipulandum is quoted as a change in postural set as this provides an additional frame of reference (Schieppati and Nardone, 1991). However, cognitive set is also affected through an increased feeling of security - reduced risk of falls by being able to grasp for support. The opposite can be said about standing on a narrow beam or a slippery support surface (Horak and Nashner, 1986). Interesting perspectives are offered by the ability to vary the amount of prior information about upcoming perturbations; this can be complete, partial - for example only the stimulus size but not its direction or absent, and the information given can be correct or incorrect (Diener et al., 1991; Maki and Whitelaw, 1993). Leaving subjects without prior information provides opportunities to study automatic habituation of postural responses through learning processes (Nashner, 1976; Hansen et al., 1988; Bloem et al., 1998). Another possibility is to evaluate the effects of specific prior instructions, for example to either use compensatory stepping responses or not. Healthy subjects who are specifically instructed not to resist a postural perturbation show a surprising ability to almost completely suppress certain automatic postural responses (Bloem et al., 1995). Two areas have received considerable attention in recent years. One of them is fear of falling, which is a common feature among subjects with balance impairment and prior falls. Recent posturography studies led to growing recognition that this fear may cause secondary deterioration of postural performance (Maki et al., 1991; Carpenter et al., 2001). For example, assessments of stiffness - derived from COM and COP measures - revealed that subjects standing at an elevated stable forceplate resort to a stiffening strategy that is not seen while standing at a much lower height (Fig. 19). In this illustration, stiffness is appropriately seen in the anteriorposterior, but not the medial-lateral plane. The amount of threat can be experimentally varied by altering the height of the platform or the subject's

distance with respect to the edge of the platform, thus modulating the perceived threat. Joint stiffening could have certain advantages, particularly to maximize stability under relatively static, unperturbed conditions such as standing at the edge of a cliff. However, posturography studies using movable platforms showed that under more dynamic conditions, subjects pay a price for their stiffness because the loss of joint flexibility makes them fall like a pushed toy soldier (Horak et al., 1992; Allum et al., 2002). Such all-or-nothing strategies are adopted by unstable patients with very different causes for their balance disorder, apparently in an attempt to control upright stance by standing relatively motionless when unperturbed, but at the expense of inflexibility in case of sudden postural perturbations. Another interesting area focuses on the effects of multiple tasking on postural performance. In daily life, many falls are related to simultaneous execution of a balance task such as walking and another usually unrelated - task, for example talking to someone or carrying an object. Following a fascinating report that inability to walk and talk at the same time accurately predicted falls in frail elderly persons (Lundin-Olsson et al., 1997), many investigators rushed to examine why dual tasking had such deleterious effects on postural control. Posturography experiments and gait studies proved helpful in quantifying the effects of different secondary tasks (Geurts et al., 1991; Shumway-Cook et al., 1997; Morris et al., 2000; Maki et al., 2001; Yardley et al., 2001). 20.6. Safety All platforms are equipped with safety measures to prevent actual falls. In many laboratories, persons are instructed to wear a safety harness that is secured with straps to an overhead ceiling. The amount of slack in the straps permits natural body movements within safe limits. Another possibility is the use of safety bars, so subjects can grasp for support in case of an imminent fall. This feature actually offers interesting research perspectives, because protective arm movements are an important element of the normal balance repertoire. Some groups have taken this to their advantage and performed interesting experiments where the distance to the safety bars was experimentally manipulated; this provided interesting insights in the automatic and cognitive

326

B.R. BLOEM ET AL.

A

B

:c ~ .......

E

z•

III III

QJ

:E 'p V'l

Anterior-posterior

Medial-lateral

Fig. 19. Illustration of cognitive set manipulation. A: The amount of perceived postural threat ("fear") can be altered by placing subjects at either ground level or at an elevation. B: Standing on the elevated platform is associated with increased stiffness in the anterior-posterior direction, but not in the medial-lateral direction. Redrawn and modified from Carpenter et al. (2001), with permission.

regulation of protective ann responses (McIlroy and MaId, 1995; Maki and McIlroy, 1997). Some groups use a movable platform with a relatively small surface. To prevent subjects from stepping of the platform, the subject's feet can be lightly strapped into heel guides fixed to the top surface off the platform (Carpenter et al., 1999; Allum et al., 2002). An additional advantage is that these heel guides - which are adjustable in the anterior-posterior direction - can be used to ensure that the ankle joint axis is always aligned with the

axis of the rotating platform. An obvious drawback is that compensatory stepping responses cannot be investigated, and this creates a "bias" towards use of compensatory trunk and ann movements. In the few studies where actual falls were induced, the support surface consisted of a thick mattress in order to cushion the impact of the fall (Hsiao and Robinovitch, 1998). Apart from these safety measures, most research groups ensure that at least one experimenter is standing next to the subject, to provide additional support if needed.

POSTUROGRAPHY

When these safety measures are adhered to, posturography is a safe procedure (Furman et al., 1993). The authors have never seen accidental falls in their own experiments - involving hundreds of healthy subjects and patients with range of balance disorders - and are not aware of such incidents in other laboratories. Patients may be fatigued after the experiments, but this can be reduced by providing sufficient seated rests in between trials. For some patients, participation in a posturography experiment can be a painful confrontation with their physical handicaps, and this occasionally leads to emotional reactions. Nausea and anxiety reactions are rare. 20.7. Limitations Several important drawbacks - mainly to dynamic posturography - are listed in Table 7. The ecological validity of posturography remains uncertain. Subjects are tested in a high-tech laboratory setting where they are mounted with various recording devices, all this after having given prior informed consent. Hence, subjects are much "prepared" to sustain physical perturbations and are more consciously aware of their equilibrium. Furthermore, balancing on a moving support surface is unusual in daily life, except for standing in accelerating buses or trains. Some feel that this "unusualness" of moving platforms could actually be seen as advantageous, because falls themselves are unusual and unpracticed events. In this respect, a potentially greater drawback is the need to average the results of Table 7 Shortcomings of dynamic posturography. • Poor ecological validity a Use of unnatural stimuli a Need for a safety measures (harness or handrails) a Laboratory setting with prior informed consent a Averaging required to separate signal from noise a Aspecific compensatory strategies • • • •

Test-retest reliability largely unknown High intra- and intersubject variability Responsiveness largely unknown Feasibility problems a Complex instrumentation a Labor intensive data analysis a Expensive

327

multiple trials to separate true signals from the noise. Stimulus repetition leads to habituation of postural responses (Nashner, 1976; Hansen et al., 1988; Bloem et al., 1998), and this obscures a direct comparison to falls in daily life which are typically single, unexpected events. The need to employ safety measures also creates problems, as this may potentially affect postural performance in various ways. For example, the safety harness may create a feeling of safety in patients who are normally fearful of falling, thereby concealing secondary changes in postural control. In addition, the harness may supply persons with additional somatosensory feedback about the position of proximal body segments. One study found no major effects of the safety harness on dynamic postural control (Hill et al., 1994), but this issue must be examined in more detail under different experimental conditions. A further disadvantage of dynamic posturography is the general emphasis on how postural control looks, rather than functional ability (Tinetti, 1986). For example, patients with chronic disease can have a markedly impaired posture when tested in the laboratory, yet maneuver safely and effectively in daily life due to development of compensatory strategies. Such compensatory strategies markedly hamper the interpretation of findings in patients with longstanding balance disorders. Most dynamic posturography tests correlate poorly to accepted clinical tests of balance control (Bloem et al., 1998; Evans and Krebs, 1999). A general problem that hampers validity studies in this field is the lack of a golden standard for postural control. Thus, in face-to-face comparisons to medical judgment, a poor sensitivity of a particular posturography test can equally be interpreted as a false positivity in the clinical assessment. Some of the observed "differences" between patients and control populations may be artifacts of variations in initial posture. For example, many patients with balance disorders tend to adopt a stooped posture - even while seated! - and often this gradually increases over time during the experiment (Hirschfeld, 1998; Bloem et al., 1999). Such changes in baseline stance lead to different movement patterns following external perturbations, and may induce changes in pre- and post-stimulus muscle activity. Investigators should therefore not rely on EMG analyses alone, but use kinetics and kinematics to control for changes in baseline body position.

328 Other basic performance measures of posturography also remain largely unknown. For example, the test-retest reliability has rarely been examined and - if available - applies only to the specific experimental conditions and populations for which it was studied (Goldie et aI., 1989; Ishizaki et aI., 1991; Hill et aI., 1994; Ford-Smith et aI., 1995; Clark et aI., 1997; Tarantola et aI., 1997; Benvenuti et aI., 1999; Corriveau et al., 2000). Various studies reported a considerable inter- and intra-individual variability (Nashner, 1976; Geurts et aI., 1993; Takala et aI., 1997; Cobb and Nichols, 1998). Such reliability measures are required to interpret findings in patients with fluctuating clinical symptoms, and are essential if posturography is to be used as an objective outcome measure during intervention studies. Many outcome measures merely provide a "keyhole view" of postural performance. For example, measurement of sway in only the sagittal plane negates the importance of roll instability. Similarly, EMG can only be recorded from a limited number of muscles at a time, hence any inferences must be restricted to these measurements alone. Also, EMG alone provides little functional information because it is difficult to derive forces from muscular activation during dynamic contractions. An occasional problem with outcome measures is the uncertainty about data interpretation. More sway is usually equated with poorer performance, but this is not necessarily so: subjects standing with eyes shut may purposely resort to larger sway oscillations because this enhances afferent feedback about postural poorer performance from other sources, such as the vestibular system (Coma et al., 1999). A more general drawback relates to the high costs and technical complexity of dynamic posturography. This prevents more widespread clinical use in clinical settings. 20.8. Physiological and clinical utility 20.8.1 Insights into normal postural control

The greatest utility of posturography remains undoubtedly in studies of normal postural control in humans. Posturography studies have been instrumental in clarifying the role of isolated afferent feedback systems, and how different afferent signals are centrally integrated and "translated" into correc-

B.R. BLOEM ET AL.

tive balance responses. Using batteries of objective and quantitative outcome measures, posturography has helped to delineate the different types of balance corrections, ranging from purely passive factors body inertia, viscoelastic properties of stretched ligaments or muscles - to complex and preprograrnmed postural strategies. In addition, posturography studies have shed new light on the interactions between cognitive processes, voluntary intent and "automatic" postural responses. 20.8.2. Patients with balance disorders

In 1993, a specialist panel agreed that static posturography was unlikely to become an efficacious diagnostic test for patients with balance disorders (Furman et al., 1993). At the same time, this panel regarded dynamic posturography as a "promising" technique for the assessment and, possibly, the management of patients with balance disorders. A decade later, the overall utility of dynamic posturography in this field remains unconvincing, mainly because of the difficulties outlined in Section 20.7. Only few studies have produced convincing results with a potential to change the diagnostic approach or management of patients with balance disorders. One example of a practically relevant result came from a study that examined the real-life accelerations of trains and buses; when elderly persons were subsequently exposed to such linear accelerations during a posturography experiment, many lost their balance (De Graaf and Van Weperen, 1995). Such findings could influence policy makers to set a legal upper limit to accelerations in public transport vehicles. However, most studies thus far produced results that, although scientifically interesting, had little bearings on everyday clinical practice. 20.8.2.1. Selective lesion studies Perhaps the greatest utility for posturography lies in studies of patients with balance disorders caused by "selective" lesions of the central or peripheral nervous system. According to classical neurological teaching, normal functions of the nervous system can be estimated from studies on patients with selective lesions in areas that normally supply a particular function. Lesion studies could therefore provide insight into normal balance control. This approach has been applied widely and produced

329

POSTUROGRAPHY

neuropathy is often used as a model of selective proprioceptive loss, but in fact such patients often show at least some degree of paresis upon careful examination.

interesting insights into the effects of selective lesions in, for example, the vestibular system or specific parts of the basal ganglia. However, such "lesioning studies" must be interpreted with caution, particularly if chronic patients with long-lasting balance impairment are used to attribute functions to the nervous system for postural control. Various factors obscure the interpretation of balance abnormalities in any study of chronically unstable patients, in particular the ability of such persons to develop striking compensatory strategies to cope with their longstanding balance disorder. Examples include a stooped posture - to protect against backward falls - and a stiffening strategy to minimize static sway. Such compensatory mechanisms seem more or less "generic" as they are used by patients with very different causes for their balance disorder. Other confounders include secondary physical changes due to immobility, for example contractures, and the ability of the nervous system for neuronal reorganization (plasticity). In addition, the question remains how selective a particular lesion is. For example, a diabetic sensory 0.6

B

A

20.8.2.2. (Early) diagnosis Dynamic posturography can reveal clear abnormalities in groups of patients with balance impairment, as compared to control groups of healthy subjects. For this reason, dynamic posturography has been advocated as a potential screening tool for mild balance abnormalities, thus providing an objective and early marker for subjects at risk of falls. However, not one study has convincingly shown that posturography is superior to the clinical opinion of experienced physicians when patients with mild dysfunction must be detected. For example, at a group level, medium latency stretch responses to sudden toe-up rotational movements of a supporting forceplate are increased in patients with Parkinson's disease (Fig. 20A). Individual analysis reveals that only a few patients with the most advanced disease are responsible for this group difference, rendering

c

0.4 "t:l Ql

.t:!

'iij

E ...

0.2

0

.s Ql

0.0

"t:l

:::l

.'!:

Q.

E -0.2

•

r

f

III

I

c

...J

~

-0.4

-0.6

• Controls o PD

Both legs

* Right leg

I

I

•

• Control s A PD, H&Y 1 V PD, H&Y 2 PD, H&Y 2.5 C PD, H&Y 3 <> PD, H&Y 4

*

Left leg

Fig. 20. Early detection of postural abnormalities in individual patients, as illustrated for medium latency (ML) stretch responses evoked by sudden toe-up rotations of a support surface. A: There is a clear increase in ML-amplitudes for patients with Parkinson's disease as a group (N=23) compared to the control group (N=24). *=P<0.OO05. B: Individual analysis of ML-amplitudes shows that only few patients are responsible for the observed group difference. Most of these patients had advanced Parkinson's disease, indicating that ML responses cannot be used for individual analyses. Abbreviations: H&Y = Hoehn and Yahr stages; PD=Parkinson's disease. Redrawn from Bloem et al. (1992), with permission.

330

B.R. BLOEM ET AL.

this abnormality completely unsuitable for early detection purposes (Fig. 20B). Even among patients with clinically overt disease, posturography cannot be used to make a reliable diagnosis in individuals, nor can it be used to localize a specific lesion in the central nervous system. Indeed, most postural abnormalities reported thus far are not specific for a particular disease. This lack of overall individual diagnostic utility can be ascribed in part to the shortcomings of posturography - see Section 20.9. The presence of aspecific compensatory strategies is also responsible for the overlap across different types of balance disorders. For example, instability under conditions 5 and 6 of the SOT (Fig. 18) is often interpreted as evidence for vestibular loss (Nashner et al., 1982; Furman, 1994). However, most unstable patients - irrespective of the cause of their postural deficit - have greatest difficulty with these two test conditions, simply because they are most demanding. Therefore, posturography cannot be used for the differential diagnosis of balance disorders. There are perhaps two exceptions. One is primary orthostatic tremor, where power spectrum analysis of body sway during quiet stance may reveal the characteristic 16 Hz oscillation that is invisible to the naked eye (Yarrow et al., 2001). Psychogenic balance disorders represent the second exception, because bizarre performance on moving platforms cannot be reconciled with known organic patterns (Allum et al., 1996).

improve the ecological validity of dynamic posturography. Postural perturbations should be sufficiently destabilizing to test subjects around their stability limits, as this will presumably lead to better insights into actual falling mechanisms. Concurrent use of multiple outcome measures - "multimodal posturography" - is needed to fully understand postural control, and to determine which elements best separate patients from controls. Recent studies have begun to show that innovative computerassisted analyses of "ordinary" outcome measures such as the COP can lead to new fundamental insights into postural regulation. Such data can be entered into models of postural control (neural networks) which, in tum, can generate new questions for posturography experiments. A true challenge for the future will be to combine posturography techniques with imaging modalities, such as functional resonance imaging, single photon emission tomography and near-infrared spectroscopy (Bloem et al., 2003). Basic performance measures such as testretest reliability and responsiveness to change can be addressed fairly easily. Care must be taken to include carefully defined and homogenous patient groups, and to distinguish primary (disease-related) postural abnormalities from aspecific secondary compensation. Such experiments may be time-consuming, but they should eventually form a knowledge basis from which simple and inexpensive experiments can be distilled for use in daily clinical practice.

20.8.2.3. Patient management In light of the aforementioned shortcomings, it is understandable that dynamic posturography presently plays a negligible role in the management of individual patients. For example, it remains unclear what to do when abnormal posturography is an isolated finding. Advocates have suggested that dynamic posturography might be used to design rehabilitation programs for patients with balance disorders, for example in the field of physiotherapy (Furman et al., 1993). This remains to be demonstrated convincingly.

20.10. Acknowledgments

20.9. Recommendations for the future The ideal posturography tool addresses the concerns outlined in Table 7. Application of multidirectional perturbations and analyses of freely moving subjects are promising developments to

We thank Dr M.G. Carpenter and L. Oude Nijhuis for their critical comments. B.R. Bloem and J.E. Visser were supported by the Prinses Beatrix Fonds and de Stichting de Drie Lichten. J.H,J. Allum was supported by a grant from the Swiss National Research Foundation (31-59'319.99).

References Allum, JH and Budingen, HJ (1979) Coupled stretch reflexes in ankle muscles: an evaluation of the contributions of active muscle mechanisms to human posture stability. Progr. Brain Res., 50: 185-195. Allum, JH and Honegger, F (1992) A postural model of balance-correcting movement strategies. J. Vestib. Res., 2: 323-347. Allum, JH and Honegger, F (1998) Interactions between vestibular and proprioceptive inputs triggering and

POSTUROGRAPHY

modulating human balance-correcting responses differ across muscles. Exp. Brain Res., 121: 478-494. Allum, JH and Mauritz, K-H (1984) Compensation for intrinsic muscle stiffness by short-latency reflexes in human triceps surae. 1. Neurophysiol., 52: 797-818. Allum, JR, Honegger, F and Pfaltz, CR (1989) The role of stretch and vestibulo-spinal reflexes in the generation of human equilibriating reactions. Progr. Brain Res., 80: 399-409. Allum, JR, Honegger, F and Schicks, H (1993) Vestibular and proprioceptive modulation of postural synergies in normal subjects. J. Vest. Res., 3: 59-85. Allum, JH, Huwiler, M and Honegger, F (1996) Identifying cases of non-organic vertigo using dynamic posturography. Gait Posture, 4: 52-61. Allum, JH, Adkin, AL, Carpenter, MG, Held-Ziolkowska, M, Honegger, F and Pierchala, K (2001) Trunk sway measures of postural stability during clinical balance tests: effects of a unilateral vestibular deficit. Gait Posture, 14: 227-237. Allum, JH, Carpenter, MG, Honegger, F, Adkin, AL and Bloem, BR (2002) Age-dependent variations in the directional sensitivity of balance corrections and compensatory arm movements in man. J. Physiol., 542: 643-663. Aruin, AS and Latash, ML (1996) Anticipatory postural adjustments during self-initiated perturbations of different magnitude triggered by a standard motor action. Electroencephalogr. CUn. Neurophysiol., 10 1: 497-503. Baloh, RW, Spain, S, Socotch, TM, Jacobson, KM and Bell, T (1995) Posturography and balance problems in older people. J. Am. Geriatr: Soc., 43: 638-644. Beckley, DJ, Bloem, BR, Remler, MP, Roos, RA and Van Dijk, JG (1991a) Long latency postural responses are modified by cognitive set. Electroencephalogr. CUn. Neurophysiol., 81: 353-358. Beckley, DJ, Bloem, BR, Van Dijk, JG, Roos, RA and Rernler, MP (1991 b) Electrophysiological correlates of postural instability in Parkinson's disease. Electroencephalogr. CUn. Neurophysiol., 81: 263-268. Beckley, DJ, Panzer, VP, Remler, MP, lIog, LB and Bloem, BR (1995) Clinical correlates of motor performance during paced postural tasks in Parkinson's disease. J. Neurol. Sci., 132: 133-138. Begbie, GH (1967) Some problems in postural sway. In: AVS de Reuck and J Knight (Eds.), Myotatic, Kinesthetic and Vestibular Mechanisms. Ciba Foundation Symposium. JA Churchill Ltd, London, pp. 80-104. Benvenuti, F, Stanhope, SJ, Thomas, SL, Panzer, VP and Hallett, M (1997) Flexibility of anticipatory postural adjustments revealed by self-paced and reaction-time arm movements. Brain Res., 761: 59-70.

331 Benvenuti, F, Mecacci, R, Gineprari, I, Bandinelli, S, Benvenuti, E, Ferrucci, L, Baroni, A, Rabuffetti, M, Hallett, M, Dambrosia, JM and Stanhope, SJ (1999) Kinematic characteristics of standing disequilibrium: reliability and validity of a posturographic protocol. Arch. Phys. Med. Rehabil., 80: 278-287. Bisdorff, AR, Bronstein, AM, Wolsley, C, Gresty, MA, Davies, A and Young, A (1999) EMG responses to free fall in elderly subjects and akinetic rigid patients. J. Neurol. Neurosurg. Psychiatry, 66: 447-455. Bloem, BR, Beckley, DJ, Van Dijk, JG, Zwinderman, AH and Roos, RA (1992) Are medium and long latency reflexes a screening tool for early Parkinson's disease? J. Neurol. Sci., 113: 38-42. Bloem, BR, Van Dijk, JG, Beckley, DJ, Zwinderman, AH, Rernler, MP and Roos, RA (1993) Correction for the influence of background muscle activity on stretch reflex amplitudes. J. Neurosci. Methods, 46: 167-174. Bloem, BR, Beckley, DJ, Rernler, MP, Roos, RA and Van Dijk, JG (1995) Postural reflexes in Parkinson's disease during 'resist' and 'yield' tasks. J. Neurol. Sci., 129: 109-119 Bloem, BR, Beckley, DJ, Van Dijk, JG, Zwinderman, AH, Rernler, MP and Roos, RA (1996) Influence of dopaminergic medication on automatic postural responses and balance impairment in Parkinson's disease. Mov. Disord., II: 509-521. Bloem, BR, Van Vugt, JP, Beckley, DJ, Remler, MP and Roos, RA (l998a) Habituation of lower leg stretch reflexes in Parkinson's disease. Electroencephalogr. Clin. Neurophysiol., 109: 73-77. Bloem, BR, Beckley, DJ, Van Hilten, II, Roos, RA (I 998b ) Clinimetrics of postural instability in Parkinson's disease. J. Neurol., 245: 669-673 Bloem, BR, Van Dijk, JG and Beckley, DJ (1999) Are automatic postural responses in patients with Parkinson's disease abnormal due to their stooped posture? Exp. Brain Res., 124: 481-488. Bloem, BR, Allum, JH, Carpenter, MG and Honegger, F (2000) Is lower leg proprioception essential for triggering human postural responses? Exp. Brain Res., 130: 375-391. Bloem, BR, Carpenter, MG, Allum, JH and Honegger, F (2001) Trunk control and protective arm responses to multidirectional postural perturbations in Parkinson's disease. In: J Duysens, BCM Smits-Engelsman and H Kingma (Eds.), Control of Posture and Gait. Symposium of the International Society for Postural and Gait Research, Maastricht, pp. 916-919. Bloem, BR, Allum, JH, Carpenter, MG, Verschuuren, JJOM and Honegger, F (2002) Triggering of postural responses and compensatory strategies in a patient with total leg proprioceptive loss. Exp. Brain Res., 142: 91-107

332 Bloem, BR, Steijns, JAG and Smits-Engelsman, BCM (2003) An update on falls. Curr. Opin. Neurol., 16: 15-26. Brandt, T, Buchele, Wand Krafczyk, S (1986) Training effects on experimental postural instability: a model for clincial ataxia therapy. In: W Bles and T Brandt (Eds.), Disorders of Posture and Gait. New York, Elsevier, pp. 353-365. Bronstein, AM, Yardley, L, Moore, AP and Cleeves, L (1996) Visually and posturally mediated tilt illusion in Parkinson's disease and in labyrinthine defective subjects. Neurology, 47: 651-656. Carpenter, MG, Allum, JH and Honegger, F (1999) Directional sensitivity of stretch reflexes and balance corrections for normal subjects in the roll and pitch planes. Exp. Brain Res., 129: 93-113. Carpenter, MG, Frank, JS, Silcher, CP and Peysar, GW (2001) The influence of postural threat on the control of upright stance. Exp. Brain Res., 138: 210-218. Clark, S, Rose, OJ, Fujimoto, K (1997) Generalizability of the limits of stability test in the evaluation of dynamic balance among older adults. Arch. Phys. Med. Rehabil., 78: 1078-1084. Clement, G, Gurfinkel, VS, Lestienne, F, Lipshits, MI and Popov, KE (1984) Adaptation of postural control to weightlessness. Exp. Brain Res., 57: 61-72. Cobb, SVG and Nichols, SC (1998) Static posture tests for the assessment of postural instability after virtual environment use. Brain Res. Bull., 47: 459-464. Collins, JJ and De Luca, CJ (1994) Random walking during quiet standing. Phys. Rev. Lett., 73: 764-767. Commissaris, DACM, Nieuwenhuijzen, PHJA, Overeem, S, De Vos, A, Duysens, J and Bloem, BR (2002) Dynamic posturography using a new movable multidirectional platform driven by gravity. J. Neurosci. Methods, 113: 73-84. Cordo, PJ and Nashner, LM (1982) Properties of postural adjustments associated with rapid arm movements. 1. Neurophysiol., 47: 287-302. Coma, S, Tarantola, J, Nardone, A, Giordano, A and Schieppati, M (1999) Standing on a continuously moving platform: is body inertia counteracted or exploited? Exp. Brain Res., 124: 331-341. Corriveau, H, Hebert, R, Prince, F and Raiche, M (2000) Intrasession reliability of the "center of pressure minus center of mass" variable of postural control in the healthy elderly. Arch. Phys. Med. Rehabil., 81: 45-48. Cresswell, AG, Oddsson, Land Thorstensson, A (1994) The influence of sudden perturbations on trunk muscle activity and intra-abdominal pressure while standing. Exp. Brain Res., 98: 336-341. Day, BL, Steiger, MJ, Thompson, PD and Marsden, CD (1993) Effect of vision and stance width on human

B.R. BLOEM ET AL.

body motion when standing: implications for afferent control of lateral sway. J. Physiol., 469: 479-499. De Graaf, B and Van Weperen, W (1995) Met behoud Van evenwicht: experimenteel onderzoek naar de grenswaarden voor versnellingen die het menselijk lichaam kan ondergaan zonder het houdingsevenwicht te verliezen. Ned. Tijdschr. Geneeskd., 139: 377-382. De Hoon, EWJ, Allum, JH, Carpenter, MG, Salis, C, Bloem, BR, Conzelmann, M and Bischoff, HA (2003) Quantitative assessment of the "stops walking while talking test" in the elderly. Arch. Phys. Med. Rehabil., 84: 832-842. Di Fabio, RP, Emasithi, A and Paul, S (1998) Validity of visual stabilization conditions used with computerized dynamic platform posturography. Acta. Oto-Laryngol., 118: 449-454. Dichgans, J and Diener, HC (1987) The use of short- and long-latency reflex testing in leg muscles of neurological patients. In: A Struppler and A Weindl (Eds.), Clinical Aspects of Sensory Motor Integration. Springer-Verlag, Berlin, pp. 165-175. Dichgans, J, Held, R, Young, LR and Brandt, T (1972) Moving visual scenes influence the apparent direction of gravity. Science, 178: 1217-1218. Diener, He, Dichgans, J, Bruzek, W and Selinka, H (1982) Stabilization of human posture during induced oscillations of the body. Exp. Brain Res., 45: 126-132. Diener, HC, Bootz, F, Dichgans, J and Bruzek, W (1983) Variability of postural "reflexes" in humans. Exp. Brain Res., 52: 423-428. Diener, HC, Dichgans, J, Guschlbauer, Band Mau, H (1984) The significance of proprioception on postural stabilization as assessed by ischemia. Brain Res., 296: 103-109. Diener, HC, Horak, FB and Nashner, LM (1988) Influence of stimulus parameters on human postural responses. J. Neurophysiol., 59: 1888-1905. Diener, HC, Horak, FB, Stelmach, G, Guschlbauer, Band Dichgans, J (1991) Direction and amplitude precuing has no effect on automatic posture responses. Exp. Brain Res., 84: 219-223. Dietz, V, Mauritz, KH and Dichgans, J (1980) Body oscillations in balancing due to segmental stretch reflex activity. Exp. Brain Res., 40: 89-95. Dietz, V, Quintem, J and Berger, W (1984) Corrective reactions to stumbling in man: functional significance of spinal and transcortical reflexes. Neurosci. Lett., 44: 131-135. Dietz, V, Horstmann, GA, Trippel, M and Gollhofer, A (1989) Human postural reflexes and gravity and underwater simulation. Neurosci. Lett., 106: 350-355. Dietz, V, Trippel, M, Ibrahim, IK and Berger, W (1993) Human stance on a sinusoidally translating platform:

POSTUROGRAPHY

balance control by feedforward and feedback mechanisms. Exp. Brain Res., 93: 352-362. Do, MC, Breniere, Y and Bouisset, S (1988) Compensatory reactions in forward fall: are they initiated by stretch receptors? Electroencephalogr. Clin. Neurophysiol., 69: 448-452. Evans, MK and Krebs, DE (1999) Posturography does not test vestibulospinal function. Otolaryngol. Head Neck Surg., 120: 164-173. Fernie, GR, Gryfe, CI, Holliday, PJ and Llewellyn, A (1982) The relationship of postural sway in standing to the incidence of falls in geriatric subjects. Age Ageing, 11: 11-16. Fitzpatrick, RC, Taylor, JL and McCloskey, DI (1992) Ankle stiffness of standing humans in response to imperceptible perturbation: reflex and task-dependent components. J. Physiol., 454: 533-547. Ford-Smith, CD, Wyman, JF, Elswick, RK, Jr., Fernandez, T and Newton, RA (1995) Test-retest reliability of the sensory organization test in noninstitutionalized older adults. Arch. Phys. Med. Rehabil., 76: 77-81. Forssberg, H and Hirschfeld, H (1994) Postural adjustments in sitting humans following external perturbations: muscle activity and kinematics. Exp. Brain Res., 97: 515-527. Furman, 1M (1994) Posturography: uses and limitations. Bail/ieres Clin. Neurol., 3: 501-513. Furman, JM, Baloh, RW, Barin, K, Hain, TC, Herdman, SJ, Konrad, HR and Parker, SW (1993) Assessment: posturography. Report of the therapeutics and technology assessment subcommittee of the American Academy of Neurology. Neurology, 43: 1261-1264. Gatev, P, Thomas, S, Kepple, T and Hallett, M (1999) Feedforward ankle strategy of balance during quiet stance in adults. 1. Physiol., 514: 915-928. Geurts, AC, Mulder, TW, Nienhuis, Band Rijken, RA (1991) Dual-task assessment of reorganization of postural control in persons with lower limb amputation. Arch. Phys. Med. Rehabil., 72: 1059-1064. Geurts, AC, Nienhuis, B and Mulder, TW (1993) Intrasubject variability of selected force-platform parameters in the quantification of postural control. Arch. Phys. Med. Rehabil., 74: 1144-1150. Gill, J, Allum, JH, Held-Ziolkowska, M, Carpenter, MG, Honegger, F and Pierchala, K (2001) Trunk sway measures of dynamic equilibrium during clinical balance tasks: effects of age. J. Gerontol. Med. Sci., 56: M438-M447. Gilles, M, Wing, AM and Kirker, SGB (1999) Lateral balance organisation in human stance in response to a random or predictable perturbation. Exp. Brain Res., 124: 137-144. Goldberg, JM, Smith, CE and Fernandez, C (1984) Relation between discharge regularity and responses to

333 externally applied galvanic currents in vestibular nerve afferents of the squirrel monkeyJ Neurophysiol., 51: 1236-1256. Goldie, PA, Bach, TM and Evans, OM (1989) Force platform measures for evaluating postural control: reliability and validity. Arch. Phys. Med. Rehabil., 70: 510-517. Greenspan, SL, Myers, ER, Kiel, DP, Parker, RA, Hayes, WC and Resnick, NM (1998) Fall direction, bone mineral density, and function: risk factors for hip fracture in frail nursing home elderly. Am. J. Med., 104: 539-545. Grey, MJ, Ladouceur, M, Andersen, JB, Nielsen, JB and Sinkjaer, T (2001) Group II muscle afferents probably contribute to the medium latency soleus stretch reflex during walking in humans. J. Physiol., 534: 925-933. Grey, MJ, Larsen, Band Sinkjaer, T (2002) A task dependent change in the medium latency component of the soleus stretch reflex. Exp. Brain Res., 145: 316322. Gurfinkel, VS, Lipshits, MI and Popov, KE (1979) On the origin of short-latency muscle responses to postural disturbances. Agressologie, 20B: 153-154. Gurfinkel, VS, Levik, YS, Popov, KE and Sme-Tanin, BN (1988) Body scheme in the control of postural activity. In: S Gurfinkel, ME Joffe, J Massion and JP Roll (Eds.), Stance and Motion Facts and Concepts. Plenum Press, New York, pp. 185-193. Hansen, PD, Woollacott, MH and Debu, B (1988) Postural responses to changing task conditions. Exp. Brain Res., 73: 627-636. Henry, SM, Fung, J and Horak, FB (1998) EMG responses to maintain stance during multidirectional surface translations. J. Neurophysiol., 80: 1939-1950. Hill, KM, Harbum, KL, Kramer, JF, Noh, S, Vandervoort, AA and Matheson, JM (1994) Comparison of balance responses to an external perturbation test, with and without an overhead harness safety system. Gait Posture, 2: 27-31. Hirschfeld, H (1998) EMG: limitations regarding analysis of postural responses following surface perturbations. In: HJ Hermens, G Rau, C Disselhorst-Klug and B Freriks (Eds.), Surface Electromyography Application Areas and Parameters. Proceedings of the third general SENIAM workshop, Aachen, Germany, May 1998. Roessingh Research and Development, pp. 76-83. Horak, FB and Nashner, LM (1986) Central programming of postural movements: adaptation to altered support surface configurations. J. Neurophysiol., 55: 13691381. Horak, FB, Diener, HC and Nashner, LM (1989) Influence of central set on human postural responses. 1. Neurophysiol., 62: 841-853.

334 Horak, FB, Nutt, JG and Nashner, LM (1992) Postural inflexibility in parkinsonian subjects. J. Neurol. Sci., 111: 46-58. Horak, FB, Frank, JS and Nutt, JG (1996) Effects of dopamine on postural control in parkinsonian subjects: scaling, set, and tone. J. Neurophysiol., 75: 23802396. Horak, FB, Earhart, GM and Dietz, V (2001) Postural responses to combinations of head and body displacements: vestibular-somatosensory interactions. Exp. Brain Res., 141: 410-414. Horstmann, GA, Gollhofer, A and Dietz, V (1988) Reproducibility and adaptation of the EMG responses of the lower leg following perturbations of upright stance. Electroencephalogr. Clin. Neurophysiol., 70: 447-452. Hsiao, ET and Robinovitch, SN (1998) Common protective movements govern unexpected falls from standing height. J. Biomech., 31: 1-9. Ishizaki, H, Pyykko, I, Aalto, H and Starck, J (1991) Repeatability and effect of instruction of body sway. Acta. Otolaryngol., 481 (Suppl.): 589-592. Jeka, 11 and Lackner, JR (1995) The role of haptic cues from rough and slippery surfaces in human postural control. Exp. Brain Res., 103: 267-276. Jongkees, LBW and Groen, 11 (1942) De standvastigheid Van den mensch. Ned. Tijdschr. Geneeskd., 86: 14011407. Keshner, EA, Woollacott, MH and Debu, B (1988) Neck, trunk and limb muscle responses during postural perturbations in humans. Exp. Brain Res., 71: 455466. Lauk, M, Chow, CC, Lipsitz, LA, Mitchell, SL and Collins, 11 (1999) Assessing muscle stiffness from quiet stance in Parkinson's disease. Muscle Nerve, 22: 635--639. Lekhel, H, Popov, KE, Anastasopoulos, D, Bronstein, AM, Bhatia, KP, Marsden, CD and Gresty, MA (1997) Postural responses to vibration of neck muscles in patients with idiopathic torticollis. Brain, 120: 583-591. Loram, ill, Kelly, SM and Lakie, M (2001) Human balancing of an inverted pendulum: is sway size controlled by ankle impedance? J. Physiol., 532: 879-891. Lord, SR, Clark, RD and Webster, IW (1991) Physiological factors associated with falls in an elderly population. J. Am. Geriatr. Soc., 39: 1194-1200. Lord, SR, Rogers, MW, Howland, A and Fitzpatrick, RC (1999) Lateral stability, sensorimotor function and falls in older people. J. Am. Geriatr. Soc., 47: 1077-1081. Lund, S and Broberg, C (1983) Effects of different head positions on postural sway in man induced by a

B.R. BLOEM ET AL.

reproducible vestibular error signal. Acta. Physiol. Scand., 117: 307-309. Lundin-Olsson, L, Nyberg, L and Gustafson, Y (1997) "Stops walking when talking" as a predictor of falls in elderly people. Lancet, 349: 617. Maki, BE and McIlroy, WE (1997) The role of limb movements in maintaining upright stance: the 'changein-support' strategy. Phys. Ther., 77: 488-507. Maki, BE and Whitelaw, RS (1993) Influence of expectation and arousal on center-of-pressure responses to transient postural perturbations. 1. Vest. Res., 3: 2539. Maki, BE, Holliday, PJ and Fernie, GR (1990) Aging and postural control. A comparison of spontaneous- and induced-sway balance tests. J. Am. Geriatr. Soc., 38: 1-9. Maki, BE, Holliday, PJ and Topper, AK (1991) Fear of falling and postural performance in the elderly. J. Gerontol., 45: MI23-M131. Maki, BE, McIlroy, WE and Perry, SD (1996) Influence of lateral destabilization on compensatory stepping responses. J. Biomech., 29: 343-353. Maki, BE, Zecevic, A, Bateni, H, Kirshenbaum, Nand McIlroy, WE (2001) Cognitive demands of executing postural reactions: does aging impede attention switching? NeuroReport, 12: 3583-3587. Martin, JP (1965) Tilting reactions and disorders of the basal ganglia. Brain, 88: 855-874. Matjacic, Z (2000) A multi-purpose rehabilitation frame: an apparatus for experimental investigations of human balance and postural control. J. Med. Eng. Technol., 24: 250-254. McDonagh, MJ and Duncan, A (2002) Interaction of preprogrammed control and natural stretch reflexes in human landing movements. J. Physiol., 544: 985-994. McIlroy, WE and Maki, BE (1995) Early activation of arm muscles follows external perturbations of upright stance. Neurosci. Lett., 184: 177-180. Mitchell, SL, Collins, 11, De Luca, CJ, Burrows, A and Lipsitz, LA (1995) Open-loop and closed-loop postural control mechanisms in Parkinson's disease: increased mediolateral activity during quiet standing. Neurosci. Lett., 197: 133-136. Moe-Nilssen, R (1998) A new method for evaluating motor control in gait under real-life environmental conditions. Part 1: The instrument. Clin. Biomech., 13: 320-327. Morris, ME, Iansek, R, Smithson, F and Huxham, F (2000) Postural instability in Parkinson's disease: a comparison with and without a concurrent task. Gait Posture, 12: 205-216. Nashner, LM (1976) Adapting reflexes controlling the human posture. Exp. Brain Res., 26: 59-72.

POSTUROGRAPHY

Nashner, LM (1982) Adaptation of human movement to altered environments. Trends Neurosci., 5: 358-361. Nashner, LM and McCollum, G (1985) The organization of human postural movements: a formal basis and experimental synthesis. Behav. Brain Sci., 8: 135-172. Nashner, LM, Black, FO and Wall, C (1982) Adaptation to altered support and visual conditions during stance: patients with vestibular deficits. 1. Neurosci., 2: 536-544. Nevitt, MC and Cummings, SR (1993) Type of fall and risk of hip and wrist fractures: the study of osteoporotic fractures. The Study of Osteoporotic Fractures Research Group. J. Am. Geriatr. Soc., 41: 1226-1234. Nieuwboer, A, Feys, P, De Weerdt, Wand Dom, R (1997) Is using a cue the clue to the treatment of freezing in Parkinson's disease? Physiother. Res. Int., 2: 125-132. Ondo, WG, Warrior, D, Overby, A, Calmes, J, Hendersen, N, Olson, S and Jankovic, J (2000) Computerized posturography analysis of progressive supranuclear palsy: a case-control comparison with Parkinson's disease and healthy controls. Arch. Neurol., 57: 1464-1469. Overstall, PW, Exton-Smith, AN, Imms, FJ and Johnson, AL (1977) Falls in the elderly related to postural imbalance. Br. Med. J., I: 261-264. Owen, N, Leadbetter, AG and Yardley, L (1998) Relationship between postural control and motion sickness in healthy subjects. Brain Res. Bull., 47: 471-474. Panzer, VP, Bandinelli, S and Hallett, M (1995) Biomechanical assessment of quiet standing and changes associated with aging. Arch. Phys. Med. Rehabil., 76: 151-157. Paulignan, Y, Dufosse, M, Hugon, M and Massion, J (1989) Acquisition of co-ordination between posture and movement in a bimanual task. Exp. Brain Res., 77: 337-348. Perennou, DA, Amblard, B, Laassel, EM and Pelissier, J (1997) Hemispheric asymmetry in the visual contribution to postural control in healthy adults. NeuroReport, 8: 3137-3141. Peterka, RJ and Black, FO (1990) Age-related changes in human posture control: sensory organization tests. 1. Vest. Res., I: 73-85. Pyykko, I, Enbom, H, Magnusson, M and Schalen, L (1991) Effect of proprioceptor stimulation on postural stability in patients with peripheral or central vestibular lesion. Acta. Oto-Laryngol., Ill: 27-35. Rademaker, GGJ and Garcin, R (1932) Rev. Neurol., 39: 637. Rietdyk, S, Patla, AE, Winter, DA, Ishac, MG and Little, CE (1999) Balance recovery from medio-Iateral perturbations of the upper body during standing. J. Biomech., 32: 1149-1158.

335 Riley, MA, Wong, S, Mitra, S and Turvey, MT (1997) Common effects of touch and vision on postural parameters. Exp. Brain Res., 117: 165-170. Rogers, MW, Wardman, DL, Lord, SR and Fitzpatrick, RC (2001) Passive tactile sensory input improves stability during standing. Exp. Brain Res., 136: 514-522. Schieppati, M and Nardone, A (1991) Free and supported stance in Parkinson's disease. Brain, 114: 1227-1244. Schieppati, M and Nardone, A (1997) Medium-latency stretch reflexes of foot and leg muscles analysed by cooling the lower limb in standing humans. J. Physiol., 503: 691-698. Schieppati, M, Hugon, M, Grasso, M, Nardone, A and Galante, M (1994) The limits of equilibrium in young and elderly normal subjects and in parkinsonians. Electroencephalogr. Clin. Neurophysiol., 93: 286-298. Shumway-Cook, A, Woollacott, MH, Kerns, KA and Baldwin, M (1997) The effects of two types of cognitive tasks on postural stability in older adults with and without a history of falls. J. Gerontol. Med. Sci., 52A: M232-M240. Speers, RA, Paloski, WH and Kuo, AD (1998) Multivariate changes in coordination of postural control following spaceflight. J. Biomech., 31: 883-889. Takala, EP, Korhonen, I and Viikari-Juntura, E (1997) Postural sway and stepping response among working population: reproducibility, long-term stability, and associations with symptoms of the low back. Clin. Biomech., 12: 429-437. Tarantola, J, Nardone, A, Tacchini, E and Schieppati, M (1997) Human stance stability improves with the repetition of the task: effect of foot position and visual condition. Neurosci. Lett., 228: 75-78. Tinetti, ME (1986) Performance-oriented assessment of mobility problems in elderly patients. J. Am. Geriatr. Soc., 34: 119-126. Tossavainen, T, Juhola, M, Aalto, H, Toppila, E, Pyykko, I, Honkavaara, P, Laurikkala, J and Laakso, J (200 I) Postural control as assessed with virtual reality. Acta. Otolaryngol., 545 (Suppl): 53-56. Winter, DA (1990) Biomechanics and motor control of human movement. John Wiley & Sons, Inc., New York. Winter, DA, Patla, AE, Prince, F, Ishac, M and GieloPerczak, K (1998) Stiffness control of balance in quiet standing. J. Neurophysiol., 80: 1211-1221. Wolfson, LI, Whipple, R, Amerman, P and Kleinberg, A (1986) Stressing the postural response. A quantitative method for testing balance. J. Am. Geriatr. Soc., 34: 845-850. Wolfson LI, Judge 10, Whipple R, King M (1995) Strength is a major factor in balance, gait, and the

336 occurrence of falls. J. Gerontal. Med. Sci., 50 Spec. No.: 64-67 Yardley, L, Gardner, M, Bronstein, A, Davies, R, Buckwell, D and Luxon, L (2001) Interference between postural control and mental task performance in

B.R. BLOEM ET AL.

patients with vestibular disorder and healthy controls. J. Neural. Neurosurg. Psychiatry, 71: 48-52. Yarrow, K, Brown, P, Gresty, MA and Bronstein, AM (2001) Force platform recordings in the diagnosis of primary orthostatic tremor. Gait Posture, 13: 27-34.

Movement Disorders Handbook of Clinical Neurophysiology, Vol. 1 M. Hallett (Ed.) © 2003 Elsevier B.V. All rights reserved

337 CHAPTER 21

Gait analysis Kenton R. Kaufman * Biomechanics Laboratory. Mayo Clinic/Mayo Foundation, Rochester, MN 55905. USA

21.1. Introduction

21.2. Historical perspective

Every health professional must evaluate and treat the patient based on data and assumptions about the relative function of the central nervous system, selective motor control, weak or spastic muscles, joint contracture and bony deformities. The usual methods for clinical assessment include physical examination, manual muscle testing and visual observation. Most of this data is qualitative and subjective. Medical professionals apply an evolving knowledge base. A treatment plan is formed based on algorithms passed on by others and modified by experience gained from personal observations. Clinical tests are useful for diagnosis but are often not sensitive enough or objective for quantitative evaluation of treatment. However, many centers now have laboratories and skilled personnel to perform objective measurements rather than rely upon intuition and assumptions. Motion analysis laboratories provide quantitative information which can be used as a basis for appropriate therapeutic intervention as well as to objectively evaluate the effectiveness and efficacy of treatment methods. The focus of this chapter is on the methods for objective assessment of persons with movement impairment. The chapter will provide a brief overview of the gait cycle, a discussion of available measurement technology, and application of this technology for treatment planning and assessment.

Motion analysis can be described as the objective, systematic analysis of movement. The concept of depicting and recording human motion began during the Renaissance Period. Giovanni Alphonso Borelli, a student of Galileo, was among the first scientists to analyze motion while developing his theory of muscle action based upon mechanical principles (Borelli, 1685). Duchenne conducted the first scientific systematic evaluation of muscle function. His findings were published in the monumental work, Physiologie des Mouvements, published in 1857 (Basmajian and De Luca, 1985). Eadweard Muybridge first performed photographic recording of human motion. Muybridge was asked to settle a bet by Governor Leeland Stanford of California regarding whether a trotting horse had all four feet off the ground at any instant in time. Muybridge placed cameras at regular intervals along a race track. Thin threads stretched across the track triggered the shutters. The horse's hooves triggered cameras in order, and a series of photographs clearly depicted the gait sequence. Muybridge subsequently compiled a detailed photographic atlas of human and animal locomotion, which was published in three volumes (Muybridge, 1979 (original work published in 1887)). At the tum of the 20th century in Germany, Braune and Fischer became interested in measuring the motion of human body segments. They placed Geissler tubes, containing a rarified nitrogen gas, on various limb segments of a human subject dressed in black. Electrical circuits connected to the tubes created incandescence and the illuminated tubes were recorded by cameras as the subject walked. Experiments were carried out at night because there was no means to darken the room in which studies were performed. It took 10 to 12 hours to put this apparatus on the subject, whereas data collection

* Correspondence to: Dr. Kenton R. Kaufman, Ph.D., P.E., Biomechanics Laboratory, Charlton North L-llON, Mayo Clinic, 200 First Street SW, Rochester, MN 55905, USA. E-mail address: [email protected] Tel.: + I (507) 284-2262; fax: + I (507) 266-2227.

338

was completed in minutes using four cameras. The images were digitized using a precision optical device. Coordinate geometry was used to extract three-dimensional coordinates. Equations needed to calculate resultant forces and moments at the joints of a 12-segment rigid body model were formulated. Although their quantitative results were published in 1895 (Braune and Fischer, 1895), their findings are still valid today. Inman and colleagues combined rudimentary motion recordings with electromyography (EMG) in the mid-1900s. Their pioneering work in limb prosthetic research laid the foundation for modem gait analysis. Their text, Human Walking, was published in 1981 and represents the seminal text in the field (Inman et aI., 1981). Since this pioneering work, much effort has been put into developing the needed technology for human movement analysis. Automated movement tracking systems have replaced hand digitization. Advances in the aerospace industry have been utilized for the development of force plates for kinetic analysis. Computerized electromyography systems have replaced hand palpation. Currently, the technology and knowledge for gait analysis has advanced to a level that permits rapid analysis. 21.3. Gait cycle Human walking is a unique form of movement that is bipedal. While there are animal examples of bipedal locomotion (bears, primates, marsupials), the bipedal gait in man is uniquely efficient and functional. During gait, a repetitive cyclic activity of limb motion is used to move the body forward while simultaneously maintaining stance stability. Movement of the body from one point to another is accomplished in a smooth, well-coordinated fashion by the reciprocal activity of two multi-segmented lower limbs and the total body mass. Lettre and Contini (Lettre and Contini, 1967) have described locomotion in three distinct stages: (1) development phase (from rest to constant velocity); (2) rhythmic phase (constant forward velocity); and (3) decay phase (coming back to rest) (Fig. 1). Most descriptions of gait have concentrated on the rhythmic stage (Stage II) of free-speed walking. Observations have confirmed that the rhythmic stage of human walking is very consistent and is directly related to the

K.R. KAUFMAN

I..

1

!

!

Developmenl phc~e

~I
Rhylhmk ~I

phau!!

i

..

O-ecay

phase

I

Speed

Fig. 1. Three phases of walking. (From: Sutherland, Kaufman, Moitoza, "Kinematics of Normal Human Walking", Chapter 2, Human Walking Eds. Rose and Gamble, Williams and Wilkins, 1994.)

optimal energy efficiency for an individual. The patterns of movement during free-speed walking are remarkably consistent between individuals and have been given a common nomenclature defined as the events and phases of the gait cycle (Sutherland and Cooper, 1981; Perry, 1992). This series of events is repeated by each limb in a reciprocal manner as the person moves from one point to another. Gait can be defined in terms of an interval of time during which one sequence of regularly recurring events occur (Fig. 2). During free-speed (self selected) walking this cycle of repeated events is called the gait cycle. Each gait cycle can be divided into two phases: stance and swing. The stance phase is defined by the period of time when the foot is in contact with the ground. The swing phase is the time when the foot is in the air for limb advancement. The two phases (stance and swing) can be described in terms of two repeated events: foot strike and toe off. The gait cycle starts when contact is first made with the ground. This instant of time can be called foot strike, heel strike, or initial contact. This mix of terminology is used because initial contact may not occur in the heel in patients with pathological gait patterns. As the foot leaves the ground at the end of stance phase, it is usual for the toe to be the last point of contact and this time is called toe off. However, just like foot strike, the toe may not be the last portion of the foot to leave the ground, so the term foot off may also be used. Since there are two extremities during normal locomotion, there are four events: foot strike (FS), opposite toe off (OTO), opposite foot strike (OFS), and toe off (TO). The entire cycle then repeats itself with the second foot strike.

339

GAIT ANALYSIS

Fig. 2. Typical normal walk cycle illustrating the events of gait. (From: Sutherland, Kaufman and Moitoza, "Kinematics of Normal Human Walking", Chapter 2, Human Walking, Eds. Rose and Gamble, Williams and Wilkins, 1994.)

The major subdivisions of the gait cycle (Table 1 and Fig. 2) describe the transitions that have to occur while the body's center of mass moves over the oscillating limbs. Stance phase is commonly divided into three periods: (1) initial double support (foot strike to opposite toe off); (2) single limb support (opposite toe off to opposite foot strike); and (3)

second double limb support (opposite foot strike to toe off). Swing phase can also be divided into three periods: (1) initial swing (toe off to foot clearance); (2) midswing (foot clearance to tibia vertical); and (3) terminal swing (limb deceleration). Initial double support (Table 2) is characterized by rapid loading onto the forward limb with shock absorption and

Table 1 Gait cycle: events, periods, and phases. Event Foot strike

% Gait cycle

Period

Phase

a Initial double limb support

Opposite toe-off

12

Opposite foot strike

50

Toe off

62

Foot clearance

75

Tibia vertical

85

Single limb support

Stance, 62% of cycle

Second double limb support Initial swing Mid Swing

Swing, 38% of cycle

Terminal swing Second foot strike

100

From: Sutherland, Kaufman and Moitoza, "Kinematics of Normal Human Walking", Chapter 2, Human Walking, Eds. Rose and Gamble, Williams and Wilkins, 1994.

340

K.R. KAUFMAN

Table 2 Gait cycle: periods and function. Period 1. Initial double limb support

% Cycle

0-12

Function

Contralateral limb

Loading, weight transfer

Unloading and preparing for swing (pres wing)

2. Single limb support

12-50

Support of entire body weight; center of mass moving forward

Swing

3. Second double limb support

50-62

Unloading and preparing for swing (preswing)

Loading, weight transfer

4. Initial swing

62-75 75-85 85-100

5. Mid swing 6. Terminal swing

Foot clearance

Single limb support

Limb advances in front of body

Single limb support

Limb deceleration, preparation for weight transfer

Single limb support

From: Sutherland, Kaufman and Moitoza, "Kinematics of Normal Human Walking", Chapter 2, Human Walking, Eds. Rose and Gamble, Williams and Wilkins, 1994.

slowing of the body's forward momentum. The foot usually progresses to foot flat and the knee acts as a shock absorber. After opposite toe off, the opposite limb is in swing and the weight bearing limb is in single limb stance. As the body passes over the fixed foot, the center of mass rises to its peak while both forward and vertical velocity decrease. During the single limb support interval the body's entire weight is resting on one extremity. The duration of single limb stance is the best index of the limb's support capability. During this time the anterior/posterior ground reaction force reverses from forward shear to aft shear, the center of mass falls, and the forward and vertical velocity increase. This transition from forward to aft shear occurs around 30% of the cycle, which is midstance in normal subjects. It is very difficult to determine this transition point precisely in the walking cycle without laboratory measurements. Once this peak in the elevation of the center of mass is achieved, the center of mass falls until the end of single limb stance at opposite foot strike (50% of the gait cycle). Second double limb support begins with floor contact by the other foot (opposite foot strike) and continues until the original stance limb is lifted for swing (toe off). This period of time can also be defined as preswing. As weight is transferred rapidly to the forward limb, the trailing limb is beginning its preparation to swing in front of the body.

The swing phase can be split into three phases. The subdivision of the swing phase can best be understood by comparing the leg to a compound pendulum. A compound pendulum can change its swing period by changing its configuration in space. The first phase involves the foot leaving the ground and accelerating forward. This is called initial swing (toe off to foot clearance). The critical event of foot clearance occurs around 75% of the gait cycle when a swinging limb passes the standing limb. As the swing limb passes the stance limb, this period is called mid swing (foot clearance the vertical tibia). Vertical tibia occurs when the tibia becomes vertical to the floor heralding the beginning of limb deceleration. Thereafter, the limb undergoes deceleration and the foot is controlled to get ready for the next foot strike. This final phase is called terminal swing (vertical tibia to foot strike). Although swing phase does not involve the large forces associated with ground contact, two important goals must be achieved for foot clearance during swing. First, the limb must be shortened in length so that the foot can clear the ground during the forward movement. This is accomplished by flexing the hip and knee and dorsiflexing the ankle. Secondly, during the swing phase, the masses of the lower limb are subjected to accelerations and decelerations which require muscular efforts at the joints. As the swing phase is shortened with increased walking velocity, the

341

GAlT ANALYSIS

Fig. 3. Important temporal distance parameters defined and measured in gait analysis.

acceleration becomes larger and this requires greater demands from the musculature. The gait cycle is defined in terms of percentages, rather than elapsed time, since the events occur in a repeatable sequence and are independent of time (Table 1). Therefore, initial foot strike is defined as 0% and the second foot strike completes the gait cycle at 100%. In normal subjects, the opposite limb repeats the same sequence of events but is 1800 out of phase so opposite foot strike occurs at 50% of the gait cycle. The average gait cycle consists of 62% stance phase and 38% swing phase. Stance phase can be further broken down into two subdivisions of double limb support, each lasting about 12% of the gait cycle. Therefore, timing for the stance phase consists of 12% for initial double-limb support, 38% for single limb stance, and 12% for second double limb support. Note that single limb stance for one limb occurs at the same time as swing in the opposite limb. Terminology has been developed to describe the linear measurements of the gait cycle (Fig. 3). Cadence is the number of steps taken in a standard time frame. The usual units are steps per minute. Stride length is defined by the distance (in ems) between two successive foot strikes of the same foot.

There are two steps in each stride (or gait cycle). Step length is defined by the distance between the same point on each foot (usually the heel) during double limb support. Right step length is the distance of the step when supported by the right leg. Similarly, left step length is the step distance while on the left limb. Walking speed or walking velocity is the average distance covered by the whole body in a given time in a particular direction (expressed in ems per second or meters per minute). The instantaneous velocity varies from moment to moment during the gait cycle, but the average velocity is a combination of the cadence and the stride length. The step width is the side to side distance between the two feet, usually measured at midpoint of the heel but sometimes at the center of the ankle joint. The step width also defines the walking base of support.

21.4. Equipment and methods 21.4.1. Observation The simplest form of gait analysis is observational gait analysis. A systematic approach for observational gait analysis was developed at the Rancho Los

342

Amigos Medical Center in Downey, California (Perry, 1992). An experienced observer can detect many gait deviations during both stance and swing phases. However, an obvious limitation of observation in gait analysis is the difficulty of observing multiple events and multiple body segments interacting concurrently. Further, it is not possible to visualize the location of force vectors in space or electromyographic activity of muscles. Events happening faster than 1/12 of a second (83 ms) cannot be perceived by the human eye (Gage and Ounpuu, 1989). More consistent observations are obtained when motion videotapes are reviewed in slow motion (Krebs et al., 1985). Three expert observers rated video footage of fifteen children who had lower limb disability and wore braces (Krebs et al., 1985). Pearson's Correlation coefficient was 0.6 within observers and less between observers. Thus, observational gait analysis is a convenient, but only moderately reliable technique. Saleh and Murdoch (Saleh and Murdoch, 1985) utilized experienced observers to study the gait of transtibial amputees. The prosthetic limbs of the amputees were intentionally misaligned in the sagittal plane. The agreement of experienced observers with a biomechanical model was 22%. In a similar study, 54 licensed physical therapists with varying amounts of clinical experience rated three patients with rheumatoid arthritis (Eastlack et al., 1991). Generalized Kappa coefficients ranged from 0.11 to 0.52 indicating that clinician assessments are only slightly to moderately reliable. Thus, it is easy to see that limitations in observational gait analysis can lead to misinterpretation of the patient's locomotion capabilities. Hence, it is important to utilize advances in gait analysis techniques to more precisely quantify the patient's functional status. Extensive instrumentation has been developed for recording the various parameters used to describe gait.

21.4.2. Movement measurement In the biomechanical analysis of motion, skeletal segments are studied as rigid links moving through space. These rigid links are assumed to be interconnected through a series of frictionless joints. Measurement systems that are aimed at capturing the spatial trajectories of body segments usually involve a camera system or an electromagnetic system that tracks a series of body-fixed markers.

K.R. KAUFMAN

With a camera, either passively reflective or actively illuminated markers are used (Fig. 4a). These markers are commonly attached to the subjects as either discrete points or rigid clusters with multiple markers on each cluster. Placement of these external markers on the surface of the body segments are aligned with particular bony landmarks. Using stereophotogrammetric principles, the planar projections of markers viewed by each camera are used to reconstruct the 3-dimensional instantaneous position of the markers relative to an inertially fixed laboratory coordinate system. If the position of at least three non-colinear points fixed to the body segment can be obtained (and the body segment is assumed to be rigid) then the six degrees-of-freedom associated with the position and orientation of each segment can be obtained. Initially, a body-fixed coordinate system is computed for each body segment (Fig. 4b). For instance, consider the markers on the shank at an instant in time. A vector, STZ' can be formed from the lateral malleolus (B) to the lateral knee marker (A). Another vector can be formed from the lateral malleolus to the marker on the shank wand (C). The vector cross product of these two vectors is a vector STX, that is perpendicular to the plane containing all three markers. The unit vector, STY' may be determined as the vector cross-product of STZ and STX' Thus, the vectors STX' Sn" and STZ form an orthogonal body fixed coordinate system, called a technical coordinate system. In a similar manner, the marker based, or technical coordinate system may be calculated for the thigh, i.e, TTX, TTY' and TTZ' Once the position of adjacent limb segments has been determined, it is possible to determine the relative angle between adjacent limb segments in three dimensions. This assumes that the technical coordinate systems reasonably approximate the anatomical axes of the body segments, e.g. TTZ approximates along axis of the thigh and STZ approximates the long axis of the shank. A more rigorous approach adapts a subject calibration procedure to relate the technical coordinate systems with pertinent anatomical landmarks (Cappozzo et al., 1995). Additional data can be collected that relates the technical coordinate system to the underlying anatomical coordinate system. The subject calibration is performed as a static trial with the subject standing. Additional markers are typically added to the medial femoral condyle and the medial maleolli during the static

343

GAIT ANALYSIS

(a)

x (b)

Fig. 4. Body-fixed reflective markers used for establishing anatomical coordinate systems. (a) Video camera motion measurement systems calculate the location of external markers placed on the body segments and aligned with specific bony landmarks. (b) A body-fixed external coordinate system is then computed from three or more markers on each body segment. (c) Subsequently, a subject calibration relates the external coordinate system with an anatomical coordinate system through the identification of anatomical landmarks, e.g. the medial and lateral femoral condyles and medial lateral malleoli.

calibration trial. These markers serve as anatomical references for the knee axis and ankle axis. The hip center location is estimated from markers placed on the pelvis (Bell et al., 1989). The technical coordinate system is then transformed into alignment with the anatomical coordinate system for each limb segment, e.g. SAX' SAY' SAZ (Fig. 4c). The marker system is coupled to a biomechanical model (Kadaba et al. 1990; Kaufman et al., 1995). Once the position of adjacent limb segments has been determined (and each body segment is assumed to be

rigid), it is possible to determine the relative angle between adjacent limb segments in 3-dimensions. The Euler system is the most commonly used method for describing 3-dimensional motion (Chao, 1980; Grood and Suntay, 1983) (Fig. 5). Recently, advances have been made in the animation industry that will have future applications in the biomechanics field. Computerized motion analysis systems have been developed that provide marker trajectory data in real time. These systems are now available for applications in biomechanics. These

344

K.R. KAUFMAN

Determination of Joint Motion

Fig. 5. Description of knee joint motion using Eulerian angle system. An axis fixed to the distal femur defines flexion! extension motion, <1>. An axis fixed to proximal tibia along its anatomical axis defines internal-external rotation, '1'. A floating axis is orthogonal to the other two axes and used to measure abduction-adduction, (3 (Reproduced with permission from Chao, 1980).

applications make it possible to obtain results of gait analysis studies much faster and will make gait analysis much more clinically available. The most recent development to be used for quantifying human motion is an electromagnetic tracking system (An et aI., 1988). Electromagnetic systems detect the motion of sensors placed on each segment using an electromagnetic field. A three-axis magnet dipole source and a three-axis magnetic sensor are used (Raab et al., 1979) (Fig. 6). The excitation of the source and the resulting sensor output are represented as vectors. The source excitation pattern is composed of three sequential excitation states, each of which produces an excitation vector that is linearly independent of the other two. The sensor is connected to a system controller through a cable. The sensor outputs are preamplified, multiplexed and transmitted to a system electronics unit. The resultant set of three sensor output vectors contains information sufficient to determine both the position and orientation of the sensor relative to the source. Thus, these systems can provide real-time six degree-of-freedom movement

data. The use of this equipment is growing in areas of human motion analysis. The instrumentation is simple to use and is insensitive to limb interference. The limitations are the cabling necessary to connect the sensors, the sampling frequency, and the sensitivity to magnetic interference from nearby ferromagnetic metallic structures such as a total joint replacement. Nonetheless, as these electromagnetic system capabilities increase, it is expected that these devices will be used more frequently for movement analysis. 21.4.3. Force measurement

Gait analysis is also concerned with the forces that cause the observed movement and the assessment of their effect on locomotion (kinetic analysis). Forces acting on the human body can be divided into internal and external forces. The external forces represent all physical interactions between the body and the environment. These forces include gravitational, ground reaction forces and inertial forces. The internal forces are those transmitted by body

345

GAIT ANALYSIS

Electromagnetic Tracking Technology

Position and Orientation Measurements

Fig. 6. System block diagram of an electromagnetic tracking system. The 3-axis magnetic source emits three sequential excitation states that are picked up by the 3-axis magnetic sensor. The resultant set of sensor excitation vectors is used to calculate the position and orientation of the sensor relative to the source.

tissues that include muscular forces, ligament forces, and forces transmitted through joint contact. Generally, only the ground reaction forces can be measured using a force plate. Current force plates typically use strain gauge or piezoelectric transducers. Force platforms can be used to define the magnitude and direction of the resultant ground reaction force (GRF) applied to the foot by the ground (Fig. 7). The GRF vector is three dimensional and consists of a vertical component plus two shear components acting along the force plate surface. The shear forces are applied parallel to the ground and require friction. These shear forces are usually resolved in the anterior-posterior and mediallateral directions. An additional variable, the center of pressure, is needed to define the location of this GRF vector. The center of pressure is defined as the point about which the distributed force has zero moment when applied to the foot. It is found by determining the line of action of the forces measured by the platform and calculating where that line intersects the surface of the force platform. This force data is combined with kinematic data using Newton's Second Law to calculate the intersegmental forces and moments causing motion (Fig. 8). The process of proceeding from known kinematic

data and external forces to obtain intersegmental joint forces and moments is called the inverse dynamics approach (Chao, 1971). The gravitational forces acting on each body segment can be determined from the relevant mass and location of the center of mass for each segment. These quantities can be calculated together with the segmental mass moments of inertia using prediction techniques from anthropometric dimensions. The inertial forces can be obtained from calculations of angular and linear position, as well as velocity and acceleration of the body segments with respect to either a fixed laboratory coordinate system or referenced to another body segment using kinematic data. This information can be combined to solve the inverse dynamics problem (Fig. 9). Joint power can also be calculated (Winter, 1990). These data provide understanding of the subtle musculoskeletal adaptations which are utilized by patients to maintain dynamic balance during gait. Kinetic data are available at the hip, knee and ankle joint. When the position of this force line with respect to joint center has been established by combining force and movement data, the extrinsic joint moment, which is the product of lever arm and the ground reaction force, plus gravity and inertia can be calculated. This moment is of

346

K.R. KAUFMAN

Fig. 7. A force plate is used to measure the location and magnitude of the ground reaction force. Transducers are located in the four corners of the plate. The ground reaction force is divided into three force (Fxe F)' F,) and three moment (M.. M)' Mz) components. F, and F, are shear forces. F, is the vertical force. Some force plates only measure the moment around the vertical axis, i.e. M, This assumes that no tensile forces are imposed on the force plate, i.e. the foot does not stick to the plate. Under this assumption, the other moments are zero.

mg

T~

F CP1035086B-4

Fig. 8. The joint dynamics, which includes the intersegmental forces and moments, is computed through the use of Newtonian mechanics. The computation accounts for the external loads applied at the foot, e.g. the ground reaction forces, F and T, the weight of the limb segment, mg, and the inertial loads, rna and R., in order to calculate the intersegmental force, F, and moment, M.

great importance because in the case of lower extremity muscles acting during load bearing, it determines the requirements for intrinsic (muscle) force. For example, when the force line falls behind the knee joint center, quadriceps muscle action is required to prevent knee collapse, and when the force line falls in front of the knee, extensor muscle force is not needed. At each joint, a state of equilibrium exists where the external joint forces are balanced by the internal joint forces. The measurement of internal forces requires sophisticated techniques which are invasive. Analytical procedures have been developed for estimating internal joint forces. These analytical approaches use classical mechanics and mathematical optimization routines (Kaufman et al., 1991). These analytical approaches require the use of simplifying assumptions about the mechanical structure and knowledge of muscle physiology principles. Thus, the accuracy of the analytical predictions depends not only on the quality of the input data but also on the validity of the assumptions. In general, it is necessary to evaluate the estimated quantities by comparing them with experimental observations. Typically, electromyographic data are obtained to provide information regarding muscle activation patterns.

347

GAIT ANALYSIS

INVERSE DYNAMIC PROBLEM Measured Displacement x y z

¢ B l/f

d dt

Velocity

Acceleration

d

~

~

~

X

Y

dt

z

¢ iJ if!

~

x ji Z ¢ (j Iii

Equations of Motion:

Lf=rnr

c

LAfc =iJc -'-"d'('ai)-~'a[:¢~" dt 8qi

........

8qi

Fig. 9. Solution process for inverse dynamics problem. Displacement information must be differentiated twice to yield acceleration. Either Newtonian or Lagrangian formulations can be used to formulate the equations of motion.

21.4.4. Electromyog raphy

Neuromuscular coordination is required to adjust the varying muscular and ligamentous forces interacting with the abundant degrees of freedom in the joints and other parts of passive locomotive system to obtain dynamic balance during gait. Electromyographic data are useful to provide information about the timing of muscle activity and the relative intensity of muscle activity. Both surface and finewire electrodes have been used for gait kinesiological electromyographic (EMG) analysis. Each type of electrode has its advantages and disadvantages. Surface electrodes are convenient, easy to apply to the skin, and do not cause pain, irritation, or discomfort to the subject. However, they pick up signals from other active muscles in the general area of application. This feature makes surface electrodes the ideal choice for analysis of global activity in superficial muscles or muscle groups. However, surface electrodes are sensitive to movement of the skin under the electrodes and have poor specificity. They are influenced by significant muscle "crosstalk," in which the electrode signals of one muscle interfere with the signals from another (Zuniga and Simons, 1969). Thus, the activity of adjacent muscle groups can interfere and lead to false results.

However, a double differential technique has been shown to reduce cross-talk in surface EMGs (Koh and Grabiner, 1992). The major advantage of finewire electrodes is selectivity to measure the activity of specific muscles. The influence of electrical activity of nearby muscles is greatly reduced. Nonetheless, a number of disadvantages are associated with fine-wire electrodes. Pain on insertion, the difficulty of accurate placement, wire movement with muscle contraction, and the need for licensure to utilize wire electrodes are some of the drawbacks. Furthermore, subjects with in-dwelling electrodes walk more slowly after insertion of the electrodes (Young et aI., 1989). Because they are to be inserted transcutaneously, needle electrodes must be sterilized and sufficiently strong to resist breakage. Commonsense considerations, such as time, expense, pain experienced during a long study, the tolerance of the subject to multiple needle insertions, and the influence of indwelling electrodes on walking, necessitate a selection of the muscles most relevant to the specific movement abnormalities. Large muscles near the surface can be studied well with surface electrodes, whereas small muscles and those surrounded by other muscles require insertion of fine-wire electrodes. Electrical stimuli are usually given to confirm the accuracy of placement. EMG

348

systems are available in either hardwired or telemetry versions. The hardwired versions now send multiple signals on a single cable. These systems are reliable and less expensive than telemetry. Telemetry systems do not encumber the subject with cables but are susceptible to electromagnetic interference. Once the EMG data are acquired, they must be processed further to provide information about the timing of muscle activity and the relative intensity of the muscle activity. The EMG data are recorded throughout the gait cycle. The gait cycle is indicated either with synchronization of the kinematic data, foot-switch information, or force plate data to indicate each foot strike and toe-off. Analysis of the EMG is done by a phase-time plot of the activity of the muscle against events of the gait cycle. The raw EMG signal can be analyzed or processed further. The most common methods of EMG signal processing are full wave rectification, linear envelope, and integration of the rectified EMG. The linear envelope in created by low-pass filtering the full wave rectified signal. Integrated processing of the rectified signal is usually performed over short duration, i.e. 2% gait cycle, and then the integration is reset and accumulated again. The electromyogram provides a means for studying muscle activity. The signals that result from action potentials and muscle fibers are stochastic and non-stationary adding to uncertainty in interpretation. While the ultimate source of locomotor activity is muscle force, no study has established that electromyographic signals represent muscle force (Perry and Becky, 1981). The EMG signal is a measure of the bioelectric events that occur in conjunction with contraction of the muscle fibers. Thus, it is a phenomenon related to the initiation of muscle contraction rather than an effect of the muscles mechanical action. There are many difficulties in correlating the EMG signal amplitude with muscle force magnitude. Both linear and nonlinear relationships between the force level of skeletal muscles and the EMG signal have been reported (Lippold, 1952; Messier et al., 1971; Komi and Buskirk, 1972; Zuniga et al., 1972; Maton and Bouisset, 1977; Matral and Casser, 1977; Moritani and de Vries, 1978; Bigland-Ritchie et al., 1980; Woods and Bigland-Ritchie, 1983) Consequently, the EMG is commonly used in clinical gait analysis to determine phasic patterns for individual muscles or muscle groups. It is possible to examine simple

K.R. KAUFMAN

on/off patterns (Sutherland et al., 1988), or the EMG can be processed to find a graduation of signal level, after which EMG patterns are examined as defined by the level of activity over the gait cycle (Shiavi and Green, 1983; Limbird et al., 1988; Wooten et al., 1990). In the latter process, it is common to normalize the signal as a percentage of voluntary maximum muscle contraction. The process of detecting when a muscle is "turned on or off" is usually one of testing whether the average level of the signal is above some predefined limit. Normalization schemes may be used to aid in analysis. The limit is often defined as a percentage of the maximum voluntary muscle contraction or maximum EMG signal during gait. The determination of on/off time is often done by calculating the EMG level and then testing for occasions when the level exceeds some threshold value. The muscle is considered to be activated when at least 5% of the maximum electrical activity obtained during a manual muscle test is present for 5% of the gait cycle (Bogey et al., 1992). EMG on/off times are generally more variable from step to step than either kinematic or kinetic gait measurements. Problems occur in dynamic situations when using electromyographic activity as a measure of muscle functional capability. A dynamic force produced by a muscle is not proportional to the degree of muscular activity. Other factors may affect the muscle force, such as a change of the muscle length, change of the contraction velocity, the rate and type of muscle contraction, joint position, and muscle fatigue. It is desirable to find an alternative measurable mechanical parameter related to muscle force. The electromyographic signal does not assess the tension produced by a muscle, because the tension reflects the sum of both the active contraction and the passive stretch. A technique which may provide information about muscle force is measurement of intramuscular pressure. Intramuscular pressure (IMP) is a mechanical variable that is proportional to muscle tension. It is possible to obtain IMP measurements during gait and relate these measurements to the timing and intensity of muscle contraction. IMP has been used to quantify muscle function during dynamic activities (Kaufman and Sutherland, 1995). Intramuscular pressure measurements obtained during walking parallel the electromyographic activity, and also account for passive stretch of the muscle (Kaufman and Sutherland, 1995) (Fig. 10).

349

GAIT ANALYSIS

FS OTO

TO OFS I I I

EMG Right Gastrocnemius (volts)

FS

TO

OTO OFS J

I I

TO

FS TO

FS

OFS I

I I

+0.5-

+0.5

IMP Right Gastrocnemius so(mm HG) 0-

Fig. 10. Raw data for a single subject during gait. Both EMG and intramuscular pressure are being recorded from the gastrocnemius muscle. The stance phase of gait occurs from FS to TO. The swing phase of gait occurs from TO to FS. Single limb stance occurs from OTO to OFS. Peaks in intramuscular pressure during gait can be correlated with peaks of active contraction and passive stretch of the gastrocnemius. (Reproduced with permission from Kaufman and Sutherland, 1995.)

21.5. Interpretation of gait data

Once the data that describe the biomechanics of the patient's gait has been collected, the most crucial step of interpreting the data remains to be performed. Based on the clinical examination and measurements performed, the data must be synthesized and integrated in order to supply clinically relevant information. Human locomotion is very complex and multifaceted. The clinical interpretation of pathological gait disorders involves holding in human memory a large number of graphs, numbers, and clinical tests from data presented on hard copy, charts, x-rays, video, and computer generated 3-D graphics from multiple trials of a subject walking. Further, comparisons must be made to data from an able-bodied normal population in order to identify the potential movement problems for a given individual. The referring clinician, who may not be an expert in gait analysis, is commonly overwhelmed by the magnitude of the number of measurements included in a typical gait report. The

person interpreting the data must integrate this information. While data collection techniques for gait analysis have continually evolved over the last 50 years, the method of data presentation has not changed over this time. The data are still reported in 2-D charts with the abscissa usually defined as a percentage of the gait cycle and the ordinate displaying the gait parameter. Recent developments in computer animation may make it possible to apply advanced methods to visualize human movements. The large volume of variables currently found in a typical clinical report could be replaced with a few graphic images that succinctly provide the needed information. It is difficult to appreciate and fully understand the relationships between motion dynamics and biomechanical variables without scientific graphic visualization. Presently, computer software packages have advanced to the stage where it is possible to provide a gait analysis report using animation of fully 3-D, realistic graphical depictions of human locomotion. The format used for reporting test

350

K.R. KAUFMAN

results is a matter of considerable importance. The data must be presented in an accurate, clear, and concise format. If the results are not communicated in an effective format, they will be of little use to the clinician regardless of quality.

21.6. Treatment planning and assessment When treatment is being planned, the main objective is to differentiate between the primary causes and compensations for the patient's functional problems. If the treatment is directed at a compensation, the patient will lose their ability to compensate and their movement problems will worsen. The patient will display adaptations in their gait pattern due to their pain, injury, deformity, instability, and/or inappropriate muscle activation patterns. The ramifications of these problems cannot be fully assessed without an instrumented gait study. Patients can undergo dynamic adaptation related to the biomechanics of walking which must be factored into the treatment algorithm. For example, it has been shown that when planning corrective osteot-

omy knee surgery, patients with the same bony deformity will have differing knee loading due to dynamic adaptations (Prodromos et al., 1985). These dynamic adaptations will have a direct effect on surgical outcome. The patients who dynamically compensated for their malalignment had a better long-term outcome (Prodromos et al., 1985; Wang et al., 1990). For patients with progressive disorders, these dynamic adaptations will also change with time as the disease progresses. It is possible to use motion analysis studies to quantify these dynamic changes in locomotor patterns.

21.7. illustrative case We are currently conducting a study to quantify gait changes in patients with progressive multiple sclerosis. Using gait analysis techniques, it is possible to show changes in joint motion which can be attributed to disease progression. An example of changes in knee motion is shown in Fig. 11. The patient had an EDSS (Kurtzke) rating of 3.0 when

KNEE FLEXION

60

~

40

>< w

20

C)

-oz ..J La.

__

.,....----------~---

Time (Weeks) --Owk 1 wk - - 26wk 52wk ... 78wk - - Normal

W W

~

0

o

20

40

60

80

100

GAIT CYCLE (%) Fig. II. Changes in knee motion over an 18-month period in a patient with primary progressive multiple sclerosis. A normal knee motion curve is given for comparison. The patient lacks knee flexion during early stance and goes into hyperextension during midstance. The knee flexion during swing is less than normal at the initial study and progressively declines over the 18-month study period. The results for the initial visit (0 weeks) and a second study done a week later (l week) demonstrate excellent measurement repeatability.

351

GAIT ANALYSIS

first studied, and progressed to an EDSS score of 5.5 at the end of the 18 months study period, thus requiring a cane for ambulation. Comparing the data from the initial (0 week) and second visit (l week), it can be seen that the data are very repeatable. The motion curves for these two timepoints differed by less than three degrees on average throughout the entire gait cycle. The data also demonstrated that the patient lacked the normal knee flexion (shock absorption) wave during early stance. Further, the patient's knee went into hyperextension in midstance. These gait characteristics were consistent for all visits during the study period. The knee flexion during swing was less than normal at the initial visit and declined further over the 18 month study period. This example clearly illustrates the objective nature of instrumented gait analysis and the ability to document subtle changes which may not be remembered over time if only clinical observation is used.

21.8. Rationale In this era of government reimbursement for medical services, the ability to document the need and effectiveness of a particular treatment will assume an increasingly important role. Managed care will require validation for many types of therapeutic interventions. Pre- and post-treatment measurement will become mandatory. Outcomes will have to be compared. Practitioners and facilities

will be rated on their outcomes. Maximizing anticipated outcomes will be required to document that a treatment plan is worthwhile. Objective gait analysis is an essential tool to meet these demands. The technology is at a level where it is both feasible and affordable to provide an objective form of patient assessment. In all areas of medical care, a need exists for instrumentation and procedures to aid in a differential diagnosis and treatment of patients. Further, information is required to document patient response to treatment objectively. The ability to diagnose, prescribe treatment, and document results is common to all areas of medical care. However, the technology available in different medical specialties varies widely. This is particularly evident when the current medical technology for treating patients with cardiovascular conditions is compared to the technology for treating patients with neuromuscular conditions (Table 3). For both types of patients the technology can be divided into three levels: static examinations, dynamic examinations, and invasive procedures. When a patient reports to a physician that they are experiencing chest pain and the risk factors for a myocardial infarction exist, the patient is monitored with an electrocardiogram. In some centers, a computed tomography (CT) scan is obtained to assess the amount of arteriosclerosis in the vessels of the heart muscle. These tests are obtained while the patient is either sitting or lying down. Hence, these

Table 3 Current medical technology. Treatment modality Level

Exam

Cardiovascular patients

Neuromuscular patients

Static

ECG EBCT

X-ray CT

MRI

MRI

II

Dynamic

III

Invasive

PET Ultrasound Nuclear imaging Stress test Echocardiography Angiography

Nuclear imaging

Motionanalysis Diagnostic EMG Kinesiological EMG

352

constitute static exams. Other static modalities include MRI, PET, ultrasound, and nuclear imaging. If the physician has a high index of suspicion, dynamic tests may be undertaken. These include a stress test or echocardiography. Finally, angiography may be utilized as an invasive procedure to further examine the heart. In stark contrast, most of the modalities for a patient with neuromuscular dysfunction are static modalities. Current modalities include x-ray, CT, MRI, nuclear imaging and diagnostic electromyography (EMG). However, all of these examinations are static and non-weight bearing. Dynamic assessment of patients with neuromuscular dysfunction can only be obtained using motion analysis techniques along with acquisition of kinesiological EMG. An objection sometimes raised is that these studies are too costly. However, this objection is unfounded. In terms of cost-benefit ratio, the most compelling consideration is the high cost of inappropriate treatment. It is important to remember that unsuccessful treatment will result in unfavorable changes in function and may require subsequent procedures to deal with the original problem. There is, of course, no assurance that gait studies performed prior to treatment planning will always ensure favorable outcome, but careful planning, based on objective data, will provide a solid foundation for decision making. Post-treatment studies will give the information required for objective evaluation of treatment results. The rate at which gait analysis technologies will become more common will depend on the market, the manufacturers, and managed care requirement. Objective patient assessment using gait analysis techniques will facilitate the identification of optimal treatment regimens, and provide a solid foundation for clinical decision making. 21.9. Summary Human function cannot be fully understood without studies of movement. The modem motion analysis laboratory that has evolved with technological advances has the potential for opening new opportunities for progress in the treatment of patients with movement disorders. Current gait analysis systems offer sophisticated automatic track-

K.R. KAUFMAN

ing systems, force platforms, and electromyographic activity measurements. When coupled with a biomechanical model, this equipment is able to provide a complete, three-dimensional, dynamic description of the patient's gait along with information on the timing and intensity of muscle activity. The function of a motion analysis laboratory is to objectively measure the dynamic aspects of an individual patient's performance that cannot be quantitatively assessed in the clinical setting. Interpretation of this data makes it possible to integrate morphology and functional adaptations in order to understand and treat the pathophysiology of movement disorders. Acknowledgment The author thanks Barbara Iverson-Literski for her careful manuscript preparation. References An, K-N, Jacobsen, MC et al. (1988) Application of a

magnetic tracking device to kinesiologic studies. J. Biomech., 21(7): 613-620.

Basmajian, JV and De Luca CL (1985) Muscles alive: their functions revealed by electromyography (5th ed.). Baltimore, MD, Williams & Wilkins. Bell, AL, Pederson, DR et al. (1989) Prediction of hip joint center location from external landmarks. Hum. Move. Sci., 8: 3-16. Bigland-Ritchie, B, Kukulka, CJ et al. (1980) Surface EMG-force relationships in human muscles of different fiber composition. J. Physiol. (Lond.), 308: 103P. Bogey, RA, Barnes, LA et al. (1992) Computer algorithms to characterize individual subject EMG profiles during gait. Arch. Phys. Med. Rehab., 73: 835-841. Borelli, GA (1685) De motor animalium. Batavis: Lugduni. Braune, Wand Fischer 0 (1895) Der Gang des Menschen (The human gait). Leipzig: BG Teubner. Cappozzo, A, Catani, F et al. (1995) Position and orientation in space of bones during movement: anatomical frame definition and determination. Clin. Biomech., 10(4): 171-178. Chao, EYS (1971) Determination of applied forces in linking systems with known displacements: with special application to biomechanics. Iowa City, Iowa, University of Iowa. Chao, EYS (1980) Justification of Triaxial Goniometer for the measurement of joint rotation. J. Biomech., 13: 989-1006.

GAIT ANALYSIS

Eastlack, ME, Arvidson, J et al. (1991) Interrater reliability of videotaped observation of gait-analysis assessments. Phys. Ther., 71(6): 465-472. Gage, JR and Ounpuu S (1989) Gait analysis in clinical practice. Sem. Orthopaed., 2: 72-87. Grood, ES and Suntay WJ (1983) A joint coordinate system for the clinical discussion of 3-dimensional motions: applications to the knee. J. Biomech. Engineer., 105: 136-144. Inman, VT, Ralston, HJ et al. (1981) Human walking. Baltimore, MD, Williams & Wilkins. Kadaba, MP, Ramakrishnan, HK et al. (1990) Measurement of lower extremity kinematics during level walking. J. Orthopaed. Res., 8: 383-392. Kaufman, KR and Sutherland DH (1995) Dynamic intramuscular pressure measurement during gait. Oper. Tech. Sports Med., 3(4): 250-255. Kaufman, K, An, K et al. (1991) Physiological prediction of muscle forces - I. Theoretical formulation. Neurosci., 40(3): 781-792. Kaufman, KR, An, KN et al. (1995) A comparison of intersegmental joint dynamics to isokinetic dynamometer measurements. J. Biomech., 28(10): 1243-1265. Koh, TJ and Grabiner MD (1992) Cross-talk in surface electromyograms of human hamstring muscles. J. Orthopaed. Res., 10: 701-709. Komi, PV and Buskirk ER (1972) Effect of eccentric and concentric muscle conditioning on tension and electrical activity of human muscle. Ergonomics, 15: 417. Krebs, DE, Edelstein, JE et al. (1985) Reliability of observational kinematic gait analysis. Phys. Ther., 65: 1027-2033. Lettre, C and Contini R (1967) Accelerographic analysis of pathologtical gait. New York, New York University School of Engineering and Science Technical Report 1368-01. Limbird, TJ, Shiavi, Ret al. (1988) EMG profiles of knee joint musclature during walking: Changes induced by anterior cruciate ligament deficiency. 1. Orthopaed. Res., 6: 630. Lippold, OCJ (1952) The relation between integrated action potentials in a human muscle and its isometric tension. J. Physiol. (Lond.), 117: 492. Maron, Band Bouisset S (1977) The distribution of activity among the muscles of a single group during isometric contraction. Eur. 1. Appl. Physiol., 37: 101-109. Matral, Sand Casser G (1977) Relationship between force and integrated EMG activity during voluntary isometric and isotonic contraction. Eur: J. Appl. Physiol., 46: 185.

353 Messier, RH, Duffy, Jet al. (1971) The electromyogram as a measure of tension in the human biceps and triceps muscles. Inter. J. Mech. Sci., 13: 585-598. Moritani, T and de Vries HA (1978) Re-examination of the relationship between the surface integrated electromyogram (IEMG) and force of isometric contraction. Am. J. Phys. Med., 57: 263-277. Muybridge, E (1979 (original work published in 1887)) Human and animal locomotion. New York, Dover. Perry, J (1992) Gait analysis: normal and pathological function. Thorofare, NJ, Slack. Prodromos, CC, Andriacchi, TP et al. (1985) A relationship between gait and clinical changes following high tibial osteotomy. J. Bone Joint Surg. (American Vol.), 67A: 1188-1194. Raab, FH, Blood, EB et al. (1979) Magnetic position and orientation tracking system. IEEE Trans. Aerospace Electron. Sys., AES-15(5): 709-718. Saleh, M and Murdoch G (1985) In defense of gait analysis. J. Bone Joint Surg., 67B: 237-241. Shiavi, R and Green N (1983) Ensemble averaging of locomotor electromyographic patterns using interpolation. Med. BioI. Eng. Comput., 21: 573. Sutherland, DH and Cooper L (1981) The events of gait. Bull. Prosthet. Res., 10-35: 281-282. Sutherland, DH, Olshen, RA et al. (1988) The development of mature walking. Oxford, England, Mac Keith Press. Wang, JW, Kuo, NN et al. (1990) The influence of walking mechanics and time on the results of proximal tibial osteotomy. J. Bone Joint Surg., 6: 905-909. Winter, DA (1990) Biomechics and Motor Control of Human Movement (2nd ed.). New York, Wiley & Sons. Woods, JJ and Bigland-Ritchie B (1983) Linear and nonlinear surface EMG/force relationships in human muscles. Am. 1. Phys. Med., 62(6): 287-299. Wooten, ME, Kadaba, MP et al. (1990) Dynamic electromyography. n. Normal patterns during gait. J. Orthopaed. Res., 8: 259. Young, CC, Rose, SE et al. (1989) The effect of surface and internal electrodes on the gait of children with cerebral palsy, spastic diplegia type. 1. Orthopaed. Res., 7(5): 732-737. Zuniga, EN and Simons DG (1969) Nonlinear relationship between averaged electromyogram potential and muscle tension in normal subjects. Arch. Phys. Med. Rehab., 50: 613. Zuniga, EN, Leavitt, LA et al. (1972) Gait patterns in the above-knee amputees. Arch. Phys. Med. Rehab., 53(8): 373-382.

Movement Disorders Handbook of Clinical Neurophysiology, Vol. I M. Hallett (Ed.) © 2003 Elsevier B.V. All rights reserved

357 CHAPTER 22

Physiologic and enhanced physiologic tremor Rodger J. Elble* Department of Neurology, Southern Illinois University School of Medicine, P.O. Box 19643, Springfield, 1L 62794-9643, USA

Healthy people exhibit rhythmic oscillations in body position and muscle contraction, and these oscillations are called physiologic tremor. The mechanisms of physiologic tremor are reviewed in this chapter.

22.1. Passive mechanical oscillation The friction and viscous damping in most joints are not sufficient to prevent passive mechanical oscillation in response to a perturbation. Skeletal muscle behaves as a low-pass filter in response to motor-unit excitation, and this filtering effect limits the fluctuations in muscle force caused by irregularities in motor unit firing (Stein and Oguztoreli, 1976). However, these irregularities are not filtered completely, and they continuously perturb associated joints into oscillation. Muscle-spindle and tendonorgan reflex loops lack the sensitivity to attenuate these small oscillations, and the underdamped mechanical properties of most joints are conducive to oscillation (Hagbarth and Young, 1979; Stiles and Hahs, 1991). Consequently, irregularities in muscle force keep most body parts in a constant state of oscillation that is barely visible to the unaided eye. This oscillation is the mechanical component of physiologic tremor. The mechanical component of physiologic tremor has a frequency w that depends upon the inertia I and stiffness K of the joint, according to the equation w =VKii (Lakie et al., 1986). Therefore, tremor frequency is increased by added stiffness and decreased by added inertia. Furthermore, the frequency of this passive mechanical oscillation will

* Correspondence to: Dr. RJ. Elble, Department of Neurology, Southern Illinois University School of Medicine, P.O. Box 19643, Springfield, IL 62794-9643, USA. E-mail address: [email protected] Tel.: 217-524-7881 (ext. 3002); fax: 217-524-1903.

vary from joint to joint, depending upon the natural inertia and stiffness. Consequently, the frequency of passive mechanical oscillation at the metacarpophalangeal joint is higher (20-30 Hz) than at the wrist (7-10 Hz) and elbow (3-5 Hz) (Stiles and Randall, 1967; Fox and Randall, 1970; Elble and Randall, 1978). The mechanical component of physiologic tremor exists during complete relaxation (rest tremor), active steady posture (postural tremor), and voluntary movement (kinetic tremor). Rest tremor occurs because the mechanical shock of cardiac systole perturbs the body into oscillation (Brumlik and Yap, 1970). These cardioballistic oscillations also contribute to physiologic action tremor (Elble and Randall, 1978), but the relative magnitude of this contribution depends upon the level of voluntary motor activity and varies with the site of recording. Postural head tremor is largely cardioballistic (Fig. 1), while cardioballistics play a minor role in postural hand tremor (Marsden et aI., 1969; Elble and Randall, 1978). The contribution of cardioballistics to physiologic tremor has been studied extensively by looking for mathematical correlation or coherence between the tremor and the timing of cardiac systole, as measured with the electrocardiogram (Marsden et aI., 1969; Elble and Randall, 1978). However, the contribution of cardioballistics is not routinely quantified in measurements of physiologic tremor. Physiologic tremor is barely visible to the unaided eye. Hand tremor is most visible in the extended fingers and may not be visible at the wrist. Physiologic hand tremor is only symptomatic during fine motor tasks requiring great precision, as in microvascular surgery (Harwell and Ferguson, 1983). Normal postural hand tremor recorded 10 em from the wrist contains rhythmic oscillations with a mean peak-to-peak amplitude of 0.009-0.153 mm displacement and 3-33 cm/s' acceleration (Elble, 1986). These mean amplitudes overlap somewhat

358

R.I. ELBLE

Lateral Acceleration

'"

~o

0

t

t -1

5

6

7

8

9

10

5

6

7

8

9

10

Time (s)

Fig. 1. Five-second recordings of head tremor and electrocardiogram (EKG) were obtained from a healthy 80-year-old woman while she sat quietly in a chair. Tremor was measured with a triaxial accelerometer attached to her forehead. Lateral (side-to-side) acceleration is shown in the upper graph. Note the transient increase in acceleration (arrows) following each QRS complex of the EKG, typical of cardioballistic oscillation.

with the lower range of mild essential tremor (Elble, 1986). 22.2. Neurogenic oscillation

Motor unit firing is largely asynchronous during steady muscle contractions. Consequently, the Fourier power spectrum of the rectified-filtered EMG is statistically flat (Fig. 2). Two or more motor units may discharge synchronously (within a few milliseconds) during steady voluntary contractions, but this short-term synchronization occurs intermittently for a few milliseconds and affects only 8% of motor unit discharges (Dietz et al., 1976; Dengler et al., 1984; Datta et al., 1991; De Luca et al., 1993; Semmler and Nordstrom, 1998). The degree to which short-term synchronization contributes to physiologic tremor has not been quantified, and the mechanism of short-term synchronization is uncertain. Some authors believe that this synchronization is due to the joint occurrence of excitatory postsynaptic potentials evoked in multiple motoneurons by branches of a corticospinal fiber (Sears and Stagg, 1976; Datta and Stephens, 1990; Datta et al., 1991; Farmer et al., 1997), while others favor a role

for central oscillatory drive to the motoneuron pool (De Luca et al., 1993). Sensory feedback is not necessary for short-term synchronization (Farmer et al., 1993; Farmer et al., 1997). Rhythmic oscillations in volitional muscle contraction were observed more than 100 years ago by Schafer et al. (1886). The 8-12 Hz component of physiologic tremor is produced by bursts of motorunit discharge at 8-12 Hz. These bursts are produced by synchronous 8-12 Hz modulation of motor unit firing, such that double (paired) discharges, with interspike intervals of 10-40 ms, tend to occur during a cycle of tremor (Elble and Randall, 1976; Kakuda et al., 1999; Wessberg and Kakuda, 1999). The mean firing frequency of participating motor units ranges from 8 to 25 spikes/so Short-term synchronization of motor units does not contribute significantly (Farmer et al., 1993). The frequency of 8-12 Hz entrainment is independent of reflex arc length and is not reduced by increased limb inertia (Fox and Randall, 1970; Stephens and Taylor, 1974; Elble and Randall, 1976; Elble and Randall, 1978). Furthermore, the stretch reflex response to joint perturbation is too weak and too delayed to account for this entrainment (Wessberg and Vallbo, 1996). Therefore, this tremor is believed to emerge from a central source of oscillation. The inferior olive is the most frequently hypothesized source of 8-12 Hz tremor, but this hypothesis is largely conjectural, based on the similarities between this tremor and the 8-12 Hz harmaline tremor in laboratory primates (Elble, 1998a). Other sources of rhythmicity (e.g. thalamus and cortex) are also possible. Koster et al. (1998) found that the 8-12 Hz tremor, enhanced by salbutamol, was coherent between the two forearms in three patients with the syndrome of persistent mirror movements. As these patients are believed to have ipsilateral and contralateral corticospinal projections, Koster's data suggest that that enhanced physiologic tremor is conducted through corticospinal pathways. Corticospinal involvement is supported by the recent finding of coherence between the 8-12 Hz tremor and EEG (Raethjen et al., 2000a). However, the primary source of oscillation is still uncertain. Elble (2003) quantified hand tremor and forearm EMG in 100 healthy adults, aged 20-40, and 100 older adults, aged 70-90, and Raethjen et al. (2000b) quantified hand tremor and finger tremor in 117 normal people, aged 20-94 years. All of these

359

PHYSIOLOGIC AND ENHANCED PHYSIOLOGIC TREMOR

Physiologic tremor

Enhanced Physiologic Tremor

No added load


60

No added load

~

.[ 45

2

4

6

8

10

2

12

500- ram load

-.~.. a..

~

w

6

8

10

12

6

8

10

12

100 ,------=-=..=....oz.:,.=~~'----_,
-~...


~

4

150

75

Q)

100

a..

~

50

50

~

25

w

2

4

6

8

10

12

Frequency (Hz)

2

4

Frequency (Hz)

Fig. 2. Autospectra of hand tremor and rectified-filtered extensor carpi radialis brevis EMG recorded with (thin lines) and without (thick lines) a 500-gram load on the hand. Tremor was recorded with an accelerometer and skin electrodes during horizontal extension of the hand, with forearm supported. The frequency of physiologic tremor (left column graphs) decreased with 500-gram loading, and there was no associated peak in the EMG spectrum (i.e. no evidence of motor-unit entrainment). The frequency of enhanced physiologic tremor in a patient with thyrotoxicosis (right column graphs) also decreased with mass loading, but this tremor was associated with a significant peak in the EMG spectrum (arrows).

normal controls exhibited the mechanical component of physiologic tremor, usually with no evidence of associated motor unit entrainment. Only about 10% of these controls exhibited 8-12 Hz motor unit entrainment that was strong enough to produce Fourier spectral peaks in the acceleration and rectified-filtered EMG spectra (Fig. 3). The frequency of this neurogenic tremor may be as low as 6-8 Hz in people older than 65 years. This motor unit entrainment is indistinguishable from mild essential tremor, but its relationship to essential tremor is unclear (Elble, 1986). Most, if not all, people exhibit 8-12 Hz bursts of EMG during slow voluntary movements, particularly in the wrist and finger extensors during slow wrist or finger flexion (Wessberg and Vallbo, 1996; Kakuda et al., 1999).

Thus, there is a tendency for 8-12 Hz motor unit entrainment to occur in everyone, but this tendency is too weak in most healthy adults to produce an EMG spectral peak during voluntary horizontal extension of the hand or finger. In a study of finger tremor, Halliday et al. demonstrated the presence of 15-30 Hz motor unit entrainment that was estimated to explain about 20% of finger tremor in this frequency band (Halliday et al., 1999). The contribution of 15-30 Hz motor unit entrainment to tremor in body parts with greater inertia (e.g. hand, forearm) is much smaller and usually not measurable, even with accelerometry. This component of physiologic tremor is believed to emerge from cortical rhythmicity, as detected with EEG and MEG (Conway et al., 1995; Baker et al.,

360

RJ. ELBLE

32-year-old woman No added mass

20

'" e:

be a manifestation of the neural mechanisms of smooth pursuit (McAuley et al., 1999a, b; Spauschus et al., 1999).

...

22.3. Enhanced physiologic tremor

'" e-

160

~

120

15 E .£.

~0

80

10 0~

If)

If)

... ...0

...0

Q.

Q.

5

E 40

~ I-

2

4

6

8

10

0 12

E

~ I-

4

'":;-

'"

-...

-... ~3

E

Q)

2 0~

~ 2

Q.

Q.

~

w

"w

1 00

::E

2

4

6

8

10

Frequency (Hz) Fig. 3. Autospectra of hand tremor and rectified-filtered extensor carpi radialis brevis EMG recorded with (thin lines; right vertical axes) and without (thick lines; left vertical axes) a 300-grarn load were obtained from a neurologically-normal 32-year-old woman. Tremor was recorded with an accelerometer and skin electrodes during horizontal extension of the hand, with forearm supported. With no mass loading, a single peak is found in the acceleration and rectified-filtered EMG spectra. Mass loading reduced the amplitude and frequency of the mechanical resonance oscillation, which disclosed the second oscillation with motor-unit entrainment at 8-12 Hz (arrows). With no mass loading, the two sources of oscillation seemed to resonate at the same frequency.

1997; Salenius et al., 1997; Halliday et al., 1998; Baker et al., 1999). Normal vocal tremor has a frequency of 6-13 Hz and is due to a combination of rhythmic neural drive and mechanical resonance of the larynx (Ludlow et al., 1986; Winholtz and Ramig, 1992). Ocular microtremor has a peak frequency of approximately 80 Hz, reflecting the high-frequency motoneuron entrainment and extraocular muscle dynamics (Bolger et al., 1999; Spauschus et al., 1999). Additional ocular microtremor at 4 and 10Hz may

Enhanced physiologic tremor occurs during fatigue, anxiety and voluntary movement and in response to beta-adrenergic drugs (Stiles, 1976; Logigian et al., 1988; Stiles and Hahs, 1991). Under these circumstances, the amplitude of tremor may increase by a factor of 5 to 20, and the EMG exhibits bursts of motor unit activity at the frequency of tremor (Hagbarth and Young, 1979; Stiles, 1980). These bursts are produced by entrainment of motor units with mean firing frequencies of 8 to 25 spikes/s, and the modulation of individual motor units is such that double (paired) discharges, with interspike intervals of 10-40 ms, tend to occur during a cycle of tremor (Logigian et al., 1988). Enhanced participation of spinal and possibly longloop stretch reflex pathways plays an important role in the entrainment of motor units. In most people, this enhanced physiologic tremor consists of a single rhythmic oscillation that decreases in frequency with inertial loading (Fig. 2). Thus, enhanced physiologic tremor is typically a mechanical-reflex oscillation in which the inertia and stiffness of the joint play a pivotal role determining tremor frequency. However, the 8-12 Hz component of physiologic tremor is also increased by fatigue, anxiety, voluntary movement (e.g. wrist extension-flexion), beta-adrenergic drugs and central nervous system stimulants (Raethjen et al., 2001), and this component of physiologic tremor may be more evident in people with enhanced tremor. The stiffness and damping of the wrist and other joints are not constant; rather they are a function of displacement and velocity. The effective damping ratio of the wrist and other joints of the limbs is less than one (underdamped), making these joints prone to oscillation. Stiffness and damping increase as the displacement and velocity of oscillation decrease (Milner and Cloutier, 1998). These nonlinear mechanical properties tend to attenuate and control oscillation at low magnitudes to which the stretch reflex is relatively insensitive. Stiffness and damping also increase with increased muscle activation (Milner and Cloutier, 1998), which explains the common

PHYSIOLOGIC AND ENHANCED PHYSIOLOGIC TREMOR

361

strategy of antagonistic muscle co-contraction when stability is threatened. The frequency of enhanced mechanical-reflex tremor decreases as the amplitude increases, and this can be explained, at least partially, by a reduction in the mechanical stiffness that occurs when the amplitude of joint oscillation increases (Gottlieb and Agarwal, 1977; Agarwal and Gottlieb, 1984; Lakie et aI., 1984; Zahalak and Pramod, 1985; Milner and Cloutier, 1998). The reduction in tremor frequency with increased amplitude produces a greater phase advance of sensory feedback on tremor, and this results in greater reflex damping, which is beneficial in the control of tremor (Stiles and Hahs, 1991). Stretch-reflex responses can destabilize the wrist and similar joints at frequencies of 7 Hz or greater (Milner and Cloutier, 1998). People with deafferented limbs exhibit irregular broad-band errors in limb position that exceed the amplitude of physiologic and enhanced physiologic tremor, but they do not exhibit the increased tremor with rhythmic motor unit entrainment seen in patients with enhanced mechanical-reflex tremor (Sanes, 1985). Thus, the presence of sensory feedback is not deleterious in terms of the overall amplitude of positional error, rather sensory feedback seems to entrain or concentrate error at a particular frequency, resulting in rhythmic oscillation. The stretch reflex probably contributes to most, if not all, pathologic tremors in this manner, even when a tremor emerges from a central source of oscillation. With greater involvement of segmental and longloop reflexes, the reflex stiffness, damping and loop time play an increasingly important role in determining tremor frequency, while the influence of limb mechanics on tremor frequency diminishes. Consequently, enhanced mechanical-reflex tremor has a frequency that is less dependent upon limb mechanics and more dependent upon reflex loop properties than normal mechanical tremor (Stiles, 1980). This is nicely illustrated in mathematical models of mechanical-reflex tremor (Stein and Oguztoreli, 1976; Bock and Wenderoth, 1999). In these models, the frequency of mechanical-reflex tremor is nearly independent of limb stiffness and inertia when mechanical damping and stiffness are very small relative to stretch-reflex damping and stiffness. Therefore, in the situation of large reflex gain, a mechanical-reflex oscillation may be difficult to

distinguish from a central oscillation on the basis of the frequency response to mechanical loading. However, the mechanical-reflex tremor frequency is still a function of reflex arc length (latency) (Bock and Wenderoth, 1999) and thereby differs from the 8-12 Hz component of physiologic tremor and from essential and Parkinson tremors (Deuschl et al., 1996). 22.4. Clinical guidelines The EMG and mechanical-loading properties of physiologic tremor are of value in the electrophysiologic analysis of action tremor of uncertain etiology. However, the following guidelines and caveats are noteworthy: (1) A tremor whose frequency varies predictably

with mechanical load or reflex arc length is produced, at least in part, by mechanical-reflex mechanisms. A tremor whose frequency is independent of mechanical load and reflex arc length most likely emerges from a central source of oscillation. (2) Tremor with the EMG, amplitude and frequency properties of physiologic tremor or enhanced mechanical-reflex tremor may be recorded from patients with intermittent or very mild action tremor of central origin (e.g. essential tremor, Parkinson tremor) (Deuschl et al., 1996; Deuschl and Elble, 2000). The motor-unit entrainment in a mild pathologic tremor of central origin can be so irregular or intermittent that the bursts of motor-unit activity simply perturb the limb, producing an enhanced mechanical-reflex oscillation (Elble, 1991). A sustained rhythmic motor-unit entrainment is needed to produce an EMG spectral peak that is frequency-invariant with mechanical loading. (3) The significance of a prominent unenhanced 8-12 Hz tremor is unclear. Identical tremor can be recorded from patients with mild essential tremor (Elble, 1986) and Parkinson disease (Lance et al., 1963), and many ostensibly normal older people (age >70) may exhibit a slightly lower tremor frequency of 6--8Hz (Elble, 1998b), making this tremor virtually indistinguishable from mild essential tremor. Since a prominent 8-12 Hz tremor is found in about 10% of controls, its presence should raise the clinical suspicion of an underlying neurologic disorder.

362

(4) Some pathologic tremors exhibit the electrophysiologic characteristics of a mechanicalreflex oscillation (e.g. cerebellar outflow tract tremors) (Elble et aI., 1984; Qureshi et aI., 1996). 22.5. Summary Normal action tremor in the upper extremity consists primarily of two rhythmic oscillations, mechanical-reflex and 8-12 Hz (Elble and Randall, 1978; Timmer et aI., 1998). The 8-12 Hz component of physiologic tremor is produced by rhythmic motor-unit entrainment at 8-12 Hz although seemingly normal elderly people exhibit frequencies as low as 6-8 Hz. The frequency of this component is not affected by changes in limb mechanics (inertia and stiffness) or reflex arc length, and it is therefore believed to emerge from a central source of oscillation, the identity of which is unknown. The mechanical-reflex component is so named because its frequency is a function of the inertia and stiffness of the limb and its reflex arc, and when the stretch reflex is enhanced, the frequency of tremor is also a function of reflex arc length. Irregularities in motor unit firing and cardioballistics provide a broadfrequency forcing to the limb, resulting in mechanical-reflex oscillation. The mechanical-reflex oscillation is associated with motor-unit entrainment when its amplitude becomes large enough to induce reflex modulation of motor-unit discharge or when the sensitivity of the reflex arc is increased by such factors as drugs, fatigue, and anxiety. This enhanced mechanical-reflex oscillation is the primary source of enhanced physiologic tremor although the 8-12 Hz component is also enhanced by these factors.

Acknowledgment This work was supported by the Spastic Paralysis Research Foundation of Kiwanis International, Illinois-Eastern Iowa District.

References Agarwal, GC and Gottlieb, GL (1984) Mathematical modeling and simulation of the postural control loop, Part III. Grit. Rev. Biomed. Eng., 12: 49-93.

R.I. ELBLE

Baker, SN, Olivier, E and Lemon, RN (1997) Coherent oscillations in monkey motor cortex and hand muscle EMG show task-dependent modulation. J. Physiol. (Lond.), 501: 225-241. Baker, SN, Kilner, JM, Pinches, EM and Lemon, RN (1999) The role of synchrony and oscillations in the motor output. Exp. Brain Res., 128: 109-117. Bock, a and Wenderoth, N (1999) Dependence of peripheral tremor on mechanical perturbations: modeling study. Biol. Cybern., 80: 103-108. Bolger, C, Bojanic, S, Sheahan, NF, Coakley, D and Malone, JF (1999) Dominant frequency content of ocular microtremor from normal subjects. Vision Res., 39: 1911-1915. Brurnlik, J and Yap, C-B (1970) Normal tremor: a comparative study. Charles C. Thomas, Springfield, IL. Conway, BA, Halliday, DM, Farmer, SF, Shahani, D, Maas, P, Weir, AI and Rosenberg, JR (1995) Synchronization between motor cortex and spinal motoneuronal pool during the performance of a maintained motor task in man. 1. Physiol. (Lond.), 489: 917-924. Datta, AK and Stephens, JA (1990) Synchronization of motor unit activity during voluntary contraction in man. J. Physiol. (Lond.), 422: 397--419. Datta, AK, Farmer, SF and Stephens, JA (1991) Central nervous pathways underlying synchronization of human motor unit firing studied during voluntary contractions. J. Physiol. (Lond.), 432: 401--425. De Luca, CJ, Roy, AM and Erim, Z (1993) Synchronization of motor-unit firings in several human muscles. J. Neurophysiol., 70: 2010-2023. Dengler, R, Wolf, W, Birk, P and Struppler, A (1984) Synchronous discharges in pairs of steadily firing motor units tend to form clusters. Neurosci. Lett., 47: 167-172. Deuschl, G, Krack, P, Lauk, M and Timmer, J (1996) Clinical neurophysiology of tremor. J. Clin. Neurophysiol., 13: 110-121. Deuschl, G and Elble, RJ (2000) The pathophysiology of essential tremor. Neurology, 54: SI4-20. Dietz, V, Bischofberger, E, Wita, C and Freund, H-J (1976) Correlation between the discharges of two simultaneously recorded motor units and physiological tremor. Electroencephalogr. Clin. Neurophysiol., 40: 97-105. Elble, RJ (1986) Physiologic and essential tremor. Neurology, 36: 225-231. Elble, RJ (1991) Inhibition of forearm EMG by palatal myoclonus. Mov. Disord., 6: 324-329. Elble, RJ (1998a) Animal models of action tremor. Mov. Disord., 13: 35-39.

PHYSIOLOGIC AND ENHANCED PHYSIOLOGIC TREMOR

363

Elble, RJ (1998b) Tremor in ostensibly normal elderly people. Mov. Disord., 13: 457-464. Elble, RJ (2003) Characteristics of physiologic tremor in young and elderly adults. Clin. Neurophysiol., 114: 624-635. Elble, RJ and Randall, JE (1976) Motor-unit activity responsible for 8- to 12-Hz component of human physiological finger tremor. J. Neurophysiol., 39: 370383. Elble, RJ and Randall, JE (1978) Mechanistic components of normal hand tremor. Electroencephalogr. Clin. Neurophysiol., 44: 72-82. Elble, RJ, Schieber, MH and Thach, WT, Jr (1984) Activity of muscle spindles, motor cortex and cerebellar nuclei during action tremor. Brain Res., 323: 330-334. Farmer, SF, Bremner, PD, Halliday, DM, Rosenberg, JR and Stephens, JA (1993) The frequency content of common synaptic inputs to motoneurons studied during voluntary isometric contraction in man. J. Physiol. (Lond.), 470: 127-155. Farmer, SF, Halliday, DM, Conway, BA, Stephens, JA and Rosenberg, JR (1997) A review of recent applications of cross-correlation methodologies to human motor unit recording. J. Neurosci. Meth., 74: 175-187. Fox, JR and Randall, JE (1970) Relationship between forearm tremor and the biceps electromyogram. J. Appl. Physiol., 29: 103-108. Gottlieb, GL and Agarwal, GC (1977) Physiological clonus in man. Exp. Neurol., 54: 616-621. Hagbarth, K-E and Young, RR (1979) Participation of the stretch reflex in human physiological tremor. Brain, 102: 509-526. Halliday, DM, Conway, BA, Farmer, SF and Rosenberg, JR (1998) Using electroencephalography to study functional coupling between cortical activity and electromyograms during voluntary contractions in humans. Neurosci. Lett., 241: 5-8. Halliday, DM, Conway, BA, Farmer, SF and Rosenberg, JR (1999) Load-independent contributions from motorunit synchronization to human physiological tremor. J. Neurophysiol., 82: 664-675. Harwell, RC and Ferguson, RL (1983) Physiologic tremor and microsurgery. Microsurgery, 4: 187-192. Kakuda, N, Nagaoka, M and Wessberg, J (1999) Common modulation of motor unit pairs during slow wrist movement in man. J. Physiol. (Lond.), 520: 929-940. Koster, B, Lauk, M, Timmer, J, Winter, T, Guschlbauer, B, Glocker, FX, Danek, A, Deuschl, G and Lucking, CH (1998) Central mechanisms in human enhanced physiological tremor. Neurosci. Lett., 241: 135-138. Lakie, M, Walsh, EG and Wright, GW (1984) Resonance at the wrist demonstrated by the use of a torque motor:

an instrumental analysis of muscle tone in man. J. Physiol. (Lond.), 353: 265-285. Lakie, M, Walsh, EG and Wright, GW (1986) Passive mechanical properties of the wrist and physiological tremor. J. Neurol. Neurosurg. Psychiatry, 49, 669-676. Lance, JW, Schwab, RS and Peterson, EA (1963) Action tremor and the cogwheel phenomenon in Parkinson's disease. Brain, 86: 95-110. Logigian, EL, Wierzbicka, MM, Bruyninckx, F, Wiegner, AW, Shahani, BT and Young, RR (1988) Motor unit synchronization in physiologic, enhanced physiologic and voluntary tremor in man. Ann. Neurol., 23: 242250. Ludlow, CL, Bassich, CJ, Connor, NP and Coulter, DC (1986) Phonatory characteristics of vocal fold tremor. J. Phonetics, 14: 509-515. Marsden, CD, Meadows, JC, Lange, GW and Watson, RS (1969) The role of the ballistocardiac impulse in the genesis of physiological tremor. Brain, 92: 647-662. McAuley, JH, Farmer, SF, Rothwell, JC and Marsden, CD (1999a) Common 3 and 10 Hz oscillations modulate human eye and finger movements while they simultaneously track a visual target. J. Physiol. (Lond.), 515: 905-917. McAuley, JH, Rothwell, JC and Marsden, CD (1999b) Human anticipatory eye movements may reflect rhythmic central nervous system activity. Neuroscience, 94: 339-350. Milner, TE and Cloutier, C (1998) Damping of the wrist joint during voluntary movement. Exp. Brain Res., 122: 309-317. Qureshi, F, Morales, A and Elble, RJ (1996) Tremor due to infarction in the ventrolateral thalamus. Mov. Disord., 11: 440-444. Raethjen, J, Lindemann, M, Dumpelmann, M, Stolze, H, Wenzelburger, R, Pfister, G, Elger, CE, Timmer, J and Deuschl, G (2000a) Cortical correlates of physiologic tremor. Mov. Disord., 15 (Suppl, 3): 90-91. Raethjen, J, Pawlas, F, Lindemann, M, Wenzelburger, R and Deuschl, G (200Gb) Determinants of physiologic tremor in a large normal population. Clin. Neurophysiol., Ill: 1825-1837. Raethjen, J, Lemke, MR, Lindemann, M, Wenzelburger, R, Krack, P and Deuschl, G (2001) Amitriptyline enhances the central component of physiological tremor. J. Neurol. Neurosurg. Psychiatry, 70: 78-82. Salenius, S, Portin, K, Kajola, M, Salmelin, R and Hari, R (1997) Cortical control of human motoneuron firing during isometric contraction. J. Neurophysiol., 77: 3401-3405. Sanes, IN (1985) Absence of enhanced physiological tremor in patients without muscle or cutaneous afferents. J. Neurol. Neurosurg. Psychiatry, 48: 645-649.

364 Schafer, EA, Canney, EL and Thnstall, JO (1886) On the rhythm of muscular response to volitional impulses in man. J. Physiol. (Lond.), 7: 1I1-117. Sears, TA and Stagg, D (1976) Short-term synchronization of intercostal motoneuron activity. 1. Physiol. (Lond.), 263: 357-381. Semmler, JG and Nordstrom, MA (1998) Motor unit discharge and force tremor in skill- and strengthtrained individuals. Exp. Brain Res., 119: 27-38. Spauschus, A, Marsden, J, Halliday, DM, Rosenberg, JR and Brown, P (1999) The origin of ocular microtremor in man. Exp. Brain Res., 126: 556-562. Stein, RB and Oguztoreli, MN (1976) Tremor and other oscillations in neuromuscular systems. Biol. Cybem., 22: 147-157. Stephens, JA and Taylor, A (1974) The effect of visual feedback on physiological muscle tremor. Electroencephalogr. Clin. Neurophysiol., 36: 457-464. Stiles, RN (1976) Frequency and displacement amplitude relations for normal hand tremor. J. Appl. Physiol., 40: 44-54. Stiles, RN (1980) Mechanical and neural feedback factors in postural hand tremor of normal subjects. J. Neurophysiol., 44: 40--59.

RJ. ELBLE

Stiles, RN and Hahs, DW (1991) Muscle-load oscillations: detection, analysis and models. In: DL Wise (Ed.), Bioinstrumentation and Biosensors. Marcel Dekker, Inc., New York, pp. 75-119. Stiles, RN and Randall, JE (1967) Mechanical factors in human tremor frequency. 1. Appl. Physiol., 23: 324-330. Timmer, J, Lauk, M, Pfleger, W and Deuschl, G (1998) Cross-spectral analysis of physiologic tremor and muscle activity. I. Theory and application to unsynchronized electromyogram. Biol. Cybern., 78: 349-357. Wessberg, J and Vallbo, AB (1996) Pulsatile motor output in human finger movements is not dependent on the stretch reflex. J. Physiol. (Lond.), 493: 895-908. Wessberg, J and Kakuda, N (1999) Single motor unit activity in relation to pulsatile motor output in human finger movements. J. Physiol. (Lond.), 517: 273-285. Winholtz, WS and Ramig, LO (1992) Vocal tremor analysis with the vocal demodulator. J. Speech Hear. Res., 35: 562-573. Zahalak, GI and Pramod, R (1985) Myoelectric response of human triceps brachii to displacement-controlled oscillations of the forearm. Exp. Brain Res., 58: 305317.

Movement Disorders Handbook of Clinical Neurophysiology, Vol. I M. Hallett (Ed.) © 2003 Elsevier B.V. All rights reserved

365

CHAPTER 23

Essential tremor and primary writing tremor Peter G. Bain* Imperial College School ofMedicine. Charing Cross Hospital Campus, Fulham Palace Road, London W68RF, UK

23.1. Essential tremor

23.1.2, Clinical features

23.1.1. Definition

Essential tremor is classified as being either hereditary or sporadic. Different studies have given the overall population prevalence of essential tremor to be between 0.08 and 220 cases per 1000 people (Louis et al., 1998). The wide variations in these figures are largely the result of differences in the observed populations and the diagnostic criteria deployed in each study. If these are standardized the prevalence estimates vary from 4.1 to 39.2 cases per 1000 (Louis et al., 1998). However, two significant diagnostic difficulties remain: firstly, distinguishing between an enhanced physiological tremor and the early stages of essential tremor, as currently there is no accepted way of making this distinction; and secondly, dissecting out essential tremor from other causes of action tremor that can mimic the condition. Essential tremor has a bimodally distributed age of onset with a median age of about 15 years. The sexes are affected with equal frequency and severity. Tremor provoked disability typically begins in the second decade and increases with age and tremor duration (Koller et al., 1986; Bain et al., 1994). The upper limbs are affected first and then in a proportion of cases tremor can spread to affect the legs, head, facial muscles, voice and tongue (Bain et aI.,1994).

In a consensus statement of the Movement Disorder Society on Tremor (Deuschl et al., 1998) the following clinical criteria were developed to make the diagnosis of classical essential tremor: Inclusion criteria: (i) Bilateral, largely symmetrical postural or kinetic

tremor involving hands and forearms that is visible and persistent. (ii) Additional or isolated tremor of the head may occur but in the absence of abnormal posturing. (This item of the inclusion criteria is controversial). Exclusion criteria: (i) Other abnormal neurologic signs, especially

dystonia. (ii) The presence of known causes of enhanced

(iii) (iv) (v) (vi) (vii) (viii) (ix)

physiological tremor, including current or recent exposure to tremorgenic drugs or the presence of a drug withdrawal state. Historic or clinical evidence of psychogenic tremor. Convincing evidence of sudden onset or evidence of stepwise deterioration of tremor. Primary orthostatic tremor. Isolated voice tremor. Isolated position-specific or task-specific tremors, including occupational tremors and primary writing tremor. Isolated tongue or chin tremor. Isolated leg tremor.

* Correspondence to: Dr. Peter G. Bain, Imperial College School of Medicine, Charing Cross Hospital Campus, Fulham Palace Road, London W6 8RF, UK. E-mail address:[email protected] Tel.: 020-8846-1182; fax: 020-8846-7487.

23.1.3. Response to alcohol and other treatments

About 50% of people with essential tremor respond to alcoholic drinks. However, the tremor invariably rebounds about 3-4 hours later in a temporarily exacerbated form (Critchley, 1949; Bain et al., 1994). Even so moderate use of alcohol can facilitate feeding and social interaction. Several controlled trials have established propranolol and primidone or a combination of both as the mainstays of treatment of essential tremor, as these drugs

366

reduce the amplitude of essential tremor. However, some disability remains and patients are often troubled by side effects (Bain, 1997). Several other medications have been advocated as treatments for essential tremor, including other ~-receptor blockers, clonazepam, alprazolam, acetazolamide, amantadine, benzhexol, trazadone, phenoxybenzamine, progabide, theophylline, clozapine, verapamil, nifedipine, nicardipine and botulinum toxin, but of these only clonazepam is widely deployed as a second line treatment, particularly for the predominantly kinetic form of essential tremor (Biary and Koller, 1987; Bain, 1997). Stereotactic surgery can be an effective treatment for patients with severe essential tremor but has inherent risks (Bain, 1997). The conventional cerebral target for alleviating contralateral essential tremor is the ventralis intermedius nucleus of the thalamus (Pahwa et al., 2000), although recently ventralis oralis posterior and zona incerta have been targeted to good effect (Aziz and Bain, personal communication). The issues of the comparative safety and efficacy of unilateral thalamotomy vs. chronic thalamic stimulation were addressed in a randomized controlled trial conducted by Schuurman et al. (2000). The results of this study showed that, six months after surgery, both techniques were equally effective at suppressing essential tremor but stimulation caused a significantly greater improvement in the patients' capacities to perform their activities of daily living and was associated with fewer adverse events. The efficacy of thalamic stimulation over a one to two-year period has also been established (Koller et aI., 1997; Limousin et al., 1999). Bilateral deep brain stimulation is considered to be preferable to bilateral thalamotomies because of the high incidence of dysarthria associated with the latter (Koller et al., 2000). The short term results obtained by gamma knife thalamotomy, which has been performed on a small number of essential tremor patients who could not undergo invasive surgery, indicate that essential tremor can be alleviated by this technique. However, serious concerns about the long term consequences of this treatment remain (Niranjan et al., 2000). 23.1.4. Genetics

The gene or genes responsible for hereditary essential tremor have not been isolated. Two studies

P.G. BAIN

have excluded linkage between familial essential tremor and the DYTI dystonia locus (Conway et aI., 1993; Durr et al., 1993). Linkage with a locus ('FETl') on chromosome 3q13 has been obtained, with a LaD score of 3.71, in an Icelandic study (Gulcher et aI., 1997) and Higgins et aI. (1997) obtained linkage with a locus ('ETM') on chromosome 2p22-25 (LaD score: 5.92). These results indicate that essential tremor may have heterogeneous genetic causes. 23.1.5. Pathophysiology

Standard neuropathological techniques have failed to demonstrate any consistent physical or biochemical changes in brains obtained from patients in whom essential tremor had been present (Rajput et aI., 1991). However, it has been reported that lesions in the nucleus ventralis intermedius of the thalamus, internal capsule, and pons abolished contralateral and a cerebellar lesion abolished ipsilateral essential tremor (Hirai et aI., 1983; Larsen and Calne, 1983; Young, 1986; Duncan et al., 1988; Dupuis et aI., 1989; Nagaratnam and Kalasabail, 1997). Spectral analysis of moderate-severe essential tremor shows a clear spectral peak in both accelerometric and EMG records, although multiple peaks may be present in the spectra obtained from patients with mild essential tremor (Bain et al., 1993). However, the frequency band of essential tremor (::=4-12 Hz) is not diagnostic, as it overlaps with that of many other types of tremor (Deuschl et al., 1998) and like normal physiologic tremor, decreases with advancing age (Marshall, 1962). It has no relationship with the length of the reflex arc (Elble, 1995). The amplitude of essential tremor, recorded at the wrist, has a logarithmic relationship with the tremor frequency (Elble, 1986). Although, changes in the amplitude and frequency of essential tremor have been detected during the performance of different tasks, for example whilst holding a cupful of water and during a joy-stick controlled tracking test, compared to a maintained posture (Bain et al., 1993). Studies involving spring and peripheral mass loading have demonstrated that essential tremor consists of two components, namely a mechanical reflex and a stable frequency component, the latter being associated with abnormal entrainment of motor unit discharges (Elble, 1986, 1995). The

ESSENTIAL TREMOR AND PRIMARY WRITING TREMOR

No Added Mass -

"'-

Tremor

--_. EMG

3000

500

"u

-III III

-... E

400

2000

300

U

Gl ~

I I

I

j'

o

D-

200

II

1000 0

a

2

4

6

100

I\. 10

8

'>

-... ~

::1

GI

~ 0

D-

a

12

Frequency (Hz)

500 gm Added Mass -

" "u

Tremor

--_. EMG

400

2000

~

III

300

E u

200

III

-... CII ~ 0

D-

ET fI

1600 1200

" I

100

j

MR

~~

O

0

2

4

6

800

\..B

400

to

~

-...

">::1 II

~ 0 ~

0 12

Frequency (Hz)

Fig. 1. Wrist tremor measured by accelerometry and extensor carpi radialis brevis EMG autospectra recorded from a patient with essential tremor before and during 500 g mass loading. Upper figure: Note that the 8 Hz essential tremor is associated with a prominent spectral peak in the carpi radialis brevis EMG spectrum, indicating motor unit entrainment. Lower figure: Adding 500 mg causes a mechanical reflex frequency component (at about 4.1 Hz) to separate from a frequency stable component (at about 8 Hz). It is this stable-frequency component that distinguishes essential tremor from normal physiological tremor (reproduced from Elble RJ, 1986).

mechanical reflex component of essential tremor is similar to that detected in normal physiological tremor (Elble, 1986). Bursts of EMG activity separated by relative silence occur in ail types of pathological and enhanced physiological tremor but not in normal, low amplitude, physiological tremor (Rothwell et al.,

367

1987). The relationship of the EMG bursts in agonist/antagonist muscle pairs is variable in patients with essential tremor and recordings from the same muscle pair may even vary within an individual patient. The following patterns have all been recorded from the forearm flexor and extensor muscle pairs: co-contraction, alternating, switching from alternating to co-contraction and agonist (antigravity activation alone) as well as other more diffuse patterns (Shahani and Young, 1976; Deuschl et aI., 1987; Rothwell et aI., 1987; Koller et al., 1992; Spieker et aI., 1998). Single motor unit discharges recorded from muscles of normal subjects show that a moderate tonic contraction produces motor unit discharge rates of about 10-15 Hz. In patients with essential tremor the motor units fire in bursts of up to 50 Hz, which are grouped into the active period of each tremor cycle and there is also another tendency for discharges from separate motor units to synchronize over a shorter time period. Single motor unit studies have shown that the recruitment order of units within each tremor cycle is disrupted in essential tremor (Rothwell et al., 1987). Two studies of simple voluntary ballistic wrist movements show conflicting results in patients with essential tremor: in one (Britton et aI., 1994) a delayed second agonist burst was found in the triphasic angonist-antagonist-agonist EMG pattern but this aberration was not detected in another study (Elble et al., 1994). Subsequently Deuschl et al. (2000) used an infra-red camera system to make quantitative assessments of a grasping movement, which was performed by: (a) two subgroups of essential tremor patients, one with predominant postural tremor and the other with predominant intention tremor; (b) healthy control subjects; and (c) patients with cerebellar disease. Analysis of the resulting data indicated that the amplitude of intention tremor was similar in essential tremor to that detected in cerebellar disease. Moreover, hypermetria and a slowed deceleration phase were found in the kinematic profiles of the intention-essential tremor subgroup as well as the patients with cerebellar disease, suggesting that there is an abnormality of cerebellar function in essential tremor (Deuschl et al., 2000). Essential tremor can be distinguished from the action tremor resulting from benign IgM paraproteinemic neuropathy as peripheral nerve con-

368

duction velocities are normal in the former but are markedly reduced 00-30 mls) in the latter (Smith et al., 1984; Bain et al., 1996). Various peripheral stimuli have been used to alter the phase of ongoing essential tremor activity relative to the tremor's phase prior to the disturbance. Initially developed as a method of determining how susceptible a tremor generator is to peripheral inputs (mechanical wrist perturbations or median nerve shocks) the technique has stumbled on methodological problems that have led to conflicting reports (Lee and Stein, 1981; Marsden, 1984; E1ble et al., 1987; Britton et al., 1992). One reason for this is that mechanical perturbations of the wrist produce transient mechanico-reflex oscillations that can be mistaken for the underlying essential tremor rhythm. Attempts to overcome this flaw by using sinusoidal forcings applied to the wrist showed that essential tremor could be entrained within a narrow frequency band (Elble et al., 1992). This approach has been further developed using magnetic cortical or electrical stimulation; the former but not the latter reset essential tremor (Britton et al., 1993; Pascual-Leone et al., 1994). Normal cortical excitability, motor latencies and thresholds were detected in essential tremor using the double pulse transcranial magnetic stimulation technique (Romeo et al., 1998). The forearm short and long latency stretch reflexes are of normal size, latency and duration in patients with essential tremor but afterwards rebound oscillations occur which are said to be underdamped (Rothwell et al., 1987). These rebound oscillations were thought to be characteristic of essential tremor but have since been observed in other types of action tremor (Bain, 1993). No abnormalities have been found in studies of reciprocal inhibition of the forearm H-reflex in patients with essential tremor (Panizza et al., 1990). Coherence analysis has been deployed for studying the pathophysiology of essential tremor by examining the relationship between brain activity, recorded using magnetoencephalopathy (MEG) or electroencephalopathy (EEG), and tremorgenic muscle bursts detected by electromyography (EMG) (Halliday et al., 2000; Hellwig et al., 2001). However, the results have been conflicting as Halliday et al. (2000) failed to find a relationship between MEG and EMG, whilst Hellwig et al. (2001) detected significant corticomuscular coherences between EEG and EMG signals at tremor

P.G. BAIN

frequency, with maximum coherences being located over the contralateral sensorimotor cortex. Coherence analysis has also been used to examine the relationship between tremorgenic muscle activities in different muscles within the same limb and also muscles in different limbs (Raethjen et al., 2000). Coherence between tremor-related activity in different muscles within the same limb was found but not from signals derived from muscles from different extremities; indicating that essential tremor is driven by multiple oscillators. Single cell recordings obtained from patients with essential tremor whilst stereotactic thalamic surgery was performed revealed coherent activity between thalamic cell bursts and tremor signals obtained from patients' limbs (Hua et al., 1998).

23.1.6. Functional imaging Positron emission tomography (PET) studies have demonstrated that the rate of 18F-DOPA uptake by the basal ganglia is in the low normal range in hereditary essential tremor patients and usually normal in sporadic cases, whereas in Parkinson's disease the rate of 18F_DOPA uptake by the putamen is asymmetrically diminished (Leenders et al., 1989; Brooks et al., 1992a). Furthermore, striatal dopamine transporter function was found to be normal using 123I-I3-CIT-SPECT in essential tremor, indicating that dopamine dysfunction is not relevant to the majority of essential tremor cases (Asenbaum et al., 1996). A PET study examining resting cerebral 18F_ 2-deoxyglucose utilization in essential tremor showed medullary hypermetabolism which was thought to be indicative of over-activity of the inferior olive (Dubinsky and Hallett, 1987), but these data await confirmation. Studies of regional cerebral blood flow (rCBF) in essential tremor patient cohorts demonstrated that there is overactivity of bilateral cerebellar circuitry whilst tremor is occurring and even during rest in essential tremor patients compared to normal subjects (Colebatch et al., 1990; Brooks et al., 1992b; Zeffiro et al., 1992; Jenkins et al., 1993; Wills et al., 1994, 1995; Boecker et al., 1996). The results of these studies also suggest that the bilateral cerebellar activation evident in essential tremor was not caused by peripheral sensory feedback and that bilateral red nucleus and thalamic activation is associated with essential tremor (Wills

ESSENTIAL TREMOR AND PRIMARY WRITING TREMOR

369

et al., 1994; Boecker et aI., 1996). No excess olivary activity was found in these studies, possibly because the inferior olive has few interneurons and thus synaptic activity (Boecker and Brooks, 1998). A study using functional magnetic resonance imaging and single-subject analysis confirmed that rubral activation was increased in essential tremor (Bucher et aI., 1997). As drugs that potentiate GABA transmission can alleviate essential tremor the finding of increased bilateral ventrolateral thalamic [11C]flumazenil binding to central GABAA-receptors using PET in essential tremor is interesting (Boecker and Brooks, 1998). Furthermore, ethanol, which potentiates GABA transmission, reduced cerebellar overactivity in patients with alcohol responsive essential tremor but not in alcohol unresponsive cases (Boecker et aI., 1996; Boecker and Brooks, 1998).

concerned a young man who presented with difficulty writing, that resulted from bursts of tremor that were evoked whenever his right forearm was pronated. Tendon taps to the volar aspect of the patient's forearm would also induce tremor bursts. The authors concluded that the patient's tremor resulted from an abnormal response to the spindle input arising from pronator teres because his writing difficulty and tremor were temporarily abolished by partial motor point anesthesia of this muscle. Subsequently a number of patients with similar symptoms have been documented (Klawans et al., 1982; Ohye et aI., 1982; Kachi et al., 1985; Ravits et aI., 1985; Koller and Martyn, 1986; Cohen et aI., 1987; Rosenbaum and Jankovic, 1988; Elble et al., 1990; Kim and Lee, 1994; Bain et aI., 1995; Jimenez-Jimenez et aI., 1998; Minguez-Castellanos et aI., 1999; Rotella et al., 1999; Berg et aI., 2000; Racette et al., 2001).

231.7. Conclusion Presently the gene and physiological fault responsible for essential tremor have not been precisely determined, although evidence from physiological and functional imaging experiments implicate the cerebellum in the genesis of this form of action tremor. However, the cerebellum may not be the origin of essential tremor, which clearly involves more diffuse circuitry in its expression. f3-receptor blocking and GABA transmission facilitating drugs (namely alcohol and primidone) reduce the severity of essential tremor, whilst lesions or high frequency stimulation of the ventralis intermedius or ventralis oralis posterior nuclei of the thalamus or the zona incerta can abolish or greatly attenuate this common condition.

23.2. Primary writing tremor 23.2.1. Definition In a consensus statement of the Movement Disorder Society on Tremor (Deuschi et al., 1998) primary writing tremor was defined as 'being present when tremor occurs only or predominantly during writing but not during other tasks in the active hand'.

23.2.2. Introduction Primary writing tremor was first described by Rothwell et aI. in 1979. The initial description

23.2.3. Clinical features Bain et aI. (1995) studied the clinical and neurophysiological features of 21 patients (20 males and 1 female) with primary writing tremor and found a degree of non-specific action tremor of the upper limbs in most cases, which was of smaller magnitude than the writing tremor. In "" 20% of cases a preceding history of trauma to the dominant arm was obtained, although there was a delay between trauma and the appearance of tremor that varied from 18 months to 12 years. Patients with primary writing tremor were found to have significantly reduced dominant hand writing speeds (letters per minute) compared to those of healthy control subjects but their speed of writing with the non-dominant hand was within normal limits. Primary writing tremor may occur sporadically or be inherited as an autosomal dominant trait and a family history of primary writing tremor can be obtained from about one third of the patients (Rothwell et aI., 1979; Klawans et al., 1982; Ohye et al., 1982; Kachi et aI., 1985; Ravits et aI., 1985; Koller and Martyn, 1986; Cohen et aI., 1987; Rosenbaum and Jankovic, 1988; Elble et aI., 1990; Bain et aI., 1995). The mean age at primary writing tremor onset is 50.1 years (Bain et aI., 1995), which is between 15 and 20 years later than that reported for writer's cramp (Sheehy and Marsden, 1982; Sheehy et al.,

370

1988; Nakashima et aI., 1989; Waddy et al., 1991), which is also a task-specific disorder of handwriting. It is also significantly older than the mean age of onset of hereditary idiopathic torsion dystonia (Fletcher et aI., 1990) and hereditary essential tremor (Bain et al., 1994). Patients with primary writing tremor have been classified into two types depending on whether tremor appeared during writing (type A: task induced tremor) or whilst writing and also on adopting the hand position normally used for writing (type B: positionally sensitive tremor). However, no significant differences between these two primary writing tremor sub-groups with regard to age, age at tremor onset or duration of tremor was found and the condition is usually non-progressive (Bain et al., 1995). Primary writing tremor produces sufficient difficulty with writing that in most instances (voluntary) adaptive techniques are employed. However, only ::05% of patients give up trying to write and in ::0 10% employment status is affected. Curiously, only few patients switch themselves to writing with the nondominant hand (Bain et al., 1995). 23.2.4. Etiology

The etiology of primary writing tremor is controversial (Fahn, 1984; Sheehy et al., 1988; Elble and Koller, 1990; Bain et al., 1995) and opinion is divided into two main schools of thought: (i) that primary writing tremor is a variant of essential tremor (Ohye et al., 1982; Kachi et al., 1985; Koller and Martyn, 1986); or (ii) that it is a type of focal dystonia, a variant of writer's cramp (Ravits et al., 1985; Cohen et aI., 1987; Elble et al., 1990). However, it is possible that primary writing tremor is a different entity from both focal dystonia and essential tremor (Bain et al., 1995; Berg et aI., 2000). Sheehy et aI. (1988) and Rosenbaum and Jankovic (1988) have suggested a scheme in which some cases of primary writing tremor were related to essential tremor and others to torsion dystonia. The evidence supporting the hypothesis that primary writing tremor is a variant of essential tremor is that both types of tremor have similar frequencies (between 4 and 8 Hz) and can be relieved (in about 30% and 50% of cases respectively) by moderate amounts of alcohol (Ohye et al., 1982; Kachi et aI., 1985; Koller and Martyn, 1986; Bain et aI., 1994, 1995). However, the unilateral

P.G. BAIN

nature of primary writing tremor, its tendency to remain focal and to appear rather than suppress during skilled manual tasks argue strongly against primary writing tremor being a variant of hereditary essential tremor; as the latter typically affects both hands, becomes increasingly severe with time, often spreads to other parts of the body, and can usually be suppressed to some extent during skilled tasks (Bain et al., 1993). Furthermore, not one example of primary writing tremor was detected in a detailed observational study of 20 families with essential tremor (Bain et al., 1994). The view that PWT is a type of writer's cramp is supported by the observation that both conditions are task specific. Furthermore, in some cases of primary writing tremor both tremor and abnormal posturing are apparent (Sheehy and Marsden, 1982; Sheehy et aI., 1988; Elble et aI., 1990). In addition tremor without any other movement disorder can be one of the manifestations of idiopathic torsion dystonia (Bundey et al., 1975; Fletcher et al., 1990, 1991) and one family has been described in which cases of writer's cramp, writing tremor and non-task specific tremor were all noted (Cohen et aI., 1987). However, this family were subsequently found to have a spinocerebellar atrophy (SCA) mutation (Hallett, personal communication). Furthermore, classification difficulties arise because some tremulous patients voluntarily employ unusual postures and excessive force to control the pen and thus appear to be dystonic. In addition, not one case of primary writing tremor was documented in several genetic studies of focal or other forms of idiopathic torsion dystonia (Bundey et aI., 1975; Forsgren et al., 1988; Fletcher et al., 1990; Waddy et al., 1991). Kim and Lee (1994) described the development of writing tremor in a 67 year old man after a discrete left frontal infarct. The patient had mild essential tremor preceeding the stroke and developed writing tremor after recovering from a right hemiparesis. 23.2.5. Effect of alcohol and other treatments

There have been no formal therapeutic trials involving patients with primary writing tremor. However, alcohol greatly improves or abolishes writing tremor in ::033% of patients but ::0 15% express concern about their own excessive alcohol consumption (Bain et aI., 1995). Anecdotal evidence suggests that ::0 80% of patients with PWT obtain

ESSENTIAL TREMOR AND PRIMARY WRITING TREMOR

371

some benefit from treatment: propranolol, primidone and anti-cholinergic drugs being the most commonly prescribed (Bain et aI., 1995). Intramuscular botulinum toxin injected into the forearm flexor and extensor muscles has also been reported to benefit some cases (Bain et aI., 1995). In appropriate cases contralateral stereotactic ventralis intermedius thalamotomy and thalamic stimulation can alleviate primary writing tremor (Ohye et al., 1982; MinguezCastellanos et aI., 1999; Racette et aI., 2001).

primary writing tremor (Kachi et aI., 1985; Ravits et aI., 1985; Elble et aI., 1990; Bain et., 1995), although skipping between an alternating and a co-contracting pattern or alternating and extensor activation only, extensor only activation and co-contraction have also been observed (Kachi et aI., 1985; Bain et aI., 1995). Bain et aI. (1995) could not detect any rhythmic EMG activity from any of their type A primary writing tremor patients when the arms were at rest, held outstretched or placed in a writing posture. Rhythmic segmentation of the EMG could be recorded from all the type B patients when the dominant hand was placed in a writing posture or partially pronated to a critical angle (positionally sensitive tremor), but in these postures the tremor frequency was usually similar to that of their standard postural tremor (i.e. above that of their writing tremor). The only other differences between these two groups were that a co-contracting EMG pattern and tremor induced by tendon taps to the volar aspect of the wrist were present in type B but not type A cases. The amplitude and latencies of the forearm stretch reflexes were normal in five patients with primary writing tremor described by Kachi et aI. (1985) but the authors had difficulty reconciling this observation with the fact that tendon taps and muscle stretches caused tremor to appear in two and six of their nine patients, respectively. Ravits et al. (1985) recorded 5-20 I-LV cerebral potentials, which could be elicited by stretching pronator teres, and also noted that C-reflexes were absent. The time course of reciprocal inhibition of the forearm median nerve H-reflex has been clearly defined for normal individuals (Day et aI., 1984) and is known to be abnormal in patients with writer's cramp (Nakashima et aI., 1989). In normal subjects, it has been shown that the time course of inhibition of the median nerve H-reflex produced by radial nerve shocks has three distinct inhibitory phases, at conditioning-test intervals of about -3 to 1, 5 to 50 (maximal at 15-20) and 50 to at least 100 ms respectively. The first (-3 to 1 ms) of these is compatible with Ia disynaptic inhibition of flexor motomeurons produced by activation of large diameter afferents from extensor muscles. The second (5 to 50 ms) is thought to reflect presynaptic inhibition of the terminals of flexor Ia afferent fibers whilst the cause of the third inhibitory period (50 to at least 100

23.2.6. Pathophysiology In normal handwriting epochs of rhythmic EMG activity (at about 4-7.7 Hz) are seen in the intrinsic hand muscles and wrist extensors of the dominant arm and can also be evident in the biceps and triceps muscles (Bain et aI., 1995). Tremor frequencies of 5-7 Hz have consistently been found in patients with primary writing tremor (Ohye et aI., 1982; Kachi et aI., 1985; Ravits et aI., 1985; Elble et aI., 1990; Bain et aI., 1995), the only exception being the original patient described by Rothwell et al. in 1979 in whom a 4-6 Hz tremor occurred in biceps, supinator and pronator teres whilst triceps fired at about 10 Hz. Accelerometric studies show that the magnitudes of the standard postural tremors in either hand or the writing tremor/oscillation in the dominant hand were similar for a group of PWT patients compared to those of healthy control subjects (Bain et aI., 1995). The frequencies of the dominant hand writing tremor/oscillation or non-dominant standard postural tremor were also found to be similar. The frequency of primary writing tremor patients' postural tremor on the dominant side was significantly lower than that of the control group. In three primary writing tremor patients, tremor appeared solely in the dominant hand when writing with the non-dominant hand (Bain et aI., 1995). A similar phenomenon, called mirror dystonia, has been described in writer's cramp (Sheehy and Marsden, 1982). During writing with the dominant hand, rhythmic EMG is present in comparable muscle groups of PWT patients and control subjects; indicating that there is no excessive "overflow" of this rhythmic EMG activity in primary writing tremor, although this activity is more sustained in this condition (Bain et aI., 1995). EMG bursts that alternate between the forearm agonist/antagonist muscles is typical of

372

P.G. BAIN

RECIPROCAL INHIBITION

w

C/)

Z

o

a..

C/)

~ ...J o a::

IZ

o o LL o W o ~

Z W

o a:: w

a..

Fig. 2. Inhibition of the forearm flexor H-reflex when tested at different times after a single motor threshold conditioning stimulus (at t=O ms) given to the radial nerve in the spiral groove. Negative timings indicate that the median nerve test shock (used to elicit flexor H-reflexes) was given before the conditioning (radial nerve) shock. Mean (+ I SEM) values of the size of the flexor H-reflex, expressed as a percentage of the control If-reflexes, are plotted. The results from 10 normal control subjects, 13 patients with PWT and 16 patients with writer's cramp (We) are shown (reproduced from Bain et al.,

1995).

ms) is not presently understood (Day et al., 1984; Berardelli et al., 1987; Nakashima et al., 1989). The reciprocal inhibition curve for 13 patients with primary writing tremor was found to be normal in every respect and no significant differences between the curves obtained from type A and B primary writing tremor patients were detected (Bain et al., 1995). However, it should be noted that patients with tremulous writers' cramp were not included in this study. In 1982 Ohye et al. recorded a very high incidence of irregular burst discharges within the thalamus and showed that contralateral stereotactic ventralis intermedius thalamotomy could successfully abolish primary writing tremor. 23.2.7. Functional imaging studies

A single PET study demonstrated increased bilateral cerebellar regional cerebral blood flow in

six patients with primary writing tremor, whilst the patients held pen to paper but did not actually write (Wills et al., 1995). In addition three patients with primary writing tremor were examined using functional MRI involving a paradigm that consisted of alternating 30-s periods of rest or writing; the results were compared to those of healthy control subjects (Berg et al., 2000). Both primary writing tremor patients and healthy control subjects were found to have significant activation of the contralateral primary sensorimotor cortex, supplementary motor area and area 44. In addition, patients with primary writing tremor were found to have activation of the contralateral premotor area and ipsilateral prefrontal area, bilateral parietal lobule and bilateral cerebellum, the latter being more pronounced on the ipsilateral side. Consequently, the view that primary writing tremor is distinct from both essential tremor and writer's cramp is supported by these findings, as the cerebral activation patterns found in primary writing tremor integrate hallmarks

ESSENTIAL TREMOR AND PRIMARY WRITING TREMOR

373

of those found in the latter two conditions (Berg et al., 2000).

CD (1996) Tremor associated with benign IgM paraproteinaemic neuropathy. Brain, 119: 789-799. Berg, 0, Preibisch, C, Hofmann, E and Naumann, M (2000) Cerebral activation pattern in primary writing tremor. J. Neurol. Neurosurg. Psychiatry, 69: 780786. Berardelli, A, Day, BL, Marsden, CD and Rothwell, JC (1987) Evidence favouring presynaptic inhibition between antagonist muscle afferents in the human forearm. J. Physiol., 391: 71-83. Biary, N and Koller, WC (1987) Kinetic predominant essential tremor: successful treatment with clonazepam. Neurology, 37: 471-474. Boecker, H and Brooks, OJ (1998) Functional imaging of tremor. Mov. Disord., 13 (Suppl. 3): 64-72. Boecker, H, Wills, AJ, Ceballos-Baumann,A et al. (1996) The effect of ethanol on alcohol-responsive essential tremor: a positron emission tomography study. Ann. Neurol., 39: 650-658. Britton, TC, Thompson, PO, Day, BL, Rothwell, JC and Findley, U (1992) "Resetting" of postural tremors at the wrist with mechanical stretches in Parkinson's disease, essential tremor and normal subjects mimicking tremor. Ann. Neurol., 31: 507-514. Britton, TC, Thompson, PO, Day, BL, Rothwell, JC, Findley, U and Marsden, CD (1993) Modulation of postural wrist tremors by magnetic stimulation of the motor cortex in patients with Parkinson's disease, essential tremor, and in normal subjects mimicking tremor. Ann. Neural., 33: 473-479. Britton, TC, Thompson, PO, Day, BL et al. (1994) Rapid wrist movements in patients with essential tremor. The critical role of the second agonist burst. Brain, 117: 39-47. Brooks, OJ, Playford, ED, Ibanez, V, Sawle, GV, Thompson, PO, Findley,U and Marsden, CD (1992a) Isolated tremor and disruption of the nigrostriatal doparninergic system: an 18F-DOPA PET study. Neurology, 42: 1554-1560. Brooks, OJ, Jenkins, IH, Bain, PG, Colebatch, JG, Thompson, PD, Findley, U and Marsden, CD (1992b) A comparison of the abnormal pattern of cerebral activation associated with neuropathic and essential tremor. Neurology, 42 (Suppl. 3): 423. Bucher, SF, Seelos, KC, Dodel, RC, Reiser, M and Oertel, WH (1997) Activation mapping in essential tremor with functional magnetic resonance imaging. Arch. Neural., 52: 299-305. Bundey, S, Harrison, MJG and Marsden, CD (1975) A genetic study of torsion dystonia. J. Med. Genetics, 12: 12-19. Cohen, LG and Hallett, MD (1988). Hand cramps: clinical features and electromyographic patterns in a focal dystonia. Neurology, 38: 1005-1011.

23.2.8. Conclusion In conclusion, there are clinical, physiological and functional imaging differences between primary writing tremor and both writer's cramp and essential tremor. The results of physiological studies have demonstrated two specific differences between primary writing tremor and writer's cramp: Firstly, during writing in PWT there is no evidence of excessive overflow of EMG activity into the proximal musculature (Bain et al., 1995), whereas overflow is characteristic of writer's cramp (Cohen and Hallett, 1988; Hughes and Mclellan, 1989). Secondly, reciprocal inhibition is entirely normal in primary writing tremor, whereas in writer's cramp the inhibitory effect of ipsilateral radial nerve stimulation on the median nerve H-reflex is diminished when the radial nerve is stimulated between 5 and 50 ms before the median nerve (Nakashima et al., 1989; Panizza et al., 1990). There have been no post-mortem studies involving patients with PWT.

References

Asenbaum, S, Brucke, T, Pirker, W, Muller, C, Wober, C and Podreka, I (1996) 12313-CIT-SPECT in Parkinson's disease and essential tremor. Bain, P (1993) A combined clinical and neurophysiological approach to the study of patients with tremor. J. Neurol. Neurosurg. Psychiatry, 56: 839-844. Bain, PG (1997) The effectiveness of treatments for essential tremor. The Neurologist, 3: 305-321. Bain, PG, Mally, J, Gresty, MA and Findley, U (1993) Assessing the impact of essential tremor on upper limb function. J. Neurol., 241: 54-61. Bain, PG, Findley, U, Thompson, PO, Gresty, MA, Rothwell, JC, Harding, AE and Marsden, CD (1994) A study of hereditary essential tremor. Brain, 117: 805824. Bain, PG, Findley, U, Britton, TC, Rothwell, JC, Gresty, MA, Thompson, PO and Marsden, CD (1995) Primary writing tremor. Brain, 118: 1461-1472. Bain, PG, Britton, TC, Jenkins, IH, Thompson, PO, Rothwell, JC, Thomas, PK, Brooks, OJ and Marsden,

374 Cohen, LG, Hallett, M and Sudarsky, L (1987) A single family with writer's cramp, essential tremor and primary writing tremor. Mov. Disord., 2: 109-116. Colebatch, JG, Findley, LJ, Frackowiak, RSJ, Marsden, CD and Brooks, DJ (1990) Preliminary report: activation of the cerebellum in essential tremor. Lancet, 336: 1028-1030. Conway, D, Bain, PG, Warner, T et a!. (1993) Linkage analysis with chromosome 9 markers in hereditary essential tremor. Mov. Disord., 8: 374-376. Critchley, M (1949) Observations on essential (heredofamilial) tremor. Brain, 72: 113-139. Day, BL, Marsden, CD, Obeso, JA and Rothwell, JC (1984). Reciprocal inhibition between the muscles of the human forearm. J. Physiol., 349: 519-534. Deuschl, G, LUcking, CH and Schenck, E (1987) Essential tremor: electrophysiological and pharmacological evidence for a subdivision. J. Neurol. Neurosurg. Psychiatry, 50: 1435-1441. Deuschl, G, Bain, P, Brin, M and an Ad Hoc Scientific Committee (1998) Consensus statement of the Movement Disorder Society on Tremor. Mov. Disord., 13 (Supp!. 3): 2-23. Dubinsky, R and Hallett, M (1987) Glucose hypermetabolism of the inferior olive in patients with essential tremor. Ann. Neurol., 22: 118. Duncan, R, Bone, I and Melville, ill (1988) Essential tremor cured by infarction adjacent to the thalamus. J. Neurol. Neurosurg. Psychiatry, 51: 591-592. Dupuis, MJ, Delwaide, PJ, Boucquey, D and Gonsette, RE (1989) Homolateral disappearance of essential tremor after cerebellar stroke. Mov. Disord., 4: 183-187. Durr, A, Stevanin, G, Jedynak, CP, Penet, C, Agid, Y and Brice, A (1993). Familial essential tremor and idiopathic torsion dystoniaare different genetic entities. Neurol., 43: 2212-2214. Elble, RI (1986) Physiologic and essential tremor. Neurology, 36: 225-231. Elble, RJ (1995) Physiological and essential tremor. In: LJ Findley and WC Koller (Eds.), Handbook of Tremor Disorders. Marcel Dekker, New York, pp. 51-62. Elble, RI and Koller, WC (1990) Task-specific tremors. In: RI Elble and WC Koller (Eds.), Tremor. Johns Hopkins University Press, Baltimore and London, pp. 146-150. Elble, RI, Higgins, C and Moody, CJ (1987) Stretch reflex oscillations and essential tremor. J. Neurol. Neurosurg. Psychiatry, 50: 691-698. Elble, RI, Moody, C and Higgins, C (1990) Primary writing tremor. A form of focal dystonia. Mov. Disord., 5(2): 118-126. Elble, RI, Hughes, L and Higgins, C (1992) Phase resetting and frequency entrainment of essential tremor. Exp. Neurol., 116: 355-361.

P.G. BAIN

Elble, RJ, Higgins, C and Hughes, L (1994) Essential tremor entrains rapid voluntary movements. Exp. Neurol., 1994: 138-143. Fahn, S (1984) Rare tremors. In: LJ Findley and R Capildeo (Eds.), Movement Disorders: Tremor. Macmillan Press, London and Basingstoke, pp. 436-437. Fletcher, NA, Harding, AE and Marsden, CD (1990) A genetic study of idiopathic torsion dystonia in the United Kingdom. Brain, 113: 379-395. Fletcher, NA, Harding, AE and Marsden, CD (1991). A case control study of idiopathic torsion dystonia. Mov. Disord., 4: 304-309. Forsgren, L, Holmgren, G, Almay, BGL and Drugge, D (1988) Autosomal dominant torsion dystonia in a Swedish family. In: S Fahn, CD Marsden and DB Calne (Eds.) Advances in Neurology, Vol. 50, Dystonia 2. Raven Press, New York, pp. 83-92. Gulcher, JR, Jonsson, P, Kong, A et a!. (1997) Mapping of a familial essential tremor gene, FETl, to chromosome 3q13. Nature Genet., 17: 84-87. Halliday, DM, Conway, BA, Farmer, SF, Shahani, D, Russel, AJC and Rosenberg, JR (2000) Coherence between low-frequency activation of the motor cortex and tremor in patients with essential tremor. Lancet, 335: 1149-1153. Hellwig, B, Haubler, S, Schelter, B, Lauk, M, Guschlbauer, B, Timmer, J and LUcking, CH (2001) Tremorcorrelated cortical activity in essential tremor. Lancet, 357: 519-523. Higgins, JJ, Pho, LT and Nee, LE (1997) A gene (ETM) for essential tremor maps to chromosome 2p22-p25. Mov. Disord., 12: 859-864. Hirai, T, Miyazaki, M, Nakajima, H, Shibazaki, T and Ohye, C (1983) The correlation between tremor characteristics and the predicted volume of effective lesions in stereotaxic nucleus ventralis intermedius thalamotomy. Brain, 106: 1001-1018. Hua, SE, Lenz, FA, Zirh, TA et a!. (1998) Thalamic neuronal activity correlated with essential tremor. J. Neurol. Neurosurg. Psychiatry, 64: 273-276. Hughes, D and McLellan, DL (1985) Increased coactivation of the upper limb muscles in writer's cramp. J. Neurol. Neurosurg. Psychiatry, 48: 782-787. Jenkins, IB, Bain, PG, Colebatch, JG, Thompson, PD, Findley, U, Frackowiak, RSJ and Brooks, OJ (1993) A PET study of essential tremor: evidence for overactivity of cerebellar connections. Ann. Neurol., 34: 82-90. Jimenez-Jimenez, FJ, Cabrera-Valdivia, F, Orti-Pareja, M, Gasalla, T, Tallon-Barranco, A and Zurdo, M (1998) Bilateral primary writing tremor. Eur. J. Neurol., 5: 613-614. Kachi, T, Rothwell, JC, Cowan, JMA and Marsden, CD (1985) Writing tremor: its relationship to benign

ESSENTIAL TREMOR AND PRIMARY WRITING TREMOR

essential tremor. J. Neurol. Neurosurg. Psychiatry, 48: 545-550. Kim, JS and Lee, MC (1994) Writing tremor after discrete cortical infarction. Stroke, 25: 2280-2282. Klawans, HL, Glantz, R, Tanner, CM and Goetz, CG (1982) Primary writing tremor: a selective action tremor. Neurology, 32: 203-206. Koller, WC and Martyn, B (1986) Writing tremor: its relationship to essential tremor. J. Neurol. Neurosurg. Psychiatry, 49: 220. Koller, WC, Busenbark, K, Gray, C, Hassanein, RS and Dubinsky, R (1992) Classification of essential tremor. Clin. Neuropharmacol., 15: 81-87. Koller, WC, Pahwa, R, Busenbark, K et al. (1997) Highfrequency unilateral thalamic stimulation in the treatment of essential and parkinsonian tremor. Ann. Neurol., 42: 292-299. Koller, WC, Pahwa, R, Lyons, KE and Wilkinson, SB (2000) Deep brain stimulationm of the Vim nucleus of the thalamus for the treatment of tremor. Neurology, 55 (Suppl. 6): 29-33. Larsen, TA and CaIne, DB (1983) Essential tremor. Clin. Neuropharmacol., 6: 185-206. Leenders, KL, Salmon, EP, Tyrrell, PJ, Perani, D, Brooks, DJ, Frackowiak, RSJ et al. (1989) Brain L-CRF)6-Fluorodopa and C'Cj-nomifensine uptake in patients with Parkinson's disease and healthy volunteers. 1. Cereb. Blood Flow Metab., 9: S722. Lee, RG and Stein, RB (1981) Resetting of tremor by mechanical perturbations: a comparison of essential tremor and parkinsonian tremor. Ann. Neurol., 10: 523-53l. Limousin, P, Speelman, JD, Gielen, F and Janssens, M (1999) Multicenter European study of thalamic stimulation in parkinsonian and essential tremor. J. Neurol. Neurosurg. Psychiatry, 66: 289-296. Louis, ED, Ottman, R and Hauser, A (1998) How common is the most common adult movement disorder? Estimates of the prevalence of essential tremor throughout the world. Mov. Disord., 13: 5-10. Marsden, CD (1984) Origins of normal and pathological tremor. In: LJ Findley and R Capildeo (Eds.), Movement Disorders: Tremor. Macmillan Press, London, pp.37-84. Marshall, J (1962) Observations on essential tremor. J. Neurol. Neurosurg. Psychiatry, 25: 122-125. Minguez-Castellanos, A, Camero-Pardo, C, GomezCamello, A, Ortega-Moreno, A, Garcia-Gomez, T and Martin Linares, JM (1999) Primary writing tremor treated by chronic thalamic stimulation. Mov. Disord., 14: 1030-1033. Nagaratnam, N and Kalasabail, G (1989) Contralateral abolition of essential tremor following a pontine stroke. J. Neurol. Sci., 149: 195-196.

375 Nakashima, K, Rothwell, JC, Day, BL, Thompson, PD, Shannon, K and Marsden, CD (1989) Reciprocal inhibition between forearm muscles in patients with writer's cramp and other occupational cramps, symptomatic hemidystonia and hemiparesis due to stroke. Brain, 112: 681-697. Niranjan, A, Kondziolka, D, Baser, S, Heyman, R and Lunsford, LD (2000) Functional outcomes after gamma knife thalamotomy for essential tremor and MS-related tremor. Neurology, 55: 443-446. Ohye, C, Miyazaki, M, Hirai, T, Shibazaki, T, Nakajima, Hand Nagaseki, Y (1982) Primary writing tremor treated by stereotactic selective thalmotomy. J. Neurol. Neurosurg. Psychiatry, 45: 988-997. Pahwa, R, Lyons, K and Koller, WC (2000) Surgical treatment of essential tremor. Neurology, 54 (Supp!. 4): 39-44. Panizza, M, Lelli, S, Nilsson, J and Hallett, M (1990) Hreflex recovery curve and reciprocal inhibition of H-reflex in different kinds of dystonia. Neurology, 40: 824-828. Pascual-Leone, A, Valls-Sole, J, Toro, C, Wasserman, EM and Hallett, M (1994) Resetting of essential tremor and postural tremor in Parkinson's disease with transcranial magnetic stimulation. Muscle Nerve, 17: 800-817. Racette, BA, Dowling, J, Randle, J and Mink, JW (2001) Thalamic stimulation for primary writing tremor. J. Neurol., 248: 380-382. Raethjen, J, Lindermann, M, Schmaljohann, H, Wenzelburger, R, Pfister, G and Deuschl, G (2000) Multiple oscillators are causing parkinsonian and essential tremor. Mov. Disord., 15: 84-94. Rajput, AH, Rozdilsky, B, Ang, L and Rajput, A (1991) Clinicopathologic observations in essential tremor: report of six cases. Neurology, 41: 1422-1424. Ravits, J, Hallett, M, Baker, M and Wilkins, D (1985). Primary writing tremor and myoclonic writer's cramp. Neurology, 35: 1387-1391. Romeo, S, Beradelli, A, Pedace, F, Inghilleri, M, Giovanelli, M and Manfredi, M (1998) Cortical excitability in patients with essential tremor. Muscle Nerve, 21: 1304-1308. Rosenbaum, F and Jankovic, J (1988) Focal task-specific tremor and dystonia: categorization of occupational movement disorders. Neurology, 38: 522-527. Rotella, DL, Darling, WG and Rizzo, M (1999) Effect of hand posture on the temporal and kinematic properties of pointing and drawing by healthy subjects and by a patient with primary writing tremor. J. Mot. Behav., 31: 190-198. Rothwell, JC, Traub, MM and Marsden, CD (1979) Primary writing tremor. J. Neurol. Neurosurg. Psychiatry, 42: 1106-1114.

376 Rothwell, JC, Kachi, T, Thompson, PD, Day, BL and Marsden, CD (1987) Physiological investigations of Parkinsonian rest tremor and benign essential tremor. In: R Benecke, B Conrad and CD Marsden (Eds.), Motor Disturbances J. Academic Press, London, pp.I-17. Shahani, BT and Young, RR (1976) Physiological and pharmacological aids in the differential diagnosis of tremor. J. Neural. Neurosurg. Psychiatry, 39: 772783. Sheehy, MP and Marsden, CD (1982) Writer's cramp - a focal dystonia. Brain, 105: 461-480. Sheehy, MP, Rothwell, JC and Marsden, CD (1988) Writer's cramp. In: S Fahn and CD Marsden (Eds.), Advances in Neurology, Vol. 50, Dystonia 2. Raven Press, New York, pp. 457-472. Smith, IS, Furness, P and Thomas, PK (1984) Tremor in peripheral neuropathy. In: U Findley and R Capildeo (Eds.), Movement Disorders: Tremor. Macmillan press, London,pp.399-406.

P.G. BAIN

Spieker, S, Boose, A, Breit, S and Dichgans, J (1998) Long-term measurement of tremor. Mov. Disord., 13 (Suppl. 3): 81-84. Waddy, HM, Fletcher, NA, Harding, AE and Marsden, CD (1991) A genetic study of idiopathic focal dystonias. Ann. Neurol., 29: 320-324. Wills, AJ, Jenkins, IH, Thompson, PD, Findley, U and Brooks, DJ (1994) Red nuclear and cerebellar but no olivary activation associated with essential tremor: a positron emission tomographic study. Ann. Neurol., 36: 636-642. Wills, AJ, Jenkins, IH, Thompson, PD, Findley, U and Brooks, DJ (1995) A positron emission tomography study of cerebral activation associated with essential and writing tremor. Arch. Neurol., 52: 299-305. Young, RR (1986) Essential-familial tremor. In: PJ Vinken and GW Bruyn (Eds.), Handbook of Clinical Neurology. Elselvier, Amsterdam, pp. 565-581. Zeffiro, TA, Liao, K-K and Hallett, M (1992) Regional cerebral blood flow abnormailities in essential tremor. Neuralogy, 42 (Suppl. 3): 423.

Movement Disorders Handbook of Clinical Neurophysiology, Vol. 1 M. Hallett (Ed.) © 2003 Elsevier B. V. All rights reserved

377 CHAPTER 24

Clinical neurophysiology and pathophysiology of Parkinsonian tremor Giinther Deuschl*, Urban Fietzek, Stephan Klebe and Jens Volkmann Department of Neurology, Christian-Albrechts-Universitdt Kiel, Niemannsweg 147. D-24105 Kiel, Germany

24.1. Introduction The pathophysiological understanding of akinesia is advanced compared with the one of tremor. The detection of the anatomy and physiology of the basal ganglia loop (Alexander et al., 1986, 1990) has been a critical step towards this aim. The ultimate confirmation of a pathophysiological mechanism may be to design a therapeutic approach. In case of akinesia this was the destruction of the subthalamic nucleus which proved to cure akinesia in monkeys (Bergman et aI., 1990) and the reversible inactivation of the subthalamic nucleus through deep-brain stimulation has meanwhile become the accepted therapy for advanced akinesia (Limousin et aI., 1995, 1998). Akinesia and tremor in Parkinson's disease seem to share some pathophysiological principles but they are by no means directly related. Thus, conclusions drawn from the explanation of akinesia should not airily be applied to tremor. The present review will summarize the clinical and paraclinical findings which are considered to be relevant for the understanding of tremor and the most recent hypothesis will be summarized. The present review is based on earlier reviews (Deuschl et aI., 2000, 2001; Bergman and Deuschl, 2002; Deuschl and Bergman, 2002). 24.2. Clinical aspects of parkinsonian tremor The clinical presentation of tremor in PO is not uniform and although rest tremor is the classical form there are other manifestations which present mainly with postural and action tremors. Because of

* Correspondence to: Prof. Dr. G. Deuschl, Department of Neurology, Christian-Albrechts-Universitat Kiel, Niemannsweg 147,0-24105 Kiel, Germany. E-mail address:[email protected] Tel.: +49(431)597-2610: fax: +49(431)-597-2712.

these variants, tremor in Parkinson's disease (PO) has been defined (Deuschl et al., 1998) to be present if: (1) the patient has Parkinson's disease according to

the brain bank criteria (Hughes et aI., 1993), i.e. bradykinesia and one of the symptoms of rest tremor, rigidity or postural instability; (2) the patient has any form of pathological tremor. Classical parkinsonian tremor is defined as a rest tremor or rest and posturallkinetic tremor with the same frequency. Mostly this tremor is inhibited during movement and may reoccur with the same frequency when adopting a posture or even when moving. There is typically a reduction of tremor amplitude during the transition from rest to posture (Deuschl et aI., 1998; Jankovic et aI., 1999). The frequency of rest tremor is mostly above 4 Hz but the upper frequency limit is not well defined. Especially in the early stages higher rest tremor frequencies of up to 9 Hz can be found (Deuschl and Lucking, 1989; Koller et al., 1989). The clinical observations fit with the hypothesis that this form of postural! kinetic tremor (with similar frequencies for both the rest and posturallkinetic tremors) is a continuation of the rest tremor under postural and or kinetic conditions. Some patients exhibit a rest and/or postural tremor without signs of bradykinesia or rigidity being significant enough to diagnose Parkinson's disease. Thus, the clinical findings are not sufficient to diagnose Parkinson's disease, although there is PET evidence, that these patients have a dopaminergic deficit (Brooks et aI., 1992). When such a syndrome persists more than two years it is labeled as monosymptomatic tremor at rest (Deuschl et al., 1998). The clinical features of this tremor form are mostly identical with classical parkinsonian tremor

378

and it is assumed that the pathophysiological basis is similar as for classical parkinsonian tremor. Some patients have rest and postural/kinetic tremors of different frequencies, with the postural/ kinetic tremor displaying a higher (> 1.5 Hz) and non-harmonically related frequency to the rest tremor (Deuschl et aI., 1998). This form has often been considered to be a combination of an essential tremor with Parkinson's disease (Lance et aI., 1963; Findley et aI., 1981; Deuschl and Lucking, 1989) and is rare « 10% of patients with Parkinson's disease) according to our experience. In these patients the tremor can be extremely disabling. Some of these patients have had their postural tremor much longer than their Parkinson's disease. A low amplitude and high frequency (8-12 Hz) kinetic tremor is present in many parkinsonian patients. This can be detected easily by analyzing slow flexion/extension movements. This tremor is not disabling by itself but the patients are disabled by akinesia and rigidity. Isolated postural and kinetic tremors do occur in PD. The tremor frequency may vary between 4 and 9 Hz. Postural tremors are quite common in the akinetic rigid variant of Parkinson's disease. Severe postural and even intention tremor are rare in Parkinson's disease. These tremors have often been considered as essential tremor variants or have been found to be indistinguishable from enhanced physiologic tremor. In the following we will mainly deal with classical parkinsonian tremor including monosymptomatic rest tremor. Rest tremor is a unique clinical feature and should not be confused with the continuation of some postural tremors when the subject is not completely relaxed. Rest tremor should be diagnosed only when it shows an increasing amplitude during mental stress and at least a diminution of its amplitude when the extremity is voluntarily activated (Deuschl et al., 1998; Jankovic et aI., 1999). If this strict definition is applied, rest tremor is almost limited to parkinsonian tremor (or tremor in nonidiopathic parkinsonian syndromes) and Holmes' tremor, a rest and intention tremor of low frequency mostly secondary to a midbrain lesion. Both of these patient groups show a dopaminergic deficit in the striatum on positron emission tomography (Remy et al., 1995; Brooks 1999). This is another strong argument favoring rest tremor as a consequence of a nigro-striatal dopamine deficit no matter if it occurs

G. DEUSCHL ET AL.

in PD, non-idiopathic parkinsonian syndromes or in Holmes' tremor. Rest tremor in PD is the most specific sign for idiopathic PD among the other cardinal symptoms as the presence of classical rest tremor has a more than 95% probability to indicate idiopathic Parkinson's disease (Deuschl and Koester, 1996). But the severity of rest tremor is not correlated with the disease severity of PD or its duration nor with the striatal depletion of dopamine measured with positron emission tomography (Brucke et aI., 2000; Leenders and Oertel, 2001). The pathologic hallmark of Parkinson's disease is the degeneration of dopaminergic cells within the substantia nigra and the subsequent dopamine depletion of the striatum. The question arises if the pathology of tremor-dominant PD differs from the one of akinesia/rigidity-dominant PD. The medial substantia nigra, especially the retrorubral area A8 is more severely affected by dopaminergic cell degeneration in the tremor dominant form in contrast to more severe damage of the lateral substantia nigra (A9) in the akinetic rigid variant (Paulus and Jellinger, 1991; Hirsch et aI., 1992; Jellinger, 1999). Hence parkinsonian tremor is likely to be associated with cell loss of the retrorubral substantia nigra. Other nuclei like the locus ceruleus, however, are also more affected in the akinetic rigid variant. Patients with tremor-dominant PD have a better prognosis concerning disease progression than those with the akinetic rigid variant (Jankovic et al., 1990; LeWitt et aI., 1997; Louis et al., 1999). Thus, it cannot be excluded that these pathologic findings are related to other features of the tremor-dominant variant of PD, but these findings are a first hint at specific pathologic abnormalities of rest tremor.

24.3. Clinical neurophysiology of parkinsonian tremor in humans 24.3.1. Recording of tremor Recording of tremor is an important tool for clinical neurophysiology. The features of PD tremor can be quantified with electromyography or accelerometry. Such studies have shown that the frequency of classical rest tremor is between 4 and 7 Hz (Zimmermann et aI., 1994; Deuschl et aI., 1996), but during the early stages of the disease higher resting tremor frequencies up to 9 Hz may occur (Deuschl and Lucking, 1989; Koller et aI., 1989). However, as reflected in the clinical definition of PD

CLINICAL NEUROPHYSIOLOGY AND PATHOPHYSIOLOGY OF PARKINSONIAN TREMOR

tremor, the spectrum is much broader. Some patients have tremor frequencies which differ for resting and postural tremor by more than 1.5 Hz (Deuschl et al., 1996; Deuschl, 1999). Especially in rigid patients slow movements are accompanied by a highfrequency oscillation at 7-12 Hz which has been called "rippling" and which can occur without any resting tremor (Findley et al., 1981). More complicated forms of action tremor have long been recognized (lung, 1941; Lance et al., 1963). The tremor frequency or amplitude are only poor criteria for separating PD tremor from other tremors (Deuschl et al., 1996). The pattern of activation in antagonistic muscles is mostly reciprocal alternating in rest tremors (Fig. 1) but this is not a reliable discriminator between PD and other tremors (Spieker et al., 1995; Boose et al., 1996; Spieker et aI., 1997). Methods of non-linear time series analysis have separated PD

379

from essential tremor on the basis of the regularity and other more complex features of the tremor movement (Timmer et aI., 1993; Deuschl et al., 1995), but in prospective studies of larger series of patients, this distinction became much less clear (Timmer et al., 2000). Usually the cases which are most difficult to diagnose on clinical grounds are also difficult to diagnose on the basis of such electrophysiological tests. Thus, we conclude that although a variety of tests show statistically highly significant differences between PD and other tremors, no fully reliable electrophysiological tool is available to separate tremor in PD from other tremors unequivocally. 24.3.2. The contribution of reflex pathways Tremors may be caused by mechanical oscillations of a limb, enhanced reflexes, central oscilla-

Clinical Neurophysiology and Pathophysiology of Parkinsonian Tremor

Fig. 1. Electromyogram of muscles involved in tremor. Upper pair shows a classical reciprocal alternating pattern of antagonists, the pattern most frequently found in classical PD tremor, and the lower pair a synchronous pattern of antagonists.

380

tions or abnormally functioning feedback-loops within the CNS. Mechanical factors do always contribute to the generation of tremor but their contribution is almost negligible for large amplitude central tremors. The contribution of reflex factors to parkinsonian tremor has been assessed in different ways. On the one hand some clinical observations demonstrate that reflexes play only a minor role for the generation of parkinsonian tremor. The removal of the dorsal roots in a patient with parkinsonian tremor did reduce the tremor amplitude but only slightly changed the frequency and did not stop the tremor (Pollock and Davis, 1930). Neurosurgeons have shown early that the section of the pyramidal tract or removal of the motor cortex abolishes parkinsonian tremor (Bucy, 1940; Putnam, 1940) indicating that reflexes could only be involved if they follow a transcortical route. On the other hand attempts have been made to analyze the contribution of reflexes with different electrophysiological and pharmacological tests. The first was to infiltrate the muscles with novocaine until rigidity and stretch reflexes were absent but force was still present (Walshe, 1924). Walshe found no significant influence of this intervention on tremor. Another attempt was to look at changes of the frequency when loads or springs were attached to the trembling limb (Rack and Ross, 1986). With these procedures the resonance frequency is changed and thus might change the tremor frequency. Later larger studies demonstrated a lack of frequency reduction following loading of the trembling limb in parkinsonian patients (Hamberg et aI., 1987; Deuschl et aI., 1996) (Fig. 2). If reflexes would play a significant role for the generation of tremor the reduced resonance frequency should be accompanied by a decrease of the EMG frequency. However, this negative result has to be interpreted with reservations as for large-amplitude tremors because the mass of the trembling limb may be too large to show a significant frequency effect. Another approach was to impose sinusoidal movements on the trembling limb demonstrating that the tremor frequency could be entrained within a small frequency window (Rack and Ross, 1986). This has been interpreted to reflect some reflex contribution. Additionally parkinsonian tremor has been investigated in patients with the joints fixed in a cast (Burne, 1987). This has also reduced the tremor

G. DEUSCHL ET AL.

amplitude and changed the tremor frequency. It even stopped tremor when the limbs were rigidly fixed. Another series of studies has dealt with resetting of the rhythm of parkinsonian tremor following different stimuli. The first of these studies did not find a significant influence of mechanical pertubations imposed on the trembling hand of PD (Lee and Stein, 1981) but a clear resetting in essential tremor. This has been reinvestigated later and a dependence of the resetting on the force or amplitude of the pertubation rather than on the etiology (parkinsonian or essential tremor) was found (Britton et aI., 1992). Subsequent studies using electrical stimulation of the median nerve showed similar resetting of the tremor for both parkinsonian and essential tremor (Britton et aI., 1993). Related results have been found for both patient groups when using transcranial magnetic stimulation of the motor cortex (Pascual-Leone et aI., 1994). In summary, most studies do not support a critical role of peripheral mechanisms in the generation of parkinsonian tremor. There are indications that PD tremor may be modulated by peripheral manipulation (resetting) but this feature may eventually also apply for tremors due to central oscillator(s). The role of the central generators seems to be much more important. 24.3.3. Central oscillators underlying parkinsonian tremor 24.3.3.1. Clinical data suggesting a major role of central oscillators Deafferentation changes the frequency of parkinsonian tremor but does not suppress it (Pollock and Davis, 1930). This strongly favors a central origin of parkinsonian tremor. It has long been known that different lesions within the central nervous system can suppress parkinsonian tremor. Early attempts removing parts of the motor cortex or lesioning of the internal capsule have been successful in suppressing tremor but have produced other unacceptable side effects (Putnam, 1940; Cooper, 1969; Das et aI., 1998). Lesions within the thalamus or the zona incerta have been successful targets for thermocoagulation during stereotactic procedures (Hassler et aI., 1960; Cooper, 1969; Hassler et al., 1979), and recently it has been demonstrated that chronic stimulation of these same thalamic targets but also of the subthalamic nucleus and the pallidum

381

CLINICAL NEUROPHYSIOLOGY AND PATHOPHYSIOLOGY OF PARKINSONIAN TREMOR

Physiological Tremor

Rest Tremor in PO

Enhanced Physiological Tremor log (mg)2 2

Without load: log (mg)2

log (mg)2

-3

~l~

:~

Ace. I

-6

I

+~ '\

''''{"VI~ :: -J \/'J -3

~i~

o

5

10

15

Ext.

rl\

20

Flexor

f (Hz) 25

25

f (Hz)

1000 9 load: log (mg)2

IOg(mg~2

)

~

log (mg)2 -1

-3 -4

-6

-5

-7

-6

log

Ace.

(~V)2

1

10g~V)~

lV~-~ ~

-~

-3

-4

o

, 5

Ext.

'-/'V

, 10

, 15

, 20

-1

, f (Hz) 25

Flexor ~--.".--r.:---'~""""':--""'J.

(Hz)

Fig. 2. Spectral analysis of tremor. The spectra of the accelerogram (Ace.) of the hand, the EMG of hand flexors (flexor) and extensors (ext) are shown for three different subjects. The upper three spectra from each row show the recordings in the unloaded condition and the lower three traces the recordings when the hand is loaded with 1000 g. In the normal subject, the accelerogram shows a peak at 9 Hz in the unloaded condition but at 6 Hz in the loaded condition (see arrows) reflecting the shift in the resonance frequency of this mechanically oscillating system. In the patient with enhanced physiologic tremor there is also a resonance peak which can be seen at 8 and 4.5 Hz (see arrows), respectively. However, there is a further peak at about 12.5 Hz both in the EMGs and in the accelerograms (see broken line) which is most likely due to a central oscillator. In the patient with Parkinson's disease there is a sharp peak at 5 Hz. This peak is invariant to load changes indicating the central tremor in this case.

are all able to efficiently suppress parkinsonian tremor (Benabid et al., 1991, 1994; Koller et al., 1997, 1999; Krack et al., 1997; Volkmann et al., 1998; Limousin-Dowsey et al., 1999). This is all evidence that the preservation of some loops within the nervous system is critical for the occurrence of PD tremor.

24.3.3.2. Coherence analysis The frequency of tremor in a given patient is often similar in different extremities (Altenburger, 1937;

lung, 1941). These observation led to the assumption that all these muscles are governed by a common oscillator (Hunker and Abbs, 1990). This has been questioned recently with new mathematical techniques which have been applied to the study of these relations. The appropriate method to analyze such relations is coherence analysis (Winfree, 1980; Tass et al., 1998; Timmer et al., 1998a, b) because it can separate between a condition when two muscles are only trembling at the same frequency or whether they share common inputs indicating the same

382 oscillator underlying the tremor in the selected muscles. With this technique it could be demonstrated for orthostatic tremor that all the trembling muscles are coherent no matter if leg muscles are compared with arm muscles on different sides or even with cranial muscles (Koster et al., 1998). When applied to the electromyographic tremor bursts of patients with Parkinson's disease, the tremor on the right and left arm were not coherent (Lauk et al., 1999a). A more detailed analysis of this phenomenon has shown that the muscles within one body part (arm, leg, head) are mostly coherent (Raethjen et al., 2000a; Ben-Paz et al., 2001) but the rhythms in different extremities almost never are (Fig. 3). These differences are indicating that different oscillators are underlying parkinsonian tremor in the different extremities. The most likely location for such oscillating neuronal processes is within the basal ganglia loop (Bergman et al., 1998a). Besides the coherence between different muscles the relation between EEG activity and muscle activity is interesting with respect to the mechanisms generating tremor. Up to now most studies have focused on the coherence between brain and muscle activity in voluntary movements of normal subjects (Baker et al., 1997, 1999; Brown et al., 1998; Hari and Salenius, 1999; Mima et al., 1999; Mima and Hallett, 1999). These studies have demonstrated that some types of muscle activity are reflected in the coherence pattern at different frequency bands (15-30 Hz, 30-50 Hz) and that at least the lower frequency activity seems to reflect pyramidal tract cell activity (Baker et al., 1999). Up to now, abnormal muscle activity has only rarely been studied. In parkinsonian patients the normal 11-rhythm (Makela et al., 1993) is suppressed during tremor periods. Furthermore, it has been shown that neuromagnetic activity coherent with the tremor can be recorded over wide areas of the frontal and parietal cortex (Volkmann et al., 1996). Diencephalic, somatosensory and frontal areas are obviously sequentially activated during a tremor cycle. Recently it has been shown that a similar coherence can also be detected with electroencephalography (Hellwig et al., 2000). Data on the phase between the cortical activity and different extremities are still lacking, but are especially interesting with respect to the generation within different oscillators. Similar

G. DEUSCHL ET AL.

findings have been observed in essential tremor (Halliday et al., 2000; Hellwig et al., 2001). 24.3.3.3. Functional imaging The pathologic hallmark of PD is the nigral cell degeneration which leads to the reduction of dopaminergic terminals in the striatum. The latter has been demonstrated in numerous studies using labeled fluorodopa in positron emission tomographic studies. However, there is no correlation between the severity of the dopaminergic deficit in the striatum and the severity of tremor (Eidelberg et al., 1995; Antonini et al., 1998; Brooks, 1999) The clinical severity of tremor is also uncorrelated with the clinical disease progression (Jankovic et al., 1990; Nieuwboer et al., 1998; Louis et al., 1999). Similarly, most studies using imaging of the presynaptic dopamine transporters with B-CIT and single photon emission tomography found a clear abnormally reduced labeling of at least the putamen in patients with resting tremor (Asenbaum et al., 1998; Brucke et al., 2000). Both methods found a clearcut dopamine deficiency in the striatum, the hallmark of Parkinson's disease. Thus, we are faced with the conflict that PD tremor depends on the nigrostriatal deficit but once PD is present the tremor does not depend on the severity of this deficit. These differences in the clinical expression of tremor may be due to differences of the degeneration in the substantia nigra. Another relevant observation has been made with fluoro-deoxy-glucose PET, a method measuring the regional cerebral blood metabolism and, thereby, presumably the activity of the corresponding brain area. In a paradigm using deep brain stimulation of the ventral intermediate nucleus of the thalamus stimulation effective, non-effective low-frequency stimulation, and no stimulation have been compared (Deiber et al., 1993). It could be demonstrated that effective stimulation reduced the cerebellar regional blood flow compared with ineffective stimulation whereas the cortical blood flow was not significantly influenced. This finding has stimulated the idea that the cerebellum may be a key structure involved in tremor genesis. However, this paradigm cannot exclude that peripheral afferents may activate the cerebellum preferentially and, thereby, this finding may eventually only reflect the somatosensory input of rhythmic muscle activity. This interpretation is further supported by the finding that almost all

383

CLINICAL NEUROPHYSIOLOGY AND PATHOPHYSIOLOGY OF PARKINSONIAN TREMOR

1000 -

IIVS

ECU

100

ECU

IIVS

100

ECU

10 10

0.1

0.1

100

IIVS

FCU

Triceps brach.

Tlb. ant.

10

0.1 0.01

coherence 0.8

0.6 0.4

+-.----+H-----+---+--+--+++-++

0.2

o 3.14

o -3.14 +-+ll-+'-'4-+~'4-J+-4--+-~-+'4-+,-+-,-.J,Ol-i-l o 5

10

15

Hz

20

0

5

10

15

Hz 20

Fig. 3. Cross-spectral analysis of three muscle combinations in a PD patient. The upper two traces show the power spectra for both muscles. The coherence and phase spectra are demonstrated in the third and fourth row. The thin lines indicate the upper and lower limits of the 95% confidence interval of the spectra. The antagonistic and not antagonistic muscle combinations within the same arm show highly significant coherencies (see columns I and 2). The coherence is near I and the phase is around pi for the ECU-FCU muscle pair and near 0 for the ECU-triceps combination. The coherence and phase plots show narrow confidence limits for the peak frequencies and the higher harmonics with the latter being a mathematical artifact. In contrast the forearm extensor and the anterior tibial muscle of the same side (C) oscillate independently from each other although they share exactly the same frequency. Abbreviations: ECU: extensor carpi ulnaris, FCU: flexor carpi ulnaris (modified from Raethjen et a\., 2000a).

tremors show such a cerebellar hyperactivity (Colebatch et al., 1990; Hallett and Dubinsky, 1993; Jenkins et al., 1993; Wills et al., 1995, 1996; Bain et al., 1996). In another study (Kassubek et al., 2001) the metabolic changes could be localized within the ventrolateral thalamus and were also found to be related to the severity of tremor.

24.4. The firing pattern of basal ganglia cells in PD tremor New insights into the ongm of parkinsonian tremor are expected from recordings of basal ganglia cells in PD, because this is believed to be the primary locus of abnormality in PD. Such recordings

384

are obtained during stereotactic lesional procedures in the thalamus and pallidum or during the implantation of deep brain electrodes into the thalamus, pallidum or the subthalamic nucleus. The recordings are difficult to interpret since these recordings are performed only in patients but not in controls. "Control" data in healthy humans are not available. Therefore, the recordings obtained in human patients can only be compared with recordings in animal models of Parkinson's disease. Therefore the "normal" discharging pattern of basal ganglia neurons in humans will remain speculative to some extent. We will focus here on rhythmic patterns demonstrated in these recordings. The majority of papers report recordings from the ventrolateral thalamus. This is because the thalamus has been the traditional stereotactic target in the past (Lenz et al., 1985, 1987, 1988, 1989, 1990, 1993, 1994; Lin and Lenz, 1993; Mandir et al., 1997; Hua et al., 1998; Raeva et al., 1999a, b, 1998; Zirh et al., 1998). The nucleus ventralis caudalis (Vc) contains mainly cells that respond to sensory stimulation (sensory cells) and can thereby be identified. Anterior to the Vc is the motor thalamus with the Vim and the Vop and this area contains cells firing in response to voluntary movement (voluntary cells) and cells which respond both to movement and somatosensory stimulation (combined cells). Tremor cells have been found among the sensory, voluntary and combined cells (Lenz et al., 1988, 1990). Voluntary and combined cells usually exhibit a phase-advance suggesting an efferent role for this firing and sensory cells exhibit a phase lag suggesting an afferent relay function for this cell group. The percentage of tremor cells among the different groups seem to be higher than for the STN and GPi. Tremor cells were found among the combined cells in 53% among the voluntary cells in 22% and among the sensory cells in 54% (Lenz et al., 1994). These studies have convincingly demonstrated that some of the Vim and Vopcells fire before the onset of the tremor burst and may therefore be causally related with tremor. On the other hand the spike-train time series can be analyzed for a rebound excitation mechanism as proposed by R. Llinas (Llinas, 1984; Jahnsen and Llinas, 1984). Up to now a single study found that such a rebound excitation can be assumed for only a minority of cells whereas the majority seems to follow a relay mode (Zirh et al., 1998). The important conclusion drawn from this study has been

G. DEUSCHL ET AL.

that the thalamus is unlikely to be the generator of parkinsonian tremor. Instead the thalamus must be regarded as a relay station transmitting the tremor burst activity from other sources. Recordings in the pallidum are frequently performed because many pallidotomies are done but relatively few are reported. As expected from the basal ganglia model, the net firing frequency of neurons differ for the external (60±21Hz) and the internal (91 ±52Hz) pallidum (Magnin et al., 2000). Different cell types can be separated in the pallidum according to their firing pattern (Hutchison et a1., 1997, 1998b; Taha et al., 1997a, b; Guridi et al., 1999; Lozano et al., 1999; ). In the pallidum 12.3% (Hutchison et al., 1997) or, 19.7% (Magnin et al., 2000), respectively, of the cells were found to fire at the peripheral tremor frequency. All these cells were reported in the lower portion of the internal globus pallidus whereas the external globus pallidus did not show tremor-related activity (Hutchison et al., 1997). The rhythm of tremor cells within the pallidum is often correlated with the tremor frequency of the tremor in the periphery (Lozano et al., 1998). However, this does not mean that their activity is coherent with the peripheral tremor bursts which would allow the conclusion that GPi cells drive peripheral tremor if a phase lead of the tremor cell would be present. Instead, when 11 GPi tremor cells were studied for their relation with the peripheral tremor only one of them exhibited coherence with the peripheral tremor (Lemstra et al., 1999). A recent study of pallidal cells in a single patient undergoing pallidotomy has shown interesting dynamics of pallidal tremor cell activity (Hurtado et al., 1999): A single tremor cell showed coherent activity with a peripheral muscle for some time and desynchronized later from the ongoing peripheral tremor. Pairs of single units in the pallidum with tremor-related activity were recorded and showed partly coherent and partly independent firing between each other. These studies have shown that some cells have a clear phase advance which is necessary to accept them as neuronal activity that is causally related with tremor. A recent study in patients with or without tremor found coherence of Gpi or Gpe cells only in patients with tremor (Levy et al., 2002). These observations fit with the assumption of independent tremor-generating circuits for the different extremities which sometimes couple their activity with

CLINICAL NEUROPHYSIOLOGY AND PATHOPHYSIOLOGY OF PARKINSONIAN TREMOR

specific muscle groups but uncouple for other periods. Further studies in this field are promising. As the subthalamic nucleus is the target for deepbrain stimulation more data are becoming available from this important nucleus. Three types of cells have been described for the subthalamic nucleus in parkinson's disease (Rodriguez et aI., 1998): (1) Tonic cells with a mean frequency of 49 Hz which are usually modulated by passive and voluntary movement. (2) Pausing cells which discharge irregularly, with high firing rates for some milliseconds and subsequent pause (mean 30 Hz). These cells are the most frequent ones, they resemble the "pauser"cells of the GPe and can also be activated by sensorimotor input. (3) tremor cells firing with a burst pattern at 4-5 Hz with a frequency of 9-66 Hz. These cells are mostly sensitive to kinesthetic stimulation. They are not very frequent and are found in less than 20% of the STN cells. Similar findings were obtained in other studies (Hutchison et al., 1998a; Krack et aI., 1998a) that additionally emphasized that tremor cells are more often seen in tremor-dominant parkinsonian patients. The available observations are compatible with the notion that rhythmic activity related to different extremities is usually separated in different channels through the basal ganglia, but this separation seems to be lost in Parkinson's disease (Bergman et al., 1998a) and would also be compatible with the assumption that different circuits do exist for the generators of parkinsonian tremor in different extremities (Hurtado et aI., 1999; Lauk et al., 1999a; Raethjen et aI., 2000a).

24.5. The lesson of stereotactic surgery Lesions which can effectively suppress tremor of various origin are located in the ventrolateral thalamus. The region which is most often targeted is the nucleus ventralis intermedius thalami (Vim) or the nucleus ventralis oralis posterior (Vop) according to Hassler's classification. Most authors agree that the Vim is the most efficient location for the treatment of tremor (Ohye et al., 1993; Lenz et al., 1994, 1995; Tasker et al., 1996; Narabayashi, 1998). In Germany, Hassler, Mundinger and Riechert developed the lesion of the zona incerta additional to the VimNop-lesion. In their hands this lesion has improved the therapeutic result (Hassler et aI., 1979) regardless whether parkinsonian or non-parkinson-

385

ian tremors al'e considered. This has been confirmed recently by Aziz and colleagues (Alusi et al., 2001). Unfortunately, the different targets have never been compared prospectively for their therapeutic efficacy on tremor and, thus, a scientific conclusion is not yet possible. It has been found that larger-amplitude tremors need larger lesions (Hirai et al., 1983; Hariz and Hirabayashi, 1997). But these studies do not really solve the question if there is one target within the ventrolateral thalamus which is better than others. Microelectrode recordings during stereotactic surgery have demonstrated that there are specific areas at least in the thalamus which show tremor-related activity and neurosurgeons use this information as well as the intraoperative effect of high frequency stimulation on the tremor to place their lesion or to locate the stimulation electrode. The Vim and the Vop differ in their afferent input as the cerebellothalamic fibers terminate in the Vim and the pallido-thalamic fibers in the Vop. The thalamus is not the only place where parkinsonian tremor can be improved by lesion or stimulation. Recent experience with pallidotomy and especially deep brain stimulation of the pallidum and subthalamic nucleus has significantly expanded this view. It is now clear that parkinsonian tremor can also be effectively suppressed by stimulation of the pallidum (or pallidotomy) (Krack et al., 1998b; Lozano et al., 1998; Pollak et aI., 1998; Volkmann et al., 1998; Lang et al., 1999) and the subthalamic nucleus (Benabid et aI., 1994; Krack et aI., 1997). Thus, parkinsonian tremor can also be effectively treated by the blockade of nuclei upstream from the thalamus within the basal ganglionic-thalamic loop.

24.6. Animal models of parkinsonian tremor Historically, the first convincing demonstration of mechanisms mediating rhythmic activity with a possible relation to tremor came from studies of Llinas and his group (Llinas and Volkind, 1973; Llinas and Yarom, 1981; Jahnsen and Llinas, 1984) and is labeled as the 'rebound-excitation hypothesis'. They have proposed the basic mechanism of central tremors to be due to specific oscillating properties of thalamic cells (Llinas, 1984). These cells can be activated in two different modes, the "relay mode" with normal summation of postsynaptic potentials at the membrane until the firing

386 threshold is reached and the "oscillatory mode". The latter is mainly driven by calcium-dependent changes of the membrane potential leading to a rebound calcium-spike which causes the next spike of a particular cell (for details see Jahnsen and LUnas, 1984; Llinas, 1984). The oscillatory mode of the thalamic cells is driven by hyperpolarization of these thalamic cells. As the pallidum internum is overactive in Parkinson's disease, the inhibitory input to the thalamus might hyperpolarize the thalamic cells and thereby cause this mechanism to be activated. Animal data have shown that thalamic neurons which are driven by this rebound excitation due to low threshold calcium spikes show a characteristic pattern of inter-spike-intervals (Steriade and Deschenes, 1984). These interspike-intervals can also be measured in thalamic tremor cells of parkinsonian patients undergoing stereotactic surgery. This was done in a reasonable number of tremor cells but the majority failed to show the typical feature of rebound excitation (Zirh et al., 1998). Thus, these recordings in patients do not support the rebound excitation hypothesis. Another hypothesis how the 4-6 Hz pattern of parkinsonian tremor could be generated within the thalamus is based on in-vitro experiments of the interaction between the pallidum and the thalamic target nuclei. Pare et al. (1990) have shown, that for some frequencies of pallidal neurons, the thalamus does have specific "filter" properties (thalamic filter hypothesis). They could show that a 12-15 Hz pattern of pallidal cells will be transformed into a 4-6 Hz pattern due to specific membrane properties of these thalamic cells. This mechanism is also not yet supported by data from patients as most of the tremor cells in the pallidum internum of patients are already firing at a low frequency (Hutchison et al., 1997, 1998b; Hurtado et al., 1999; Lemstra et al., 1999) and the 12-15 Hz range seems to be not specifically more frequent in these PD patients. The above mentioned models are all based on invitro models but in-vivo models may be more prone to be compared with the situation in patients. A variety of such in-vivo animal models of tremors exist (Wilms et al., 1999). The best model of parkinsonian tremor seems to be the l-methyl4-phenyl-l,2,3,6-tetrahydropyridine hydrochloride (MPTP) model of Parkinson's disease. However, not all species develop tremor and if they do the animals

G. DEUSCHL ET AL.

only rarely have classical resting tremor (Wilms et al., 1999). This has been especially studied by Bergman et al. (Wichmann et al., 1994; Nini et al., 1995; Raz et al., 1996; Bergman et al., 1998b) in the rhesus and vervet monkey. Studies of this animal modelled to the 'loss of segregation hypothesis' (see Fig. 4) for parkinsonian tremor (Bergman et al.,

Fig. 4. The loss-of-segregation hypothesis of parkinsonian tremor. The current model of the basal ganglia connectivity is displayed. The hypothesis is based on the fact that different regions within the main basal-ganglia nuclei fire independently in the healthy monkey and become coherent after MPTP treatment (Bergman et al., 1998a; Raz et al., 2000).This loss of segregation between parallel basal ganglialoops might lead to an enhancement of rhythmic firing resulting in parkinsonian tremor. This well documented phenomenon implies that it is not a single structure of the basal ganglia but the whole basal ganglialoop is responsible for tremorgeneration. (Abbreviations: SNC: substantia nigra, pars compacta; SNR: substantia nigra pars reticularis; OPE: external pallidum; GPI: internal pallidum; STN: subthalamic nucleus; VIM: ventral intermediate nucleus; YOP: ventral oral posterior nucleus; PPN: pedunculopontine nucleus; -: oscillating activity.)

CLINICAL NEUROPHYSIOLOGY AND PATHOPHYSIOLOGY OF PARKINSONIAN TREMOR

1998a; Raz et aI., 2000) which explains most of the findings for PD tremor. This hypothesis is based on the present view of basal ganglia function which assumes that the basal ganglia select specific motor programs and thereby inhibit others. This is believed to be accomplished with segregated loops connecting all the nuclei involved in this loop (cortex, putamen, pallidum, subthalamic nucleus and thalamus). The dopamine depletion of the striatum may lead to an inability to suppress competing motor programs and thereby may slow down the desired motor program. Single cell recordings in the normal situation show uncorrelated firing between different cells along this pathway. When the animals have been rendered parkinsonian with MPTP the discharge characteristics show synchronous firing between different cells. Pairs of cells which are connected to different extremities tend to show coherence that cannot be found in the normal state. Moreover, correlated activity was found in tremulous monkeys with a phase around 0 degrees whereas such stable phase relations were lacking in monkeys without tremor although they also showed rhythmic and correlated activity. The tremor cells in the pallidum were often correlated with the peripheral arm tremor. According to this hypothesis, tremor is the result of the loss of segregation within the basal ganglia loops for the different topographic regions and within each of these information channels. These findings provide the anatomic and physiologic basis for the separation of the oscillating activity for the different extremities and, thus, the basis for multiple oscillators of parkinsonian tremor. Many of the features typical for parkinsonian tremor as the rest component or the waxing and waning can be hypothetically explained by this mechanism. Thus, it provides an interesting candidate for future hypothetically driven tremor research of parkinsonian tremor. Finally, other tremor models must be mentioned. Numerous cholinergic substances have been shown to induce tremor in animals (Wilms et aI., 1999) which, however, do not match well with rest tremor. The frequency of this tremor is mostly above 8 Hz. The striatum might play an important role for the development of this form of tremor because local injection of oxotremorine can elicit this tremor (Everett et aI., 1956; Ankier et aI., 1971). It is tempting to speculate that this form of tremor is related to the action tremor seen in some

387

patients with Parkinson's disease which has a frequency indistinguishable from enhanced physiologic tremor. Successful surgical attempts have been undertaken to produce resting tremor in animals (especially monkeys) (for review see Wilms et aI., 1999). It might be deduced from these studies, that not a single nucleus but combined destructions of nuclei or pathways are necessary to produce resting tremor. Moreover, isolated rest tremor seems to be almost impossible to produce by isolated lesions, because the most consistent models have always also shown postural or/and intention tremor. Especially the destruction of three structures seems to be crucial to the induction of rest and intention tremor in the primate model: The parvocellular division of the red nucleus, cerebello-thalamic fibers and nigro-striatal fibers (Ohye et aI., 1988). In these tremor models, rhythmic discharges could be recorded at the thalamocortical and corticospinal level (Lamarre, 1984). In another animal model using lesions of the midbrain including the substantia nigra and the red nucleus resting tremor has been produced resembling parkinsonian tremor. Stimulation of the subthalamic nucleus in this condition has demonstrated tremor relief under circumstances comparable to the human situation (Gao et aI., 1999).

24.8. Conclusions 24.8.1. Diagnosis ofparkinsonian tremor with electrophysiological methods

Parkinson's disease can be diagnosed with clinical methods but the gold standard for confirmation is neuropathologic assessment. Electrophysiologic methods can further improve the accuracy of the diagnosis. There is not a single electrophysiologic method to diagnose parkinsonian tremor. However, when putting clinical and electrophysiological data together a very high degree of accuracy may be obtained. Controlled studies on this issue are not yet available. Therefore, the following remarks may be interpreted as the way how the authors use these electrophysiological techniques in their daily practice. Almost every patient with tremor of unknown origin or with suspected Parkinson's disease gets a formalized tremor assessment in our laboratory including recording of rest tremor, postural tremor

388

without and postural tremor with the hands loaded with 1 kg each (for details see Deuschl et al. (1996). Recording includes the EMG of the hand flexors and extensors and the accelerogram of the hand. This is subsequently automatically analyzed for each hand separately and for the following variables (Lauk et al., 1999b): tremor frequency, tremor amplitude (total power between 2 and 25 Hz), some more detailed measures of the spectra for accelerogram and EMGs. Normal values for our procedure have been published (Raethjen et al., 2000b) or can be obtained for other methods of data analysis. The frequency can also be calculated from the raw EMGs (especially if the EMG is rectified) or accelerograms. Several significant contributions to the final diagnosis can be made with electromyography: Diagnosis of tremor in a muscle: Electromyography is the most potent method to diagnose tremor and may be very helpful for the diagnosis of PD tremor, especially when combined with spectral analysis. It can show reciprocal alternating patterns of muscle activity which is very common in classical parkinsonian tremor, but when the recordings are analyzed in more details and for longer time periods other phase relations between the two antagonists can sometimes be found (Boose et al., 1996), however, with a clear-cut preponderance of the - 180 deg angle (reciprocal alternating activity). Diagnosis of a pathologic tremor: A pathologic tremor is present if: (a) the peak frequency of the tremor is outside the normal values (6-13 Hz); and/ or (b) if the amplitude is outside the normal range; and/or (c) if the comparison of spectral analysis in the different recording conditions shows features of a central or reflex-driven tremor. The tremor frequency can be obtained without spectral analysis by simple counting of the EMG bursts per minute. However, in the cases which are referred to with this question, it is often hard to detect the tremor from the raw traces at all. The amplitude is even more difficult to estimate from the raw traces of the accelerogram and we found no other way than to use the total power of the spectrum. Diagnosis of a centrally driven or reflex activated tremor: It has been mentioned earlier, that central tremors can be identified by looking at the characteristics of the dominant frequency within the EMG during postural activity with and without load (see Fig. 2). Central tremors have an EMG peak which is invariant to loading - even in the rare case when the

G. DEUSCHL ET AL.

peak of the accelerogram shows a decreasing frequency. This is found in almost every patient with classical parkinsonian tremor. A reflex-driven tremor is identified by the decreasing tremor frequency when the hand is loaded. The latter case occurs rarely in PD and may indicate another underlying disease. Anyway, if a central tremor can be diagnosed this is confirmation for the clinical diagnosis of parkinsonian tremor according to the clinical definition. Diagnosis of rest tremor: As mentioned above, the demonstration of rest tremor is a crucial sign for PD and has a very high diagnostic specificity. Clinical neurophysiology can establish the diagnosis of rest tremor in several ways. It can demonstrate that tremor is present at rest and that the tremors disappears or is reduced in amplitude when the arm is voluntarily activated (Deuschl and Krack, 1998). This has been labeled as re-emergent tremor and was found to be typical for classical PD tremor (Jankovic et al., 1999). Rest tremor is beginning with very small amplitudes until the patient really perceives the tremor himself. Therefore, the diagnosis of subclinical (rest) tremor is often a significant help for the clinician: Invisible tremor of a muscle can be detected with EMG but much better with spectral analysis of the rectified EMG. This can be a very powerful tool to diagnose tremor in a patient with no obvious signs for parkinsonian tremor but with clinical suspicion for PD. Very subtle tremor can be detected with both accelerogram and EMG. Maneuvers to provoke the tremor in the resting condition (counting backwards) are recommended in complicated cases. The total power of tremor may be within normal limits but the synchronization may be abnormal leading to a sharp peak. Another frequent question is if in a patient with a clinically unilateral parkinsonian tremor the other side is involved or not. In this situation spectral analysis of the rectified EMG is much better than the spectral analysis of the accelerogram because the latter is less specific and can be influenced by rhythmic activity from the other side. Tremor frequency as a confirming sign or a red flag: The frequency is not a sign with high confirmatory value, especially if it is within the most frequent range between 4 and 6 Hz as many patients with other tremors (e.g. essential tremor, dystonic tremor) share these frequencies. When it is below

CLINICAL NEUROPHYSIOLOGY AND PATHOPHYSIOLOGY OF PARKINSONIAN TREMOR

4 Hz the patient must undergo a further clinical exam for the presence of additional clinical or paraclinical signs for a complicating brain disease (e.g. vascular disease). Resting tremor frequencies above 7 Hz do occur in PD but mostly in early stage patients. Whenever they occur further clinical analysis of the tremor is required. Diagnostic information of differing frequencies for rest and postural tremor: The definition of tremors in PD is dependent on the frequencies under rest and postural conditions. Even in the case of classical parkinsonian tremor the frequency of postural tremor is somewhat higher than the one of tremor at rest. However, according to our experience this frequency difference is rarely above 0.8 Hz and almost never above 1.5 Hz. This has been defined as the cut-off value between classical parkinsonian tremor (type I) and type II tremor with differing frequency of rest and postural tremor. This difference is usually clinically visible but may need further confirmation which is a typical task for the clinical neurophysiology of tremors. We'd like to stress that the history of type II tremor is usually different from the one of classical parkinsonian tremor as the type II patients have a history of preceding postural tremor and often a positive family history for essential tremor. Further neurophysiological tests for the routine analysis of parkinsonian tremor: Several features have been discussed in the pathophysiology section of this chapter which can separate parkinsonian from other tremors. None of these methods has a very high specificity and they were not really established for clinical routine purposes. In conclusion the clinical diagnosis of PD and PD tremor can be significantly improved with electrophysiological methods. 24.8.2. The pathophysiology ofparkinsonian tremor

Many data support the hypothesis that parkinsonian symptoms are related to malfunction of the basal ganglia. The cerebellum is believed to modify the frequency and activation condition of parkinsonian tremor. Spinal reflex mechanisms are unlikely to playa major role for the generation of parkinsonian symptoms, but they are believed to modify the frequency and amplitude characteristics of the parkinsonian tremor. These statements are based on the data which are reviewed here but still represent a

389

hypothesis more than a scientific conclusion. Some of them need still further confirmation but they can be amalgamated into a working hypothesis which is based both on clinical and experimental data and may provide a framework for future research. The limbs of parkinsonian patients tremble independent from each other, but nevertheless often at a similar frequency. If spinal mechanisms might account for these observations, strong reflex effects would be expected on the rhythm and amplitude of parkinsonian tremor which have never been demonstrated. The easiest resetting of the tremor seems to be possible with transcranial magnetic stimulation which could be interpreted as a direct resetting of the cortico-basal ganglia circuits that produce parkinsonian tremor. The studies on the coherence of cortical and muscle activity suggest, that cortical events seem to precede tremor in the muscles and, thus, the cortex may be leading a sequence of events which ultimately produces tremor. The cortex itself is unlikely to produce the rhythmic discharges while the basal ganglia seem to be crucially involved. It is now well established both with anatomical studies and cell recordings that the basal ganglia are topographically organized and that the loops of the basal ganglia cortical circuitry can be coupled. Under normal conditions, the topographic compartments of these circuits are highly separated, but this segregation seems to be abolished under pathologic conditions. This could provide the anatomical background for the development of parkinsonian tremor and possibly other Parkinson symptoms. In patients tremor cells were found in the subthalamic nucleus, the internal pallidum and the thalamus. They could either be the generators themselves or they may be an integral part of an unstable oscillating network. In this case not a single nucleus by itself is responsible but synchronized action of the cells regulating one functional region or even from different regions get synchronized because the dopaminergic input is believed to keep these cells separated under normal conditions. How does this idea match with the results of functional neurosurgery? The excellent response of parkinsonian patients to lesions or stimulation within the GPi or the STN might be due to a desynchronization of the rhythmic activity within the oscillating loops. An explanation for the excellent improvement of rest tremor following Vim lesions or stimulation is more difficult to provide as this lesion

390

is not within the anatomic target of the basal ganglialoop in the thalamus but within the cerebellothalamo-cortical loop. A simple view could be that either the anatomic projection of pallido-thalamic fibers partially project to the Vim, or that the pallido-thalamic pathway passes through the Vim. Alternatively, interrupting the cerebello-thalamic circuitry may block parkinsonian tremor, suggesting a critical role for complex basal ganglia-cerebellum relationships at the level of the thalamus or cortex. These observations leave us with the paradox that parkinsonian symptoms are generated within the basal ganglia loop, but they can be successfully treated not only by blockade of the cortico-basal ganglia-cortical loop but also by blockade of the cerebello-thalamo-cortical loop. The beneficial effect of the Vim-lesion/stimulation must then be due to an interaction between the basal gangliathalamo-cortical and the cerebello-thalamo-cortical projection. As this interaction is unlikely to occur at thalamic level because the two projections are believed to be separated up to the cortex (Strick, 1985; Kitano et aI., 1998; Middleton and Strick, 2000), we must assume that such interaction takes place at cortical level. If so, it must be further hypothesized that cortical stimulation at the cortical targets of the basal ganglia and cerebello-thalamocortical loops would ameliorate the tremor and also other parkinsonian signs. This has indeed been shown in several cases of Parkinson's disease responding to epidural motor cortex stimulation (Canavero and Paolotti, 2000). Taking all the evidence together, the basal ganglia are the most likely candidate producing parkinsonian tremor. The segregated basal ganglia loops may be influenced by cerebellar mechanisms as well. Neurosurgical procedures may not necessarily destroy or reversibly block the 'oscillator' of parkinsonian tremor but they may desynchronize abnormally coupled neuronal pathways. Acknowledgment

This work was supported by a grant of the German Bundesministerium fur Bildung und Forschung (Kompetenznetz Parkinson) to GD.

References Alexander, GE, DeLong, MR and Strick, PL (1986) Parallel organization of functionally segregated circuits

G. DEUSCHL ET AL.

linking basal ganglia and cortex. Ann. Rev. Neurosci., 9: 357-381. Alexander, GE, Crutcher, MD and DeLong, MR (1990) Basal ganglia-thalamocortical circuits: parallel substrates for motor, oculomotor, "prefrontal" and "limbic" functions. Progr. Brain Res., 85: 119-146. Altenburger, H (1937) Elektrodiagnostik. In: OFO Bumke (Ed.), Handbuch der Neurologie. Band m. Berlin: Julius Springer, pp. 1036-1086 Alusi, SH, Aziz, TZ, Glickman, S, Jahanshahi, M, Stein, JF and Bain, PG (2001) Stereotactic lesional surgery for the treatment of tremor in multiple sclerosis: a prospective case-controlled study. Brain, 124: 15761589. Ankier, SI, Brittain, RT and Jack, D (1971) Investigation of central cholinergic mechanisms in the conscious mouse. Br. J. Pharmacol., 42: 127-136. Antonini, A, Moeller, JR, Nakamura, T, Spetsieris, P, Dhawan, V and Eidelberg, D (1998) The metabolic anatomy of tremor in Parkinson's disease. Neurology, 51: 803-810. Asenbaum, S, Pirker, W, Angelberger, P, Bencsits, G, Pruckmayer, M and Brucke, T (1998) (1231)beta-CIT and SPECT in essential tremor and Parkinson's disease. J. Neural. Trans., 105: 1213-1228. Bain, PG, Britton, TC, Jenkins, IH, Thompson, PD, Rothwell, JC, Thomas, PK, Brooks, DJ and Marsden, CD (1996) Tremor associated with benign IgM paraproteinaemic neuropathy. Brain, 119: 789-799. Baker, SN, Olivier, E and Lemon, RN (1997) Coherent oscillations in monkey motor cortex and hand muscle EMG show task-dependent modulation. J. Physiol. (Lond.), 501: 225-241. Baker, SN, Kilner, JM, Pinches, EM and Lemon, RN (1999) The role of synchrony and oscillations in the motor output. Exp. Brain Res., 128: 109-117. Benabid, AL, Pollak, P, Gervason, C, Hoffmann, D, Gao, DM, Hommel, M, Perret, JE and de Rougemont, J (1991) Long-term suppression of tremor by chronic stimulation of the ventral intermediate thalamic nucleus. Lancet, 337: 403-406. Benabid, AL, Pollak, P, Gross, C, Hoffmann, D, Benazzouz, A, Gao, DM, Laurent, A, Gentil, M and Perret, J (1994) Acute and long-term effects of subthalamic nucleus stimulation in Parkinson's disease. Stereotact. Funct. Neurosurg., 62: 76-84. Ben-Paz, H, Bergman, H, Goldber, JA, Giladi, N, Hansel, D, Reches, A and Simon, ES (2001) Synchrony of rest tremor in multiple limbs in Parkinson's disease: evidence for multiple oscillators. J. Neural. Trans., 108: 287-296. Bergman, H and Deuschl, G (2002) Pathophysiology of Parkinson's disease: From clinical neurology to basic neuroscience and back. Mov. Disord., 17: S28-40.

CLINICAL NEUROPHYSIOLOGY AND PATHOPHYSIOLOGY OF PARKINSONIAN TREMOR

Bergman, H, Wichmann, T and DeLong, MR (1990) Reversal of experimental parkinsonism by lesions of the subthalamic nucleus. Science, 249: 1436-1438. Bergman, H, Feingold, A, Nini, A, Raz, A, Siovin, H, Abeles, M and Vaadia, E (I 998a) Physiological aspects of information processing in the basal ganglia of normal and parkinsonian primates. Trends Neurosci., 21: 32-38. Bergman, H, Raz, A, Feingold, A, Nini, A, Nelken, I, Hansel, D, Ben-Pazi, Hand Reches, A (l998b) Physiology of MPTP tremor. Mov. Disord., 13: 29-34. Boose, A, Spieker, S, Jentgens, C and Dichgans, J (1996) Wrist tremor: investigation of agonist-antagonist interaction by means of long-term EMG recording and cross-spectral analysis. Electroencephalogr. Clin. Neurophysiol., 101: 355-363. Britton, TC, Thompson, PD, Day, BL, Rothwell, JC, Findley, LJ and Marsden, CD (1992) "Resetting" of postural tremors at the wrist with mechanical stretches in Parkinson's disease, essential tremor, and normal subjects mimicking tremor. Ann. Neurol., 31: 507514. Britton, TC, Thompson, PD, Day, BL, Rothwell, JC, Findley, LJ and Marsden, CD (1993) Modulation of postural tremors at the wrist by supramaximal electrical median nerve shocks in essential tremor, Parkinson's disease and normal subjects mimicking tremor. J. Neurol. Neurosurg. Psychiatry, 56: 10851089. Brooks, DJ (1999) Functional imaging of Parkinson's disease: is it possible to detect brain areas for specific symptoms? J. Neural. Trans. 56 (Supp!.): 139-153. Brooks, DJ, Playford, ED, Ibanez, V, Sawle, GV, Thompson, PD, Findley, LJ and Marsden, CD (1992) Isolated tremor and disruption of the nigrostriatal dopaminergic system: an 18F-DOPA PET study (see comments). Neurol.,42: 1554-1560. Brown, P, Salenius, S, Rothwell, JC and Hari, R (1998) Cortical correlate of the Piper rhythm in humans. J. Neurophysiol., 80: 2911-2917. Brucke, T, Djamshidian, S, Bencsits, G, Pirker, W, Asenbaum, Sand Podreka, I (2000) SPECT and PET imaging of the dopaminergic system in Parkinson's disease. J. Neurol., 247 (Suppl. 4): IVJ2-7. Bucy, PC (1940) Cortical extirpation in the treatment of involuntary movements. Assoc. Res. Nerv. Ment. Dis., 21: 551-595. Burne, JA (1987) Reflex origin of parkinsonian tremor. Exp. Neurol., 97: 327-339. Canavero, Sand Paolotti, R (2000) Extradural motor cortex stimulation for advanced Parkinson's disease: case report. Mov. Disord., 15: 169-171. Colebatch, JG, Findley, LJ, Frackowiak, RS, Marsden, CD and Brooks, DJ (1990) Preliminary report: activation of

391

the cerebellum in essential tremor (published erratum appears in Lancet 1990, Nov 24;336(8726): 1330). Lancet, 336: 1028-1030. Cooper, I (1969) Neurosurgical alleviation of intention tremor of multiple sclerosis and cerebellar disease. N. Engl. J. Med., 263: 441--444. Das, K, Benzil, DL, Rovit, RL, Murali, Rand Couldwell, WT (1998) Irving S. Cooper (1922-1985): a pioneer in functional neurosurgery. 1. Neurosurg., 89: 865-873. Deiber, MP, Pollak, P, Passingham, R, Landais, P, Gervason, C, Cinotti, L, Friston, K, Frackowiak, R, Mauguiere, F and Benabid, AL (1993) Thalamic stimulation and suppression of parkinsonian tremor. Evidence of a cerebellar deactivation using positron emission tomography. Brain, 116: 267-279. Deuschl, G (1999) Differential diagnosis of tremor. J. Neural. Trans., 56 (Suppl.): 211-220. Deuschl, G and Bergman, H (2002) Pathophysiology of non-parkinsonian tremors. Mov. Disord., 17: S41--48. Deuschl, G and Koester, B (1996) Diagnose und Behandlung des Tremors. In: B Conrad and AO CeballosBaumann (Eds.), Bewegungsstorungen in der Neurologie. Stuttgart/New York: Thieme Verlag, pp. 222-253. Deuschl, G and Krack, P (1998) Tremors: Differential diagnosis, neurophysiology and pharmacology. In: JJaE Tolosa (Ed.), Parkinson's Disease and Movement Disorders. Baltimore: Williams and Wilkins, pp. 419452. Deuschl, G and LUcking, CH (1989) Tremor and electrically elicited long-latency reflexes in early stages of Parkinson's disease. In: HP Riederer (Ed.), Early Diagnosis and Preventive Therapy of Parkinson's Disease. Wien, New York: Springer Verlag, pp. 103110. Deuschl, G, Lauck, M and Timmer, M (1995) Tremor classification and tremor time series. Chaos, 5: 48-51. Deuschl, G, Krack, P, Lauk, M and Timmer, J (1996) Clinical neurophysiology of tremor. J. Clin. Neurophysiol., 13: 110-121. Deuschl, G, Bain, P, Brin, M and Scientific-Committee (1998) Consensus statement of the movement disorder society on tremor. Mov. Disord., 13: 2-23. Deuschl, G, Raethjen, J, Baron, R, Lindemann, M, Wilms, H and Krack, P (2000) The pathophysiology of parkinsonian tremor: a review. J. Neurol., 247 (Supp!. 5): V33--48. Deuschl, G, Raethjen, J, Lindemann, M and Krack, P (2001) The pathophysiology of tremor. Muscle Nerve, 24: 716-735. Eidelberg, D, Moeller, JR, Ishikawa, T, Dhawan, V, Spetsieris, P, Chaly, T, Robeson, W, Dahl, JR and Margouleff, D (1995) Assessment of disease severity

392 in parkinsonism with fluorine-18-fluorodeoxyglucose and PET. J. Nucl. Med., 36: 378-383. Everett, GM, Blockus, LE and Shepperd, 1M (1956) Tremor induced by tremorine and its antagonim by anti-parkinsonian drugs. Science, 124: 79. Findley, LJ, Gresty, MA and Halmagyi, GM (1981) Tremor, the cogwheel phenomenon and clonus in Parkinson's disease. J. Neural. Neurosurg. Psychiatry, 44: 534-546. Gao, OM, Benazzouz, A, Piallat, B, Bressand, K, llinsky, lA, Kultas-Ilinsky, K and Benabid, AL (1999) Highfrequency stimulation of the subthalamic nucleus suppresses experimental resting tremor in the monkey. Neuroscience, 88: 201-212. Guridi, J, Gorospe, A, Ramos, E, Linazasoro, G, Rodriguez, MC and Obeso, JA (1999) Stereotactic targeting of the globus pallidus intemus in Parkinson's disease: imaging vs. electrophysiological mapping. Neurosurgery, 45: 278-287; discussion 287-289. Hallett, M and Dubinsky, RM (1993) Glucose metabolism in the brain of patients with essential tremor. J. Neurol. Sci., 114: 45-48. . Halliday, OM, Conway, BA, Farmer, SF, Shahani, U, Russell, AJ and Rosenberg, JR (2000) Coherence between low-frequency activation of the motor cortex and tremor in patients with essential tremor. Lancet, 355: 1149-1153. Hari, Rand Salenius, S (1999) Rhythmical corticomotor communication. NeuroReport, 10: R1-1O. Hariz, MI and Hirabayashi, H (1997) Is there a relationship between size and site of the stereotactic lesion and symptomatic results of pallidotomy and thalamotomy? Stereotact. Funct. Neurosurg., 69: 28-45. Hassler, R, Riechert, TFM, Umbach, W and Gangelberger, A (1960) Physiological observations in stereotactic operations in extrapyramidal motor disturbances. Brain, 83: 337-350. Hassler, R, Mundinger, F and Riechert, T (1979) Stereotaxis in Parkinson syndrome. Berlin: Springer Verlag. Hellwig, B, Haussler, S, Lauk, M, Guschlbauer, B, Koster, B, Kristeva-Feige, R, Timmer, J and Lucking, CH (2000) Tremor-correlated cortical activity detected by electroencephalography. Clin. Neurophysiol., Ill: 806-809. Hellwig, B, Haussler, S, Schelter, B, Lauk, M, Guschlbauer; B, Timmer, J and Lucking, CH (2001) Tremorcorrelated cortical activity in essential tremor. Lancet, 357: 519-523. Hirai, T, Miyazaki, M, Nakajima, H, Shibazaki, T and Ohye, C (1983) The correlation between tremor characteristics and the predicted volume of effective lesions in stereotaxic nucleus ventralis intermedius thalamotomy. Brain, 106: 1001-1018.

G. DEUSCHL ET AL.

Hirsch, EC, Mouatt, A, Faucheux, B, Bonnet, AM, Javoy, AF, Graybiel, AM and Agid, Y (1992) Dopamine, tremor, and Parkinson's disease. Lancet, 340: 125126. Homberg, V, Hefter, H, Reiners, K and Freund, HJ (1987) Differential effects of changes in mechanical limb properties on physiological and pathological tremor. J. Neurol. Neurosurg. Psychiatry, 50: 568-579. Hua, SE, Lenz, FA, Zirh, TA, Reich, SG and Dougherty, PM (1998) Thalamic neuronal activity correlated with essential tremor. J. Neurol. Neurosurg. Psychiatry, 64: 273-276. Hughes, AJ, Daniel, SE, Blankson, S and Lees, AJ (1993) A clinicopathologic study of 100 cases of Parkinson's disease. Arch. Neurol., 50: 140-148. Hunker, CJ and Abbs, JH (1990) Uniform frequency of parkinsonian resting tremor in the lips, jaw, tongue, and index finger. Mov. Disord., 5: 71-77. Hurtado, JM, Gray, CM, Tamas, LB and Sigvardt, KA (1999) Dynamics of tremor-related oscillations in the human globus pallidus: a single case study. Proc. Nati. Acad. Sci. USA, 96: 1674-1679. Hutchison, WO, Lozano, AM, Tasker, RR, Lang, AE and Dostrovsky, 10 (1997) Identification and characterization of neurons with tremor-frequency activity in human globus pallidus. Exp. Brain Res., 113: 557563. Hutchison, WO, Allan, RJ, Opitz, H, Levy, R, Dostrovsky, 10, Lang, AE and Lozano, AM (1998a) Neurophysiological identification of the subthalamic nucleus in surgery for Parkinson's disease. Ann. Neurol., 44: 622628. Hutchison, WO, Benko, R, Dostrovsky, JO, Lang, AE and Lozano, AM (1998b) Coherent relation of rest tremor and pallidal tremor cells in Parkinson's disaese patients. Mov. Disord., 13 (Suppl. 2): P3218. Jahnsen, H and Llinas, R (1984) Ionic basis for the electro-responsiveness and oscillatory properties of guinea-pig thalamic neurons in vitro. J. Physiol. (Lond.), 349: 227-247. Jankovic, J, McDermott, M, Carter, J, Gauthier, S, Goetz, C, Golbe, L, Huber, S, Koller, W, Olanow, C, Shoulson, I et al. (1990) Variable expression of Parkinson's disease: a base-line analysis of the DATATOP cohort. The Parkinson Study Group. Neurology, 40: 1529-1534. Jankovic, J, Schwartz, KS and Ondo, W (1999) Reemergent tremor of Parkinson's disease. 1. Neurol. Neurosurg. Psychiatry, 67: 646-650. Jellinger, KA (1999) Post mortem studies in Parkinson's disease - is it possible to detect brain areas for specific symptoms? J. Neural. Trans., 56 (Suppl.): 1-29. Jenkins, IH, Bain, PG, Colebatch, JG, Thompson, PD, Findley, LJ, Frackowiak, RS, Marsden,CD and

CLINICAL NEUROPHYSIOLOGY AND PATHOPHYSIOLOGY OF PARKINSONIAN TREMOR

Brooks, DJ (1993) A positron emission tomography study of essential tremor: evidence for overactivity of cerebellar connections. Ann. Neurol., 34: 82-90. Jung, R (1941) Physiologische Untersuchungen uber den Parkinsontremor und andere Zitterformen beim Menschen. Zsch. ges Neurol. Psychiat., 173: 263-332. Kassubek, J, Juengling, FD, Hellwig, B, Knauff, M, Spreer, J and LUcking, CH (2001) Hypermetabolism in the ventrolateral thalamus in unilateral Parkinsonian resting tremor: a positron emission tomography study. Neurosci. Lett., 304: 17-20. Kitano, H, Tanibuchi, I and Jinnai, K (1998) The distribution of neurons in the substantia nigra pars reticulata with input from the motor, premotor and prefrontal areas of the cerebral cortex in monkeys. Brain Res., 784: 228-238. Koller, W, Pahwa, R, Busenbark, K, Hubble, J, Wilkinson, S, Lang, A, Tuite, P, Sime, E, Lazano, A, Hauser, R, Malapira, T, Smith, D, Tarsy, D, Miyawaki, E, Norregaard, T, Kormos, T and Olanow, CW (1997) High-frequency unilateral thalamic stimulation in the treatment of essential and parkinsonian tremor. Ann. Neurol., 42: 292-299. Koller, WC, Vetere, OB and Barter, R (1989) Tremors in early Parkinson's disease. Clin. Neuropharmacol., 12: 293-297. Koller, WC, Lyons, KE, Wilkinson, SB and Pahwa, R (1999) Efficacy of unilateral deep brain stimulation of the VIM nucleus of the thalamus for essential head tremor. Mov. Disord., 14: 847-850. Koster, B, Lauk, M, Timmer, J, Winter, T, Guschlbauer, B, Glocker, FX, Danek, A, Deuschl, G and LUcking, CH (1998) Central mechanisms in human enhanced physiological tremor. Neurosci. Lett., 241: 135-138. Krack, P, Pollak, P, Limousin, P, Benazzouz, A and Benabid, AL (1997) Stimulation of subthalamic nucleus alleviates tremor in Parkinson's disease. Lancet, 350: 1675. Krack, P, Benazzouz, A, Pollak, P, Limousin, P, Piallat, B, Hoffmann, D, Xie, J and Benabid, AL (1998a) Treatment of tremor in Parkinson's disease by subthalamic nucleus stimulation. Mov. Disord., 13: 907914. Krack, P, Pollak, P, Limousin, P, Hoffmann, D, Xie, J, Benazzouz, A and Benabid, AL (1998b) Subthalamic nucleus or internal pallidal stimulation in young onset Parkinson's disease. Brain, 121: 451-457. Lamarre, Y (1984) Animal models of physiological, essential and parkinsonian-like tremors. In:LJ Findley and R Capildeo (Eds.), Movement Disorders: Tremor. London: Macmillan, pp. 183-194. Lance, JW. Schwab, RS and Peterson, EA (1963) Action tremor and the cogwheel phenomenon in Parkinson's disease. Brain, 86: 95-110.

393

Lang, AE, Duff, J, Saint-Cyr, JA, Trepanier, L, Gross, RE, Lombardi, W, Montgomery, E, Hutchinson, Wand Lozano, AM (1999) Posteroventral medial palIidotomy in Parkinson's disease. J. Neurol., 246 (Supp\. 2): 112841. Lauk, M, Koster, B, Timmer, J, Guschlbauer, B, Deuschl, G and LUcking, CH (1999a) Side-to-side correlation of muscle activity in physiological and pathological human tremors. Clin. Neurophysiol., 110: 1774-1783. Lauk, M, Timmer, J, LUcking, CH, Honerkamp, J and Deuschl, G (l999b) A software for recording and analysis of human tremor. Comput. Meth. Prog. Biomed., 60: 65-77. Lee, RG and Stein, RB (1981) Resetting of tremor by mechanical perturbations: a comparison of essential tremor and parkinsonian tremor. Ann. Neurol., 10: 523-531. Leenders, KL and Oertel, WH (2001) Parkinson's disease: clinical signs and symptoms, neural mechanisms, positron emission tomography, and therapeutic interventions. Neural. Plast., 8: 99-110. Lemstra, AW, Verhagen-Metman, L, Lee, Jl, Dougherty, PM and Lenz, FA (1999) Tremor-frequency (3-6 Hz) activity in the sensorimotor arm representation of the internal segment of the globus pallidus in patients with Parkinson's disease. Neurosci. Lett., 267: 129-132. Lenz, FA, Tasker, RR, Kwan, HC, Schnider, S, Kwong, R and Murphy, JT (1985) Cross-correlation analysis of thalamic neurons and EMG activity in parkinsonian tremor. Appl. Neurophysiol., 48: 305-308. Lenz, FA, Schnider, S, Tasker, RR, Kwong, R, Kwan, H, Dostrovsky, 10 and Murphy, JT (1987) The role of feedback in the tremor frequency activity of tremor cells in the ventral nuclear group of human thalamus. Acta. Neurochir., 39 (Supp!.): 54-56. Lenz, FA, Tasker, RR, Kwan, HC, Schnider, S, Kwong, R, Murayama, Y, Dostrovsky, 10 and Murphy, JT (1988) Single unit analysis of the human ventral thalamic nuclear group: correlation of thalamic "tremor cells" with the 3-6 Hz component of parkinsonian tremor. J. Neurosci., 8: 754-764. Lenz, FA, Kwan, HC, Dostrovsky, JO and Tasker, RR (1989) Characteristics of the bursting pattern of action potentials that occurs in the thalamus of patients with central pain. Brain Res., 496: 357-360. Lenz, FA, Kwan, HC, Dostrovsky, JO, Tasker, RR, Murphy, IT and Lenz, YE (1990) Single unit analysis of the human ventral thalamic nuclear group. Activity correlated with movement. Brain, 113: 1795-1821. Lenz, FA, Vitek, JL and DeLong, MR (1993) Role of the thalamus in parkinsonian tremor: evidence from studies in patients and primate models. Stereotact. Func. Neurosurg., 60: 94-103.

394 Lenz, FA, Kwan, HC, Martin, RL, Tasker, RR, Dostrovsky, JO and Lenz, YE (1994) Single unit analysis of the human ventral thalamic nuclear group. Tremorrelated activity in functionally identified cells. Brain, 117: 531-543. Lenz, FA, Normand, SL, Kwan, HC, Andrews, D, Rowland, LH, Jones, MW, Seike, M, Lin, YC, Tasker, RR, Dostrovsky, JO et al. (1995) Statistical prediction of the optimal site for thalamotomy in parkinsonian tremor. Mov. Disord., 10: 318-328. Levy, R, Hutchison, WD, Lozano, AM and Dostrovsky, JO (2002) Synchronized neuronal discharge in the basal ganglia of parkinsonian patients is limited to oscillatory activity. J. Neurosci., 22: 2855-2861. LeWitt, P, Oakes, D and Cui, L (1997) The need for levodopa as an end point of Parkinson's disease progression in a clinical trial of selegiline and alphatocopherol. Parkinson Study Group. Mov. Disord., 12: 183-189. Limousin, P, Pollak, P, Benazzouz, A, Hoffmann, D, Le, BJ, Broussolle, E, Perret, JE and Benabid, AL (1995) Effect of parkinsonian signs and symptoms of bilateral subthalamic nucleus stimulation. Lancet, 345: 91-95. Limousin, P, Krack, P, Pollak, P, Benazzouz, A, Ardouin, C, Hoffmann, D and Benabid, AL (1998) Electrical stimulation of the subthalamic nucleus in advanced Parkinson's disease. N. Engl. J. Med., 339: 1105-1111. Limousin-Dowsey, P, Pollak, P, VanBlercom, N, Krack, P, Benazzouz, A and Benabid, A (1999) Thalamic, subthalamic nucleus and internal pallidum stimulation in Parkinson's disease. J. Neurol., 246: S42-45. Lin, YC and Lenz, FA (1993) Effective response evoked by microstimulation of thalamus nuclei in patients with tremor. Chin. Med. J. Engl., 106: 372-374. Llinas, Rand Volkind, RA (1973) The olivo-cerebellar system: functional properties as revealed by harmalineinduced tremor. Exp. Brain Res., 18: 69-87. Llinas, R and Yarom, Y (1981) Electrophysiology of mammalian inferior olivary neurons in vitro. Different types of voltage-dependent ionic conductances. J. Physiol. (Lond.), 315: 549-567. Llinas, R (1984) Rebound excitation as the physiological basis for tremor: a biophysical study of the oscillatory properties of mammalian central neurons in vitro. In: LJ Findley and R Capildeo (Eds.), Movement Disorders, Tremor. London: Macmillan, pp. 339-351. Louis, ED, Tang, MX, Cote, L, Alfaro, B, Mejia, Hand Marder, K (1999) Progression of parkinsonian signs in Parkinson disease. Arch. Neurol., 56: 334-337. Lozano, AM, Lang, AB and Hutchison, WD (1998) Pallidotomy for tremor. Mov. Disord., 13: 107-110. Lozano, AM, Hutchison, WO, Tasker, RR, Lang, AB, Junn, F and Dostrovsky, JO (1999) Microelectrode

G. DEUSCHL ET AL.

recordings define the ventral posteromedial pallidotomy target. Stereotact. Funct. Neurosurg., 71: 153-163. Magnin, M, Morel, A and Jeanmonod, D (2000) Singleunit analysis of the pallidum, thalamus and subthalamic nucleus in parkinsonian patio Neuroscience, 96: 549-564. Makela, JP, Hari, P, Karhu, J, Salrnelin, Rand Teravainen, H (1993) Suppression of magnetic mu rhythm during parkinsonian tremor. Brain Res., 617: 189-193. Mandir, AS, Rowland, LH, Dougherty, PM and Lenz, FA (1997) Microelectrode recording and stimulation techniques during stereotactic procesures in the thalamus and pallidum. In: JA Obeso, MR Delong and C Ohye (Eds.), The Basal Ganglia and New Surgical Approaches for Parkinson's Disease 74. Philadelphia: Lippicott-Raven, pp. 159-165. Middleton, FA and Strick, PL (2000) Basal ganglia and cerebellar loops: motor and cognitive circuits. Brain Res. Rev., 31: 236-250. Mirna, T and Hallett, M (1999) Corticomuscular coherence: a review. J. Clin. Neurophysiol., 16: 501-511. Mirna, T, Simpkins, N, Oluwatimilehin, T and Hallett, M (1999) Force level modulates human cortical oscillatory activities. Neurosci. Lett., 275: 77-80. Narabayashi, H (1998) Stereotactic surgery and Parkinson's disease. Stereotact. Funct. Neurosurg., 70: 114121. Nieuwboer, A, De Weerdt, W, Dom, R and Lesaffre, E (1998) A frequency and correlation analysis of motor deficits in Parkinson patients. Disabil. Rehabil., 20: 142-150. Nini, A, Feingold, A, Slovin, H and Bergman, H (1995) Neurons in the globus pallidus do not show correlated activity in the normal monkey, but phase-locked oscillations appear in the MPTP model of parkinsonism. J. Neurophysiol., 74: 1800--1805. Ohye, C, Shibazaki, T, Hirai, T, Wada, H, Kawashima, Y, Hirato, M and Matsumura, M (1988) A special role of the parvocellular red nucleus in lesion-induced spontaneous tremor in monkeys. Behav. Brain Res., 28: 241243. Ohye, C, Shibazaki, T, Hirai, T, Kawashima, Y, Hirato, M and Matsumura, M (1993) Tremor-mediating thalamic zone studied in humans and in monkeys. Stereotact. Funct. Neurosurg., 60: 136-145. Pare, D, Curro'Dossi, Rand Steriade, M (1990) Neuronal basis of the parkinsonian resting tremor: a hypothesis and its implications for treatment. Neuroscience, 35: 217-226. Pascual-Leone, A, Valls-Sole, J, Toro, C, Wassermann, EM and Hallett, M (1994) Resetting of essential tremor and postural tremor in Parkinson's disease with

CLINICAL NEUROPHYSIOLOGY AND PATHOPHYSIOLOGY OF PARKINSONIAN TREMOR

transcranial magnetic stimulation. Muscle Nerve, 17: 80D-807.

Paulus, W and Jellinger, K (1991) The neuropathologic basis of different clinical subgroups of Parkinson's disease. J. Neuropathol. Exp. Neurol., 50: 743-755. Pollak, P, Benabid, AL, Krack, P, Limousin, P and Benazzouz, A (1998) Deep brain stimulation. In: J Jankovic and E Tolosa (Eds.), Parkinson's Disease and Movement Disorders (3rd ed.). Baltimore: Williams & Wilkins, pp. 1085-1102. Pollock, LJ and Davis, L (1930) Muscle tone in parkinsonian states. Arch. Neurol. Psychiat., 23: 303-309. Putnam, TJ (1940) The operative treatment of diseases characterized by involuntary movements (tremor, athetosis). Assoc. Res. Nerv. Ment. Dis., 21: 666-696. Rack, PM and Ross, HF (1986) The role of reflexes in the resting tremor of Parkinson's disease. Brain, 109: 115-141. Raethjen, J, Lindemann, M, Schmaljohann, H, Wenzelburger, R, Pfister, G and Deuschl, G (2000a) Multiple oscillators are causing parkinsonian and essential tremor. Mov. Disord., 15: 84-94. Raethjen, J, Pawlas, F, Lindemann, M, Wenzelburger, R and Deuschl, G (2000b) Determinants of physiologic tremor in a large normal population. Clin Neurophysiol., 111: 1825-1837. Raeva, SN, Vainberg, NA and Dubynin, VA (1998) Spontaneous activity of individual neurons in the human ventrolateral thalamus during changes in the functional state of the brain. Neurosci. Behav. Physiol., 28: 513-520. Raeva, S, Vainberg, Nand Dubinin, V (1999a) Analysis of spontaneous activity patterns of human thalamic ventrolateral neurons and their modifications due to functional brain changes. Neuroscience, 88: 365-376. Raeva, S, Vainberg, N, Tikhonov, Y and Tsetlin, I (1999b) Analysis of evoked activity patterns of human thalamic ventrolateral neurons during verbally ordered voluntary movements. Neuroscience, 88: 377-392. Raz, A, Feingold, A, Zelanskaya, V, Vaadia, E and Bergman, H (1996) Neuronal synchronization of tonically active neurons in the striatum of normal and parkinsonian primates. J. Neurophysiol., 76: 20832088. Raz, A, Vaadia, E and Bergman, H (2000) Firing patterns and correlations of spontaneous discharge of pallidal neurons in the normal and the tremulous l-methyl4-phenyl-l,2,3,6-tetrahydropyridine vervet model of parkinsonism. 1. Neurosci., 20: 8559-8571. Remy, P, De Recondo, A, Defer, G, Loc'h, C, Amarenco, P, Plante-Bordeneuve, V, Dao-Castellana, MH, Bendriem, B, Crouzel, C, Clanet, M et a!. (1995) Peduncular

395

'rubral' tremor and dopaminergic denervation: a PET study (see comments). Neurol., 45: 472-477. Rodriguez, MC, Guridi, OJ, Alvarez, L, Mewes, K, Macias, R, Vitek, J, DeLong, MR and Obeso, JA (1998) The subthalamic nucleus and tremor in Parkinson's disease. Mov. Disord., 13: 111-118. Spieker, S, Boose, A, Jentgens, C and Dichgans, J (1995) Long-term tremor recordings in parkinsonian and essential tremor. J. Neural. Trans., 46 (Supp!.): 339-349. Spieker, S, Strole, V, Sailer, A, Boose, A and Dichgans, J (1997) Validity of long-term electromyography in the quantification of tremor. Mov. Disord., 12: 985-991. Steriade, M and Deschenes, M (1984) The thalamus as a neuronal oscillator. Brain Res., 320: 1-63. Strick, PL (1985) How do the basal ganglia and cerebellum gain access to the cortical motor areas? Behav. Brain Res., 18: 107-123. Taha, JM, Favre, J, Baumann, TK and Burchiel, KJ (1997a) Tremor control after pallidotomy in patients with Parkinson's disease: correlation with microrecording findings. J. Neurosurg., 86: 642-647. Taha, JM, Favre, J and Burchiel, KJ (1997b) Infrequent types of pallidal discharges during pallidotomy; Stereotact. Funct. Neurosurg., 68: 231-235. Tasker, RR, DeCarvalhom, GC, Li, CS and Kestle, JR (1996) Does thalamotomy alter the course of Parkinson's disease? Adv. Neurol., 69: 563-583. Tass, P, Rosenblum, MG, Weule, J, Kurths, J, Pikovsky,A, Volkmann, J, Schnitzler, A and Freund, H-J (1998) Detection of n: m phase locking from noisy data: application to magnetoencephalography. Phys. Rev. Lett., 81: 3291-3294. Timmer, J, Gantert, C, Deuschl, G and Honerkamp, J (1993) Characteristics of hand tremor time series. Biol. Cybem., 70: 75-80. Timmer, J, Lauk, M, Pfleger, Wand Deuschl, G (1998a) Cross-spectral analysis of physiological tremor and muscle activityI Theory and application to unsynchronized electromyogram. BioI. Cybem., 78: 349-357. Timmer, J, Lauk, M, Pfleger, W and Deuschl, G (1998b) Cross-spectral analysis of physiological tremor and muscle activity. II. Application to synchronized electromyogram. BioI. Cybem., 78: 359-368. Timmer, J, HauBler, S, Lauk, M and LUcking, CH (2000) Pathological tremors: deterministic chaos or nonlinear stochastic oscillators? Chaos, 10: 278-288. Volkmann,J, Joliot, M, Mogilner, A, Ioannides, AA, Lado, F, Fazzini, E, Ribary, U and Llinas, R (1996) Central motor loop oscillations in parkinsonian resting tremor revealed by magnetoencephalography. Neurology, 46: 1359-1370.

396 Volkmann, J, Sturm, V, Weiss, P, Kappler, J, Voges, J, Koulousakis, A, Lehrke, R, Hefter, H and Freund, HJ (1998) Bilateral high-frequency stimulation of the internal globus pallidus in advanced Parkinson's disease. Ann. Neurol., 44: 953-961. Walshe, FMR (1924) Observations on the nature of the muscular rigidity of paralysis agitans, and on its relationship to tremor. Brain, 47: 159-177. Wichmann, T, Bergman, H and DeLong, MR (1994) The primate subthalamic nucleus. Ill, Changes in motor behavior and neuronal activity in the internal pallidum induced by subthalamic inactivation in the MPTP model of parkinsonism. 1. Neurophysiol., 72: 521530. Wills, AJ, Jenkins, IH, Thompson, PD, Findley, U and Brooks, DJ (1995) A positron emission tomography

G. DEUSCHL ET AL.

study of cerebral activation associated with essential and writing tremor. Arch. Neurology, 52: 299-305. Wills, AJ, Thompson, PD, Findley, U and Brooks, DJ (1996) A positron emission tomography study of primary orthostatic tremor. Neurol., 46: 747-752. Wilms, H, Sievers, J and Deuschl, G (1999) Animal models of tremor. Mov. Disord., 14: 557-571. Winfree, AT (1980) The geometry of biological time. New York: Springer Verlag. Zimmermann, R, Deuschl, G, Hornig, A, Schulte-Monting, J, Fuchs, G and Lucking, CH (1994) Tremors in Parkinson's disease: Symptom analysis and rating. Clin. Neuropharmacol., 17: 303-314. Zirh, TA, Lenz, FA, Reich, SG and Dougherty, PM (1998) Patterns of bursting occurring in thalamic cells during parkinsonian tremor. Neuroscience, 83: 107-121.

Movement Disorders Handbook of Clinical Neurophysiology, Vol. I M. Hallett (Ed.) © 2003 Elsevier B. V. All rights reserved

397 CHAPTER 25

Uncommon tremors C.H. Lticking* and B. Hellwig Neurologische Universitiitsklinik, Neurozentrum, Breisacher Str. 64, D-79106 Freiburg, Germany

Besides physiological, essential and parkinsonian tremor, there are tremor disorders which are less common in the clinical routine of neurologists. Nevertheless, these tremors deserve attention, since they may be disabling for the patient. Moreover, as in the case of drug-induced tremors, they may be quite frequently encountered outside departments of neurology. These 'uncommon tremors' are reviewed in the following chapters.

25.1. Primary orthostatic tremor Primary orthostatic tremor is a disorder of middleaged or elderly persons characterized by a feeling of unsteadiness in the legs and a fear of falling while standing. There is a barely visible tremor of around 16 Hz in the leg muscles. The disorder is frequently mistaken for somatofonn or psychogenic. Heilman (1984) first described this unusual tremor and coined the term "orthostatic" tremor. It is more common than generally believed. The clinical features are a feeling of unsteadiness, irregular swaying, and mostly only fine ripples of leg muscles when standing. Occasionally a postural hand tremor of high frequency and small amplitude can be seen. The unsteadiness can be overcome by taking some steps. Walking is unaffected. The predominant electrophysiological finding is a highly synchronized rhythmic EMG activity of around 16 Hz (13-18 Hz) in all leg and trunk muscles while standing (Fig. 1). Muscles of the upper limbs and cranial muscles may also be involved (Koster et al., 1999). EMG bursts are biphasic or short and polyphasic, their duration

* Correspondence to: Prof. Dr. Dr. h.c. C.H. LUcking, Neurologische Universitatsklinik, Neurozentrum, Breisacher Str. 64, D-79106 Freiburg, Germany. E-mail address:[email protected] Tel.: ++497612705306; fax: ++49761 2705210.

ranges between 10 and 80 ms (Deuschl et aI., 1987; Wu et al., 2001). The activation of agonists and antagonists may be both alternating and synchronous (Thompson et aI., 1986; Deuschl et al., 1987; Britton et al., 1992). McAuley et al. (2000) demonstrated that the phase relation between EMG bursts in agonists and antagonists is posture-dependent and may change from patient to patient, but is constant for the same posture in the same patient. The tremor bursts are highly correlated in all muscles of the right and left side, of the upper and lower limbs as well as in neck and cranial muscles (Figs. 1 and 2) (Britton et aI., 1992; Koster et aI., 1999). A mechanical tremor analysis at the patella may reveal a subharmonic 8 Hz tremor caused by alternating large and small amplitudes of the 16 Hz tremor bursts. Single motor units mostly fire at a frequency of 8 Hz, but only at the time of tremor bursts (Deuschl et aI., 1987). There are a number of hypotheses concerning the nature of the unsteadiness in primary orthostatic tremor. On one hand, unsteadiness might be caused by partially fused muscle contractions due to the high frequency of the tremulous muscle activity. On the other hand, unsteadiness could be mainly related to the mechanical 8 Hz tremor that can be recorded from the patella (Deuschl et al., 1987). Finally, it has been suggested that the sensation of unsteadiness arises from the tremulous disturbance of proprioceptive afferent activity from the legs (Fung et al., 2001). The occurrence of the rhythmic EMG activity is not strictly related to stance. Identical bursting EMG can be recorded from leg muscles when sitting or lying if the muscles are isometrically activated. The same is true for isometrically activated arm and hand muscles as well as neck and cranial muscles /Uncini et aI., 1989; Walker et aI., 1990; Boroojerdi et al., 1999; Koster et aI., 1999). The major role of isometric innervation makes the term "orthostatic"

398 Quadriceps muscle - right side

~

::l

:!

l::~a

':!.... 1.E+Ol

~e.

s Quadriceps muscle - left side

::l

~

f

-50

Ie.

S

0.5

1.E-01 1.E-02 - - - - . - - - r - - - - - ' - - - , - - , - - - , - - - - - - - " Hz 15 25 o 30 10 20 5

d

rn

~

50

o

C

':!.... 1.E+Ol

1~~ -150

I

Spectra of the EMG time series

Cl

.s

-50 ~w

:> a

1.E+00

C.H. LUCKING AND B. HELLWIG

1

1.5

1.E+00 1.E-Ol 1.E-02 +---,---..,--.:.....,----.---.-------, Hz

o

2

5

10

15

20

25

30

Coherence between the two EMGs

e

1

2l

0.8

c:

0.6

CIl

0.4

l!1 s: 0

o

0.2 0 0

5

10

15

20

25

30

Hz

Fig. 1. (a, b) Surface EMG recordings from the right and left quadriceps muscle in a patient with primary orthostatic tremor. (c, d) Spectra of both EMG time series showing a distinct peak at about 15 Hz. (e) Coherence between the two EMG time series showing a highly significant peak at the tremor frequency. The horizontal line indicates the level of significance

(p
and the linkage to stance regulating spinal mechanisms somewhat doubtful. But, as the main complaint of patients is linked to standing, the clinical use of the term "orthostatic tremor" remains valuable. The pathophysiology of primary orthostatic tremor is yet unclarified. A peripheral type of tremor due to spontaneous rhythmic firing of motoneurons (Uncini et al., 1989) is not very likely. In contrast to the 16 Hz tremor, single motor units are mostly firing at only 8 Hz. Underlying oscillatory activity of peripheral loops is also unlikely because there is no evident tremor resetting by electrical stimulation of the peroneal or tibial nerve (Thompson et al., 1986; Deuschl et al., 1987; Britton et al., 1992; Sander et al., 1998). Resetting studies using transcranial magnetic stimulation are somewhat controversial. In

some studies, a modulation of the timing of tremor bursts by motor cortex stimulation could be demonstrated (Tsai et al., 1998; Manto et al., 1999; Pfeiffer et al., 1999), in others not (Mills and Nithi, 1997; Wu et al., 2001). Pfeiffer et a1. (1999) suggested that the use of different stimulus intensities might explain why transcranial magnetic stimulation is not always able to modulate orthostatic tremor. The high coherence of tremor bursts in all muscles supports the assumption of a common central pacemaker (Koster et al., 1998). PET studies revealed abnormal bilateral overactivity of cerebellar structures which is in line with findings in other tremulous disorders and does not provide any specific information about the tremor generator (Wills et al., 1996). However, the finding that orthostatic tremor can be reset by electrical stimulation of the posterior fossa was also

399

UNCOMMON TREMORS

gastrocnemius / sternocleidomastoid muscle

sternocleidomastoid / paravertebral muscle, lumbar level

0.8

0.8

c

06

06

CD

0.4

0.4

0 (J

0.2

0.2

CD

o ~

s:

5

10

15

20

25

30

5

gastrocnemius / genioglossus muscle

0.8

08

0.6

0.6

CD s: 0.4 0.2

0.4

o

c::

15

20

25

30

genioglossus muscle / paravertebral muscle, lumbar level d

b CD

10

~

0 (J

0.2 5

10 15 20 frequency [Hz]

25

30

5

10

15 20 frequency [Hz]

25

30

Fig. 2. Coherences between EMG time series recorded from various limb, trunk and cranial muscles of a patient with primary orthostatic tremor. Highly significant peaks are visible at the tremor frequency (13 to 14 Hz), in most cases also at its first harmonic. The horizontal line indicates the level of significance (p
considered as evidence that cerebellar or brainstem structures are involved in the generation of orthostatic tremor (Wu et al., 2001). It is still a matter of debate whether primary orthostatic tremor is a subtype of essential tremor. It has been observed in patients with additional postural hand tremor of lower frequency and with a broad family history (Koller et al., 1987; Papa and Gershanik, 1988; FitzGerald and Jankovic, 1991). But the clinical and physiological properties are quite different between essential tremor and primary orthostatic tremor (Walker et al., 1990; Britton et al., 1992; McManis and Sharbrough, 1993). There are large families of essential tremor without any case of primary orthostatic tremor (Bain et aI., 1994). The frequency of essential tremor is clearly slower, the coherence of EMG bursts between the left and right side and the upper and lower limbs is low in contrast to primary orthostatic tremor. In addition, the pharmacological effects are different and alcohol responsiveness has not been recognized. The com-

plaints of patients are only related to unsteadiness when standing. Primary orthostatic tremor is successfully treated in many cases with clonazepam and less with primidone, whereas beta blockers are rarely effective (Koller et aI., 1987; FitzGerald and Jankovic, 1991; Willeit et aI., 1991). There are some positive reports on gabapentin (Evidente et aI., 1998; Onofrj et al., 1998), L-DOPA(Wills et aI., 1999), and pramipexole (Finkel, 2000).

25.2. Holmes' tremor Holmes' tremor is an infrequent movement disorder which was first reported by Benedikt in 1889. It has been classified in the past as midbrain tremor (Sarnie et aI., 1990), rubral tremor (MossutoAgatiello et al., 1993), peduncular tremor (Remy et al., 1995), thalamic tremor (Miwa et aI., 1996), Benedikt's syndrome and myorhythmia (Masucci et al., 1984). In order to get a common denomination

400

and to avoid a strong and misleading topographical assignment the Movement Disorder Society (Deuschl et al., 1998a) recommended the term "Holmes' tremor". Holmes (1904) described a kinetic tremor caused by a dysfunction of the cerebello-rubral system. Holmes' tremor appears at rest as well as during maintained posture and during goal-directed movements and may be described as a resting tremor with marked postural and kinetic components. The frequency is slower than in most other tremor types and ranges between 2.5 and 4.5 Hz (Fig. 3a--c). Rhythm and manifestation are often irregular. The EMG burst duration is on average 150 to 170 ms (Milanov, 2002). Agonist and antagonist muscles are usually activated in an alternating fashion (Ghika et al., 1994; Milanov, 2002). Weight loading to the outstretched hands does not decrease the tremor frequency. Holmes' tremor is in most cases a major

c.a

LUCKING AND B. HELLWIG

disabling disease. In addition to the tremor, patients suffer from cerebellar, midbrain, and/or thalamic signs such as ataxia, diplopia, paresis, hemianesthesia, dystonic posture, and rigidity. The complex tremor syndrome develops after lesions in the midbrain area involving the red nucleus and/or the superior cerebellar peduncle or in the posterior thalamus (Mossuto-Agatiello et al., 1993). Most often there is clinical and radiographic evidence of vascular infarct, tumor, head trauma, multiple sclerosis, arteriovenous malformation, or abscess. There is a typical delay of weeks, months or even years between the time of the lesion and the occurrence of tremor. The pathophysiological mechanisms of Holmes' tremor are not well understood. Lesions involving the superior cerebellar peduncle in the midbrain have been shown to induce a dopaminergic denervation in the striatum, presumably caused by damage to

Fig. 3. (a) Surface EMG recording under resting condition from the right wrist extensor of a patient with Holmes' tremor. (b, c) The spectra of the accelerometer and the EMG show peaks at 3 Hz. (d, e, f) After deep brain stimulation in the left nucleus ventralis intermedius of the thalamus the 3 Hz tremor has disappeared.

UNCOMMON TREMORS

nigrostriatal dopaminergic fibers (Remy et al., 1995). This might contribute to the rest component of the tremor. The postural component of Holmes' tremor is probably caused by lesions to the rubroolivo-cerebello-rubral loop (Narabayashi, 1986) which has been demonstrated also by experiments in monkeys (Larochelle et al., 1970). Isolated lesions of the red nucleus are not sufficient to induce Holmes' tremor. There are even reports about cases without obvious involvement of the red nucleus. For instance, Holmes' tremor was observed in a patient several months after pontine hemorrhage and subsequent hypertrophic olivary degeneration (Shepherd et al., 1997). The typical delayed onset of tremor months or years after the lesion favors the role of secondary degeneration which leads to olivary hypertrophy. In general, a lesion to both the dopaminergic and the cerebello-thalamic system seems to be responsible for the manifestation of Holmes' tremor (Deuschl et al., 1999). However, there are exceptions to this rule. Miwa et al. (1996) reported two patients with a circumscribed lesion of the postero-lateral thalamus that provoked a contralateral Holmes' tremor without any apparent involvement of brainstem, cerebellum, or the cerebellar outflow tract to the ventro-lateral thalamic nucleus. Therapeutic approaches are based on these hypothesized pathomechanisms. Dopaminergic and anticholinergic substances have shown good results (Findley and Gresty, 1980; Berkovic and Bladin, 1984; Samie et al., 1990; Krack et al., 1994). Clonazepam as benzodiazepine derivative and GABA agonist has led to a considerable improvement of all aspects of tremor in a patient who previously did not respond to levodopa (Jacob and Pratap Chand, 1998). Stereotactic lesions or chronic stimulation of the ventrolateral thalamic nucleus can totally alleviate Holmes' tremor (Fig. 3d-t). This approach might interrupt the tremorogenic oscillation of the cerebello-thalamic system (Andrew et al., 1982; Shepherd et al., 1997; Pahwa et al., 2002). 25.3. Palatal tremor Palatal tremor consists of rhythmic movements of the soft palate. It has also been called: rhythmic palatal myoclonus, oculo-palatal myoclonus, palatal nystagmus, brainstem myorhythmia, or segmental

401

cranial tremor. The Movement Disorder Society in 1990 and 1998 classified it as tremor and proposed the term palatal tremor (Deuschl et al., 1998a). Palatal tremor is a rare disease. Nearly 300 cases have been reported in the literature until 1990 (Deuschl et al., 1990). Palatal tremor can be subdivided into two types: essential (idiopathic) and symptomatic palatal tremor in a relation of 1:3. The mean age of onset of the symptomatic form is 45 years, of the essential type 25 years. The age distribution in symptomatic palatal tremor extends from childhood to old age, the onset of the essential type is rarely after the age of 40 years (Deuschl et al., 1990). The predominant features of palatal tremor are spontaneous, rhythmic, jerky movements of the soft palate; the soft palate and the uvula are drawn upward and backward whereas the posterior wall of the pharynx moves forward. In cases of unilateral tremor, the palate and uvula are drawn to one side. Other brainstem-innervated muscles can be involved, occasionally there is a postural tremor of the limbs in clear synchronization to the jerks of the palatal tremor. The palatal tremor itself is usually a minor annoyance and most often ignored by the patient, but related symptoms such as oscillopsia, ear click, and postural tremor can provoke major complaints; laryngeal, palatal, pharyngeal and diaphragmatic involvement may produce a broken speech pattern. The two types of palatal tremor differ in various aspects: essential palatal tremor is most often restricted to the soft palate and mainly involves the tensor veli palatini muscles which are innervated by the trigeminal nerve (Strutz et al., 1988; Deuschl et al., 1991, 1994a). Occasionally pharyngeal, rarely other muscle groups are involved. About 90% of the patients presenting essential palatal tremor suffer from psychologically distressing ear clicks which can be heard without amplification by the examiner. The ear clicks are probably caused by a sudden opening of the eustachian tube. The tremor frequency ranges mainly between 60 and 180 per minute with a peak below 120 jerks per minute. Symptomatic palatal tremor mainly involves the levator veli palatini muscles which are innervated by the glossopharyngeal or facial nerve (Strutz et al., 1988; Deuschl et al., 1994a). In addition to the soft palate, other groups of muscles of the pharynx, lower face (chin, perioral), eyes (pendular nys-

402 tagmus), larynx, diaphragm, and rarely the upper face, trunk, and limbs may be affected. Some of these muscles are easily accessible by surface electromyography. Related and disabling symptoms such as ear clicks, oscillopsia, or postural tremor accompany the symptomatic palatal tremor in less than 10%. The main tremor frequency is between 100 and 200 per minute with a peak above 120 jerks per minute. In less than 20% the movements are strictly unilateral. Sleep, coma and barbiturate anesthesia can modify both types; symptomatic palatal tremor is distinctly more resistant to these conditions. The etiology of palatal tremor is still not clarified. In essential palatal tremor no preceding diseases are present. Magnetic resonance imaging shows no structural abnormalities, particularly the medulla oblongata and the cerebellum appear normal. Stimulation of trigeminal afferents does not modulate the tremor rhythm. Remote effects on the tonic electromyographic activity of the upper and lower extremities could not be demonstrated. Monosynaptic and oligosynaptic brain stem reflexes as well as motor learning tasks involving the cerebellum are essentially normal (Deuschl et aI., 1994b, 1996). Symptomatic palatal tremor is preceded by cerebellar or brain stem lesions caused predominantly by cerebrovascular disease and to a lesser extent by infections, multiple sclerosis, neoplasm, trauma, malformations or degenerative disorders. The tremor mostly appears with a delay of weeks or months after the onset of the underlying neurological disorder. The lesion causing symptomatic palatal tremor is thought to affect the dentato-olivary tract, i.e. a pathway originating in the dentate nucleus, crossing to the contralateral side via the superior cerebellar peduncle, passing the red nucleus, and then descending ipsilaterally through the central tegmental tract to the inferior olive (Lapresle and Ben Hamida, 1965, 1968, 1970). Lesions of the dentato-olivary tract give rise to (transsynaptic) hypertrophic degeneration of the inferior olive which can be visualized by magnetic resonance imaging (Yokota et aI., 1989; Deuschl et al., 1994b). In cases of unilateral palatal tremor, olivary hypertrophy is usually contralateral to the tremor side, in bilateral palatal tremor, both unilateral and bilateral olivary hypertrophy have been described (Rondot and Ben Hamida, 1968; Deuschl et aI., 1994b). Olivary

C.H. LUCKING AND B. HELLWIG

hypertrophy is obviously not necessary for the maintenance of symptomatic palatal tremor, since the initial olivary hypertrophy seems to be progressively transformed into olivary atrophy despite unchanged clinical tremor manifestation (Nishie et al.,2002). The significance of olivary hypertrophy on the pathophysiological level is not precisely known. However, recordings in brainstem slices isolated from cerebellar afferents have demonstrated that olivary neurons tend to oscillate (Llinas, 1984). Such oscillatory discharges could be responsible for tremor generation. They might also be the basis of the increased glucose metabolism found in the medulla of patients with symptomatic palatal tremor (Dubinsky et aI., 1991). It has been suggested that rhythmic activity in the inferior olive spreads to motor neurons via cerebelloreticular and cerebellospinal connections (Deuschl et aI., 1994b). However, there is also evidence for the involvement of the thalamus and the motor cortex in symptomatic palatal tremor. Rondot et al. (1965) found rhythmic activity related to palatal tremor in the posterior ventral thalamus. Guillain (1938) observed that the palatal tremor as well as the accompanying tremor of other muscle groups disappear after damage of the corticonuclear or corticospinal tract. Moreover, symptomatic palatal tremor can be modulated by magnetic stimulation of the motor cortex (Chen et aI., 2000). Thus, it is conceivable that the generation of symptomatic palatal tremor involves a pathway originating in the inferior olive, reaching the ipsilateral thalamus via the contralateral cerebellum, leading via thalamocortical fibers to the ipsilateral motor cortex and finally reaching contralateral motor neurons via the corticonuclear and corticospinal tracts. This would explain the fact that a unilateral palatal tremor appears ipsilateral to the cerebellar lesion side and contralateral to the side of the hypertrophic inferior olive. Stimulation of trigeminal afferents does not modulate symptomatic palatal tremor. Brain stem reflexes and motor learning are abnormal, probably due to the underlying lesions of brain stem or cerebellum. In contrast to essential palatal tremor, symptomatic palatal tremor has been shown to exert remote effects on muscles of the upper and lower extremities even in cases in which a limb tremor is not apparent clinically (Nagaoka and Narabayashi, 1984; Deuschl et aI., 1994b).

UNCOMMON TREMORS

As the palatal tremor itself does not provoke any major complaint the therapeutic approaches are mainly directed to relieve related disorders like broken speech pattern, oscillopsia, and especially ear clicks. There are no consistent results of drug administrations for both types of palatal tremor: phenytoin, barbiturates, benzodiazepines, 5-HDP, and trihexyphenidyl have occasionally shown some relief, double blind studies are not available. Surgical interventions (tamponade of eustachian tube, perforation of tympanic membrane, cutting of levator or tensor veli palatini muscle) were inconclusive for the relief of the distressing ear clicks. Successful treatment with local injection of botulinum toxin into the tensor veli palatini muscle for the relief of ear clicks has been reported (Deuschl et aI., 1991). Summarizing, the clinical diagnosis of palatal tremor is based on the observation of spontaneous, rhythmic, tremor-like, low frequency movements of the soft palate and of adjacent muscle groups which occasionally involve also trunk and limb muscles. Essential palatal tremor is characterized by the restriction to palatal and pharynx muscles, distressing ear clicks in the majority of cases and the absence of preceding brain stem lesions. Symptomatic palatal tremor is more often accompanied by tremor of other muscle groups, only rarely by ear clicks and oscillopsia. Typically, the symptomatic form is preceded by brain stem disorders, and magnetic resonance imaging may reveal signs of a hypertrophic inferior olive. Essential and symptomatic palatal tremor are mostly longstanding and probably life long diseases. Only in rare cases of the essential type of tremor remissions have been described. 25.4. Cerebellar tremor

According to the consensus statement of the Movement Disorder Society on tremor (Deuschl et al., 1998a), pure or dominant intention tremor which may occur uni- or bilaterally is the hallmark of cerebellar tremor. Intention tremor is characterized by increasing tremor amplitudes in the termination phase of a target-directed movement. Other types of tremor such as postural hand tremor (Fahn, 1984), leg tremor (Silfverskiold, 1977), head tremor (Finsterer et al., 1996), axial postural tremor (Brown et aI., 1997) or stance tremor (Diener et al., 1984) may be present. However, they are only considered to be of

403

cerebellar ongm if additional clinical signs of cerebellar disease such as dysmetria or ataxia are obvious. Rest tremor is not a feature of cerebellar tremor. The frequency of cerebellar tremor is usually below 5 Hz (Fig. 4a-c) (Deuschl et al., 1998a). The duration of EMG bursts ranges between 50 and 300 ms, agonists and antagonists may be activated both in a synchronous and an alternating fashion (Rondot and Bathien, 1995). The frequency of cerebellar postural tremor does not decrease with mass loading to the hands. A differential diagnosis of cerebellar tremor may be essential tremor with marked intention tremor. Intention tremor occurs in about 50% of the patients with essential tremor (Deuschl et al., 2000). Other signs of cerebellar dysfunction, e.g. an ataxic gait (Stolze et al., 2001) or dysmetric ballistic arm movements (Koster et al., 2002), may also be present in essential tremor patients. Thus, cerebellar and essential tremor may be partly overlapping disorders. The pathophysiology of cerebellar tremor has not been clarified. Some evidence can be gained from studies on target-directed movements using the paradigm of ballistic arm movements. EMG analysis of ballistic arm movements (e.g. flexion of the elbow joint by a certain angular distance) shows a typical triphasic pattern: the initial activation of the agonist muscle is followed by an EMG burst in the antagonist muscle and a second activation of the agonist (Hallett et al., 1975). In patients with cerebellar disease, this triphasic EMG pattern is disturbed. The first activation of the agonist is prolonged, and the onset of the antagonist activity is delayed (Hallett et al., 1991; Hore et aI., 1991). Similar findings have been obtained in a monkey model of intention tremor in which cerebellar dysfunction is reversibly induced by cooling the dentate nucleus (Flament and Hore, 1986). The mistiming in the sequential activation of agonists and antagonists during target-directed movements is also found in cerebellar patients with limb ataxia but without intention tremor (Hallett et al., 1991). Thus, intention tremor cannot be solely explained on the basis of the disturbed triphasic EMG activation described above. Additionally, an oscillatory process has to be induced somehow. In this context, transcortical long loop reflexes which are enhanced and delayed in cerebellar disease may

404

C.H. LOCKING AND B. HELLWIG

Postural Tremor

Fig. 4. (a) Surface EMG recording under postural condition from the left wrist extensor of a patient with cerebellar dysfunction due to multiple sclerosis. (b, c) The spectra of the accelerometer and the EMG show peaks slightly below 3 Hz. (d, e, f) After deep brain stimulation in the right nucleus ventralis intermedius of the thalamus the 3 Hz tremor has disappeared. play an important role (Vilis and Hore, 1980; Mauritz et al., 1981; Diener and Dichgans, 1992). If one assumes that disturbed long latency reflexes contribute to the generation of cerebellar tremor, it should be possible to modulate cerebellar tremor by peripheral manipulation. This is indeed the case. For instance, altering the mechanical loads applied to a limb leads to a change in frequency and amplitude of cerebellar tremor (Hewer et al., 1972; Vilis and Hore, 1977). The influence of visual information on cerebellar tremor is less clear. While the removal of visual feedback did not change tremor characteristics in the above-mentioned monkey model of cerebellar tremor (Flament et al., 1984), intention tremor was enhanced in cerebellar patients who used visual guidance in target-directed movements (Sanes et al., 1988). The modulation of cerebellar tremor by

peripheral manipulation should, however, not be taken as evidence for an exclusively peripheral tremor generation. Central mechanisms seem also important, since tremor evoked by cerebellar ablation in the monkey is preserved despite deafferentation by dorsal root sectioning (Gilman et al., 1976). The treatment of cerebellar intention tremor is difficult, pharmacotherapy is often ineffective. However, there are some reports in the literature about beneficial effects of drugs such as serotonergic agents or clonazepam (Wasielewski et al., 1998). It has also been suggested that attaching weights to the wrists of patients with cerebellar tremor reduces tremor amplitudes and increases functional capacities (Hewer et al., 1972). Cerebellar tremor can be significantly reduced by stereotactic lesioning or

UNCOMMON TREMORS

deep brain stimulation in the contralateral nucleus ventralis intermedius of the thalamus (Fig. 4d-t) (Goldman et aI., 1992; Geny et aI., 1996). The cerebellar ataxia, however, remains, limiting the utility of this operation.

25.5. Dystonic tremor Dystonic tremor is defined as tremor in a body part that is affected by dystonia. Dystonic muscle activity is frequently associated with tremulous movements. In a series of 272 patients with cervical dystonia, about 60% were affected by head tremor (Jankovic et aI., 1991). Another survey of 266 patients with cervical dystonia revealed head tremor in 28% of the patients (Chan et aI., 1991). Since dystonia is often focal in nature, dystonic tremor is also typically localized. Clinical observation reveals that dystonic tremor is a postural or action tremor whose amplitude and frequency can be irregular. This is corroborated by electromyographic recordings showing that the amplitude and duration of each burst of muscle contraction may be variable and may differ from preceding or succeeding bursts (Jednyak et aI., 1991). Tremor bursts last between 50 and 300 ms (Deuschl et aI., 1992). Tremor frequencies also vary widely, frequency ranges of 3-7 Hz (Deus chi et aI., 1992) and 4-12 Hz (Jednyak et aI., 1991) have been reported. Bursts of muscle activity in agonist and antagonist mucles do not alternate, but seem to be imperfectly synchronized (Jednyak et aI., 1991). Dystonic tremor is typically absent during complete rest, while it is exacerbated by muscle contraction. Dystonic tremor should be distinguished from tremor that is associated with dystonia, i.e. that occurs in a body part not affected by dystonia. For instance, a postural tremor of the hands may accompany craniocervical dystonia (Koller et aI., 1994). In the 272 patients reviewed by Jankovic et al. (1991), 27% were affected by hand tremor. A similar percentage was found in the series of 266 patients with torticollis reviewed by Chan et al. (1991). The hand tremor in craniocervical dystonia is usually considered as nonspecific. Using clinical and electromyographic criteria, it cannot be distinguished from enhanced physiological tremor or essential tremor (Jankovic et aI., 1991; Deuschl et aI., 1997). However, Miinchau et al. (2001) suggested on the basis of more specialized neuro-

405

physiological investigations that arm tremor in craniocervical dystonia may have a mechanism different from that in essential tremor. In ballistic wrist flexion movements, for instance, the onset of the second agonist EMG burst was delayed in patients with essential tremor, but not in dystonic patients. Miinchau et al. (2001) distinguished also two subgroups of patients with craniocervical dystonia and arm tremor, one with a late and simultaneous onset of arm tremor and torticollis, and another with early onset of arm tremor and later development of torticollis. Tremor may be an isolated finding in patients with dystonic relatives. For instance, head tremor may occur in a patient with relatives affected by spasmodic torticollis (Yanagisawa et aI., 1972). This type of tremor can be referred to as dystonia geneassociated tremor. A specific example of dystonic tremor is tremulous spasmodic torticollis. The reduction of tremor amplitudes by a geste antagoniste is an important diagnostic feature (Fig. 5) which can be used to distinguish dystonic head tremor from essential head tremor. Dystonic tremor of the extremities responds less favorably to gestes anatagonistes. Another form of dystonic tremor is writing tremor associated with focal dystonia of the upper limb (Sheehy and Marsden, 1982; Elble et aI., 1990). This tremor must be distinguished from primary writing tremor which, according to the consensus statement of the Movement Disorder Society on tremor (Deuschl et aI., 1998a), is not associated with dystonia. Tremor may also occur in the focal dystonia of golfers, the yips (McDaniel et aI., 1989). The pathophysiology of dystonic tremor is unknown. The basal ganglia are likely to be involved in tremor generation, since the underlying dystonia seems to be a basal ganglia disorder. Due to the focal nature of dystonic tremor, injections of botulinum toxin are a possible treatment (Jankovic et aI., 1991; Wissel et aI., 1997).

25.6. Tremor in peripheral neuropathy According to the consensus statement of the Movement Disorder Society on tremor (Deuschl et aI., 1998a), a tremor syndrome is thought to be caused by a peripheral neuropathy if the tremor develops in association with the peripheral neuropathy, and if other neurological diseases leading to

406

C.H. LOCKING AND B. HELLWIG

Fig.5. Surface EMGrecordings in a patientwithtremulous torticollis spasmodicus. Tremoris almostcompletely abolished by a geste antagoniste. SCM r, SCM I: right and left sternocleidomastoid muscle. SPL r, SPL I: right and left splenius muscle. Trapr, Trap I: right and left trapezius muscle.

tremor are absent. Neuropathic tremor is usually postural or kinetic. There is no obvious correlation between tremor severity and the degree of proprioception loss or muscle weakness (Dalakas et aI., 1984; Smith, 1994). Neuropathies leading to tremor may be caused by a broad spectrum of underlying diseases such as diabetes, alcoholism, uremia, vasculitis or amyloidosis (Said et aI., 1982). Tremor is a frequent feature in hereditary motor-sensory neuropathies (HMSN) where about 40% of the patients may be affected by tremor (Cardoso and Jankovic, 1993). Tremor is also present in up to 90% of the demyelinating neuropathies associated with paraproteinemia (Dalakas et aI., 1984; Yeung et aI., 1991; Smith, 1994; Bain et aI., 1996; Nobile-Orazio et aI., 2000). There is one case report in which a tremor of the 4th and 5th fingers presumably caused by distal ulnar neuropathy was relieved by surgical decompression of Guyon's canal (Streib, 1990). Tremor frequencies in neuropathic tremor appear to cover a wide range between 4 and 11 Hz (Deuschl et aI., 1998a). In contrast to essential and Parkinsonian tremor, the tremor frequency in IgM paraprotein-

emic neuropathy tends to be lower in distal than in proximal muscle groups. In distal muscles, for instance in the m. abductor poIlicis brevis, frequencies as low as 2.8 Hz may be found. The different frequencies of rhythmic activity in proximal and distal muscles of the same ann are not simple harmonics of each other (Bain et aI., 1996). EMG activation patterns of agonists and antagonists in neuropathic tremor may be both cocontracting and alternating (Bain et aI., 1996; Pedersen et aI., 1997). Tremor associated with benign IgM paraproteinemic neuropathy can be modulated by peripheral inputs, e.g. by mechanical perturbations of the wrist or by electric median nerve shocks (Bain et aI., 1996). The mechanism of tremor generation in peripheral neuropathy is not precisely known. Central mechanisms seem to be involved, since mass loading to the extended arms does not decrease the tremor frequency (Fig. 6) (Pedersen et aI., 1997). It has been hypothesized that distortion and mistiming of peripheral inputs caused by demyelination of peripheral nerves disturb a central processor and lead thus to a central tremor generation. The cerebellum is likely to be involved, as suggested by the kinematic

407

UNCOMMON TREMORS

Postural tremor: unloaded

loading:1000g

Fig. 6. (a) Surface EMG recording under postural condition from the right wrist extensor of a patient with neuropathic tremor. (b, c) The spectra of the accelerometer and the EMG show peaks slightly below 5 Hz. (d, e, f) The tremor frequency remains unchanged after loading of lOOO g to the right hand.

and electromyographic analysis of ballistic wrist movements which show abnormalities similar to those found in cerebellar disease (Bain et al., 1996). Treatment of neuropathic tremor consists mainly in treating the underlying neuropathy. Moreover, improvement of neuropathic tremor by administration of propanolol, gabapentin, clonazepam and primidone has been reported (Dalakas et al., 1984; Bain et al., 1996; Saverino et al., 2001). 25.7. Drug-induced and toxic tremor syndromes

Drug-induced or toxic tremors are the consequence of drug intake or intoxication. These tremor syndromes are considerably heterogeneous, their clinical presentation covers a large spectrum. The most frequent drug-induced tremor is enhanced

physiological tremor which may, for instance, be caused by sympathomimetics or antidepressants. Neuroleptic drugs may lead to a rest tremor similar to Parkinsonian tremor. Substances such as lithium or alcohol can induce cerebellar tremor. Druginduced and toxic tremors are often accompanied by other symptoms and signs of eNS disturbance such as dizziness, unsteady gait or diplopia. The number of drugs and toxins leading to tremor is large. Among the more important ones are alcohol, neuroleptics, antidepressants (especially tricyclics), lithium, sympathomimetics, theophylline, caffeine, valproate, antiarrhythmics (amiodarone), thyroid hormones, cytostatics or immunosuppressants (cyclosporin A). A more complete list of tremorinducing agents can be found in the consensus statement of the Movement Disorder Society on tremor (Deuschl et al., 1998a).

408

C.H. LOCKING AND B. HELLWIG

The same drug may lead to different types of tremor. Neuroleptics may not only induce rest tremor as a sign of drug-induced parkinsonism. They can also cause tardive tremor syndromes, e.g. the Rabbit syndrome which consists in tremulous movements of the mouth and chin (Villeneuve, 1972). Chronic treatment with neuroleptics may also lead to tardive limb tremor which is of low frequency (3-5 Hz) and occurs mainly in the postural condition, but to a lesser extent also at rest and during goal-directed movements (Stacy and Jankovic, 1992). Tardive tremor continues after cessation of neuroleptic therapy. Another drug leading to various kinds of tremor is alcohol. Almost 50% of the patients with chronic alcoholism complain of a postural hand tremor (Koller et al., 1985b). This tremor as well as the tremor during alcohol withdrawal has been described as enhanced physiological tremor (Aisen et al., 1992; Milanov et al., 1996). In addition, alcohol can induce cerebellar and neuropathic tremor. The neurophysiological mechanisms leading to drug-induced and toxic tremors are largely unknown. Some evidence exists for the most frequent drug-induced tremor, enhanced physiological tremor. Marsden et al. (1967) showed that syrnpathomimetics such as adrenaline and isoprenaline increase physiological tremor by acting on peripheral beta-adrenergic receptors. In some drug-induced tremors central mechanisms are likely to be involved. Parkinsonian-like rest tremor caused by neuroleptics, for instance, is probably due to a blockade of striatal dopamine receptors. Alcohol, on the other hand, induces cerebellar tremor by damaging neuronal structures in the cerebellum.

et al., 1996). The etiology of task- and positionspecific tremors is unclear. A relation to both essential tremor and focal dystonia has been suggested (Rosenbaum and Jankovic, 1988). Another example of a task-specific tremor is isolated voice tremor. According to the consensus statement of the Movement Disorder Society on tremor, isolated voice tremor can be diagnosed if vocalization is tremulous but no other parts of the body show tremor (Deuschl et al., 1998a). Voice tremor is most prominent during prolonged sustained phonation of a vowel. It leads to rhythmic alterations in loudness and pitch, the frequency ranging between 4 and 8 Hz (Brown and Simonson, 1963). Tomoda et al. (1987) provided evidence that voice tremor is an action tremor of voluntary expiratory muscles. A relation to both essential tremor (Koller et al., 1985a; Massey and Paulson, 1985) and focal dystonia, i.e. spasmodoc dysphonia, (Aminoff et al., 1978; Blitzer et al., 1985; Pool et al., 1991) has been discussed. Treatment of voice tremor is difficult. There is evidence for positive effects of clonazepam or diazepam, while the efficacy of propanolol is controversial (Koller et al., 1985a; Tomoda et al., 1987). In severe cases, injections of botulinum toxin may be a therapeutic option. According to the consensus statement of the Movement Disorder Society on tremor, hereditary chin trembling or geniospasm should not be considered as a form of tremor (Deuschl et al., 1998a). Geniospasm is a rare, autosomal-dominant disorder which appears shortly after birth. It is characterized by an intermittent, often stress-induced involuntary trembling of the chin lasting for several minutes (Danek, 1993).

25.8. Task- and position-specific tremors

25.9. Psychogenic tremor

Task- and position-specific tremors are activated only during certain tasks or in certain positions. Primary writing tremor, first described by Rothwell et al. (1979), is the most frequent task-specific tremor and has been most thoroughly studied. A detailed account of primary writing tremor can be found in this book in the chapter by Bain. Task- and position-specific tremors may also occur in musicians and athletes as well as during weight-holding and smiling (Kachi et al., 1985; Jacome and Yanez, 1987; Cleeves et al., 1994; Lang et al., 1995; Soland

Psychogenic tremor (see also the chapter on psychogenic movement disorders) was observed as a common condition in the first half of the last century, especially in Europe during and after World War I. It was labeled in the German literature as "Kriegszittern" (war trembling) and in the Anglo-American literature as the "shell shock syndrome". Since then psychogenic tremor has become a relatively rare condition which is not always easily distinguishable from organic tremor. The prevalence of psychogenic tremor among tremor patients is estimated as nearly

409

UNCOMMON TREMORS

2% (Lempert et al., 1990) and around 50% among patients with psychogenic movement disorders (Factor et al., 1995). The age distribution shows peaks between 10 and 30 and more pronounced between 40 and 60 years. Females are slightly more represented (Deuschi et al., 1998b). Diagnosis of psychogenic tremor is usually based on negative criteria (Koller et al., 1989), i.e. the absence of any other symptom indicating an underlying organic disturbance. In general, the diagnosis "psychogenic" requires a related acute or chronic neurotic conflict as a positive diagnostic criterion. But psychogenic tremor itself presents with typical signs and symptoms which allow a positive diagnosis (Koller et al., 1989; Deuschl et aI., 1998b). One of the main diagnostic criteria is the sudden onset of unilateral tremor in the dominant hand or of bilateral tremor during rest as well as during movements. The clinical course shows large fluctuations with periods of progression and improvement as well as complete remission. The most valid clinical sign is the co-contraction phenomenon (Deuschl et al., 1998b): simultaneous activation of agonists and antagonists is the precondition of voluntary trembling movements; in fully relaxed muscles trembling cannot be activated (Fig. 7). Relaxation and co-contraction can be clinically tested by the examiner. Psychogenic tremor is found mainly in posture and action and less in resting conditions. A resting tremor sensu strictu with increase of amplitude by mental stress and suppression of tremor due to postural innervation does not fit with and would exclude a psychogenic tremor.

Most cases of psychogenic tremor exhibit a clear distractibility with decrease of tremor amplitude or change of frequency and regularity if the patient performs rapid alternating movements or difficult sequences of movements at the contralateral limb or during complex mental arithmetic tasks. A history of conversion disorders and multiple somatizations as well as the presence of secondary gain may support the assumption of a psychogenic origin. Improvement or resistance to all kinds of medication and psychotherapy are less indicative. Underlying psychodynamic aspects are more related to difficulties in coping with every day life than to defined intrapersonal conflicts. A peculiar case has been reported as "noise-induced psychogenic tremor associated with post-traumatic stress disorder" (Walters and Hening, 1992). Electromyographic recording and quantitative accelerometry reveal greater fluctuations of amplitudes and of frequencies (between 4 and 10Hz) than in Parkinsonian and essential tremor (Koller et al., 1989; Deuschl et al., 1998b; O'Suilleabhain and Matsumoto, 1998). The most striking feature consists in constant or increased tremor amplitude during loading with 500 or 1000 g which is unusual in other tremor types (Fig. 8). In psychogenic tremor as opposed to organic tremors, repetitive voluntary movements, e.g. tapping a beat, with the limb contralateral to the tremulous limb will lead to the dissipation of tremor or to a shift of the tremor frequency to the frequency of tapping (O'Suilleabhain and Matsumoto, 1998). Thus, frequency analysis and coherence measurement of tremor and

accelerometer

flexor muscle

extensor muscle

initial cocontraction 1000ms

Fig. 7. Accelerometer recording from the hand and surface EMG recordings from wrist flexors and extensors in a patient with psychogenic tremor. Tremor starts after initial cocontraction of wrist flexors and extensors (modified from Deuschl et al.,1998b).

410

C.H. LOCKING AND B. HELLWIG

Fig. 8. (a) Spectra from accelerometer recordings in a patient with parkinsonian tremor and in a patient with psychogenic tremor. While tremor amplitudes decrease with mass loading to the hand in parkinsonian tremor, they increase in psychogenic tremor. (b) Frequency of increased and reduced tremor amplitudes under mass loading in patients with parkinsonian, essential and psychogenic tremor (modified from Deuschl et al., 1998b).

voluntary tapping on both sides might be helpful in separating organic from psychogenic tremor (McAuley et aI., 1998). The simultaneous occurrence in separate muscle groups of tremors with different frequencies is suggestive of organic tremor. In

psychogenic tremor, such a frequency dissociation between different muscle groups is not present (O'SuilIeabhain and Matsumoto, 1998). Therapeutic approaches should be directed to psychological treatment, suggestion, physical treat-

UNCOMMON TREMORS

ment and placebo as well as to social arrangements like settlement of personal injury claim. References Aisen, ML, Adelstein, BD, Romero, J, Morris, A and Rosen, M (1992) Peripheral mechanical loading and the mechanism of the tremor of chronic alcoholism. Arch. Neurol., 49: 740-742. Aminoff, MJ, Dedo, HH and Izdebski, K (1978) Clinical aspects of spasmodic dysphonia. J. Neural. Neurosurg. Psychiatry, 41: 361-365. Andrew, J, Fowler, CJ and Harrison, MJG (1982) Tremor after head injury and its treatment by stereotaxic surgery. J. Neural. Neurosurg. Psychiatry, 45: 815819. Bain, PG, Findley, U, Thompson, PD, Gresty, MA, Rothwell, JC, Harding, AE and Marsden, CD (1994) A study of hereditary essential tremor. Brain, 117: 805-824. Bain, PG, Britton, TC, Jenkins, IH, Thompson, PD, Rothwell, JC, Thomas, PK, Brooks, DJ and Marsden, CD (1996) Tremor associated with benign IgM paraproteinaemic neuropathy. Brain, 119: 789-799. Benedikt, M (1889) Tremblement avec paralysie croisee du moteur oculaire commun. Bull. Med. (Paris), 3: 547-548. Berkovic, SF and Bladin, PF (1984) Rubral tremor: Clinical features and treatment of three cases. CUn. Exp. Neurol., 20: 119-128. Blitzer, A, Lovelace, RE, Brin, MF, Fahn, S and Fink, ME (1985) Electromyographic findings in focal laryngeal dystonia (spastic dysphonia). Ann. Otol. Rhinol. Laryngol., 94: 591-594. Boroojerdi, B, Ferbert, A, Foltys, H, Kosinski, CM, Noth, J and Schwarz, M (1999) Evidence for a nonorthostatic origin of orthostatic tremor. J. Neurol. Neurosurg. Psychiatry, 66: 284-288. Britton, TC, Thompson, PD, Van der Kamp, W, Rothwell, JC, Day, BL, Findley, U and Marsden, CD (1992) Primary orthostatic tremor: further observations in six cases. J. Neurol., 239: 209-217. Brown, JR and Simonson, J (1963) Organic voice tremor: a tremor of phonation. Neurology, 13: 520-525. Brown, P, Rothwell, JC, Stevens, JM, Lees, AJ and Marsden, CD (1997) Cerebellar axial postural tremor. Mov. Disord., 12: 977-984. Cardoso, FE and Jankovic, J (1993) Hereditary motorsensory neuropathy and movement disorders. Muscle Nerve, 16: 904-910. Chan, J, Brin, MF and Fahn, S (1991) Idiopathic cervical dystonia: clinical characteristics. Mov. Disord., 6: 119126.

411 Chen, JT, Yu, HY, Wu, ZA, Kao, KP, Hallett, M and Liao, KK (2000) Modulation of symptomatic palatal tremor by magnetic stimulation of the motor cortex. Clin. Neurophysiol., Ill: 1191-1197. Cleeves, L, Findley, U and Marsden, CD (1994) Odd tremors. In: CD Marsden and S Fahn (Eds.), Movement Disorders 3. Butterworth-Heinemann, Oxford, pp.434-458. Dalakas, MC, Teravainen, H and Engel, WK (1984) Tremor as a feature of chronic relapsing and dysgammaglobulinemic polyneuropathies. Incidence and management. Arch. Neurol., 41: 711-714. Danek, A (1993) Geniospasm: hereditary chin trembling. Mov. Disord., 8: 335-338. Deuschl, G, Lucking, CH and Quintern, J (1987) Orthostatischer tremor: Klinik, pathophysiologie und therapie. EEG-EMG, 18: 13-19. Deuschl, G, Mischke, G, Schenck, E, Schulte-Monting, J and Lucking, CH (1990) Symptomatic and essential rhythmic palatal myoclonus. Brain, 113: 1645-1672. Deuschl, G, Lohle, E, Heinen, F and Lucking, CH (1991) Ear click in palatal tremor: its origin and treatment with botulinum toxin. Neurol., 41: 1677-1679. Deuschl, G, Heinen, F, Kleedorfer, B, Wagner, M, Lucking, CH and Poewe, W (1992) Clinical and polymyographic investigation of spasmodic torticollis. J. Neurol., 239: 9-15. Deuschl, G, Toro, C and Hallett, M (1994a) Symptomatic and essential palatal tremor. 2. Differences of palatal movements. Mov. Disord., 9: 676-678. Deuschl, G, Toro, C, Valls-Sole, J, Zeffiro, T, Zee, DS and Hallet, M (1994b) Symptomatic and essential palatal tremor: I. Clinical, physiolgical and MRI analysis. Brain, 117: 775-788. Deuschl, G, Toro, C, Valls-Sole, J and Hallett, M (1996) Symptomatic and essential palatal tremor. 3. Abnormal motor learning. J. Neurol. Neurosurg. Psychiatry, 60: 520-525. Deuschl, G, Heinen, F, Guschlbauer, B, Schneider, S, Glocker, FX and Lucking, CH (1997) Hand tremor in patients with spasmodic torticollis. Mov. Disord., 12: 547-552. Deuschl, G, Bain, P and Brin, M (1998a) Consensus statement of the Movement Disorder Society on Tremor. Ad Hoc Scientific Committee. Mov. Disord., 13 (Suppl. 3): 2-23. Deuschl, G, Koster, B, Lucking, CH and Scheidt, C (1998b) Diagnostic and pathophysiological aspects of psychogenic tremors. Mov. Disord., 13: 294-302. Deuschl, G, Wilms, H, Krack, P, Wurker, M and Heiss, WD (1999) Function of the cerebellum in parkinsonian rest tremor and Holmes' tremor. Ann. Neurol., 46: 126-128.

412 Deuschl, G, Wenze1burger, R, Loftier, K, Raethjen, J and Stolze, H (2000) Essential tremor and cerebellar dysfunction: clinical and kinematic analysis of intention tremor. Brain, 123: 1568-1580. Diener, HCand Dichgans, J (1992) Pathophysiology of cerebellar ataxia. Mov. Disord., 7: 95-109. Diener, HC, Dichgans, J, Bacher, M and Gompf, B (1984) Quantification of postural sway in normals and patients with cerebellar diseases. Electroencephalogr. Clin. Neurophysiol.,57: 134-142. Dubinsky, RM, Hallet, M, Di Chiro, G, Fulham, M, Schwankhaus, J (1991) Increased glucose metabolism in the medulla of patients with palatal myoclonus. Neurol., 41: 557-562. Elble, RJ, Moody, C and Higgins, C (1990) Primary writing tremor. A form offocal dystonia? Mov. Disord., 5: 118-126. Evidente, VG, Adler, CH, Caviness, IN and Gwinn, KA (1998) Effective treatment of orthostatic tremor with gabapentin. Mov. Disord., 13: 829-831. Factor, SA, Podskalny, GD and Molho, ES (1995) Psychogenic movement disorders: frequency, clinical profile, and characteristics. J. Neurol. Neurosurg. Psychiatry, 59: 40~12. Fahn, S (1984) Cerebellar tremor: clinical aspects. In: U Findley and R Capildeo (Eds.), Movement Disorders: Tremor. Macmillan, London, pp. 355-363. Findley, U and Gresty, MA (1980) Suppression of "rubral" tremor with levodopa. Brit. Med. J., 281: 1043. Finkel, MF (2000) Pramipexole is a possible effective treatment for primary orthostatic tremor (shaky leg syndrome). Arch. Neurol., 57: 1519-1520. Finsterer, J, Muellbacher, Wand Mamoli, B (1996) Yes/ yes head tremor without appendicular tremor after bilateral cerebellar infarction. J. Neurol. Sci., 139: 242-245. FitzGerald, PM and Jankovic, J (1991) Orthostatic tremor: an association with essential tremor. Mov. Disord., 6: 60-64. Flament, D and Hore, J (1986) Movement and electromyographic disorders associated with cerebellar dysmetria. J. Neurophysiol., 55: 1221-1233. Flament, D, Vilis, T and Hore, J (1984) Dependence of cerebellar tremor on proprioceptive but not visual feedback. Exp. Neurol., 84: 314-325. Fung, VSC, Sauner, D and Day, BL (2001) A dissociation between subjective and objective unsteadiness in primary orthostatic tremor. Brain, 124: 332-330. Geny, C, Nguyen, JP, Pollin, B, Feve, A, Ricolfi, F, Cesaro, P and Degos, JD (1996) Improvement of severe postural cerebellar tremor in multiple sclerosis by chronic thalamic stimulation. Mov. Disord., II: 489494.

c.a LUCKING AND B. HELLWIG Ghika, J, Bogousslavsky, J, Henderson, J, Maeder, P and Regli, F (1994) The "jerky dystonic unsteady hand": a delayed motor syndrome in posterior thalamic infarctions. J. Neurol., 241: 537-542. Gilman, S, Carr, D and Hollenberg, J (1976) Kinematic effects of deafferentation and cerebellar ablation. Brain, 99: 311-330. Goldman, MS and Kelly, PJ (1992) Symptomatic and functional outcome of stereotactic ventralis lateralis thalamotomy for intention tremor. 1. Neurosurg., 77: 223-229. Guillain, G (1938) The syndrome of synchronous and rhythmic palate pharyngo-Iaryngo-oculo-diaphragmatic myoclonus. Proc. R. Soc. Med., 31: 1030-1038. Hallett, M, Shahani, BT and Young, RR (1975) EMG analysis of stereotyped voluntary movements in man. J. Neurol. Neurosurg. Psychiatry, 38: 1154-1162. Hallett, M, Berardelli, A, Matheson, J, Rothwell, J and Marsden, CD (1991) Physiological analysis of simple rapid movements in patients with cerebellar deficits. J. Neurol. Neurosurg. Psychiatry, 54: 124-133. Heilman, KM (1984) Orthostatic tremor. Arch. Neurol., 41: 880-881. Hewer, RL, Cooper, R and Morgan, MH (1972) An investigation into the value of treating intention tremor by weighting the affected limb. Brain, 95: 579-590. Holmes, G (1904) On certain tremors in organic cerebral lesions. Brain, 27: 325-375. Hore, J, Wild, B and Diener, HC (1991) Cerebellar dysmetria at the elbow, wrist, and fingers. J. Neurophysiol., 65: 563-571. Jacob, PC and Pratap Chand, R (1998) Posttraumatic rubral tremor responsive to Clonazepam, Mov. Disord., 13: 977-978. Jacome, DE and Yanez, GF (1987) Tremors of the smile. J. Neurol. Neurosurg. Psychiatry, 50: 489-490. Jankovic, J, Leder, S, Warner, D and Schwartz, K (1991) Cervical dystonia: clinical findings and associated movement disorders. Neurol., 41: 1088-1091. Jedynak, CP, Bonnet, AM and Agid, Y (1991) Tremor and idiopathic dystonia. Mov. Disord., 6: 230-236. Kachi, T, Rothwell, JC, Cowan, JMA and Marsden, CD (1985) Writing tremor: its relationship to benign essential tremor. J. Neurol. Neurosurg. Psychiatry, 48: 545-550. Koller, W, Graner, D and Mlcoch, A (1985a) Essential voice tremor: treatment with propranolol. Neurol., 35: 106-108. Koller, W, O'Hara, R, Dorus, W and Bauer, J (1985b) Tremor in chronic alcoholism. Neurol., 35: 16601662. Koller, WC, Glatt, S, Biary, N and Rubino, FA (1987) Essential tremor variants: effect of treatment. Clin. Neuropharmaco I., 10: 342-350.

UNCOMMON TREMORS

Koller, W, Lang, A, Vetere-Overfield, B, Findley, L, Cleeves, L, Factor, S, Singer, C and Weiner, W (1989) Psychogenic tremors. Neurol., 39: 1094-1099. Koller, WC, Busenbark, K and Miner, K (1994) The relationship of essential tremor to other movement disorders: report on 678 patients. Essential Tremor Study Group. Ann. Neurol., 35: 717-723. Koster, B, Lauk, M, Timmer, J, Poersch, M, Guschlbauer, B, Deuschl, G and LUcking, CH (1999) Involvement of cranial muscles and high intermuscular coherence in orthostatic tremor. Ann. Neurol., 45: 384-388. Koster, B, Deuschl, G, Lauk, M, Timmer, J, Guschlbauer, B and LUcking, CH (2002) Essential tremor and cerebellar dysfunction: abnormal ballistic movements. J. Neurol. Neurosurg. Psychiatry, 73: 1-6. Krack, P, Deuschl, G, Kaps, M, Warnke, P, Schneider, S and Traupe, H (1994) Delayed onset of "rubral tremor" 23 years after brainstem trauma. Mov. Disord., 9: 240-242. Lang, AE, Jog, M and Ashby, P (1995) "Weight-holding tremor": an unusual task-specific form of essential tremor? Mov. Disord., 10: 228-229. Lapresle, J and Ben Hamida, M (1965) Correspondance somatotopique, secteur par secteur, des degenerescences de l' olive bulbaire consecutives a des lesions lirnitees du noyau dentele centro-lateral. Etude de 4 observations anatomiques. Rev. Neurol., 113: 439448. Lapresle, J and Ben Hamida, M (1968) Contributions a la connaissance de la voie dento-olivaire: etude anatomique de deux cas de degenerescence hypertrophique de l' olive bulbaire secondaire a un ramollissement limite de la calotte mesencephalique, Presse. MM., 76: 1226-1230. Lapresle, J and Ben Hamida, M (1970) The dentatoolivary pathway. Somatotopic relationship between the dentate nucleus and the contralateral inferior olive. Arch. Neurol., 22: 135-143. Larochelle, L, Bedard, P, Boucher, R and Poirier, LJ (1970) The rubro-cerebello-rubral loop and postural tremor in the monkey. J. Neurol. Sci., 11: 53-64. Lempert, T, Dieterich, M, Huppert, D and Brandt, T (1990) Psychogenic disorders in neurology: frequency and clinical spectrum. Act. Neurol. Scand., 82: 335340. Llinas, RR (1984) Rebound excitation as the physiological basis for tremor: a biophysical study of the oscillatory properties of mammalian central neurons in vitro. In: LJ Findley and R Capildeo (Eds.), Movement Disorders: Tremor. Macmillan, London, pp. 165-182. Manto, MD, Setta, F, Legros, B, Jacquy, J and Godaux, E (1999) Resetting of orthostatic tremor associated with cerebellar cortical atrophy by transcranial magnetic stimulation. Arch. Neurol., 56: 1497-1500.

413 Marsden, CD, Foley, TH, Owen, DAL and McAllister, RG (1967) Peripheral beta-adrenergic receptors concerned with tremor. Clin. Sci., 33: 53-65. Massey, EW and Paulson, GW (1985) Essential vocal tremor: clinical characteristics and response to therapy. South Med. J., 78: 316-317. Masucci, EF, Kurtzke, IF and Saini, N (1984) Myorhythmia: A widespread movement disorder. Brain, 107: 53-79. Mauritz, KH, Schmitt, C and Dichgans, J (1981) Delayed and enhanced long latency reflexes as the possible cause of postural tremor in late cerebellar atrophy. Brain, 104: 97-116. McAuley, JH, Rothwell, JC, Marsden, CD and Findley, LJ (1998) Electrophysiological aids in distinguishing organic from psychogenic tremor. Neurol., 50: 18821884. McAuley, JH, Britton, TC, Rothwell, JC, Findley, LJ and Marsden, CD (2000) The timing of primary orthostatic tremor bursts has a task-specific plasticity. Brain, 123: 254-266. McDaniel, KD, Cummings, JL and Shain, S (1989) The "yips": a focal dystonia of golfers. Neurology, 39: 192-195. McManis, PG and Sharbrough, FW (1993) Orthostatic tremor: clinical and electrophysiological characteristics. Muscle Nerve, 16: 1253-1260. Milanov, I (2002) Clinical and electromyographic examinations of patients with midbrain and cerebellar tremor. Electromyogr. Clin. Neurophysiol., 42: 105112. Milanov, I, Toteva, Sand Georgiev, D (1996) Alcohol withdrawal tremor. Electromyogr. Clin. Neurophysiol., 36: 15-20. Mills, KR and Nithi, KA (1997) Motor cortex stimulation does not reset primary orthostatic tremor. J. Neurol. Neurosurg. Psychiatry, 63: 553. Miwa, H, Hatori, K, Kondo, T, Imai, H and Mizuno, Y (1996) Thalamic tremor: Case reports and implications of the tremor-generating mechanism. Neurol., 46: 75-79 Mossuto-Agatiello, L, Puccetti, G and Castellano, AE (1993) Rubral tremor after thalamic haemorrhage. J. Neurol., 241: 27-30. Munchau, A, Schrag, A, Chuang, C, MacKinnon, CD, Bhatia, KP, Quinn, NP and Rothwell, JC (2001) Arm tremor in cervical dystonia differs from essential tremor and can be classified by onset age and spread of symptoms. Brain, 124: 1765-1776. Nagaoka, M and Narabayashi, H (1984) Palatal myoclonus - its remote influence. J. Neurol. Neurosurg. Psychiatry, 47: 921-926. Narabayashi, H (1986) Tremor, its generating mechanisms and treatment. In: PJ Vinken, GW Bruyn and HL

414 Klawans (Eds.), Handbook of Clinical Neurology (Vol. 49) Extrapyramidal Disorders. Elsevier, Amsterdam, pp.597-607. Nishie, M, Yoshida, Y, Hirata, Y and Matsunaga, M (2002) Generation of symptomatic palatal tremor is not correlated with inferior olivary hypertrophy. Brain, 126: 1348-1357. Nobile-Orazio, E, Meucci, N, Baldini, L, Di Troia, A and Scarlato, G (2000) Long-term prognosis of neuropathy associated with anti-MAG IgM M-proteins and its relationship to immune therapies. Brain, 123: 71D--717. Onofrj, M, Thomas, A, Paci, C and D' Andreamatteo, G (1998) Gabapentin in orthostatic tremor: results of a double-blind crossover with placebo in four patients. Neurology, 51: 880-882. O'Suilleabhain, PE and Matsumoto, JY (1998) Timefrequency analysis of tremors. Brain, 121: 2127-2134. Pahwa, R, Lyons, KE, Kempf, L, Wilkinson, SB and Koller, WC (2002) Thalamic stimulation for midbrain tremor after partial hemangioma resection. Mov. Disord., 17: 40~07. Papa, SM and Gershanik, OS (1988) Orthostatic tremor: an essential tremor variant? Mov. Disord., 3: 97-108. Pedersen, SF, Pullman, SL, Latov, N and Brannagan, TH, III (1997) Physiological tremor analysis of patients with anti-myelin-associated glycoprotein associated neuropathy and tremor. Muscle Nerve, 20: 38--44. Pfeiffer, G, Hinse, P, Humbert, T and Riemer, G (1999) Neurophysiology of orthostatic tremor. Influence of transcranial magnetic stimulation. Electromyogr. Clin. Neurophysiol., 39: 49-53. Pool, KD, Freeman, FJ, Finitzo, T, Hayashi, MM, Chapman, SB, Devous, MD, Sr, Close, LG, Kondraske, GV, Mendelsohn, D, Schaefer, SD and Watson, BC (1991) Heterogeneity in spasmodic dysphonia. Neurologic and voice findings. Arch. Neurol., 48: 305-309. Remy, P, De Recondo, A, Defer, G, Loc'h, C, Amarenco, P, Plante-Bordeneuve, V, Dao-Castellana, M-H, Bendriem, B, Crouzel, C, Clanet, M, Rondot, P and Samson, Y (1995) Penduncular rubral tremor and doparninergic denervation: a PET study. Neurology, 45: 472-477. Rondot, P and Bathien, N (1995) Cerebellar tremors: physiological basis and treatment. In: LJ Findley and WC Koller (Eds.), Handbook of Tremor Disorders. Marcel Dekker, New York, pp. 371-385. Rondot, P and Ben Hamida, M (1968) Myoclonies du voile et myoclonies squelettiques. Etude clinique et anatomique. Rev. Neurol., 119: 59-83. Rondot, P, Guiot, G, Arfel, G and Derome, P (1965) Syndrome myoclonique complexe. Etude clinique et derivation des activites thalamiques. La Semaine des Hopitaux, 41: 457-464.

C.H. LUCKING AND B. HELLWIG

Rosenbaum, F and Jankovic, J (1988) Focal task-specific tremor and dystonia: categorization of occupational movement disorders. Neurology, 38: 522-527. Rothwell, JC, Traub, MM and Marsden, CD (1979) Primary writing tremor. J. Neural. Neurosurg. Psychiatry, 42: 1106--1114. Said, G, Bathien, N and Cesaro, P (1982) Peripheral neuropathies and tremor. Neurology, 32: 480-485. Samie, MR, Selhorst, JB and Koller, WC (1990) Posttraumatic midbrain tremors. Neurology, 40: 62-66. Sander, HW, Masdeu, JC, Tavoulareas, G, Walters, A, Zimmermann, T and Chokroverty, S (1998) Orthostatic tremor: an electrophysiological analysis. Mov. Disord., 13: 735-738. Sanes, IN, Lewitt, PA and Mauritz, KH (1988) Visual and mechanical control of postural and kinetic tremor in cerebellar system disorders. J. Neurol. Neurosurg. Psychiatry, 51: 934-943. Saverino, A, Solaro, C, Capello, E, Trompetto, C, Abbruzzese, G and Schenone, A (200 1) Tremor associated with benign IgM paraproteinaemic neuropathy successfully treated with gabapentin. Mov. Disord., 16: 967-968. Sheehy, MP and Marsden, CD (1982) Writers' cramp - a focal dystonia. Brain, 105: 461-480. Shepherd, GM, Tauboll, E, Bakke, SJ and Nyberg-Hansen, R (1997) Midbrain tremor and hypertrophic olivary degeneration after pontine hemorrhage. Mov. Disord., 12: 432-437. Silfverskiold, BP (1977) A 3 Cis leg tremor in a "cerebellar" syndrome. Acta. Neurol. Scand., 55: 385393. Smith, IS (1994) The natural history of chronic demyelinating neuropathy associated with benign IgM paraproteinaemia. A clinical and neurophysiological study. Brain, 117: 949-957. Soland, VL, Bhatia, KP, Volonte, MA and Marsden, CD (1996) Focal task-specific tremors. Mov. Disord., 11: 665-670. Stacy, M and Jankovic, J (1992) Tardive tremor. Mov. Disord., 7: 53-57. Stolze, H, Petersen, G, Raethjen, J, Wenzelburger, R and Deuschl, G (200 1) The gait disorder of advanced essential tremor. Brain, 124: 2278-2286. Streib, EW (1990) Distal ulnar neuropathy as a cause of finger tremor: a case report. Neurology, 40: 153-154. Strutz, J, Hammerich, T and Amedee, R (1988) The motor innervation of the soft palate. An anatomical study in guinea pigs and monkeys. Arch. Otorhinolaryngol., 245: 18D---l 84. Thompson, PD, Rothwell, JC, Day, BL, Berardelli, A, Dick, JPR, Kachi, T and Marsden, CD (1986) The physiology of orthostatic tremor. Arch. Neurol., 43: 584-587.

UNCOMMON TREMORS

Tomoda, H, Shibasaki, H, Kuroda, Y and Shin, T (1987) Voice tremor: dysregulation of voluntary expiratory muscles. Neurology, 37: 117-122. Tsai, CH, Semmler, JG, Kimber, TE, Thickbroom, G, Stell, R, Mastaglia, FL and Thompson, PD (1998) Modulation of primary orthostatic tremor by magnetic stimulation over the motor cortex. J. Neurol. Neurosurg. Psychiatry, 64: 33-36. Uncini, A, Onofrj, M, Basciani, M, Cutarella, R and Gambi, D (1989) Orthostatic tremor: report of two cases and an electrophysiological study. Act. Neural. Scand.,79: 119-122. Vilis, T and Hore, J (1977) Effects of changes in mechanical state of limb on cerebellar intention tremor. J. Neurophysiol., 40: 1214-1224. Vilis, T and Hore, J (1980) Central neural mechanisms contributing to cerebellar tremor produced by limb perturbations. J. Neurophysiol., 43: 279-291. Villeneuve, A (1972) The rabbit syndrome. A peculiar extrapyramidal reaction. Can. Psychiatr: Assoc. J., 17 (Suppl. 2): SS69-SS72. Walker, FO, McCormick, GM and Hunt, VP (1990) Isometric features of orthostatic tremor: an electromyographic analysis. Muscle Nerve, 13: 918-922. Walters, AS and Hening, WA (1992) Noise-induced psychogenic tremor associated with post-traumatic stress disorder. Mov. Disord., 7: 333-338. Wasielewski, PG, Burns, JM and Koller, WC (1998) Pharmacologic treatment of tremor. Mov. Disord., 13 (Suppl. 3): 90-100.

415 Willeit, J, Deisenhammer, F, Ransmayr, G and Gerstenbrand, F (1991) Orthostatic tremor. Dtsch. Med. Wochenschr., 116: 1509-1512. Wills, AJ, Thompson, PD, Findley, LJ and Brooks, DJ (1996) A positron emission tomography study of primary orthostatic tremor. Neurology, 46: 747-752. Wills, AJ, Brusa, L, Wang, HC, Brown, P and Marsden, CD (1999) Levodopa may improve orthostatic tremor: case report and trial of treatment. J. Neurol. Neurosurg. Psychiatry, 66: 681-684. Wissel, J, Masuhr, F, Schelosky, L, Ebersbach, G and Poewe, W (1997) Quantitative assessment of botulinum toxin treatment in 43 patients with head tremor. Mov. Disord., 12: 722-726. Wu, YR, Ashby, P and Lang, AE (2001) Orthostatic tremor arises from an oscillator in the posterior fossa. Mov. Disord., 16: 272-279. Yanagisawa, N, Goto, A and Narabayashi, H (1972) Familial dystonia musculorum deformans and tremor. J. Neurol. Sci., 16: 125-136. Yeung, KB, Thomas, PK, King, RH, Waddy, H, Will, RG, Hughes, RA, Gregson, NA and Leibowitz, S (1991) The clinical spectrum of peripheral neuropathies associated with benign monoclonal IgM, IgG and IgA paraproteinaemia. Comparative clinical, immunological and nerve biopsy findings. J. Neurol., 238: 383391. Yokota, T, Hirashima, F, Furukawa, T, Tsukagoshi, H and Yoshikawa, H (1989) MRI findings of inferior olives in palatal myoclonus. J. Neurol., 236: 115-116.

Movement Disorders Handbook of Clinical Neurophysiology, Vol. 1 M. Hallett (Ed.) © 2003 Elsevier B.V. All rights reserved

417

CHAPTER 26

Diseases and treatments: Parkinson's disease John C. Rothwell* Sobell Department, Institute of Neurology, Queen Square, London WCIN 3BG, UK

In Parkinson's disease, as in other conditions, the techniques of clinical neurophysiology are used to address two types of question: (1) to quantify clinical deficits in order to provide objective information about the effectiveness of treatment; and (2) to explore the physiology of the basal ganglia by using Parkinson's disease as a relatively pure model of disordered motor circuits. A variety of techniques can be applied to the four main symptoms of tremor, rigidity, bradykinesia and postural instability. Tremor is covered in the chapter by Deuschl; this chapter concentrates on the neurophysiology of bradykinesia, rigidity and postural control.

cally this coincides with the idea that the basal ganglia and its projection targets in the secondary motor areas (e.g. supplementary motor area) are the main areas involved in planning movement commands, whereas the primary motor cortex, brainstem and spinal cord are more closely linked to execution of motor commands. A multitude of different techniques have been used to probe bradykinesia. In the text below, studies have been grouped according to the methods used, from behavioral (reaction time and EMG recording) tasks that measure the amount of slowness in moving, to EEG and MEP studies that try to tackle some of the underlying mechanisms.

26.1. Bradykinesia Bradykinesia refers to the slowness that is so characteristic of all movements in Parkinson's disease. It is often used to encompass the terms hypokinesia (smallness of movement, as in micrographia) and akinesia (lack of movement, as in freezing or lack of facial expression). Nevertheless, although the three symptoms are linked, they often occur to different extents in different patients, and therefore must have, at least to a limited extent, separate mechanisms (Evarts et al., 1981). Whatever the subvariety of bradykinesia, physiological techniques are usually designed either to measure the amount of slowness, or to give some insight into the underlying mechanisms that produce bradykinesia. Most workers separate the latter into those that result from a problem in producing the commands to move and those that result from a problem in executing those commands. Anatomi-

* Correspondence to: John C. Rothwell, Sobell Department, Institute of Neurology, Queen Square, London WC1N 3BG, UK. E-mail address:[email protected] Tel.: +44 (0)20 78298725; fax: +44 (0)20 7278 9836.

26.1.1. Behavioral tasks: descriptive measures of bradykinesia 26.1.1.1. Reaction time tasks

A reaction time task involves three main processes, identification of a sensory triggering input, association of that input with an appropriate response, and recruitment of the response. The purest measure of reaction time is EMG onset. This is because, if any movement is required, for example, to press a button or to move a stylus, then the measure of reaction time will also involve some measure of movement speed. In cases with severe slowing of movement, the latter can add a considerable extra delay. Simple reaction time tasks involve subjects making the same response to a given stimulus on every trial. In most studies, simple reaction times have been found to be slowed in Parkinson's disease (Evarts et al., 1981; Jahanshahi et al., 1993), however, the improvement after L-DOPA or neurosurgical treatments has often been found to be quite minor (Limousin et al., 1999; Muller et al., 2001), and not well linked to clinical estimates of bradykinesia.

418 As discussed in the chapter by Jahanshahi, simple reaction times are always shorter than choice reaction times. This is because in a choice task some element of movement planning must occur after the reaction signal, whereas in a simple reaction task, the movement is always known fully in advance. There has been an enormous amount of controversy over these two types of reaction time in Parkinson's disease. Initially it was claimed that simple reaction times were slower than normal whereas choice reaction times were the same as normal. Others said that both sets of reaction times were slow in Parkinson's disease, but that the usual difference between simple and choice times was absent (Bloxham et al., 1981; Evarts et al., 1981; Kutukcu et al., 1999). These data were interpreted as showing that patients failed fully to pre-program their movements in the simple reaction task and they were therefore slower than normal. However, in the choice reaction time, this was not such a disadvantage and so their reaction times were less affected. Unfortunately, not all data follow this pattern (Pullman et al., 1988; Brown et al., 1993). Some of the variety in results may be due to differences in patient groups; some due to elements in the design of the tasks, such as stimulus-response compatibility that may be handled differently in patients with Parkinson's disease (Brown et al., 1993). Whether or not patients fail to take advantage of advance information in the setting of a reaction time task, a less controversial finding is that patients are undoubtedly slow at utilising information that they are given. An example of this is the precued choice reaction task discussed by Jahanshahi. Here, information about the nature of a forthcoming choice reaction is given at different times prior to the imperative signal. The more advance information is given, the quicker the reaction time. In Parkinson's disease, patients take more time to incorporate this information into their movement plans (Jahanshahi et al., 1992). This measure is improved by treatment with L-DOPA as well as by pallidotomy (Limousin et al., 1999).

26.1.1.2. Movement execution A range of different factors can contribute to slowness of movement in patients with Parkinson's disease. Three of these, weakness, rigidity and movement inaccuracy, can be regarded as secondary

J.e. ROTHWELL or contributing factors since they do not occur in all patients or at all joints.

26.1.1.2.1. Muscle weakness. Several studies have compared muscle strength of patients and healthy age-matched subjects and found that patients are marginally weaker than expected (Stelmach et al., 1989; Jordan et al., 1992). However, it is difficult to match groups for amounts of daily exercise and physical build, and much larger differences can be observed in some muscle groups when the same patients are tested ON and OFF their medication. Corcos et al. (Corcos et al., 1996) showed that there was a 30% loss of maximum muscle strength in triceps after overnight withdrawal of treatment, and a smaller loss of 10% in biceps. One of the reasons for this is that in the OFF condition, patients have more action tremor than ON. This tends to synchronize motor unit firing at around 8 Hz, preventing the contraction from being fully fused (Brown et al., 1997, 1998). The loss of maximal strength is linked to a slowing in the rate at which patients can recruit muscle force. In some individuals it may take several seconds when OFF treatment to reach maximum contraction levels. 26.1.1.2.2. Rigidity. If stretch reflexes are enhanced in patients (see below) it could be that stretch reflexes in antagonist muscles interferes with attempted rapid movements. However, in most patients there is no evidence from EMG recordings that such activation occurs to any significant extent, so that rigidity is unlikely to be a major contributing factor to bradykinesia. 26.1.1.2.3. Movement variability. The movements of patients with Parkinson's disease are less accurate than normal particularly if they have to move as fast as possible (Sanes, 1985; Sheridan and Flowers, 1990; Phillips et al., 1994). In other words, the speed-accuracy trade-off is less efficient in patients than in healthy subjects. Sheridan and Flowers (1990) suggested that bradykinesia might be due to an active strategy of patients to move more slowly in order to improve their accuracy. Although a plausible strategy, other factors must also be involved since bradykinesia remains in tasks where spatial accuracy constraints have been removed (Teasdale et al., 1996).

DISEASES AND TREATMENTS: PARKINSON'S DISEASE

26.1.1.3. Ballistic movements at a single joint Simple, self-terminated elbow or wrist movements are slower than normal in Parkinson's disease (Berardelli et aI., 1996a). Although some of this may be due to weakness or slowness in recruiting muscle power, this cannot always be the case. In many patients, such movements are still accompanied by the usual triphasic pattern of activity in agonist, antagonist and agonist muscle (Hallett and Khoshbin, 1980). The bursts of activity have the same durations and time relations as in healthy subjects, suggesting that rapid recruitment of muscle power is not a primary problem. The major deficit is that the amplitude of the first agonist burst of EMG is smaller than normal (Berardelli et aI., 1986). As a result, patients often add further bursts of activity after the usual triphasic burst in order to take the joint to the target position (Hallett and Khoshbin, 1980)(Fig. 1). A clue to what is happening comes when patients are required to make movements over a variety of different distances. In healthy subjects, the amplitude and duration of the first agonist EMG burst increases as the amplitude of the movement increases, and the same happens in patients (Berardelli et aI., 1984, 1986). The consequence is that although patients may use an undersized AG1 burst when making a small amplitude movement, they can in fact produce a larger AG1 burst when attempting a larger movement. If the burst for the large movement had been used to make the small movement, the movement speed may have been normal. The conclusion from this is that an important factor in producing slowness of movement is an underscaling of dynamic muscle force to the parameters of movement (Fig. 2). Treatment with L-DOPA, pallidotomy and DBS of subthalamus all improve the speed of simple single joint movements, although in some cases, the effect has been relatively minor (e.g. Berardelli et aI., 1986), leading authors to measure performance in more complex or natural tasks. 26.1.2. Complex movements If any additional complexity is added to a simple movement, either by repeating the movement, or combining it with other tasks, then bradykinesia becomes more prominent. Clinical tests of bradykinesia often make use of this phenomenon. Repetitive

419

sequential movement involving isolated finger movements, hand opening/closing or wrist pronation/supination become smaller and slower with repetition of the movement ("fatigue"). Schwab et al. (1954) asked patients to squeeze a sphygmomanometer bulb in one hand and outline a drawing with the other. They had much more difficulty if they had to do both tasks together than if each one was performed alone. Indeed, in most cases, patients tended to alternate between the tasks rather than perform them at the same time. Experimental studies have analyzed these features of bradykinesia in some detail. In essence they show that bradykinesia is more than the slowness seen in simple single movements. There are additional problems in combining or sustaining complex movements. Benecke et al. (1986, 1987a) examined rapid elbow flexion movements combined with a simultaneous or sequential hand movement performed with either the same or the opposite arm. In contrast to normal subjects, in whom there was no decrement of performance when two tasks were combined, patients with Parkinson's disease showed: (i) a marked slowing of movement over and above that seen in each task alone when both had to be performed together; and (ii) a longer pause between each element of a sequential task. Indeed, these two extra deficits correlated better with clinical measures of bradykinesia, and improved more impressively after treatment with L-DOPA (Benecke et aI., 1987b) than the slowness in each simple movement. Similar problems in performing simultaneous movements have been described in bilateral reaching (Stelmach and Worringham, 1988; Castiello and Bennett, 1997) or cranking tasks (Johnson et aI., 1998). In sequential movements, prolonged pauses between each element have been observed in everyday movements such as rising from a chair to pick up an object or drinking from a cup (Bennett et aI., 1995). Elements of "fatigue" have also been reported in longer lasting sequences of movements (Agostino et aI., 1992; Dettmers et aI., 1995; Stewart et aI., 2001). Complex movements are often performed better when guided by external cues. Clinically it can readily be seen that stripes drawn on the floor or the sound of martial music can improve walking in patients with Parkinson's disease; micrographia can be ameliorated if patients are asked to write their letters between horizontal lines drawn on the paper.

r.c. ROTHWELL

420 Normal~t

15°

Velocity

[ 300"'. FF

FEJ~

~[O.2mY 15

Off

Parkinson's diHasr

Posi~ [ 75",.

FF~

[0." mY

FE-------- ~ 15 Fig. 1. Rapid movements performed at a single joint in a normal subject (15° and 60° of movement amplitude) and in a patient with Parkinson's disease (OFF and ON therapy, 15° of movement amplitude). The normal subject performed the movement with a single triphasic EMG pattern in the agonist and antagonist muscles, whereas the patient performed the movement with multiple EMG bursts. EMG activity was recorded from the forearm flexor (FF) and extensor (FE) muscles (From Berardelli et aI., 1999.)

Experimentally, several studies have shown that the performance of sequential button pressing (Georgiou et al., 1993, 1994; Sheppard et al., 1996) or drawing tasks is improved if visual cues are given during the

task, rather than being performed from memory. In both cases, the speed of the individual movements increases and the pauses between individual elements of the sequence become shorter. The

421

DISEASES AND TREATMENTS: PARKINSON'S DISEASE

II) <, CJ)

at

'a

>-

~

u 0

'ii

>

.lIl: ftI

NO LOAD 0 ." 0

WITH LOAD

0

II)

,..

Normal

0 ,.. 0 0

0 0

"

0

It) l")

I

+~---

at

a.

_-

- :.- --

0

0

30

-~

"

ON OFF

60

P.O.

01

:1

~---

0

_::/~ 30

60

Movement amplitude (deg) Fig. 2. Underscaling of movement velocity in Parkinson's disease. Subjects made rapid self-paced, self-terminated flexion movements of the wrist in their own time through 15° or 60° either without (left) or with an opposing force (right). Note how the peak velocity of movement increases with distance moved in both patients and healthy controls. However, the velocity at anyone movement amplitude is smaller in patients. Note also that treatment with L-DOPA only improves velocity mildly in the patient group for this simple task. (Data from Berardelli et al., 1986.)

conclusion is that parkinsonian patients have more difficulty in performing sequential movements that rely more on an internally determined than an externally triggered mode (Dettmers et al., 1996).

would consume more of the processing resource, and lead to difficulties in performing several tasks at once, or in switching between tasks. When required to perform more than one task at once, this would become a limiting factor.

26.1.2.1. What is the nature of the extra deficits in performance of complex movements? The problem of combining tasks or switching from one task to another is not confined to movement. It can be observed in cognitive tasks or combined cognitive and motor tasks (Brown and Marsden, 1991; Oliveira et al., 1998). Such observations are important since they indicate that the extra deficit seen in complex movements is not necessarily a purely motor problem. They raise the possibility that global processing mechanisms, perhaps involving attention, are also a factor. Brown and Marsden (1991) suggested that patients either have a limited processing "resource" that interferes with their ability to run more than one task at the same time, or that they have difficulty in switching this resource between tasks. An alternative is that the global resource is the same in patients, but that tasks are performed less automatically than normal. Effectively, patients may try to compensate for lack of basal ganglia input by devoting more resources to each single task they perform. In this case, each task

26.1.3. Sensorimotor processing The studies above have focused primarily on details of motor tasks, but recent work also suggests that there may be minor abnormalities of sensorimotor processing in patients with Parkinson's disease that could affect the preparation of instructions to move. Schneider et al. (1986) reported that two point discrimination and proprioceptive position sense was reduced, whilst Klockgether et al. (1995) in a study that tested integration of visual and kinesthetic information in an arm reaching movement found that peripheral feedback was compromized in patients. This point was taken further in a study by Demirci et al. (1997) who devized a task where patients had to match their visual and kinesthetic impressions of the position of their fingers. When kinesthesia was used to match to a visual reference, patients underestimated distance. If we assume that the patients' visual sense was normal, then this implies a "reduced" sense of kinesthetic input in Parkinson's disease. The authors

422

suggested that this may relate to the underscaling of motor output discussed above. The unexpected conclusion from the latter was that in a simple ballistic movement, patients' motor system might be able to produce a large initial burst of agonist EMG but did not do so. The result was that their movements were slower than they need have been. Demciri et al. (1997) suggest that a larger motor command was not selected because patients may not actually have perceived their movements to be slow due to the underscaling of their sensory input. 26.1.4. EEG studies: physiological mechanisms of bradykinesia 26.1.4.1. Bereitschaftspotential (BP) As described in the chapter by Shibasaki, the Bereitschaftspotential gives a measure of brain activation in the period immediately preceding a selfpaced voluntary movement. Since the latter are slow in Parkinson's disease, it is clearly of interest to investigate whether there is an accompanying abnormality in the pattern of brain activation prior to such

J.e. ROTHWELL movements. In early studies, there was some debate over whether the BP was abnormal in Parkinson's disease. Some of the controversy was resolved by Dick et al. (1987), who showed that L-DOPA could affect the amplitude of the BP both in normal subjects and in patients, and that a difference between patients with PD and normals was crucially dependent on the level of dopaminergic function. They went on to show that the BP in patients OFF therapy was reduced in the early part (BPI), whilst it was larger than normal in the later part (BP2). The net effect was that the peak BP was virtually the same in the patients as in normals (Fig. 3). They suggested that underactivity of a source in the SMA was responsible for the reduction of the early component, and that this was compensated by overactivity in lateral motor areas nearer the time of onset of movement. Recent studies have tended to confirm this idea. For example, Jabanshahi et al. (1995) noted that premovement EEG activity was normal in patients with Parkinson's disease when they performed an externally triggered task compared with the reduction that is seen in self-paced movements (Fig. 1). Cunnington et al. (1995, 1997)

Fig. 3. Grand average premovement EEG activity in eight healthy age matched subjects and eight patients with Parkinson's disease after overnight withdrawal of their normal medication. The movement was a self-paced extension of the index finger, with 50-100 trials averaged in each subject. Note that in the patients (PD), the amplitude of the early portion of the BP is smaller than in normals (N), particularly in the midline and ipsilateral leads. (Data from Dick et al., 1987.)

DISEASES AND TREATMENTS: PARKINSON'S DISEASE

proposed that underactivation of SMA in PD patients made them more reliant on external cues so that they did not use predictive models when cues were available. Premovement activity could, however, be induced in patients if they were asked to attend to the time of the next movement, and this improved task performance (Cunnington et al., 1999). The authors suggested that attentional processes allow lateral premotor systems, less impaired by basal ganglia dysfunction, to compensate for deficiencies in midline motor systems that normally are active in internally generated tasks. Inherent noise in the signal means that the BP is not a reliable measure in individual cases. However, group studies have shown that it is sensitive to the effects of treatment, being larger, particularly in the second part, in patients after pallidotomy (Limousin et al., 1999). These early studies involved subjects repeating the same movement over many trials. Later studies have required that subjects chose to make different movements (e.g. moving a joystick up, down, left or right) on each trial. In healthy subjects, the BP is much larger than when subjects make the same movement on each occasion, perhaps reflecting the additional processing necessary to choose between movements on each trial (Touge et al., 1995; Praamstra et al., 1996a, b). Praamstra et al. (1998) used dipole modeling to show that the most likely source for the extra activation was the SMA. This extra activation is lacking in patients with Parkinson's disease, consistent with the model outlined above. 26.1.4.2. Lateralized readiness potential (LRP) Normal functioning or even compensation for underactive midline motor areas by the lateral motor cortical areas is supported by LRP studies. Praamstra et al. (1998) employed a task using incompatible target and distractor elements to precue which hand was to be used for a forthcoming movement. Sometimes the precue made subjects expect that they would have to respond with the wrong hand. In these trials the reaction time was longer than if the precue biased responses to the correct hand. EEG recording of the lateralized readiness potential (the difference between motor cortical activity in the two hemispheres) showed that the distractor cue often caused early activation of the motor cortex controling the wrong response hand. This activity was

423

larger in patients, and was associated with a longer delay in producing a correct behavioral reaction. The interpretation is that external sensory stimuli in the precue can bias activity towards one or other side of the brain in expectation of a subsequent imperative stimulus. In patients with Parkinson's disease this system seems to work more efficiently than in healthy subjects, and is compatible with the idea that there is increased access of external cues to motor areas of cerebral cortex. Praamstra and Plat (2001) recorded the LRP in a task where the visual instruction to press a right or left hand key was presented to one or other side of a central fixation spot. Subjects therefore had to shift attention to one or other side before they could react. This attentional shift elicited an attention-related ERP component just preceding the movementrelated LRP, i.e. the N2pc. The N2pc was distributed broadly on the scalp from occiput to frontal areas, and the authors suggested it represented simultaneous attention-related activity in occipital and motor areas of the cortex. Interestingly, the attentionrelated activity attributed to motor areas was of significantly higher amplitude in Parkinson's disease patients compared to age-matched controls. This suggests that changes in motor excitability produced by shifts of attention are larger in patients, and this again may be one mechanism that is employed to compensate for decreased activity in basal ganglia output. 26.1.4.3. Contingent negative variation (CNV) Although the CNV and the BP share some common mechanisms, it is usually found that the CNV is more depressed in Parkinson's disease than the BP (Ikeda et al., 1997; Gerschlager et al., 1999). An important difference between the tasks may be that the S2 stimulus in the CNV paradigm acts as a stimulus for the next movement, whereas in the BP, no cue for movement is given. We might speculate that healthy subjects prepare for the forthcoming movement to the same extent in the CNV task as they do in the BP task. This yields a large amplitude potential in both situations. Patients with Parkinson's disease may prepare less effectively in the BP situation, but in the CNV, they may tend to rely on S2 as an external cue to trigger release of movement. Thus, in the CNV paradigm, there will be much more difference in the level of preparation prior to S2 in patients and normals than in the BP task. The

424

CNV improves significantly after DBS of the subthalamic nucleus (Gerschlager et al., 1999) (Fig. 4). 26.1.4.4. Event related synchronization! desynchronization (ERS/ERD) Abnormalities in cortical activation prior to and during movement have been also found with the technique of event-related desynchronization (ERD) (Defebvre et al., 1996; Magnani et al., 1998). The amount of power in the alpha (10 Hz) and beta (20 Hz) ranges of EEG activity decreases about one second before onset of movement, and remains lower than at rest while movement occurs. It has been suggested that the 10-20 Hz rhythm occurs because the activity of cortical neurons tends to become synchronized during periods of relative inactivity. If so, ERD is a measure of cortical activation that reflects uncoupling of the population activity into more discrete temporal and spatial patterns. The duration of the ERD prior to voluntary movements is shorter in patients with Parkinson's disease and the pattern of movement related attenuation of the alpha and beta rhythms during various

J.e. ROTHWELL

types of motor tasks is abnormal. Brown and Marsden (1999) found that dopaminergic stimulation in PD restores the movement-related attenuation of the alpha and beta rhythms. This effect was specific for the motor areas involved in the motor task and correlated with the improvement of bradykinesia. Similar findings were present in the study of Wang et al. (1999) on simple and complex movements. The above studies suggest that the basal ganglia have a role in releasing cortical elements from idling rhythms during voluntary movement. In a recent report, Brown et al. (2001) carried this idea one stage further by examining the effect of dopaminergic stimulation on the coherence between activity at different frequencies in different nuclei of the basal ganglia. They recorded local potentials from patients with implanted electrodes in STN and internal pallidum, and found that when OFF medication there was clear coherence between activity in the 20-30 Hz range, whereas ON medication this changed to 60-70 Hz. Interestingly, the coherence at 60-70 Hz increased during movement whereas that in the OFF state in the 20-30 Hz range decreased (Cassidy et al., 2002).

Fig. 4. Mean CNV traces for Parkinson's disease subjects with subthalamic nucleus stimulation on (thick line) compared with off (thin line) stimulation. Potentials are shown from 200 ms before the warning stimulus until 600 ms after the imperative stimulus (total duration 2.8 s). Note the clear increase in all leads during stimulation. (From Gerschlager et al., 1999.)

DISEASES AND TREATMENTS: PARKINSON'S DISEASE

26.1.4.5. Somatosensory evoked potentials Neurophysiological studies of sensory input have generally found there to be no change in the parietal components of the SEP, but a number of studies have shown that the frontal N30 may be reduced in amplitude (De Mari et aI., 1995; Garcia et aI., 1995; Onofrj et aI., 1995; Traversa et aI., 1995; Drory et aI., 1998; Bostantjopoulou et aI., 2000) Treatment with apomorphine or pallidal stimulation through chronic implanted electrodes tends to increase the amplitude (De Mari et aI., 1995; Pierantozzi et aI., 1999). If part of the N30 is related to activation in SMA, then the reduced amplitude in patients may be another sign of the underactivation of midline motor areas in Parkinson's disease. In addition, the enhancement of the N30 may be related to the enlarged long latency stretch reflexes seen in patients (Ashbridge et aI., 1997). 26.1.5. Transcranial magnetic stimulation; physiological mechanisms of bradykinesia 26.1.5.1. Thresholds, cortical inhibition and silent period Corticomotoneuronal conduction and motor threshold are both normal in Parkinson's disease (Dick et aI., 1984; Priori et aI., 1994; Valls-Sole et aI., 1994; Ridding et aI., 1995). Indeed, the fact that movements elicited by direct stimulation of the motor cortex are the same whether the stimulus is given when patients are immobile and OFF therapy or dyskinetic and ON therapy confirms that bradykinesia is not primarily the result of any deficit in the final output pathways of the motor areas of cortex. Despite the lack of change in threshold, the slope of the relationship between stimulus intensity and response size is steeper than normal when tested at rest (Valls-Sole et aI., 1994; Filippi et aI., 2001). Perhaps as a result of this, voluntary contraction facilitates responses less than in normal subjects (Valls-Sole et aI., 1994). The implication is that the distribution of cortical excitability at rest is skewed towards higher values than normal. This seems unlikely to be the result of a primary basal ganglia deficit, and may well be an attempt to compensate for slow recruitment of commands to move by making it easier to recruit activity from a resting state. There are also changes in the excitability of cortical inhibitory circuits. The silent period is

425

shorter in bradykinetic patients (Priori et aI., 1994; Deuschl, 1999), and normalized by treatment with LDOPA. Short latency inhibition as tested at rest using the double pulse paradigm of Kujirai et aI. (1993) is also smaller in patients than normal (Ridding et aI., 1995), and normalized by administration of apomorphine (Pierantozzi et aI., 2001) (Fig. 5). Kleine et al. (2001) recently suggested that this may increase the number of I wave volleys recruited by each TMS pulse and lead to a longer depolarization of spinal motoneurons than normal. It is difficult to know whether these changes contribute to bradykinesia or whether they represent some form of compensatory process. Ridding et aI. (1995) suggested that activity in cortical inhibitory circuits is normally used to reinforce spatial and temporal patterns of cortical excitability appropriate for a forthcoming or ongoing task. Even though basal ganglia projections to primary motor cortex are smaller than those to midline motor areas, such activity could still contribute to patterns of cortical inhibition. If so, the deficits in cortical inhibition might be a primary factor in bradykinesia since one source of information about the pattern of cortical excitability for a task would have been lost. An alternative explanation is that the reduced inhibition is a compensatory mechanism that makes it easier for motor commands to access cortical output. Bilateral stimulation of the subthalamic nucleus can normalize the amount of short latency paired pulse inhibition (Cunic et aI., 2002). A final type of interaction between two suprathreshold stimuli was tested by Berardelli et aI. (1996b). They performed these experiments during a slight voluntary contraction and found that the test response was inhibited more than normal at 100 and 150 ms. It may be that during contraction, PD patients lack the facilitation seen in healthy subjects. The inhibition normalized after treatment with LDOPA. 26.1.5.2. TMS before movement TMS has also been employed to examine the buildup of corticospinal excitability prior to a reaction time movement (e.g. Starr et aI., 1988). Responses to stimuli given early in the reaction period (up to 70 ms after the imperative stimulus in a simple reaction task) elicit responses with the same probability or the same amplitude as immediately before the reaction. Thereafter, the probability or

426

J.e. ROTHWELL

Fig. 5. The time course of ICI in PD patients, as tested by a conditioning-test paired-pulse TMS protocol at 1-6 ms ISIs. Motor responses were obtained from a representative PD patient recorded in off DBS (A), during bilateral GPi DBS (B), bilateral STN DBS (C) and finally, during apomorphine infusion (D). It is worth noting that both STN and GPi DBS reversed the reduced ICI observed in off DBS mimicking the apomorphine effect. In each panel, the top two traces show the response to the conditioning and the test stimuli given alone. In this figure, each trace is the average of four sweeps. TMS artefacts correspond to the vertical bars at constant and variable position, respectively. (From Pierantozzi et al., 2001.)

427

DISEASES AND TREATMENTS: PARKINSON'S DISEASE

amplitude of responses increases gradually until the onset of overt EMG activity (Pascual-Leone et al., 1992). The latter period represents build-up of excitability in the corticospinal system (motor cortex and spinal cord) that finally causes visible movement. The former is usually taken as a measure of the time taken to activate the commands to move. In bradykinetic patients with Parkinson's disease, the duration of the initial period is unchanged whereas the latter period of increasing excitability is longer than in healthy subjects (Pascual-Leone et al., 1994). According to this interpretation, commands are selected as fast as in healthy subjects, but the initial execution of the commands is slow. However, there are two other possible explanations of these results. First, programming of movement commands may continue during the initial part of the execution of a task. If this process were slower than normal then it would increase the duration of the premovement increase in corticospinal excitability. A second possibility is that two processes are involved in movement execution. One may create the increase in corticospinal excitability prior to movement, and the other may be the movement command itself. Again, the first phase could be slow in patients and delay the onset of movement. It is not clear what system could act as a modulator of excitability prior to movement. However, it is interesting to note that the excitability of the reticulospinal system, tested either by the startle response (Vidailhet et al., 1992) (see also chapter by Valls-Sole) or audiospinal facilitation of H-reflexes (Delwaide et al., 1993), is decreased in Parkinson's disease. If this were involved in setting excitability prior to movement then it could account for the slow build up of responses to TMS. 26.1.5.3. TMS interruption of movement Cunnington et al. (1996) asked patients to make a rapid sequence of finger movements from left to right along a tapping board. They gave single pulse TMS at maximum intensity through a 5 em figure eight coil over the SMA randomly with respect to the button presses. They found that if the stimuli occurred in the early part of the interval between two presses, then the next button press was slower than expected. There was no effect if the stimulus occurred in the late part of the button pressing interval, and no effect in healthy subjects. They suggested, in line with data from PET experiments

(Jenkins et al., 1992) that this indicated that the SMA function was compromized in patients and therefore easier to disrupt than in healthy subjects. 26.2. Rigidity

Meara and Cody (1992, 1993) showed that rigidity at the wrist is related to the level of stretch evoked EMG activity in wrist flexor/extensor muscles during manipulation of the wrist joint. However several pathways, from monosynaptic Ia to group II and long loop responses can potentially contribute to such reflexes, and there is evidence that Parkinson's disease may affect transmission in all of them. Clinically, the tendon jerk is usually said to be normal in Parkinson's disease, but measurements of EMG responses evoked in the FDI muscle by mechanical muscle stretch showed that the tendon jerk component of the response was small or absent compared with normal (Wenzelburger et al., 2000). Since the EMG response to electrical stimulation of muscle afferents was the same as normal, Noth et al. (1988) suggested that the muscle spindle was less sensitive to phasic muscle stretch in Parkinson's disease. They speculated that increased static gamma drive might be responsible and increase the longer latency responses due to activity in group II afferents. There is some evidence that group II reflexes are enhanced in Parkinson's disease. Berardelli et al. (1983) examined the long-latency stretch reflexes evoked in the triceps surae and tibialis anterior muscle during passive movements of the ankle. These responses were tonic, lasting as long as the stretch was applied, and the authors considered them to be of group II origin. They were larger in patients than in healthy subjects. Long loop reflexes are also enhanced in Parkinson's disease (Rothwell et al., 1983; Ashbridge et al., 1997). These reflexes, which are best seen in muscles that work on the fingers and wrist employ a transcortical pathway that operates in parallel with the spinal tendon jerk pathway. The afferents responsible are thought to be Ia fibers from primary spindle endings. In general, the degree of enhancement is not related to clinical measures of muscle stiffness, but in individual patients, pallidotomy has been shown to decrease both the long latency response and the clinical rigidity (Limousin et al., 1999; Hayashi et al., 2001) (Fig. 6).

J.e. ROTHWELL

428

Fig. 6. Averaged rectified EMG responses from the wrist flexors and kinematics changes obtained by 20 perturbations from one patient to displacements of the wrist joint in extensor direction. The patient was instructed not to respond to the perturbation (passive) or to oppose it (active). Dotted vertical line indicates the time at which the torque motor was turned on to initiate the displacement. Dotted horizontal line shows the initial hand position, a level at which hand velocity was zero, or an electrically silent level in the flexor muscle activity. Vertical lines indicate the onset of Ml (30 ms), M2 (60 ms), and the time at 90 ms, from left to right, respectively. Two traces in each figure were superimposed (thin trace obtained before operation, thick trace after pallidotomy which is indicated by arrowhead). From top to bottom, hand trajectory, its velocity, and the integrated EMG of wrist flexormuscles. (From Hayashi et al., 2001.)

It should be noted that increased stretch reflexes could potentially contribute to bradykinesia if they were elicited in an antagonist muscle during an active isotonic contraction of the agonist. Johnson et al. (1991) tested this hypothesis by using a torque motor to stretch muscles unexpectedly during active sinusoidal movements of the wrist. They showed that reflexes elicited in the antagonist muscle were not suppressed as much as in normal subjects, and that the degree of abnormality was related to the amount of clinical bradykinesia. The one flaw in this argument was that the amount of activity in the antagonist muscle during unperturbed flexion/extension movements was no greater than that seen in normal subjects. Thus, there was no evidence in the actual movements tested that antagonist cocontraction could have been a limiting factor. Indeed, cocontraction has never been described as an important feature even in very rapid movements where the triphasic ballistic movement EMG pattern has been analyzed in some detail. The conclusion must be that the role of rigidity in bradykinesia has yet to be proven conclusively. In addition to these changes in stretch reflex pathways, there are changes in two other spinal reflex circuits that could potentially contribute to rigidity. Delwaide et al. (1991) used H-reflex testing in calf muscles to show that the excitability of the Ib inhibitory pathway from gastrocnemius to soleus was reduced in patients with Parkinson's disease. The degree of reduction correlated with clinical estimates of ankle rigidity. The authors also drew attention to the increase in excitability of the Ia inhibitory pathway that had been described between tibialis anterior and soleus in Parkinson's disease and suggested that both were due to overactivity of the descending projections from the nucleus reticularis gigantocellularis.

26.3. Posture Three types of measure are generally applied to postural control in Parkinson's disease: balance control during quiet stance, EMG and force responses to both externally and self-initiated disturbances of balance, and assessment of initiation and maintenance of gait.

26.3.1. Quiet stance The simplest measure of stability in quiet stance is a timed balance test in which subjects have to stand

DISEASES AND TREATMENTS: PARKINSON'S DISEASE

429

Fig. 7. Mean 95% confidence ellipses, computed from bidimensional center of foot pressure excusions of control subjects and subjects with Parkinson's disease in the four test conditions. AP, anteroposterior; DBS, under deep brain stimulation only; DBS + DOPA, under deep brain stimulation and levodopa; DOPA, under levodopa only; ML, mediolateral; OFF, no treatment. (From Rocchi et al., 2002.)

unsupported for 30 s or so. Only in cases of severe postural instability are such tests sensitive enough to detect changes due to treatment or disease progression. Better measures can be obtained, as described in the chapter by Bloem, using a force platform to quantify motion of the center of pressure or center of gravity in particular stance conditions (e.g. feet together). These can be supplemented with 3D measures of the motion of body segments. Postural sway has been reported to be normal (Schieppati and Nardone, 1991), reduced (Horak et aI., 1992) or increased (Rocchi et aI., 2002) in patients with Parkinson's disease. The variability probably reflects both patient selection as well whether subjects were tested ON or OFF medication. In a recent study, Rocchi et aI. (2002) found that sway was larger than normal when OFF therapy, and that it increased further when patients were ON medication (Fig. 7). Part of the difference may have

been due to increased dyskinesias when ON treatment, but this was not thought to be a major factor. Instead, the authors proposed that when OFF, patients were stiffer at the ankle than when ON treatment. This led to a reduced sway path and also may have been responsible for the higher frequencies of sway seen when OFF. Interestingly, DBS of pallidum or subthalamus reduced sway, and may be one factor in the improvement of postural control seen during DBS. Some workers have emphasized that medio-lateral control is particularly compromised in Parkinson's disease (Mitchell et aI., 1995), and this would be consistent with the observation of Maki arid colleagues that increased lateral away is associated with increased risk of falling in the elderly (Maki et aI., 1994; Maki and Mcilroy, 1996). Finally, there is some evidence that patients are more reliant on visual information to stabilize their posture than

r.c. ROTHWELL

430

normal. Bronstein et al. (1990) investigated the body sway that can be provoked by slow movement of the visual surround. In Parkinson's disease, the amount of sway was larger than normal and did not habituate as quickly on repeated presentations of the same stimulus. 26.3.2. Responses to perturbations External disturbances to standing posture can be applied using a moveable platform. Toe up rotations provoke a medium latency (ML) stretch reflex response in the soleus, and a later long latency (LL) response in the tibialis anterior. The former tends to destabilize balance whilst the latter is a corrective response that pulls the body forwards (see chapter by Bloem). In Parkinson's disease, the ML response is larger than normal, and the LL response is less sensitive to the amplitude of the disturbance and less influenced by anticipatory set than in healthy subjects (Beckley et al., 1993; Bloem et al., 1994, 1996). Both factors will tend to make patients less stable in the face of perturbations than normal. Treatment with L-DOPA reduces the ML response but has little effect on the LL response. The latter failure may be related to the poor response of postural instability to drug treatment. Perturbations can also be applied by pushing or pulling on other parts of the body. Traub et al. (1980) used a motor to deliver brisk pulls to the wrist in standing subjects. This evoked short latency reflex responses in leg muscles that opposed the direction of pull. Since the responses occurred before any sway of the body was evident, they were thought to be driven by input from receptors in the arm. Such responses were reduced in Parkinson's disease particularly in patients with greater clinical instability. Voluntary movement also perturbs posture, but to reduce the movement of the center of mass, all voluntary limb movements in healthy unsupported subjects are accompanied, or even preceded, by activation of postural muscles. For example, the act of raising the arms in front of the body produces activity in posterior leg and back muscles that causes the body to sway backwards and reduces the forwards displacement of the center of mass produced by the arm movement. Similarly, the act of standing onto tip toe begins initially with reduced activity of the triceps surae and activation of the tibialis anterior to move the center of mass forwards

followed by the main activation of triceps surae to plantarflex the ankle. The timing and amplitude of such responses can be measured relatively easily using EMG. Calculation of the force produced can be performed if force plate data is available. In Parkinson's disease, the timing of anticipatory activity is approximately normal or prolonged (Dick et al., 1986; Frank et al., 2000), but the magnitude of the activity is reduced (Rogers et al., 1987). Much of the latter is accounted for by the fact that the amplitude of anticipatory activity depends on the speed of the prime movement (Lee et al., 1987), and this is reduced in Parkinson's disease. 26.3.3. Gait Gait can be analyzed in a number of ways from timed walking to the biomechanics of force development during initiation of a step. The most frequently used measures are the cadence (the number of steps per minute) and the stride length during unperturbed walking in a straight line. In healthy subjects there is a linear relation between the stride length and the walking speed. The slope of this relationship is the same in patients with Parkinson's disease, but the intercept is higher. Effectively, for any given cadence the stride length is reduced (Ebersbach et al., 1999; Morris et al., 2001). Treatment with L-DOPA improves walking speed by increasing stride length (Defebvre et al., 2002) for any given cadence. Force platform and EMG measures have been used to investigate the initiation of gait. However, the general pattern of shifts in center of pressure and center of mass are normal when normalized to the gait velocity (Halliday et al., 1998), suggesting that the program to initiate gait is intact. References Agostino, R, Berardelli, A, Formica, A, Accomero, N and Manfredi, M (1992) Sequential ann movements in patients with Parkinson's disease, Huntington's disease and dystonia. Brain, 1l5: 1481-1495. Ashbridge, E, Walsh, V and Cowey, A (1997) Temporal aspects of visual search studied by transcranial magnetic stimulation. Neuropsychologia, 35: 1l21-1l31. Beckley, OJ, Bloem, BR and Remler, MP (1993) Impaired scaling of long latency postural reflexes in patients with Parkinson's disease. Electroencephalogr. Clin. Neurophysiol., 89: 22-28. Benecke, R, Rothwell, IC, Dick, JP, Day, BL and Marsden, CD (1986) Performance of simultaneous

DISEASES AND TREATMENTS: PARKINSON'S DISEASE

movements in patients with Parkinson's disease. Brain, 109: 739-757. Benecke, R, Rothwell, JC, Dick, JP, Day, BL and Marsden, CD (1987a) Disturbance of sequential movements in patients with Parkinson's disease. Brain, 110: 361-379. Benecke, R, Rothwell, JC, Dick, JP, Day, BL and Marsden, CD (1987b) Simple and complex movements off and on treatment in patients with Parkinson's disease. J. Neurol. Neurosurg. Psychiatry, 50: 296303. Berardelli, A, Sabra, AF and Hallett, M (1983) Physiological mechanisms of rigidity in Parkinson's disease. J. Neurol. Neurosurg. Psychiatry, 46: 45-53. Berardelli, A, Rothwell, JC, Day, BL, Kachi, T and Marsden, CD (1984) Duration of the first agonist EMG burst in ballistic arm movements. Brain Res., 304: 183-187. Berardelli, A, Dick, JP, Rothwell, JC, Day, BL and Marsden, CD (1986) Scaling of the size of the first agonist EMG burst during rapid wrist movements in patients with Parkinson's disease. J. Neurol. Neurosurg. Psychiatry, 49: 1273-1279. Berardelli, A, Hallett, M, Rothwell, JC, Agostino, R, Manfredi, M, Thompson, PD and Marsden, CD (1996a) Single-joint rapid arm movements in normal subjects and in patients with motor disorders. Brain, 119: 661-674. Berardelli, A, Rona, S, Inghilleri, M and Manfredi, M (1996b) Cortical inhibition in Parkinson's disease - a study with paired magnetic stimulation. Brain, 119: 71-77. Bloem, BR, Beckley, DJ, Van Dijk, JG, Zwinderman, AH, Remler, MP, Langston, JW and Roos, RA (1994) Medium latency stretch reflexes in young-onset Parkinson's disease and MPTP-induced parkinsonism. J. Neurol. Sci., 123: 52-58. Bloem, BR, Beckley, DJ, Van Dijk, JG, Zwinderman, AH, Remler, MP and Roos, RA (1996) Influence of dopaminergic medication on automatic postural responses and balance impairment in Parkinson's disease. Mov. Disord., 11: 509-521. Bloxham, CA, Dick, DJ and Moore, M (1987) Reaction times and attention in Parkinson's disease. J. Neurol. Neurosurg. Psychiatry, 50: 1178-1183. Bostantjopoulou, S, Katsarou, Z, Zafiriou, D, Gerasimou, G, Alevriadou, A, Georgiadis, G, Kiosseoglou, G and Kazis, A (2000) Abnormality of N30 somatosensory evoked potentials in Parkinson's disease: a multidisciplinary approach. Neurophysiol. Clin., 30: 368376. Bronstein, AM, Hood, JD, Gresty, MA and Panagi, C (1990) Visual control of balance in cerebellar and parkinsonian syndromes. Brain, 113: 767-779.

431 Brown, P and Marsden, CD (1999) Bradykinesia and impairment of EEG desynchronization in Parkinson's disease. Mov. Disord., 14: 423-429. Brown, P, Corcos, DM and Rothwell, JC (1997) Does parkinsonian action tremor contribute to muscle weakness in Parkinson's disease? Brain, 120: 401-408. Brown, P, Corcos, DM and Rothwell, IC (1998) Action tremor and weakness in Parkinson's disease: a study of the elbow extensors. Mov. Disord., 13: 56-60. Brown, P, Oliviero, A, Mazzone, P, Insola, A, Tonali, P and Di, L, V (200 I) Dopamine dependency of oscillations between subthalamic nucleus and pallidum in Parkinson's disease. J. Neurosci., 21: 1033-1038. Brown, RG and Marsden, CD (1991) Dual task performance and processing resources in normal subjects and patients with Parkinson's disease. Brain, 114: 215-231. Brown, RG, Jahanshahi, M and Marsden, CD (1993) Response choice in Parkinson's disease. The effects of uncertainty and stimulus-response compatibility. Brain, 116: 869-885. Cassidy, M, Mazzone, P, 01iviero, A, Insola, A, Tonali, P, Di, LV and Brown, P (2002) Movement-related changes in synchronization in the human basal ganglia. Brain, 125: 1235-1246. Castiello, U and Bennett, KM (1997) The bilateral reachto-grasp movement of Parkinson's disease subjects. Brain, 120: 593-604. Corcos, DM, Chen, CM, Quinn, NP, McAuley, J and Rothwell, JC (1996) Strength in Parkinson's disease: relationship to rate of force generation and clinical status. Ann. Neurol., 39: 79-88. Cunic, D, Roshan, L, Khan, FI, Lozano, AM, Lang, AE and Chen, R (2002) Effects of subthalamic nucleus stimulation on motor cortex excitability in Parkinson's disease. Neurology, 58: 1665-1672. Cunnington, R, Iansek, R, Bradshaw, JL and Phillips, JG (1995) Movement-related potentials in Parkinson's disease presence and predictability of temporal and spatial cues. Brain, 118: 935-950. Cunnington, R, Iansek, R, Thickbroom, GW, Laing, BA, Mastaglia, FL, Bradshaw, JL and Phillips, JG (1996) Effects of magnetic stimulation over supplementary motor area on movement in Parkinson's disease. Brain, 119: 815-822. Cunnington, R, Iansek, R, Johnson, KA and Bradshaw, JL (1997) Movement-related potentials in Parkinson's disease. Motor imagery and movement preparation. Brain, 120: 1339-1353. Cunnington, R, Iansek, R and Bradshaw, JL (1999) Movement-related potentials in Parkinson's disease: external cues and attentional strategies. Mov. Disord., 14: 63-68.

432 De Mari, M, Margari, L, Lamberti, P, Iliceto, G and Ferrari, E (1995) Changes in the amplitude of the N30 frontal component of SEPs during apomorphine test in parkinsonian patients. J. Neural Transm., Suppl. 45: 171-176. Defebvre, L, Bourriez, JL, Destee, A and Guieu, JED (1996) Movement related desynchronization pattern preceding voluntary movement in untreated Parkinson's disease. J. Neurol. Neurosurg. Psychiatry, 60: 376-382. Defebvre, LJ, Krystkowiak, P, Blatt, JL, Duhamel, A, Bourriez, JL, Perina, M, Blond, S, Guieu, JD and Destee, A (2002) Influence of pallidal stimulation and levodopa on gait and preparatory postural adjustments in Parkinson's disease. Mov. Disord., 17: 76-83. Delwaide, PJ, Pepin, JL and Maertens de Noordhout, A (1991) Short-latency autogenic inhibition in patients with parkinsonian rigidity. Ann. Neurol., 30: 83-89. Delwaide, PJ, Pepin, JL and Maertens de Noordhout, A (1993) The audiospinal reaction in parkinsonian patients reflects functional changes in reticular nuclei. Ann. Neurol., 33: 63-69. Demirci, M, Grill, S, McShane, L and Hallett, M (1997) A mismatch between kinesthetic and visual perception in Parkinson's disease. Ann. Neurol. 41: 781-788. Deuschl, G (1999) Neurophysiological tests for the assessment of tremors. Adv. Neurol., 80: 57-65. Dick JP, Cowan 1M, Day BL, Berardelli A, Kachi T, Rothwell, JC and Marsden, CD (1984) The corticomotoneuron connection is normal in Parkinson's disease. Nature, 310: 407-409. Dick, JP, Rothwell, JC, Berardelli, A, Thompson, PD, Gioux, M, Benecke, R, Day, BL and Marsden, CD (1986) Associated postural adjustments in Parkinson's disease. 1. Neurol. Neurosurg. Psychiatry, 49: 13781385. Dick, JP, Cantello, R, Buruma, 0, Gioux, M, Benecke, R, Day, BL, Rothwell, JC, Thompson, PD and Marsden, CD (1987) The Bereitschaftspotential, L-DOPA and Parkinson's disease. Electroencephalogr. Clin. Neurophysiol., 66: 263-274. Donoghue, JP (1987) Cholinergic modulation of sensory responses in cerebral cortex. NIDA Res. Monogr., 78: 169-188. Drory, YE, Inzelberg, R, Groozman, GB and Korczyn, AD (1998) N30 somatosensory evoked potentials in patients with unilateral Parkinson's disease. Acta Neurol. Scand., 97: 73-76. Ebersbach, G, Sojer, M, Valldeoriola, F, Wissel, J, Muller, J, Tolosa, E and Poewe, W (1999) Comparative analysis of gait in Parkinson's .disease, cerebellar ataxia and subcortical arteriosclerotic encephalopathy. Brain, 122: 1349-1355.

J.e. RafHWELL Evarts, EV, Teravainen, H and Calne, DB (1981) Reaction time in Parkinson's disease. Brain, 104: 167-186. Filippi, MM, Oliveri, M, Pasqualetti, P, Cicinelli, P, Traversa, R, Vernieri, F, Palmieri, MG and Rossini, PM (2001) Effects of motor imagery on motor cortical output topography in Parkinson's disease. Neurology, 57: 55-61. Frank, JS, Horak, FB and Nutt, J (2000) Centrally initiated postural adjustments in parkinsonian patients on and off levodopa. J. Neurophysiol., 84: 2440-2448. Garcia, PA, Aminoff, MJ and Goodin, DS (1995) The frontal N30 component of the median-derived SEP in patients with predominantly unilateral Parkinson's disease. Neurology, 45: 989-992. Georgiou, N, Iansek, R, Bradshaw, JL, Phillips, JG, Mattingley, JB and Bradshaw, JA (1993) An evaluation of the role of internal cues in the pathogenesis of parkinsonian hypokinesia. Brain, 116: 1575-1587. Georgiou, N, Bradshaw, JL, Iansek, R, Phillips, JG, Mattingley, JB and Bradshaw, JA (1994) Reduction in external cues and movement sequencing in Parkinson's disease. J. Neurol. Neurosurg. Psychiatry, 57: 368-370. Gerschlager, W, Alesch, F, Cunnington, R, Deecke, L, Dimberger, G, Endl, W, Lindinger, G and Lang, W (1999) Bilateral subthalamic nucleus stimulation improves frontal cortex function in Parkinson's disease. An electrophysiological study of the contingent negative variation. Brain, 122: 2365-2373. Hallett, M and Khoshbin, S (1980) A physiological mechanism of bradykinesia. Brain, 103: 301-314. Halliday, SE, Winter, DA, Frank, JS, Patla, AE and Prince, F (1998) The initiation of gait in young, elderly, and Parkinson's disease subjects. Gait Posture, 8: 8-14. Hayashi, R, Hashimoto, T, Tada, T and Ikeda, S (2001) Relation between changes in long-latency stretch reflexes and muscle stiffness in Parkinson's disease comparison before and after unilateral pallidotomy. Clin. Neurophysiol., 112: 1814-1821. Horak, FB, Nutt, JG and Nashner, LM (1992) Postural inflexibility in parkinsonian subjects. J. Neurol. Sci., Ill: 46-58. Ikeda, A, Shibasaki, H, Kaji, R, Terada, K, Nagarnine, T, Honda, M and Kimura, J (1997) Dissociation between contingent negative variation (CNV) and Bereitschaftspotential (BP) in patients with parkinsonism. Electroencephalogr. Clin. Neurophysiol., 102: 142151. Jahanshahi, M, Brown, RG and Marsden, CD (1992) Simple and choice reaction time and the use of advance information for motor preparation in Parkinson's disease. Brain, 115: 539-564. Jahanshahi, M, Brown, RG and Marsden, CD (1993) A comparative study of simple and choice reaction time

DISEASES AND TREATMENTS: PARKINSON'S DISEASE

in Parkinson's, Huntington's and cerebellar disease. J. Neurol. Neurosurg. Psychiatry, 56: 1169-1177. Jahanshabi, M, Jenkins, IH, Brown, RG, Marsden, CD, Passingham, RE and Brooks, DJ (1995) Self-initiated vs. externally triggered movements. I. An investigation using measurement of regional cerebral blood flow with PET and movement-related potentials in normal and Parkinson's disease subjects (see comments). Brain, 118: 913-933. Jenkins, IH, Fernandez, W, Playford, ED, Lees, AJ, Frackowiak, RJ, Passingham, RE and Brooks, DJ (1992) Impaired activation of the supplementary motor area in Parkinson's disease is reversed when akinesia is treated with apomorphine. Ann. Neurology, 32: 749757. Johnson, MT, Kipnis, AN, Lee, MC, Loeweson, RB and Ebner, TJ (1991) Modulation of the stretch reflex during volitional sinusoidal tracking in Parkinson's disease. Brain, 114: 443-460. Jordan, N, Sagar, HJ and Cooper, JA (1992) A component analysis of the generation and release of isometric force in Parkinson's disease. J. Neurol. Neurosurg. Psychiatry, 55: 572-576. Kleine, BU, Praamstra, P, Stegeman, DF and Zwarts, MJ. Impaired motor cortical inhibition in Parkinson's disease: motor unit responses to transcranial magnetic stimulation. Exp. Brain Res., 138: 477-483. Klockgether, T, Borutta, M, Rapp, H, Spieker, S and Dichgans, J (1995) A defect of kinaesthesia in Parkinson's disease. Mov. Disord., 10: 460-465. Kujirai, T, Caramia, MD, Rothwell, JC, Day, BL, Thompson, PD, Ferbert, A, Wroe, S, Asselman, P and Marsden, CD (1993) Corticocortical inhibition in human motor cortex. J. Physiol. Lond., 471: 501-519. Kutukcu, Y, Marks, WJ Jr., Goodin, DS and Aminoff, MJ (1999) Simple and choice reaction time in Parkinson's disease. Brain Res., 815: 367-372. Lee, WA, Buchanan, TS and Rogers, MW (1987) Effects of arm acceleration and behavioral conditions on the organization of postural adjustments during arm flexion. Exp. Brain Res., 66: 257-270. Limousin, P, Brown, RG, Jahanshahi, M, Asselman, P, Quinn, NP, Thomas, D, Obeso, JA and Rothwell, JC (1999) The effects of posteroventral pallidotomy on the preparation and execution of voluntary hand and arm movements in Parkinson's disease. Brain, 122: 315327. Magnani, G, Cursi, M, Leocani, L, Volonte, MA, Locatelli, T, Elia, A and Comi, G (1998) Event-related desynchronization to contingent negative variation and self-paced movement paradigms in Parkinson's disease. Mov. Disord., 13: 653-660. Maki, BE and McIlroy, WE (1996) Postural control in the older adult. Clin. Geriatr. Med., 12: 635-658.

433 Maki, BE, Holliday, PJ and Topper, AK (1994) A prospective study of postural balance and risk of falling in an ambulatory and independent elderly population. J. Gerontol., 49: M72-M84. Meara, RJ and Cody, FW (1992) Relationship between electromyographic activity and clinically assessed rigidity studied at the wrist joint in Parkinson's disease. Brain, 115: 1167-1180. Meara, RJand Cody, FW (1993) Stretch reflexes of individual parkinsonian patients studied during changes in clinical rigidity following medication. Electroencephalogr. Clin. Neurophysiol., 89: 261-268. Mitchell, SL, Collins, JJ, De, LC, Burrows, A and Lipsitz, LA (1995) Open-loop and closed-loop postural control mechanisms in Parkinson's disease: increasedmediolateral activity during quiet standing. Neurosci. Lett., 197: 133-136. Morris, ME, Huxham, FE, McGinley, J and Iansek, R (2001) Gait disorders and gait rehabilitation in Parkinson's disease. Adv. Neurol., 87: 347-361. Muller, T, Benz, Sand Bornke, C (2001) Delay of simple reaction time after levodopa intake. Clin. Neurophysiol., 112: 2133-2137. Oliveira, RM, Gurd, JM, Nixon, P, Marshall, JC and Passingham, RE (1998) Hypometria in Parkinson's disease: automatic vs. controlled processing. Mov. Disord., 13: 422-427. Onofrj, M, Fulgente, T, Malatesta, G, Ferracci, F, Thomas, A, Curatola, L, Bollettini, F and Ragno, M (1995) The abnormality of N30 somatosensory evoked potential in idiopathic Parkinson's disease is unrelated to disease stage or clinical scores and insensitive to dopamine manipulations. Mov. Disord., 10: 71-80. Pascual-Leone, A, Valls-Sole, J, Wassermann, EM, Brasil, NJ, Cohen, LG and Hallett, M (1992) Effects of focal transcranial magnetic stimulation on simple reaction time to acoustic, visual and somatosensory stimuli. Brain, 115: 1045-1059. Pascual-Leone, A, Valls-Sole, J, Brasil-Neto, J, Cohen, LG and Hallett, M (1994) Akinesia in Parkinson's disease. I. Shortening of simple reaction time with focal, singlepulse transcranial magnetic stimulation. Neurology, 44: 884-889. Phillips, JG, Martin, KE, Bradshaw, JL and Iansek, R (1994) Could bradykinesia in Parkinson's disease simply be compensation. J. Neurol., 241: 439-447. Pierantozzi, M, Mazzone, P, Bassi, A, Rossini, PM, Peppe, A, Altibrandi, MG, Stefani, A, Bernardi, G and Stanzione, P (1999) The effect of deep brain stimulation on the frontal N30 component of somatosensory evoked potentials in advanced Parkinson's disease patients (see comments). Clin. Neurophysiol., 110: 1700-1707.

434 Pierantozzi, M, Palmieri, MG, Marciani, MG, Bernardi, G, Giacomini, P and Stanzione, P (2001) Effect of apomorphine on cortical inhibition in Parkinson's disease patients: a transcranial magnetic stimulation study. Exp. Brain Res., 141: 52-62. Praamstra, P and Plat, FM (2001) Failed suppression of direct visuomotor activation in Parkinson's disease. J. Cogn. Neurosci., 13: 31-43. Praamstra, P, Cools, AR, Stegeman, DF and Horstink, MW (1996a) Movement-related potential measures of different modes of movement selection in Parkinson's disease. J. Neurol. Sci., 140: 67-74. Praamstra, P, Meyer, AS, Cools, AR, Horstink, MW and Stegeman, DF (1996b) Movement preparation in Parkinson's disease. Time course and distribution of movement-related potentials in a movement precueing task. Brain, 119: 1689-1704. Praamstra, P, Stegeman, DF, Cools, AR and Horstink, MW (1998) Reliance on external cues for movement initiation in Parkinson's disease. Evidence from movement-related potentials. Brain, 121: 167-177. Priori, A, Berardelli, A, Inghilleri, M, Accornero, N and Manfredi, M (1994) Motor cortical inhibition and the dopaminergic system. Pharmacological changes in the silent period after transcranial brain stimulation in normal subjects, patients with Parkinson's disease and drug-induced parkinsonism. Brain, 117: 317-323. Pullman, SI, Watts, RL, Juncos, JL, Chase, TN and Sanes, IN (1988) Dopaminergic effects on simple and choice reaction time performance in Parkinson's disease. Neurology, 38: 249-254. Ridding, MC and Rothwell, JC (1999) Afferent input and cortical organisation: a study with magnetic stimulation. Exp. Brain Res., 126: 536-544. Ridding, MC, Inzelberg, R and Rothwell, JC (1995) Changes in excitability of motor cortical circuitry in patients with Parkinson's disease. Ann. Neurol., 37: 181-188. Rocchi, L, Chiari, L and Horak, FB (2002) Effects of deep brain stimulation and levodopa on postural sway in Parkinson's disease. J. Neurol. Neurosurg. Psychiatry, 73: 267-274. Rogers, MW, Kukulka, CG and Soderberg, GL (1987) Postural adjustments preceding rapid ann movements in parkinsonian subjects. Neurosci. Lett., 75: 246-251. Rothwell, JC, Obeso, JA, Traub, MM and Marsden, CD (1983) The behaviour of the long-latency stretch reflex in patients with Parkinson's disease. J. Neurol. Neurosurg. Psychiatry, 46: 35-44. Sanes, IN (1985) Information processing deficits in Parkinson's disease during movement. Neuropsychologia, 23: 381-392. Schieppati, M and Nardone, A (1991) Free and supported stance in Parkinson's disease. The effect of posture and

J.c. RafHWELL 'postural set' on leg muscle responses to perturbation, and its relation to the severity of the disease. Brain, 114: 1227-1244. Schneider, JS, Diamond, SG and Markham, CH (1986) Deficits in orofacial sensorimotor function in Parkinson's disease. Ann. Neurol., 19: 275-282. Sheppard, D, Bradshaw, JL, Phillips, JG, Iansek, R, Cunnington, R, Georgiou, N and Bradshaw, JA (1996) Cueing of movement in Parkinson's disease. Neuropsychiat. Neuropsychol. Behav. Neurol., 9: 91-98. Sheridan, MR and Flowers, KA (1990) Movement variability and bradykinesia in Parkinson's disease. Brain, 113: 1149-1161. Siebner, HR (2000) Simultaneous repetitive transcranial magnetic stimulation does not speed fine movement in PD. Neurology, 54: 272. Starr, A, Caramia, MD, Zarola, F and Rossini, PM (1988) Enhancement of motor cortical excitability in humans by non-invasive electrical stimulation appears prior to voluntary movement. Electroencephalogr. Clin. Neurophysiol., 70: 26-32. Stelmach, GE and Worringham, CJ (1988) The preparation and production of isometric force in Parkinson's disease. Neuropsychology, 26: 93-103. Stelmach, GE, Teasdale, N, Philips, J and Worringham CJ (1989) Force production characteristics in Parkinson's disease. Exp. Brain Res., 76: 165-172. Teasdale, N, Phillips, J and Stelmach, GE (1990) Temporal movement control in patients with Parkinson's disease. J. Neurol. Neurosurg. Psychiatry, 53: 862-868. Touge, T, Werhahn, KJ, Rothwell, JC and Marsden, CD (1995) Movement-related cortical potentials preceding repetitive and random-choice hand movements in Parkinson's disease. Ann. Neurol., 37: 791-799. Traub, MM, Rothwell, JC and Marsden, CD (1980) Anticipatory postural reflexes in Parkinson's disease and other akinetic-rigid syndromes and in cerebellar ataxia. Brain, 103: 393-412. Traversa, R, Pierantozzi, M, Semprini, R, Loberti, M, Cicardi, MC, Bassi, A and Stanzione, P (1995) N30 wave amplitude of somatosensory evoked potentials from median nerve in Parkinson's disease: a phannacological study. J. Neural Transm., Suppl. 45: 177-185. Valls-Sole, J, Pascual-Leone, A, Brasi1-Neto, JP, Cammarota, A, McShane, L and Hallett, M (1994) Abnormal facilitation of the response to transcranial magnetic stimulation in patients with Parkinson's disease. Neurology, 44: 735-741. Vidailhet, M, Rothwell, JC, Thompson, PD, Lees, AJ and Marsden, CD (1992) The auditory startle response in the Steele-Richardson-Olszewski syndrome and Parkinson's disease. Brain, 115: 1181-1192.

DISEASES AND TREATMENTS: PARKINSON'S DISEASE

Wang, He, Lees, AJ and Brown, P (1999) Impairment of EEG desynchronisation before and during movement and its relation to bradykinesia in Parkinson's disease. J. Neurol. Neurosurg. Psychiatry, 66: 442-446.

435 Wenzelburger, R, Raethjen, J, Loffler, K, Stolze, H, Illert, M and Deuschl, G (2000) Kinetic tremor in a reach-tograsp movement in Parkinson's disease. Mov. Disord., 15: 1084-1094.

Movement Disorders Handbook of Clinical Neurophysiology, Vol. 1 M. Hallett (Ed.) © 2003 Elsevier B.Y. All rights reserved

437 CHAPTER 27

Parkinson-plus conditions Josep Valls-Sole" and Francese Valldeoriola Unitat d'EMG, Servei de Neurologia, Departament de Medicina, Hospital Clinic, lnstitut d'lnvestigacion Biomedica August Pi i Sunyer (IDIRAPS), Universitat de Barcelona, Villarroel170, Barcelona 08036, Spain

27.1. Introduction Diseases associated with abnormal movement control have been for a long time a subject of interest for neurologists and clinical neurophysiologists. Neurophysiological examination can provide useful cues for understanding the pathophysiological mechanisms of some symptoms and signs characteristic of specific conditions. Neurologists could use some of the neurophysiological observations for reassurance in their clinical diagnosis as well as for follow-up and objective evaluation of treatment. The paradigmatic condition presenting as an hypokinetic movement disorder is parkinsonism (Table 1). The concept of parkinsonism applies to the clinical observation of rigidity and bradykinesia. However, other features such as tremor, postural instability, gait disorders, and abnormalities in facial expression, are also considered part of the syndrome). The most frequent disorder manifesting with parkinsonism is idiopathic Parkinson's disease (IPD), which is dealt with elsewhere in this book. In this chapter, we review some of the most relevant neurophysiological findings in patients with the socalled Parkinson-plus syndromes, neurodegenerative diseases featuring parkinsonism together with other symptoms and signs. Parkinsonism might result from a dysfunction in the basal ganglia, leading to a hyperactivity in the internal part of the globus pallidus (GPi). In idiopathic Parkinson's disease (IPD), pallidal hyperactivity might be responsible for reduced activation of thalamocortical projections, as well as for abnormal control of brainstem circuits, when executing a

* Correspondence to: Dr. Josep Valls-Sole, Unitat d'EMG, Servei de Neurologia, Hospital Clinic, Villarroel 170, Barcelona 08036, Spain. E.mail address: [email protected] Tel.: 34-93-2275413; fax: 34-93-2275783.

voluntary movement (Hallett and Khoshbin, 1980; Berardelli et al., 1983; Delwaide et al., 2000). However, the situation might be slightly different in Parkinson-plus conditions, in which basal ganglia pathology is accompanied by neuronal loss and atrophy in many nuclei of the brainstem and cerebellum, and dysfunctions might be present in Table 1 Neurodegenerative diseases manifesting with parkinsonism - Idiopathic Parkinson's disease - Progressive supranuclear palsy - Multiple system atrophy - Cortico-basal ganglionic degeneration - Diffuse Lewy body disease - Parkinson-Dementia complex of Guam - Atypical parkinsonism of French West Indies - Hallervorden-Spatz syndrome - Machado-Joseph's disease - Huntington's disease (Westphal's variant) - Herniparkinson-herniatrophy syndrome - Primary pallidal atrophy - Psychogenic parkinsonism Known causes of parkinsonism - pharmacological - toxic - post-encephalitic - post-vaccination - traumatic - vascular lesions - space occupying lesions - metabolic disorders - Wilson's disease - basal ganglia calcification - parathyroid metabolism abnormalities - infections - Creutzfeldt-Jakob disease - Gerstmann-Straussler-Scheinker syndrome

438

various circuits. Regardless of whether the pathophysiological mechanisms underlying parkinsonian signs are similar or not in IPD and Parkinson-plus conditions, clinical neurophysiological manifestations of parkinsonism are common to both types of disorders. In addition, in Parkinson-plus conditions, clinical neurophysiological evaluation may show a few specific symptoms and signs, distinctive of each syndrome. In the following review, we will first describe neurophysiological features that are common to various parkinsonian syndromes, and then highlight the differential aspects between Parkinsonplus disorders.

27.2. Neurophysiological observations common to parkinsonian syndromes 27.2.1. Bradykinesia and rigidity

The term bradykinesia is used to describe slowness of movement, while the term akinesia refers to absence of movement or delay in onset of a voluntary movement (Hallett and Khoshbin, 1980; Hallett, 1990; Pascual-Leone et al., 1994). Clinical manifestations related to akinesia/bradykinesia in patients with parkinsonism include facial hypoactivity, decreased spontaneous blinking rate, hypomimia, micrographia, loss of synkinetic arm movements while walking, postural instability, difficulties in rising from a chair or in turning in bed, etc. Normal blink rate depends largely on the level of the subject's attention, and the activity being performed. However, counting the blink rate can still furnish useful information in normal subjects and patients with CNS disorders. The resting blinking rate, of 24 per minute in normal controls, was found to be reduced in most patients with parkinsonism (Karson et al., 1984), but significantly more so in patients with progressive supranuclear palsy (PSP), whose mean blinking frequency was 4 blinks per minute. Karson et al. (1984) suggested that blinking rate may be an expression of the level of dopamine activity. The most universal neurophysiological method used in the assessment of bradykinesia is the reaction time task. By recording the EMG activity of forearm flexor and extensor muscles, Hallett and Khoshbin (1980) found that IPD patients were unable to appropriately scale the size of the first agonist EMG burst to the requirements of a ballistic

1. VALLS-SOLE AND F. VALLDEORIOLA

movement, and proposed that such defect represented a physiological mechanism of bradykinesia. Since then, simple and complex reaction time task paradigms have been thoroughly used in the study of patients with IPD. However, only a few studies have included such tests in the neurophysiological examination of patients with Parkinson-plus conditions (Valldeoriola et al., 1998). The term rigidity is used to describe the increased muscle tone of patients with parkinsonism (Berardelli et al., 1983). Clinical manifestations of rigidity are the increased limb stiffness and resistance to passive movements. The stooped posture, with flexion of the neck and trunk may also be a manifestation of larger rigidity in flexor than in extensor muscles, although there could also be a component or weakness in the paraspinal muscles (Djaldetti et al., 1999). Patients with parkinsonism have difficulties to completely relax their muscles, which often show some degree of tonic background EMG activity. Several neurophysiological tests have been used to assess rigidity, but direct cliniconeurophysiological correlation has proven more difficult than with bradykinesia. In IPD, rigidity has been demonstrated neurophysiologically by observations such as an increased size of the F wave (Abbruzzese et al., 1985), increased size of long loop reflex responses to stretch (Lee and Tatton, 1975; Rothwell et al., 1983) or electrical stimuli (Deuschl and Lucking, 1990), reduced inhibition in cutaneo-muscular reflexes (Fuhr et al., 1992), decreased duration of the silent period induced by transcranial magnetic stimulation (Cantello et al., 1991; Valls-Sole et al., 1994), reduced reciprocal inhibition (Bathien and Rondot, 1977; Lelli et al., 1991), and reduced autogenic inhibition of the soleus H-reflex (Delwaide et al., 1991). None of these techniques have been applied to patients with Parkinson-plus conditions, with the single exception of the examination of autogenic inhibition in patients with PSP, which interestingly showed the opposite result to patients with IPD, i.e. an enhancement of Ib inhibition (Fine et al., 1998). . Clinical manifestations of parkinsonism are rather conspicuous in patients with Parkinson-plus syndromes. Patients with PSP exhibit marked facial hypoactivity which usually combines with some dystonic features and ocular palsy to give the appearance of astonishment, typical of these patients. Their most important functional disabilities

439

PARKINSON-PLUS CONDITIONS

Table 2 Reaction time in control subjects and in patients with parkinsonism. EMG

Movement

219 (81)

256 (86)

IPD

342 (172)

400 (197)

864 (316)

PSP

443 (170)

521 (178)

1120 (443)

MSA

262 (45)

302 (48)

728 (482)

CS

Task 589 (124)

Figures are the mean, and one standard deviation in parenthesis, of the reaction time (in ms from the imperative signal). CS = control subjects; IPD=Idiopathic Parkinson's disease; PSP=Progressive supranuclear palsy, and MSA=multiple system atrophy.

are severe postural instability, loss of equilibrium and frequent falls. In our experience (Table 2), reaction time is significantly slower in PSP than in other parkinsonisms (Valldeoriola et al., 1998). In multiple system atrophy (MSA), parkinsonism might not be present at the moment of the diagnosis (i.e. in patients with predominating cerebellar dysfunction), but it develops in most patients, reaching 91% of cases in the advanced stages of the disease (Wenning et al., 1994). In Wenning's et al, series, bradykinesia and rigidity were asymmetric in 74% of patients. When compared to patients with IPD, patients with MSA became disabled at a significantly faster rate, with more than 50% of patients being markedly disabled or wheelchair bound within 5 years of onset of parkinsonism. Reaction time in patients with the rigido-akinetic type of MSA is not markedly delayed (Valldeoriola et al., 1998), and their parkinsonian features do not differ from those of patients with IPD. Response to L-DOPA therapy has been reported in up to 40% of patients with MSA (Fearnley and Lees, 1990). Therefore, differences with other parkinsonian syndromes have to be found in other clinical features such as autonomic disturbances or sphincter denervation (Quinn, 1989). In cortico-basal ganglionic degeneration (CBGD), parkinsonism may not be the first manifestation although it is present, even though sometimes at subclinical level, in 100% of patients (Rinne et al., 1994). It is usually asymmetric, and is accompanied by other more conspicuous clinical signs, such as the alien hand, myoclonus or dystonia, which are

the ones bringing the patient to the neurologist. Secondary parkinsonism (to neuroleptics or other medications, intoxications, infections, trauma, metabolic disorders or vascular lesions) is not essentially different from the parkinsonism seen in IPD. Its diagnosis is made through the analysis of the patient's history, observation of accompanying disorders, or serological determinations.

27.2.2. Disorders of brainstem reflexes and functions Short latency brainstern reflex responses (jawjerk, Rl of the blink reflex, and SPI of the masseter inhibitory reflex) are unaltered in patients with parkinsonism. This observation indicates that the afferent and efferent fibers of the reflex arc and the brain stem mono- or oligosynaptic circuits are not abnormal in these diseases. In contrast, long latency reflexes, which follow polysynaptic pathways subject to a strong suprasegmental influence, often show abnormal excitability. Kimura (1973) was first to show such abnormal excitability in patients with IPD, using the most paradigmatic brainstem reflex, the blink reflex. The basal ganglia modulate the excitability of the blink reflex through the output signals arising from the GPi and the substantia nigra pars reticulata (SNr). According to Basso et al. (1996) and Basso and Evinger (1996), the GPi/SNr complex sends inhibitory inputs to the superior colliculus (sq, which sends excitatory inputs to the nucleus raphe magnus (nRM) which, in tum, inhibits the trigeminal neurons of the spinal nucleus. In PO, there is increased GPi/SNr inhibition of the SC which, as a consequence, reduces its excitatory inputs to the nRM. The less active nRM induces less inhibition of the spinal trigeminal nucleus, which become dis-inhibited (hyperexcitable). As in bradykinesia, it is difficult to know whether the same mechanisms apply to Parkinson-plus conditions. We found similar interneuronal brainstem excitability enhancement in PSP and MSA patients, but not in patients with CBGD (Valls-Sole et al., 1997a). Figure 1 shows the graphical results obtained in examining the excitability recovery curve of patients with different parkinsonisms. Another brainstem reflex is the involuntary motor reaction to a startling stimulus. This is an interesting method to test the function of the reticulo-spinal system (Brown et al., 1991). The startle reaction has

440

AND F. VALLDEORIOLA

1. VALLS-SOLl~

100 80 ~

cv

> o 0

60

~

o Controls

'#. 40

M

IPD

• MSA

20

a +--~-=Y-----r--"---"",--",,,,-""""---r--T"" a 100 200 300

*

*

• PSP CBGD

A

500

800

Inter-stimulus interval (ms)

Fig. 1. Blink reflex excitability recovery curves in control subjects and in patients with parkinsonism. The asterisks mark the intervals with statistically significant differences between control subjects and patients with Parkinson's disease, multiple system atrophy and progressive supranuclear palsy. Patients with cortico-basal ganglionic degeneration were not different from control subjects.

been found moderately delayed in IPD (Vidailhet et aI., 1992), severely reduced or absent in PSP (Vidailhet et aI., 1992), normal or enhanced in MSA (Valldeoriola et al., 1997; Kofler et aI., 2001), and reduced and delayed in patients with Lewy-body disease (Kofler et aI., 2001). It has not been studied in patients with CBGD. 27.3. Neurophysiological observations of interest for specific Parkinson-plus conditions 27.3.1. Progressive supranuclear palsy

This is a degenerative disorder which features parkinsonism with a severe equilibrium disturbance and frequent falls, pseudobulbar palsy, dystonia, and oculomotor gaze palsy predominating in the vertical direction (Steele et aI., 1964). Clinical criteria for probable PSP include: a progressive disorder, beginning at age 40 or later, featuring vertical oculomotor palsy and instability with falls during the first year, with no evidence of other diseases that can explain the syndrome (Litvan et al., 1996). 27.3.1.1. Eye and eyelid movement disorders Some of the most striking features differentiating PSP from Parkinson's disease regard facial expression and gaze disturbances (Golbe et aI., 1988).

Spontaneous blinking rate is significantly more reduced in patients with PSP than in other patients with parkinsonism (Karson et al., 1984; Golbe et al., 1989). Clinical evidence of eye movement difficulties are not always present when other clinical features lead to the suspicion of PSP (Golbe et aI., 1989). At this time, recording of eye movements by electrooculography might be of some help (Chu et aI., 1979; Vidailhet et aI., 1994). Surface electrodes are placed in the upper, lower, nasal and temporal edges of the orbit and the subject is requested to make horizontal and vertical eye movements. Electrooculogram recordings may show characteristic abnormalities in patients with PSP, such as slowness of vertical eye movements, absent Bell's phenomenon, square wave jerks and microsaccades (Fig. 2). In the study reported by Vidailhet et aI., 9 out of 10 patients with PSP had vertical gaze paralysis with preserved reflex eye movements. Vidailhet et al. (1984) also showed slowness of horizontal eye movement and microsaccades that can be helpful in distinguishing patients with PSP from those with other parkinsonisms. Facial reflex responses to electrical stimulation of the median nerve have a distinct pattern of abnormalities in patients with PSP (Valls-Sole et aI., 1997a). Mentalis muscle responses can be elicited

441

PARKINSON-PLUS CONDITIONS A

----------~--------

L B

Fig. 2. Electrooculogram of horizontal eye movements in a healthy control (A) and in a patient with progressive supranuclear palsy (B). The upper graphs were recorded after asking the subjects to keep their gaze fixed to the center of a blank screen, located at a distance of 70 em in front of their eyes. The lower graphs show the recordings done when subjects were asked to move their eyes to the right and left sides of the screen. Note the square-wave jerks and the microsaccades in the patient's recordings. Horizontal calibration bar is 500 ms. Vertical calibration bar is 5° for the first and third traces and 15° for the 2nd and 4th traces.

with somatosensory stimuli applied to the cutaneous territory of the median nerve, as in the so-called palmomental reflex (Dehen et al., 1975). The same stimulus can generate other facial reflex responses, such as those appearing in the orbicularis oculi (Miwa et al., 1995). As part of a study of brainstern reflexes in patients with parkinsonism, we applied electrical stimuli to the median nerve and recorded facial responses simultaneously in the mentalis and orbicularis oculi muscles. The study included patients with IPD, PSP, MSA, CBGD, and healthy volunteers. Responses in the mentalis muscle were found in most patients and in 2 out of 10 normal subjects. In all of them, whenever there were responses in the mentalis muscle, there were also responses in the orbicularis oculi muscle. The exception were the patients with PSP who had no

orbicularis oculi responses even if the responses of the mentalis muscle were not different from those observed in the other groups of patients (Valls-Sole et al., 1997a). This abnormality adds to the other eye and eyelid motility disorders of patients with PSP (Golbe et al., 1989). The possibility to find a dissociation between mentalis and orbicularis oculi responses suggests that these two responses are mediated through two different circuits. The mentalis response could be conveyed through the cortico-nuclear tract, since this tract innervates predominantly lower facial motoneurons (Jenny and Saper, 1987) and a transcortical loop has been suggested because of the contiguity between thumb and chin areas in the brain sensorimotor region (Dehen et al., 1975). The orbicularis oculi response can be the result of activating the brainstem reticular formation as a kind of somatosensory startle (Gokin and Karpukhina, 1985). The selective damage of the pontine reticular formation in patients with PSP would then be responsible for the absence of the orbicularis oculi response. Enhancement of mentalis response may occur because of disinhibition of thalamo-cortical connections from their striatal control (Maertens de Noordhout and Delwaide, 1988). Table 3 summarizes the abnormalities of eye and eyelid movements reported on patients with PSP. 27.3.1.2. The startle reaction and reaction time task experiments Patients with PSP have absent or severely reduced auditory startle reactions (Vidailhet et al., 1992). From studies carried out on rats, the startle reaction is known to be generated in the nucleus reticularis pontis caudalis (nRPC), which activates the reticulospinal tract inducing muscle responses in facial and spinal motoneurons (Davis et al., 1982). In humans, the startle reaction is also thought to originate in corresponding nuclei of the brainstem, and spread caudally and rostrally to limb and facial muscles. Neuronal loss in patients with PSP involves specifically the cholinergic neurons of the lower pontine reticular formation, including those of the pedunculo-pontine tegmental nucleus and the nucleus reticularis pontis caudalis (Zweig et al., 1987; Juncos et al., 1991; Malessa et al., 1991). Therefore, it is not surprising that patients with PSP have abnormalities in the startle reaction to auditory stimuli. In the study carried out by Vidailhet et al.

442

1. VALLS-SOLl~

AND F. VALLDEORlOLA

Table 3 Eye and eyelid movement abnormalities observed in patients with PSP. Observation

Reference

Reduced spontaneous blinking

Karson (1984)

Lid retraction

Maher and Lees (1986)

Blepharospasm

Jackson et al. (1983)

Supranuclear palsy of eyelid opening

Lepore and Duvoisin (1985)

Supranuclear palsy of eyelid closing

Golbe et al. (1989)

Reduced voluntary suppression ofVOR

Golbe et al. (1989)

Eyelid retraction (Cowper's sign)

Golbe et al. (1989)

Square-wave jerks

Chu et al. (1979)

Absent eyelid responses to acoustic stimuli

Vidailhet et al. (1992)

Absent eyelid responses to median nerve stimuli

Valls-Sole et al. (1997)

(1992), the response was absent in 3 out of 8 patients, and it was small and delayed in the other 5 patients. Patients with PSP have a significantly longer simple reaction time than other patients with parkinsonism, and their reaction time does not speed up when an additional startling stimulus is given together with the 'go' signal (Valldeoriola et al., 1998). Delay in simple reaction time has been equated to akinesia, and attributed to the fact that it takes longer than normal to sufficiently energize the structures needed for the execution of the motor task, including the primary motor cortex (Hallett, 1990). Another explanation comes from the implication of subcortical motor structures in the preparation and execution of ballistic movements (Valls-Sole et aI., 1999). Reaching a certain degree of motor preparation is necessary for the execution of ballistic movements. Motor preparation might be mediated by subcortical motor structures such as the reticulospinal tract, which is probably an important part of the execution channel for a ballistic movement. Congruent with this hypothesis, external activation of the reticulospinal tract by a startling auditory stimulus causes significant acceleration of reaction time in normal human subjects (Valls-Sole et al., 1995, 1999). The significantly longer delay in the reaction time of patients with PSP with respect to other parkinsonisms could be due to faulty preparation of the execution channel as a consequence of their abnormalities in the reticular formation.

An interesting collateral observation of the startle induced shortening of the reaction time (StartReact effect) was its reduced habituation in successive trials, which can be an advantage for clinical applicability in comparison to testing the startle reaction alone. Valldeoriola et al. (1998) applied this method to the investigation of patients with parkinsonism. These authors found a significant StartReact effect in all patients except in those with PSP (Fig. 3), and suggested that this method might have clinical applicability in early differential diagnosis between IPD and PSP. More recently, Molinuevo et al. (2000) have shown that the same results (i.e. absence of reaction time shortening in PSP patients but significant shortening in PD patients) can be obtained using transcranial magnetic stimulation (TMS). Shortening of simple reaction time with subthreshold TMS has been already reported in normal subjects and IPD patients (Pascual-Leone et al., 1992, 1994).Apart from confirming the abnormalities of PSP patients in the execution of ballistic movements, the observations reported by Molinuevo et al. (2000) give some cues on the mechanisms by which TMS induces a reaction time shortening effect: Since PSP patients do not have significant motor cortical disturbances, the effect would have been present if it were dependent on the activation of the corticospinal tract. The absence of the effect points to the possibility that TMS shortens reaction time by way of activating the reticulospinal structures through various sources: The sound stim-

443

PARKINSON-PLUS CONDITIONS

A

B

VL

Movement of wrist

----;,----~/'

Task 'go' signal

lOOms

'go' signal

-+-~~~ --t---~-----/~~~ 'go' signal 'go' signal + startle

+

startle

Fig. 3. Effects of a startle on reaction time in a control subject (A), and in a representative patient with progressive supranuclear palsy (B).

ulus inevitably delivered together with the discharge of the coil, the trigeminal inputs caused by activation of scalp and jaw muscles, and the direct activation of cortico-reticular pathways (Molinuevo et aI., 2000). 27.3.2. Multiple system atrophy

Multiple system atrophy is a progressive neurodegenerative disease featuring autonomic disorders together with parkinsonism, cerebellar or pyramidal dysfunctions (Quinn, 1989). Clinical diagnosis of probable MSA requires the presence of autonomic failure (including urinary dysfunction), and either a poor levodopa responsive parkinsonism or cerebellar dysfunction (Gilman et aI., 1999). 27.3.2.1. Autonomic dysfunction Autonomic nervous system dysfunction is the key to the diagnosis in patients with MSA who present with parkinsonism or cerebellar syndromes (Quinn,

1989), and is presently a required criterion for the diagnosis of probable MSA (Gilman et aI., 1999). Autonomic nuclei of the brainstem, the intermediolateral column, and the sacral parasympathetic neurons are some of the structures related to the autonomic nervous system in which pathological examination has revealed loss of neurons and glial cytoplasmic inclusions in patients with MSA. Clinically relevant autonomic dysfunctions in these patients are orthostatic hypotension, urinary and faecal incontinence, erectile dysfunction in males, sudomotor disregulation, and abnormalities in respiratory control during sleep. Orthostatic hypotension may be due to the inability to increase sympathetic activity with standing. This can be shown as an abnormal regulation of baroreflex responses to different stimuli (Benarroch et al., 1993). Using readily available electrophysiological equipment, it is also possible to monitor heart beat frequency. Recording the R-R interval variation by means of the signal trigger and the delay line

444

unit of an electromyograph shows graphically the reduced adaptation of the heart beat rate to a postural change or to the Valsalva maneuver (Valls-Sole, 2000). The main drawback of this test is that patients with severe bradykinesia might be unable to perform adequately the maneuvers. Reduced R-R interval variation could then be due to insufficient stimulation. For this reason, we have examined the possibilities to test R-R interval variation using methods that do not require the patient's cooperation. One of these methods is based on the observation of normal startle responses in patients with MSA (Valldeoriola et aI., 1997), and on the fact that a startle accelerates heart-beat frequency in normal subjects (Roland et aI., 1999). We have examined the behavior of such an effect in normal subjects and patients with MSA (Valls-Sole et aI., 2002). In spite of having a normal motor component of the startle reaction in chest muscles, patients with MSA had significantly smaller change of the R-R interval after the acoustic stimulus (Fig. 4). The sympathetic sudomotor skin response, or SSR (Shahani et aI., 1984), may reveal dysfunctions in the autonomic control of sudomotor reflexes. Loss of sympathetic neurons of the intermediolateral column might explain the finding of frequently abnormal SSRs in patients with MSA (Bordet et aI., 1996). Other tests of sudomotor function, such as the evaluation of the amount of sweat production to direct gland stimulation with intradermal methacholine, have also demonstrated a decreased sweat response in patients with MSA (Baser et aI., 1991). Sleep disorders are frequent in patients with MSA. Some of these disorders might be related to autonomic dysfunction. Thirty-five of 39 patients with MSA had REM sleep behavior disorders (Plazzi et aI., 1993), which preceded the diagnosis in 44% of the cases. Polysomnographic studies revealed subclinical obstructive sleep apnea in six patients, laryngeal stridor in eight patients, and periodic leg movements during sleep in ten patients. Laryngeal stridor, due to vocal cord abductor paralysis during sleep, is probably caused by selective denervation atrophy of the cricoarytenoid muscle due to selective loss of neurons in the nucleus ambiguous (Bannister et aI., 1981), and may lead to choking and death in advanced stages of MSA. This can be prevented with tracheostomy or with continuous positive air pressure (lsozaki et al., 1996; Iranzo et aI., 2000).

J. VALLS-SOLJ~

AND F. VALLDEORIOLA

A

, I

B

Auditory stimulus

Fig. 4. Modulation of heart rate by an auditory startle in a control subject (A) and in a patient with multiple system atrophy (B). Note the EMG activity of the chest muscles just after the stimulus artifact, and the reduced effect on heart rate in the patient.

27.3.2.2. Sphincter EMG In MSA patients, manifestations of autonomic dysfunctions such as erectile impotence are usually

445

PARKINSON-PLUS CONDITIONS

accompanied by urinary frequency and urgency, leading soon to incontinence, associated with large residual urine volumes (Kirby et aI., 1986). The severity of urinary symptoms is one main red flag that should warn the neurologist of the possibility that the parkinsonian patient thought to have IPD is actually facing the diagnosis of probable MSA (Quinn, 1989). Urinary incontinence in MSA patients might be due to autonomic dysfunction, loss of pontine control of micturition, striatal sphincter denervation, or a combination of all of them. Striatal sphincter denervation is attributed to the selective loss of motoneurons in the nucleus of Onuff at the S2-S3 medullary segments. Needle electromyography of the external anal sphincter, therefore, is considered an important neurophysiological test in the assessment of patients with parkinsonism, as most patients with MSA show denervation-reinnervation signs (Sakuta et aI., 1978; Eardley et aI., 1989). We and others have confirmed that anal sphincter denervation is prominent in patients with MSA, although similar type of abnormalities have been found in a large proportion of patients with PSP as well as in some patients with Parkinson's disease (Valldeoriola et aI., 1995; Giladi et aI., 2000). Therefore, the utility of anal or vesical sphincter needle EMG in the diagnosis of MSA is still under debate (Rodi et aI., 1996; Libelius and Johansson, 2000). Chronic constipation, local trauma related to delivery, and other pudendal nerve longstanding lesions may give rise also to sphincter denervation (Kiff and Swash, 1984; Podnar and Vodusek, 2000), which may diminish the validity of the sign as a true marker of motoneuronal loss. Recently, a consensus statement has been published in which sphincter EMG abnormalities are considered as a supportive laboratory finding, but not a criterion, for the diagnosis of MSA (Gilman et aI., 1999). 27.3.2.3. Minipolymyoclonus Tremor has been reported in up to 74% of MSA patients (Wenning et aI., 1994). However, this figure included several types of tremor, with only a few patients exhibiting the resting tremor typical of IPD, and a large proportion of unclassifiable hands and finger 'jerky' tremors. Electrophysiological studies of these latter movements have shown that their characteristics are closer to myoclonus than to tremor (Salazar et aI., 2000). A piezoelectric accelerometer was used to record finger movements and

analyze the frequency spectrum of the signal through fast Fourier transformation. This procedure showed that movements of MSA patients were rather nonrhythmic in comparison to those of patients with other forms of tremor. Salazar et al. (2000) suggested the term minipolymyoclonus to be used to describe these small amplitude, irregular, jerk-like abnormal movements. Other forms of myoclonus have also been reported in a few MSA patients (Chen et aI., 1992; Gouider-Khouja et al., 1995), which might have their origin in a reduced inhibition of the strio-palido-thalamo-cortical circuit (Patel and Slater, 1987). 27.3.3. Cortico-basal ganglionic degeneration

Patients with CBGD characteristically present with a progressive neurodegenerative disorder, featuring asymmetric ideomotor apraxia, cortical sensory deficit or alien limb, together with a rigidakinetic syndrome and either dystonic postures or spontaneous, action or reflex myoclonus (Gibb et aI., 1989; Riley et aI., 1990; Lang, 1996). 27.3.3.1. Myoclonus Myoclonus in CBGD is thought to be of cortical origin despite of the fact that neurophysiological evidence is lacking. The expected findings of cortical myoclonus, such as giant somatosensory evoked potentials and jerk-locked EEG potentials, are typically absent in CBGD. This is attributed to the marked frontoparietal cortical atrophy and neuronal degeneration characteristic of these patients (Gibb et aI., 1989; Brunt et aI., 1995). Cortical atrophy could involve predominantly inhibitory neurons, leading to an enhanced (disinhibited) motor cortex excitability. The 'C' wave, or focal reflex myoclonus (Sutton and Mayer, 1974), is a response seen in forearm muscles after electrical stimulation of ipsilateral cutaneous nerves of the hand. This response is thought to be mediated by fast conducting afferent and efferent pathways and might have a latency as short as 43.1±3.2 ms (Thompson et al., 1994). In some patients, focal reflex myoclonus might be elicited by stimuli of an intensity below perception threshold, which suggests a direct connection from the thalamic nuclei to the motor cortex (Mauguiere et aI., 1983). The 'C' response should not be mistaken for the long latency excitatory response of the cutaneo-muscular reflexes (Caccia et aI., 1973;

446

1. VALLS-SOLl~

AND F. VALLDEORIOLA

Fig. 5. Maps of cortical representation of hand muscles in a healthy volunteer (A) and in a patient with cortico-basal ganglionic degeneration (B).

Jener and Stephens, 1982). The cutaneo-muscular reflex can be elicited during a sustained tonic voluntary contraction of the forearm muscles. The long latency excitatory component of the cutaneomuscular reflex might be abnormally enhanced in patients with IPD or MSA (Chen et al., 1992). However, the latency of such a response is longer than that of the 'C' reflex. 27.3.3.2. Alien hand sign The lack of voluntary control of limb movements, or 'alien-hand' syndrome, that is often seen in

patients with CBGD, suggests involvement of cortical motor pathways, with dysfunctions localized either at the frontal lobe (Goldberg et al., 1981) or at the corpus callosum (Feinberg et al., 1992). Accordingly, the study of corticospinal tract functions could show abnormalities in patients with CBGD exhibiting the alien hand phenomenon. A focal, figure-of-8, coil was used to activate limited regions of the motor cortex of only one hemisphere. In this way, a motor map of the area of representation of a given muscle can be constructed by applying the coil to a grid of points in the scalp separated by 1 em. Six out of 10

PARKINSON-PLUS CONDITIONS

CBGD patients with alien hand sign had bilateral responses to focal, unilateral, TMS applied to the side contralateral to the alien hand (Fig. 5). Ipsilateral responses were delayed with respect to the contralateral ones by a mean of 7.7±2.2 ms, a time allowing for conduction through the corpus callosum. Such abnormality was not found in any of 10 normal subjects, eight patients with Alzheimer's disease, or six patients with IPD presenting with predominantly unilateral rigidity. This finding points again to an enhanced motor cortex excitability in the hemisphere contralateral to the alien hand, which may be unable to inhibit transcallosal excitatory inputs from the other hemisphere.

References Abbruzzese, G, Vische, M, Ratto, S, Abbruzzese, M and Favale, E (1985) Assessment of motor neuron excitability in parkinsonian rigidity by the F wave. J. Neurol., 232: 246-249. Alexander, GE and Crutcher, MD (1990) Functional architecture of basal ganglia circuits: neural substrates of parallel processing. Trends Neurosci., 13: 266-271. Albin, RL, Young, AB and Penney, JB (1989) The functional anatomy of the basal ganglia disorders. Trends Neurosci., 12: 366-374. Bannister, R, Gibson, W, Michaels, L and Oppenheimer, DR (1981) Laryngeal abductor paralysis in multiple system atrophy. A report on three necropsied cases, with observation on the laryngeal muscles and the nuclei ambigui. Brain, 104: 351-368. Baser, SM, Meer, J, Polinsky, RJ and Hallett, M (1991) Sudomotor function in autonomic failure. Neurology, 41: 1564-1566. Basso, MA and Evinger, C (1996) An explanation for reflex blink hyperexcitability in Parkinson's disease. II. Nucleus raphe magnus. J. Neurosci., 16: 7318-7330. Basso, MA, Powers, AS and Evinger, C (1996) An explanation for reflex blink hyperexcitability in Parkinson's disease. I. Superior colliculus. J. Neurosci., 16: 7308-7317. Bathien, Nand Rondot, P (1977) Reciprocal continuous inhibition in rigidity of parkinsonism. J. Neurol. Neurosurg. Psychiatry, 40: 20-24. Benarroch, EE and Chang, FLF (1993) Central autonomic disorders. J. Clin. Neurophysiol., 10: 39-50. Berardelli, A, Sabra, AF and Hallett, M (1983) Physiological mechanisms of rigidity in Parkinson's disease. J. Neurol. Neurosurg. Psychiatry, 46: 45-83 Brown, P, Rothwell, JC, Thompson, PD, Day, BL and Marsden, CD (1991) New observations on the normal auditory startle reflex in man. Brain, 114: 1891-1902.

447 Brunt, ERP, VanWeerden, TW, Pruim, J and Lakke, lWPF (1995) Unique myoclonic pattern in corticobasal degeneration. Mov. Disord., 10: 132-142. Caccia, MR, McComas, AJ, Upton, ARM and Blogg, T (1973) Cutaneous reflexes in small muscles of the hand. J. Neurol. Neursurg. Psychiatry, 36: 960-977 Cantello, R, Gianelli, M, Bettucci, D, Civardi, C, De Angelis, MS and Mutani, R (1991) Parkinson's disease rigidity: Magnetic MEPs in a small hand muscle. Neurology, 41: 1449-1456. Chen, R, Ashby, P and Lang, AE (1992) Stimulussensitive myoclonus in akinetic rigid syndromes. Brain, 115: 1875-1888. Chu, FC, Reingold, DB, Cogan, DG and Williams, AC (1979) The eye movement disorders of progressive supranuclear palsy. Ophthalmology, 86: 422-'1-28. Davis, M, Gendelman, DS, Tischler, MD and Gendelman, PM (1982) A primary acoustic startle circuit: lesion and stimulation studies. J. Neurosci., 2: 791-805. Dehen, H, Bathien, N and Cambier, J (1975) The palomomental reflex. An electrophysiological study. Eur. Neurol., 13: 395-'1-04. DeLong, MR (1990) Primate model of movement disorders of basal ganglia origin. Trends Neurosci., 13: 281-285. Delwaide, P, Pepin, JL and Maertens de Noordhout, A (1991) Short latency autogenic inhibition in patients with parkinsonian rigidity. Ann. Neurol., 30: 83-89. Delwaide, P, Pepin, JL, DePasqua, V and Maertens de Noordhout, A (2000) Projections from the basal ganglia to tegmentum: a subcortical route for explaining the pathophysiology of Parkinson's disease signs? J. Neurol., 247 (Supp!. 2): 75-81. Deuschl, G and LUcking, CH (1990) Physiology and clinical applications of hand muscle reflexes. Electroencephalogr. Clin. Neurophysiol., Suppl41: 84-101. Djaldetti, R, Mosberg-Galili, R, Sroka, H, Merims, D and Melamed, E (1999) Camptocormia (bent spine) in patients with Parkinson's disease. Characterization and possible pathogenesis of an unusual phenomenon. Mov. Disord., 14: 443-'1-47 Eardley, I, Quinn, NP, Fowler, CJ, Kirby, RS, Parkhouse, HF, Marsden, CD and Bannister, R (1989) The value of urethral sphincter electromyography in the differential diagnosis of parkinsonism. Br. J. Urol., 64: 360-362. Fearnley, JM and Lees, AJ (1990) Striatonigral degeneration: a clinico-pathologic study. Brain, 113: 18231832. Feinberg, TE, Schindler, RJ, Flanagan, NG and Haber, LD (1992) Two alien hand syndromes. Neurology, 42: 19-24. Fine, EJ, Hallett, M, Litvan, I, Tresser, N and Katz, D (1998) Dysfunction of Ib (autogenic) spinal inhibition

448 in patients with progressive supranuclear palsy. Mov. Disord., 13: 668-672. Fuhr, P, Zeffiro, T and Hallett, M (1992) Cutaneous reflexes in Parkinson's disease. Muse. Nerve, 15: 733-739. Gibb, WRG, Luthert, PJ and Marsden, CD (1989) Corticobasal degeneration. Brain, 112: 1171-1192. Giladi, N, Simon, ES, Korczyn, AD, Groozman, GB, Orlov, Y, Shabtai, H and Drory, VE (2000) Anal sphincter EMG does not distinguish between multiple system atrophy and Parkinson's disease. Muse. Nerve, 23: 731-734. Gilman, S, Low, PA, Quinn, N, Albanese, A, Ben-Shlomo, Y, Fowler, CJ, Kaufmann, H, Klockgether, T, Lang, AE, Lantos, PL, Litvan, I, Mathias, CJ, Oliver, E, Robertson, D, Schatz, I and Wenning, GK (1999) Consensus statement on the diagnosis of multiple system atrophy. J. Neurol. Sci., 163: 94-98. Golbe, LI, Davis, PH, Schoenberg, BS and Duvoisin, RC (1988) Prevalence and natural history of progressive supranuclear palsy. Neurology, 38: 1031-1034. Golbe, LI, Davis, PH and Lepore, FE (1989) Eyelid movement abnormalities in progressive supranuclear palsy. Mov. Disord., 4: 297-302. Goldberg, G, Mayer, NH and Toglia, JU (1981) Medial frontal cortex infarction and the alien hand sign. Arch. Neurol., 38: 683-686. Gokin, AP and Karpukhina, MV (1985) Reticular structures in the cat brain involved in startle to somatic stimuli of various modalities. Neurophysiology, 17: 278-286. Gouider-Khouja, N, Vidailhet, M, Bonnet, AM, Pichon, J and Agid, Y (1995) 'Pure' striatonigral degeneration and Parkinson's disease: a comparative clinical study. Mov. Disord., 10: 288-294. Hallett, M (1990) Clinical neurophysiology of akinesia. Rev. Neurol., 146: 585-590. Hallett, M and Khoshbin, SA (1980) A physiological mechanism of bradykinesia. Brain, 103: 301-314. Holand, S, Girard, A, Laude, D, Meyer-Bisch, C and Elghozi, JL (1999) Effects of an auditory startle stimulus on blood pressure and heart rate in humans. J. Hypertens., 17: 1893-1897. Iranzo, A, Santamaria, J, Tolosa, E, Aguilar, F, Vilaseca, I, Morello, A, Valldeoriola, F, Marti, MJ, Valls-Sole, J, Munoz, E, Molinuevo, JL and Pou, A (2000) Continuous positive air pressure eliminates nocturnal stridor in multiple system atrophy. The Lancet, 356: 1329-1330. Isozaki, E, Naito, A, Horiguchi, S, Kawamura, R, Hayashida, T and Tanabe, H (1996) Early diagnosis and stage classification of vocal cord abductor paralysis in patients with multiple system atrophy. J. Neurol. Neurosurg. Psychiatry, 60: 399-402.

1. VALLS-SOLl~

AND F. VALLDEORIOLA

Jackson, JA, Jankovic, J and Ford, J (1983) Progressive supranuclear palsy: clinical features and response to treatment in 16 patients. Ann. Neurol., 13: 273-278 Jenny, AB and Saper, CB (1987) Organization of the facial nucleus and corticofacial projection in the monkey: a reconsideration of the upper motor neuron facial palsy. Neurology, 37: 930-939. Juncos, JL, Hirsch, EC, Malessa, S, Duyckaerts, C, Hersh, LB and Agid, Y (1991) Mesencephalic cholinergic nuclei in progressive supranuclear palsy. Neurology, 41: 25-30. Jenner, JR and Stephens, JA (1982) Cutaneous reflex responses and their central nervous pathways studied in man. J. Physiol., 333: 405-419. Karson, CN, Burns, S, LeWitt, P, Foster, NL and Newman, RP (1984) Blink rates and disorders of movement. Neurology, 34: 677-678. Kiff, ES and Swash, M (1984) Slowed conduction in the pudendal nerves in idiopathic (neurogenic) faecal incontinence. Br. J. Surg., 71: 614-616. Kimura, J (1973) Disorders of interneurons in Parkinsonism. The orbicularis oculi reflex to paired stimuli. Brain, 96: 87-96. Kirby, R, Fowler, CJ, Gosling, J and Bannister, R (1986) Urethro-vesical dysfunction in progressive autonomic failure with multiple system atrophy. J. Neurol. Neurosurg. Psychiatry, 49: 554-562. Kofler, M, MUller, J, Wenning, GK, Reggiani, L, Hollosi, J and Poewe, W P, Bosch, S, Ransmayr, G, Valls-Solt~, (2001) The auditory startle reaction in parkinsonian disorders. Mov. Disord., 16: 62-71. Lang, AE (1996) Cortico-basal ganglionic degeneration: an update. Annual Courses of the American Academy of Neurology (Vol. IV), 440: 39-53. Lee, RG and Tatton, WG (1975) Motor responses to sudden limb displacements in primates with specific CNS lesions and in human patients with motor system disorders. Can. J. Neurol. Sci., 2: 285-293. Lelli, S, Panizza, M and Hallett, M (1991) Spinal cord inhibitory mechanisms in Parkinson's disease. Neurology, 41: 553-556. Lepore, FE and Duvoisin, RC (1985) "Apraxia" of eyelid opening: an involuntary levator inhibition. Neurology, 35: 423-427. Libelius, R and Johansson, F (2000) Quantitative electromyography of the external anal sphincter in Parkinson's disease and multiple system atrophy. Muse. Nerve, 23: 1250-1256. Litvan, I, Agid, Y, Calne, D, Campbell, G, Dubois, B, Duvoisin, RC, Goetz, CG, Golbe, LI, Grafman, J, Growdon, JH, Hallett, M, Jankovic, J, Quinn, NP, Tolosa, ES and Zee, DS (1996) Clinical research criteria for the diagnosis of progressive supranuclear palsy (Steele-Richardson-Olszewski syndrome): report

PARKINSON-PLUS CONDITIONS

of the NINDS-SPSP international workshop. Neurology, 47: 1-9. Maertens de Noordhout, A and Delwaide, PJ (1988) The palmomental reflex in Parkinson's disease. Comparisons with normal subjects and clinical relevance. Arch. Neurol., 45: 425--427. Maher, ER and Lees, AJ (1986) The clinical features and natural history of the Steele-Richardson-Olszewski syndrome (progressive supranuclear palsy). Neurology, 36: 1005-1008 Malessa, S, Hirsch, EC, Cervera, P, Javoy-Agid, F, Duyckaerts, C, Hauw, 11 and Agid, Y (1991) Progressive supranuclear palsy: Loss of cholinergic acetyltransferase-like immunoreactive neurons in the pontine reticular formation. Neurology, 41: 15931597. Mauguiere, F, Desmedt, JE and Courjon, J (1983) Astereognosis and dissociated loss of frontal or parietal components of somatosensory evoked potentials in hemispheric lesions: detailed correlations with clinical signs and computerized tomographic scanning. Brain, 106: 271-311. Miwa, H, Imamura, N, Kogahara, K, Ohori, T and Mizuno, Y (1995) Somatosensory evoked blink response: findings in patients with Miller Fisher syndrome and in normal subjects. J. Neurol. Neurosurg. Psychiatry, 58: 95-99. Molinuevo, JL, Valls-Sole, J and Valldeoriola, F (2000) The effect of transcranial magnetic stimulation on reaction time in progressive supranuclear palsy. Clin. Neurophysiol., 111: 2008-2013. Parent, A and Hazrati, LN (1995) Functional anatomy of the basal ganglia. I. The cortico-basal ganglia-thalamocortical loop. Brain Res., 20: 91-127. Pascual-Leone, A, Valls-Sole, J, Wassermann, EM, BrasilNeto, JP, Cohen, LG and Hallett, M (1992) Effects of focal transcranial magnetic stimulation on simple reaction time to acoustic, visual, and somatosensory stimuli. Brain, 115: 1045-1059. Pascual-Leone, A, Valls-Sole, J, Brasil-Neto, JP, Cohen, LG and Hallett, M (1994) Akinesia in Parkinson's disease. I. Shortening of simple reaction time with focal, single pulse transcranial magnetic stimulation. Neurology, 44: 884-891. Patel, S and Slater, P (1987) Analysis of the brain regions involved in myoclonus produced by intracerebral picrotoxin. Neuroscience, 20: 687-693. Podnar, Sand Vodusek, DB (2000) Standardization of anal sphincter electromyography: effect of chronic constipation. Muse. Nerve, 23: 1748-1751. Quinn, N (1989) Multiple system atrophy - the nature of the beast. J. Neurol. Neursurg. Psychiatry, Suppl.: 7889.

449 Riley, DE, Lang, AE, Lewis, A, Resch, L, Ashby, P, Hornykiewicz, 0 and Black, S (1990) Cortical-basal ganglionic degeneration. Neurology, 40, 1203-1212. Rinne, JO, Lee, MS, Thompson, PD and Marsden, CD (1994) Cortico-basal degeneration. A clinical study of 36 cases. Brain, 117: 1183-1196. Rodi, Z, Denislic, M and Vodusek, DB (1996) External anal sphincter electromyography in the differential diagnosis of parkinsonism. J. Neurol. Neurosurg. Psychiatry, 60: 460--461. Rothwell, JC, Obeso, JA, Traub, MM and Marsden, CD (1983) The behavior of the long latency stretch reflex in patients with Parkinson's disease. J. Neurol. Neurosurg. Psychiatry, 46: 35--44. Sakuta, M, Nakanishi, T and Toyokura, Y (1978) Anal muscle electromyograms differ in amyotrophic lateral sclerosis and Shy-Drager syndrome. Neurology, 28: 1289-1293. Salazar, G, Valls-Sole, J, Marti, MJ, Chang, H and Tolosa, ES (2000) Postural and action myoclonus in patients with parkinsonian type multiple system atrophy. Mov. Disord., 15: 77-83. Shahani, BW, Halperin, 11, Boulu, P and Cohen, J (1984) Sympathetic skin response: a method of assessing unmyelinated axon dysfunction in peripheral neuropathies. J. Neurol. Neurosurg. Psychiatry, 47: 536-542. Steele, JC, Richardson, JC and Olszewski, J (1964) Progressive supranuclear palsy: a heterogenous degeneration involving the brainstem, basal ganglia and cerebellum, with vertical gaze and pseudobulbar palsy, nuchal dystonia and dementia. Arch. Neurol., 10: 333-359. Sutton, GG and Mayer, RF (1974) Focal reflex myoclonus. J. Neurol. Neurosurg. Psychiatry, 37: 207-217. Thompson, PD, Day, BL, Rothwell, JC, Brown, P, Britton, TC and Marsden, CD (1994) The myoclonus in corticobasal degeneration. Evidence for two forms of cortical reflex myoclonus. Brain, 117: 1197-1208. Valldeoriola, F, Valls-Sole, J, Tolosa, ES and Marti, MJ (1995) Striated anal sphincter denervation in patients with progressive supranuclear palsy. Mov. Disord., 10: 550-555. Valldeoriola, F, Valls-Sole, J, Tolosa, E, Nobbe, FA, Munoz, JE and Marti, MJ (1997) The acoustic startle response is normal in patients with multiple system atrophy. Mov. Disord., 12: 697-700. Valldeoriola, F, Valls-Sole, J, Tolosa, E, Ventura, PJ, Nobbe, FA and Marti, MJ (1998) The effects of a startling acoustic stimulus on reaction time in patients with different parkinsonian syndromes. Neurology, 51: 1315-1320.

450 J (2000) Neurophysiological characterization of parkinsonian syndromes. Neurophysiol. CUn., 30: 352-367. Valls-Sole, J, Pascual-Leone, A, Brasil-Neto, JP, McShane, L and Hallett, M (1994) Abnormal facilitation of the response to transcranial magnetic stimulation in patients with Parkinson's disease. Neurology, 44: 735-741. Valls-Sole, J, Sole, A, Valldeoriola, F, Munoz, E, Gonzalez, LE and Tolosa, ES (1995) Reaction time and acoustic startle in normal human subjects. Neurosci. Lett., 195: 97-100. Valls-Sole, J, Valldeoriola, F, Tolosa, E and Marti, MI (1997a) Distinctive abnormalities of facial reflexes in patients with progressive supranuclear palsy. Brain, 120: 1877-1883. Valls-Sole, J, Valldeoriola, F, Tolosa, E and Nobbe, F (1997b) Habituation of the auditory startle reaction is reduced during preparation for execution of a motor task in normal human subjects. Brain Res., 751: 155-159. Valls-Sole, J, Rothwell, JC, Goulart, F, Cossu, G and Munoz, IE (1999) Patterned ballistic movements Valls-Solt~,

J. VALLS-SOLI:;: AND F. VALLDEORIOLA

triggered by a startle in healthy humans. J. Physiol., 516: 931-938. Valls-Sole, J et al. (2002) Effects of a startle on heart rate in patients with multiple system atrophy. Mov. Disord., 17: 546-549. Veciana, M, Valls-Sole, J, Valldeoriola, F, Munoz, E and Tolosa, ES (2000) Startle effects on the R-R interval in normal controls and patients with multisystem atrophy. Mov. Disord., 15: 78 (abstract). Vidailhet, M, Rothwell, IC, Thompson, PO, Lees, AJ and Marsden, CD (1991) The auditory startle response in the Steele-Richardson-Olszewsky syndrome and Parkinson's disease. Brain, 115: 1181-1192. Vidailhet, M, Rivaud, S, Gouider-Khouja, N, Pillon, B, Bonnet, AM, Gaymard, B, Agid, Y and PierrotDeseilligny, C (1994) Eye movements in parkinsonian syndromes. Ann. Neurol., 35: 420-426. Wenning, GK, Ben Shlomo, Y, Magalhaes, M, Daniel, SE and Quinn, NP (1994) Clinical features and natural history of multiple system atrophy. An analysis of 100 cases. Brain, 117: 835-845. Zweig, RM, Whitehouse, PJ, Casanova, MF, Walker, LC, Jankel, WR and Price, DL (1987) Loss of pedunculopontine neurons in progressive supranuclear palsy. Ann. Neural., 22: 18-25.

Movement Disorders Handbook of Clinical Neurophysiology, Vol. I M. Hallett (Ed.) © 2003 Elsevier B.Y. All rights reserved

451 CHAPTER 28

Dystonia Ryuji Kaji* Department of Clinical Neuroscience, Hospital of the University of Tokushima, 2-Chome 5-1, Kuramotocho, Tokushima City, Tokushima 770-8503, Japan

28.1. Introduction: diagnosis of dystonia Dystonia is defined as a syndrome of sustained muscle contractions causing twisting or repetitive movements or abnormal postures (Fahn, 1988). It is classified as focal, segmental, multifocal, generalized or herni-dystonia according to the distribution of the symptoms. Primary or idiopathic dystonia represents a condition with no known cause, and usually lacks clinical signs other than dystonic muscle contractions. Some are found clearly genetic, and rarely accompany parkinsonism and hyperreflexia (DOPA-responsive dystonia or DRD). Dystonia may be secondary to focal lesions, other neurological illnesses or prolonged use of major tranquilizers or related drugs (secondary dystonia). Although widely used, the above definition does not fully represent the salient clinical features of this disorder, and some additional points deserve mention. Writer's cramp is a focal dystonia of the hand, which typically affects only writing at its onset (Fig. 1). This task-specificity may also be seen in other forms of dystonia. Abnormal muscle contractions involve the same set of muscles in the same way (Yanagisawa and Goto, 1971) such that the movements are stereotyped; patients with cervical dystonia, for example, always present involuntary head turning to a fixed direction. This distinguishes dystonia from other involuntary movements such as seen in chorea or athetosis, which lack stereotypy.

* Correspondence to: Prof. Ryuji Kaji, Department of Clinical Neuroscience, Hospital of the University of Tokushima, 2-Chome 5-1, Kuramotocho, Tokushima City, Tokushima 770-8503, Japan. E-mail address:[email protected] or [email protected] Tel.: 81-88-633-7206; fax: 81-88-633-7208.

Another feature of dystonia is the importance of sensory input in modifying the abnormal muscle contractions. The most characteristic is the sensory trick, in which a fixed cutaneous or proprioceptive input nearby the affected region helps the patient reestablish normal posture or movement. For instance, a patient with cervical dystonia uses touching a part of the face or the skin in the neck with his or her hand as a method to correct abnormal posture. Such a phenomenon is not lirnited to the somatosensory modality. For instance, patients with blepharospasm frequently complain of excessive brightness in the sun that leads to more eye closure, and benefit from wearing dark sunglasses. For clinical evaluation of patients, surface electromyography (EMG) studies give most valuable information. These should be recorded in different motor tasks with or without sensory trick. They may

Fig. 1. Surface EMGs recorded from wrist flexor, extensor, biceps, triceps muscles during waving (left) and writing (right) in a patient with writer's cramp. Adapted from Rothwell et al. (1983).

452

be repeated to see if the pattern of muscle activation is reproducible, or stereotyped. Normal phasic voluntary movement comprises reciprocal activation of local agonist and antagonist muscles. In dystonia, surface EMGs frequently show simultaneous contraction of agonist and antagonist muscles (cocontraction) or contraction of surrounding muscles (oveiflow) (Fig. 1). If abnormal phasic muscle contractions intervene, surface EMGs should tell whether they are rhythmic tremor or irregular myoclonic jerks. Serial activation of muscles appropriate for performing a task is also impaired, so that two muscles contract not sequentially but simultaneously (Rothwell et al., 1983). These features are useful in diagnosing dystonia, and suggest that dystonia is a disorder of a motor program for performing a routine act which probably utilizes sensory input for motor control (Kaji et aI., 1995d).

28.2. Historical perspective Writer's cramp was first recognized as scrivener's palsy as early as 1855, and appeared in the Encyclopaedia Britannica in 1877. During the Victorian era, the world's great commerce center in London created a large number of scriveners who were responsible for copying many papers of contract and other business documents by hand. Many of them were required to use a quill, whose thin shaft had to be gripped firmly. Some of them soon developed a strange motor disability that first affected only writing, but later involved other tasks. In 1878, Samuel Wilks wrote on scrivener's palsy under the title of "local spasms" in his textbook (Wilks, 1878). "Local spasms" included wry neck (or spasmodic torticollis), spasm of the facial muscles (currently known as blepharospasm or Meige syndrome), and spasmodic contraction of the jaw (mandibular dystonia). All these disorders are now known as focal dystonias. The first complete account of cervical dystonia was given by Destarac (1901), who clearly showed the sensory trick or geste antagoniste as the cardinal feature. DennyBrown was puzzled by this sensory phenomenon (Gilman et aI., 1999), and viewed generalized dystonia as resulting from imbalance of reflex responses to natural sensory inputs (Denny-Brown et al., 1964; Denny-Brown, 1965). Despite these early observations, it was as late as the 1980s when all these seemingly different clinical presentations of

R. KAJI

focal dystonias were first recognized as sharing common features (Marsden, 1976; Sheehy and Marsden, 1982). On the other hand, a hereditary form of generalized dystonia was first described as dystonia musculorum deformans by Oppenheim (Oppenheim, 1911). Genetic studies recently showed a GAG deletion in the DYTl gene in this condition, now called primary torsion dystonia or Oppenheim dystonia.

28.3. Sensory and subcortical abnormalities in dystonia The sensory phenomena in dystonia suggest that adjusting the link between sensory input and movement allows motor commands to be issued more effectively from the brain. Repetitive exposure to bright light may induce blepharospasm (Kaji et al., 1999), and as mentioned, wearing dark sunglasses often relieves symptoms of blepharospasm. Abnormal postures in focal dystonia are often maintained after botulinum toxin injections into active muscles; in patients with cervical dystonia, previously silent agonist muscles (e.g. the anterior margin of the trapezius muscle) come into action after a hyperactive muscle (e.g. the sternocleidomastoid muscle on the same side) has been weakened with botulinum toxin. This phenomenon indicates that an abnormal body image but not individual muscle hyperactivity is pre-programmed in cervical dystonia, and underscores the crucial role of proprioceptive sensory input in causing dystonia. These sensory aspects of dystonia have attracted increasing attention (Hallett, 1995). Recently, direct evidence of sensory abnormality was obtained at the cortical level. Byl and colleagues (Byl et aI., 1996) successfully produced a model of hand dystonia in the monkey, where they found a markedly disorganized representation of the digits in the primary somatosensory cortex. Interestingly, these investigators produced the model by imposing highly repetitive and demanding task of hand opening and closing (Topp and Byl, 1999), which was similar to that of a scrivener. Subsequently, Bara-Jimenez and colleagues (Bara-Jimenez et al., 1998) found analogous abnormalities of hand representation at the cortex in patients with writer's cramp using brain mapping techniques. Recently, this hand

453

DYSTONIA

Dystonia

Normal

... u

"+U

IlU

IlU 151'VI

...u IlU

... u MU

... u MU

I

.hU 51'V

I

MU

51' V

Nt

5m. 5m.

Fig. 2. Median and ulnar SEPs recorded from a normal subject (left) and a patient with hand dystonia (right). Algebraic sum (M + U) and waveforms after simultaneous stimulation (MU) are shown for each recording site (from above, frontal, parietal, cervical and brachial plexus). Double-headed arrows indicate the difference between M+U and MU, which was lost in dystonia. Adapted from Tinazzi et al. (2000).

representation was found abnormal not only in the affected but also in the non-affected hand in dystonia (Meunier et al., 2001). This speaks in favor of the disorganized sensory cortex as a cause rather than a consequence of the disease. Moreover, a trauma to the affected body part, which apparently disturbs the input-output link, has been recognized as a risk factor to develop dystonia (Jankovic, 1994). These findings strengthened the case for abnormal sensorymotor integration. Clinical sensory deficits were also documented in hand dystonia. Spatial and temporal discrimination of somesthetic stimuli were found abnormal (Bara-Jimenez et al., 2000a, b). In line with these observations, temporal and spatial summation of upper-limb somatosensory evoked potentials were found excessively large in hand dystonia (Tinazzi et al., 2000; Frasson et al., 2001) (Fig. 2). The latter findings lead to the notion that "overflow" is seen in the sensory as well as in the motor domain, and surround and collateral inhibitory mechanisms in the sensory system seem deficient (Fig. 3). Murase and colleagues (Murase et al., 2000) studied the attenua-

tion (gating) of somatosensory evoked potentials (SEPs) before and during hand movements in patients with writer's cramp, and found a lack of Cortex

--

_

Excitation

c:J Inhibition

Fig. 3. A diagram to show surround and collateral inhibition at various sensory relay nuclei. In dystonia, this inhibitory mechanism seems to be deficient. Adapted from Brodal (1981).

454

R. KAJI

Fig. 4. Premovement gatingof median SEPs in a normal subject(left) and a patientwith hand dystonia(right).Frontal N30 a and b are normally attenuated prior to movement, but are unchangedin dystonia.Adapted from Muraseet al. (2000). normal gating before movements (Fig. 4). This indicates an abnormality of utilizing sensory input in preparation for a movement in dystonia.

was interpreted as reduced presynaptic inhibition of Ia terminals. This deficient presynaptic inhibition may lead to the enhanced TVR in affected muscles, whereas blocking Ia input may compensate for the

28.4. Is dystonia a reflex abnormality? As was predicted by the reflex theory of DennyBrown (1965), dystonic contractions in writer's cramp can be reproduced by stimulating group Ia afferents by high-frequency vibration (tonic vibration reflex or TVR) and are abolished by blocking them with diluted lidocaine (muscle afferent block) (Kaji et al., 1995b, c). Multiple mechanisms must underlie this phenomenon. In normal subjects, one of the effects of vibration is to increase the preexisting presynaptic inhibition of group Ia afferent terminals, while simultaneously facilitating the TVR (Fig. 5). Using H-reflex recovery curves, Nakashima and colleagues (Nakashima et aI., 1989) demonstrated the lack of reciprocal inhibition at 20 ms conditioning-test stimuli interval in writer's cramp, possibly contributing to the co-contraction in dystonia. This

lstim

(H wave)

Fig. 5. An illustration to show the spinal stretchreflex,the gamma efferentand presynaptic inhibition.

455

DYSTONIA

lack of presynaptic inhibition. In this experiment, the sensory trick of applying cutaneous input to the affected hand was equally effective for abolishing abnormal TVR and dystonic symptoms. Of note is that cutaneous input is known to enhance presynaptic inhibition. An alternative hypothesis is that muscle spindle responsiveness is enhanced through abnormal activity of the gamma motor fibers, which may be effectively blocked by diluted lidocaine. Those muscles with increased gamma activity may be prone to develop a TVR because of increased spindle sensitivity. The dysfunction of presynaptic inhibitory mechanisms was also indicated by a study using tendon stimulation (Lorenzano et al., 2000). Several studies have explored the role of vibratory stimuli in dystonia. Grunewald and colleagues (Grunewald et al., 1997) found that vibratoryinduced illusion of hand motion was abnormally perceived in hand dystonia patients who showed normal position sense at rest, which suggests impaired processing of muscle spindle afferents in dystonia. Using transcranial magnetic stimulation of the primary motor cortex (M1) as a probe, Rosenkranz and colleagues (Rosenkranz et al., 2000) found the excitability of M1, which normally increases during vibratory stimulation, is unchanged in dystonia patients. This accords with the previous observation of the lack of cerebral blood flow increases during vibrotactile stimulation in hand dystonia (Tempel and Perlmutter, 1990, 1993). Serrien and colleagues (Serrien et al., 2000) examined the effect of vibratory stimulation in the affected and unaffected hand during precision grip, and found the task was impaired in dystonics but not in normals. As for other subcortical abnormalities, an abnormality in the recovery function of the blink reflex was first found in blepharospasm (Berardelli et al., 1985). The blink reflex R2 component is normally inhibited after electric conditioning pulses. This inhibition becomes much less in blepharospasm than in normals. Similar findings of abnormal recovery function were later confirmed in other types of dystonia as well as in Parkinson's disease. In normal subjects, similar but less inhibition of R2 recovery was found after photic conditioning stimuli. Patients with blepharospasm show much less effect of photic conditioning than normals (Katayama et al., 1996). This indicates that abnormal excitability of blink reflex pathways is not appropriately inhibited by

light, and this may underlie the clinically observed sensitivity to brightness in these patients.

28.5. Abnormal brain activity preceding movements in dystonia Brain potentials preceding spontaneous movements, the Bereitschaftspotential (BP) have been studied in patients. The second component, from 650 ms to 50 ms preceding the hand movement (NS' or NS2), was found decreased in writer's cramp (Deuschl et al., 1995). Using the same raw data and the event-related desynchronization technique, Toro and colleagues (Toro et al., 2000) demonstrated less reduction of 20-30 Hz EEG power preceding hand movements in dystonics than in normals. Yazawa and colleagues (Yazawa et al., 1999) recorded BP preceding muscle relaxation in hand dystonia, and found more extensive abnormality before relaxation than before initiating movement in hand dystonia. Hummel and colleagues (Hummel et al., 2002) reported that the increase in synchrony of cortical activity, reflecting inhibitory cortical control, was defective for hand movements in dystonia. These may correspond to recent observations showing defective inhibitory cortical mechanisms as discussed later. EEGs before cued movements are generally called contingent negative variation (CNV). Its late component, representing movement preparation, was abnormally decreased specifically for head turning in cervical dystonia (Kaji et al., 1995a), and for hand movement in writer's cramp (Hamano et al., 1999) (Fig. 6). These task-specific brain activities clearly point to defective motor control for specific tasks in dystonia.

28.6. Transcranial magnetic stimulation Transcranial magnetic stimulation (TMS) of the primary motor cortex allows direct assessment of the corticospinal projection. Using different stimulus intensities and facilitation levels, Ikoma and colleagues (Ikoma et al., 1996) found abnormal excitability curves for dystonia patients. Inhibition in motor cortex is also deficient in patients with hand dystonia. Ridding and colleagues (Ridding et al., 1995) studied intracortical inhibition using a "double pulse" paradigm, in which a subthreshold conditioning pulse precedes a test pulse. They found less

R. KAJI

456

Fig. 6. Contingent negative potential (CNV) prior to head rotation (upper) and finger extension (lower) in patients with cervical dystonia (left column) and hand dystonia (right column). Grand-averages are shown and thick traces are from the patients, thin ones from normal controls. 51: warning, 52: imperative signal. Recorded from Cz. Adapted from Kaji et al. (1995) and Hamano et al. (1999).

inhibition of conditioned motor evoked potentials (MEPs) at 1-5 ms intervals in dystonia than in normals. TMS can also suppress ongoing voluntary muscle activation (silent period). The duration of this silent period is another index of the cortical excitability. Chen and colleagues (Chen et al., 1997) reported this silent period was abnormally short in dystonia, again suggesting an abnormal cortical inhibitory mechanism. These abnormalities of cortical inhibition were also found in obsessive compulsive disorder (Greenberg et al., 2000), which possibly is a psychological feature of dystonia (Kubota et al., 2001). These findings are best explained by deficient GABAergic inhibitory interneurons in motor cortex, because intrathecal baclofen, a GABA agonist, prolongs silent period in generalized dystonia (Siebner et al., 1998). More direct evidence comes from MR spectroscopy. GABA levels were significantly reduced in the sensorimotor cortex and the lentiform nucleus contralateral to the dystonic hand (Levy and Hallett, 2002). Slow (l Hz) repetitive TMS (rTMS) was shown to reduce cortical excitability or to reinforce cortical inhibition. Siebner and colleagues (Siebner et al., 1999) demonstrated that rTMS could prolong silent

periods and increase intracortical inhibition in writer's cramp. This technique may therefore prove clinically useful for treating these patients. 28.7. Treatment of dystonia Anticholinergic agents such as trihexyphenidyl and GABA agonists like baclofen have been the mainstay for medical treatment of dystonia until botulinum toxin (BTX) injections revolutionized the therapy of focal dystonia. These agents are still used in conjunction with BTX injection with success. BTX injections are superior to phenol motor point injections because of their relatively selective action on muscle fibers with increased activities. For instance, twitching muscles are more sensitive to BTX than non-twitching ones in patients with hemifacial spasm, so that twitchings can be abolished without causing complete facial paralysis. This may be due to the fact that BTX is taken up more in active muscles. The precise mechanism of this selectivity is still unknown, but recent progress in the basic toxin research has revealed that type B toxin must recognize both ganglioside GTlb and synaptotagmin II for being internalized to the motor nerve ternimals (Fig. 7) (Nishiki et al., 1996; Kozaki et al.,

457

DYSTONIA

A.

I

GTIb SynaplDIagmtn II

Ach

Fig. 7. An illustration to show the acceptors of botulinum toxin type B (GTI band synaptotagmin II) for internalization of the toxin. Synaptotagmin II are mainly located at the inner surface of synaptosomes.

1998). Because synaptotagmin II is normally present in the inner surface of the synaptic vesicle, BTX is selectively taken up into those terminals with active vesicle release, or those at the activated muscles. Another important mechanism of BTX in dystonia is through its effect on intrafusal as well as extrafusal muscle fibers. Rosales and colleagues (Rosales et aI., 1996) examined the effect of BTX injections into rat muscles. They compared intrafusal fibers innervated by gamma efferent to the muscle spindles with extrafusal fibers, and found that both fiber types showed atrophy to a similar extent. Hence BTX will affect muscle afferent actively. Gilio and colleagues (Gilio et aI., 2000) studied the effect of botulinum toxin type A on intracortical inhibition in patients with dystonia. They demonstrated that BTX transiently altered the excitability of the cortical motor areas possibly through its action on muscle afferents (Hallett, 2000). Likely by utilizing this phenomenon, blocking gamma efferent fibers with diluted lidocaine and ethanol has been used to block muscle afferents (muscle afferent block or MAB) to treat writer's cramp (Kaji et aI., 1995b, c), spasmodic torticollis (Kubori et aI., 2000), oromandibular dystonia (Yoshida et aI., 1998). The advantages of MAB over botulinum toxin injection are the lack of weakness because of the relative sparing of the extrafusal

fibers, and the low cost, although it requires repeated injection trials to obtain durable effects. A triaI of MAB for blepharospasm was unsuccessful, probably because of the lack of muscle spindles in the facial muscles. Another advance in the treatment of dystonia is stereotactic surgery. There is an increasing consensus that bilateral stimulation of the internal globus pallidus (GPi) with high frequencies is effective in generalized dystonia including those with DYTl gene mutation (Coubes et aI., 2000; Tronnier and Fogel, 2000; Vercueil et aI., 2001). Deep brain stimulation of GPi was also applied in treating other types of dystonia (Krauss et aI., 1999, 2002). Because the basic abnormality in dystonia is sensory-motor disintegration, it would be reasonable to use the sensory input appropriate for motor performance as a tool of training. In fact, Zeuner and colleagues (Zeuner et aI., 2002) were successful in treating patients with writer's cramp by training those patients with braille reading. Another approach is the limb immobilization with a cast for extended periods (Priori et aI., 2001). This could possibly break down abnormal sensory-motor link in hand dystonia.

28.8. Dystonia as a basal ganglia disorder There is now increasing awareness that dystonia is a basal ganglia disorder, because focal lesions in the basal ganglia or their connections produce hemidystonia including the hand on the contralateral side (Marsden et aI., 1985), and because trihexyphenidyl, an anticholinergic agent used in Parkinson's disease, is also effective in dystonia. The basal ganglia lie between the cerebral cortex and the thalamus, with which they have dense fiber connections forming 4-5 distinct circuits allowing parallel processing of information (Alexander and Crutcher, 1990). The most studied is the motor loop, which has direct and indirect pathways. The direct pathway disinhibits the powerful inhibition of GPiI substantia nigra pars reticulata upon thalamic ventrolateral nuclei (Vop) with a net facilitation on the motor cortex. By contrast, the indirect pathway exerts an inhibitory effect. This dual system provides a center (excitatory)-surround (inhibitory) mechanism to focus its effect on selected cortical neurons. Despite considerable knowledge of the chemical

458

messengers in these projections, the functional role of the loop in motor control is not precisely understood. The sensory trick in dystonia highlights sensory aspects implicated in this basal ganglia disorder. Although the basal ganglia are commonly regarded as a center for motor control, its sensory role has been under-emphasized. Lenz and colleagues (Lenz et al., 1999) made intraoperative recordings in dystonia patients and showed that neurons in Vop discharge at peak EMG spectrum frequencies of dystonic contractions. They also found abnormal sensory units responding to more than one joint in Vim, the cerebellar relay nucleus of the thalamus. This enlarged receptive area in the thalamus is probably associated with the disorganized primary sensory cortex in dystonia (Bara-Jimenez et al., 1998). Lenz and colleagues also demonstrated the lack of correlation between the discharge of jointsensitive Vim neurons and EMG of its effector muscles, suggesting a mismatch between the proprioceptive input and the effector muscle to be controlled. Physiological studies in the monkey showed that numerous neurons in the supplementary motor or premotor cortex and the basal ganglia discharge in response to a sensory cue long before the movement onset (Romo et al., 1993). However, the exact pathway through which the sensory input reaches the basal ganglia remains elusive. Animal studies indicate that sensory inputs reaching the basal ganglia significantly differ from those in the lemniscal system in that they show encoding of information that appeared to be relevant for motor control (Lidsky et aI., 1985). Indeed the basal ganglia appear to 'gate' sensory inputs at various levels (Schneider et aI., 1986; Tinazzi et aI., 2000). Stimulation of the basal ganglia inhibits auditory and visual cortical evoked responses. Lemniscal and extraleminiscal components of the somatosensory system are modified by the basal ganglia. For example, the sensory responsiveness of second order neurons in the trigeminal sensory nucleus is altered by activation of the caudate nucleus or the globus pallidus. Such gating of sensory input is known to use the centersurround inhibition mechanism. All these pieces of evidence tempt us to speculate that dystonia is a disorder of a frequently used motor program or subroutine, in which motor output is matched to a fixed sensory input (Kaji et aI., 1995d).

R. KAJI

Probably this motor subroutine is stored as connectivities in the motor loop. If the center-surround inhibitory mechanism in the motor loop is disrupted, the lack of inhibition of antagonist or surrounding muscles causes co-contraction or overflow phenomena. 28.9. Conclusion

Lesions of the basal ganglia mostly affect automatic movements that need sensory guidance. It is therefore likely that the basal ganglia control automatic or highly trained movements in relation to relevant sensory inputs. Tasks impaired in writer's cramp or other dystonias could be among these. Indeed most of recent works in the physiology of dystonia converge upon disturbed sensory-motor integration and the lack of inhibitory motor control. If extraneous sensory input is fed back for subsequent movement, the mismatch would set up a vicious cycle causing further delapidation of motor control, resulting in dystonic movements. References Alexander, GE and Crutcher, MD (1990) Functional architecture of basal ganglia circuits: neural substrates of parallel processing. Trends Neurosci., 13(7): 266271. Bara-Jimenez, W, Catalan, MJ, Hallett, M and Gerloff, C (1998) Abnormal somatosensory homunculus in dystonia of the hand. Ann. Neurol., 44(5): 828-831. Bara-Jimenez, W, Shelton, P and Hallett, M (2000a) Spatial discrimination is abnormal in focal hand dystonia. Neurology, 55(12): 1869-1873. Bara-Jimenez, W, Shelton, P, Sanger, TD and Hallett, M (2000b) Sensory discrimination capabilities in patients with focal hand dystonia. Ann. Neurol., 47(3): 377380. Berardelli, A, Rothwell, JC, Day, BL and Marsden, CD (1985) Pathophysiology of blepharospasm and oromandibular dystonia. Brain, 108(3): 593-608. Byl, NN, Merzenich, MM and Jenkins, WM (1996) A primate genesis model of focal dystonia and repetitive strain injury: 1. Learning-induced dedifferentiation of the representation of the hand in the primary somatosensory cortex in adult monkeys. Neurology, 47(2): 508-520. Chen, R, Wassermann, EM, Canos, M and Hallett, M (1997) Impaired inhibition in writer's cramp during voluntary muscle activation. Neurology, 49(4): 1054-1059.

DYSTONIA

Coubes, P, Roubertie, A, Vayssiere, N, Hemm, S and Echenne, B (2000) Treatment of DYTl-generalized dystonia by stimulation of the internal globus pallidus. Lancet, 355(9222): 2220--2221. Denny-Brown, D (1965) The nature of dystonia. Bull. NY Acad. Med., 41: 858-869. Denny-Brown, DE, Gilman, S and Van der Meulen, J (1964) Patterns of cortical ablations leading to dystonic postures. Trans. Am. Neurol. Assoc., 89: 117-121. Destarac (1901) Torticolis spasmodique et spasmes functionels. Rev. Neurol., 9: 591-597. Deuschl, G, Toro, C, Matsumoto, J and Hallett, M (1995) Movement-related cortical potentials in writer's cramp. Ann. Neurol., 38(6): 862-868. Fahn, S (1988) Concepts and classification of dystonia. Adv. Neurol., 50: 1-8. Frasson, E, Priori, A, Bertolasi, L, Mauguiere, F, Fiaschi, A and Tinazzi, M (2001) Somatosensory disinhibition in dystonia. Mov. Disord., 16(4): 674-682. Gilio, F, Curra, A, Lorenzano, C, Modugno, N, Manfredi, M and Berardelli, A (2000) Effects of botulinum toxin type A on intracortical inhibition in patients with dystonia. Ann. Neurol., 48(1): 20--26. Gilman, S, Vilensky, JA, Morecraft, RW and Cook, JA (1999) Denny-Brown's views on the pathophysiology of dystonia. J. Neurol. Sci., 167(2): 142-147. Greenberg, BD, Ziemann, D, Cora-Locatelli, G, Harmon, A, Murphy, DL, Keel, JC et al. (2000) Altered cortical excitability in obsessive-compulsive disorder. Neurology, 54(1): 142-147. Grunewald, RA, Yoneda, Y, Shipman, JM and Sagar, HJ (1997) Idiopathic focal dystonia: a disorder of muscle spindle afferent processing? Brain, 120(12): 21792185. Hallett, M (1995) Is dystonia a sensory disorder? Ann. Neurol., 38(2): 139-140. Hallett, M (2000) How does botulinum toxin work? Ann. Neurol., 48(1): 7-8. Hamano, T, Kaji, R, Katayama, M, Kubori, T, Ikeda, A and Shibasaki, H et al. (1999) Abnormal contingent negative variation in writer's cramp. Clin. Neurophysiol., 110(3): 508-515. Hummel, F, Andres, F, Altenmuller, E, Dichgans, J and Gerloff, C (2002) Inhibitory control of acquired motor programs in the human brain. Brain, 125(2): 404-420. Ikoma, K, Samii, A, Mercuri, B, Wassermann, EM and Hallett, M (1996) Abnormal cortical motor excitability in dystonia. Neurology, 46(5): 1371-1376. Jankovic, J (1994) Post-traumatic movement disorders: central and peripheral mechanisms. Neurology, 44(11): 2006-2014. Kaji, R, Ikeda, A, Ikeda, T, Kubori, T, Mezaki, T, Kohara, N et al. (1995a) Physiological study of

459 cervical dystonia. Task-specific abnormality in contingent negative variation. Brain, 118(2): 511-522. Kaji, R, Kohara, N, Katayama, M, Kubori, T, Mezaki, T, Shibasaki, H et al. (l995b) Muscle afferent block by intramuscular injection of lidocaine for the treatment of writer's cramp. Muscle Nerve, 18(2): 234-235. Kaji, R, Rothwell, JC, Katayama, M, Ikeda, T, Kubori, T, Kohara, N et al. (l995c) Tonic vibration reflex and muscle afferent block in writer's cramp. Ann. Neurol., 38(2): 155-162. Kaji, R, Shibasaki, H and Kimura, J (l995d) Writer's cramp: a disorder of motor subroutine? Ann. Neurol., 38(6): 837-838. Kaji, R, Katayama-Hirota, M, Kohara, N, Kojima,Y,Yang, Q and Kimura, J (1999) Blepharospasm induced by an LED flashlight. Mov. Disord., 14(6): 1045-1047. Katayama, M, Kohara, N, Kaji, R, Kojima,Y, Shibasaki, H and Kimura, J (1996) Effect of photic conditioning on blink reflex recovery function in blepharospasm. Electroencephalogr. Clin. Neurophysiol., 101(5): 446-452. Kozaki, S, Kamata, Y,Watarai, S, Nishiki, T and Mochida, S (1998) Ganglioside GTlb as a complementary receptor component for Clostridium botulinum neurotoxins. Microb. Pathog., 25(2): 91-99. Krauss, JK, Pohle, T, Weber, S, Ozdoba, C and Burgunder, 1M (1999) Bilateral stimulation of globus pallidus internus for treatment of cervical dystonia. Lancet, 354(9181): 837-838. Krauss, JK, Loher, rr. Pohle, T, Weber, S, Taub, E, Barlocher, CB et al. (2002) Pallidal deep brain stimulation in patients with cervical dystonia and severe cervical dyskinesias with cervical myelopathy. J. Neurol. Neurosurg. Psychiatry, 72(2): 249-256. Kubori, T, Kaji, R, Mezaki, T et al. (2000) Muscle afferent block for cervical dystonia:a controlled trial with botulinum toxin. Neurology, 54: 198-199. Kubota, Y, Murai, T, Okada, T, Hayashi, A, Toichi, M and Sakihama, M et al. (2001) Obsessive-compulsive characteristics in patients with writer's cramp. J. Neurol. Neurosurg. Psychiatry, 71(3): 413-414. Lenz, FA, Jaeger, CJ, Seike, MS, Lin, YC, Reich, SG, DeLong, MR et al. (1999) Thalamic single neuron activity in patients with dystonia: dystonia-related activity and somatic sensory reorganization. J. Neurophysiol., 82(5): 2372-2392. Levy, LM and Hallett, M (2002) Impaired brain GABA in focal dystonia. Ann. Neurol., 51(1): 93-101. Lidsky, TI, Manetto, C and Schneider, JS (1985) A consideration of sensory factors involved in motor functions ofthe basal ganglia. Brain Res., 356(2): 133146. Lorenzano, C, Priori, MA, Curra, A, Gilio, F, Manfredi, M and Berardelli, A (2000) Impaired EMG inhibition

460 elicited by tendon stimulation in dystonia. Neurology, 55(12): 1789-1793. Marsden, CD (1976) Blepharospasm-oromandibular dystonia syndrome (Brueghel's syndrome). A variant of adult-onset torsion dystonia? J. Neurol. Neurosurg. Psychiatry, 39(12): 1204-1209. Marsden, CD, Obeso, JA, Zarranz, JJ and Lang, AE (1985) The anatomical basis of symptomatic hemidystonia. Brain, 108(2): 463-483. Meunier, S, Garnero, L, Ducorps, A, Mazieres, L, Lehericy, S, du Montcel, ST et al. (2001) Human brain mapping in dystonia reveals both endophenotypic traits and adaptive reorganization. Ann. Neurol., 50(4): 521527. Murase, N, Kaji, R, Shimazu, H, Katayama-Hirota, M, Ikeda, A, Kohara, N et al. (2000) Abnormal premovement gating of somatosensory input in writer's cramp. Brain, 123(9): 1813-1829. Nakashima, K, Rothwell, JC, Day, BL, Thompson, PD, Shannon, K and Marsden, CD (1989) Reciprocal inhibition between forearm muscles in patients with writer's cramp and other occupational cramps, symptomatic hemidystonia and hemiparesis due to stroke. Brain, 112(3): 681-697. Nishiki, T, Tokuyama,Y, Kamata, Y, Nemoto, Y,Yoshida, A, Sato, K et al. (1996) The high-affinity binding of Clostridium botulinum type B neurotoxin to synaptotagmin II associated with gangliosides GTlb/GDla. FEBS Lett., 378(3): 253-257. Oppenheim, H (1911) Uber eine eigenartige Kampfkrankheit des kindlichen und jugendlichen Alters (Dysbasia lordotica progressiva, Dystonia musculorum deformans). Neurol. Zentbl., 30: 1090-1107. Priori, A, Pesenti, A, Cappellari, A, Scarlato, G and Barbieri, S (2001) Limb immobilization for the treatment of focal occupational dystonia. Neurology, 57(3): 405-409. Ridding, MC, Sheean, G, Rothwell, JC, Inzelberg, Rand Kujirai, T (1995) Changes in the balance between motor cortical excitation and inhibition in focal, task specific dystonia. J. Neurol. Neurosurg. Psychiatry, 59(5): 493-498. Romo, R, Ruiz, S, Crespo, P, Zainos, A and Merchant, H (1993) Representation of tactile signals in primate supplementary motor area. J. Neurophysiol., 70(6): 2690-2694. Rosales, RL, Arimura, K, Takenaga, S and Osame, M (1996) Extrafusal and intrafusal muscle effects in experimental botulinum toxin-A injection. Muscle Nerve, 19(4): 488-496. Rosenkranz, K, Altenmuller, E, Siggelkow, S and Dengler, R (2000) Alteration of sensorimotor integration in

R. KAJI

musician's cramp: impaired focusing of proprioception. CUn. Neurophysiol., 111(11): 2040-2045. Rothwell, J, Obeso, J, Day, B and Marsden, C (1983) Pathophysiology of dystonia. Adv. Neurol., 39: 851863. Schneider, JS, Diamond, SG and Markham, CH (1986) Deficits in orofacial sensorimotor function in Parkinson's disease. Ann. Neurol., 19(3): 275-282. Serrien, DJ, Burgunder, JM and Wiesendanger, M (2000) Disturbed sensorimotor processing during control of precision grip in patients with writer's cramp. Mov. Disord., 15(5): 965-972. Sheehy, MP and Marsden, CD (1982) Writers' cramp - a focal dystonia. Brain, 105(3): 461-480. Siebner, HR, Dressnandt, J, Auer, C and Conrad, B (1998) Continuous intrathecal baclofen infusions induced a marked increase of the transcranially evoked silent period in a patient with generalized dystonia. Muscle Nerve, 21(9): 1209-1212. Siebner, HR, Tormos, JM, Ceballos-Baumann, AO, Auer, C, Catala, MD, Conrad, B et al. (1999) Low-frequency repetitive transcranial magnetic stimulation of the motor cortex in writer's cramp. Neurology, 52(3): 529-537. Tempel, LW and Perlmutter, JS (1990) Abnormal vibration-induced cerebral blood flow responses in idiopathic dystonia. Brain, I 13(Pt 3): 691-707. Tempel, LW and Perlmutter, JS (1993) Abnormal cortical responses in patients with writer's cramp. Neurology, 43(11): 2252-2257. Tinazzi, M, Priori, A, Bertolasi, L, Frasson, E, Mauguiere, F and Fiaschi, A (2000) Abnormal central integration of a dual somatosensory input in dystonia. Evidence for sensory overflow. Brain, 123(1): 42-50. Topp, KS and Byl, NN (1999) Movement dysfunction following repetitive hand opening and closing: anatomical analysis in Owl monkeys. Mov. Disord., 14(2): 295-306. Toro, C, Deuschl, G and Hallett, M (2000) Movementrelated electroencephalographic desynchronization in patients with hand cramps: evidence for motor cortical involvement in focal dystonia. Ann. Neurol., 47(4): 456-461. Tronnier, VM and Fogel, W (2000) Pallidal stimulation for generalized dystonia. Report of three cases. J. Neurosurg., 92(3): 453-456. Vercueil, L, Pollak, P, Fraix, V, Caputo, E, Moro, E, Benazzouz, A et al. (200 I) Deep brain stimulation in the treatment of severe dystonia. J. Neurol., 248(8): 695-700. Wilks S (1878) Lectures on diseases of the nervous system. London, Churchill.

DYSTONIA

Yanagisawa, Nand Goto, A (1971) Dystonia musculorurn deformans. Analysis with electromyography. J. Neurol. Sci., 13(1): 39-65. Yazawa, S, Ikeda, A, Kaji, R, Terada, K, Nagamine, T, Toma, K et al. (1999) Abnormal cortical processing of voluntary muscle relaxation in patients with focal hand dystonia studied by movement-related potentials. Brain, 122(7): 1357-1366.

461 Yoshida, K, Kaji, R, Kubori, T, Kohara, N, Iizuka, T and Kimura, J (1998) Muscle afferent block for the treatment of oromandibular dystonia. Mov. Disord., 13(4): 699-705. Zeuner, KE, Bara-Jimenez, W, Noguchi, PS, Goldstein, SR, Dambrosia, JM and Hallett, M (2002) Sensory training for patients with focal hand dystonia. Ann. Neurol., 51(5): 593-598.

Movement Disorders Handbook of Clinical Neurophysiology, Vol. 1 M. Hallett (Ed.) © 2003 Elsevier B.V. All rightsreserved

463 CHAPfER29

Stiffness with continuous motor unit activity P. Brown* Sobell Department of Motor Neuroscience and Movement Disorders, Institute of Neurology, London WCI N 3BG, UK

29.1. Introduction

Table 2

Stiffness is a common feature of many disorders of the motor system. However, only rarely does it persist, regardless of relaxation or change in posture, as evidenced by continuous motor unit activity (CMUA) on EMG. Although rare, causes are varied and can involve central and peripheral conditions. Table I summarizes those features helpful in separating peripheral and central causes of stiffness in the presence of CMUA.

Central causes of stiffness with CMUA. Stiff person syndrome I Stiff person plus syndromes' I. Subacute': 'Progressive encephalomyelitis with rigidity' 2. Chronic": Brainstem form - Includes the 'jerking stiff man syndrome' Spinal form 'Stiff limb syndrome' Focal lesions of the spinal cord Intrinsic neoplasms Syringomyelia Traumatic Vascular Paraneoplastic segmental myelitis

29.2. Central stiffness with CMUA

The central causes of stiffness with CMUA are summarized in Table 2.

Infective/toxic causes Acute poliomyelitis Borreliosis Encephalomyelitis lethargica Acute or chronic tetanus Strychnine

Table I Features distinguishing central and peripheral causes of stiffness with CMUA. Central

Peripheral

++ ++

+/-

See Table 3 for clinicalfeatures. See Table 4 for clinicalfeatures. 3 Death within 3 years. Long tract signs present. 4 Survive > 3 years. Long tract signs absent/few. I

Spasm painful Reflex exacerbations Clinical myokymia! fasciculations Abnormal exteroceptive reflexes EMG evidence of neuromyotonia

++ ++ ++

* Correspondence to: Dr. P. Brown, Sobell Department of Neurophysiology, Institute of Neurology, Queen Square, London WCIN 3BG, UK. E-mail address:[email protected] Tel.: 0207 829 9836; fax: 020 7278 9836.

2

29.2.1. Tetanus

Tetanus is suggested by the classical symptoms of trismus, dysphagia and muscular rigidity (Habermann, 1978). The peripheral silent period is abnormal in contrast to the stiff person syndrome. Treatment consists of wound management, neutralization of tetanus toxin by immunoglobulins, antibiotics, treatment of muscle spasm and instability of the autonomic nervous system, and supportive care. Acute tetanus may be mimicked by strychnine poisoning (Hardin and Griggs, 1971).

464

P.BROWN

29.2.2. Focal spinal cord pathology

Focal spinal cord pathology that preferentially involves the grey matter may cause stiffness, sometimes termed alpha rigidity in the literature, based on the hypothesis that it may arise from loss of inhibitory interneuronal inputs to alpha motoneurons (Gelfan and Tarlov, 1959, 1963). Intrinsic tumors (Rushworth et al., 1961; Lourie, 1968), syringomyelia (Tarlov, 1967), vascular insufficiency (Davis et al., 1987) and paraneoplastic myelitis (Roobol et al., 1987) may be associated with stiffness, postural abnormality, CMUA, and reflex and action induced spasms of the trunk and limbs (Fig. 1). There may be additional wasting, weakness, absence of tendon reflexes and signs of denervation upon EMG in the affected myotomes. The relatively selective destruction of spinal intemeurons in some of these patients has been confirmed histologically (Rushworth et al., 1961). Effective stimuli need not be restricted to somesthetic stimulation below the level of the spinal cord pathology. Paradoxically, jerks and spasms of the legs can be precipitated in patients with spinal lesions by startle inducing stimuli, including sounds (Sikes et al., 1959; Davis et al., 1987). Presumably this represents an excessive response at the segmental level to descending reticulospinal activity. 29.2.3. Stiffperson syndrome

The term stiff man or stiff person syndrome was first introduced by Moersch and Woltman in 1956 in their original description of axial rigidity and spasm, often in association with diabetes mellitus. Histology was unremarkable in the one case that came to post-mortem. Since then a number of cases have been reported, some very similar to those of Moersch and Woltman, but others atypical in that additional signs or limb involvement were found that were absent in the original report. Some of these atypical cases have had evidence of a polio encephalomyelitis at post-mortem. It is debatable whether these cases form a continuum with the classical stiff person syndrome or whether the classical axial stiff person syndrome forms a discrete entity with atypical cases resulting from other pathologies with differing features and prognosis. The limited histology available would support the latter view and this is the one taken here.

Fig. 1. Abnormal fixed posture of arms and shoulder girdle due to rigidity with CMUA in a man with a cervical astrocytoma. Reprinted with permission from Rushworth et al. (1961).

29.2.3.1. Cases without encephalomyelitis: the classical stiffperson syndrome The classical stiff person syndrome, as described by Moersch and Woltman (1956) is characterized by paraspinal and abdominal rigidity with an exaggerated lumbar lordosis, and superimposed spasms precipitated by movement, emotional upset, peripheral stimulation or auditory startle (Fig. 2). The proximal legs are involved in some cases, but this is often only apparent on walking when the patient has a stiff wooden gait. The calf and foot muscles are rarely, if ever, involved (Moersch and Woltman,

STIFFNESS WITH CONTINUOUS MOTOR UNIT ACTIVITY

465

Fig. 2. Abnormally excessive activity of the paraspinal muscles causing folding of the skin and lumbar hyperlordosis in a patient with anti-GAD antibody positive classical axial stiff person syndrome.

1956; Spehlmann and Norcross, 1979; Lorish et al., 1989). There is no weakness, sensory loss, sphincter involvement or clinical evidence of brainstem disturbance. Diagnostic criteria for the stiff person syndrome have been adapted from those of Lorish et al. (1989) and are summarized in Table 3. Cases defined in this way respond to diazepam and baclofen and have a good prognosis. For example, Lorish et al. (1989) followed up 12 patients over a mean of nine years, all of whom remained ambulant until last seen, and experience has been similar in a recent series (Barker et aI., 1998). Up to 70% of such selected patients are diabetic (Lorish et aI., 1989) and 90%

have antibodies against the GABA-synthesizing enzyme glutamic acid decarboxylase (GAD), usually in very high titer (Barker et aI., 1998); further evidence that the general criteria of Lorish et al. (1989) can identify a disease that is fairly homogenous in both clinical and immunological terms. Nevertheless, it is clear that anti-GAD antibodies are not exclusive to the stiff person syndrome, in the same way as rheumatoid factor is not exclusively seen in rheumatoid arthritis (Nemni et aI., 1994; Saiz et aI., 1997). Apart from diabetes and anti-GAD antibodies a number of other auto-immune diseases and organ specific antibodies are commonly found in the classical stiff person syndrome, such as thyroid

466 Table 3 Criteria for the diagnosis of the classical stiff person syndrome*. I. Stiffness and rigidity in axial muscles (proximal limb muscles may also be sometimes involved). 2. Abnormal axial posture (usually an exaggeration of the normal lumbar lordosis). 3. Superimposed spasms precipitated by voluntary movement, emotional upsets and unexpected auditory and somesthetic stimuli. 4. Absence of brainstem, pyramidal, extrapyramidal and lower motor neuron signs, sphincter and sensory disturbance, and cognitive involvement.** 5. CMUA in at least one axial muscle.

* Adapted from Lorish et al. (1989). They also considered a positive response to intravenous or oral diazepam a necessary prerequisite for the diagnosis of the stiff man syndrome. ** Epilepsy may occur. disease and thyroid antibodies and pernicious anemia and antibodies to intrisic factor and gastric parietal cells. The cerebrospinal fluid may contain bands of oligoclonal IgG, but the cell count is not elevated. The pathophysiology of the stiff man syndrome remains uncertain. Moersch and Woltman found no significant pathological changes at post-mortem. This finding has since been confirmed (Trethowan et aI., 1960; Martinelli et aI., 1978), with one exception, which suggested loss of GABAergic cells in the cerebellar cortex (Warich-Kirches et al., 1997). An immune etiology seems likely in view of the autoantibody profile of many of the patients (Solimena et aI., 1990). More specifically, Guilleminault et aI. (1973) have suggested a functional imbalance between descending aminergic, possibly reticulospinal, projections to the cord, facilitating flexor reflex pathways, and inhibitory GABAergic systems. This idea receives some support from the widespread enhancement of exteroceptive reflexes, including blink reflexes (Meinck et aI., 1984), and from pharmacological studies (Guilleminault et aI., 1973; Meinck et aI., 1984). The common presence of antibodies against the GABA synthetic enzyme GAD would also be consistent with this hypothesis, although the pathogenic importance of these antibodies remains unclear, as Purkinje cells, known to contain high amounts of GAD, may be unaffected (Ishzawa et aI., 1999) and high titers of the antibody

P. BROWN

are being recognized in other conditions, particularly cerebellar ataxia (Saiz et aI., 1997). 29.2.3.2. Cases with encephalomyelitis: stiffperson plus syndromes Several cases of stiff people have now been reported with a recognizable, fairly uniform, pathology, that varies in severity (Campbell and Garland, 1956; Kasperek and Zebrowski, 1971; Lhermitte et al., 1973; Whitely et aI., 1976; Howell et aI., 1979; Fenzi et aI., 1988; Batemen et aI., 1990; Meinck et al., 1994; Armon et al., 1996; Barker et aI., 1998). This consists of a subacute or chronic encephalomyelitis with prominent involvement of the gray matter (polioencephalomyelitis). The spinal cord and brainstem are most severely affected. Hemispheric structures are relatively spared, with the frequent exception of the limbic areas. The additional clinical features associated with this pathology are listed in Table 4, and can be usefully thought of as findings suggestive of the stiff man plus syndromes. Perhaps the core feature is that rigidity is not confined to the trunk, but also involves the distal limb, often exclusively. It seems likely that the pathological changes described in the stiff man plus syndromes directly account for the curious rigidity. Similar preferential involvement of the gray matter of the spinal cord is seen in dogs with experimentally induced ischemic damage to the cord. These animals develop a form of Table 4 Clinical features in stiff people with histological evidence of a polioencephalomyelitis*. I. Rigidity and abnormal posturing of one or more limbs, that includes the hand or foot 2. Myoclonus involving all four limbs 3. Brainstem signs 4. Long tract signs 5. Lower motor neuron signs 6. Cognitive changes, especially memory impairment 7. Autononic involvement 8. CSF pleocytosis

* Subacute or chronic encephalomyelitis which predominantly affects gray matter. Findings drawn from Campbell and Garland, 1956; Kasperek and Zebrowski, 1971; Lhermitte et al., 1973; Whitely et al., 1976; Howell et al., 1979; Fenzi et aI., 1988; Batemen et aI., 1990; Meinck et al., 1994; Armon et al., 1996; Barker et aI., 1998.

STIFFNESS WITH CONTINUOUS MOTOR UNIT ACTIVITY

467

rigidity, termed alpha rigidity, that is due to the isolation of motoneurons from the action of spinal inhibitory interneurons (Gelfan and Tarlov, 1959, 1963). The selective loss of spinal interneurons has been confirmed in progressive encephalomyelitis with rigidity (Howell et al., 1979).

encephalomyelitis with rigidity (Fenzi et aI., 1988). A comparable situation exists in other 'paraneoplastic conditions'; most cases are rapidly progressive, but occasional patients are seen that survive many years without the appearance of malignancy, raising the possibility that similar pathology may be triggered by both tumor and autoimmune diathesis (Nemni et aI., 1993). Certainly, one case of jerking stiff man syndrome has been reported in whom antiacetylcholine receptor, antinuclear, gastric parietal, thyroid microsomal, thyroglobulin and anti-GAD antibodies were positive (Bum et aI., 1991), and striking therapeutic success has been achieved with plasmapheresis and immunosuppression (Fogan, 1996). The remaining chronic cases have delayed and mild, or no signs of brainstem dysfunction, and do not develop generalized myoclonus. Instead, the clinical picture is dominated by rigidity, postural abnormality and painful spasms of the limbs, especially distally. The legs are most commonly involved and there is a relative or total sparing of the trunk. As such the condition was initially termed the stiff leg or limb syndrome (Brown et aI., 1997). Others have suggested the term focal stiff-man syndrome on the basis that some have antibodies to GAD (Saiz et aI., 1998). However, publication of reports of single cases or short series may tend to positively discriminate against cases without GAD antibodies, where the diagnosis may be seen as less certain. In a recent large series anti-GAD antibody positive patients represented only 15% of cases of stiff limb syndrome (Barker et al., 1998). Indeed, the available, but limited, histology would be more in keeping with the other stiff man plus syndromes (see below). We prefer the term stiff limb syndrome as the upper limbs may occasionally be involved (Fig. 3), and no pathological relationship with the classical stiff man syndrome is implied. The EMG activity recorded in the limb spasms has an unusual segmented appearance in three quarters of such patients (Brown et aI., 1997; Barker et aI., 1998), distinct from the normal looking interference pattern recorded in the spasms of the stiff man syndrome (Moersch and Woltman, 1956). The segmented EMG is due to the abnormally synchronous discharge of motor units (Fig. 4). To date, pathology has only been reported in one case, in whom there were striking changes in the lumbar spinal cord, consisting of perivascular cuff-

29.2.3.3. Three types of stiffperson plus syndromes The clinical features listed in Table 4 are found in three related syndromes, that vary in the severity and distribution of pathology. The first of these stiff man plus syndromes is subacute, and has generally been termed progressive encephalomyelitis with rigidity. Histologically, it may differ from more indolent cases in the presence of demyelination. The latter characteristically spares the corticospinal tracts (Kasperek and Zebrowski, 1971; Whitely et aI., 1976; Howell et al., 1979). Clinically, the condition is characterized by widespread rigidity, painful myoclonus and spasms, and long tract and brainstem signs (Campbell and Garland, 1956; Kasperek and Zebrowski, 1971; Lhermitte et al., 1973; Whitely et al., 1976; Howell et aI., 1979). Patients survive less than three years, regardless of treatment. The relentless progression and the histology suggest a paraneoplastic etiology, and this has been confirmed in occasional cases (Batemen et al., 1990). Some authors have also suggested a viral etiology, drawing parallels with the spinal form of encephalitis lethargica (Howell et al., 1979). The remaining stiff man plus syndromes are chronic and, unlike progressive encephalomyelitis with rigidity, are remarkable for the relative absence of long tract signs, despite florid rigidity. These cases are further classified according to whether or not the clinical picture is dominated by brain stem signs. The most striking of the latter is a brainstem myoclonus that involves all four limbs, and has given rise to the term jerking stiff man syndrome (Leigh et aI., 1980). The jerks may occur in paroxysms which compromise respiration and may be fatal (Fenzi et aI., 1988). Tracheostomy and assisted ventilation may be necessary before the jerks respond to clonazepam and other antimyoclonic agents (Kullmann et aI., 1996). With this proviso, patients with the jerking stiff man syndrome may survive 10 years or more (Leigh et aI., 1980). The latter makes a paraneoplastic etiology seem unlikely, despite the fact that the basic pathological findings are similar to those found in progressive

468

P. BROWN

Fig. 3. Abnormal hand posture due to rigidity with CMUA in an anti-GAD antibody negative patient with the stiff limb variant of the stiff person syndrome.

ing, dense inflammatory infiltration of the anterior horns and diffuse astrocytosis of the surrounding gray matter (Armon et al., 1996). In addition, there was a much less marked infiltrate of the anterior horns at the cervical level, and mild perivascular cuffing in the rest of the cord, brains tern, thalamus, hippocampus and amygdala. The long tracts were spared. These findings are consistent with the idea that the stiff limb syndrome is due to pathology of the gray matter concentrated at the spinal cord level rather than the brainstem (Brown et al., 1997), a contention supported by the fact that a very similar clinical picture may arise from focal lesions that involve the central cord. The duration of the stiff limb syndrome is often measured in decades, with approximately half of cases becoming wheelchair bound (Barker et aI., 1998). Three quarters have relapses and remissions, and many have auto-antibodies, raising the possibil-

ity of an autoimmune etiology (Barker et al., 1998). Nevertheless, the autoimmune profile in these patients is reasonably distinct from that in the stiff man syndrome. Diabetes mellitus is not a feature, positive rheumatoid factor is common, and antiGAD antibodies are only found in 15% (Barker et al., 1998). Occasionally, rigidity of the limbs (sometimes with later spread to the trunk) is seen in the setting of breast or small-cell lung carcinoma. Such patients often have antibodies against the presynaptic vesicle associated 128-kD protein amphiphysin, which may have a role in synaptic vesicle endocytosis and is expresssed in breast and small-cell lung cancer tissue (De Camilli et al., 1993; Folli et aI., 1993; Dropcho, 1996). More rarely patients with tumors and rigidity may have antibodies against GAD (Silverman, 1999) or gephyrin, a cytosolic protein selectively concentrated at the postsynaptic mem-

469

STIFFNESS WITH CONTINUOUS MOTOR UNIT ACTIVITY STIFF MAN SYNDROME

STIFF LIMB SYNDROME

Fig. 4. Surface EMG recordings during spontaneous spasms in the stiff person syndrome. Left: Patient with classical axial rigidity and positive anti-GAD antibodies. Note spasm EMG is indistinguishable from that recorded in voluntary contraction. Right: Patient with rigidity of the distal lower limbs who was anti-GAD antibody negative. Spasm is confined to lower limbs and in the left (L) tibialis anterior tends to segment into large, but brief discharges. ECG artefact is arrowed. Vertical calibration is 100 f.L V and 500 f.L V for the lower and upper 4 channels respectively. Reprinted with permission from Barker et al. (1998).

brane of inhibitory synapses (Butler et al., 2000). Rarely, Borrelia burgdorferi myelitis can cause stiffness with CMUA (Martin et al., 1990). 29.2.4. Treatment of central stiffness with CMUA

Treatment of the cause, be it tetanus, focal cord pathology or paraneoplastic myelitis, should be attempted whenever possible. Around half of the stiff people with anti-amphiphysin antibody improve with prednisolone and treatment of the underlying tumor (De Camilli et al., 1993; Folli et al., 1993; Dropcho, 1996). When symptomatic treatment is necessary then oral diazepam and baclofen are usually sufficient, albeit in relatively large and frequent doses. The response of those stiff person syndrome patients with distal limb involvement tends to be less satisfactory than in those with the classical axial distribution of stiffness (Barker et al., 1998). Clonazepam, valproate, vigabatrin, gabapentin and tizanidine may occasionally be helpful. Carbamazepine and phenytoin are unhelpful, in contrast to their beneficial effects in CMUA of peripheral nerve origin. Both stiff person syndrome and stiff person-plus syndromes have been successfully treated with intrathecal baclofen delivered by pump (Stayer et al., 1997). Sudden dosage reduction

through spasm-induced rupture of the catheter, catheter dislocation or pump malfunction may, however, be fatal. Only a minority of stiff person syndrome patients have refractory disease that needs immunomodulatory treatment. Steroids often in combination with plasma exchange have been reported to be beneficial in the stiff person syndrome (Fegan, 1996), although such treatment is by no means always successful (Harding et al., 1989). Many centers now use parenteral immunoglobulin therapy as the principal treatment for severe stiff person syndrome, and its use has been supported by a randomized placebo controlled trial in 16 patients with stiff person syndrome (Dalakas et al., 2001). Two single case reports have described clinical improvement with immunoglobulin in patients with progressive encephalomyelitis with rigidity (Dropcho, 1996; Molina et al., 2000) and one case report has described improvement with immunoglobulin in the stiff limb syndrome (Souza-Lima et al., 2000). 29.2.5. Differential diagnosis of central stiffness

Spasticity due to cerebral, brainstem or spinal pathology can usually be distinguished by the velocity dependence of resistance to passive stretch, and other clinical evidence of an upper motor neuron

470

syndrome. CMUA is absent, and central motor conduction times are delayed. However, diagnostic problems can arise in patients with focal cord pathology or encephalomyelitis with rigidity, in whom tone increases often consist of a combination of stiffness with CMUA and spasticity. The rigidity seen in parkinsonism is usually readily distinguished by the presence of other clinical signs, such as bradykinesia, tremor and supranuclear ophthalmoplegia. There are, however, two extrapyramidal syndromes that can cause difficulties. The first is the neuroleptic malignant syndrome, a subacute illness characterized by generalized rigidity, autonomic dysfunction that typically follows the use of major tranquilizers or sudden withdrawal of dopaminergic treatment. It is the drug history that therefore distinguishes this from a subacute encephalomyelitis with rigidity. Second is corticobasal degeneration, which can be difficult to distinguish from the distal variant of the stiff person syndrome. Corticobasal degeneration often starts as painful rigidity in the foot and may show reflex and action induced spasm and jerks. In time, however, patients develop supranuclear ophthalmoplegias and cortical phenomena such as sensory inattention, dyspraxia and alien limb behavior. Dystonia may cause stiffness and spasm, but without the reflex spasms that characterize the stiff person syndrome. Tardive dystonia, following exposure to neuroleptics and other drugs, may often involve the axial muscles, raising the posssibility of axial stiff person syndrome. However, pain is not a prominent feature, spasm resolves in the supine position or with relaxation and there is no true CMUA. On the other hand post-traumatic dystonia is often fixed and peripheral, leading to confusion with the distal variant of the stiff person syndrome. Cases with post-traumatic dystonia are often associated with complex regional pain syndromes and have no reflex spasms. Severe rigidity may be a feature of catatonia, usually in the setting of mutism and waxy flexibility. It may complicate schizophrenia, affective disorders or occur secondary to brain pathology such as central pontine and extrapontine myelinolysis. Finally, stiffness similar to that seen in the stiff person syndrome can be psychogenic. Clinical features suggestive, but not diagnostic, of a psychogenic origin are the sudden onset, the lessening of spasm and postural abnormality when distracted

P.BROWN

and, conversely, a dramatic increase in severity during direct observation, settling again when no longer under direct observation. More convincing is the disappearance of the movement disorder when supposedly unobserved or following suggestion and placebo. Nevertheless, the diagnosis is a notoriously difficult one. The typical indicators of a conversion disorder, such as psychological precipitants, multiple somatizations and secondary gain mayor may not be present. Psychological factors are only discovered in a third of patients with somatization in general neurological practice (Mace and Trimble, 1991), and are, of course, not limited to those with psychogenic disease. Moreover, the distinction between primary psychological factors and those secondary or consequent to illness can be difficult.

29.3. Peripheral stiffness with CMUA 29.3.1. Neuromyotonia As stressed by Thompson (1994), this area has become confusing as terms used for electromyographic descriptions are often used to refer to clinical phenomena. Perhaps this could be resolved by the use of the qualifying terms clinical and discharge. Thus, clinical myokymia consists of the wave-like rippling of muscle, whereas myokymic discharge describes the finding of regular motor unit discharges. Usually the discharge is a brief, repetitive firing of single units at 2-60 Hz for a short period, repeated in the same sequence after a few seconds or so. Doublets or triplets are typical. Less commonly, the potential recurs continuously at a fairly uniform rate (1-5 Hz). Clinical neuromyotonia is the combination of clinical myokymia with delayed muscle contraction, whereas neuromyotonic discharge consists of bursts of motor unit action potentials at high frequency for a few seconds. Although the amplitude of the response typically wanes, the frequency is maintained unlike myotonic discharges. Neuromyotonic discharges may be spontaneous or follow needle insertion, voluntary contraction or ischemia or percussion of a nerve. Clinical neuromyotonia (Isaacs' syndrome) is a rare and heterogeneous syndrome of CMUA of peripheral nerve origin (Fig. 5). It manifests as various combinations of muscle stiffness, cramps, twitching, weakness and delayed muscle relaxation (Isaacs, 1961; Isaacs and Heffron, 1974). Unlike the stiff person syndrome, sleep and general anesthesia

STIFFNESS WITH CONTINUOUS MOTOR UNIT ACTIVITY

471

Fig. 6. Distal muscle wasting in a patient with neuromyotonia and peripheral neuropathy. Reprinted with permission from Lance et al. (1979).

Fig. 5. A boy with neuromyotonia (Isaac's syndrome) showing stiff posture and excessive muscular contraction, especially evidentin trapezius. Reprinted with permission from Isaac (1961). do not abolish motor hyperactivity in neuromyotonia (although muscle activity is eliminated by neuromuscular blockade). Neuromyotonia may be seen in the setting of inherited disease, where patients may or may not have clear clinical evidence of neuropathy (Lance et al., 1979; Fig. 6). Alternatively neuromyotonia may be acquired as an immune

mediated channelopathy in which autoantibodies to voltage gated potassium channels produce the peripheral motor nerve hyperexcitability that leads to clinical neuromyotonia (Newsom Davis, 1997). Some of these cases are paraneoplastic, usually occurring in association with thymoma or oat cell carcinoma of the lung (Partanen et al., 1980; Lee et al., 1998). Others are seen in association with autoimmune neuropathies (Vasilescu et al., 1984) and penicillamine treatment (Reeback et al., 1979). Although stiffness can be distal and associated with abnormal hand and foot postures, pain is not a prominent feature, reflex spasms are absent and myokymia, as well as fasciculations, are evident clinically, distinguishing clinical neuromyotonia

472

P.BROWN

from the distal limb variant of the stiff person syndrome. Profuse sweating and tachycardia may occur. The tendon reflexes are usually absent, occluded by the continuous muscle activity, but may return after successful treatment. The presence of sensory symptoms and signs depends on the nature of the underlying neuropathy. Involvement of the face may occur and is absent in the classical stiff person syndrome and exceptional in the stiff-man plus syndromes. Neuromyotonia may also be focal rather than generalized. For example, isolated ocular myotonia may be seen after radiation therapy to the brainstem and hypothalamus. Fasciculations and the grouped discharges of myokymia are evident on EMG. Prolonged bursts of motor units of normal appearance and complex repetitive discharges (sometimes referred to as bizarre high frequency discharges) may occur during attempts to relax following voluntary contraction or after electrical stimulation of motor nerves (Figs. 7 and 8). High frequency discharges may also be unprovoked. Evidence of muscle denervation and reinnervation and abnormalities of nerve conduction may be found according to the presence of an underlying neuropathy.

29.3.2. Treatment of stiffness with CMUA Clinical neuromyotonia responds well to phenytoin or carbamazepine. Some patients with acquired neuromyotonia have responded to plasma-exchange. Diazepam is not helpful, in contrast to central conditions responsible for stiffness with CMUA. The prognosis in those without underlying neuropathy or neoplasm may be good, with reports of remission after several years (Isaacs and Heffron, 1974).

29.3.3. Differential diagnosis ofperipheral stiffness Other causes of stiffness of peripheral origin are distinguished by the absence of CMUA. They also persist after nerve or neuromuscular blockade (with the exception of some cases of Schwartz-Jampel syndrome). Clinical myotonia is the delayed relaxation after muscle contraction giving rise to stiffness and limitation of movement. It is due to hyperactivity of the muscle cell membrane, and may also be seen focally after percussion of muscle. Percussion myotonia is not a clinical feature of neuromyotonia. Stiffness is particularly evident after voluntary movements, but under these circumstances

Fig. 7. A: Flexor digitorum profundus EMG in a patient with neuromyotonia during voluntary contraction for the duration of the horizontal bar. Note that thereafter there is spontaneous activity. Upper trace: 'Integrated' EMG. Lower trace: Raw surface EMG. B: High frequency discharge of a complex motor potential recorded with a needle electrode at the point of the vertical arrow in A. Upper trace: Morphology of the spontaneously discharging potential. Time calibration 10 ms. Middle section: Instantaneous frequency plot of the potential showing high frequency but irregular rate > 30 Hz. Lower trace: discharge of the spontaneous EMG potential. Time calibration 100 ms. Reprinted with permission from Lance et al. (1979).

the exacerbations generally last less than 1 min and are much briefer than in the stiff person syndrome or its variants. Myotonic discharge consists of repetitive discharges at rates of 20-80 Hz. They may be biphasic spike potentials less than 5 ms in duration and resembling fibrillation potentials or positive waves of 5-20 ms duration resembling positive sharp waves. Both are due to independent

STIFFNESS WITH CONTINUOUS MOTOR UNIT ACTIVITY

473

Fig. 8. Spontaneous EMG of flexor digitorum profundus during brief ischemia of the muscle in a patient with neuromyotonia. Note large, irregular fasiculation potential and the smaller high-frequency discharge, the amplitude of which progressively decreases. Reprinted with permission from Lance et al. (1979).

repetitive discharges of single muscle fibers and may occur after needle insertion, voluntary contraction or muscle percussion. Their most important characteristic is that the amplitude and frequency of the potentials wax and wane. This is responsible for the typical 'dive bomber' sound in the audio display of the EMG.

29.3.3.1. Schwartz-Jampel syndrome Schwartz-Jampel syndrome or chondrodystrophic myotonia is a rare autosomal recessive condition first described in 1962 (Schwartz and Jampel, 1962) and characterized by clinical myotonia and osteoarticular deformities. Patients have a short stature and a peculiar 'pinched' and fixed facial expression with narrow eyes, puckered mouth and dimpled chin caused by facial myotonia (Fig. 9). Bone dysplasias include kyphoscoliosis, platyspondyly, hip dysplasia, bowing of the leg diaphyses and irregular epiphyses. It is unclear whether these deformities are a primary defect or secondary to myotonia. The Schwartz-Jampel syndrome has three subtypes. Type 1A and B present in childhood, with osteoarticular deformities only being prominent in type lB. Type 2 is a severe and usually lethal condition with onset during pregnancy or at birth, clinically indistinguishable from the Stuve-Wiedemann syndrome. Types lA and B have been mapped to chromosome 1p35-p36.1 (Nicole et al., 1995).

It is not known at present whether the myotonia in the Schwartz-Jampe! syndrome is neurogenic or myogenic in origin, although current evidence from single fiber EMG studies (Fig. 10) suggests that most spontaneous activity is due to muscle fiber action potentials (Spaans et al., 1990; Arimura et al., 1996). The response of muscle activity to neuromuscular blockade by D-tubocurarine has been mixed. Lehmann-Hom et aI., (1990) demonstrated impaired muscle Na" channel inactivation in muscle biopsies, but the subsequent identification of the Gly1306Glu mutation in the Na" channel gene in their case established that this patient suffered from myotonia permanens rather than the SchwartzJampel syndrome. Routine nerve conduction studies are normal, but concentric needle EMG demonstrates two types of spontaneous activity. The first consists of typical myotonic high frequency discharges with increment and decrement of firing rates and amplitudes. The second type of spontaneous activity consists of high frequency discharges without variation in frequency or amplitude (Arimura et aI., 1996). The Schwartz-Jampel syndrome may respond to carbamazepine or procainamide.

29.3.3.2. Other causes ofperipheral stiffness Clinical myotonia is most commonly seen in dystrophia myotonica. This is a dominantly inherited neuromuscular disease, highly variable and multi-

474

Fig. 9. Facial appearance in a boy with Schwartz-Jampel syndrome showing blepharospasm due to myotonia when he attempts to open the eyes after forceful closure. Note also the puckered mouth and dimpled chin. Reprinted with permission from Spaans et al. (1990).

systemic, which is caused by the expansion of a CTG repeat located in the 3' untranslated region of the DMPK gene. Normal alleles show a copy number of 5-37 repeats on normal chromosomes, amplified 100100-fold on dystrophia myotonica chromosomes. Characteristic clinical features include distal atrophy, frontal balding, cataracts, testicular atrophy, diabetes and family history. More rarely myotonia is due to myotonia congenita, and distinguished by diffuse muscle hypertrophy. This can be either autosomal dominant (Thomsen's disease) or autosomal recessive (Becker form). Also rare is the cold induced myotonia of paramyotonia congenita. Creatine phosphokinase may be elevated

P.BROWN

and the EMG distinctive, with waxing and waning myotonic disharges after voluntary movement, percussion, cold, needle insertion and electrical stimulation of muscle. The metabolic myopathies such as McArdle's disease and phosphofructokinase deficiency, may also produce painful muscle cramps after and during exercise. These are electrically silent. Diagnosis is confirmed by an excessive rise in lactate during ischemic muscle exercise and by muscle biopsy. Hypothyroidism may cause stiffness and cramps, and Addison's disease has been associated with stiffness, although the exact mechanism is not clear from the literature. Acute hypertonia may be seen, together with pyrexia and rhabdomyolysis, in malignant hyperthermia, an autosomal dominant disorder triggered in susceptible people by volatile anesthetics and depolarizing skeletal muscle relaxants. Malignant hyperthermia susceptibility is usually diagnosed by the in vitro contracture test performed on fresh muscle biopsies exposed to caffeine and halothane, respectively. Around 50% of affected families are linked to the ryanodine receptor gene. The gene maps to chromosome 19q13.1 and encodes a protein that acts as a calcium-release channel from the sarcoplasmic reticulum. Finally, it should not be forgotten that stiffness can be purely mechanical in origin, due to myopathy associated muscle contracture, scleroderma or arthritis.

29.4. Conclusions Stiffness in the setting of CMUA can be due to a variety of causes, involving the central and peripheral nervous system. Prominent pain and the presence of reflex spasms usually distinguishes those with central causes. Worldwide the most important cause of central stiffness and spasm is tetanus. In developed countries most patients with a central cause for their stiffness with CMUA tum out to have either focal spinal cord pathology, such as tumor, or the stiff person syndrome. Several subgroups can be recognized in the latter. If care is taken to adhere to the diagnostic criteria summarized in Table 3, treatment response and prognosis in the classical axial stiff person syndrome are excellent. Such cases have axial rigidity and about 90% are found to have anti-GAD antibodies, suggesting that the stiff man

475

STIFFNESSWITH CONTINUOUS MOTOR UNIT ACTIVITY OFF

2

3 4

5

6

OFF

:

.. ~ .- -- - - -----~' .

A

.

:

J\':

..,.

7

',,--------~--------~'""'20""O

2mS~~

........~V AMP 1 OFF

Fig. lO. Spontaneous single fiber activity in biceps brachii of a patient with Schwartz-Jampel syndrome. Upper trace: Raw record (10 ms per division). Lower trace: Average of 50 potentials showing a typical single fiber potential (2 ms per division). Reprinted with permission from Spaans et al. (1990).

syndrome defined in this way represents a remarkably homogenous disease. Stiff people with "plus" signs are unlikely to have the classical stiff man syndrome. Those in whom pathology becomes available usually have an encephalomyelitis with prominent involvement of the gray matter. These cases can be divided into three groups according to the aggressiveness of the pathology, and its relative distribution - encephalomyelitis with rigidity, the jerking stiff person syndrome and the stiff limb syndrome. Some tum out to have a paraneoplastic syndrome, while a non-malignant autoimmune basis seems likely in others. Stiffness and spasm with CMUA may also be peripheral in origin, due to neuromyotonia. This may be inherited or acquired and mayor may not occur in the setting of evident neuropathy. Some of those without neuropathy may have an immune mediated channelopathy with autoantibodies to voltage gated potassium channels. References Arimura, K, Takenaga, S, Nakagawa, M, Osame, M and Stalberg, E (1996) Stimulation single fiber EMG study in a patient with Schwartz-Jampel syndrome. J. Neurol. Neurosurg. Psychiatry, 61: 425-426. Armon, C, Swanson, JW, McLean, JM, Westbrook, PR, Okazaki, H, Kurtin, PJ, Kalyan-Raman, UP and Rodriguez, M (1996) Subacute encephalomyelitis presenting as stiff-person syndrome: Clinical, polygraphic and pathologic correlations. Mov. Disord., 11: 701709. Barker, RA, Revesz, T, Thom, M, Marsden, CD and Brown, P (1998) A review of 23 patients affected by the stiff man syndrome: clinical subdivision into stiff trunk (man) syndrome, stiff limb syndrome and

progressive encephalomyelitis with rigidity. J. Neurol. Neurosurg. Psychiatry, 65: 633-640. Bateman, DE, Weller, RO and Kennedy, P (1990) Stiff man syndrome: a rare paraneoplastic disorder? J. Neurol. Neurosurg. Psychiatry, 53: 695-696. Brown, P, Rothwell, JC and Marsden, CD (1997) The stiffleg syndrome. J. Neurol. Neurosurg. Psychiatry, 62: 31-37. Bum, DJ, Ball, J, Lees, AJ, Behan, PO and MorganHughes, JA (1991) A case of progressive encephalomyelitis with rigidity and positive antiglutamic acid antibodies. J. Neurol. Neurosurg. Psychiatry, 54: 449-451. Butler, MH, Hayashi, A, Ohkoshi, N, Villrnann, C, Becker, CM, Feng, G, De Camilli, P and Solimena, M (2000) Autoimmunity to gephyrin in stiff-man syndrome. Neuron., 26: 307-312. Campbell, AMG and Garland, H (1956) Subacute myoclonic spinal intemeuronitis. J. Neurol. Neurosurg. Psychiatry, 19: 268-274. Dalakas, MC, Fujii, M, Li, M, Lutfi, B, Kyhos, J and McElroy, B (2001) High-dose intravenous immune globulin for stiff-person syndrome. N. Eng. J. Med., 345: 187()....1876. Davis, SM, Murray, NMF, Diengoh, N, Galea-Debono, A and Kocen, RS (1987) Stimulus-sensitive spinal myoclonus. J. Neurol. Neurosurg. Psychiatry, 50: 628-631. De Camilli, P, Thomas, A and Cofiell, R (1993) The synaptic vesicle-associated protein amphiphysin is the l28_kD autoantigen of stiff-man syndrome with breast cancer. J. Exp. Med., 178: 2219-2223. Dropcho, EJ (1996) Antiamphiphysin antibodies with small-cell lung carcinoma and paraneoplastic encephalomyelitis. Ann. Neurol., 39: 659-667. Fenzi, F, Bongiovanni, G, Fincati, E, Pampanin, M, Tomelleri, G and Rizzuto, N (1988) Anatomical and clinical study of a case of subacute encephalomyelitis

476 with hyperekplexia syndrome. Ital. J. Neurol. Sci., 9: 505-508. Fogan, L (1996) Progressive encephalomyelitis with rigidity responsive to plasmapheresis and immunosuppression. Ann. Neurol., 40: 451-453. Folli, F, Solimena, M, Cofieli, R, Austoni, M, Tallini, G, Fasseta, G, Bates, D, Cartlidge, N,Bottazzo, GF, Piccolo, G and De Camilli, P (1993) Autoantibodies to a 128_kd synaptic protein in three women with the stiffman syndrome and breast cancer. N. Engl. J. Med., 328: 546-551. Gelfan, S and Tarlov, 1M (1959) Interneurons and rigidity of spinal origin. J. Physiol., 146: 594-617. Gelfan, Sand Tarlov, 1M (1963) Altered neuron population in L7 segment of dogs with experimental hind-limb rigidity. Am. J. Physiol., 205: 606-616. Guilleminault, C, Sigwald, J and Castaigne, P (1973) Sleep studies and therapeutic trial with L-DOPA in a case of stiff man syndrome. Eur. Neurol., 10: 89-96. Habermann, E (1978) Tetanus. In: PJ Vinken and GW Bruyn (Eds.), Handbook of Clinical Neurology (Vol 33). Elsevier, Amsterdam, pp. 491-547. Hardin, JA and Griggs, RC (1971) Diazepam in a case of strychnine poisoning. Lancet, ii: 372-373. Harding, AE, Thompson, PD, Kocen, RS, Batchelor, JR, Davey, N and Marsden, CD (1989) Plasma exchange and immunosuppression in the stiff-man syndrome. Lancet, ii: 915. Howell, DA, Lees, AJ and Toghill, PJ (1979) Spinal internuncial neurons in progressive encephalomyelitis with rigidity. J. Neural. Neurosurg. Psychiatry, 42: 773-785. Isaacs, H (1961) A syndrome of continuous muscle-fiber activity. J. Neurol. Neurosurg. Psychiatry, 24: 319325. Isaacs, H and Heffron, JJA (1974) The 'syndrome of continuous muscle-fiber activity' cured: further studies. J. Neurol. Neurosurg. Psychiatry, 37: 2131-2135. Ishizawa, K, Komori, T, Okayama, K, Qin, X, Kaneko, K, Sasaki, S and Iwata, M (1999) Large motor neuron involvement in Stiff-man syndrome: a qualitative and quantitative study. Acta. Neuropathol. (Berl.), 97: 6370. Kasperek, S and Zebrowski, S (1971) Stiff-man syndrome and encephalomyelitis: Report of a case. Arch. Neurol., 24: 22-31. Kullmann, DM, Howard, RS, Miller, DH, Hirsch, NP, Brown, P and Marsden, CD (1996) Brainstem encephalopathy with stimulus-sensitive myoclonus leading to respiratory arrest - a description of two cases and review of the literature. Mov. Disord., 11: 715-718. Lance, JW, Burke, DE and Pollard, J (1979) Hyperexcitability of motor and sensory neurons in neuromyotonia. Ann. Neurol., 5: 523-532.

P.BROWN

Lee, EK, Maselli, RA, Ellis, WG and Agius, MA (1998) Morvan's fibrillary chorea: a paraneoplastic manifestation of thymoma. J. Neurol. Neurosurg. Psychiatry, 65: 857-862. Lehmann, HF, Iaizzo, PA, Franke, C, Hatt, H and Spaans, F (1990) Schwartz-Jampe! syndrome: II. Na+ channel defect causes myotonia. Muscle Nerve, 13: 528-535. Leigh, PN, Rothwell, JC, Traub, M and Marsden, CD (1980) A patient with reflex myoclonus and muscle rigidity: "jerking stiff-man syndrome." J. Neural. Neurosurg. Psychiatry, 43: 1125-1131. Lhennitte, F, Chain, F, Escourolle, R, Chedru, F, Guilleminault, G and Francoual, M (1973) Un nouveau cas de contracture tetaniforme distinct du "stiff man syndrome": etude pharmacologique et neuropathologique d'un cas d'encephalomyelite a predominance medullaire. Rev. Neural. (Paris), 128: 3-21. Lorish, TR, Thorsteinsson, G and Howard, PM (1989) Stiff-man syndrome updated. Mayo. Clin. Proc., 64: 629-636. Lourie, H (1968) Spontaneous activity of alpha motor neurons in intramedullary spinal cord tumor. J. Neurosurg., 29: 573-580. Mace, CJ and Trimble, MR (1991) "Hysteria", "functional" or "psychogenic"? A survey of British neurologists preferences. J. R. Soc. Med., 84: 471-475. Martin, R, Meinck, HM, Schulte-Mattler, W, Ricker, K and Mertens, HG (1990) Borrelia burgdorferi myelitis presenting as partial stiff man syndrome. J. Neurol., 237: 51-54. Martinelli, P, Pazzaglia, P, Montagna, P, Coccagna, G, Rizzuto, N, Simonati, Sand Lugaresi, E (1978) Stiffman syndrome associated with nocturnal myoclonus and epilepsy. J. Neural. Neurosurg. Psychiatry, 41: 458-462. Meinck, H-M, Ricker, K and Conrad, B (1984) The stiffman syndrome: new pathophysiological aspects from abnormal exteroceptive reflexes and the response to clomipramine, clonidine, and tizanidine. J. Neural. Neurosurg. Psychiatry, 47: 280-287. Meinck, HM, Ricker, K, Htilser, PJ, Schmid, E, Peiffer, J and Solimena, M (1994) Stiff man syndrome: clinical and laboratory findings in eight patients. J. Neurol., 24: 157-166. Moersch, FP and Woltman, HW (1956) Progressive fluctuating muscular rigidity and spasm ("stiff-man syndrome"): Report of a case and some observations in 13 other cases. Mayo. Clin. Proc., 31: 421-427. Molina, JA, Porta, J, Garcia-Morales, I, Bermejo, F and Jimenez-Jimenez, FJ (2000) Treatment with intravenous prednisolone and immunoglobulin in a case of progressive encephalomyelitis with rigidity. J. Neurol. Neurosurg. Psychiatry, 68: 395.

STIFFNESS WITH CONTINUOUS MOTOR UNIT ACTIVITY

477

Nemni, R, Camerlingo, M, Fazio, R, Casto, L, Quattrini, A, Mamoli, D, Lorenzetti, I, Canal, N and Mamoli, A (1993) Serum antibodies to Purkinje cells and dorsal root ganglia in sensory neuronopathy without malignancy. Ann. Neurol., 34: 848-854. Nernni, R, Braghi, S, Natali-sora, MG et al. (1994) Autoantibodies to glutamic acid decarboxylase in palatal myoclonus and epilepsy. Ann. Neurol., 36: 665-667. Newsom Davis, J (1997) Autoimmune neuromyotonia (Isaacs' syndrome): an antibody-mediated potassium channelopathy. Ann. NY Acad. Sci., 835: I I 1-119. Nicole, S, Ben Hamida, C, Beighton, P, Bakouri, S, Belal, S, Romero, N, Viljoen, D, Ponsot, G, Sammoud, A, Weissenbach, J, Fardeau, M, Ben Hamida, M, Fontaine, B and Hentati, F (1995) Localization of the Schwartz-Jampel syndrome (SJS) locus to chromosome Ip34-p36.l by homozygosity mapping. Hum. Mol. Genet., 4: 1633-1636. Partanen, VSJ, Soininen, H, Saksa, M et al. (1980) Electromyographic and nerve conduction findings in a patient with neuromyotonia, normocalcemic tetany and small-cell lung cancer. Acta. Neurol. Scand., 6I: 216-226. Reeback, J, Benton, S, Swash, M et al. (1979) Penicillamine induced neuromyotonia. BMJ, I: 14641465. Roobol, TH, Kazzaz, BA and Vecht, CHJ (1987) Segmental rigidity and spinal myoclonus as a paraneoplastic syndrome. J. Neurol. Neurosurg. Psychiatry, 50: 628631. Rushworth, G, Lishman, WA, Trevor Hughes, J and Oppenheimer, DR (1961) Intense rigidity of the arms due to isolation of motomeurons by a spinal tumor. J. Neurol. Neurosurg. Psychiatry, 24: 132-142. Saiz, A, Arpa, J, Sagasta, A, Casamitjana, R, Zarranz, Jl, Tolosa, E and Graus, F (1997) Autoantibodies to glutamic acid decarboxylase in three patients with cerebellar ataxia, late-onset insulin dependent diabetes mellitus, and polyendocrine autoimmunity. Neurology, 49: 1026-1030. Saiz, A, Graus, F,Valldeoriola, F,Valls-Sole,J and Tolosa, E (1998) Stiff-leg syndrome: a focal form of the stiffman syndrome. Ann. Neurol., 43: 400-403. Schwartz, 0 and Jampel, RS (1962) Congenital blepharophimosis associated with a unique generalized myopathy. Arch. Ophthalmol., 68: 82-87.

Sikes, ZS (1959) Stiff-man syndrome. Dis. Nerv. System, 29: 254-258. Silverman, IE (1999) Paraneoplastic stiff limb syndrome. J. Neurol. Neurosurg. Psychiatry, 67: 126-127. Solimena, M, Folli, F, Aparisi, R, Pozza, G and De Camilli, P (1990) Autoantibodies to GABA-ergic neurons and pancreatic beta cells in stiff-man syndrome. N. EngLJMed., 322: 1555-1560. Souza-Lima, CPL, Ferraz, HB, Braz, CA, Araujo, AM and Manzano, GM (2000) Marked improvement in a stifflimb patient treated with intravenous immunoglobulin. Mov. Disord., 15: 358-359. Spaans, F, Theunissen, P, Reekers, AD, Smit, L and Veldman, H (1990) Schwartz-Jampel syndrome: I. Clinical, electromyographic, and histologic studies. Muscle Nerve, 13: 516-527. Spehlmann, R and Norcross, K (1979) Stiff-man syndrome. In: HL Klawans (Ed.), Clinical Neuropharmacology (Vol. 4). Raven Press, New York, pp. 109-121. Stayer, C, Tronnier, V, Dressnandt, J, Mauch, E, Marquardt, G, Rieke, K, Muller-Schwefe, G, Schumm, F and Meinck, HM (1997) Intrathecal baclofen therapy for stiff-man syndrome and progressive encephalomyelopathy with rigidity and myoclonus. Neurology, 49: 1591-1597. Tarlov, 1M (1967) Rigidity in man due to spinal interneuron loss. Arch. Neurol., 16: 536-543. Thompson, PD (1994) Stiff people. In: CD Marsden and S Fahn (Eds.), Movement Disorders 3. ButterworthHeinemann, Oxford, pp. 373-405. Trethowan, WH, Allsop, JL and Turner, B (1960) The stiff man syndrome. Arch. Neurol., 3: 448-456. Vasilescu, C, Alexianu, M and Dan, A (1984) Muscle hypertrophy and a syndrome of continuous motor unit activity in prednisolone: responsive Guillain-Barre polyneuropathy. J. Neurol., 231: 276-279. Warich-Kirches, M, Von Bossanyi, P,Treuheit, T, Kirches, E, Dietzmann, K, Feistner, H and Wittig, H (1997) Stiff-man syndrome: possible autoimmune etiology targeted against GABA-ergic cells. Clin. Neuropath., 16: 214-219. Whiteley,AM, Swash, M and Urich, H (1976) Progressive encephalomyelitis with rigidity: Its relation to 'subacute myoclonic spinal intemeuronitis' and the 'stiff man syndrome.' Brain, 99: 27-42.

Movement Disorders Handbook of Clinical Neurophysiology, Vol. 1 M. Hallett (Ed.) © 2003 Elsevier B. V. All rights reserved

479 CHAPTER 30

Hyperekplexia P. Brown* Sobel! Department of Motor Neuroscience and Movement Disorders, Institute of Neurology, London WCI N 3BG, UK

30.1. Introduction Hyperekplexia, or startle disease, is derived from the Greek word "eK-1T}.TJO'O'W" which means 'to startle excessively' (Suhren et al., 1966). The abnormal startle consists of an exaggerated response to unexpected stimuli, particularly sounds. The classification of the startle disorders has hitherto largely relied on electrophysiological criteria, and its clinical utility has therefore been limited. Table 1 gives a clinical classification of the abnormal startle. In this, patients are separated according to whether the clinical picture is dominated by brief body jerks that clinically seem to follow the stimulus almost

immediately or by spasms that are of visibly longer latency and last a second or more. Often these spasms follow a normal or exaggerated jerk to the stimulus. This chapter will focus on one cause of short latency body jerks following unexpected stimuli; hyperekplexia, which may be hereditary or sporadic. Upon examination the hallmark ofhyperekplexia is a brief body jerk of short latency following unexpected stimuli. Such stimuli may be visual, auditory or somesthetic. Somesthetic stimuli are most effective when applied to the mantle area, particularly the face, when the response that results is sometimes termed a head retraction reflex.

Table I

30.2. Hereditary hyperekplexia

Classification of the abnormal clinical startle response.

30.2.1. Clinical aspects

* Correspondence to: P. Brown, Sobel! Department of Neurophysiology, Institute of Neurology, London WCIN 3BO, UK. E-mail address:[email protected]

Hereditary hyperekplexia was first described by Kirstein and Silfverskiold in 1958. A detailed and landmark description of a large Dutch family followed in 1966 (Suhren et aI., 1966). The clinical picture is characterized by three major features. The first is generalized stiffness immediately after birth, remitting during the first years of life. The stiffness increases with handling and disappears during sleep. It is likely that the so-called "hereditary stiff-baby syndrome" (Klein et aI., 1972; Sander et aI., 1980) is the early presentation of hereditary hyperekplexia (Lingam et aI., 1981; Weaver et aI., 1982). Second, patients suffer from an excessive startle reflex, to unexpected, particularly auditory, stimuli. The severity and frequency of the excessive startle reflex can increase due to nervousness, fatigue, and the expectation of being frightened. Third, is a temporary generalized stiffness following the startle response, causing patients to fall 'as stiff as a stick' (Suhren et al., 1966). Attacks of stiffness last a few seconds (Fig. 1) and are by no means elicited by every stimulus presentation. Consciousness is preserved.

480

P.BROWN

Fig. 1. EMG record of the tonic spasm occurring after an unexpected sound in a patient with symptomatic hyperekplexia. The tonic spasm starts about 2 s after the stimulus, and is clearly separate from the very rapid and brief startle response (seen 2 s into the trace). left (L), right (R). Reprinted from Brown et al., 1991a with by permission.

Nevertheless, these spasms frequently culminate in injury. These tonic responses are distinct from the tonic spasms of multiple sclerosis, which are usually painful, unilateral and rarely stimulus sensitive. Several additional clinical features have been described in patients with hereditary hyperekplexia, although these are not obligatory for the diagnosis. Periodic limb movements in sleep and hypnagogic myoclonus are frequently reported (Kirstein and Silfverskiold, 1958; Gastaut and Villeneuve, 1967; Andermann et al., 1980; Morley et al., 1982; Kurczynski, 1983; Saenz et al., 1984; Brown et al., 1991a; Hayashi et al., 1991; Shahar et al., 1991; Matsumoto et al., 1992; Pascotto and Coppola, 1992). Other features are inguinal, umbilical or epigastric herniae (Suhren et al., 1966; Klein et al., 1972; Lingam et al., 1981), congenital dislocations of the hip (Hayashi et al., 1991), epilepsy (Suhren et al., 1966; Saenz et al., 1984), feeding and breathing problems in newborns (Shahar et al., 1991; Giacoia and Ryan, 1994), and sudden infant death (Suhren et al., 1966; Giacoia and Ryan, 1994). Most patients have normal intelligence, but some mildly retarded patients have been reported (Shahar et al., 1991; Ryan et al., 1992). On neurological examination, neonates have a markedly increased muscle tone and head retraction reflex. The baby is alert, but shows marked hypoki-

nesia (Tijssen and Brouwer, 1999). Adults with the major form of hyperekplexia often walk with a stifflegged, mildly wide-based gait without signs of ataxia (Suhren et al., 1966). Tendon reflexes and tone are normal, or slightly increased, without clear evidence of a pyramidal syndrome (Suhren et al., 1966). Tapping the nose induces an exaggerated head-retraction reflex in most patients (Suhren et al., 1966; Andermann et al., 1980; Kurczynski, 1983; Shahar et al., 1991; Matsumoto et al., 1992). This consists of a brisk, involuntary backward jerk of the head. It occurs only infrequently in normal subjects (Wartenberg, 1941). 30.2.2. Genetics of hereditary hyperekplexia

Linkage analysis has mapped a major gene for this disorder to chromosome 5q33-35 (Ryan et al., 1992). Different missense mutations in the GLRAI (glycine receptor) gene, Pr0250Thr (Saul et al., 1999), Gln266His (Milani et al., 1996), Arg271Leu (Shiang et al., 1993), Arg271Gln (Shiang et al., 1993; Rees et al., 1994; Schorderet et al., 1994; Shiang et al., 1995; Tijssen et al., 1995; Bernasconi et al., 1996; Elmslie et al., 1996), Lys276Glu (Seri et al., 1997), and Tyr279Cys (Shiang et al., 1995), have been identified in families with the autosomal dominant form of hyperekplexia. Besides the domi-

HYPEREKPLEXIA

nant form of hyperekplexia, two recessive cases, both offspring of consanguineous parents, have been described (Rees et al., 1994; Brune et al., 1996). In one patient a recessive point mutation, Ile244Asn (Rees et al., 1994), was detected while the other patient carried a homozygous deletion encompassing exon 1 to 6 of the GLRA1 gene (Brune et al., 1996). In a further family, two patients compound heterozygous for two mutations, Arg252His and Arg392His, showed the hyperekp1exia phenotype (Vergouwe et al., 1999). Glycine is an inhibitory neurotransmitter and the glycine receptor is a hetero-oligomeric ligand-gated chloride channel, mainly located in post-synaptic membranes in vertebrate brainstem and spinal cord (Betz, 1991). A high concentration of glycine receptors is found in the intemeurons; particularly Renshaw cells and Ia inhibitory intemeurons (Fyffe, 1991). The receptor consists of five subunits (30: and 213) forming a ring with a central ion-conducting pore. The glycine receptor has different isofonns, comprising different variants of the ligand-binding 0: subunits. The mutations described in hyperekplexia are located in the alpha-l subunit. This consists of a large extracellular N-terminal domain, four transmembrane segments (M1-M4) and a short extracellular C terminus (Grenningloh et al., 1990; Rajendra et al., 1995). Autosomal dominant mutations in the GLRAI gene, located within the intracellular M1-M2 loop, M2 domain or the extracellular M2-M3 loop, disrupt signal transduction. Normally, chloride influx through the channel antagonizes membrane depolarization. Decreased chloride permeability of the neuronal membrane therefore results in diminished inhibition of neuronal firing. 30.2.3. The minor form

In the original Dutch family, described in 1966, two clinical forms of the disorder were recognized (Suhren et al., 1966). These two forms have also been described in a Canadian pedigree and named the major and minor form of hyperekplexia (Andermann et al., 1980). The major form is characterized by the triad of neonatal stiffness, excessive startle and tonic spasms described above. The minor form consists of excessive startle responses without any signs of stiffness, neither in relation to the startle response, nor in the neonatal period. The occasional

481

occurrence of the minor form has been confirmed in other families (Brown et al., 1991a; Pascotto and Coppola, 1992), and a large pedigree with 10 affected patients has been described, suffering from both major and minor forms of hyperekplexia, although clinical details were scanty (Saul et al., 1999). Opinion is presently divided as to whether the minor form represents a variation in expression of the same gene defect as the major form. In particular, in the Dutch pedigree described by Suhren, only family members with tonic spasms were found to carry the mutation (Tijssen et al., 1995, 1996), suggesting that in at least some families the 'minor' form may be a partial phenocopy. 30.3. Sporadic hyperekplexia Sporadic cases of hyperekplexia do not seem to have a genetic basis (Gastaut and Vileneuve, 1967; Vergouwe et al., 1997); indeed many are symptomatic and usually a consequence of brainstem pathology such as infarct, hemorrhage or encephalitis (Table 2). They too may exhibit tonic spasms and episodes of sustained myoclonic jerks which are therefore not unique to hereditary hyperekplexia (Gastaut and Vileneuve, 1967; Brown et al., 1991a). 30.4. Differential diagnosis Hyperekplexia is distinguished from the normal startle reflex by its lower threshold, greater extent and resistance to habituation (Brown et al., 1991a; Chokroverety et al., 1992; Matsumoto et al., 1992). The normal startle response rarely involves the lower limbs in a sitting subject, whereas this is almost always the case in hyperekplexia. The normal startle response habituates within 1-5 trials of auditory stimulation repeated every 20 s or so, leaving only an auditory blink reflex, whereas in hyperekplexia extensive jerks persist. The pathological differential diagnosis is as follows: 30.4.1. Excessive responses to startling stimuli without stiffness in between 30.4.1.1. Startle-induced epilepsy Startle epilepsy, as described by Alajouanine and Gastaut (1955) and reviewed by Chauvel et al.

482

(1992), can readily be distinguished from hyperekplexia, although similar pathophysiological mechanisms may operate in the tonic episodes of each condition. Startle epilepsy is most often seen in the setting of early brain damage, usually perinatal anoxia. Most patients have a hemiparesis, and mental retardation is common. However, these seizures may also occur without infantile hemiplegia (Manford et al., 1996). Seizures begin in childhood or adolescence, and tend to be frequent. They consist of tonic spasms, lasting up to 30 s, with preservation of consciousness. The spasms are typically asymmetrical and predominantly involve the paretic limbs. They may be elicited by unexpected auditory, visual or somesthetic stimulation, but, in contrast to Table 2 Causes of symptomatic hyperekplexia. J. Static encephalopathies Static perinatal encephalopathy without tonic spasms (Shiamura, 1973) Post-traumatic encephalopathy (Duensing, 1952; Krauss et al., 1997) Post-anoxic encephalopathy (Brown et aI., 1991a)

2. Brainstem encephalitis Paraneoplastic (Duensing, 1952) Sarcoidosis (Brown et aI., 1991a) Jerking stiff person syndrome (Leigh et al., 1980; Brown et aI., 1991a) Viral encephalomyelitis (Fenzi et al., 1988) 3. Demyelination (Duensing, 1952; Brown et aI., 1991a) 4. Vascular lesions Occlusion of the posterior thalamic arteries (Fariello et al., 1983) Brainstem hemorrhage/infarct (Duensing, 1952; Kohara et aI., 1988; Shibasaki et al., 1988; Kimber and Thompson, 1997) 5. Structural lesions Cervico-medullar compression (Winston, 1983) 6. Gilles de la Tourette syndrome (Stell et aI., 1995) 7. Post-traumatic stress disorder (Howard and Ford, 1992)

P.BROWN

hyperekplexia, are also spontaneous. Other seizure types occur in about a quarter of patients (Chauvel et al., 1992). Cranial imaging is abnormal in the majority of patients, usually showing unilateral atrophy involving the lateral central and pericentral cortex. The interictal EEG is generally abnormal with localized or diffuse slow waves and spikes. Ictal scalprecorded EEG shows a fast low amplitude discharge often preceded by a high voltage spike at the vertex. Using depth electrodes the tonic seizures have been shown to originate in the motor or supplementary motor cortex (Bancaud et al., 1967; Chauvel et al., 1992). 30.4.1.2. Reflex myoclonus The clinical features distinguishing hyperekplexia from brainstem reticular reflex myoclonus are summarized in Table 3. Propriospinal myoclonus can cause reflex axial jerks but the face is always spared, and sensitivity to sound is rarely seen. Both brainstem reticular reflex and propriospinal myoclonus have frequent spontaneous jerks that also serve to distinguish them from hyperekplexia (Hallett et al., 1977; Brown et al., 1991b). 30.4.1.3. Culture-bound syndromes The "Jumping Frenchmen of Maine" (Stevens, 1965; Howard and Ford, 1992), Latah (Yap, 1951) and Myriachit (Yap, 1951) are culture-bound syndromes including excessive startle responses, echolalia, and echopraxia. Stiffness has not been described in these patients. 30.4.1.4. Gilles de la Tourette syndrome In Gilles de la Tourette syndrome excessive startling has occasionally been described (Stell et al., 1995), but more frequently a normal startle reflex induces motor tics and obsessive compulsive behavior (Lees et al., 1984; Eapen et al., 1994; Tijssen et al., 1999). 30.4.1.5. Conversion disorder Distinguishing psychogenic jerks from hyperekplexia may be difficult. The former are suggested by "la belle indifference," distractability, suggestibility and an inconsistent pattern of EMG activity in the jerks (Thompson et al., 1992). The latency of reflex responses can be very long (> 100 ms).

483

HYPEREKPLEXIA

Table 3 Differential diagnosis of hyperekplexia and brainstem reticular reflex myoclonus. Jerks to auditory stimuli

Hyperekplexia' Brainstem reticular reflex myoclonus

+ +

Greatest stimulus sensitivity to taps

Presence of spontaneous and action-induced jerks

Duration of individual EMG bursts'

+

<75ms

Mantle area' Distal limb

>75 ms

I Some, but not all of these patients have infrequent tonic spasms and a family history of startle which also serve to distinguish them from brainstem reticular reflex myoclonus. 2 The head and upper chest and back. 3 Note that this refers to the duration of individual EMG bursts; bursts can be repetitive in brainstem reticular reflex myoclonus.

30.4.2. Excessive responses to startling stimuli with stijTness in between 30.4.2.1. Stiff-person syndrome (stiff-man syndrome) Stiff-person syndrome (Barker et al., 1998) is characterized by spasms and stiffness of the axial muscles. The spasms can be induced by unexpected stimuli. The face is spared unlike the startle of hyperekplexia. The axial stiffness is continuous and usually associated with an exaggerated lumbar lordosis. Antibodies against glutamic acid decarboxylase occur in most patients. 30.4.2.2. Tetanus and strychnine poisoning These can be distinguished from hyperekplexia by their acute course. In addition, pronounced trismus and autonomic instability are prominent features in acute tetanus. 30.4.2.3. Continuous stiffness in the neonatal period Here the most important differential diagnosis is perinatal asphyxia (Aicardi, 1992). Pyramidal signs and irritability occur, unlike in hyperekplexia. Congenital generalized muscle hypertonia has also been described in a Mexican family as an autosomal recessive disorder (Cantu and Cuellar, 1974). These children also suffered from cardiopulmonary distress. Extrapyramidal signs, including stiffness, may occur in children born to mothers using phenothiazines (Hill et al., 1966) or cocaine (Chiriboga et al., 1995).

30.5. Treatment

Several drugs have been used in hyperekplexia: diazepam (Klein et al., 1972; Morley et al., 1982), clobazepam (Brown et al., 1991a), chlordiazepoxide (Suhren et al., 1966), carbamazepine (Brown et al., 1991a), phenytoin (Brown et al., 1991a), valproate (Saenz et aI., 1984; Dooley and Andermann, 1989; Dubowitz et aI., 1992), 5-hydroxytryptophan (Saenz et al., 1984), piracetam (Saenz et al., 1984) and phenobarbital (Andermann et aI., 1980; Dubowitz et aI., 1992). However, the most effective drug is clonazepam (Andermann et al., 1980; Morley et al., 1982; Markand et al., 1984; Saenz et al., 1984; Pascotto and Coppola, 1992; Ryan et al., 1992; Tijssen et al., 1997). Clonazepam binds to specific high affinity binding sites for benzodiazepines and potentates the inhibitory neurotransmitter gammaaminobutyric acid, although this may not be its only action. Vigabatrin proved ineffective in a doubleblind placebo-controlled study (Tijssen et aI., 1997). 30.6. Physiology

The startle response in hyperekplexia has variously been considered to have a cortical or brainstem origin, with most current evidence favoring the latter. Two lines of evidence have been used to support a cortical origin for the startle response - the presence of giant cortical evoked potentials (Markand et al., 1984) and the presence of cortical neuronal loss on magnetic resonance spectroscopy (Bernasconi et aI., 1998). However, giant evoked

484

P.BROWN

potentials are only present in the minority of patients and the study that demonstrated cortical neuronal loss did so in patients with epilepsy and without a known mutation (Bernasconi et al., 1998). Magnetic resonance spectroscopy has been repeated in patients with a mutation in the alpha) subunit of the glycine receptor and has been found to be normal (Tijssen et al., 2000). The same patients were also tested with transcutaneous stimulation of the motor cortex and found to have normal stimulus response curves, cortical inhibition and facilitation. On the other hand there are several lines of evidence pointing to a brainstem origin for the startle response in hyperekplexia: (l) Pathology in symptomatic cases is often con-

fined to the brainstem (Brown et al., 199Ia). (2) Familial cases are due to mutations in the alpha, subunit of the glycine receptor and glycine receptors are particularly concentrated in the brainstem (and spinal cord) of the mammalian CNS (Betz, 1991; Shiang et al., 1993). (3) Slowing of horizontal saccadic eye movements may be found in familial hyperekplexia and is evidence of disturbed activity in the pontine reticular formation (Tijssen et al., 1995). (4) The latency of EMG responses to taps to the head/face is often <20 ms (Fig. 2) and is in these cases only compatible with relay within the brainstem (Brown et al., 1991a; Matsumoto et al., 1992). (5) The recruitment of cranial nerve innervated muscles is caudo-rostral in the startle, as in brainstem reticular reflex myoclonus (Brown et al., 1991a). The caudo-rostral recruitment of cranial nerve innervated muscles in the startle is the most contentious observation supporting a brainstem origin for the hyperekplectic startle response. There is little doubt that the shortest latency responses to taps to the face may follow this rule, so that the earliest muscle activity is recorded in sternocleidomastoid and later in orbicularis oculi and masseter (Fig. 2). However, the picture is more complicated in the startles of longer latency that follow auditory stimulation (or responses to taps in some patients). Here the earliest EMG activity is in orbicularis oculi, with sternocleidomastoid, then masseter following (Fig. 3). But this early response in orbicularis oculi may be due to a simultaneously induced physio-

Fig. 2. EMG activity in the abnormal startle response elicited by taps to the head in a patient with symptomatic hyperekplexia. The unrectified EMG activity in three single trials is superimposed. Each trial was started at the point of tapping. EMG activity was recorded first in sternocleidomastoid, and then later in orbicularis oculi, masseter, trunk and limb muscles. The latencies to the intrinsic hand muscles of the hand and foot were disproportionately long. The vertical and horizontal calibration lines are 0.5 mV and 20 ms, respectively. Reprinted from Brown et aI., 1991a with permission.

logical blink reflex and separate from the true generalized startle response in this muscle. Certainly two components in the EMG response in this muscle are common (Fig. 4), one perhaps due to the blink reflex and a later one due to the pathological startle reflex (Colebatch et al., 1990; Brown et al., 1991a; Chokroverety et al., 1992; Matsumoto et al., 1992). There are other features in the recruitment order of muscles in the startle of hyperekplexia that merit comment. EMG activity in trunk and limb muscles follows that in sternocleidomastoid, usually at intervals that are a few milliseconds longer than those

HYPEREKPLEXIA

485

Fig. 4. EMG during startles in a patient with idiopathic exaggerated startle following a tap to the back. Traces begin at the time of stimulus presentation. Note the two components in orbicularis oculi (0.0). Trapezius (Trap) EMG saturates at 2 mV and gain for deltoid (delt), biceps (bic), triceps (tric) and forearm flexors (ER.) is similar. 0.0. gain is doubled. The horizontal calibration lines are in ms. Reprinted from Colebatch et aI., 1990 with permission.

Fig. 3. EMG acnvity in the abnormal startle response elicited by auditory stimulation in a patient with symptomatic hyperekplexia. The unrectified EMG activity in three single trials is superimposed. Each trial was started at the point of presentation of a 124 dB tone. Following the normal auditory blink reflex, EMG activity was recorded first in sternocleidomastoid, and then later in masseter and trunk and limb muscles. The latencies to the intrinsic hand muscles of the hand and foot were disproportionately long. The horizontal calibration line is 20 ms. Vertical calibration is as Fig 2, except for the lower five channels for which the line in Fig 2 represents 3 mV. Reprinted from Brown et aI., 1991a with permission.

seen after synchronous activation of muscles through the pyramidal tract, either as in cortical myoclonus or as follows transcutaneous stimulation of the motor cortex (Brown et al., 1991a). The responses recorded in the intrinsic hand and foot muscles (Figs. 2 and 3) are particularly delayed relative to

more proximal limb muscles (Brown et al., 1991a; Matsumoto et al., 1992). The pattern of muscle recruitment is therefore similar to that established for the physiological startle reflex in the human, which, like that in animals, is thought to arise in the caudal reticular formation of the brainstem, particularly within the nucleus reticularis pontis caudalis (Davis et al., 1982; Brown et al., 1991c). Studies of the recruitment order of muscles in the startle response must be carefully controlled, as stimulus parameters such as intensity, repetitition rate and site may all have an influence on the absolute latency of the response (Matsumoto et al., 1992). Even more importantly the patient's posture can have a profound effect on the pattern of the startle in hyperekplexia (Brown et al., 199Id). Postural and other influences may operate independently on different components of a series of waves of bulbospinal activity, resulting in reflex responses of differing latency. Variation in absolute latency, despite preservation of the overall pattern of muscle

486

recruitment, is a distinctive feature of both the normal and hyperekplectic startle response. The abnormal startle response is not the only motor phenonenon in hyperekplexia. The pathophysiology of the tonic spasms in hyperekplexia is not known. They may be a brainstem phenomena or the result of activity in the motor and supplementary motor cortex, as in startle epilepsy. On the other hand the stiff gait seen in hyperekplexia may relate to changes in the spinal cord. The first period of spinal reciprocal inhibition is deficient in such patients, consistent with its mediation by the glycinergic Ia inhibitory interneuron (Floeter et aI., 1996), and spinal flexor reflexes may be exaggerated (Matsumoto et aI., 1992). 30.7. Conclusion In summary, hyperekplexia is characterized by an abnormal startle response that is relayed in the brainstem and likely involves a pathological exaggeration of the physiological startle reflex. In hereditary hyperekplexia, this, the startle-induced stiffness, and other features are the result of a lack of interneuronal inhibition in the brainstem and spinal cord due to the reduced chloride permeability of the mutated glycine receptor. In symptomatic hyperekplexia, features arise from functional disturbance of the brainstem startle relay. References Aicardi, J (1992) Neurological diseases in the perinatal period. In: J Aicardi (Ed.), Diseases of the Nervous System in Childhood. Lavenham Press, Suffolk, pp.47-105. Alajouanine, T and Gastaut, H (1955) La syncinesiesursaut et l' epilepsie-sursaut a declanchement sensoriel ou sensitif inopineI Les faits anatomo-cliniques (15 observations). Rev. Neurol., 93: 29-41. Andermann, F, Keene, DL, Andermann, E and Quesney, LF (1980) Startle disease or hyperekplexia; further delineation of the syndrome. Brain, 103: 985-997. Bancaud, J, Talairach, J and Bonis, A (1967) Physiopathogenie des epilepsies-sursaut: a propos d'une epilepsie de I'aire motrice supplementaire. Rev. Neurol., 117: 441-453. Barker, RA, Revesz, T, Thorn, M, Marsden, CD and Brown, P (1998) A review of 23 patients affected by the stiff man syndrome: clinical subdivision into stiff trunk (man) syndrome, stiff limb syndrome and

P. BROWN

progressive encephalomyelitis with rigidity. J. Neural. Neurosurg. Psychiatry, 65: 633-640. Bernasconi, A, Regli, F, Schorderet, DF and Pescia, G (1996) Familial hyperekplexia: startle disease. Clinical, electrophysiological and genetic study of a family. Rev. Neurol., 152: 447-450. Bernasconi, A, Cendes, F, Shoubridge, EA, Andermann, E, Li, LM, Arnold, DL and Andermann, F (1998) Spectroscopic imaging of frontal neuronal dysfunction in hyperekplexia. Brain, 121: 1507-1512. Betz, H (1991) Glycine receptors: heterogeneous and widespread in the mammalian brain. Trends Neurosci., 14: 458-461. Brown, P, Thompson, PD, Rothwell, JC, Britton, TC, Day, BL and Marsden, CD (1991a) The hyperekplexias and their relationship to the normal startle reflex. Brain, 114: 1903-1928. Brown, P, Thompson, PD, Rothwell, JC, Day, BL and Marsden, CD (1991b) Axial myoclonus of propriospinal origin. Brain, 114: 197-214. Brown, P, Rothwell, JC, Thompson, PD, Britton, TC, Day, BL and Marsden, CD (1991c) New observations on the normal auditory startle reflex in man. Brain, 114: 1891-1902. Brown, P, Day, BL, Rothwell, JC, Thompson, PD and Marsden, CD (1991d) The effect of posture on the normal and pathological auditory startle reflex. J. Neural. Neurosurg. Psychiatry, 54: 892-897. Brune, W, Weber, RG, Saul, B, von-Knebel, DM, Grond, GC, Kellerman, K, Meinck, HM and Becker, CM (1996) A GLRAI null mutation in recessive hyperekplexia challenges the functional role of glycine receptors. Am. J. Hum. Genet., 58: 989-997. Cantu, JM and Cuellar, A (1974) Congenital severe generalized muscle hypertonia during wakefulness: a distinct autosomal recessive disorder. Clin. Genet., 6: 32-35. Chauvel, P, Trottier, S, Vignal, JP and Bancaud, J (1992) Somatomotor seizures of frontal lobe origin. Adv. Neurol.,57: 185-232. Chiriboga, CA, Vibbert, M, Malouf, R, Suarez, MS, Abrams, EJ, Heagarty, MC, Brust, JC and Hauser, WA (1995) Neurological correlates of fetal cocaine exposure: transient hypertonia of infancy and early childhood. Pediatrics, 96: 1070-1077. Chokroverety, S, Wakzak, T and Hening, W (1992) Human startle reflex: technique and criteria for abnormal response. Electroencephalogr. Clin. Neuraphysiol., 85: 236-242. Colebatch, JG, Barrett, G and Lees, AJ (1990) Exaggerated startle reflexes in an elderly woman. Mov. Disord., 5: 167-169.

HYPEREKPLEXIA

Davis, M, Gendelman, DS, Tischler, MD and Gendelman, PM (1982) Primary acoustic startle circuit: lesion and stimulation studies. J. Neurosci., 2: 791-805. Dooley, JM and Andennann, F (1989) Startle disease or hyperekplexia: adolescent onset and response to valproate. Pediatr. Neurol., 5: 126-127. Dubowitz, LM, Bouza, H, Hird, MF and Jaeken, J (1992) Low cerebrospinal fluid concentration of free gammaaminobutyric acid in startle disease. Lancet, 340: 8Q.-81. Duensing, F (1952) Schreckreflex und schreckreaktion als hirnorganische zeichen. Arch. Psychiatric Nervenkr., 188: 162-192. Eapen, V, Moriarty, J and Robertson, MM (1994) Stimulus induced behaviours in Tourette's syndrome. J. Neurol. Neurosurg. Psychiatry, 57: 853-855. Elmslie, FV, Hutchings, SM, Spencer, V, Curtis, A, Covanis, T, Gardiner, RM and Rees, M (1996) Analysis of GLRAI in hereditary and sporadic hyperekplexia: a novel mutation in a family cosegregating for hyperekplexia and spastic paraparesis. J. Med. Genet., 33: 435-436. Fariello, RG, Schwartzman, RJ and Beall, SS (1983) Hyperekplexia exacerbated by occlusion of posterior thalamic arteries. Arch. Neurol., 40: 244-246. Fenzi, F, Bongiovanni, G, Fincati, E, Pampanin, M, Tomelleri, G and Rizzuto, N (1988) Anatomical and clinical study of a case of subacute encephalomyelitis with hyperekplexia syndrome. Ital. J. Neurol. Sci., 9: 505-508. Floeter, MK, Andermann, F, Andermann, E, Nigro, M and Hallett, M (1996) Physiological studies of spinal inhibitory pathways in patients with hereditary hyperekplexia, Neurology, 46: 766-772. Fyffe, RE (1991) Glycine-like immunoreactivity in synaptic boutons of identified inhibitory interneurons in the mammalian spinal cord. Brain Res., 547: 175-179. Gastaut, H and Villeneuve, A (1967) The startle disease or hyperekplexia; Pathological surprise reaction. J. Neurol. Sci., 5: 523-542. Giacoia, GP and Ryan, SG (1994) Hyperekplexia associated with apneoa and sudden infant death syndrome. Arch. Pediatr. Adolesc. Med., 148: 54Q.-543. Grenningloh, G, Schmieden, V, Schofield, PR, Seeburg, PH, Siddique, T, Mohandas, TK, Becker, CM and Betz, H (1990) Alpha subunit variants of the human glycine receptor: primary structures, functional expression and chromosomal localization of the corresponding genes. EMBO J., 9: 771-776. Hallett, M, Chadwick, D, Adam, J and Marsden, CD (1977) Reticular reflex myoclonus: a physiological type of human post-hypoxic myoclonus. J. Neurol. Neurosurg. Psychiatry, 40: 253-264.

487 Hayashi, T, Tachibana, H and Kajii, T (1991) Hyperekplexia: pedigree studies in two families. Am. J. Med. Genet.,40: 138-143. Hill, RM, Desmond, MM and Kay, JL (1966) Extrapyramidal dysfunction in an infant of a schizophrenic mother. J. Pediatr., 69: 589-595. Howard, R and Ford, R (1992) From the jumping Frenchmen of Maine to post-traumatic stress disorder: the startle response in neuropsychiatry. Psycho I. Med., 22: 695-707. Kimber, TE and Thompson, PD (1997) Symptomatic hyperekplexia occurring as a result of pontine infarction. Mov. Disord., 12: 814-816. Kingsmore, SF, Giros, B, Suh, D, Bieniarz, M, Caron, MG and Seldin, MF (1994) Glycine receptor beta-subunit gene mutation in spastic mouse associated with LINE-l element insertion. Nat. Genet., 7: 136-141. Klein, R, Haddow, JE and DeLuca, C (1972) Familial congenital disorder resembling stiff-man syndrome. Am. J. Dis. Child., 124: 730-731. Kohara, N, Ugawa, Y, Kuzuhara, Sand Yamanouchi, H (1988) An electrophysiological study on spinobulbospinal reflex in three brainstern stroke patients. Rinsho. Shinkeigaku., 28: 137-146. Krauss, JK, Trankle, Rand Kopp, KH (1997) Posttraumatic movement disorders after moderate or mild head injury. Mov. Disord., 12: 428-431. Kuhse, J, Kuryatov, A, Maulet, Y, Malosio, ML, Schrnieden, V and Betz, H (1991) Alternative splicing generates two isofonns of the alpha 2 subunit of the inhibitory glycine receptor. FEBS Lett., 283: 73-77. Kurczynski, TW (1983) Hyperekplexia. Arch. Neurol., 40: 246-248. Lees, AJ, Robertson, M, Trimble, MR and Murray, NM (1984) A clinical study of Gilles de la Tourette syndrome in the United Kingdom. J. Neural. Neurosurg. Psychiatry, 47: 1-8. Leigh, PN, Rothwell, Jc, Traub, M and Marsden, CD (1980) A patient with reflex myoclonus and muscle rigidity: "jerking stiff-man syndrome." J. Neurol. Neurosurg. Psychiatry, 43: 1125-1131. Lingam, S, Wilson, J and Hart, EW (1981) Hereditary stiff-baby syndrome. Am. 1. Dis. Child., 135: 909-911. Manford, MR, Fish, DR and Shorvon, SD (1996) Startle provoked epileptic seizures: features in 19 patients. J. Neurol. Neurosurg. Psychiatry, 61: 151-156. Markand, ON, Garg, BP and Weaver, DD (1984) Familial startle disease (hyperexplexia). Arch. Neurol., 41: 7174. Matsumoto, J, Fuhr, P, Nigro, M and Hallett, M (1992) Physiological abnormalities in hereditary hyperekplexia. Ann. Neurol., 32: 41-50.

488 Milani, N, Dalpra, L, Del, PA, Zanini, R and Larizza, L (1996) A novel mutation (Gln266 - c-His) in the alpha 1 subunit of the inhibitory glycine-receptor gene (GLRAl) in hereditary hyperekplexia. Am. 1. Hum. Genet., 58: 420-422. Morley, OJ, Weaver, DD, Garg, BP and Markand, a (1982) Hyperexplexia: an inherited disorder of the startle response. Clin. Genet., 21: 388-396. Pascotto, A and Coppola, G (1992) Neonatal hyperekplexia: a case report. Epilepsia, 33: 817-820. Rajendra, S, Lynch, JW, Pierce, KD, French, CR, Barry, PH and Schofield, PR (1995) Mutation of an arginine residue in the human glycine receptor transforms betaalanine and taurine from agonists into competitive antagonists. Neuron., 14: 169-175. Rees, MI, Andrew, M, Jawad, S and Owen, MJ (1994) Evidence for recessive as well as dominant forms of startle disease (hyperekplexia) caused by mutations in the alpha 1 subunit of the inhibitory glycine receptor. Hum. Mol. Genet., 3: 2175-2179. Ryan, SG, Dixon, MJ, Nigro, MA, Kelts, KA, Markand, ON, Terry, JC, Shiang, R, Wasmuth, JJ and O'Connell, P (1992) Genetic and radiation hybrid mapping of the hyperekplexia region on chromosome 5q. Am. J. Hum. Genet., 51: 1334-1343. Saenz, LE, Herranz-Tanarro, FJ, Masdeu, JC and Chacon Pena, JR (1984) Hyperekplexia: a syndrome of pathological startle responses. Ann. Neurol., 15: 36-41. Sander, JE, Layzer, RB and Goldsobel, AB (1980) Congenital stiff-man syndrome. Ann. Neurol., 8: 195197. Saul, B, Kuner, T, Sobetzko, D, Brune, W, Hanefeld, F, Meinck, HM and Becker, CM (1999) Novel GLRAI missense mutation (P250T) in dominant hyperekplexia defines an intracellular determinant of glycine receptor channel gating. J. Neurosci., 19: 869-877. Schorderet, DF, Pescia, G, Bernasconi, A and Regli, F (1994) An additional family with Startle disease and a G1192A mutation at the alpha I subunit of the inhibitory glycine receptor gene. Hum. Mol. Genet., 3: 1201. Seri, M, Bolino, A, Galietta, U, Lerone, M, Silengo, M and Romeo, G (1997) Startle disease in an Italian family by mutation (K276E): The alpha-subunit of the inhibiting glycine receptor. Hum. Mutat., 9: 185-187. Shahar, E, Brand, N, Uziel, Y and Barak, Y (1991) Nose tapping test inducing a generalized flexor spasm: a hallmark of hyperekplexia. Acta. Paed. Scand., 80: 1073-1077. Shiang, R, Ryan, SG, Zhu, Y, Hahn, AF, O'Connell, P and Wasmuth, JJ (1993) Mutations in the alpha! subunit of the inhibitory glycine receptor cause the dominant neurologic disorder, hyperekplexia. Nature Genet., 5: 351-357.

P.BROWN

Shiang, R, Ryan, SG, Zhu, YZ, Fielder, TJ, Allen, RJ, Fryer, A, Yamashita, S, O'Connell, P and Wasmuth, JJ (1995) Mutational analysis of familial and sporadic hyperekplexia. Ann. Neurol., 38: 85-91. Shibasaki, H, Kakigi, R, ada, K and Masukawa, S (1988) Somatosensory and acoustic brain stem reflex myoclonus. J. Neurol. Neurosurg. Psychiatry, 51: 572-575. Shimamura, M (1973) Neural mechanisms of the startle reflex in cerebral palsy, with special reference to its relationship with spino-bulbo-spinal reflexes. In: JE Desmedt (Ed.), New Developments in Electromyography and Clinical Neurophysiology (Vol. 3). Basel, Karger, pp. 761-766. Stell, R, Thickbroom, GW and Mastaglia, FL (1995) The audiogenic startle response in Tourette's syndrome. Mov. Disord., 10: 723-730. Stevens, H (1965) 'Jumping Frenchman of Maine': myarichit. Arch. Neurol., 12: 311-314. Suhren, 0, Bruyn, GW and Tuynman, JA (1966) Hyperexplexia: a hereditary startle syndrome. J. Neurol. Sci., 3: 577--605. Thompson, PD, Colebatch, JG, Brown, P, Rothwell, JC, Day, BL, Obeso, JA and Marsden, CD (1992) Voluntary stimulus-sensitive jerks and jumps mimicking myoclonus or pathological startle syndromes. Mov. Disord., 7: 257-262. Tijssen, MAJ and Brouwer, OF (2000) Hyperekplexia in the first year of life. Mov. Disord., 15: 1293-1296. Tijssen, MA, Bollen, E, Van Exel, E and Van Dijk, JG (1995a) Saccadic eye movements in hyperekplexia. Mov. Disord., 10: 749-753. Tijssen, MA, Shiang, R, Van Deutekom, J, Boerman, RH, Wasmuth, JJ, Sandkuijl, LA, Frants, RR and Padberg, GW (1995b) Molecular genetic re-evaluation of the Dutch hyperekplexia family. Arch. Neurol., 52: 578582. Tijssen, MA, Padberg, GW and Van Dijk, JG (1996) The startle pattern in the minor form of hyperekplexia. Arch. Neurol., 53: 608--613. Tijssen, MA, Schoemaker, HC, Edelbroek, PJ, Roos, RA, Cohen, AF and van, OJ (1997) The effects of clonazepam and vigabatrin in hyperekplexia. J. Neurol. Sci., 149: 63-67. Tijssen, MAJ, Brown, P, Morris, HR and Lees, AJ (1999) Late onset startle-induced tics. J. Neurol. Neurosurg. Psychiatry, 67: 782-784. Tijssen, MAJ, Meyer, BU, Davie, C, Rothwell, JC and Brown, P (2000) Motor cortex function in hereditary hyperekplexia is normal. Mov. Disord. (in press). Vergouwe, MN, Tijssen, MA, Shiang, R, Van Dijk, JG, al Shahwan, S, Ophoff, RA and Frants. RR (1997) Hyperekplexia-like syndromes without mutations in the GLRAI gene. Clin. Neurol. Neurosurg., 99: 172178.

HYPEREKPLEXIA

Vergouwe, MN, Tijssen, MA, Peters, AC, Wielaard, R and Frants, R (1999) Hyperekplexia phenotype due to compound heterozygosity for GLRAI gene mutations. Ann. Neurol., 46: 634-638. Wartenberg, R (1941) Head retraction reflex. Am. J. Med. Sc., 201: 553-561.

489 Weaver, DD, Morley, DJ, Garg, BP and Markand, 0 (1982) Hyperexplexia: not hereditary stiff-baby syndrome. Am. J. Dis. Child., 136: 562. Winston, K (1983) Hyperekplexia relieved by surgical decompression of the cervicomedullary region. Neurosurgery, 13: 708-710.

Movement Disorders Handbook of Clinical Neurophysiology, Vol. 1

M. Hallett (Ed.) © 2003 Elsevier B.V.All rights reserved

491 CHAPTER 31

Cerebellar ataxias Mario-Ubaldo Manto* Fonds National de la Recherche Scientifique, B-1070 Brussels, Belgium

Introduction

Massaquoi and Topka, 2001). This theory will be one of the leads throughout this chapter.

The term "cerebellar ataxia" encompasses various forms of cerebellar disorders encountered during daily practice. The scope of this chapter is to give an overview of the neurophysiological tests which provide important information for the understanding and the management of cerebellar ataxias. The relevant neuroanatomy of cerebellar pathways and the clinical symptoms of cerebellar disorders will be reviewed in the first and second part, respectively. In the third part, results of neurophysiological studies in cerebellar patients will be detailed. In particular, the results of investigations dedicated to motor control and learning will be discussed. Rather than a principal driver of movement, the cerebellum is viewed currently as a tonic reinforcer and an adaptative modulator or controller. Controllers are intended to optimize the movements, enhancing the stability of a moving system (Massaquoi and Topka, 2002). They are used in machines to diminish perturbation-induced oscillations or to improve the smoothness and accuracy during the execution of a task. They playa critical role for fast movements when several interlinked segments are involved. Because cerebellar deficits are more prominent in quick movements involving several joints, the theory of the controller fits with clinical observations in cerebellar patients. In addition, cerebellum plays a determinant role during motor learning (Gilbert and Thach, 1977; Ito, 1984; Hallett and Grafman, 1997). Therefore, it works as an adaptative controller (Goodwin and Sin, 1984;

PART I. NEUROANATOMICAL PATHWAYS

* Correspondence to: Mario-Ubaldo Manto, MD, Charge de Recherches FNRS, Neurologie, 808 Route de Lennik, 1070 Brussels, Belgium. E-mail address:[email protected] Tel.: 32-2-555.67.47; fax: 32-2-555.39.42.

This section aims to reconsider the relevant neuroanatomy for the understanding of clinical and neurophysiological findings. The reader is referred to Colin et al. for an in-depth review (Colin et al., 2002). From a structural point of view, the cerebellum is made of pairs of deep nuclei which are embedded in white matter and are surrounded by the cerebellar cortex. The fastigial nucleus is medial, the globosus and emboliform nuclei are intermediate and the dentate nucleus is lateral (Fig. 1). Globosus and emboliform nuclei are grouped under the terminology of interpositus nuclei. Vermal cortex projects to the fastigial nuclei, the intermediate cortex projects to the interpositus nuclei and the lateral cerebellar cortex projects to the dentate nuclei. Afferents and efferents enter and leave the cerebellum through three pairs of cerebellar pedDEEP CEREBELLAR NUCLEI

Fig. 1. Deep cerebellar nuclei. The cerebellum is shown in an unfolded presentation.

M.-U.MANTO

492

FISSURE AND LOBULES .,...............~"""'

Central lobule .............."Anterior quadrang. lobule Posterior quadrang. lobule Superior semilunar lobule

Inferior semilunar lobule Gracile lobule

~~~d.~l.~~~5:~Blventer

lobule Paraflocculus Flocculus

Fig. 2. Fissures and lobules. The primary fissure demarcates the anterior lobe from the posterior lobe of the cerebellum. The posterolateral fissure separates the posterior lobe from the flocculonodular lobe.

uncles: the inferior (ICP; restiform body), the middle (MCP; brachium pontis) and the superior cerebellar peduncles (SCP; brachium conjonctivum) (Allen and Tsukahara, 1974). Jansen divided the cerebellum in three main parts: the anterior lobe, the posterior lobe and the paraflocculus/flocculus area (Jansen, 1969). Larsell proposed a different nomenclature, which is used in animal studies. Some authors apply this nomenclature to humans, sometimes inappropriately. Figure 2 illustrates the division of the cerebellum in lobules and Fig. 3 indicates the equivalence between COMPARATIVE NOMENCLATURE

Human cerebellum

Mammalian cerebellum lobule simplex

ansiform lobule

---+ posterior quadrangulate lobule crus I ---+ superior semilunar lobule

i

crus "---+ inferior semilunar lobule

paramedian lobule ~ dorsal paraflocculus

-4

gracile lobule biventral lobule tonsilla

ventral paraflocculus

---+ accessory paraflocculus

flocculus

---+ flocculus

Fig. 3. Comparison of mammalian and human nomenclature of the cerebellum.

the mammalian nomenclature and the human nomenclature.

31.1. Cerebellar afferents Afferents enter the cerebellum mainly via the ICP and the MCP, although some fibers enter via the SCPo There are three main types of afferents: the climbing fibers, the mossy fibers and a third group including cholinergic and monoaminergic afferents. Climbing fibers emerge from a unique source: the contralateral inferior olive, whereas mossy fibers project to the cerebellum from a large spectrum of ipsilateral and contralateral sources. Climbing fibers are thin, myelinated, relatively slow conducting (about 20 mls) and fire at frequency of about 1.1 Hz (Ito, 1984). Mossy fibers are large, myelinated, fast conducting and have high frequency discharges. The projections from climbing fibers to the Purkinje cells and from mossy fibers to the granule cells are excitatory (Fig. 4). Figure 5 indicates the origin of afferent pathways for each of the three sagittal subdivisions of the cerebellum: the vermis, the intermediate cerebellum and the lateral cerebellum. The anterior vermis receives a greater somatosensory input than the posterior vermis. Vestibular inputs are much greater for the posterior vermis (see also Section 31.6 of Part III for the clinical and neurophysiological implications).

493

CEREBELLARATAXIAS

OUTFLOW TRACTS Fig. 4. Excitatory input from climbing fibers and mossy fibers. Climbing fibers project upon cerebellar nuclei, inhibitory intemeurons in the cerebellar cortex and make synapse with the dendritic tree of the Purkinje cells, which inhibit the deep nuclei. Mossy fibers participate with granule cells and Golgi cells in glomeruli. Parallel fibers emerge from granule cells and make synapse with the Purkinje cells.

31.2. Cerebellar cortex

Three layers make up the cerebellar cortex: the outer molecular layer, the ganglionic layer made of Purkinje cells and the inner granular layer (Fig. 6). Axons of Purkinje cells project upon the deep cerebellar nuclei and vestibular nuclei. They have an inhibitory effect, which is mediated by the neurotransmitter gamma-aminobutyric acid (GABA). Purkinje cells have also an inhibitory effect upon intra-cortical intemeurons (basket cells, stellate cells, Golgi cells, Lugaro cells), which are themselves inhibitory (Eccles et al., 1967; Ito, 1984). Purkinje cells have a single primary dendrite emerging from the outer pole. The dendritic tree is perpendicular to the folium axis. Terminal portions are characterized by numerous spines connected with the parallel fibers originating from the granule cells (see below). A single climbing fiber makes synapse with the Purkinje cell. This is the most powerful excitatory synapse in the brain. A single

spike induces a large EPSP of about 9-11 ms, followed by repetitive discharges of 3 to 10 spikes in the soma of the Purkinje cell. Thach called the extracellular response the "complex spike" (Thach, 1967). However, climbing fibers have also inhibitory effects: (1) after selective destruction of climbing fibers by 3-acetylpyridine (Colin et al., 1980), the spontaneous discharge rate of Purkinje cells increases; (2) when the climbing fiber is stimulated simultaneously with parallel fibers, the responsiveness of parallel fibers declines. This inhibition persists for a long time (LTD: long-term depression). Granule cells are the main cells in the granular layer. They have 4 to 5 dendrites. They make synapses with enlarged terminals of mossy fibers ("rosettes"). Rosettes are part of complexes called glomeruli. The unmyelinated thin axon of the granule cells ascends in the molecular layer and bifurcates in 2 branches (parallel fibers) running in opposite directions. The length of one branch has

M.-U.MANTO

494

fibers send collaterals to the granular layer. These nucleo-cortical projections reach mostly the areas from which they receive Purkinje cell axons. 31.4. The inferior olive

Fig. 5. Cerebellar afferents. The medial and intermediate parts of the cerebellum receive major inputs from the spinal cord, the vestibular system, the trigeminal system, as well as auditory and visual inputs. The lateral cerebellum receives major inputs from the cerebral cortex. The vestibulocerebellum (flocculonodular lobe) is a main site of projection of vestibular pathways.

been estimated to 2 to 7 mm. Another type of neuron in the granular layer is the unipolar brush cell. These cells are mainly found in the vestibulocerebellum (Mugnaini et al., 1997). 31.3. Cerebellar nuclei The dentate nucleus is divided into a rostromedial region (magnocellular) and a posterolateral region (parvocellular). The dentate is composed mainly of large multipolar neurons. This is also the case for the emboliform nucleus. Globosus nucleus and fastigial nucleus contain both large and small multipolar neurons. Nuclear cells receive a dense inhibitory Purkinje cell innervation. The excitatory inputs on the nuclear cells come from collaterals of climbing and mossy afferents.Projection fibers from the nuclei are excitatory (glutamatergic cells), with the exception of those targetting the inferior olive. All efferent

Olivary neurons have special features (Llinas et al., 1974). A strong electrotonic coupling between cells has been identified. The olivary complex is made of functional units with a tendency to fire synchronously. Five to six spines from different cells cluster in glomerular structures, making gap junctions. Harmaline increases the spontaneous tendency of olivary cell membrane potential to oscillate around 10 Hz. The inferior olive receives directly information from the spinal cord, the brainstern and the motor cortex. The dentato-rubro-olivary tract (also called Guillain-Mollaret triangle; see Section 31.4.1 of Part III) is involved in tremorgenesis. An interesting discovery was the observation that the white matter of the cerebellum is divided into long longitudinal strips (Voogd, 1964). Degeneration studies and retrograde tracing techniques have revealed eight major bands (A, X, B, C1, C2, C3, D1, D2). Each band receives climbing fibers from specific zones of the inferior olive and sends projections to well identified regions of the deep cerebellar nuclei (Voogd and Glickstein, 1998). The anatomical and functional unit of the olivocerebellar system consists of a sagittal band of cortex which receives climbing fibers from a small part of the inferior olive and which projects to a confined zone of the nuclei. This coupling between afferents and efferents does not exist for the mossy pathways. 31.5. Mossy fibers They have numerous origins: somesthetic, cortical, vestibular, acoustic, visual. 31.5.1. Spinocerebellar pathways

The dorsal spinocerebellar tract (DSCT; Flechsig's tract) originates from Clarke's column. DSCT is uncrossed. Animal studies have demonstrated that it conveys information mainly from the hind limbs. DSCT enters in the cerebellum via the ICP. DSCT cells are either proprioceptive or exteroceptive. Proprioceptive neurons are stimulated by la, II and

CEREBELLARATAXIAS

495

Fig. 6. Stereodiagram of the cerebellar cortex.

Ib endings and have a discharge frequency which increases linearly with the length of the muscle. The exteroceptive cells are monosynaptically activated by pressure receptors. The cuneocerebellar tract (CCT) is the equivalent of the DSCT for the forelimb. The ventral spinocerebellar tract (VSCT; Gower's tract) conveys information from the hind limbs. Axons end in the anterior lobe of the cerebellum, after crossing the midline first in the spinal cord and second at the level of the SCPo VSCT receives polysynaptic input from the ipsilateral flexor reflex afferents (FRA), defined as the myelinated fibers which induce the flexor reflex in the spinal preparation. They include cutaneous fibers, groups II and III muscle afferents and joints afferents. The rostral spinocerebellar tract (RSCT) is the equivalent of the VSCT for the forelimb. Axons project bilaterally in the anterior lobe, entering in the cerebellum through the ICP and the SCPo

31.5.2. Brainstem reticular nuclei

There are powerful excitatory loops between reticular nuclei and deep cerebellar nuclei. Fibers originating from the lateral reticular nuclei (LRN) run through the SCPo Projections are bilateral, predominating ispilaterally. LRN receives afferents from the pyramidal and rubrospinal tracts, the superior colliculus, the lateral vestibular nucleus. The nucleus gigantocellularis reticularis occupies the area dorsal and medial to the inferior olivary complex (Carpenter, 1985). It has characteristic large cells. The nucleus reticularis tegmenti pontis (NRTP) receives projections from Brodmann's areas 1 to 6, from vestibular nuclei and from pretectal areas. In addition, NRTP receives inputs from cerebellar nuclei via the SCP and projects back to the cerebellum through the MCP. The nucleus prepositus hypoglossi projects to the oculomotor nuclei through collaterals of fibers reaching the

496 tlocculonodular lobe. It plays an important role in the control of gaze. 31.5.3. Pontine nuclei

Pontine nuclei represent the most substantial collection of precerebellar nuclei (Carpenter, 1985). They are a critical relay in the conduction of impulses from the cerebral cortex to the cerebellum. Corticopontine fibers originate from the cerebral cortex of 4 lobes. In monkeys, the areas 4, 6, 3, 1,2, 5 and the visual cortex give rise to the main corticopontine fibers. Some computation is likely to occur in the pontine nuclei, but the low convergence (2/1) of corticopontine fibers argues in favor of a rather direct transfer of neuronal information. Brodmann's areas project to one or two slabs oriented rostro-caudally in the pons. Slabs project to the contralateral cerebellar hemisphere via the MCP. Less than 10% of the pontocerebellar projections are ipsilateral. The nodulus seems to be the only part of the cerebellum with no pontine projection. One hallmark of the mossy fiber system is the so-called "fractured somatotopy", which refers to the fact that a small area of cerebral cortex may project to several points in the slab and on the cerebellum. However. a sagittal orientation of the projections towards the cerebellar cortex has been found. 31.5.4. Vestibular projections

There are projections directed towards the ipsilateral vestibulocerebellum (tlocculus, paratlocculus, nodulus, uvula) and also fibers from the vestibular nuclei which project diffusely bilaterally to the vermis, tlocculus/paratlocculus, paramedian lobule and fastigiallinterpositus nuclei. 31.5.5. Acoustic, visual and trigeminal afferents

The audiovisual representation includes lobules VI, VII and vm. Posterior lobe receives afferents from the trigeminal nerve.

M.-U.MANTO

31.6.1. Fastigial nucleus (FN)

FN projections do not emerge via the SCP and a large part of them cross within the cerebellum (Carpenter, 1985). FN projects in particular on vestibular and reticular nuclei bilaterally. FN efferences modulate the activity of vestibulospinal and reticulospinal tracts. 31.6.2. Interpositus nucleus (IN)

The nucleus is divided into an anterior part (emboliform nucleus in man) and a posterior part. The anterior part projects to the contralateral thalamus (NVL: ventrolateral nucleus, NVA: ventral anterior nucleus, 'area X', NVPL: ventral posterolateral nucleus). There are collaterals towards the magnocellular portion of red nucleus and a contingent of fibers are dedicated to the NRTP, the pontine nuclei and the superior colliculus. The posterior part of the interpositus nucleus has efferents towards the thalamus and the red nucleus, though more limited. There is a somatotopy between paravermal cortex and spinal cord (Carpenter, 1985). Indeed, fibers from the anterior IN project somatotopically upon cells of the contralateral red nucleus via the SCPo Rubrospinal fibers cross the midbrain and descend to spinal levels. There is a debate concerning the role of this cerebello-rubro-spinal system in human. According to some authors, it might be involved in some facilitation of ipsilateral tlexor muscle tone in upper limbs. 31.6.3. Dentate nucleus (DN)

Neurons of the lateral nucleus projects to the NVL, NVPL, 'area X' of the contralateral thalamus, and also to the intralaminar thalamic nuclei (CLN: central lateral nucleus). Collaterals towards the red nucleus are connected with the parvocellular portion.

31.6. Cerebellar efferents

31.6.4. Projections from thalamic nuclei to the cerebral cortex

Cerebellar output emerges from deep nuclei, with the exception of vermal Purkinje neurons projecting to the lateral vestibular nuclei. Figure 7 illustrates the efferent pathways of the cerebellum.

Neurons from the NVL, NVPL and NVA project to the cerebral cortex. In monkeys, fastigial nucleus projects bilaterally to the hindlimb area of the motor cortex and the parietal cortex. The interpositus

CEREBELLAR ATAXIAS

497

Fig. 7. Cerebellar efferents. The lateral cerebellar cortex has major connections with contralateral motor and premotor cortices, via the thalamic nuclei. The intermediatecerebellar cortex influences the activity of contralateral motor cortex and is connected with lateral descending systems. The vermis projects upon medial descending systems. DN: dentate nucleus; IN: interpositus nucleus; FN: fastigial nucleus; mf: mossy fiber; cf: climbing fiber; NVL: ventrolateral nucleus; RN: red nucleus; dD/vD: vestibular nuclei; RF: reticular formation. Adapted from Eccles et al., 1967. nucleus is linked with the trunk area of the motor cortex and premotor cortex. The dentate nucleus projects contralaterally to the forelimb area of the motor cortex, the premotor cortex and prefrontal association cortex. Anatomical and functional output channels from different regions of the dentate nuclei to distinct cerebral cortical areas have been suggested (Middleton and Strick, 1997). Ventral areas of the dentate nucleus tend to project upon the prefrontal cortex via the thalamic relays, whereas dorsal regions are related to Ml area in cerebral cortex (Fig. 8). Of interest is the fact that areas 9 and 46 of prefrontal cortex are targets of cerebellar efferents. These areas are involved in working memory and the guidance of behavior based on transiently stored information.

31.7. Connections with autonomic centers The cerebellum is connected with autonomic centers in the central nervous system (eNS), especially with the hypothalamus. These connections are

OUTPUTS OF DENTATE NUCLEUS TO CORTICAL AREAS

Fig. 8. Representation of output channels in the dentate nucleus. Discrete areas of the dentate nucleus project upon distinct regions of the cerebral cortex. Dorsal parts of the dentate nucleus have a preferential projection towardsarea 4, whereas ventral parts of the dentate nucleus tend to project more on the prefrontal cortex.

M.-U.MANTO

498

reciprocal. It has been demonstrated that the cerebellum is involved in the control of the visceromotor system (Reis and Golanov, 1997).

PART II. CEREBELLAR ATAXIAS IN CLINICAL PRACTICE 31.1. Definition of ataxia Ataxia literally means 'without order'. The term ataxia was initially used to describe the loss of coordination in Tabes dorsalis. Nowadays, ataxia refers mainly to the inaccuracy of a movement towards a target (dysmetria), to rhythmic limb movements either during a sustained posture (postural tremor) or when the patient attempts to reach a target (kinetic tremor), to the inability to perform smooth alternate movements (dysdiadochokinesia) and to the loss of coordination of muscle groups in multi-joint movements (asynergia/decomposition).

of the cerebellar lesion. Recently, subtle neuropsychological deficits have been shown in some cerebellar patients, but they will not be discussed here. As underlined in Part I, different regions of the cerebellum are connected with distinct input/output areas. Therefore, it is not suprising that clinical deficits will be different according to the topography of the lesion. The terminology of functional compartmentalization has been coined (Dichgans, 1984). Lesions of the flocculo-nodular lobe and fastigial nuclei induce oculomotor deficits. Lesions of the anterior lobe generate postural ataxia. Lesions in the lateral cerebellum induce dysmetria and tremor ipsilaterally in the limbs.

31.2. Clinical signs of cerebellar disorders The most important clinical signs of cerebellar dysfunction fall into four categories: abnormal oculomotor control, dysarthric speech, inaccuracy in limb movements (dysmetria/tremor) and ataxia of posture and gait (Holmes, 1917; Gilman et al., 1981). Clinically, the most common oculomotor signs are impaired fixation especially square wave jerks, nystagmus, saccadic dysmetria and impaired vestibuloocular responses (VOR). Skew deviation (static ocular misalignment with one eye higher than the other) is not rare. Table 1 indicates the main oculomotor deficits as a function of the localization

31.3. Disorders of the cerebellum in daily practice The most common causes of cerebellar disorders encountered in daily practice are listed in Table 2. Patients present isolated cerebellar signs or a combination with extra-cerebellar signs, resulting most often from a brainstem involvement.

31.3.1. Sporadic forms ofcerebellar ataxias There are difficulties of nosologic delimitation when one tries to classify the various forms of degenerative sporadic cerebellar atrophy (see Berciano, 2002). Clinically, sporadic forms of cerebellar cortical atrophy (CCA) usually present with isolated cerebellar signs. Radiologically, atrophy is restricted to cerebellar structures. When additional extra-

Table I Oculomotor deficits: topography of the lesion in the cerebellum. Flocculus/paraflocculus

Nodulus/ventral uvula

Dorsal vermis/fastigial nucleus

Nystagmus Primary position Rebound Centripetal Impaired pursuit Glissades VORcancellation Abnormal adaptation of VOR gain

Positional nystagmus Periodic alternating nystagmus Impairedtilt suppression

Saccadic dysmetria Macrosaccadic oscillations Smoothpursuit deficit

499

CEREBELLARATAXIAS

Table 2 Common causes of cerebellar disorders in an ataxic clinic. Stroke (infarction, hemorrhage) Degenerative Sporadic (cerebellar cortical atrophy CCAllate onset cerebellar atrophy LOCA, sporadic olivopontocerebellar atrophy sOPCA, multiple system atrophy MSA) Hereditary (Friedreich ataxia FA, spinocerebellar ataxia SCA, ataxia-telangiectasia, mitochondrial diseases) Toxic (alcohol, phenytoin, lithium salts) Immune (multiple sclerosis MS) Neoplastic disorder (cerebellar tumor, metastatic disease)

provide important information for the diagnosis. Unfortunately, sphincter denervation is not pathognomonic of MSA. Electrophysiological studies may also reveal a sublinical polyneuropathy (Golbe, 1998). Paraneoplastic cerebellar syndromes have typically a subacute course. Most cases are associated with a breast or ovarian cancer in women. More and more reports of cerebellar atrophy associated with immune diseases appear in the literature. Since cerebellar ataxia might improve following administration of steroids or other treatments acting on the immune system, it is important to exclude these diseases before making a diagnosis of idiopathic cerebellar atrophy (Table 3).

Paraneoplastic disease (mainly breast, ovarian, lung carcinoma)

31.3.2. Inherited ataxias

Structural disease (Chiari malformations, Dandy-Walker malformations, agenesis, hypoplasia)

In daily practice, the most common causes of inherited ataxias are Friedreich ataxia (FA) and spinocerebellar ataxias (SCAs). The first is a recessive disorder, whereas SCAs are autosomal dominant diseases. FA has a prevalence of about 1/50,000. The essential features of typical cases are the following: onset before the age of 25, progressive limb and gait ataxia, absence of tendon reflexes in lower limbs evidence of axonal sensory neuropathy (Pandolfo: 2002). Within five years of onset, dysarthria, tendon areflexia in four limbs, distal loss of position and vibration sense, and pyramidal signs are observed. Atypical cases have been described, including Acadian families who have a milder course, onset after the age of 25 (LOFA: late onset Friedreich ataxia), and Friedreich ataxia with retained reflexes (FARR). Cardiomyopathy, pes cavus, kyphoscoliosis, dia-

Infectious/parainfectious diseases (cerebellar abscess, cerebellitis) Trauma

cerebellar signs are present in combination with brainstem atrophy on MRI, then the patient presents a sporadic form of olivopontocerebellar atrophy (sOPCA). The term of multiple system atrophy (MSA) encompasses sOPCA, striatonigral degeneration (SND) and Shy-Drager syndrome (SDS). MSA is defined as a sporadic degenerative disease of the nervous system causing clinically various combinations of extra-pyramidal, pyramidal, cerebellar and autonomic symptoms (Quinn, 1994; Berciano, 2002). Gilman et al. have recommended to divide MSA into MSA-P when parkinsonian features predominate, and into MSA-C when cerebellar signs predominate (Gilman et al., 1989). The mean age of onset of MSA is 42.2±9.0 years (range: 31 to 78) and the mean age of death is 60.5 ± 8.7 years (range: 34 to 84) (Wenning et al., 1994; Wenning et al., 1997). Cerebellar signs will be noted in 47% of patients throughout the clinical course. Isolated cerebellar signs will be prominent signs in 9% of the cases at the beginning of the disease. Autonomic failure is nearly constant in MSA (Berciano, 2002). Search for postural hypotension using a tilt table and for denervation of the urethral sphincter (EMG) may

Table 3 Differential diagnosis of sporadic cerebellar ataxias associated with an auto-immune disease. -

multiple sclerosis (MS) cerebellar ataxia with anti-GAD antibodies celiac ataxia systemic lupus erythematosus Sjogren syndrome Cogan syndrome paraneoplastic

500

M.-U.MANTO

Table 4 Clinical features of spinocerebellar ataxias*. Clinical feature

SCA

Age at onset

young adult: SCA 1,2,3; older adult: SCA 6; childhood: DRPLA, SCA 7, SCA 13

Anticipation

Most SCA; marked in DRPLA and SCA 4

Normal lifespan

SCA 6, SCA II

Upper motor neurons signs

SCA 1,3,7,12; some SCA 6,8; rare: SCA 2

Slow saccades

Prominent in SCA 2; late in SCA 1,3

Downbeat nystagmus

SCA6

Extra-pyramidal signs

SCA 3; SCA 12; DRPLA

Areflexia

SCA 2, SCA 3, SCA 4

Maculopathy

SCA 7

Seizures

SCA 10; SCA 7; DRPLA

Dementia

DRPLA,; SCA 2; SCA 7

Myoclonus

DRPLA; SCA 14; SCA 2

Head tremor

SCA 12, SCA 16

Mental retardation

SCA 13; DRPLA

* Adapted from Subramony and Filla, 2001.

betes mellitus, hearing loss and optic atrophy occur in some patients. The main differential diagnoses of FA are early onset cerebellar ataxia with retained reflexes (EOCA), ataxia with vitamin E deficiency (AVED), progressive metabolic ataxias and ataxia with hypogonadism. Preservation of knee jerks is the main clinical hallmark which separates EOCA from FA (Pandolfo, 2002). Patients do not present with peripheral neuropathy and cardiomyopathy is absent. The molecular test remains the gold standard to distinguish EOCA from FA. Patients with AVED may retain reflexes in lower limbs and nerve conduction velocities may be normal. Somatosensory evoked potentials (SEP) show a disproportionate involvement of the central vs. the peripheral axonal branch of primary sensory neurons. Serum vitamin E levels are markedly reduced or not detectable by classical techniques. The disease is treatable by vitamin E given orally. Patients with a SCA exhibit a pancerebellar syndrome, often associated with extra-cerebellar signs. Table 4 lists the main features helpful in the differential diagnosis of SCAs.

PART III: NEUROPHYSIOLOGICAL

ASPECTS 31.1. Eye movements 31.1.1. Smooth pursuit movements Cerebellar structures playing a key role in ocular pursuit are the flocculus/paraflocculus and dorsal vermis/caudal fastigial nucleus. Smooth pursuit is typically irregular in cerebellar patients. Severe impairment of smooth pursuit in monkeys following cerebellectomy suggests that the cerebellum is involved in the initiation phase (first 100 ms) (Burde et al., 1975). Using infrared oculography, Straube et al. examined the consequences of unilateral cerebellar lesions (n= 10 patients) on the initial (0-20 ms) and later (80-100 ms) periods as well as later responses (200-300 ms) of horizontal smooth pursuit (Straube et al., 1997). No lesion affected the flocculo-nodular lobe. Results were compared with those obtained in 17 healthy subjects. Smooth pursuit was elicited using a step-ramp paradigm with

501

CEREBELLAR ATAXIAS

randomized horizontal directions. Lesions affected smooth pursuit initiation toward the lesioned side and reduced ipsiversive/contraversive horizontal smooth pursuit velocities. Velocities ranged from 10 to 30 degrees/so In the first 20 ms, velocity was 22% lower towards the side of the lesion than away from it. In the period of 80-100 ms, velocity was 16% lower. The authors proved that human cerebellum participates in pursuit initiation.

31.1.2. Saccades Physiologically, saccades are the fast eye movements used to bring images of objects onto the fovea. The command to the eye muscles is a phasic-tonic or pulse-step change in the innervation (Lewis and Zee, 1993). The pulse is reflected in a burst of motoneuron activity, to overcome the viscosity in orbital tissues. For horizontal movements, the pulse is generated by burst neurons in the paramedian reticular formation of the pons. For vertical and torsional movements, the signal the command is generated in the midbrain. The command is sent to motoneurons and to the oculomotor integrator (step of innervation). The cerebellum helps to maintain adequate calibration of the amplitudes for the pulse and the step of innervation. In clinical practice, cerebellar patients exhibit dysmetria in both eyes (Zee et aI., 1986; Lewis and Zee, 1993). However, this dysmetria may be asymmetrical (disconjugate). Saccadic dysmetria will result from an abnormal amplitude of the saccadic pulse. The abnormal amplitude of the saccadic step will lead to the post-saccadic drift. The regions of the cerebellum involved in the regulation of the saccadic pulse are the dorsal vermis (lobules VI and VII, mainly) and the fastigial nuclei. Stimulation of the vermis during visually guided saccades induces hypometric contralateral saccades (Keller et aI., 1983), highlighting that vermis is involved in the "on-line" control of the pulse amplitude. Lesions in the fastigial nuclei are usually associated with pulse hypermetria.

31.1.3. VOR and OKR As underlined in Part I, the flocculus, the nodulus and the paraflocculus are highly interconnected with the vestibular nuclei, suggesting key functions in the adaptation of the vestibulo-ocular responses (VOR) and optokinetic responses (OKR). The VOR helps to

stabilize retinal image during head motion. In most patients exhibiting cerebellar atrophy, the severity of the deficits correlate with the degree of atrophy (Fetter et aI., 1994). Patients exhibit an increased gain of the VOR (the gain of the VOR represents the ratio between head and eye velocities). In addition, suppression of the VOR by fixating a target moving simultaneously with the head is reduced. 31.2. Limb movements

31.2.1. Bereitschaftpotential (BP) Self-paced voluntary movements are preceded by a negative EEG potential (BP). This negative shift is divided into an early and a late component. The late component is found in the hemisphere contralateral to the moving arm, starting about 400 ms before motion onset. It is called NS2 or NS I or even the "contralateral preponderance of negativity" according to different authors. The early component of the BP (BP proper) may begin about 2 s before movement onset. BP proper is not lateralized and has its maximal intensity over the vertex (Deecke and Kornhuber, 1978). In simple motor tasks, such as finger flexion or extension, a depression in the amplitude of BP has been found in patients with a cerebellar stroke and in patients presenting dentate nucleus lesions, such as tumors or the Ramsay-Hunt syndrome. BP may even be completely absent (Kitamura et aI., 1999; Ikeda et al., 1994). More complex tasks have also been analyzed. In 1994, Wessel et aI. studied patients with autosomal dominant or idiopathic cerebellar atrophy. Subjects performed self-paced movements in two different paradigms: a keys task (goal-directed) and a mouse task (sequential task) (see Wessel et aI., 1994 for the details of the technique). The authors selected these tasks that were expected to be more sensitive to cerebellar deficits than simple flexions or extensions of one finger. Wessel and colleagues performed a detailed analysis of the BP by differentiating an early NS1 component (600 to 800 ms prior to movement onset) preceding an NS2 component. A diffuse and bilateral distribution of the potential close to movement onset (MP) was found. The early component NS1 was increased, having its largest amplitude at Cz, and peak amplitude and NS2 amplitudes were reduced. Figure 9 shows the reduction of NS2 amplitude in three of our patients exhibiting a disabling cerebellar syndrome asso-

502

M.-D.MANTO

ciated with cerebellar atrophy as compared to three control subjects. These findings confirm the results obtained in animal studies showing that the primary motor cortex receives a strong excitatory input from the cerebellar efferent system (Asanuma et al., 1983; see also Part I). Cerebellar limb ataxia induced by the cooling of the dentate nucleus in monkeys is associated with a decreased phasic discharge of motor cortex neurons (Hore and Flament, 1988). This defective cortical phasic discharge is correlated with a less abrupt onset and a smaller intensity of EMG activity. In patients, involvement of the dentato-thalamo-cortical pathways results in an inadequate activation of motor cortex, hence the decrease in NS2 amplitude. Dentate nuclei exert a facilitatory effect upon the generation of BP proper and the negative slope. The more diffuse and MOVEMENT-RELATED CORTICAL POTENTIALS (BP) Healthy subjects .:

NS2

:. . ,: .,

:

:

:

••••• j • • • • • •: • • • • • ', ••••• : • • • • • • ;. • • • • •

: ..

:

-Cz· - Fz

·2

·3

·1

0

+1

+2

+3

lime (S80)

Cerebellar patients

..... i······:······:·· ....;... "'.:". ..... . . , ,

.,

·3

N~

,

. , . ,

·2

·1

0

.'., . . . +1

+2

.

- Cz - Fz

+3

Time (sec)

Fig. 9. Averages of movement-related cortical potentials in three healthy subjects and in three patients presenting a severe cerebellar syndrome associated with a marked cerebellar atrophy. The task is a sequential task. Subjects are asked to push rapidly on keys of a keyboard. Peak amplitudes of Bereitschaftspotential (BP) are reduced in ataxic patients, suggesting that cerebellar structures are critical to generate adequate intensities of movementrelated cortical potentials.

bilateral distribution of the MP could be related to a more diffuse cortical activation in cerebellar disorders. The increase in NS1 amplitude is presumably due to an attempt of compensation of motor deficit by a more extended cortical activation before the movement (Wessel et al., 1994).

31.2.2. Reaction time (RT) and movement time (MT)

RT (e.g. index finger release from a key) evaluates cognitive processing speed, also called speed of information processing (SIP). MT is a measurement of motor completion (e.g. finger pressing of an adjacent key). Experimental studies in monkeys have demonstrated that unilateral dentate nucleus lesions induce increased auditory RT in the immediate post-lesion phase, which gradually declines in the following weeks (Spidalieri et al., 1983). Early after a cerebellar stroke in humans, RTs are increased and MTs tend to be larger than in controls. RT determinations in patients who recovered partially or completely from a unilateral cerebellar stroke have yielded conflicting results. Some authors found normal values, whereas others observed increased RT (Wallesch and Hom, 1990; Botez-Marquard and Botez, 1997). In our experience, there is a good correlation between recovery of pure acute cerebellar deficits and return to normal values of simple visual RT. Patients with a degenerative disorder take a longer time to initiate a movement in response to a specific stimulus. Simple visual and auditory RT and MT are significantly larger in OPCA and in FA (BotezMarquard and Botez, 1997). There is a relationship between the severity of brainstemlcerebellar atrophy in OPCA and the lengthening of both RT and MT. In choice RT,RT increases in OPCA and controls as the task becomes more complex. Serial reaction time test (SRTT) will be discussed in the motor learning section.

31.2.3. Single-joint movements 31.2.3.1. Accuracy of the movement Cerebellar dysmetria is a cardinal cerebellar sign (Holmes, 1917). Hypermetria is the most common type of dysmetria in human. Less frequently, the

CEREBELLAR ATAXIAS

movement is characterized by a premature arrest before the target. This undershoot is called hypometria. Hypermetria is highly suggestive of a cerebellar disease, though not entirely specific. For instance, lesions involving thalamic nuclei or thalamo-cortical projections may mimic a cerebellar hypermetria. Kinematic profiles of single-joint movements are characterized by asymmetry, the ratio of the acceleration peak divided by the deceleration peak being lower than in controls. Other kinematic abnormalities have also been reported. In particular, a prolongation of the acceleration time as compared to the deceleration time has been observed (Hallett and Massaquoi, 1993). A prolongation of the deceleration time has been described in patients who exhibited clinically mild cerebellar deficits (Brown et al., 1990). Fast movements are normally launched by a burst of EMG activity in the agonist muscle and braked by a burst of EMG activity in the antagonist muscle (Hallett et al., 1975). The antagonist EMG burst is followed by a second burst in the agonist muscle. Several EMG deficits have been found in cerebellar patients exhibiting hypermetria. The most critical EMG defect is a delayed onset latency of the antagonist EMG activity (Conrad and Brooks, 1974; Massaquoi and Topka, 2002). The rate of rise in agonist and antagonist EMG activities is more gradual (Hallett et al., 1975; Hore et al., 1991; Manto et al., 1996). Studies with transcranial magnetic stimulation (TMS) over the motor cortex indicate that abnormal EMG activities are related to a reduction in the facilitatory drive exerted by the cerebellar output to the motor cortex (Spidalieri et al., 1983; Wessel et al., 1996). Increased durations of both the agonist and the antagonist EMG bursts have also been detected, confirming that timing computation is distorted in patients presenting a cerebellar ataxia. The three principal mechanisms underlying hypometria of fast single-joint distal movements, when present, are: (1) an imbalance between the rate of rise of the agonist EMG activity and the rate of rise of the antagonist EMG activity; (2) a reduction in the intensity of the agonist EMG activity which is associated with an increased duration of the antagonist EMG activity; and (3) a prolongation of the antagonist EMG activity (Manto et al., 1998a). This latter deficit was initially reported by Hallett et al. (1975).

503

Fig. 10. This figure illustrates the inability of cerebellar patients to adapt appropriately to increased inertia of the moving limb. Fast flexions of the wrist are illustrated, before and after addition (dotted lines) of a mass of 500 g at the level of the hand. The hypermetria which is observed in the normal state of inertia is increased after addition of an extra-mass. Aimed target: 50 degrees.

31.2.3.2. Effects of inertia The timing aspects of the EMG activities of the triphasic pattern are not the sole deficits in cerebellar patients. The cerebellum is also involved in the tuning of the intensity of the antagonist EMG activity. Facing an increased inertia, healthy subjects are able to increase the intensity of both the agonist and the antagonist EMG activities. By contrast, patients presenting a lesion in the lateral cerebellum increase the intensity of the agonist EMG activity, but are unable to tune appropriately the intensity of the antagonist EMG activity (Manto et al., 1994). As a result, the hypermetria is increased (Fig. 10). This test can be used to unravel silent cerebellar lesions (Manto et al., 1995a). 31.2.3.3. Effects of hyperventilation It has been shown that hyperventilation changes the characteristics of the nystagmus in cerebellar patients (Walker and Zee, 1999). Hyperventilation might influence the oculomotor control through a metabolic effect on cerebellar calcium channels, which play a critical role in the firing properties of cerebellar neurons. Patients presenting with spinocerebellar ataxia type 6, a disease associated with a polyglutamine expansion in the alpha I-A voltagedependent calcium channel, exhibit dysmetria during the performance of proximal and distal movements. It was shown in a small number of these patients that

504

hyperventilation may increase the overshoot of fast distal single-joint movements (Manto, 2001b). By contrast, hypermetria remains unchanged following hyperventilation in patients with idiopathic lateonset cerebellar degeneration (ILOCA). This might be a useful test to identify patients with SCA-6. The electric properties of calcium channels are defective in SCA-6. Matsuyama et al. have demonstrated that the abnormal calcium channels show a hyperpolarizing shift in the voltage dependence of inactivation, reducing to a great extent the available channel population at a resting membrane potential (Matsuyama et al., 1999). The result is in an impairment of the calcium influx in particular into Purkinje cells whose behavior is highly dependent upon calcium transfers, leading to a neuronal dysfunction and finally to cell death. Hyperventilation might aggravate the defective calcium transfers in SCA-6, resulting in an impairment of the calcium influx in particular into Purkinje neurons involved in the control of fast goal-directed voluntary movements. 31.2.3.4. Compensation and decompensation of cerebellar hypermetria It is well known that hypermetria due to a cerebellar injury may be followed by recovery (Dow and Moruzzi, 1958; Goldberger and Gordon, 1973; Kase et al., 1993). We recorded fast goal-directed wrist movements in patients who experienced an acute cerebellar hypermetria following a stroke and who subsequently recovered clinically (Manto et aI., 1995b). Movements and the associated agonist (flexor carpi radialis) and antagonist (extensor carpi radialis) EMG activities were recorded before and after addition of inertia throughout the recovery period. On the basis of the presence or absence of hypermetria in the basal state and of the effect of load addition on movement amplitudes, four qualitative successive stages emerged. Stage 1 was characterized by a hypermetria in the basal state which was unaffected by addition of masses. In stage 2, the hypermetria observed in the basal state increased with extra masses. At stage 3, the patients had recovered clinically. Movements were accurate in the normal state of inertia. Addition of a load induced a hypermetria. Stage 4 was characterized by accurate movements both in the basal condition and when a mass was added. In terms of EMG parameters controlling the execution of rapid and

M.-U. MANTO

accurate movements, the stage-by-stage recovery of hypermetria could be explained by a differential recovery. Immediately after an acute cerebellar injury, the patient presents several defects in the control of ballistic single-joint movements: (1) facing an increased inertial load, the patient cannot increase the agonist activity; (2) the onset latency of the antagonist EMG activity is delayed; (3) the patient is unable to increase appropriately the intensity of the antagonist EMG activity when the inertia of the moving limb is increased. The return of an adequate intensity of the agonist muscle in the presence of increased inertia precedes the return of a normal onset latency of the antagonist activity, which precedes the return to normal of an adequate intensity of the antagonist muscle in the presence of an increased inertial load. Experimental studies, as well as observations in humans, suggest that the use of alternative pathways is one of the mechanisms underlying the recovery of cerebellar deficits. Lesioning of the sensory cortex, the dorsal column system and the premotor cortex following recovery from a cerebellar injury decompensates cerebellar deficits (Aring and Fulton, 1936; Carpenter and Correll, 1961; Mackel, 1987). We observed a patient who presented cerebellar signs following a first cerebellar stroke, who recovered and who exhibited similar cerebellar signs after a second stroke involving the posterior parietal association cortex (Manto et aI., 1999). The first stroke was a left cerebellar infarction in the territory of the postero-inferior cerebellar artery. The second stroke was a cortico-subcortical infarct in the territory of right angular artery occurring 20 months later. The posterior parietal association area is characterized by an extensive convergence from several sensory systems, receiving afferents from high-order somatic, visual and auditory areas. Its role is to integrate this information, combining elemental information into more complex perceptions. It also acts as a sensori-motor interface whose properties might allow compensation of cerebellar deficits. 31.2.3.5. Aberrant recovery Clinical recovery may not occur following a cerebellar lesion. The reasons for this absence of recovery are poorly understood (Dow and Moruzzi, 1958; Gilman et al., 1981). Inadequate recovery might be misdiagnosed as an absence of recovery (Manto et aI., 1998b). Successive stages can be

505

CEREBELLARATAXIAS

identified in old patients presenting an aberrant recovery following a cerebellar stroke. At stage 1, movement is hypermetric. The onset latency of the antagonist EMG activity is delayed and both the intensities in the agonist and antagonist EMG activities are depressed. Stage 2 is characterized by terminal oscillations around the target. The onset latency of the antagonist EMG activity is normal, but the intensities of both the agonist and antagonist EMG activities are still depressed. At stage 3, movements are hypometric. The intensity of the antagonist EMG activity has returned to normal values, but the intensity of the agonist EMG activity is still depressed. At stage 2, mean movement amplitudes are within the normal range, but the movement is abnormal because of atypical dysmetric movements consisting of terminal oscillations around the target. Such movements having an increased variability of end position have been described in the monkey following cooling of the dentate nucleus (Flament and Hore, 1986). They are probably due to a mechanical instability of the limb during the holding phase around the target. Flament and Hore found that addition of masses to the handle tended to abolish these terminal oscillations. This atypical dysmetria turned into a typical hypermetria after increasing the inertia of the handle, in association with an increasing onset latency of the antagonist EMG activity, supporting the hypothesis of a stretch reflex participating in the antagonist EMG burst. What are the possible reasons for this aberrant recovery occurring in the elderly? There are three hypothesis which can be considered. First, an aberrant modification would occur during the acute neuronal reorganizational mechanisms which follow a cerebellar injury. Experimental studies have shown that lesions of the cerebellar nuclei are followed by sprouting of axon terminals in the contralateral motor cortex. These physical changes are characterized by a preferential increase of terminals making synapses with dendritic spines of pyramidal neurons (Keller et al., 1990). However, abnormal neuronal connections might impair performance. Neuronal reorganizational processes might generate maladaptative behavior, instead of a behavioral recovery of function (Finger and Almli, 1985). For instance, hemicerebellectomy in rats is associated with aberrant connections from remaining cerebellar nuclei to red nuclei and thalamic nuclei (Leong,

1978). A second hypothesis is that age influences recovery. It is known that the CNS plasticity is more prominent in developing brains and that the range of potential mechanisms of recovery decreases with age (Murakami et al., 1992; Hallett, 1995). Neuronal population diminishes with age. This phenomenon is also observed in Purkinje cells, whose total number decreases in the elderly (Hall et al., 1975). A third possible mechanism is that recovery may start in an erroneous way and therefore will stop prematurely. Indeed, the return to normal of the intensity of the antagonist activity precedes the return to normal of the onset latency of the antagonist EMG activity in patients who recover from a stroke. The reverse was observed in case of aberrant recovery. Such mistake might put the recovery process at a deadlock. 31.2.4. Multi-joint movements in the upper limb 31.2.4.1. Angular kinematics Abnormal curvature of hand trajectories and increased hand paths have been repeatedly found in cerebellar patients (Topka et aI., 1998a). Peak hand acceleration and deceleration are normal during slow and moderate velocity movements, but tend to be smaller than in control subjects for fast movements. Hypermetria is larger as movement velocity increases (Massaquoi and Hallett, 1996). Impairment in scaling motion variables (such as joint acceleration and joint deceleration) with movement speed is a key feature in some cerebellar ataxic patients (Topka et aI., 1998a). This defect is associated with abnormal coupling of shoulder and elbow joint rotations. Compared with healthy subjects, the patients present larger amplitudes of shoulder and elbow angular motion as velocity increases. This is illustrated in Fig. 11 in a patient presenting an ILOCA. The overshoot is usually associated with a hyperextension of the elbow during later phases of movement. The decomposition index suggested by Bastian et al. is a quantification of the degree to which elbow and shoulder joint move simultaneously during a reaching movement (Bastian et aI., 1995). Ideally, the shoulder and the elbow move simultaneously. Ataxic movements are associated with motion of one joint while the other joint angle is relatively stable. The temporal decomposition of shoulder and elbow joint movement causes an increased curvature of the hand path.

506

M.-U.MANTO

Fig. 11. Effects of increasing velocities on the kinematics of a pointing task in a healthy subject and in a patient presenting an idiopathic late-onset cerebellar cortical atrophy (ILOCA). In the patient, defects in angular motion increase with increasing velocity. The black line and the gray line correspond to angular position for the elbow and the shoulder respectively. Elb: elbow angle; sh: shoulder angle.

31.2.4.2. Dynamics Kinematic analysis is an important step to analyze movement defects in cerebellar patients. However, this technique does not take into account the mechanical properties of the human arm or the magnitude and orientation of the gravitational field (Topka et al., 1998b). As opposed to single-joint movements, force and acceleration of limb segments are no longer proportional for multi-joint movements, but are influenced by dynamic interactions (Kaminski and Gentile, 1989). Mechanically, the upper limb is made of several segments. An adequate coordination between the segments is a prerequisite to perform accurate movements (Topka et aI., 1998b). Forces acting at a given single joint arise from the muscles acting on this joint and also from other sources. In particular, forces linked to gravity and to the interactions between segments have to be taken into account. Using an inverse dynamics

approach, active and passive torques can be estimated. A detailed analysis of the equations is outside the scope of this chapter. Briefly, the upper limb is modelled as three interconnected rigid links (Topka et al., 1998b) with frictionless joints. Net torques (NET), gravitational torques (GRA), interactive torques (INT), muscle torques (MUS) are computed with this method. NET is given by: NET=GRA+INT+MUS

Defective control of dynamic movement variables is a factor of kinematic movement abnormalities in cerebellar limb ataxia (Topka et al., 1998b). Impairment in generating sufficient levels of MUS torques contributes to the patients' difficulties in controlling interaction forces during multijoint movements. Bastian et aI. have shown a mismatch between muscle torques and interaction torques in cerebellar patients (Bastian et aI., 1996). Topka et aI. (l998b)

507

CEREBELLARATAXIAS

have observed a reduction in peak muscular torques during the initiation of fast movements and a marked reduction in peak torques rate for shoulder and elbow at the initiation and termination of fast movements. The theory of impaired dynamics in cerebellar patients is very attractive, but remains a matter of controversy (Topka, 1999). In particular, it remains unclear how the cerebellum would be able to monitor rapid modifications in movement dynamics as they occur during limb motion. No peripheral receptor that would measure the force components of different origin has been identified so far. In addition, this theory does not explain the oculomotor deficits observed in cerebellar patients, considering the small amount of inertia for the eye and the absence of several segments.

analyze the spatial tuning curves. In healthy subjects, a privileged direction is observed for the peak EMG activity in each muscle. This direction is identical between successive recordings. In cerebellar patients, not only the directions associated with the maximal intensities of EMG activities are wrong but the shape of polar plots is also characterized by multiple peaks for proximal muscles. Experimental studies have shown a significant percentage of cell populations with directional responses in the cerebellum (Fortier et aI., 1989). The overall activity of the cerebellar population during reaches generated a signal varying with the direction of the proximal arm. The motor cortex might interact with the cerebellum to select the direction which will be associated with the largest intensity of the activity in proximal muscles.

31.2.4.3. Spatial tuning of EMG activities in the vertical plane Spatial tuning of motor activity is a critical phenomenon for the eNS (Georgopoulos, 1991). The role of the motor cortex has been studied extensively. One major finding was the discovery that movement direction is encoded by populations of neurons in the motor cortex (Georgopoulos, 1991). The cerebellum contributes also to the spatial tuning of EMG activities. In particular, cerebellar patients exhibit defects in the phasic spatial tuning of proximal EMG activities (Manto et al., 2002a). To test the hypothesis that the cerebellum is involved in the spatial tuning of EMG activities, a method similar to that described by Flanders et al. (1996) has been used. Briefly, 12 targets are located in the sagittal plane at a distance of 25-30 em from a central starting position. Fast pointing movements are repeatedly recorded towards each target. Surface EMG activities of the brachioradialis (BR), biceps (Bi), medial head of triceps (MT), long head of triceps (LoT), anterior deltoid (AD), posterior deltoid (PD) and latissimus dorsi (LaD) are recorded, rectified and averaged. To extract the phasic aspects of EMG activities, a subtraction procedure is applied. The averaged EMG activity related to slow movements is digitally compressed into the time frame of the fast EMG trace. Thereafter, the fast EMG trace and the slow EMG trace are subtracted. The result is the phasic trace. For each target and for each muscle, the peak EMG activity is identified. Polar plots of peak EMG activities are obtained to

31.2.5. Visuo-motor coordination

The accuracy of a limb movement towards visual targets implies a spatio-temporal coordination of both the oculomotor system and the moving limb (Brown et al., 1993). During motor tasks, the eyes normally foveate on the target well before completion of the motion of the limb. Retinal and extra-retinal information concerning eye position and position of the limb are involved in the ongoing control of oculomanual movements (Brown et al., 1993). The cerebellum has a strategic function in these tasks. Brown et al. have demonstrated that the initiation of both the saccadic and limb motor systems is delayed in cerebellar patients, and that this delay is dependent upon the nature of the visual feedback available during the movement. Removal of the visual feedback increases eye onset times and arm onset times in cerebellar patients (Fig. 12). The cerebellum might serve as the coordination center linking the oculomotor and limb motor systems during tasks requiring coordinated eye and limb movements. 31.3. Motor learning The cerebellum participates in various aspects of motor and non-motor learning. As underlined by Hallett and Grafman, motor learning is a complex phenomenon with many components (Hallett and Grafman, 1997). It can be defined as a change in motor performance with practice, as an increase in the repertoire of motor behavior or as a new behavior

508

M.-U. MANTO

HEALTHY SUBJECTS

U

1>o

z

~

450 400 350 300 250 200 150 100 50 0

NoVF

VF

CEREBELLAR PATIENTS 450 . , - - - - - - - - - - - - . ,

400 350 300 250 200 150 100 50

VF

*

NoVF

**

**

o

Fig. 12. Effects of removal of visual feedback on eye and arm onset movement times in a control group (n=5) and in a group of cerebellar patients (n=6). The task is a visuomotor coordination paradigm. Subjects have to flex the elbow and to make a horizontal saccade in response to a displacement of a target on a screen. In controls, removal of the visual information related to the target location does not influence saccades latencies and arm onset times. In patients, eye onset times and arm onset times are significantly increased as compared to controls (*intergroup difference: p<0.005 in both conditions) under the visual feedback condition. In the absence of visual feedback, movement onsets are increasingly delayed (** Visual effect in patients group: p<0.005 for both eye onset times and arm onset times). VF: visual feedback; No VF: absence of visual feedback. Means ± SD in both groups are shown.

maintained over a period of time (Hallett and Grafman, 1997). Motor adaptation learning is defined as a modification in motor performance without a change in "operating characteristics", whereas motor skill learning is defined as a change in motor performance with a modification in the operating characteristics. This means a new capability of the motor system. Let us consider Fitts' law which relates movement velocity and accuracy. If movements from a target A to a target B is considered, lower velocities are associated with increased precision. If the movement is performed both quickly and with greater accuracy, this is an apparent violation of Fitts' law (Hallett and Grafman, 1997). This means a change in the operating characteristic. Likely, skill learning cannot stand alone separate from adaptation learning, but it is important to look for changes in the operating characteristics in the assessment of motor learning. The cerebellum is a key structure in adaptation learning, whereas its implication is less important for skill learning. In this case, cerebral cortical structures are predominant (Hallett and Grafman, 1997). Several tests have been developed to evaluate the learning process in cerebellar patients. In particular, eye blink conditioning, adaptation to prisms, changes in visuomotor gain, and estimation of motor skill learning are amongst the most reliable tests. 31.3.1. Motor adaptation learning

The prototype of adaptation learning is the change in the gain of the VOR, which refers to the amplitude of the eye movements due to head motion. It is well established that posterior fossa structures playa key role in the adaptation of the VOR. The selection of the adequate gain is a learning procedure, but no apparent new skill is required when the gain changes (Hallett and Grafrnan, 1997). 31.3.1.1. Eye blink conditioning This is recognized by the majority of authors as a form of motor learning. The cerebellum is involved in the classical conditioning of withdrawal reflexes, a basic form of learning (Bracha et al., 1997). It is a part of the circuitry responsible for the anticipatory control protecting the body against potential harmful stimuli. Memory traces are formed in the cerebellum. The potential roles of the two major inputs to the cerebellar cortex (climbing fibers and parallel

509

CEREBELLARATAXIAS

Comeol puff

Tone

US

CS

Fig. 13. Circuitry involved in eyeblink conditioning. The nucleus interpositus is critical for the conditioned eyeblink response. UR: unconditioned response; CR: conditioned response; U.S.: unconditioned stimulus; CS: conditioned stimulus. Adapted from Woodruff-Pak, 1997.

fiber-mossy fiber system) have been demonstrated in animal studies. In particular, electrical stimulation of the pontine nuclei (conditioned stimulus) and of the inferior olive (unconditioned stimulus) has been used in classical conditioning of the rabbit nictitating membrane response (Woodruff-Pak et al., 1988). The nucleus interpositus is critical for the conditioned eyeblink response (Fig. 13). Lesions of the intermediate cerebellum are associated with an inability to acquire and retain classically conditioned eyeblinks. In humans, it has been shown that patients presenting a pure cerebellar cortical atrophy or olivopontocerebellar atrophy cannot acquire new classically conditioned eyeblinks (Topka et aI., 1993). In patients with bilateral lesions, few conditioned eyeblink responses are produced with the ipsilateral and contralateral eyes (Woodruff-Pak, 1997). In patients with unilateral lesions, conditioned responses (CR) are relatively normal in the contralateral eye and few or no CRs occur in the ipsilateral eye. Bracha et al. have tested control subjects and cerebellar patients for their capacity to acquire, retain and express conditioned eyeblink responses (Bracha et aI., 1997). They identified patients who could not acquire new classically conditioned eyeblinks and tested subsequently these patients for the retention of conditioned responses acquired prior to their cerebellar pathology. In acquisition testing, subjects were trained in a delay classical conditioning paradigm. The conditioned stimulus was a tone

and the unconditioned stimulus a midline forehead tap. Healthy subjects developed anticipatory eyeblinks to the tone in one session, whereas cerebellar patients could not acquire conditioned responses in four consecutive training sessions. In this study, the conditioning deficit was bilateral even in patients with a unilateral cerebellar disorder. The authors investigated the same groups of subjects for the presence of eyeblinks to a visual threat. A rubber ball was suspended in a pendulum-like fashion in front of the subjects. The ball hit the forehead of the subjects. Both groups responded when they saw an object approaching the face. These eyeblinks to a visual threat are likely to be naturally acquired, since they extinguish in control subjects if they are not reinforced by the unconditioned cutaneous stimulus. Moreover, seeing the approaching object blocks the acquisition of classically conditioned eyeblinks to a new conditioned stimulus in healthy subjects. Therefore, these findings show that cerebellar patients who are unable to acquire new classically conditioned responses are able to retain conditioned eyeblinks which were acquired prior to the beginning of the disease. This demonstrates that the neuronal networks involved in learning new conditioned reflexes are not required for the storage of naturally learned conditioned responses. 31.3.1.2. Adaptation to prisms

Prisms glasses induce a lateral displacement of vision. In pointing tasks, they cause a discrepancy

510

M.-U. MANTO

between direction of movement and the location of target seen by the eyes. With practice, healthy subjects improve in performance, by adjusting the direction of the movement. When the prisms are removed, there is a typical pointing movement in the opposite direction. With practice, the subject is able to make appropriate pointing again. Cerebellar patients exhibit a poor adaptation (Weiner et al., 1983). 31.3.1.3. Change in the visuomotor gain Abnormal learning has been shown in cerebellar patients (Deuschl et al., 1996a, b). The following paradigm has been used by Deuschl and colleagues (Deuschl et al., 1996a). The subject has to match a ballistic movement of the forearm to a target alternating between two positions on a computer screen. The cursor follows the target motions when the subjects perform a single rapid elbow flexion or extension. A first block of trials is recorded. Subsequently, the gain of the system is changed and a second block of trials is recorded. The absolute movement amplitude and movement duration are measured and an exponential function is fitted to the amplitude values obtained after the change of the gain, according to a least squares algorithm. The exponent T of this exponential function is computed to estimate the speed of learning. T value is a measure of the number of trials necessary to reach the baseline error. Healthy subjects rapidly adapt to the new gain, as compared to cerebellar patients. T values are higher in patients than in healthy subjects. Figure 14 illustrates that the rate of adaptation is much slower in a cerebellar patient (ILOCA) than in a control subject in a similar task. 31.3.2. Motor skill learning

As pointed out earlier, skill learning ("procedural memory") refers to the ability to increase performance through practice. Skill learning may be quantified by estimating the drop in errors. Multijoint arm movements are typical examples of motor skills. Movements become more accurate and are made quicker with learning. The cerebellum plays a major role when sequencing is critical during the task. Several studies have revealed deficits in motor skill learning in cerebellar patients. An exemplative experiment consists of making repetitive multijoint arm movements at various speeds. Patients are asked to increase the accuracy of movements without

Fig. 14. Adaptation to a change in the gain during a visuomotor task in a healthy subject and in a cerebellar patient. Wrist flexions on the right side with a visual feedback on a screen. The gain is changed suddenly at trial 31. The adaptation is much quicker in the healthy subject (dotted line) than in the patient presenting a idiopathic late onset cerebellar atrophy (ILOCA).

changing the durations of movements. They are still able to improve in performance with practice (change in operating characteristic), but the ability to learn is speed related (Hallett and Grafman, 1997). Skill learning is worse when a fast speed is required. 31.3.3. Serial reaction time test (SRIT)

The SRTT is very useful to investigate learning of sequences of motor actions (Pascual Leone et al., 1995). Typically, the paradigm is a choice reaction time with four possible responses (Hallett and Grafman, 1997). There are a series of stimuli to which the subjects must respond as rapidly as possible. The subject is not informed that the stimuli consist of a repeating sequence of about 10 elements. In healthy subjects, the responses become faster with practice, even without knowledge that there is a sequence. The subject may subsequently recognize that there is indeed a sequence. This is the shift from implicit learning to explicit learning. In the next step, the subject will be able to identify the sequence, leading to an improvement in performance. Implicit learning is abnormal in cerebellar degenerative disorders (Pascual-Leone et al., 1993). Patients do not improve in RT or develop explicit knowledge. Even when clues about the sequences are given, RT does not improve. Other structures known to play a key role in SRTT are the primary motor cortex (implicit learning) and the premotor/ parietal areas (explicit learning).

511

CEREBELLARATAXIAS

31.4. Tremors

Cerebellar tremors are highly complex. Four forms of cerebellar tremor are identified in human: palatal tremor (also classified as a myoclonus), postural tremor, kinetic tremor and titubation. Lesions of cerebellar outflow pathways are responsible for the most severe forms of kinetic and postural tremor. Postural tremor is evoked when the patient is asked to maintain the arms against gravity (Gilman et aI., 1981). The oscillations may be facilitated by body displacement. A fatigue-induced postural tremor has also been identified recently (see below). Kinetic tremor is tested clinically using the finger-tonose or the heel-to-shin tests. It is most severe during the final phase of the movement, when the limb approaches the target. Titubation is a low-frequency tremor of the head and/or the trunk. It might represent a peculiar form of postural tremor affecting neck and upper truncal muscles (Gilman et aI., 1981; Topka and Massaquoi, 2002). 31.4.1. Palatal tremor

Palatal tremor (PT; also called palatal myoclonus) is characterized by rhythmic movements of the soft palate at a frequency of 100-160/min. Two forms of PT are identified (Deuschl et al., 1996b): a symptomatic form (SPT) and an essential form (EPT). They differ clinically and pathophysiologically (Deuschl et al., 1994a, b). In SPT, the movements of the soft palate are usually asymptomatic. Visual inspection of the soft palate shows activation of different palatal muscles in the two disorders (Deuschl et al., 1994a). The levator veli palatini muscle is rhythmically active in SPT, whereas contractions of the tensor veli palatini are observed in EPT. In SPT, movements of the soft palate may be associated with synchronous movements of the eyes, the face, the tongue, limb muscles, and the diaphragm. EPT is associated with a clicking noise, but there are no other neurological signs. SPT is caused by a lesion located in the dentato-rubro-olivary tract. Anatomically, SPT is associated with a lesion in the central tegmental tract of the pons, a lesion in the dentate nucleus, the superior cerebellar peduncle, the decussation of the superior cerebellar peduncle or a lesion in the inferior olive. Currently, it is considered that the pathway between the inferior olive and the contralateral dentate nucleus through the inferior cerebellar peduncle is always preserved in SPT.

Indeed, no lesion in the inferior cerebellar peduncle has been found to be associated with SPT. The causes of SPT are cerebrovascular diseases, brainstem tumors, multiple sclerosis (MS), encephalitis, brainstem trauma and obstructive hydrocephalus. PT has also been described in association with progressive supranuclear palsy, and after removal of a low-grade cerebellar tumor. The main hypothesis concerning the physiopathogenesis of SPT is that olivary cells oscillate at 1-3 Hz, producing the involuntary movements at the same frequency. The inferior olive receives inhibitory projections from the dentate nucleus (see Sections 31.3 and 31.4 of Part I). A lesion of the dentato-rubro-olivary tract leads to decreased inhibition of the inferior olive and to hypersynchronization of the inferior olivary neurons. Olivary output produces synchronous firing of Purkinje cells in the cerebellar cortex via the olivo-cerebellar pathway. Hypertrophy of the olive may occur and be recognizable on MRI. This peculiar hypertrophic degeneration results from trans-synaptic degeneration or from hypersynchronization due to disinhibition. 31.4.2. Kinetic tremor

The mechanisms underlying kinetic tremor have not yet been discovered. Several lines of evidence support the idea that cerebellar tremors are generated within central loops (Topka and Massaquoi, 2002). Cooling the dentate nuclei in monkeys is associated with a terminal tremor (Flament and Hore, 1986). It is hypothesized that to avoid limb oscillations around the target, the central nervous system generates bursts of muscle activity in order to counteract the oscillations. These predictive bursts would be impaired by cerebellar dysfunction. They would lose not only their timing, but also their appropriate intensities. As a result, stretch reflexes at the spinal and supraspinal level would become prominent, causing additional unfavorable effects on the accuracy of movement. 31.4.3. Cerebellar postural tremor

Cerebellar patients may exhibit a postural tremor (Holmes, 1904, 1917). Five types can be distinguished on the basis of the underlying lesion, the precipitant factors, the frequency and the proximal! distal distribution (Table 5).

512

M.-U.MANTO

Table 5 Types of postural tremor in cerebellar patients *. Tremor type

Pathology

Precipitants

Frequency (Hz)

Distribution

Midbrain tremor

Cerebellar outflow tracts Nigrostriatal lesion

Any posture

2-5

Distal> proximal

Asthenic cerebellar tremor

Cerebellar hemisphere

Fatigue/weakness

Irregular

Proximal + distal

Precision cerebellar tremor

Cerebellar nuclei and brachium conjonctivum

Accurate placements

2-5

Distal

Cerebellar axial postural tremor

Cerebellarolivary system

Any posture

2-10

Proximal> distal

Cerebellar proximal exertional tremor

Cerebellar malformation

Prolonged exercise

3-4

Proximal> distal

* Adapted

and modified from Brown et aI., 1997.

Brown et al. have reported three patients with a predominantly axial tremor (Brown et al., 1997). One patient presented an ILOCA and another had a history of hemangioblastoma. Tremor was characterized by a frequency varying from 3 to 10 Hz, often jumping from one frequency to another. This tremor affected head and shoulder muscles much more than distal muscles, having a similar amplitude in repose or with the arms outstreched. In one case, EMG analysis revealed a subclinical modulation of palatal muscle activity with the truncal tremor. We have reported another form of proximal tremor, observed in patients presenting a posterior fossa malformation (Manto and Jacquy, 200la). One patient presented a hypoplasia of one cerebellar hemisphere and another patient presented a hypoplasia of the cerebellum associated with a large retrocerebellar cyst (Dandy-Walker-Blake continuum). Upper limb tremor was triggered by exertion. In patient 1, fatigue induced co-contracting bursts at a frequency of 3.2 Hz in the anterior deltoid, the biceps and the triceps ipsilaterally to the cerebellar lesion (Fig. 15). In patient 2, fatigue induced a 3.9 Hz tremor in the anterior deltoid on both sides, predominating on the right side. Frequency of the tremor was suggestive of a lesion of cerebellar pathways. The higher gain of long-latency stretch reflexes that is observed in cerebellar patients during

postural tasks may be a key factor in the genesis of cerebellar proximal exertional tremor. Indeed, long latency EMG responses are abnormal in cerebellar patients (Fig. 16). The early response (Ml: spinal component) is not affected in terms of latency and size, but later responses show normal latencies and higher amplitudes (M2/M3 components) as compared to controls (Friedmann et al., 1987). In control subjects, it is usually hard to separate the M2 from the M3 response in hand muscles. M3 tends to occur about 10 to 15 ms later than the M2 response (Wiesendanger and Miles, 1982). In patients, the most salient problem concerns M3, indicating that the cerebellum modulates the amplitude of this response. The increase in the intensity of the M3 component is mostly observed when cerebellar cortex is damaged and when cerebellar nuclei are spared, suggesting a loss of inhibition from Purkinje cells. In FA, the Ml response is delayed and the M2/M3 components are either delayed or not recognizable (Friedmann et al., 1987). Cerebellar postural tremor is only mildly improved by drugs. Stimulation of the thalamic ventral intermediate nucleus (VIM) may improve the tremor. In 13 patients with a disabling tremor due to MS, tremor amplitude decreased in about two thirds of the patients with this technique (Geny et al., 1996). Of the eight patients presenting a severe

CEREBELLARATAXIAS

Fig. 15. Cerebellarproximal exertionaltremor. Full waverectified surface EMG recording of the right anterior deltoid in a patient presenting a hypoplasia of a cerebellar hemispherebefore (A) and after exertion (B). Vertical bar: 0.25 mY. C: spectral analysis, estimated from a l-minute recording of data. Thin trace: before exercise; thick trace: after exertion. Adapted from Manto and Jacquy, in press. tremor, seven recovered the possibility to catch an object and to use it. VIM stimulation may help patients with MS who suffer from a disabling postural tremor.

31.4.4. TMS and cerebellar tremors Trains of focal magnetic stimuli over primary motor cortex induce a complex pattern of inhibition and excitation processes, depending on the intensity and the frequency of repetitive TMS (Pascual-Leone et aI., 1994). It was shown recently that trains of repetitive TMS over motor cortex in healthy subjects could induce a cerebellar-like tremor (Topka et al., 1999). Tremor occurred at the end of targeted voluntary movements and during maintained posture. The likelihood that tremor would occur

513

Fig. 16. Long latency EMG responses to stretches of the first dorsal interosseous on the right side. A: Comparison between a patient exhibiting a cerebellar syndrome on the right side following a right ischemic cerebellar stroke sparing the dentate nucleus (black line) and a control subject (gray line). Latencies of averaged rectified EMG responses are normal, but the M2 response is increased in the patient. The increase is even larger for the M3 response. B: Grand average in three patients with a unilateral cerebellar lesion and in four control subjects. Similar observations. Stimuli consisted in brief dorsiflexions of the index finger. The subject was asked to exert a constant force of 5 to 7% of maximum force, using a visual feedback and not to react to the perturbation. Surface EMG rectified and averaged 100 times. Responses calibrated in arbitrary units (a.u.). increased with increasing stimulus intensities or frequencies. It was suggested that tremor was caused by disruption of cortical processes involved in terminating a voluntary movements or maintaining a posture. TMS seems to enhance the activity of nonanticipatory transcortical reflex loops.

31.5. Myoclonus Cortical myoclonus leads to focal/multifocal myoclonic jerks, often precipitated during action and

M.-U.MANTO

514

disinhibiting the motor cortex (Tijssen et al., 2000). Tijssen et al. have studied cases of proven cortical myoclonus with cerebellar pathology. In two patients over three, the myoclonic ataxic syndrome was associated with a celiac disease. Enlarged SEP and time-locked cortical potentials preceding the action myoclonus were found (Fig. 17). Abnormal response to paired magnetic shocks over the motor cortex was observed. An increased facilitation occurred at interstimulus intervals of 7, 10, 15 ms. A tendency to rhythmic cortical and muscle discharges with a frequency around 40 Hz was found, arguing in favor of a role of the cerebellum in the modulation of rhythmic cortical discharges. Quantitative study demonstrated the loss of Purkinje cells in all lobules of the cerebellum. The atrophy of the cerebellar

which can be stimulus sensitive (Hallett et al., 1979). The most common causes are the Lance-Adams syndrome, Unverricht-Lundborg disease, mitochondrial disorders, celiac disease, DRPLA and spinocerebellar degenerations (Marsden et al., 1990; Bathia et al., 1995). Interestingly, pathologic findings are usually found in the cerebellum (Tijssen et al., 2000). Myoclonus has been observed in patients presenting various degrees of degeneration of spinocerebellar tracts and dentate nuclei, loss of Purkinje cells. For instance, in dyssynergia cerebellaris myoclonic a, cerebellar pathways are typically involved, with atrophy of dentate nuclei (Hunt, 1921). In cerebellar patients, cortical myoclonus might result from a selective lesion disrupting the inhibition of cerebellar cortex upon deep cerebellar nuclei, B

A

1

2 - _..."

120UV

1 20 UV

3~

3 - . ...,1\

-

AHB _-'...._ ....J"........_ _ 1250 uV 50ms

c

1~ 2 ~ AHB ~

-

1250 uV

100ms

1~

2~120UV 3~ AHB

--A.--

1250 uV

SO;;;; Fig. 17. Cortical myoclonus associated with cerebellar atrophy. Giant SEP and time-locked cortical correlate in a patient with a 5 year history of myoclonus associated with celiac disease. A: stimulation of the right tibial nerve at the ankle. TWo averages of 512 stimuli are superimposed. Amplitude of PIN2: 22 J.L V (N < 3.4). The initial reflex EMG burst is at 90 ms and is highly synchronous. A second burst occurs about 25 ms later. AHB: abductor hallucis brevis. B: average of 24 recordings following taps to the right foot. Taps evoke a giant cortical SEP, followed by repetitive EMG bursts in the AHB at a frequency around 40 Hz. C: Back-averaging of the EEG activity preceding action myoclonus induced by dorsiflexion and plantar flexion of the right foot. Action myoclonus consisted of repetitive short duration EMG bursts (20 to 50 ms). A positive cortical potential precedes EMG bursts. This patient died and neuropathological examination revealed a preservation of Betz cells in the motor cortex. No abnormality was found in basal ganglia, brainstem nuclei or in the spinal cord. The cerebellum showed symmetrical atrophy, predominating in the superficial aspects of the folia. Purkinje cell loss was present in the anterior lobe and in the posterior lobe, without evidence of granule cell loss. The dentate nuclei showed minimal gliosis. From Tijssen et al., 2000, with permission.

515

CEREBELLARATAXIAS

cortex contrasted with the relative sparing of dentate nuclei. Axial myoclonus with trunk extension has been observed during pregnancy in ataxic patients (Manto and Gille, 2002b). These jerks are not associated with giant SEP or C-reflex responses. They spare facial and limb muscles. These axial myoclonic discharges might be related to the overactivity of cerebello-reticular loops or cerebello-olivo-cerebellar loops. Stimulation of the rostral and dorsal part of the nucleus gigantocellularis reticularis produces excitatory postsynaptic potentials in motor neurons supplying axial muscles in the neck and in the back (Peterson, 1980). The same nucleus has been implicated in the genesis of other forms of myoclonic discharges. Palatal myoclonus (palatal tremor) was discussed in Section 31.4.1.

31.6. Posture and gait Posture and gait difficulties are among the first clinical symptoms to occur in sporadic and hereditary progressive cerebellar disorders. Postural adjustments are faulty when patients are asked to perform limb movements during quiet stance (Diener et aI., 1992). Cerebellar gait is irregular and staggering. Patients exhibit reduced step and stride length, a tendency to a reduced cadence and an increased variability in all gait parameters. 31.6.1. Damage to the anterior lobe

This is typically observed in patients with a history of alcoholism. Patients exhibit an increased body sway in the anterior-posterior direction, whereas lower vermal lesions are associated with an omnidirectional body sway. When the patients are lying on their back and are asked to maintain the hips and the knee flexed with an angle of 90 degrees, a highly suggestive 3 Hz tremor is observed, with a waxing and waning amplitude which gives a spindlelike aspect to the tremor. 31.6.2. Posterior vermal split syndrome

Bastian et al. have analyzed movements in children who underwent a surgical split of the posterior vermis in order to remove a fourth ventricle tumor (Bastian et al., 1998). In the procedure, the posterior cerebellar vermis was resected. None of the

patients had deficits attributable to brainstem damage. Lesion extension ranged from cerebellar lobule VI to X. Tandem gait was impaired in all children. Patients often fell or, when they could take a few consecutive steps, they exhibited lower speeds, lower cadence, increased variability of stride. By contrast, reaching, pinching, and kicking were preserved. Romberg test and regular gait was relatively unimpaired. Surgical splitting of the posterior vermis disrupted the parallel fibers crossing the midline. The authors have suggested that these fibers might be critical for the continuous bilateral coordination of the legs and trunk (Thach et aI., 1992). Tandem gait heavily relies on vestibular and visual coordination of legs and trunk (Bastian et al., 1998). The posterior vermis is therefore well suited to monitor tandem gait.

31.7. Assessment of cerebellar function using cerebellar stimulation Cerebellar stimulation may be a useful tool to distinguish: (1) cerebellar lesions; and (2) cerebellar ataxia from sensory ataxia (Ugawa et al., 1994; Matsunaga et aI., 2001). Ugawa et al. have shown that electrical stimulation of the cerebellum can transiently suppress the EMG responses evoked by TMS and that this effect may be absent in cerebellar patients (Ugawa et aI., 1994). The same authors have also developed a similar test using TMS over the cerebellum, which is less painful (Ugawa et al., 1995). In healthy subjects, stimulation over the basal occiput suppresses EMG responses evoked by magnetic cortical stimulation at an interstimulus interval (lSI) of 5, 6 and 7 ms. This suppressive effect is absent in patients presenting a lesion of the cerebello-thalamo-cortical pathway, whereas it is preserved in patients who present a lesion of cerebellar afferent pathways. Moreover, this procedure can be used in the follow-up of patients presenting LOCA (Ugawa et al., 1997). Recently, Matsunaga et al. have used the technique of Ugawa et al. in a follow-up study including two patients who developed an acute cerebellar ataxia and subsequently gradually improved (Matsunaga et al., 2001). The authors stimulated the cerebellum electrically and, at various times afterwards, stimulated the motor cortex. In healthy subjects, the conditioning stimuli over the cerebellum reduced the size of MEPs to the test shock

516

M.-U.MANTO Second study

Firststudy

control

C-T interval 3ma

4ma

Sma

6ms

7m.

10m.

u~

L" ,

1'\ '

i

v

-t-}V-

~r\r-

- '--.-.--1t:

i\

,

"

I,....----J\~

,v

FRA

GABA ICP ILOCA IN lSI LOFA LTD

_\J~~~

I

FN

..~

1U\rLJ\v-

FA

FARR

LRN

\r.--O.4r'lV

~~\.---

'~\r-- V

:'.---!\r--

NVA

" ....

Fig. 18. Averaged control and conditioned motor evoked potentials (MEPs) at each C-T interval in the first (acute phase) and second (recovery phase) studies in a patient presenting acute cerebellar ataxia following an upper respiratory tract infection. Recording in the first dorsal interosseous muscle. In the acute phase of the disease, conditioning stimulation over the cerebellum does not reduce the size of the MEP to test magnetic stimulation of the motor cortex at C-T intervals of 5-7 ms. In the recovery stage, normal suppression is observed. From Matsunaga et al., 2001 with permission.

at C-T intervals of 5,6,7 ms. At the acute stage of the ataxic syndrome, there was no or little suppression of MEPs at these intervals (Fig. 18). In the recovery stage, normal suppression of MEP re-occurred. Cerebellar stimulation was the only objective finding suggesting cerebellar dysfunction.

Abbreviations AVED BP CCA CCT CLN CNS CS CR C-T interval DN DSCT EOCA

ataxia with vitamin E deficiency Bereitschaftpotential cerebellar cortical atrophy cuneocerebellar tract central lateral nucleus central nervous system conditioned stimulus conditioned response conditioning-test interval dentate nucleus dorsal spinocerebellar tract early onset cerebellar ataxia

MCP MEP MS MSA MT NRTP NVL NVPL OKR

OPCA PT

RSCT RF RT SCA SCP SDS SEP

SIP SND SRTT TMS UR US VOR VSCT

Friedreich ataxia Friedreich ataxia with retained reflexes fastigial nucleus flexor reflex afferents gamma-arninobutyric acid inferior cerebellar peduncle idiopathic late-onset cerebellar atrophy interpositus nucleus interstimulus interval late onset Friedreich ataxia long-term depression lateral reticular nuclei middle cerebellar peduncle motor evoked potential multiple sclerosis multiple system atrophy movement time nucleus reticularis tegmenti pontis ventral anterior nucleus ventrolateral nucleus ventral posterolateral nucleus optokinetic reflex olivopontocerebellar atrophy palatal tremor rostral spinocerebellar tract reticular formation reaction time spinocerebellar ataxia superior cerebellar peduncle Shy-Drager syndrome somatosensory evoked potentials speed of informaion processing striatonigral degeneration serial reaction time test transcranial magnetic stimulation unconditioned response unconditioned stimulus vestibuloocular reflex ventral spinocerebellar tract

Acknowledgment Supported by the Fonds National de la Recherche Scientifique, Belgium.

References Allen, GI and Tsukahara, N (1974) Cerebrocerebellar communication systems. Physiol. Rev., 54: 957-1006. Aring, CD and Fulton, JF (1936) Relation of the cerebrum to the cerebellum. II. Cerebellar tremor in the monkey and its absence after removal of the principal excitable areas of the cerebrum (areas 4 and 6). m. Accentuation of the cerebellar tremor following lesions of the premotor area (area 6, upper part). Arch. Neurol. Psychiatry, 35: 439-466. Asanuma, C, Thach, WT and Jones, EG (1983) Distribution of cerebellar terminations and their relation to

CEREBELLAR ATAXIAS other afferent terminations in the ventral lateral thalamic region of the monkey. Brain Res., 5: 237-265. Bastian, AJ and Thach, WT (1995) Cerebellar outflow lesions: a comparison of movement deficits resulting from lesions at the levels of the cerebellum and thalamus. Ann. Neurol., 38: 881-892. Bastian, AJ, Martin, TA, Keating, JG and Thach, WT (1996) Cerebellar ataxia: abnormal control of interaction torques accross multiple joints. J. Neurophysiol., 76: 492-509. Bastian, AJ, Mink, JW, Kaufman, BA and Thach, WT (1998) Posterior vermal split syndrome. Ann. Neurol., 44: 601-610. Berciano, J (1982) Olivopontocerebellar atrophy. A review of 117 cases. 1. Neurol. Sci., 53: 253-272. Berciano, J (2002) Multiple system atrophy and idiopathic late onset cerebellar ataxia. In: M Manto and M Pandolfo (Eds.), The Cerebellum and its Disorders. Cambridge University Press, Cambridge, U.K., 178-197. Bhathia, KP, Brown, P, Gregory, R et al. (1995) Progressive myoclonic ataxia associated with celiac disease. The myoclonus is of cortical origin, but the pathology is in the cerebellum. Brain, 118: 10871093. Botez-Marquard, T and Botez, MI (1997) Olivopontocerebellar atrophy and Friedreich's ataxia: neuropsychological consequences of bilateral vs. unilateral cerebellar lesions. In: JD Schmahmann (Ed.), The Cerebellum and Cognition. Academic Press, San Diego, pp. 387-410. Bracha, V, Zhao, L, Wunderlich, A, Morrissy, SJ and Bloedel, JR (1997) Patients with cerebellar lesions cannot acquire but are able to retain conditioned eyeblink reflexes. Brain, 120: 1401-1413. Brown, SH, Hefter, H, Mertens, M and Freund, HJ (1990) Disturbances in human arm movement trajectory due to mild cerebellar dysfunction. J. Neurol. Neurosurg. Psychiatry, 53: 306-313. Brown, SH, Kessler, KR, Hefter, H, Cooke, JD and Freund, HJ (1993) Role of the cerebellum in visuomotor coordination. Exp. Brain Res., 94: 478-488. Brown, P, Rothwell, JC, Stevens, JM, Lees, AJ and Marsden, CD (1997) Cerebellar axial postural tremor. Mov. Disord., 12: 977-984. Burde, RM, Stroud, MH, Roper-Hall, G et al. (1975) Oculomotor dysfunction in total and hemicerebellectomized monkeys. Br. J. Ophtalmol., 59: 560-565. Carpenter, M (1985) Core text of neuroanatomy. Williams and Wilkins, Baltimore. Carpenter, MB and Correll, JW (1961) Spinal pathways mediating cerebellar dyskinesia in rhesus monkey. J. Neurophysiol., 24: 535-551.

517 Colin, F, Manil, J and Desclin, JC (1980) The olivocerebellar system. I. Delayed and slow inhibitory effects: an overlooked salient feature of the cerebellar climbing fibers. Brain Res., 187: 3-27. Colin, F, Ris, Land Godaux, E (2002) Neuroanatomy of the cerebellum. In: M. Manto and M Pandolfo (Eds.), The Cerebellum and its Disorders. Cambridge University Press, Cambridge, U.K., 6-29. Conrad, B and Brooks, VB (1974) Effects of dentate cooling on rapid alternating arm movements. J. Neurophysiol., 37: 792-804. Deecke, L, Scheid, P and Kornhuber, HH (1969) Distribution of readiness potential, premotion positivity, and motor potential of the human cerebral cortex preceding voluntary finger movements. Exp. Brain Res., 7: 158-168. Deecke, L and Kornhuber, HH (1978) An electrical sign of participation of the mesial "supplementary" motor cortex in human voluntary finger movement. Brain Res., 159: 473-476. Deuschl, G, Mischke, G, Schenck, E, Schulte-Monting, J and LUcking, CH (1990) Symptomatic and essential rhythmic palatal myoclonus. Brain, 113: 1645-1672. Deuschl, G, Toro, C and Hallett, M (1994a) Symptomatic and essential palatal tremor. 2. Differences of palatal movements. Mov. Disord., 9: 676-678. Deuschl, G, Toro, C, Valls-Sole, J, Zeffiro, T, Zee, DS and Hallett, M (1994b) Symptomatic and essential palatal tremor. I. Clinical, physiological and MRI analysis. Brain, 117: 775-788. Deuschl, G, Toro, C, Zeffiro, T, Massaquoi, S and Hallett, M (l996a) Adaptation motor learning of arm movements in patients with cerebellar disease. J. Neurol. Neurosurg. Psychiatry, 60: 515-519. Deuschl, G, Toro, C, Valls-Sole, J and Hallett, M (1996b) Symptomatic and essential palatal tremor. Abnormal motor learning. J. Neurol. Neurosurg. Psychiatry, 60: 520-525. Dichgans, J (1984) Clinical symptoms of cerebellar dysfunction and their topodiagnostical significance. Hum. Neurobiol., 2: 269-279. Diener, HC, Dichgans, J, Guschlbauer, B, Bacher, M, Rapp, Hand Klockgether, T (1992) The coordination of posture and voluntary movement in patients with cerebellar dysfunction. Mov. Disord., 7: 14-22. Dow, RS and Moruzzi, G (1958) The physiology and pathology of the cerebellum. University of Minnesota Press, Minnesota. Eccles, J, Ito, M and Szentagothai, J (1967) The cerebellum as a neuronal machine. Springer-Verlag. Fetter, M, Klockgether, T, Schultz, JB, Faiss, J, Koenig, E and Dichgans, J (1994) Oculomotor abnormalities and MRI findings in idiopathic cerebellar ataxia. J. Neurol., 241: 234-241.

518 Finger, S and Almli, CR (1985) Brain damage and neuroplasticity: mechanisms of recovery or development? Brain Res. Rev., 10: 177-186. Flament, J and Hore, J (1986) Movement and electromyographic disorders associated with cerebellar dysmetria. J. Neurophysiol., 55: 1221-1233. Flanders, M, Pellegrini, 11 and Geisler, SD (1996) Basic features of phasic activation for reaching in vertical planes. Exp. Brain Res., 110: 67-79. Fortier, PA, Kalaska, JF and Smith, AM (1989) Cerebellar neuronal activity related to whole-arm reaching movements in the monkey. J. Neurophysiol., 62: 198-211. Friedmann, HH, Noth, J, Diener, HC and Bacher, M (1987) Long latency EMG responses in hand and leg muscles: cerebellar disorders. J. Neurol. Neurosurg. Psychiatry, 50: 71-77. Geny, C, Nguyen, JP, Pollin, B, Feve, A, Ricolfi, F, Cesaro, P, Degos, JD (1996) Improvement of severe postural cerebellar tremor in multiple sclerosis by chronic thalamic stimulation. Mov. Disord., 5: 489-494. Georgopoulos, AP (1991) Higher order motor control. Annu. Rev. Neurosci., 14: 361-377. Gilbert, PE and Thach, WT (1977) Purkinje cell activity during motor learning. Brain Res., 128: 309-328. Gilman, S, Bloedel, JR and Lechtenberg, R (1981) Disorders of the cerebellum. Contemporary neurology series. Philadelphia, Davis. Gilman, S, Quinn, NP, Albanese, A, Ben Shlomo, Y, Fowler, C, Kaufmann, H et al. (1999) Consensus statement on the diagnosis of multiple system atrophy. J. Neurol. Sci., 163: 94-98. Golbe, LI (1998) Multiple system atrophy. In: S Gilman, GW Goldstein and SG Waxman (Eds.), Neurobase. Arbor Publishing, San Diego. Goldberger, ME and Growdon, FH (1973) Pattern of recovery following cerebellar deep nuclear lesions in monkey. Exp. Neurol., 39: 307-322. Goodwin, GC and Sin, KS (1984) Adaptive filtering prediction and control. Englewood Cliffs, PrencticeHall. Hall, TC, Miller, AKH and Corsellis, JAN (1975) Variations in the human Purkinje cell population according to sex and age. Neuropath. Appl. Neurol., 1: 267-292. Hallett, M (1995) The plastic brain. Ann. Neurol., 38: 4-5. Hallett, M and Grafman, J (1997) Executive function and motor skill learning. In: J. Schmahrnann (Ed.), The Cerebellum and Cognition. Academic Press, San Diego, pp. 297-323. Hallett, M and Massaquoi, S (1993) Physiologic studies of dysmetria in patients with cerebellar deficits. Can. J. Neurol. Sci. (Suppl. 3): 83-92.

M.-U.MANTO

Hallett, M, Shakani, BT and Young, RR (1975) EMG analysis of patients with cerebellar deficits. J. Neurol. Neurosurg. Psychiatry, 38: 1163-1169. Hallett, M, Chadwick, D and Marsden, CD (1979) Cortical reflex myoclonus. Neurology, 29: 1107-1125. Holmes, G (1904) On certain tremors in organic cerebral lesions. Brain, 27: 327-375. Holmes, G (1917) The symptoms of acute cerebellar injuries due to gunshot injuries. Brain, 40: 461-535. Hore, J and Flament, D (1988) Changes in motor cortex neural discharge associated with the development of cerebellar limb ataxia. J. Neurophysiol., 60: 12851302. Hore, J, Wild, B and Diener, HC (1991) Cerebellar dysmetria at the elbow, wrist and fingers. J. Neurophysiol., 65: 563-571. Hunt, JR (1921) Dyssynergia cerebellaris myoclonica primary atrophy of the dentate system. Brain, 44: 490-538. Ikeda, A, Shibasaki, H, Nagamine, T, Terada, K, Kaji, R, Fukuyama, H and Kimura, J (1994) Dissociation between contingent negative variation (CNY) and Bereitschaftpotential (BP) in a patient with cerebellar efferent lesion. Electroencephalogr. Clin. Neurophysiol., 90: 359-364. Ito, M (1984) The cerebellum and motor control. Raven Press, New York. Jansen, J (1969) On cerebellar evolution and organization from the point of view of a morphologist. In: R Llinas (Ed.), Neurobiology of Cerebellar Evolution and Development. Chicago, American Medical Association, pp. 881-893. Kaminski, TR and Gentile, AM (1989) A kinematic comparison of single and multijoint pointing movements. Exp. Brain Res., 78: 547-556. Kase, CS, Norving, B, Levine, SR et al. (1993) Cerebellar infarction. Clinical and anatomic observations in 66 cases. Stroke, 24: 76-83. Keller, EL, Siakey, DP and Crandall, WF (1983) Microstimulation of the primate cerebellar vermis during saccadic eye movements. Brain Res., 288: 131-143. Keller, A, Arissian, K and Asanuma, H (1990) Formation of new synapses in the cat motor cortex following lesions of the deep cerebellar nuclei. Exp. Brain Res., 80: 23-33. Kitamura, J, Shabasaki, H, Terashi, A and Tashima, K (1999) Cortical potentials preceding voluntary movement in patients with focal cerebellar lesion. Clin. Neurophysiol., 110: 126-132. Leong, SK (1978) Plasticity of cerebellar afferents after neonatal lesions in albino rats. Neurosci. Lett., 7: 281-289. Lewis, RF and Zee, DS (1993) Ocular motor disorders associated with cerebellar lesions: pathophysiology

CEREBELLAR ATAXIAS

and topical localization. Rev. Neurol. (Paris), 149: 665-677. Llinas, R, Baker, R and Sotelo, C (1974) Electrotonic coupling between neurons in cat inferior olive. J. Neurophysiol., 37: 560-571. Mackel, R (1987) The role of the monkey sensory cortex in the recovery from cerebellar injury. Exp. Brain Res., 66: 638-652. Manto, M, Godaux, E and Jacquy, J (1994) Cerebellar hypermetria is larger when the inertial load is artificially increased. Ann. Neurol., 35: 45-52. Manto, M, Godaux, E and Jacquy, J (1995a) Detection of silent cerebellar lesions by increasing the inertial load of the moving hand. Ann. Neurol., 37: 344-350. Manto, M, Jacquy, J, Hildebrand, J and Godaux, E (1995b) Recovery of hypermetria after a cerebellar stroke occurs as a multi-stage process. Ann. Neurol., 38: 437-445. Manto, M, Godaux, E, Jacquy, J and Hildebrand, J (1996) Cerebellar hypermetria associated with a selective decrease in the rate of rise of the antagonist EMG activity. Ann. Neurol., 39: 271-274. Manto, M, Setta, F, Jacquy, J, Godaux, E, Hildebrand, J, Roland, H, Blum, Sand Brohee, P (1998a) Different types of cerebellar hypometria associated with a distinct topography of the lesion in cerebellum. J. Neurol. Sci., 158: 88-95. Manto, M, Hildebrand, J and Jacquy, J (l998b) Shift from hypermetria to hypometria in an aberrant recovery following cerebellar infarction. J. Neurol. Sci., 157: 42-51. Manto, M, Setta, F, Jacquy, J and Godaux, E (1999) Cerebellar decompensation following a stroke in contralateral posterior parietal cortex. J. Neurol. Sci., 167: 117-120. Manto, M and Jacquy, J (200la) Proximal exertional tremor in posterior fossa malformations. Mov. Disord., 976-980. Manto, M (200Ib) Effects of hyperventilation on fast goal-directed limb movements in spinocerebellar ataxia type 6. Eur. J. Neurol., 401-406 Manto, M, Jacquy, J, Legros, B and Bosse, P (2002a) Shift from hypermetria to hypometria in multiple system atrophy: analysis of distal and proximal movements. Neurol. Res., 249-258. Manto, M and Gille, M (2002b) Acute cerebellar ataxia with axial myoclonus during pregnancy. Eur. Neurol., 55-58. Marsden, CD, Harding, AE, Obeso, JA and Lu, CS (1990) Progressive myoclonic ataxia (the Ramsay Hunt syndrome). Arch. Neurol., 47: 1121-1125. Massaquoi, S and Hallett, M (1996) Kinematics of initiating a two-joint arm movement in patients with cerebellar ataxia. Can. J. Neurol. Sci., 23: 3-14.

519 Massaquoi, Sand Topka, H (2002) Models of cerebellar function. M Manto and M Pandolfo (Eds.). Cambridge University Press, Cambridge, U.K. (69-94). Matsunaga, K, Uozumi, T, Hashimoto, T and Tsuji, S (2001) Cerebellar stimulation in acute cerebellar ataxia. Clin. Neurophysiol., 112: 619-622. Matsuyama, Z, Wakamori, M, Mori, Y, Kawakami, H, Nakamura, S and Imoto, K (1999) Direct alteration of the P/Q-type Ca2 + channel property by polyglutamine expansion in spinocerebellar ataxia 6. J. Neurosci., 19: 14. Middleton, FA and Strick, PL (1997) Cerebellar output channels. In: JD Schmahmann (Ed.), The Cerebellum and Cognition. Academic Press, San Diego, pp.6182. Mugnaini, E, Dino, MR and Jaarsma, D (1997) The unipolar brush cells of the mammalian cerebellum and cochlear nucleus: cytology and microcircuitry. Prog. Brain Res., 114: 131-150. Murakami, F, Song, WJ and Katsumaru, H (1992) Plasticity of neuronal connections in developing brains of mammals. Neurosci. Res., 15: 235-253. Pandolfo, M (2002) Friedreich ataxia. In: M Manto and M Pandolfo (Eds.), The Cerebellum and its Disorders. Cambridge University Press, Cambridge, U.K., pp.505-518. Pascual-Leone, A, Grafman, J, Clark, K, Stewart, M, Massaquoi, S, Lou, JS and Hallett, M (1993) Procedural learning in Parkinson's disease and cerebellar degeneration. Ann. Neurol., 34: 594-602. Pascual-Leone, A, Valls-Sole, J, Wasserman, EM and Hallett, M (1994) Responses to rapid-rate transcranial magnetic stimulation of the human motor cortex. Brain, 117: 847-858. Pascual-Leone, A, Grafman, J and Hallett, M (1995) Procedural leaming and prefrontal cortex. Ann. NY Acad. Sci., 769: 61-70. Peterson, B (1980) Reticulo-motor pathways: their connections and possible roles in motor behavior. In: H Asanuma and VJ Wilson (Eds.), Integration of the Nervous System. Igaku Shoin, Tokyo, pp. 185-201. Quinn, N (1994) Multiple system atrophy. In: CD Marsden and S Fahn (Eds.), Movement Disorders 3. London, Butterworths, pp. 262-281. Reis, DJ and Golanov, EV (1997) Autonomic and vasomotor regulation. In: JD Schmahmann (Ed.), The Cerebellum and Cognition. Academic Press, San Diego, pp. 121-149. Schmahmann, JD (1997) The cerebellum and cognition. Academic Press, San Diego. Spidalieri, G, Busby, L and Lamarre, Y (1983) Fast ballistic arm movements triggered by visual, auditory, and somesthetic stimuli in the monkey. II. Effects of unilateral dentate lesion on discharge of precentral

520 cortical neurons and reaction time. J. Neurophysiol., 50: 1359-1379. Straube, A, Scheuerer, W and Eggert, T (1997) Unilateral cerebellar lesions affect initiation of ipsilateral smooth pursuit eye movements in human. Ann. Neurol., 42: 891-898. Subramony, S and Filla, A (2001) Autosomal dominant spinocerebellar ataxias ad infinitum? Neurology, 56: 287-289. Thach, WT (1967) Discharge of Purkinje and cerebellar nuclear neurons during rapidly alternating arm movements in the monkey. J. Neurophysiol., 31: 785-796. Thach, WT, Goodkin, HG and Keating, JG (1992) Cerebellum and the adaptative coordination of movement. Annu. Rev. Neurosci., 15: 403-442. Tijssen, MAJ, Thom, M, Ellison, DW, Wilkins, P, Barnes, D, Thompson, PD and Brown, P (2000) Cortical myoclonus and cerebellar pathology. Neurology, 54: 1350-1356. Topka, H (1999) Motor and non-motor functions of the cerebellum. Report from the joint meeting of EuroAtaxia and the Fondation de I' Ataxie Cerebelleuse, Westende, pp. 9-19. Topka, Hand Massaquoi, S (2002) Pathophysiology of clinical cerebellar signs. In: M Manto and M Pandolfo (Eds.), The Cerebellum and its Disorders. Cambridge University Press, pp. 121-135. Topka, H, Valls-Sole, J, Massaquoi, S and Hallett, M (1993) Deficit in classical conditioning in patients with cerebellar degeneration. Brain, 116: 961-969. Topka, H, Konczak, J and Dichgans, J (l998a) Coordination of multi-joint arm movements in cerebellar ataxia: analysis of hand and angular kinematics. Exp. Brain Res., 119: 483-492. Topka, H, Konczak, J, Schneider, K, Boose, A and Dichgans, J (l998b) Multijoint arm movements in cerebellar ataxia: abnormal control of movement dynamics. Exp. Brain Res.; 119: 493-503. Ugawa, Y, Genba-Shimizu, K, Rothwell, JC, Iwata, M and Kanazawa, I (1994) Suppression of motor cortical excitability by electrical stimulation over the cerebellum in ataxia. Ann. Neurol., 36: 90-96. Ugawa, Y, Ueseka, Y, Terao, Y, Hanajima, R and Kanazawa, I (1995) Magnetic stimulation over the cerebellum in humans. Ann. Neurol., 37: 703-713. Ugawa, Y, Terao, Y, Hanajima, R, Sakai, K, Furubayashi, T, Machii, K and Kanazawa, I (1997) Magnetic stimulation over the cerebellum in patients with ataxia. Electroencephalogr. CUn. Neurophysiol., 104: 453458.

M.-U.MANTO

Voogd, J (1964) The cerebellum of the cat. Thesis, University of Leiden. Voogd, J and Glickstein, M (1998) The anatomy of the cerebellum. Trends Neurosci., 21: 370-375. Walker, MF and Zee, DS (1999) The effect of hyperventilation on downbeat nystagmus in cerebellar disorders. Neurology, 53: 1576-1579. Wallesch, CV and Hom, A (1990) Long-term effects of cerebellar pathology on cognitive functions. Brain Cogn., 14: 19-25. Weiner, MJ, Hallett, M and Funkenstain, HH (1983) Adaptation to lateral displacement of vision in patients with lesions of the central nervous system. Neurology, 33: 766-772. Wenning, GK, Ben-Shlomo, Y, Magalhaes, M, Daniel, SE and Quinn, NP (1994) Clinical features and natural history of multiple system atrophy. An analysis of 100 cases. Brain, 177: 835-845. Wenning, GK, Tison, F, Ben-Shlomo, Y, Daniel, SE and Quinn, NP (1997) Multiple system atrophy: a review of 203 pathologically proven cases. Mov. Disord., 12: 133-147. Wessel, K, Verleger, R, Nazarenus, D, Vieregge, P and Kempf, D (1994) Movement-related cortical potentials preceding sequential and goal-directed finger and arm movements in patients with cerebellar atrophy. Electroencephalogr. CUn. Neurophysiol., 92: 331-341. Wessel, K, Tegenthoff, M, Vorgerd, M, Otto, V, Nitschke, MF and Malin, JP (1996) Enhancement of inhibitory mechanisms in the motor cortex of patients with cerebellar degeneration: a study with transcranial magnetic stimulation. Electroencephalogr. CUn. Neurophysiol., 101: 273-280. Wiesendanger, M and Miles, TS (1982) Ascending pathway of low-threshold muscle afferents to the cerebral cortex and its possible role in motor control. Physiol. Rev., 62: 1234-1270. Woodruff-Pak, DS, Steinmetz, JE and Thompson, RF (1988) Classical conditioning of rabbits 2~ to 4 years old using mossy fiber stimulation as a CS. Neurobiol. Aging, 9: 187-193. Woodruff-Pak, DS (1997) Classical conditioning. In: JD Schmahmann (Ed.), The Cerebellum and Cognition. Academic Press, San Diego, pp. 341-366. Zee, DS, Yee, RD and Cogan, DG (1986) Ocular motor abnormalities in hereditary cerebellar ataxia. Brain, 99: 207-234.

Movement Disorders Handbook of Clinical Neurophysiology, Vol. I

M. Hallett (Ed.)

© 2003 Elsevier B.V. All rights reserved

521

CHAPTER 32

The clinical neurophysiology of myoclonus John N. Caviness* Department of Neurology, Mayo Clinic Scottsdale, 13400 East Shea Blvd., Scottsdale, AZ 85259, USA

32.1. Clinical concepts 32.1.1. Definition and introduction Myoclonus is a clinical symptom (or sign) defined as sudden, brief, shock-like, involuntary movements caused by muscular contractions or inhibitions. Myoclonus movements have now been recognized to have many possible etiologies, anatomical sources, and pathophysiologic features (Caviness, 1996). When including all etiologies, myoclonus has an average annual incidence of 1.3 cases per 100,000 (Caviness et aI., 1999). There are two major methods for the classification of myoclonus: etiological and physiological. Both methods offer different advantages and insights. Over time, the myoclonus literature has created strong and useful associations between clinical, etiological, and neurophysiological findings (Shibasaki, 2000). Thus, the various etiologies of myoclonus have had "clinico-electrophysiologic" types described for them. It is now very clear that two different examples of myoclonus etiology may have an identical clinical appearance but very different source localization and neurophysiological characteristics. As a result, a clinical neurophysiological examination of myoclonus is necessary to determine what type or types of myoclonus are present in a given patient. The evaluation of the patient with myoclonus must begin with the clinical history and physical exam. Important information obtained from the history should include mode of myoclonus onset, presence of other neurological problems, history of seizures, drug or toxin exposure, past or current medical problems, and family history. The basic

* Correspondence to: John N. Caviness, M.D., Department of Neurology, Mayo Clinic Scottsdale, 13400 East Shea Blvd., Scottsdale, AZ 85259, USA. E-mail address: [email protected] Tel.: 480-301-7989; fax: 480-301-8451.

parts of the myoclonus exam include defining distribution, temporal profile, and activation characteristics. The distribution of myoclonus can be focal, multifocal, segmental, or generalized. A multifocal myoclonus distribution may have bilaterally synchronous movements. The temporal pattern can be continuous or intermittent as well as rhythmic or irregular. If intermittent, the myoclonus can occur sporadically or in trains. The activation of the myoclonus may be at rest (spontaneous), induced by voluntary movement (action myoclonus), induced by reflex stimuli (reflex myoclonus), or some combination of these factors. It is important to remember that patients may exhibit more than one type of myoclonus.

32.1.2. Etiological classification The major categories of myoclonus in the popular etiological classification scheme of Marsden et al. are as follows: physiologic, essential, epileptic, and symptomatic (secondary) (Marsden et aI., 1982) (Table 1). Each of the major categories is associated with different clinical circumstances. Most cases of myoclonus are in the symptomatic category, followed by epileptic and essential. The first task is to determine which major category reflects the clinical circumstances of the patient. The second task is to match the patient's specific clinical characteristics and results from diagnostic studies with those of a diagnosis under the corresponding category.

32.1.2.1. Physiologic myoclonus Physiologic myoclonus occurs in neurologically normal people. There is minimal or no associated disability and the physical exam reveals no relevant abnormality. Jerks during sleep are the most familiar examples of physiologic myoclonus. Types of sudden movement that occur during sleep or sleep

J.N. CAVINESS

522 Table 1

Table 1

Classification of myoclonus.

Continued.

I. Physiologic myoclonus (normal subjects) A. Sleep jerks (hypnic jerks) B. Anxiety induced C. Exercise induced D. Hiccough (singultus) E. Benign infantile myoclonus with feeding II. Essential myoclonus (no known cause and no other gross neurologic deficit) A. Hereditary (autosomal dominant) B. Sporadic lll. Epileptic myoclonus (seizures dominate and no encephalopathy, at least initially) A. Fragments of epilepsy 1. Isolated epileptic myoclonic jerks 2. Epilepsia partialis continua 3. Idiopathic stimulus-sensitive myoclonus 4. Photosensitive myoclonus 5. Myoclonic absences in petit mal epilepsy B. Childhood myoclonic epilepsy 1. Infantile spasms 2. Myoclonic astatic epilepsy (Lennox-Gastaut) 3. Cryptogenic myoclonus epilepsy (Aicardi) 4. Awakening myoclonus epilepsy of Janz (Juvenile myoclonic epilepsy) C. Benign familial myoclonic epilepsy (Rabot) D. Progressive myoclonus epilepsy: Baltic myoclonus (Unverricht-Lundborg) IV. Symptomatic myoclonus (progressive or static encephalopathy dominates) A. Storage disease 1. Lafora body disease 2. GM2 gangliosidosis (late infantile, juvenile) 3. Tay-Sachs disease 4. Gaucher's disease (noninfantile neuronopathic form) 5. Krabbe's leukodystrophy 6. Ceroid-lipofuscinosis (Batten) 7. Sialidosis ("cherry-red spot") (types 1 and 2) B. Spinocerebellar degenerations 1. Ramsay-Hunt syndrome 2. Friedreich's ataxia 3. Ataxia telangiectasia 4. Other spinocerebellar degenerations

IV. Continued C. Basal ganglia degenerations 1. Wilson's disease 2. Torsion dystonia 3. Hallervorden-Spatz disease 4. Progressive supranuclear palsy 5. Huntington's disease 6. Parkinson's disease 7. Multisystem atrophy 8. Corticobasal degeneration 9. Dentatorubropallidoluysian atrophy D. Dementias 1. Creutzfeldt-Jakob disease 2. Alzheimer's disease 3. Lewy body disease E. Infections/postinfectious 1. Subacute sclerosing panencephalitis 2. Encephalitis lethargica 3. Arborvirus encephalitis 4. Herpes simplex encephalitis 5. HTLV-I 6. Human immunodeficiency virus (HIV) 7. Post-infectious encephalitis 8. Malaria 9. Syphilis 10. Cryptococcus F. Metabolic 1. Hyperthyroidism 2. Hepatic failure 3. Renal failure 4. Dialysis syndrome 5. Hyponatremia 6. Hypoglycemia 7. Nonketotic hyperglycemia 8. Multiple carboxylase deficiency 9. Biotin deficiency 10. Mitochondrial dysfunction G. Toxic and drug-induced syndromes H. Physical encephalopathies 1. Posthypoxia (Lance-Adams) 2. Post-traumatic 3. Heat stroke 4. Electric shock 5. Decompression injury

THE CLINICAL NEUROPHYSIOLOGY OF MYOCLONUS

Table I

Continued.

Iv. Continued I.

J.

K. L. M.

N.

Focal nervous system damage 1. Central nervous system (a) Poststroke (b) Post-thalamotomy (c) Tumor (d) Trauma (e) Inflammation (e.g. multiple sclerosis) 2. Peripheral nerve lesions Malabsorption 1. Celiac disease 2. Whipple's disease Eosinophilia-myalgia syndrome Paraneoplastic encephalopathies Opsoclonus-myoclonus syndrome I. Idiopathic 2. Paraneoplastic 3. Infectious 4. Other Exaggerated startle syndromes Hereditary Sporadic

From Marsden et al., 1982, with modification.

transitions are "fragmentary myoclonus" and "sleep starts" (hypnic jerks). A classification of physiologic myoclonus enables the clinician to reassure all involved persons that there is no underlying pathology and that no treatment is necessary. 32.1.2.2. Essential myoclonus Essential myoclonus refers to a syndrome in which the myoclonus is usually the most prominent or only clinical finding. Thus, the myoclonus is an isolated or "essential" phenomenon, from which the patient usually experiences some, even if mild, disability. Essential myoclonus is often idiopathic and progresses slowly or not at all. Hereditary and sporadic forms exist. Hereditary essential myoclonus has been characterized by: (1) onset before age 20 years; (2) dominant inheritance with variable severity; (3) a benign course compatible with an active life and normal longevity; (4) absence of other neurologic deficits; and (5) normal electroencephalography (EEG). The myoclonus is usually distributed

523

throughout the upper body, exacerbated by muscle activation, and decreased with alcohol ingestion. The variable observation of dystonia in some cases has raised concerns about the nosology and homogeneity of this disorder. We know from genetic studies that multiple gene loci are involved in hereditary essential myoclonus. Sporadic essential myoclonus is more clinically heterogeneous than is hereditary essential myoclonus. This "entity" has the nonspecific inclusion criteria of any idiopathic case of isolated myoclonus that cannot be included in any other myoclonic category. This category likely exists because of various heterogeneous yet undiscovered causes of myoclonus without other findings, and false-negative findings on family histories. 32.1.2.3. Epileptic myoclonus Epileptic myoclonus refers to the presence of myoclonus in the setting of epilepsy - that is, a chronic seizure disorder. Myoclonus can occur as only one component of a seizure, the only seizure manifestation, or one of multiple seizure types within an epileptic syndrome. Seizures usually dominate the clinical picture in epileptic myoclonus, and the disorder is generally idiopathic. The presence of associated seizure manifestations, such as lapses in consciousness and epileptiform abnormalities on an EEG, can help to identify the underlying epileptic syndrome. The myoclonus in epileptic syndromes is presumed to be of cortical or cortical-subcortical origin. Epilepsia partialis continua usually resembles spontaneous myoclonus; it occurs irregularly or regularly at intervals no longer than 10 s, is confined to one part of the body, and continues for a period of hours, days, or weeks. The EEG usually, but not always, shows a focal abnormality appropriate for the affected region. Myoclonic seizures are epileptic seizures in which the motor manifestation is myoclonus. The myoclonus is accompanied by a generalized ictal epileptiform EEG discharge, but the myoclonus itself may be generalized, segmental, or focal. Interictal EEG abnormalities help to delineate the underlying epileptic syndrome. Juvenile myoclonic epilepsy (awakening myoclonus of Janz) is the classic idiopathic syndrome in which myoclonic seizures may occur in conjunction with generalized tonicclonic or absence seizures (or both), but without other neurologic disability in the patient.

J.N. CAVINESS

524

32.1.2.4. Symptomatic myoclonus Symptomatic (secondary) myoclonus manifests in the setting of an identifiable underlying disorder, neurologic or non-neurologic. These symptomatic syndromes are the most common causes of myoclonus. Multiple clinical manifestations exist in these patients. Often there is clinical or pathological evidence of diffuse nervous system involvement. The other clinical manifestations that constitute the syndrome are significant and may even be more prominent than the symptom of myoclonus. Chronic clinical progression suggests symptomatic myoclonus. Mental status abnormalities and ataxia are common clinical associations in symptomatic myoclonic syndromes, and the cortex is the most commonly proven source of the myoclonic jerks. Symptomatic causes of myoclonus comprise a widely diverse group of disease processes and include storage diseases, degenerative conditions, toxic-metabolic states, physical processes, infections, focal nervous system damage, and paraneoplastic syndromes as well as other medical illnesses. 32.1.3. Physiological classification

Etiological classification provides a framework to match a patient's myoclonus to an etiology from a comprehensive list of disorders. However, there are at least three advantages to classifying the myoclonus with regard to its physiology. First, physiology can provide localizing information for the myoclonus and thus can provide at least partial localization for diagnosis of the underlying process. Second, some physiological myoclonus types are characteristic for certain disorders, so identifying their presence can aid in identifying certain underlying diagnoses. Third, certain myoclonus treatments are known to have variable success, depending on the physiological myoclonus type being treated. Piracetam, for example, can be effective for a cortical myoclonus physiology, but not for subcortical or segmental physiology. In addition, comparing and contrasting the myoclonus physiology in various disorders provides insights about the disease processes that create them. The myoclonus physiology in the essential or epileptic categories is fairly similar within the clinical category, but the types of myoclonus physiology seen in symptomatic myoclonus are much more variable. Thus, as with many

other clinical problems, neurophysiology acts as a useful extension of the clinical evaluation. 32.2. Methodological considerations in the clinical neurophysiological study of myoclonus

The methods used in the neurophysiological study of myoclonus for a specific patient are guided by the clinical circumstances. However, it is useful to perform a battery of methods that includes multichannel surface electromyography (EMG) recording with testing for long latency EMG responses to nerve stimulation (Table 2), EEG, EEG-EMG polyTable 2 Outline of the surface EMG clinical neurophysiological examination of myoclonus. I. Basic EMG properties A. Identification B. Duration C. Variability D. Positive vs. negative myoclonus II. Spatial characteristics A. Distribution 1. Focal 2. Segmental 3. Multifocal 4. Generalized B. Recruitment order I. Ascending and descending patterns 2. Occurrence of bilateral synchrony Ill. Timing characteristics A. Continuous vs. intermittent B. Rhythmic vs. non-rhythmic C. Frequency D. Recruitment speed IV. Activation characteristics and long latency EMG responses A. Rest - with and without mental activation B. Action - static posture and directed movement C. Task (e.g. handwritting) D. Associated movement remote from the myoclonus E. Mental state F. Reflex properties

525

THE CLINICAL NEUROPHYSIOLOGY OF MYOCLONUS

775

vs

Normal Ballistic Pattern

1750

Myoclonus

A

f-

B

_._~

... ..

..

~

...

................

~ -755

-1500

f-

,

I

o

50

100

150

200

250 "

0 Milliseconds

50

100

150

200

250

Fig. I. A shows the surface EMG pattern from a normal voluntary ballistic movement of wrist flexion. The subject was instructed to perform the wrist flexion as quick and as brief as possible. In B, there is a myoclonus surface EMG wrist flexor discharge from a patient with multifocal action myoclonus. Despite the fact that the normal ballistic movement was performed as brief as possible, note that there is still a more gradual build-up of activity when compared to the involuntary myoclonus EMG discharge. Modified from Caviness, 1996.

graphy with back-averaging, and evoked potentials. Positive and negative findings from these methods can then be used to provide evidence for determining the physiological type of myoclonus. 32.2.1. Multichannel surface EMG recording

The multichannel surface EMG recording is the most basic and important part of the myoclonus neurophysiological examination. Careful inspection of the recording can provide some initial pattern identification as well as determine the course of further neurophysiological study. The design of the multichannel surface EMG recording must be based on the clinical characteristics of the myoclonus as well as the technique's ability to add to the information that was provided by the clinical evaluation. Recording samples should be taken during rest, postural activation (e.g. arms out-

stretched), kinetic activation (e.g. finger to nose, handwriting), mental activation (e.g. counting backwards), and while performing associated movements (e.g. contralateral repetitive hand movements). Important attributes of the myoclonus discharges that need to be ascertained during the multichannel surface EMG recording are outlined below. 32.2.1.1. Basic EMG properties

Identification of the myoclonus EMG discharge begins with its correlation to the myoclonus movement. Myoclonus occurs as a brief burst of increased EMG activity and usually, but not always, can be distinguished from normal voluntary ballistic activity (Fig. 1). It appears as a highly synchronous discharge of multiple motor units and clearly stands out of background EMG activity (Fig. 2). The duration of the myoclonus EMG discharge is most

526

J.N. CAVINESS

Myoclonus EMG Discharges +200

I

Microvolts

Right Deltoid

o

200

600

400

800

1000

MIIIIs.conda

Fig. 2. Multichannel surface EMG recording from the right deltoid during postural activation. Arrows point to prominent myoclonus EMG discharges. Note how the myoclonus EMG discharges distinctly arise from the background activity and interrupt the smooth tonic EMG activity.

commonly 100 ms or less. However,jerks with EMG discharge duration in the 100-250 ms range can still fulfill the clinical definition for myoclonus. It is unusual for a myoclonus EMG discharge duration to be greater than 250 ms. This is because the clinical definition describes a jerk that is "brief, lightninglike," and muscle contractions that last longer than 250 ms usually do not produce such a short lasting movement. Periodic limb movements of sleep have been called myoclonus in the past. However, the EMG discharges associated with these movements last 1-2 s or longer, and the movements themselves are far too long to be considered "brief, lightninglike." Thus, the term "periodic limb movements" is preferred. On the other hand, all EMG discharges less than 100 ms are not associated with myoclonus. The EMG duration from a single discharge of exaggerated physiologic tremor, other tremors, chorea, or other movement disorders may be less than 100 ms. This observation demonstrates the impor-

tance of confirming that the suspected myoclonus EMG discharge is occurring in a patient who has the clinical definition of myoclonus. The myoclonus EMG discharge often has a stereotyped appearance. It can occur in an agonistonly pattern or can demonstrate agonist-antagonist co-contraction. When contiguous muscle segments are involved, the discharges usually appear nearly synchronous but still have a small time lag depending on the physiology of the recruitment pattern. The muscle involvement at a particular limb location may shift slightly from jerk to jerk. There is typically an irregular time interval from one discharge to the next when repetitive jerks occur. A short period of EMG silence may be present before and after the EMG discharges, particularly when the myoclonus is occurring repetitively in trains. When the discharges have a very short duration, such as less than 50 mscec, there is little variability in duration from discharge to discharge. However, when the average

THE CLINICAL NEUROPHYSIOLOGY OF MYOCLONUS

discharge duration is greater (e.g. > 100 ms), the variability in duration can be much greater. 32.2.1.2. Spatial characteristics Distribution is important to assess in the multichannel surface EMG recording. Surface EMG coverage should be sufficiently extensive to encompass and define the borders of the affected areas if possible, even if those muscles are not visibly jerking. Because the EMG may better clarify the distribution of involvement than what was apparent on the clinical exam, it can provide more accurate information as to the nature and extent of the nervous system pathology. This is not to suggest that a direct correlation between myoclonus distribution and the pathology distribution always exist. For example, focal myoclonus may not mean that the nervous system pathology only focal, and generalized myoclonus has the potential to arise from a very circumscribed location. The most common distribution patterns of myoclonus are multifocal, generalized, and segmental. Focal myoclonus can also occur. Except for focal motor seizures, a hemi-distribution of myoclonus is unusual. Diffuse encephalopathic processes commonly create a multifocal distribution of myoclonus. Recruitment order is very important for determining the physiological type of myoclonus. If certain muscles reproducibly begin the recruitment order, then this could suggest that the myoclonus generation site is closest to where the lower motor neurons for that muscle or muscles originate. For example, consistent initial recruitment of trapezius and sternocleidomastoid muscles, followed by simultaneous, sequential, ascending and descending recruitment of brainstem and spinal cord activated muscles may suggest a brainstem generator. A cortical source should produce only a descending order of recruitment. When a certain muscle or group of muscles is bilaterally recruited in an approximately simultaneous manner, this is referred to as bilateral synchrony. Even when myoclonus is occurring predominantly in a multifocal distribution, bilateral synchrony can exist. For example, one mechanism for bilateral synchrony is fast interhemispheric spread from a unilateral cortical generator. For the most accurate assessment of recruitment order, or when it is difficult to determine, peripheral conduction times should be measured by using

527

magnetic stimulation. In this way, the recruitment order may be appropriately adjusted for the differential peripheral conduction time. 32.2.1.3. Timing characteristics Gross timing characteristics are usually apparent on clinical examination. However, many timing aspects can be confirmed or more accurately assessed by examining the multichannel surface EMG recording. By common sense, myoclonus must be continuous or intermittent. If repetitive EMG discharges occur either continuously or in trains, the degree of rhythmic character or lack thereof should be noted. If rhythmic or semi-rhythmic, the frequency or variability of frequency is recorded. These timing characteristics add specific information for determining the physiological type of myoclonus being examined. The recruitment speed, i.e. how fast the activation of one muscle is followed by the activation of the next, can reveal important clues about the myoclonus generation. This is because certain central nervous system conducting pathways, both descending and ascending are thought to have different conduction speeds. Since such pathways often arise from characteristic generation sites (pyramidal vs. non-pyramidal), information on the myoclonus generation site may be obtained. The same type of adjustments using magnetic stimulation techniques can be made to recruitment speeds as those mentioned above for recruitment order. 32.2.1.4. Activation characteristics and long latency EMG responses The circumstances under which myoclonus increases, decreases, or stays the same has long been used by clinicians as a major descriptive characteristic. The multichannel surface EMG recording may be used to document and extend these observations. Recording for 1-5 minutes in the resting state may identify small amplitude EMG discharges that produce small jerks which otherwise can be missed by visual inspection alone. The increase in myoclonus that often accompanies muscle action is usually easily appreciated clinically. However, surface EMG can sometimes better capture the differences between the various methods of eliciting the muscle action. In addition, surface EMG is useful for making certain that the subject is in the desired state of muscle activation (rest vs. action). Movement of

528

one body part may influence the activation of myoclonus in another limb. The determination of any influence that mental state may have on the myoclonus usually necessitates simultaneous EEG (EEG-EMG polygraphy). The ability of certain stimuli to elicit myoclonus has been appreciated for several decades. Touch, muscle stretch, light, sound, and startle are all stimuli that have been shown to elicit myoclonus. The multichannel surface EMG recording, combined with the trigger signal displayed on one channel, can reliably document the presence/absence and consistency of the reflex sensitivity for a certain case of myoclonus. Electrical stimulation of the median and! or tibial nerve and digital cutaneous stimulation can prove useful for eliciting myoclonus EMG discharges, even in cases where no reflex myoclonus is found by clinical examination. A "long latency" EMG response characteristically begins between 40 and 60 ms after median nerve stimulation. Such reflex properties elicited by electrical stimulation have characteristic findings for certain types of myoclonus physiology and their associated etiologies. 32.2.2. Electroencephalography

A routine EEG with standard 10-20 electrode placement and technique according to accepted guidelines is a necessary part of the myoclonus neurophysiological evaluation (Sharborough et al., 1991). The EEG recording is necessary to: (1) assess background rhythm; (2) presence or absence of normal/abnormal rhythms and transient phenomena; (3) effect of different mental states (e.g. sleep) on the myoclonus; and (4) the effect of routine activating procedures on the myoclonus including mental activation, hyperventilation, and photic stimulation. These findings have an impact on the diagnosis of the patient's disorder as well as the physiological classification of the myoclonus. Even if more formal EEG-EMG polygraphy for the myoclonus will be performed, adding one or more channels for surface EMG monitor during the routine EEG recording is useful for detecting gross relationships (or lack thereof) between EEG events and the myoclonus. In the case of epileptic myoclonic syndromes, the routine EEG is often abnormal. Absence syndromes, juvenile myoclonic epilepsy, Lennox-Gastaut syndrome, and infantile spasms all have ictal and

J.N. CAVINESS

interictal EEG characteristics that support their diagnosis. The EEG can also reveal other abnormal rhythms and discharges typical of certain symptomatic myoclonus etiologies. Examples include the periodic discharges of subacute sclerosing panencephalitis (SSPE) and Creutzfeldt-Jakob disease (CJD) (Westmoreland, 1987; Brenner and Schaul, 1990). Slow background rhythms, central fast rhythms, and central midline spikes are nonspecific findings that have been identified in symptomatic myoclonus (Kelly et al., 1978; Witte et al., 1988). These EEG findings, even in the context of epileptic syndromes, may just be variably associated or never seen at the approximate time of the myoclonus. In the 1930s, it was observed that myoclonus could grossly correlate with abnormal EEG activity. During a single myoclonic jerk, an epileptogenic discharge such as a spike, sharp wave, or spike and wave can sometimes be identified. Rhythmic or semi-rhythmic discharges can occur simultaneously with trains of repetitive jerking. These discharges can appear epileptogenic or sinusoidal as "central fast rhythms" (Kelly et al., 1978). However, a key concept in the use of evaluating myoclonus with EEG is that the routine EEG may be completely normal or have changes with questionable correlation and significance with respect to the myoclonus. To clarify the relationship between EEG and myoclonus EMG discharges, EEG-EMG polygraphy with back-averaging is crucial.

32.2.3. EEG-EMG polygraphy with back-averaging

This technique brought a major advance to the neurophysiological examination of myoclonus (Shibasaki and Kuriowa, 1975). The EEG-EMG polygraphy allows simultaneous observation of both EEG and EMG events, thus enabling the examiner to observe the correlation or lack thereof between myoclonus EMG discharges and EEG events such as sharp waves, mental state, etc. Back-averaging has two major functions: (1) to increase signal-to-noise ratio in order to permit detection of low amplitude EEG events; and (2) to provide evidence for a timelocked relationship between the myoclonus EMG discharges and EEG events. For recording myoclonus, one hundred epochs are usually necessary to achieve a suitable averaged tracing. At least 200

529

THE CLINICAL NEUROPHYSIOLOGY OF MYOCLONUS

epochs or more should be averaged for a high quality recording, or to provide evidence that no transient is present when using EEG-EMG back-averaging. The number of EEG electrodes that should be used is at least six or seven to cover appropriate primary sensory and motor areas and have an averaged ear or mid-frontal (e.g. Fz) reference. Standard 10-20 positions are now available for placement over primary motor and sensory areas (Sharbrough et aI., 1991). The EMG discharges are sometimes manually marked off-line for trigger placement, or alternatively, the rectified EMG discharge is used to trigger automatically, on-line or off-line. The time epoch should cover at least 200 ms before and 200 ms after the trigger line, but longer intervals may be appropriate in certain circumstances. There are caveats and possible pitfalls when backaveraging is used. The myoclonus EMG discharge that is used as a trigger should be stereotyped and allow for the obvious placement of the trigger mark, rather than having a gradual or uncertain beginning of the discharge. Thus, if the EMG event is not welldefined or reproducible, the averaged EEG event may have uncertain significance or correlate to a poor average of very different EMG phenomena. In addition, just because an averaged EEG waveform is present before the EMG discharge, it may not mean that the waveform occurred in the same morphology, amplitude, and distribution every time. In order to evaluate the averaged waveform, partial averages should be evaluated individually, such as two 50-epoch averages for a 100-epoch tracing or two 100-epoch averages to establish the reproducibility of a 200-epoch averaged tracing. The morphology and distribution of the EEG transient should fit the physiology being proposed and be consistent with a definable dipole field. The recording to be analyzed should be as artifactfree as possible since many sources of artifact do not completely disappear with averaging. One example is muscle artifact in the EEG. The source of the muscle activity may be myoclonus occurring around the facelhead region, or tensing of the jaw or forehead muscles. If the subject must activate a muscle or concentrate in order to produce the myoclonus, it may be more difficult to obtain EEG recording free of muscle or movement artifact. Eye movements should be monitored with eye leads as such artifact (e.g. blink) can significantly disturb averaged waveforms.

32.2.4. Evoked potentials Evoked potential studies have characteristic findings for various types of myoclonus physiology. The most studied evoked potential in myoclonic disorders is the somatosensory evoked potential (SEP). The SEP is recorded using median nerve stimulation for the upper extremities and tibial nerve stimulation for the lower extremities. The cortical waves over the contralateral centro-parietal (CP) electrodes, produced by median nerve stimulation, reveal a negative wave at 20 ms (N20), positive wave at 25 ms (P25), and another negative wave at 33 ms (N33) (Allison et aI., 1991). The enlargement of the parietal cortical SEP waves supports a cortical origin for the myoclonus (Obeso et aI., 1986; Shibasaki, 2000). The P25-N33 deflection is usually the enlarged wave in this instance, while the N20-P25 wave is usually within normal limits. If multiple types of myoclonus are occurring in an individual, the SEP finding may not have relevance for all the types of myoclonus physiology being observed. During the SEP recording, EMG leads may show long latency responses. Photic flash stimulation may provide enlarged averaged potentials at occipital and central electrodes. Conventional visual evoked potentials may either be enlarged or reduced in some diseases with myoclonus, and abnormal electroretinograms are seen in certain disorders but probably do not influence the myoclonus physiology (Berkovic et aI., 1986; Rapin, 1986). 32.3. Physiological classification of myoclonus using clinical neurophysiology

Major categories of the physiological classification used here primarily refer to the neuroanatomic source of the myoclonus physiology types (Table 3). Further subdivision is based on other physiological properties as well as the clinical syndrome and/or the specific disease in which the myoclonus occurs. In practical terms, the classification for a particular example of myoclonus is derived from the clinical neurophysiological findings as well as an appreciation of the clinical context in which they occur. The main physiological classification categories for myoclonus are cortical, cortical-subcortical, subcortical, segmental, and peripheral. One should be aware that multiple myoclonus physiology types might occur in the same patient. As with any classification system, this physiological classification of myoclonus

530 Table 3 Physiological classification of myoclonus. • Cortical 1. Cortical reflex myoclonus 2. Cortical origin myoclonus without reflex activation 3. Focal motor seizures 4. Alzheimer's disease 5. Creutzfeldt-Jakob disease 6. Subacute sclerosing panencephalitis 7. Myoclonus of corticobasal degeneration 8. Asterixis • Cortical-subcortical 9. Absence seizures 10. Primary generalized myoclonic seizures 11. Primary generalized epileptic myoclonus • Subcortical 12. Essential myoclonus 13. Reticular reflex myoclonus 14. Opsoclonus-myoclonus syndrome 15. Propriospinal myoclonus 16. Focal subcortical reflex myoclonus • Segmental 17. Brainstem 18. Spinal cord 19. Combined brainstem and spinal • Peripheral 20. Hemifacial spasm 21. Other

* More than one mechanism may be present in a given patient! disease. reflects and is dependent upon the current state of knowledge and technical abilities. It may be that such a classification does not reflect the true complexity of myoclonus physiology, because of our current inability to detect and understand contributions from multiple sites in the nervous system, as well as the details about neurochemical influences and neuronal circuits. The only thing that one may consider to be certain is that our understanding of myoclonus physiology will evolve. 32.3.1. Cortical 32.3.1.1. General comments The cerebral cortex is the most common origin for myoclonus. The jerks are most often multifocal, but

IN. CAVINESS

focal, segmental, and generalized myoclonus can also occur. Action myoclonus is very common in these patients and provides most of the disability. At rest, myoclonus will usually be less prominent unless the major clinical manifestation is a focal motor seizure. Myoclonus induced by reflex stimulation often occurs and its characterization is important for physiological classification. It is the most common situation for a patient to have myoclonus with a combination of action and reflex precipitants, and presence at rest. The vast majority of cortical myoclonus patients have one or more of the three major cortical physiology types: (1) cortical reflex myoclonus; (2) cortical origin myoclonus without reflex activation; and (3) focal motor seizures. More unusual physiological descriptions have been reported for Alzheimer's disease, Creutzfeldt-Jakob disease, subacute sclerosing panencephalitis, corticobasal degeneration, and asterixis. All of these physiological descriptions of cortical origin myoclonus will be discussed below. The establishment of a cortical origin for the myoclonus may have diagnostic implications. Such etiologies that may demonstrate cortical myoclonus include post-hypoxic syndrome, progressive myoclonus epilepsy syndromes, drugs and toxins, neurodegenerative syndromes, various dementias, focal lesions, and other entities of unknown cause, both sporadic and familial (Table 4). No doubt, as the techniques discussed here are applied more commonly, more etiologies will be cited to cause cortical myoclonus. If cortical myoclonus is identified, there are treatment implications (Brown, 1995). For example, Piracetam is only indicated for cortical origin myoclonus. Valproic acid and clonazepam are also usually very useful in the treatment of cortical myoclonus. However, the drug tetrabenazine and anticholinergic agents may suppress subcortical and segmental types of myoclonus, but they are usually not effective in cortical myoclonus. 32.3.1.2. Cortical reflex myoclonus Dawson observed an exaggerated cortical reflex response in conjunction with myoclonus in 1947 (Dawson, 1947). In a seminal article, Hallett et al. coined the term, "cortical reflex myoclonus" in 1979 (Hallett et al., 1979). This type of cortical myoclonus physiology is the predominant type in post-hypoxic myoclonus or Lance-Adams syndrome, progressive myoclonus epilepsy syndromes, toxic and drug-

THE CLINICAL NEUROPHYSIOLOGY OF MYOCLONUS

Table 4

531 Back-Averaged Myoclonus CorticalCorrelate

Etiologies for which cortical myoclonus has been described. Post-hypoxic myoclonus Progressive myoclonic epilepsy syndromes - Unverricht-Lundborg disease - Mitochondrial disease - Ceroid lipofuscinosis - Lafora body disease - Sialidosis Drugs and toxins - Tricyclic anti-depressant medication - Lithium - Levodopa - Methyl bromide Neurodegenerative syndromes - Alzheimer's disease - Parkinson's disease - Multiple system atrophy - Spinocerebellar degeneration - Huntington's disease Creutzfeldt-Jakob disease Subacute sclerosing panencephalitis Celiac disease Rett's syndrome Down's syndrome Angleman's syndrome Focal lesions from numerous causes Syphilis Traumatic encephalopathy Unknown-sporadic Unknown-familial

induced myoclonus, and in many other etiologies. Cortical reflex myoclonus is defined by the demonstration of a focal time-locked cortical transient that precedes the myoclonus by a short latency «40 ms for arm) in association with evidence for exaggerated reflex cortical phenomena. This may include one or more of the following: (1) enlarged cortical SEP waves; (2) reflex-induced myoclonus; and (3) enhanced long latency EMG responses to electrical nerve stimulation. The myoclonus EMG discharge duration is usually 50 ms or less, and agonistantagonist co-contraction is common. A case of cortical reflex myoclonus is demonstrated in Fig. 3. Although spikes and/or sharp waves are sometimes present in the gross EEG, back-averaging of the EEG-EMG polygraph is the preferred method for demonstrating a time-locked cortical transient pre-

....

•

"

...I

..:.-

~.~, 1IO 0 III

•

"

..

t I'l

~1:~:t1

~

1II

~

EnlargedSEP

LongLatency EMGReflexes

,",.,.--,r--.---rnrr----------,

10

..

Fig. 3. A shows the back-averaging of a focal cortical transient preceding averaged left arm myoclonus EMG discharges in a patient with cortical reflex myoclonus. B shows the enlarged cortical SEP from the same patient. Note the similarity of the P25-N33 wave dipole in 3B to that of the pre-myoclonic back-averaged transient in 3A. In both instances, there is positive wave in the CP4 electrode with a simultaneous FC4 negative wave with an averaged ear electrode reference. C shows an enhanced Abductor Pollicis Brevis long latency EMG reflexes at rest to median nerve stimulation at 53 ms and a smaller wave at 84 ms. Thus, the features of A-C demonstrate the classic electrophysiological triad of cortical reflex myoclonus. FC and CP represent standard 10-20 EEG electrode fronto-central and centro-parietal positions, respectively. Modified from Caviness and Kurth, 1997.

532

ceding a myoclonus EMG discharge. Although such patients usually have reflex jerks, it is generally easier to collect many myoclonus events by muscle activation of the limb. The transient is typically a focal, biphasic or triphasic spike beginning with a positive deflection that precedes the onset of the myoclonic discharge by 6-22 ms in the upper extremity: the more distal muscle the myoclonus is recorded from, the longer the time interval (Shibasaki, 2000). The duration of the transient is 15-40 ms. The conduction of the spike to motomeuron pools is presumed to occur by corticospinal (pyramidal) pathways. The maximum of the transient is usually located over the sensorimotor cortex at the central or centro-parietal electrode according to anatomical somatotopic mapping, contralateral to the myoclonus EMG discharge (Fig. 3A). Enlargement of the cortical SEP P25-N33 parietal wave from median nerve stimulation is important evidence for cortical reflex myoclonus physiology (Fig. 3B). The establishment of normal values for a particular laboratory is encouraged with consistent methods and electrode derivations being used. Shibasaki et al. published an upper limit for P25N33 amplitude at the post-central electrode of 8.6 f.LV using an ear reference, Ugawa used 10.8 f.LV with Fz reference, and the normal upper limit value in our laboratory is 11.1 f.LV with Fz reference (Shibasaki et al., 1977; Ugawa et al., 1991). Sometimes the cortical SEP waves are "giant" and deviate from the morphology and distribution seen in normal individuals. The definition of "giant" SEP is arbitrary but> 20 f.LV is a commonly used value. In addition to the P25-N33 wave, the parietal N20-P25 and/or frontal P22-N30 are enlarged less often. A key characteristic of the enlarged P25-N33 wave is the similar morphology and topography to the averaged time-locked EEG transient that precedes the myoclonus EMG discharge elicited by muscle action or at rest (Fig. 3A and B). Additionally, the interval between the P25 peak and the onset of any long latency EMG response, which is simultaneously recorded, is usually similar to the latency from the back-averaged EEG transient to the onset of the myoclonus EMG discharge. The reflex myoclonus may be clinically demonstrable by touch or muscle stretch. Often, in the case of upper extremity myoclonus, briskly abducting the thumb will evoke a reflex myoclonic jerk. It is useful to confirm this with EEG-EMG polygraphy. In this

J.N. CAVINESS

way, one can be certain that there is an appropriate latency between the stimulus and myoclonus EMG discharge that is consistent with a transcortical reflex. A reproducible gross EEG transient mayor may not precede the myoclonus EMG discharge with each stimulus. It is usually easier to document the exaggerated reflex myoclonus by testing for long latency EMG responses to electrical nerve stimulation (Fig. 3C). For the hand, median nerve stimulation can show EMG discharges at 50 ms latency or greater (range 40-60) from the stimulus artifact trigger mark (Shibasaki, 2000). Repetitive discharges may be seen, at intervals of 20-40 ms (Caviness and Kurth, 1997). At rest, in a normal individual, no response should be present. Care must be taken that the arm muscles are relaxed so as to avoid a false positive response. In the cerebellar presentation of multiple system atrophy, a photic cortical reflex myoclonus has been described (Rodriguez et aI., 1994). In these cases, the occipital potentials have normal amplitude and precede the bilateral frontal potentials that are time-locked before the generalized myoclonus (Artieda et al., 1993). Obeso et ai. pointed out three types of commonly encountered associations between enlarged SEPs and reflex-induced myoclonus (Obeso et aI., 1986). The first type is a unilaterally enlarged SEP in focal limb reflex myoclonus. Thus, the abnormal physiology is limited to a small area of cortex. The second type is bilaterally enlarged SEPs in multifocal or generalized myoclonus. This is the most common situation and occurs in the setting of a diffuse encephalopathy. The third type is enlarged SEPs without reflex myoclonus. This shows that there can be dissociation between the two mechanisms responsible for the production of each. In addition, the SEP may not fulfill enlargement criteria even though there is reflex-induced myoclonus or enhanced long latency EMG responses. Brown et al. examined bilateral and generalized jerks in the setting of cortical reflex myoclonus (Brown et al., 1991b). They found that intrahemispheric and interhemispheric spread in a grossly somatotopic fashion from a focus in one hemisphere can produce these bilateral and/or generalized jerks. Because of the fast spread, the clinical jerking appears almost synchronous (Fig. 4). The intrahemispheric spread was posited to occur in the primary sensorimotor cortex and the interhemi-

533

THE CLINICAL NEUROPHYSIOLOGY OF MYOCLONUS

Myoclonus EMG Discharges with Bilateral Synchrony Left DIIloId Left~

J ',1

.1f .I'f

or

Of

II

I

Left_Polllc........

, I

Left FlrwlIlofMI1 _

II

,

T

I I .II I

,

I

,

I

.,

1

I

Left AbductDrD1g1ll Qulntl

I

I

,

Left WNt Edonaorw

iI

f , r

LeftTrteep LeftWNtF..xo..

,.

r

I

.

rv:

I'

I

P

r

Right Deltoid

I

Microvolla

I

f

I

+1000 I

., n, .IT .r'I I

.1

Rlghtllcepe RlghtT...... I

RIghtWNt F..SON RIght WNt ExIen_ RlghtAbd_ Polllc........

/'-

J

RIght AbeluclllrD1g1ll Qulntl

I

~

I

RlghtFI..t l l o f M l l _

•A

............ V¥~~

o

..J.

~~t

j

I

Y

I

~ 3

2

Seconds

Long Latency EMG Reflexes

Left Abductor PoIllcla 1_"

Left FII8tDorul 1 _ 1------"11

Left_tFlexora

v--------------tf

I - - - - -J n- - - - - - - - - - - - - - r

Left Wrist ElllieIlaora

RIght Abductor PoIIIcla &NY"

1----------------------"

R1ghtFII8tDolMII_I----------------------' Rlght-.Flexora I--------------------~ R1ghtWrlatEllliellaora

1-

-' l2J.O

Milliseconds

Fig. 4. A shows the multichannel surface EMG recording from a patient with cortical reflex myoclonus during postural activation of the arms. Some of the myoclonus EMG discharges appear to occur with almost synchronous timing bilaterally (bilateral synchrony). From the same patient, B shows the occurrence of myoclonus EMG discharges elicited by median nerve stimulation of the left abductor pollicis brevis. Note that contralateral discharges occur in contralateral homologous muscles with an approximate lO-ms delay in the same patient. This suggests that interhemispheric transfer produces the contralateral myoclonus EMG discharges in A and B.

534

spheric spread to occur by myelinated transcallosal fibers. This concept may also have relevance to the spread of seizure activity in these patients. Cortical tremor refers to relatively rhythmic distal upper extremity EMG discharges during action at approximately 9 Hz and duration around 50 ms (Ikeda et al., 1990). Despite the phenotypic designation of "tremor," these discharges were found to fit all the criteria of cortical reflex myoclonus physiology. In the consensus statement on tremor, the position was taken that this entity would be more accurately referred to as myoclonus (Deuschl et al., 1998). This is an argument of phenomenology and visual appearance vs. mechanism, as well as the dividing line between rhythmic vs. arrhythmic. The significance of cortical reflex myoclonus and its associated reflex features has been debated (Obeso et al., 1986; Shibasaki et al., 1986). There is strong evidence for the importance of the transcortical mechanism in eliciting reflex jerks. However, whether or not hyperexcitable cortical reflex mechanisms play a role in the action myoclonus or myoclonus at rest in the same individuals with cortical reflex myoclonus is unclear (Deuschl et al., 1991). There has also been dissociation documented between the enlarged cortical SEP observation and reflex myoclonus in individuals with cortical reflex myoclonus physiology (Obeso et al., 1986).

32.3.1.3. Cortical origin myoclonus without reflex activation The establishment of cortical reflex myoclonus as a distinct cortical physiology was an important step for the study of myoclonus mechanism. However, it has become apparent that myoclonus may have a focal time-locked cortical transient that precedes the myoclonus but is unassociated with clinical reflex myoclonus, enhanced long latency EMG responses to electrical nerve stimulation, or enlarged cortical SEP waves (Fig. 5). This physiology has been seen with myoclonus occurring in Parkinson's disease, dementia with Lewy bodies, hereditary diffuse Lewy body disease, drugs, and other conditions (Caviness et al., 1998,2000; Evidente and Caviness, 1999). This physiological type of cortical myoclonus usually occurs in the setting of small amplitude jerks. One explanation for the absence of reflex myoclonus in this instance may be that the reflex cortical hyperexcitability in these cases is so mild that it is not detectable with current methods, despite

J.N. CAVINESS

the fact that the physiology is still identical to that seen in cortical reflex myoclonus. Although possible, there are arguments against this concept. Small amplitude jerks, such as in the cases referred to as "cortical tremor," can demonstrate cortical reflex myoclonus physiology (Ikeda et al., 1990). Further, the size of the EEG correlates reported without reflex sensitivity or enlarged SEP's are just as large as in cortical reflex myoclonus. This remains an area for further clarification.

32.3.1.4. Focal motor seizures This myoclonus manifestation is perhaps more often thought of as partial epilepsy with motor symptomatology. Nevertheless, the basic movement phenotype is usually of focal myoclonus, either occurring as paroxysms of repetitive focal jerks, or as epilepsia partialis continua when occurring for extended periods of time. Focal or more widespread cerebral cortical processes can cause focal motor seizures. There are a variety of ictal EEG changes that may be seen in the contralateral motor area appropriate for the focal motor seizure manifestation. Repetitive focal spike, spike and wave, sharp wave, rhythmic theta or delta activity, or desynchronization may occur. In many cases, no grossly observable EEG activity is seen, and back-averaging may uncover a transient is some of those cases. In the case of epilepsia partialis continua, the above-mentioned transients will be periodic and may even occur in the pattern of periodic lateralizing epileptiform discharges (PLEDS). The EMG discharge duration is usually less than 100 ms. In some cases that have been studied in detail, enlarged cortical SEP waves and enhanced long latency reflexes have been found (Obeso et al., 1985). However, these cases had other clinical presentations of myoclonus in addition to focal motor seizures. 32.3.1.5. Alzheimer's disease The myoclonus in Alzheimer's disease is usually multifocal, although it can be generalized. The occurrence of the jerks may be at rest, with action, or stimulus induced. It is common for all the abovementioned phenotypic characteristics to occur in a single patient. A few different electrophysiological types of myoclonus have been described in Alzhei-

THE CLINICAL NEUROPHYSIOLOGY OF MYOCLONUS

535

Myoclonus in Parkinson's Disease

LoIlWrilt Ext.nlOra

+2BB.B

I

Microvolts

LollAbductor

Poillcil B...II

B

3BB

6BB

9BB

12BB

lSBB

18B8

21BB

24BB

3BBB

278B

MIIIIsBconds

Back-Averaged Myoclonus Cortical Correlate

L Wrist Extensors

+200

EMG

I

Mlcrovoha

'r .1..

.....

SOms

I

-10 EEG

+

Mlcro.olla

SOms

Fig. 5. A shows multichannel surface EMG recording from a patient with autopsy confirmed Parkinson's disease during postural activation of the left arm. Myoclonus EMG discharges are less than 100 ms duration. The time interval between the arrows denotes a train of myoclonus EMG discharges. B shows the back-averaging of a focal cortical transient prior to the average myoclonus EMG discharge. The SEP in this patient was not enlarged and no long latency EMG responses were present. Modified from Caviness et al., 1998.

536

mer's disease (Wilkins et al., 1984; Ugawa et al., 1987). The gross EEG can show background slowing and abnormal slow waves. Focal sharp waves or sharp and slow waves may occur. Periodic or quasi-periodic sharp waves sometimes occur with similarity to Creutzfeldt-Jakob disease. A relationship of these gross EEG changes and events to myoclonus is usually not clear. The myoclonus EMG discharges are less than 100 ms, and may occur in an agonist-only pattern or with co-contraction in antagonists and other muscles. Enlargement of the cortical SEP waves and the presence of long latency EMG responses to median nerve stimulation are variable. Correlation between the time-locked EEG event and the myoclonus EMG discharge usually exists, but the characteristics have been reported to vary. The most commonly reported instance is a focal contralateral central negativity with onset 20-40 ms pre-myoclonus EMG latency and 40-80 ms duration. Because of these characteristics, Wilkins et al. reported that the differentiation from CreutzfeldtJakob disease could be useful (Wilkins et al., 1984). However, even though Ugawa and coworkers also found a short latency and duration of the EEG correlate in their series, they also reported other cases where the latency to the myoclonus EMG discharge was 60-100 ms and the duration was 180-290 ms. Ugawa et al. pointed out that these cases with longer latencies and longer EEG correlate duration do overlap with the physiology reported for Creutzfeldt-Jakob disease (see below) (Ugawa et al., 1987). A bifrontal negativity with a latency preceding the myoclonus EMG discharge by 50-170 ms and lasting 100-180 ms has been seen in a few cases of Alzheimer's disease (see below) (Wilkins et al., 1985). In addition, one case was reported which showed no time-locked EEG event whatsoever. The meaning of the varied myoclonus physiology seen in Alzheimer's disease is not clear. Other factors, such as degree of severity, duration of illness, other system involvement (e.g. parkinsonism) may play a role. Even though the most common physiology described for Alzheimer's disease is believed to resemble cortical reflex myoclonus, there are notable differences. In cortical reflex myoclonus, the duration of the EEG correlate is shorter, the association of enlarged cortical SEP waves and enhanced long latency reflexes is stronger, and the presence of a positive peak

IN. CAVINESS

preceding the negativity is a more prominent finding than in Alzheimer's disease. 32.3.1.6. Creutzfeldt-Jakob disease The myoclonus in Creutzfeldt-Jakob disease can occur in early, middle, or late stages. Its clinical presentation can vary, and focal, multifocal, or generalized jerks may occur. The jerks can be rhythmic or arrhythmic, and stimulus sensitivity (somatosensory, startle, light) is common. The gross EEG findings of an abnormal slow and/or suppressed background and generalized periodic sharp wave discharges are well known. The EMG duration is <50 ms and an agonist-only pattern or with cocontraction in antagonists and other muscles is observed. There is a variable correlation between the timing of the myoclonus and the sharp wave discharges on routine EEG. When back-averaging is used, a broadly distributed contralateral negative transient is seen (Shibasaki et al., 1981). This EEG correlate has 100-160 ms duration and latency to the myoclonus EMG discharge of 50-85 ms. Enlargement of the cortical SEP waves and enhanced long latency reflexes is variable (Kelly et al., 1981; Ugawa et al., 1991). A photic cortical reflex myoclonus physiology has also been described in patients with Creutzfeldt-Jakob disease (Shibasaki and Neshige, 1987). 32.3.1.7. Subacute sclerosing panencephalitis These patients can show periodic movements that appear as a jerk followed by a momentary sustained position and then gradually melt away to the static position. These movements often occur in the upper extremities. An EMG burst duration of greater than 200 ms can be seen for this "dystonic myoclonus" of subacute sclerosing panencephalitis (SSPE). In contrast to Creutzfeldt-Jakob disease, the jerks have a consistent relationship to periodic complexes on routine EEG. These complexes consist of high voltage (300-1500 J.L V), repetitive, polyphasic and sharp and slow wave complexes ranging from 500-2000 ms in duration, usually recurring every 4-15 s or sometimes longer (Westmoreland, 1987). The complex nature of the discharge makes it difficult to measure latency between the EEG discharge and the jerk EMG discharge. The complexes are typically widespread and synchronous, although asymmetry occurs and times lags have been measured from side to side and/or front to back. The

THE CLINICAL NEUROPHYSIOLOGY OF MYOCLONUS

discharges and jerks are thought to be fairly resistant to stimuli, but exceptions occur (Westmoreland, 1987). The SEP and long latency EMG responses have not been adequately studied in SSPE. A slow negative potential shift has been found to precede the jerk and EEG complex (Oga et al., 2000). 32.3.1.8. Myoclonus of corticobasal degeneration Myoclonus is an important feature of corticobasal degeneration and occurs in 50% of cases. Its clinical presentation parallels that of the overall syndrome with a focal distribution in the arm (sometimes leg) associated with other focal limb manifestations that can include apraxia, rigidity, dystonia, and alien limb phenomenon. The myoclonus is prominent with muscle action and often has reflex activation to cutaneous stimuli and deep tendon reflexes of the affected limb. The significant rigidity and dystonia can make it difficult to produce a resting muscle state in the affected limb, and myoclonus at absolute rest is uncommon. The EMG discharge duration for the myoclonus in corticobasal degeneration is 25-50 ms with a cocontraction pattern present when antagonists and/or other local muscles are involved. This pattern, along with a rostral to caudal activation pattern is characteristic for the physiology seen commonly in cortical myoclonus. However, the myoclonus physiology in corticobasal degeneration has distinctive features (Thompson, et al., 1994). The EEG shows no correlate to the myoclonus EMG discharges, even when back-averaging is performed. Parietal SEP cortical waves are either normal or poorly formed, and frontal P22-N30 components are usually intact. Enhanced long latency EMG responses in corticobasal degeneration show a different pattern when compared to those seen in cortical reflex myoclonus. Median nerve stimulation reveals an EMG reflex at about 40 ms, rather than the 50 ms seen in classical cortical reflex myoclonus. The physiology in corticobasal degeneration shows a response to digital nerve stimulation at 50 ms. This response is thought to be quite characteristic, but probably not specific, for corticobasal degeneration (Chen et aI., 1992; Thompson et aI., 1994). These responses, discharge duration, rostral to caudal recruitment, and along with numerous bits of circumstantial evidence, has created the consensus that this myoclonus is of cortical origin. By using magnetoencephalography, cortical activity was found preceding the myoclonus

537

(Mirna et al., 1998a). Thompson has hypothesized that the reflex myoclonus in corticobasal degeneration involves the pathological enhancement of responses in motor cortical areas to direct sensory input (Thompson et aI., 1994). 32.3.1.9. Asterixis Most myoclonus is "positive," meaning the jerk results from increase EMG activity. Negative myoclonus refers to a decrease in tonic EMG activity. The term, "asterixis," is considered to be equivalent to negative myoclonus. Negative myoclonus correlates with an average EMG silence duration of 50-200 ms. Three types of EMG patterns have been described (Ugawa et al., 1989; Toro et aI., 1995). Type I consists of abrupt onset and offset of EMG silence during voluntary muscle activation. A type I primarily negative event that is associated with a brief, discrete burst of EMG activity that precedes the silence characterizes type II. Type III silent periods are those that follow typical positive myoclonus, especially in trains. The EMG silences of negative myoclonus usually have a multifocal distribution. Renal failure, hepatic failure, and the drug valproic acid are notable causes of negative myoclonus in patients. Negative myoclonus can be seen in many of the same conditions that cause positive myoclonus. Thus, it is not unusual for negative and positive myoclonus to coexist in the same patient, such as in a case of posthypoxic myoclonus or progressive myoclonic epilepsy. The EEG correlate of type II negative myoclonus is similar to that of positive cortical myoclonus. Accordingly, Shibasaki has described "cortical reflex negative myoclonus" associated with a silent period after median nerve stimulation and enlarged cortical SEP (Shibasaki et al., 1994). Type I negative myoclonus does not have an EEG correlate and may have a subcortical generator. 32.3.1.10. Significance of cortical myoclonus and its relation to epilepsy The establishment of a cortical physiology for myoclonus, through one or more of the patterns above, implies the involvement of sensorimotor cortex in the pathophysiology of the myoclonus. However, one must carefully consider this principle in proper context. The details of what specific abnormalities exist in the neuronal circuitry of the sensorimotor cortex in myoclonus, or how they

538 might arise, are currently unknown. Myoclonus arising from the "sensorimotor cortex," then, may result purely from the sensorimotor cortex itself, purely from abnormal remote input from subcortical and/or other cortical areas, or some combination of these. The different types and levels of pathology that can occur in each of these instances may account for the varying characteristics seen among patients and diseases with cortical myoclonus. Much attention has been given to the relationship between cortical myoclonus and epilepsy, given the presence of a presumed hyperexcitability of sensorimotor cortex in both instances. The classic paper by Obeso et al. examined the overlap of clinical and neurophysiological findings in patients with cortical myoclonus (Obeso et al., 1985). These patients, when taken together, exhibited a "spectrum" of varying combinations of stimulus-sensitive myoclonus, spontaneous myoclonus, epilepsia partiailis continua, focal motor seizures, and secondarily generalized convulsions. The authors suggested that patients along this spectrum represented subtle differences in the site of abnormality in sensorimotor cortical neuronal mechanisms. Hallett has divided myoclonus into epileptic and non-epileptic (Hallett, 1985). He put forth that cortical reflex myoclonus is a fragment of partial epilepsy, reticular reflex myoclonus is a fragment of generalized epilepsy, and primary generalized epileptic myoclonus is a fragment of primary generalized epilepsy. Whether a patient with myoclonus is a point along a spectrum or represents a fragment of epilepsy, advancing our understanding of cortical myoclonus still depends on knowing more details about sensorimotor circuits and those brain areas that influence them.

32.3.2. Cortical-subcortical 32.3.2.1. General comments There is strong evidence that some generalized seizure phenomena arise from paroxysmal abnormal and excessive oscillation in bidirectional connections between cortical and subcortical sites. The term "cortical-subcortical" myoclonus refers to myoclonus arising from this type of physiology and other similar phenomena. For these entities, the abnormal influence of the subcortical input is critical. Despite the subcortical involvement, the cortical discharge precedes and drives the myoclonus event. This myoclonus usually occurs in paroxysms from rest

J.N. CAVINESS

and can be associated with other seizure phenomena that may even be more clinically significant than the myoclonus itself. The myoclonus is often generalized or bilaterally synchronous, but focal or multifocal distributions occur as well. The most useful drugs for cortical-subcortical myoclonus are usually valproic acid and clonazepam, although other drugs are known to have efficacy as well.

32.3.2.2. Absence seizures An absence seizure is characterized by a paroxysmal loss of consciousness of sudden onset and sudden termination that is associated with bursts of bilaterally synchronous spike and wave discharges recorded on the EEG and occurs in both children and adults (Lockman, 1985; Panayiotopoulos et al., 1992). Myoclonus is associated with the absence seizure in up to 45% of cases (Penry et al., 1975). Most of the time, the myoclonus is located in the eyelid, other facial or midline muscles, or less commonly in the limbs. The myoclonus can be correlated to generalized spike and wave discharges on the EEG. Less commonly, spike, polyspike, or polyspike and wave discharges can be seen (Appleton et al., 1993). In classic absence, the frequency of the spike and wave discharges is 2.5-4 Hz and this is only an ictal pattern. In atypical absence, as seen in Lennox-Gastaut syndrome, 1-2.5 Hz occurs and may be seen as an ictal or interictal pattern. With the myoclonic absence that occurs in older children and adolescents, 4-6 Hz is usually seen and may be ictal or interictal. Photosensitivity may occur. Information on somatosensory evoked potentials and long latency reflexes is limited. Valproic acid can provide effective treatment. The basic underlying mechanism of the absence seizure, and presumably of the associated myoclonus, is the generation of abnormal oscillatory rhythms from cortical and thalamic networks (Snead, 1995). 32.3.2.3. Primary generalized myoclonic seizures Myoclonic seizures occur as bilateral, synchronous myoclonic jerks affecting mainly shoulders and upper limbs. The jerks may be single or repetitive. Generalized spike, polyspike, or 4-6 Hz spike and wave discharges, alone or in combination, are associated with the jerks. These discharges have a frontal maximum and may be seen interictally. Photosensitivity is common. Somatosensory evoked potentials are enlarged in a portion of patients

THE CLINICAL NEUROPHYSIOLOGY OF MYOCLONUS

(Atakli et aI., 1999). Information on long latency EMG responses is limited. This myoclonus occurs most commonly as part of the juvenile myoclonic epilepsy syndrome. Valproic acid can provide effective treatment. The basic abnormality that causes the EEG discharges and myoclonus has been proposed as pathological changes in the cerebellothalamocortical loop (Savic et aI., 1994). One possible animal model for myoclonic seizures is the type A myoclonus of Papio papio, where the EMG discharges have been documented to follow the cortical discharges (Brailowsky, 1991). 32.3.2.4. Primary generalized epileptic myoclonus Wilkins et al. described eleven cases of "primary generalized epileptic myoclonus" that were proposed to have a physiology similar to that of primary generalized epilepsy (Wilkins et aI., 1985). In another manuscript, the relationship was viewed as a "fragment" of primary generalized epilepsy (Hallett, 1985). These cases demonstrated two patterns. One pattern was a bifrontal/bifrontocentral cerebral negativity of relativity long average duration (100-250 ms) and of variable pre-myoclonus onset (5-500 ms). The intra-individual variability was less than the inter-individual variability. The second pattern consisted of an invariant cerebral potential time-locked to the myoclonus. Most of these cases had a bifrontal negativity of shorter duration (30-100 ms), the duration was fixed for each case, and had stereotyped onset (40-60 ms) pre-myoclonus. The diagnoses of these eleven cases included Lennox-Gastaut syndrome, Alzheimer's disease, familial progressive myoclonic epilepsy, and a progressive degenerative syndrome of unknown cause. Clinical seizures occurred in some of these cases in addition to the myoclonus. The myoclonus in these patients mostly presented as small, multifocal twitches most evident in the fingers and hands. These twitches occurred at rest but some enhancement with muscle action could be seen. Reflex myoclonus was not described. The authors used the term, "minipolymyoclonus" to describe the small amplitude myoclonus movements. 32.3.3. Subcortical 32.3.3.1. General comments The clinical and neurophysiological characteristics of subcortical myoclonus are more variable

539

than for those in cortical or cortical-subcortical myoclonus. The myoclonus EMG duration may be longer than in cortical or cortical-subcortical myoclonus. Simultaneous rostral and caudal recruitment of the myoclonus in the EMG channels supports a subcortical generator. As one might expect, there is no evidence for abnormal cortical excitability (e.g. EEG spikes, enlarged cortical SEP waves) in subcortical myoclonus that can be tightly correlated to the myoclonus. Assignment of a case to the subcortical category can be problematic if it is based largely on absence of evidence for abnormal cortical excitability or circumstantial findings rather than direct evidence. As techniques advance, more direct evidence for subcortical sources hopefully will be obtained. Clonazepam, tetrabenazine, and anticholinergic drugs have some efficacy for the treatment of subcortical myoclonus.

32.3.3.2. Essential myoclonus Characterizing the electrophysiology of essential myoclonus is problematic secondary to the clinical heterogeneity that has been described under its definition. In addition, many cases, both hereditary and sporadic, have had dystonia described in the same patients as well. Disagreement and changing views on the nosology of these cases has made consensus difficult. The terms overlapping with essential myoclonus have included myoclonic dystonia, dystonic myoclonus, myoclonus-dystonia syndrome, and ballistic movement overflow myoclonus. In most cases, the myoclonus EMG discharge duration is 50-200 ms (Fig. 6). Longer discharge duration values have been described, especially in those cases with dystonia. The EMG patterns can be agonist-only or co-contracting agonist-antagonist. Often, the discharges are irregular with respect to amplitude, duration, and timing between agonist and antagonist. Stimulus sensitivity has been reported but is unusual (Chokroverty et al., 1987). Somatosensory evoked potentials and long latency EMG responses are normal. The routine EEG is normal. Some cases have had generalized, symmetrical, negative or positive EEG waves back-averaged 50-70 ms before myoclonus onset (Quinn et aI., 1988). No focal EEG correlates have been reported, and most EEG-EMG back-averaging has been unrevealing.

IN. CAVINESS

540

Myoclonus-Dystonia Syndrome

Right Wrist Flexors +400.0

I1

Mictovo/lS

Right Wrist

Extensors

o

300

600

900

1200

1500

1800

2100

2400

2700

3000

Milliseconds

Fig. 6. Multichannel surface EMG recording from the right wrist in a patient with myoclonus-dystonia syndrome during myoclonic jerking with postural activation of the right arm. Note the long duration of the shortest discharges (100-200 ms) as well as longer duration discharges. The discharges are irregular with respect to amplitude, duration, and timing between agonist and antagonist.

32.3.3.3. Reticular reflex myoclonus This myoclonus appears clinically as generalized jerks that occur at rest and are stimulus sensitive (Fig. 7A). The stimulus sensitivity may be to touch, muscle stretch, deep tendon reflex, or noise. The post-hypoxic state and uremia are known causes. The EMG discharges are less than 50 ms (Hallett et al., 1977). A major characteristic of the EMG discharges is a simultaneous bilateral rostral and caudal recruitment originating from the area of the medullary reticular formation (Fig. 7B). Axial and proximal limb muscles are primarily involved. Jitter can been observed from one myoclonic jerk to the next as far as relative timing between muscles. The EEG may show that epileptiform changes are associated, but they follow the first EMG activation, vary from jerk to jerk, and do not show a time-locked relationship to the myoclonus. The SEP is normal (Rothwell et al., 1986). 32.3.3.4. Opsoclonus-myoclonus syndrome The opsoclonus-myoclonus syndrome consists of opsoclonus, myoclonus, and variably other manifestations such as ataxia, tremor, and behavior problems. It usually has a subacute time course and

may be due to infectious, paraneoplastic, drug or toxin induced, idiopathic, or other etiologies (Caviness et al., 1995). The myoclonus is usually multifocal and predominantly induced by action. The EMG discharges are less than 100 ms, often occur in trains, and can show agonist-only or agonist-antagonist co-contraction (Fig. 8). The EEG is usually normal and if abnormalities do exist, they have no relationship to the myoclonus. Backaveraging of the myoclonus shows no EEG correlate (Gwinn and Caviness, 1997). The SEP cortical waves and long latency EMG responses are normal. Nonspecific abnormalities of the brainstem auditory evoked potential have been reported (Araki et al., 1989).

32.3.3.5. Propriospinal myoclonus This myoclonus occurs as trunk flexion or extension with axial muscle activation. Proximal limb muscles are often involved in the jerk bilaterally, but the predominant action is in the axial muscles (Brown et al., 1991a). These jerks can occur from rest and/or activated by stimuli such as touch, deep tendon reflex, or muscle stretch. Single or repetitive jerks may occur. The EMG discharge lasts typically

THE CLINICAL NEUROPHYSIOLOGY OF MYOCLONUS

541

Fig. 7. A shows multichannel surface EMG recording of multiple muscles that demonstrates the paroxysmal occurrence of generalized myoclonus EMG discharges in a patient with reticular reflex myoclonus. The generalized myoclonus was elicited by touching the shoulder. B shows an expanded time scale that shows a simultaneous ascending and descending order of recruitment from a lower brainstem source. Note that the sternomastoid muscle is recruited first. The EMG discharges are rectified and averaged from 20 trials. Time zero marks the time that the examiner touched the shoulder.

J.N. CAVINESS

542

Idiopathic Opsoclonus-Myoclonus Syndrome Case 3 (Rightside)

Case2 (Leftside)

Case1 (Right side)

+

Wria1

3000.0 IJV

Flex...

Wria1 Exten&oI1

1000

2000

0

1000

2000

0

seconds

1000

2000

5econds

Fig. 8. Multichannel surface EMG recordings from three patients with idiopathic opsoclonus-myoclonus syndrome during postural activation of the arms. Myoclonus EMG discharges with less than 100 ms duration are evident in a multifocal distribution. Modified from Gwinn and Caviness, 1997.

50-300 ms but sometimes longer. Both reciprocal and co-contracting agonist-antagonist relationships have been observed. A major characteristic of the EMG discharges is a simultaneous bilateral rostral and caudal recruitment originating from the area of the spinal cord origin (Brown et aI., 1991a). The activation speed of consecutive muscles is thought to be slower than for the corticospinal (pyramidal) pathway and is likely propriospinal. Jitter occurs from jerk to jerk with regard to the relative timing of the various EMG discharges within a given episode of myoclonus. No EEG abnormalities in the routine recording or with back-averaging have been reported. Somatosensory evoked potentials are normal. 32.3.3.6. Focal subcortical reflex myoclonus This entity, "focal subcortical reflex myoclonus," was proposed by Cantello et al. (1997). They studied myoclonus in three patients with myoclonus epi-

lepsy with ragged red fibers syndrome (MERRF) and two patients with progressive myoclonic ataxia. Reflex EMG discharges to median nerve stimulation produced latencies that were too short to be explained by a transcortical pathway. Cantello et al. posited that this phenomenon is generated from the afferent stimulus arriving at a subcortical site, then producing both ascending discharges to the cortex (producing an EEG potential) and to the spinal cord to produce the myoclonus. The subcortical site involved in this mechanism is unknown. 32.3.4. Segmental

Segmental myoclonus has its generator at a particular segment or contiguous segments of the brainstem and/or spinal cord. This segmental generator produces movements at a particular segment or contiguous segments. There is usually fairly persistent, rhythmical activation of muscles corresponding to the brainstemlspinal segment(s)

543

THE CLINICAL NEUROPHYSIOLOGY OF MYOCLONUS

involved. This myoclonus is known to be relatively unaffected by state of consciousness, motor activity, or stimulus. However, exceptions do occur. The most common type of segmental myoclonus is palatal myoclonus. A variety of etiologies may cause segmental myoclonus but vascular, tumor, trauma, infectious, and idiopathic and/or "essential" diagnoses account for most cases. The type of pathophysiology involved, and thus the reason why a certain brainstemlspinallesion mayor may not cause segmental myoclonus, is unknown. Many treatments have had reported effectiveness, but unimpressive and inconsistent results are usually seen. Anticholinergic medication, clonazepam, tetrabenazine, and botulinum toxin injections are the most commonly used. The EMG usually shows synchronous activation of the affected muscles. The typical frequency is in the range of 0.5-3 Hz and the typical EMG discharge duration varies widely between 50 and 500 ms. The EEG and somatosensory evoked potential are normal. Brainstem auditory evoked potentials (BAEP) have had abnormal findings in some individuals with palatal myoclonus (Westmoreland, 1983). These inconsistent BAEP abnormalities probably represent the same lesion type, but not the same exact location, as that responsible for the palatal myoclonus pathophysiology. In spinal segmental myoclonus, mixed nerve stimulation can evoke EMG discharges in the affected muscles at latencies longer than 40 ms, but such findings are variable and the latency values vary from case to case. These reflex discharges may reflect hyperexcitability of polysynaptic pathways that contribute to the generation of the myoclonus (Hopkins and Michael, 1974; Kono et aI., 1994). Palatal myoclonus is the most common type of segmental myoclonus. The movement is rhythmic and usually bilateral with a rate between 1-4 Hz, with other rates being less common. Because of its smooth oscillatory nature, many clinicians choose the term "palatal tremor" over palatal myoclonus. A distinction is made between "essential palatal myoclonus" (EPM) and "symptomatic palatal myoclonus" (SPM) (Deuschl et aI., 1990). EPM is associated with activation of the tensor veli palatini muscle, no identifiable MRI lesion, and is unlikely to involve other muscles. SPM is associated with activation of the levator veli palatini muscle, an identifiable MRI brainstem lesion, involvement of

other muscles, cerebellar dysfunction, and an older age of onset. EPM patients are more likely to state that their ear click is the chief complaint, whereas SPM patients are more concerned with the other associated neurological problems rather than the palatal movements per se. Important differences in neurophysiological testing have also been found (Deuschl et aI., 1994). EPM shows a complete suppression with sleep, but sleep only produces mild variations in rate with SPM. The palatal movement cycle only exerts remote effects on tonic EMG activity of extremity muscles in SPM. As shown by studies of blink reflex activity, jaw jerk, and masseteric silent period, EPM had only polysynaptic brainstem reflex abnormalities, whereas SPM patients can have abnormalities of monosynaptic, oligosynaptic, and polysynaptic brainstem reflexes.

32.3.5. Peripheral Peripheral myoclonus refers to myoclonic jerks that are driven from a peripheral site. The bestdocumented example is hemifacial spasm. Such EMG discharges are characterized by marked duration variability from discharge to discharge (Fig. 9). The appearance and timing of EMG discharges, which are supplied by the same nerve, are similar. In peripheral myoclonus, the spectrum of EMG discharge duration may merge continuously with those EMG discharges that are responsible for movements that are longer lasting. It should be recognized that the literature contains other uses of the term "peripheral myoclonus." For example, some studies report myoclonus associated with peripheral nervous system lesions, but posit that the myoclonus itself is centrally generated and results from "central reorganization." Botulinum injections are very useful for hemifacial spasm. Little information is available for other treatments of peripheral myoclonus.

32.4. Psychogenic myoclonus Jerks that arise from a psychogenic mechanism may appear quick enough to overlap with myoclonus. Some authors would maintain that myoclonus only pertains to involuntary movements. Nevertheless, the term "psychogenic myoclonus" has gained some acceptance. Criteria for diagnosing psychogenic movement disorders are available (Williams et aI., 1995). Monday and Jankovic have

IN. CAVINESS

544

jerk strongly favors involuntary myoclonus for that particular type of movement. The presence of an enlarged cortical SEP, abnormal long latency reflexes, back-averaging of a short duration « 100 ms) EEG wave preceding the jerk suggest involuntary myoclonus. A psychogenic myoclonus EMG discharge may show the known voluntary reciprocal biphasic/triphasic agonist-antagonist/agonist pattern with the duration of each burst being 50-100 ms (Hallett, 1986). Longer EMG discharges are possible in psychogenic myoclonus and the discharge pattern may have a non-stereotyped appearance. Stimulusevoked jerks or jumps with a mean latency in excess of 100 ms suggest voluntary or psychogenic jerks (Brown and Thompson, 2001). It is important to remember that involuntary and psychogenic myoclonus as well as other movement disorders may coexist in the same patient. The Bereitschaftspotential (BP) is a back-averaged negative EEG cortical potential that occurs

outlined characteristics of psychogenic myoclonus: (1) inconsistent character of the movements (amplitude, frequency, and distribution) and other features incongruous with typical "organic" myoclonus; (2) associated psychogenic symptomatology; (3) marked reduction of the myoclonus with distraction; (4) exacerbation and relief with placebo and suggestion; (5) spontaneous periods of remission; (6) acute onset and sudden resolution; and (7) evidence of underlying psychopathology (Monday and Jankovic, 1993). Clinical neurophysiology may offer the clinician some assistance in evaluating these patients, but one must exercise caution when interpreting the results. It is considered unwise to use such techniques alone to "prove" or "rule out" a psychogenic basis for myoclonus. Results must always be used in their clinical context and the possible pitfalls should be realized. High amplitude EMG discharges < 50 ms duration that correlate with a moderate amplitude

Peripheral Myoclonus • Hemifacial Spasm +200

I

f-

Right Orbicularis Oria

M

-'"

I

f-

!

",

f-

fiT

l'

~

.. ,1 "

..... 'r

Microvolts ,

1

I

-

-

Mentalis

1-

I~~

I~

1.

'I

...

.I..

I

'

~

~

...

...

l

J.

'\1

9

399

999

1299

1S99

1899

2199

2499

2799

3999

Milliseconds

Fig, 9. Multichannel surface EMGrecording from the facein a patientwith hemifacial spasmduring a seriesof right facial myoclonic jerks. EMGdischarge duration variability of thefacialjerkingis evident. This samepatientalsohad muchlonger discharges associated with sustained spasms.

545

THE CLINICAL NEUROPHYSIOLOGY OF MYOCLONUS

before self-paced, voluntary phasic movements. The first phase of the BP, from approximately 2000 to 1000 ms to approximately 650 to 450 ms before voluntary movement, is widespread in distribution, maximum in the vertex, and has a slowly increasing negative slope (Evidente et al., 1999). After 650 to 450 ms before the voluntary movement, the potential becomes more lateralized contralateral to the movement. Terada et al. found that the presence of a BP preceding psychogenic myoclonus to be supportive evidence for a psychogenic etiology (Terada, 1995). However, because recording movements from a psychogenic individual may be technically challenging or inadequate, absence of the BP should not be used to indicate that voluntary mechanisms have been ruled out. At times, recording an insufficient number of artifact-free jerks «50), problems with determining a reliable trigger point, averaging movements elicited from external cues, and using movements that are two seconds or less apart will prevent the discovery of a BP. 32.5. Advanced and research methods Various other techniques have been used to study cortical mechanisms of myoclonus generation. However, they are difficult to apply routinely due to either the expense or elaborate nature of the study. Magnetoencephalography (MEG) has the ability to perform better localization and amplitude sensitivity when compared to EEG. Uesaka et al. found that among six subjects with cortical myoclonus and one with epilepsia partialis continua: (1) the dipole was located at the precentral gyrus in the case with epilepsia partialis continua; (2) dipoles were present both on the precentral and postcentral gyrus in a cortical myoclonus case; and (3) five cases localized to the postcentral gyrus only (Uesaka et al., 1996). In this study, the initial cortical somatosensory evoked magnetic fields localized to the postcentral gyrus. Uesaka et al. suggested that patients with enlarged SEPs are likely to have myoclonus arise from the postcentral gyrus. In contrast, Mirna et al. found the MEG cortical correlate for all six of their myoclonus subjects to localize to the precentral gyrus (Mirna et al., 1998a). In another report, Mirna et al. found that enlarged cortical somatosensory evoked magnetic fields localized to the precentral gyrus in four subjects and to the postcentral gyrus in one subject (Mirna et al., 1998b). More studies will be needed to

sort out whether these separate studies actually conflict or if case to case variability can truly be seen in cortical myoclonus. Electrodes placed on the cortical surface are used extensively in epilepsy surgery, but there are only rare opportunities to do this in myoclonus. Ashby et al. analyzed cortical recordings in a case of severe myoclonus associated with celiac disease (Ashby et al., 1999). The myoclonus origin as well as the enlarged SEP wave was localized to the motor cortex. Brown et al. have found changes in EEG-EMG and EMG-EMG coherence patterns for subjects with myoclonus (Brown et al., 1999). They have suggested that myoclonus patients show pathological exaggerations of physiological central rhythmicity relating to movement and that the precise pattern of coherence has possible diagnostic value. Both somatosensory and motor evoked potentials can be applied in order to add information beyond what their routine use provides. Somatosensory evoked potentials using double stimulation over a range of interstimulus intervals can detect periods of excitability enhancement and depression of cortical activity (Ugawa et al., 1991). Long latency EMG responses can be investigated in a similiar manner. Other modalities besides median nerve stimulation may be explored in this way. In a technique known as jerk-locked evoked potential, the stimulus is presented just at the time of, or at varying intervals after, the onset of the myoclonic EMG discharge (Shibasaki et al., 1985). This method allows the study of cortical excitability after the myoclonus itself. Motor evoked potentials elicited with magnetic stimulation, either using single or double stimulation protocols, is another technique used to probe cortical activity associated with myoclonus cases (Brown et al., 1996). Caution is necessary when using stimulation techniques in general, since such methods have the potential to evoke seizures. Shibasaki does not recommend the routine clinical use of transcranial magnetic stimulation in patients at risk for generalized convulsions (Shibasaki, 2000). References Allison, T, McCarthy, G, Wood, CC and Jones, SJ (1991) Potentials evoked in human and monkey cerebral cortex by stimulation of the median nerve. Brain, 114:

2465-2503.

546 Appleton, RE, Panayiotopoulos, CP, Acomb, BA and Beirne, M (1993) Eyelid myoclonia with typical absences: an epilepsy syndrome. J. Neurol. Neurosurg. Psychiatry, 56: 1312-1316. Araki, K, Veda, Y, Masaki, H and Tatsuro, T (1989) Opsoclonus-Myoclonus syndrome with abnorrnal magnetic resonance imaging and brainstem auditory evoked potentials. Japan. J. Med., 28(6): 753-756. Artieda, J and Obeso, JA (1993) The pathophysiology and pharmacology of photic cortical reflex myoclonus. Ann. Neurol., 34: 175-184. Ashby, P, Chen, R, Wennberg, R, Lozano, AM and Lang, AE (1999) Cortical reflex myoclonus studied with cortical electrodes. Clin. Neurophysiol., 110: 15211530. Atakli, D, Soysal, A, Atay, T, Altintas, H, Arpaci, Band Baybas, S (1999) Somatosensory evoked potentials and EEG findings in siblings of juvenile myoclonic epilepsy patients. Epilept. Disord., 1(3): 173-177. Berkovic, SF, Andermann, F, Carpenter, S and Wolfe, LS (1986) Progressive myoclonus epilepsies: specific causes and diagnosis. N. Eng/. J. Med., 315(5): 296-305. Brailowsky, S (1991) Myoclonus in Papio papio. Mov. Disord., 6(2): 98-104. Brenner, RP and Schaul, N (1990) Periodic EEG patterns: Classification, clinical correlation, and pathophysiology. J. Clin. Neurophysiol., 7(2): 249-267. Brown, P (1995) Myoclonus: A practical guide to drug therapy. CNS Drugs, 3: 22-29. Brown, P and Thompson, PD (2001) Electrophysiological aids to the diagnosis of psychogenic jerks, spasms, and tremor. Mov. Disord., 16: 595-599. Brown, P, Day, BL, Rothwell, IC, Thompson, PD and Marsden, CD (199Ia) Intrahemispheric and interhemispheric spread of cerebral cortical myoclonic activity and its relevance to epilepsy. Brain, 114: 2333-2351. Brown, P, Thompson, PD, Rothwell, IC, Day, BL and Marsden, CD (1991b) Axial myoclonus of propriospinal origin. Brain, 114: 197-214. Brown, P, Ridding, MC, Werhahn, KJ, Rothwell, IC and Marsden, CD (1996) Abnormalities of the balance between inhibition and excitation in the motor cortex of patients with cortical myoclonus. Brain, 119: 309-317. Brown, P, Farmer, SF, Halliday, DM, Marsden, J and Rosenberg, IR (1999) Coherent cortical and muscle discharge in cortical myoclonus. Brain, 122: 461-472. Cantello, R, Gianelli, M, Civardi, C and Mutani, R (1997) Focal subcortical reflex Myoclonus. Arch. Neurol., 54: 187-196. Caviness, IN (1996) Myoclonus. Mayo Clio Proc., 71: 679-688.

J.N. CAVINESS

Caviness, IN and Kurth, M (1997) Cortical Myoclonus in Huntington's Disease associated with an enlarged somatosensory evoked potential. Mov. Disord., 12(6): 1046-1051. Caviness, IN, Forsyth, PA, Layton, DD and McPhee, TJ (1995) The movement disorder of adult opsoclonus. Mov. Disord., 10: 22-27. Caviness, IN, Adler, CH, Newman, S, Caselli and Muenter MD (1998) Cortical myoclonus in Levodopa-responsive parkinsonism. Mov. Disord., 13: 540-544. Caviness, IN, Alving, LI, Maraganore, DM, Black, RA, McDonnell, SK and Rocca, WA (1999) The incidence and prevalence of Myoclonus in Olmsted County, Minnesota. Mayo Clin. Proc., 74: 565-569. Caviness, IN, Gwinn-Hardy, KA, Adler, CH and Muenter, MD (2000) Electrophysiological observations in hereditary parkinsonism-dementia with Lewy body pathology. Mov. Disord., 15: 140-145. Chen, R, Ashby, P and Lang, AE (1992) Stimulussensitive myoclonus in akinetic-rigid syndromes. Brain, 115: 1875-1888. Chokroverty, S, Manocha, MK and Duvoisin, RC (1987) A physiologic and pharmacologic study in anticholinergic-responsive essential myoclonus. Neurology, 37: 608-615. Dawson, GD (1947) Investigations on a patient subject to myoclonic seizures after sensory stimulation. J. Neurol. Neurosurg. Psychiatry, 10: 141-162. Deuschl, G, Mischke, G, Schenck, E, Schulte-Monting, J and Lucking, CH (1990) Symptomatic and essential rhythmic palatal myoclonus. Brain, 113: 1645-1672. Deuschl, G, Ebner, A, Hammers, R and Lucking, CH (1991) Differences of cortical activation in spontaneous and reflex myoclonias. Electroencephalogr. Clin. Neurophysiol., 80: 326-328. Deuschl, G, Toro, C, Balls-Sole, J, Zeffiro, T, Zee, DS and Hallett (1994) Symptomatic and essential palatal tremor I. Clinical, physiological and MRI analysis. Brain, 117: 775-788. Deuschl, G, Bain, P and Brin, M (1998) Consensus statement of the movement disorder society on tremor. Mov. Disord., 13 (Suppl 3): 2-23. Evidente, VGH and Caviness, IN (1999) Focal cortical transient preceding myoclonus during lithium and tricyclic antidepressant therapy. Neurology, 52: 211213. Evidente, VGH, Caviness, IN, Jamieson, B, Weaver, A and Joshi, N (1999) Intersubject variability and intrasubject reproducibility of the bereitschaftspotential. Mov. Disord., 14: 313-319. Gwinn, KA and Caviness, IN (1997) Electrophysiological observations in idiopathic Opsoclonus-Myoclonus syndrome. Mov. Disord., 12(3): 438-442.

THE CLINICAL NEUROPHYSIOLOGY OF MYOCLONUS

Hallett, M (1985) Myoclonus. Relation to epilepsy. Epilepsia, 26 (Suppl. 1): S67-S77. Hallett, M (1986) Electrophysiologic evaluation of tremor and central disorders movement. In: MJ Aminoff (Ed.), Electrodiagnosis in Clinical Neurology (2nd ed.). Churchill Livingstone, New York, Chap. 12, pp. 385401. Hallett, M, Chadwick, D, Adam, J and Marsden, CD (l977a) Reticular reflex myoclonus: A physiological type of human post-hypoxic myoclonus. J. Neurol. Neurosurg. Psychiatry, 40: 253-264. Hallett, M, Chadwick, D and Marsden, CD (1977b) Cortical reflex myoclonus. Neurology, 29: 1107-1125. Hanazono, T, Kohara, N, Kaji, R and Kimura, J (1994) Cortical reflex negative myoclonus. Brain, 117: 477486. Hopkins, AP and Michael, WF (1974) Spinal myoclonus. J. Neurol. Neurosurg. Psychiatry, 37: 1112-1115. Ikeda, A, Kakigi, R, Funai, N, Neshige, R, Kuroda, Y and Shibasaki, H (1990) Cortical tremor: A variant of cortical reflex myoclonus. Neurology, 40: 1561-1565. Kelly, n, Sharbrough, FW and Westmoreland, BF (1978) Movement-activated central fast rhythms: an EEG finding in action myoclonus. Neurology, 28: 10371040. Kelly, n, Sharbrough, FW and Daube, JR (1981) A clinical and electrophysiological evaluation of myoclonus. Neurology, 31: 581-589. Kono, I, Ueda, Y, Araki, K, Nakajima, K and Shibasaki, H (1994) Spinal myoclonus resembling belly dance. Mov. Disord., 9(3): 325-329. Lockman, LA (1985) Absence seizures and variants. Neurol. cu«. 3(1): 19-29. Marsden, CD, Hallett, M and Fahn, S (1982) The nosology and pathophysiology of myoclonus. In: CD Marsden and S Fahn (Eds.), Movement Disorders. Butterworth, London, Chap. 13, pp. 196-248. Mirna, T, Nagamine, T, Ikeda, A, Yazawa, S, Kimura, J and Shibasaki, H (1998a) Pathogenesis of cortical myoclonus studied by magnetoencephalography. Ann. Neurol., 43: 598-607. Mirna, T, Nagamine, T, Mikuni, N, Ikeda, A, Fukuyama, H, Takigawa, T, Kimura, J and Shibasaki, H (1998b) Cortical myoclonus. Sensorimotor hyperexcitability. Neurology, 50: 933-942. Obeso, JA, Rothwell, JC and Marsden, CD (1985) The spectrum of cortical myoclonus. Brain, 108: 193-224. Obeso, JA, Rothwell, JC and Marsden, CD (1986) Somatosensory evoked potentials in myoclonus. In: S Fahn, CD Marsden and M Van Woert (Eds.), Myoclonus. Raven Press, New York, Vol. 43, pp. 373-384. Oga, T, Ikeda, A, Nagamine, T, Sumi, E, Matsumoto, R, Akiguchi, I, Kimura, J and Shibasaki, H (2000) Mov. Disord., 15(6): 1173-1183.

547 Panayiotopoulos, CP, Chroni, E, Daskalopoulos, C, Baker, A, Rowlinson, S and Walsh, P (1992) Typical absence seizures in adults: clinical, EEG, video-EEG findings and diagnostic/syndromic considerations. J. Neurol. Neurosurg. Psychiatry, 55: 1002-1008. Penry, JK, Porter, RK and Dreifuss, FE (1975) Simultaneous recordings of absence seizures with videotape and EEG. Brain, 98: 427-440. Quinn, NP, Rothwell, JC, Thompson, PD and Marsden, CD (1988) Hereditary myoclonic dystonia, hereditary torsion dystonia and hereditary essential myoclonus: An area of confusion. In: S Fahn et al. (Eds.), Advances in Neurology, Dystonia 2. Raven Press, New York,Vol. 50, pp. 391-401. Rapin, I (1986) Myoclonus in neuronal storage and Lafora diseases. In: S Fahn, CD Marsden and M Van Woert (Eds.), Myoclonus. Raven Press, New York, Vol. 43, pp.65-85. Rodriguez, ME, Artieda, J, Zubieta, JL and Obeso, JA (1994) Reflex myoclonus in olivopontocerebellar atrophy.1. Neurol. Neurosurg. Psychiatry, 57: 316-319. Rothwell, JC, Obeso, JA and Marsden, CD (1986) Electrophysiology of somatosensory reflex myoclonus. In: S Fahn, CD Marsden and M Van Woert (Eds.), Myoclonus. Raven Press, New York, Vol. 43, pp. 385-398. Savic, I, Pauli, S, Jan-Olof, T and Blomqvist, G (1994) In vivo demonstration of altered benzodiazepine receptor density in patients with generalized epilepsy. J. Neurol. Neurosurg. Psychiatry, 57: 797-804. Sharbrough, F, Chartrian, G, Lesser, RP, Luders, H, Neuwer, M and Picton, TW (1991) American Electroencephalographic Society guidelines for standard electrode position nomenclature. J. Clin. Neurophysiol., 8(2): 200-202. Shibasaki, H. Electrophysiologic studies of myoclonus (2000) AAEE Minimonograph #30. Muscle Nerve, 23: 321-335. Shibasaki, Hand Kuroiwa, Y (1975) Electroencephalographic correlates of myoclonus. Electroencephalalogr. Clin. Neurophysiol., 39: 455-463. Shibasaki, Hand Neshige, R (1987) Photic cortical reflex myoclonus. Ann. Neurol., 22: 252-257. Shibasaki, H, Yamashita, and Tsuji, S (1977) Somatosensory evoked potentials. Diagnostic criteria and abnormalities in cerebral lesions. J. Neurol. Sci., 34: 427-439. Shibasaki, H, Motomura, S, Yamashita, Y, Shii, Hand Kuroiwa, Y (1981) Periodic synchronous discharge and myoclonus in Creutzfeldt-Jakob Disease: Diagnostic application of jerked-locked averaging method. Ann. Neurol.,9: 150-156

548 Shibasaki, H, Neshige, Rand Hashiba, Y (1985) Cortical excitability after myoclonus: jerk-locked somatosensory evoked potentials. Neurol., 35: 36-41. Shibasaki, H, Yamashita, Y, Tobimatsu and Neshige R (1986) Electroencephalographic correlates of myoclonus. In: S Fahn, CD Marsden and M Van Woert (Eds.), Myoclonus. Raven Press, New York, Vol. 43, pp. 357-

372. Shibasaki, H, Ikeda, A, Nagamine, T, Mirna, T, Terada, K, Nishitani, N, Kanda, M, Takano, S, Hanazono, T, Kohara, N, Kaji, R and Kimura, J (1994) Cortical reflex negative myoclonus. Brain, 117: 477-486. Snead, DC (1995) Basic mechanisms of generalized absence seizures. Ann. Neurol., 37: 146-157. Terada, K, Ideda, A, Van Ness, PC, Nagamine, T, Kaji, R, Kimura, J and Shibasaki, H (1995) Presence of Bereitschaftspotential preceding psychogenic myoclonus: clinical application of jerk-locked back averaging. J. Neurol. Neurosurg. Psychiatry, 58: 745-747. Thompson, PD, Day, BL, Rothwell, JC, Brown, P, Britton and Marsden, CD (1994) The myoclonus in corticobasal degeneration. Evidence for two forms of cortical reflex myoclonus. Brain, 117: 1197-1207. Toro, C, Hallett, M, Rothwell, JC, Asselman, PT and Marsden, CD (1995) Physiology of negative myoclonus. In: S Fahn, M Hallett, HD Luders and CD Marsden (Eds.), Negative Motor Phenomena, Advances in Neurology (Vol. 67). Lippincott-Raven Publishers, Philadelphia, pp. 211-217. Uesaka, Y, Terao, Y, Ugawa, Y, Yumoto, M, Hanajima, R and Kanazawa, I (1996) Magnetoencephalographic

J.N. CAVINESS

analysis of cortical myoclonic jerks. Electroencephalogr. Clin. Neurophysiol., 99: 141-148. Ugawa, Y, Kohara, N, Hirasawa, H, Kuzuhara, S, Iwata, M and Mannen, T (1987) Myoclonus in Alzheimer's disease. J. Neurol., 235: 90-94. Ugawa, Y, Shimpo, T and Mannen, T (1989) Physiological analysis of asterixis: silent period locked averaging. J. Neurol. Neurosurg. Psychiatry, 52: 89-93. Ugawa, Y, Genba, K, Shimpo, T and Mannen, T (1991) Somatosensory evoked potential recovery (SEP-R) in myoclonic patients. Electroencephalogr. Clin. Neurophysiol., 80: 21-25. Westmoreland, BF (1987) The EEG in cerebral inflammatory processes. In: E Niedermeyer and F Lopes da Silva (Eds.), Electroencephalography (2nd ed.). Urban & Schwarzenberg, Baltimore, Chap. 17, pp. 264-266. Westmoreland, BF, Sharbrough, FW, Stockard, JJ and Dale, AI (1983) Brainstem auditory evoked potentials in 20 patients with palatal myoclonus. Arch. Neurol., 40: 155-158. Wilkins, DE, Hallett, M, Berardelli, A, Walshe, T and Alvarez, N (1984) Physiologic analysis of the myoclonus of Alzheimer's disease. Neurology, 34: 898-903. Wilkins, DE, Hallett, M and Erba, G (1985) Primary generalized epileptic myoclonus: a frequent manifestation of minipolymyoclonus of central origin. J. Neurol. Neurosurg. Psychiatry, 48: 506-516. Witte, DE, Niedermeyer, E, Arendt, G and Freund, HJ (1988) Post-hypoxic action (intention) myoclonus: a clinico-electroencephalographic study. J. Neurol., 235: 214-218.

Movement Disorders Handbook of Clinical Neurophysiology, Vol. I

M. Hallett (Ed.) © 2003 Elsevier B.V. All rights reserved

549 CHAPTER 33

Tics Mark Hallett* Human Motor Control Section, National Institute of Neurological Disorders and Stroke, NIH, Bethesda, MD 20892, USA

33.1. Clinical features Tics are a feature of a number of disorders. The most common tic disorder is "transient tic of childhood." This is a single type of tic that affects a child for several weeks for as long as a year and then disappears. There is also a "chronic tic" disorder that begins in childhood or adult life. Characteristic of this disorder is a stereotyped tic with a rather constant frequency over the years. A distinctive tic disorder that has been studied extensively is the Gilles de la Tourette syndrome (Tourette syndrome, TS). It is a chronic multiple tic disorder with childhood onset. Features include age of onset between 5 and 10 years, male to female ratio of 3-4 to 1, often positive family history, waxing and waning tic frequency over months and years, and a high incidence of soft neurological signs. Vocal tics such as grunting noises are necessary for the diagnosis, and a characteristic feature is coprolalia although it is found in only a quarter of patients. While tics are ordinarily referred to as involuntary movements, it is clear that they are not completely involuntary. Most patients describe a psychic tension that builds up inside of them and which can be relieved by the tic movement. Patients "let the tic happen" or even "make the tic" in order to relieve the tension. It is fair to say that some patients are not even able to say whether they are voluntary or involuntary. The tics can be voluntarily suppressed for some period of time at the expense of letting the psychic tension rise. While even chorea is at times somewhat suppressible, tics are the class of involun-

* Correspondence to: Dr. Mark Hallett, M.D., Human Motor Control Section, NIH, Building 10, Room 5N226, 10 Center Drive, MSC 1428, Bethesda, MD 20892-1428, USA. E-mail address: [email protected] Tel.: 301/496-9526; fax: 301/480-2286.

tary movement that is most suppressible. A rebound exacerbation of tics usually occurs after a period of voluntary suppression. Perhaps the distinction between voluntary and involuntary is not appropriate. On another scale used to describe movement, that of more to less automatic, they are clearly more automatic. Tics can be simple or complex movements. Many tics are stereotyped, and patients can have only a single type of tic or multiple tics. Patients can have the multiple tics serially or at the same time. A patient affected by multiple tics simultaneously might look like a patient with multifocal myoclonus or chorea. Many movements look like quick voluntary movements, but some are more prolonged and have been called dystonic tics. All tic movements look like movements that can be made voluntarily and typically can be mimicked easily by the patients. In many patients, tics are preceded by sensory symptoms often called sensory tics. These sensory symptoms drive the motor act which is typically directed to the region of the sensation. The motor act stops the sensory symptom, which then may quickly recur. Scratching an itch might be analogous. The feeling of inner tension described by the patients can be sometimes fully explained by these sensory symptoms. Some patients say that their movements are entirely voluntary and directed to deal with the sensory symptom. Further, if the sensory symptoms were abolished, there would be no movements. Sensory symptoms may be a more concrete version of "psychic tension" arising from a different part of the brain. There is a very close relationship between Tourette syndrome (TS) and obsessive-compulsive disorder (OCD) (Singer and Walkup, 1991; Leonard et al., 1992), many patients showing both disorders. From a psychological or physiological point of view, there may be a similarity between tic and an

M. HALLETT

550

obsession or compulsive behavior. Both the urge to tic and an obsessive thought arise involuntarily, then increase an inner feeling of tension until either the movement is made or the compulsive behavior is carried out. Physiological findings in these patients may relate to the TS, the OCD or both. There is also a frequent relationship between tic and attention-deficit hyperactivity disorder (ADHD) (Knell and Comings, 1993). Patients with both TS and ADHD may have a different disorder from those with TS alone. Physiological studies have not always differentiated these two groups appropriately. 33.2. Physiology of movement A good deal of information has been amassed on the physiology of tics that will be reviewed in the next sections. When taken together an interesting picture emerges, but no one test or group of tests is sufficiently specific to be used as a clinical test for either tics or TS. Tic movements usually look like quick voluntary movements electromyographically. EMG bursts vary from 50 to 200 ms in duration and may have the triphasic pattern of voluntary ballistic movements (Fig. 1). The EMG has more complex patterns and longer bursts of activity associated with dystonic or complex tics. The EMG pattern of a dystonic tic is

L S.C.I1. BICEPS TRICEPS

I'

~,...

....- - -

------Aft-

F.FLEX. F.EXT. A.P.B.

----------=.Jo.. "' '00 "$

Fig. 1. Surface EMG recording of an arm tic in a patient withTS. The pattern of activity is identical to the triphasic pattern of normal voluntary rapid movements, with successive bursts in agonist, antagonist and agonist. From Hallett andMarsden (1981) withpermission.

not clearly separable from a dystonic movement. These findings differ from certain types of myoclonus, for example, where the EMG bursts are shorter than can be produced voluntarily. They also differ from chorea which is characterized by EMG bursts that appear with random length in random muscles, and is associated with paroxysmal lapses in EMG activity when there is an attempted tonic contraction. In making voluntary drawing movements, patients with TS show less asymmetry between dominant and nondominant hand compared with control subjects (Georgiou et al., 1997). With short strokes, controls showed the predicted right hand superiority in movement time more strongly than patients with TS. The right hand of controls was less force efficient with long strokes and more force efficient with short strokes, whereas either hand of patients with TS was equally force efficient, irrespective of stroke length, with an overall performance profile similar to but better than that of the controls' left hand. Patients with TS actually performed in certain respects better than controls. Grip and load forces have been investigated in patients when grasping objects and moving them up and down. In one investigation, there was a disorder of the grip to load force ratio (Serrien et al., 2002), while in another study it was normal (Flanagan et al., 1999). In the latter study, of just one patient, the subject often had tic movements in similar directions to the voluntary movements. Normal coordination of the grip to load force ratio was seen with both tic and voluntary movements. This indicates a normal incorporation of set and sensory information into tic movements. Physiology of reaction time has been investigated in patients with TS. The performance of boys with TS and control subjects was compared on a continuous performance test (CPT) (Shucard et aI., 1997). TS children demonstrated a normal capacity for discriminating targets from nontargets during the task, but showed significantly slower reaction times than controls. In a different experimental paradigm, TS patients and controls had to make a sequence of movements along a path with variable amounts of advance information (Georgiou et al., 1995a). Patients with TS were found to be more reliant than controls on external visual cues to execute motor programs. If no advance information was provided before each move, movement execution was slower

551

TICS

than that of controls. When there was a marked reduction in advance information - that is, a visual pathway to be followed was extinguished well in advance of each successive movement - executions progressively slowed as the sequence was traversed. On the other hand, the movement initiation and execution times of patients with TS were similar to those of controls when advance visual information was available. The authors interpreted their results as indicating that patients with TS may have difficulty with internal switching mechanisms or may require more time to plan and program each movement, and under such conditions may require external visual cues to direct attention effectively to given targets. TS patients also have difficulty with spatially incongruent stimulus-response configurations, called the Simon effect (Georgiou et al., 1995b; Cope et al., 1996). In this situation, subjects have to shift their attention away from naturally expected stimulusresponse relationships. For example, an arrow pointing to the right would be the stimulus for a movement to the left. This result indicates that TS patients have difficulties in making attentional shifts, or in inhibiting inappropriate responses. There is a similar deficit with saccadic eye movements. In an antisaccade task the subject must look in the opposite direction of the location of the stimulus. In such a task, patients with TS have more errors and an abnormally long latency (Straube et al., 1997; Farber et al., 1999; Dursun et al., 2000; LeVasseur et al., 2001). While some aspects of saccades are normal in patients with TS (Farber et al., 1999), there are other abnormalities including delayed latencies even for prosaccades, looking in the direction of the stimulus (LeVasseur et al., 2001; Mostofsky et al., 2001). Another instructive abnormality is in a delayed saccade task where subjects wait fqr an instructional stimulus before generating a planned movement. Patients often move earlier than they should suggesting a loss of inhibition in withholding execution (LeVasseur et al., 2001), although this may be due to co-morbid ADHD (Mostofsky et al., 2001). The conclusion from these studies is that TS patients have some difficulty generating movements from internal storage and become abnormally dependent on external stimuli. When the external stimuli are incongruent with the motor response, they have particular difficulty.

33.3. EEG and movement related cortical potentials

Minimal, non-specific EEG abnormalities have been noted in some studies (Krumholz et al., 1983; Lees et al., 1984; Neufeld et al., 1990; Drake et al., 1991; Gunther et al., 1996). This does not appear to be significant in general. In an EEG study by Hyde et al. (1994) of monozygotic twins with TS, more EEG abnormalities and lower neuropsychological testing scores were typically found in the twin with lower birth weight, and this twin also typically had worse disease. These findings are consistent with the concept that the severity of TS may arise from an interaction of genetic background and the environment, and that EEG might be a marker of the environmental component. Some abnormalities are also noted with auditory evoked potentials (Drake et al., 1992b), contingent negative variation (CNV) (Drake et al., 1992b), the P3 or P300 potential (Van Woerkom et al., 1994; Oades et al., 1996; Johannes et al., 200lb), and the NoGo-anteriorization (Johannes et al., 200la), but it is difficult to learn a consistent message from these results. The EEG activity associated with tics should provide important information. Obeso et al. (1981) reported that there was no normal premotor activity prior to simple tics, but the premotor activity was normal for similar movements produced voluntarily by the same patients. In five of six patients there was no activity at all prior to the tics, and in the sixth patient only a very small potential was recognized. These findings were interpreted to mean that tics differed from voluntary movements because of their lack of cortical preparatory activity. Karp et al. (1996) repeated this study and found similar results. Three of five patients showed no premotor activity and the other two showed a small negativity prior to tics (Fig. 2). The negativity when present was brief, and it is possible to interpret their findings that there was an NS I, but no BP. The NS I is likely to be related very closely to the generation of the motor command. The function of the BP is not clear, and hence its absence prior to tic movements is not yet interpretable. In normal reaction time movements (triggered by sensory stimuli), there is no BP either, so that it certainly is possible for the brain to generate normal movements without a BP.

M. HALLETT

552

(spontaneou~

(VOlunhru) patient

IVJ~

2

3

I

i 5

",••

1

t"w..... t"..J·~

ouv"1 . L._. _ 500 m,

Fig. 2. Movement related cortical potentials accompanying voluntarily imitated and spontaneous tics. All tracings are from Cz referenced to linked ears. The horizontal line represents the baseline and the vertical line indicates the time of EMG onset. Only in patients land 3 is there a brief negativity preceding the tic movement. From Karp et al. (1996) with permission.

From a practical point of view, the absence of the BP may be useful in some clinical circumstances. With psychogenic involuntary movements, studies have shown that there is a normal looking BP in most patients (Terada et al., 1995). On the other hand, absence of the BP is a feature of some other involuntary movements such as blepharospasm (Berardelli et al., 1985).

33.4. External triggering of tics and the startle reflex Tics can be triggered by external stimuli (Tijssen et al., 1999). In some cases, this is like a conditioned

response to a specific behavioral context (Commander et al., 1991). Echolalia may be encompassed by this concept. One of the patients reported by Commander et al. (1991), for example, coughed when he heard a cough. This phenomenon has not been extensively studied. Patients with TS are some times thought to have exaggerated startle (Corbett, 1976; Lees etal., 1984; Commander et al., 1991; Nishida et al., 1994). Three of 53 patients reported by Lees et al, (1984), for example, were said to have exaggerated startle. In all three, a loud noise produced a blink which was followed by a whole body jerk. This whole body jerk was said to closely resemble the patients' tics in

553

TICS

these cases. Corbett (1976) has made some observations using high speed photography and video suggesting that some patients have exaggerated startle. Three groups have studied startle quantitatively. Sachdev et al. (1997) studied IS TS patients and 15 control subjects. The TS patients did not differ from controls in the onset latency, amplitude, and first peak latency of the reflex response in any of the muscles. Rates of habituation studied at 3 second intervals in the orbicularis oculi muscle were widely variable across the subjects, and the two groups did not differ overall. Stell et al. (1995) reported on eight patients with TS and 15 normal controls. Stimulation was given at 10 second intervals, which is likely a more sensitive rate to detect abnormality. Two patients failed to habituate with repetition demonstrating that some patients with TS do have exaggerated audiogenic startle responses. The authors note that phenomenon may be clinically asymptomatic. Gironell et al. (2000) studied 10 patients with a matching number of controls. Patients showed a significantly higher amplitude, a major degree of spread and less habituation. 33.5. Blinking and the blink reflex Spontaneous blink rate is frequently abnormal in disorders of dopamine. For example, blink rate is depressed in Parkinson's disease and normalizes with levodopa treatment. An etiological role for dopaminergic hyperactivity has been proposed in TS, and an increased blink rate would be consistent. Two studies are consistent with this idea (Bonnet, 1982; Tulen et aI., 1999). In one of these studies (Tulen et aI., 1999), nine TS patients were observed during periods of rest, conversation, and video watching. In comparison with controls, the patients showed a significantly higher blink rate during rest and video watching. In five patients, a significant positive correlation between blink rate and eye tic frequency was found. On the other hand, in a third study, 19 patients had blink rates during placebo treatment identical with those of 49 controls during reading and quiet sitting (Karson et aI., 1985). In this study, however, blink rate did correlate with both the number and severity of tics. The blink reflex recovery cycle is a measure of brainstem inhibitory mechanisms. Smith and Lees (1989) studied the blink reflex recovery cycle in 26

patients with TS and 10 controls. The amplitude of the R2 response following paired shocks was 11%, 40% and 52% of the conditioning stimulus at intervals of 200 ms, 500 ms and 1 s in the patients, compared with 10%, 17% and 32% respectively in the controls. These results suggest increased excitability of brainstem intemeurons in TS. A similar loss of inhibition is found in patients with blepharospasm making it difficult to separate tics and blepharospasm with this test (Berardelli et aI., 1985). The blink can also be studied with prepulse inhibition. This effect is a reduction of the amplitude of the blink reflex produced by a preceding stimulus. Studies of patients with OCD alone have shown a deficiency of prepulse inhibition so that studies need to be controlled for this co-morbid condition. Patients with both TS and ADHD showed a deficiency of prepulse inhibition (Castellanos et aI., 1996). Since prior studies in patients with isolated tics or ADHD alone had been normal, the authors concluded that this combination of disorders would lead to defective prepulse inhibition. More recently, another study of prepulse inhibition in patients with TS alone has demonstrated a reduction (Swerdlow et aI., 2001). The effect was greater with a tactile than an auditory prepulse demonstrating that the exact technique is important. Since prepulse inhibition indicates the ability of sensory stimuli to inhibit motor behavior, it is therefore sometimes referred to as sensorimotor gating. If sensory stimuli are less effective in gating motor behavior, this could be a factor in how sensory tics trigger motor tics. 33.6. Sleep Both motor and verbal tics may persist during non-REM and REM sleep (Glaze et aI., 1983). The frequency of tics is particularly high in REM sleep (Cohrs et aI., 2001). The tics have a similar appearance to those during wakefulness. This indicates that the generating mechanism for the tics does not require volition, an ambiguity when patients are awake. Additionally, the generator is not fully suppressed by sleep. The tics may cause arousals and disturb sleep (Drake et aI., 1992a). There is a correlation betweeen tic severity and the amount of sleep disruption (Cohrs et aI., 2001). Periodic leg movements (PLMS) during sleep are increased in patients with TS (Voderholzer et aI.,

554

1997). Seven drug-free patients with TS and seven controls were studied. Abnormally frequent PLMS were found in five, and periodic arm movements were seen in four patients. PLMS may be due to decreased inhibition of the spinal cord and thus shares the theme of decreased inhibition with TS. However, PLMS responds to dopaminergic therapy while TS responds to dopamine blockers. 33.7. Cortical inhibition Ziemann et al. (1997) tested motor cortex excitability directly using transcranial magnetic stimulation. Twenty patients with TS and 21 controls were studied. Focal transcranial magnetic stimulation was applied to the left motor cortex, and surface EMG was recorded from the right abductor digiti minimi muscle. Motor threshold, cortical silent period, and intracortical inhibition and facilitation were studied. Motor threshold and peripheral motor excitability were normal in the TS patients, but the cortical silent period was shortened and the intracortical inhibition reduced. A subgroup analysis of the patients with TS revealed that these abnormalities were seen mainly when tics were present in the EMG target muscle or in patients without neuroleptic treatment. Age, sex, ADHD, OCD, and sensory urges had no significant effect on motor excitability. The authors concluded that their findings were consistent with the hypothesis that tics in TS originate either from a primarily subcortical disorder affecting the motor cortex through disinhibited afferent signals or from impaired inhibition directly at the level of the motor cortex or both. These findings are not specific to TS since they are also seen in other disorders such as dystonia (Hallett, 1998). 33.8. Neuroimaging PET studies of basal metabolism show abnormalities. Baxter (1990) and Baxter et al. (1990) have studied patients with OCD including some with tics and found abnormalities of orbital prefrontal cortex and striatum. In one patient with tics and OCD studied with oxygen metabolism, there was hypermetabolism in the caudate which normalized after a successful limbic leucotomy (Sawle et al., 1993). Other studies of tic with or without OCD have shown increased blood flow in the right frontal cortex similar to patients with OCD alone (George et

M. HALLETI

al., 1992). In a study of 16 drug-free TS patients, Braun et al. (1993) found decreases in metabolic rates in paralimbic and ventral prefrontal cortices, particularly in orbitofrontal, inferior insular, and parahippocampal regions, and in subcortical regions, including the ventral striatum (nucleus accumbensl ventromedial caudate) and in the midbrain. There were also bilateral increases in metabolic activity in the supplementary motor, lateral premotor, and Rolandic cortices. Eidelberg et al. (1997) studied the functional brain networks in TS by using j8F_ fluorodeoxyglucose-PET employing a statistical model of regional metabolic covariation (Scaled Subprofile Model, SSM). While global and regional metabolic rates were normal in TS, SSM analysis identified two TS-related brain networks. One pattern was characterized by covariate bilateral metabolic increases in lateral premotor and supplementary motor association cortices and in the midbrain. A second pattern was characterized by covariate decreases in caudate and thalamic metabolism associated with smaller reductions in lentiform and hippocampal metabolic activity. One study has looked at performance of a fingertapping task with fMRI (Biswal et al., 1998). Results from 5 patients with TS were compared with 5 controls. The average number of pixels activated in the sensorimotor cortices and supplementary motor area was greater in the patients. Peterson et al. studied the physiology of tic suppression compared with "allowing tics to occur freely" using fMRI (Peterson et al., 1998). They found a complex pattern of changes including bilateral decreases in basal ganglia and thalamus that correlated inversely with the severity of the tic disorder. Stem et al. (2000) studied tic production with event-related, blood flow PET techniques. Brain regions in which activity was significantly correlated with tic occurrence included medial and lateral premotor cortices, anterior cingulate cortex, dorsolateral-rostral prefrontal cortex, inferior parietal cortex, putamen, and caudate, as well as primary motor cortex, Broca's area, superior temporal gyrus, insula, and claustrum. These studies are difficult to interpret and suggest a large network of the brain is responsible for tics. In an object manipulation task, patients with TS showed less activation than normal in secondary motor areas (Serrien et al., 2002). As the activation is measured with respect to baseline, the investigators

TICS

suggested that the secondary motor areas are constantly active, perhaps since they are generating the urges to move. The findings with FDG-PET reported above are consistent with a baseline overactivity of secondary motor areas.

33.9. Treatment The standard treatments for TS have been the dopamine blocking drugs and clonidine. The newer neuroleptic drugs such as olanzepine can be effective as well (Stamenkovic et al., 2000; Budman et aI., 2001). Kwak et aI. (2000) treated TS patients with botulinum toxin (BTX) in the sites of their most problematic tics. The authors noted the surprising result that in addition to improving the motor component of tics, BTX also provided relief of premonitory sensations. Marras et aI. reported another double blind controlled trial (Marras et aI., 2001). While there was objective evidence of decreased frequency of tics, the patients did not report a subjective benefit. The physiology of the effect BTX on both motor and sensory tics deserves further study.

33.10. Synthesis and speculation I have previously proposed this synthesis and speculation (Hallett, 2000). Tics are repetitive, stereotyped movements similar to voluntary movements. A significant difference appears to be in the way they are triggered, although both are generally triggered internally rather than externally. Voluntary movements are triggered by the poorly understood processes of volition. Tics are triggered by an unknown generator that is also characterized by production of "sensory tics" and "psychic tension" and that produces a sense of "relief' after a movement is made. The tic generator does not make a BP component in the movementrelated cortical potentials. There is considerable analogy between tics and compulsive acts, and possibly between itching and scratching in normal individuals. An important feature of the tic generator is the "need to move." This feature is also seen in other movement disorders including akathisia and the restless legs syndrome. Although the clinical manifestations differ in these other conditions, they all share a possible relationship with dopamine metabolism.

555

TS patients also differ from normal subjects by having reduced inhibition in their motor systems at multiple levels. This is appreciated from direct cortical investigations with magnetic stimulation, blink reflex recovery curves, decreased prepulse inhibition, increased startle reflexes, sensory triggering of the tics, and release during sleep. While the tic movements appear to be produced rather easily, there are some difficulties the patients have in producing voluntary movement. In particular, there is a prolonged reaction time, an increased reliance on external stimuli and difficulty in switching motor programs. Internal triggering of movement appears to employ a network of motor structures with a special role for medial motor areas, including the supplementary motor area (Deiber et aI., 1996). A similar set of regions is activated during the urge to scratch (Hsieh et al., 1994), and direct electrical stimulation of the supplementary motor area can produce the urge to make a movement (Fried et aI., 1991). Indication of dysfunction of this network in TS comes from neuroimaging. These regions are under the influence of the basal ganglia which play an important role in movement generation and the level of brain inhibition. The basal ganglia network is likely the source of the problem, but the nature of the disorder is not clear. The basal ganglia malfunction may be in the "complex loop" between the frontal cortex and caudate (Alexander and Crutcher, 1990) as well as in the "motor loop" between the motor areas and the putamen.

Acknowledgment This review is similar to and updated from an earlier review (Hallett, 2000). Work of the U.S. government, it has no copyright.

References Alexander, GE and Crutcher, MD (1990) Functional achitecture of basal ganglia circuits: neural substrates of parallel processing. Trends Neurosci., 13: 266-271. Baxter, LR (1990) Brain imaging as a tool in establishing a theory of brain pathology in obsessive compulsive disorder. 1. Clin. Psychiatry, 51 (Suppl.): 22-25. Baxter, LRJ, Schwartz, JM, Guze, BH, Bergman, K and Szuba, MP (1990) PET imaging in obsessive compulsive disorder with and without depression. J. Clin. Psychiatry, 51 (Suppl): 61-69.

556 Berardelli, A, Rothwell, JC, Day, BL and Marsden, CD (1985) Pathophysiology of blepharospasm and oromandibular dystonia; Brain, 108: 593-608. Biswal, B, Ulmer, JL, Krippendorf, RL, Harsch, HH, Daniels, DL, Hyde, JS et al. (1998) Abnormal cerebral activation associated with a motor task in Tourette syndrome. AJNR Am. J. Neuroradiol., 19: 1509-1512. Bonnet, KA (1982) Neurobiological dissection of Tourette syndrome: a neurochemical focus on a human neuroanatomical model. In: AJ Friedhoff and TN Chase (Eds.), Gilles de la Tourette Syndrome. Raven Press, New York, pp. 77-82. Braun, AR, Stoetter, B, Randolph, C, Hsiao, JK, Vladar, K, Gernert, J et al. (1993) The functional neuroanatomy of Tourette's syndrome: an FDG-PET study. 1. Regional changes in cerebral glucose metabolism differentiating patients and controls. Neuropsychopharmacology, 9: 277-291. Budman, CL, Gayer, A, Lesser, M, Shi, Q and Bruun, RD (2001) An open-label study of the treatment efficacy of olanzapine for Tourette's disorder. J. Clin. Psychiatry, 62: 290-294. Castellanos, FX, Fine, EJ, Kaysen, D, Marsh, WL, Rapoport, JL and Hallett, M (1996) Sensorimotor gating in boys with Tourette's syndrome and ADHD: Preliminary results. Biol. Psychiatry, 39: 33-41. Cohrs, S, Rasch, T, Altmeyer, S, Kinkelbur, J, Kostanecka, T, Rothenberger, A et al. (2001) Decreased sleep quality and increased sleep related movements in patients with Tourette's syndrome. 1. Neurol. Neurosurg. Psychiatry, 70: 192-197. Commander, M, Corbett, J, Prendergast, M and Ridley, C (1991) Reflex tics in two patients with Gilles de la Tourette syndrome. Br. 1. Psychiatry, 159: 877-879. Cope, MT, Georgiou, N, Bradshaw, JL, Iansek, Rand Phillips, JG (1996) Simon effect and attention in Parkinson's disease: a comparison with Huntington's disease and Tourette's syndrome. J. Clin. Exp. Neuropsychol., 18: 276--290. Corbett, JA (1976) The nature of tics and Gilles de la Tourette's syndrome. In: F Abuzzahab and F Anderson (Eds.), Gilles de la Tourette Syndrome International Registry. Mason Publishing Co., St. Paul. Deiber, M-P, Ibanez, V, Sadato, N and Hallett, M (1996) Cerebral structures participating in motor preparation in humans: a positron emission tomography study. J. Neurophysiol., 75: 233-247. Drake, ME, Jr, Hietter, SA, Padamadan, H and Bogner, JE (1991) Computerized EEG frequency analysis in Gilles de la Tourette syndrome. Clin. Electroencephalogr., 22: 250-253. Drake, ME, Jr, Hietter, SA, Bogner, JE and Andrews, JM (1992a) Cassette EEG sleep recordings in Gilles de la

M.HALLETT

Tourette syndrome. Clin. Electroencephalogr., 23: 142- 146. Drake, ME, Jr, Hietter, SA, Padamadan, H, Bogner, JE, Andrews, JM and Weate, S (1992b) Auditory evoked potentials in Gilles de la Tourette syndrome. Clin. Electroencephalogr., 23: 19-23. Dursun, SM, Burke, JG and Reveley, MA (2000) Antisaccade eye movement abnormalities in Tourette syndrome: evidence for cortico-striatal network dysfunction? J. Psychopharmacol., 14: 37-39. Eidelberg, D, Moeller, JR, Antonini, A, Kazumata, K, Dhawan, V, Budman, C et al. (1997) The metabolic anatomy of Tourette's syndrome. Neurology, 48: 927934. Farber, RH, Swerdlow, NR and Clementz, BA (1999) Saccadic performance characteristics and the behavioural neurology of Tourette's syndrome. J. Neurol. Neurosurg. Psychiatry, 66: 305-312. Flanagan, JR, Jakobson, LS and Munhall, KG (1999) Anticipatory grip adjustments are observed in both goal-directed movements and movement tics in an individual with Tourette's syndrome. Exp. Brain Res., 128: 69-75. Fried, I, Katz, A, McCarthy, G, Sass, KJ, Williamson, P, Spencer, SS et al. (1991) Functional organization of human supplementary motor cortex studied by electrical stimulation. J. Neurosci., 11: 3656--3666. George, MS, Trimble, MR, Costa, DC, Robertson, MM, Ring, HA and Ell, PJ (1992) Elevated frontal cerebral blood flow in Gilles de la Tourette Syndrome: A 99Tc ffi _ HMPAO SPECT study. Psychiatry Res. Neuroimaging, 45: 143-151. Georgiou, N, Bradshaw, JL, Phillips, JG, Bradshaw, JA and Chiu, E (1995a) Advance information and movement sequencing in Gilles de la Tourette's syndrome. J. Neurol. Neurosurg. Psychiatry, 58: 184-191. Georgiou N, Bradshaw JL, Phillips JG, Bradshaw JA, Chiu E (1995b) The Simon effect and attention deficits in Gilles de la Tourette's syndrome and Huntington's disease. Brain, 118, 1305-1318. Georgiou N, Bradshaw JL, Phillips JG, Cunnington R, Rogers M (1997) Functional asymmetries in the movement kinematics of patients with Tourette's syndrome. J. Neurol. Neurosurg. Psychiatry, 63, 188-195. Gironell, A, Rodriguez-Fomells, A, Kulisevsky, J, Pascual, B, Riba, J, Barbanoj, M et al. (2000) Abnormalities of the acoustic startle reflex and reaction time in Gilles de la Tourette syndrome. Clin. Neurophysiol., 111: 1366--1371. Glaze, DG, Frost, JD, Jankovic, J (1983) Sleep in Gilles de la Tourette's syndrome: disorder of arousal. Neurology, 33: 586--592. Gunther, W, Muller, N, Trapp, W, Haag, C, Putz, A and Straube, A (1996) Quantitative EEG analysis during

TICS

motor function and music perception in Tourette's syndrome. Eur. Arch. Psychiatry Clin. Neurosci., 246: 197-202. Hallett, M (1998) The neurophysiology of dystonia. Arch. Neurol., 55: 601-603. Hallett, M (2000) Neurophysiology of tics. In: DJ Cohen, CG Goetz and J Jankovic (Eds.), Tourette Syndrome. Lippincott, Williams and Wilkins, Philadelphia, pp. 237-244. Hallett, M and Marsden, CD (1981) Physiology and pathophysiology of the ballistic movement pattern. In: JE Desmedt (Ed.), Progress in Clinical Neurophysiology. Karger, Basel, pp. 331-346. Hsieh, JC, Hagermark, 0, Stahle-Backdahl, M, Ericson, K, Eriksson, L, Stone-Elander, S et al (1994) Urge to scratch represented in the human cerebral cortex during itch. J. Neurophysiol., 72: 3004-3008. Hyde, TM, Emsellem, HA, Randolph, C, Rickler, KC and Weinberger, DR (1994) Electroencephalographic abnormalities in monozygotic twins with Tourette's syndrome. Br. J. Psychiatry, 164: 8II-817. Johannes, S, Wieringa, BM, Mantey, M, Nager, W, Rada, D, Muller-Vahl, KR et al. (2001 a) Altered inhibition of motor responses in Tourette Syndrome and ObsessiveCompulsive Disorder. Acta. Neurol. Scand., 104: 3643. Johannes, S, Wieringa, BM, Nager, W, Muller-Yahl, KR, Dengler, Rand Munte, TF (2001b) Electrophysiological measures and dual-task performance in Tourette syndrome indicate deficient divided attention mechanisms. EurJ Neurol., 8: 253-260. Karp, BI, Porter, S, Toro, C and Hallett, M (1996) Simple motor tics may be preceded by a premotor potential. J. Neurol. Neurosurg. Psychiatry, 61: 103-106. Karson, CN, Kaufmann, CA, Shapiro, AK and Shapiro, E (1985) Eye-blink rate in Tourette's syndrome. J. Nerv. Ment. Dis., 173: 566-569. Knell, ER and Comings, DE (1993) Tourette's syndrome and attention-deficit hyperactivity disorder: evidence for a genetic relationship. J. Clin. Psychiatry, 54: 331-337. Krumholz, A, Singer, HS, Niedermeyer, E, Burnite, Rand Harris, K (1983) Electrophysiological studies in Tourette's syndrome. Ann. Neurol., 14: 638-641. Kwak, CH, Hanna, PA and Jankovic, J (2000) Botulinum toxin in the treatment of tics. Arch. Neurol., 57: 1190-1193. Lees, AJ, Robertson, M, Trimble, MR and Murray, NMF (1984) A clinical study of Gilles de la Tourette syndrome in the United Kingdom. 1. Neurol. Neurosurg. Psychiatry, 47: 1-8. Leonard, HL, Lenane, MC, Swedo, SE, Rettew, DC, Gershon, ES and Rapoport, JL (1992) Tics and Tourette's disorder: a 2- to 7-year follow-up of 54

557 obsessive-compulsive children. Am. J. Psychiatry, 49: 1244-1251. LeVasseur, AL, Flanagan, JR, Riopelle, RJ and Munoz, DP (200 I) Control of volitional and reflexive saccades in Tourette's syndrome. Brain, 124: 2045-2058. Marras, C, Andrews, D, Sime, E and Lang, AE (2001) Botulinum toxin for simple motor tics: a randomized, double-blind, controlled clinical trial. Neurology, 56: 605-610. Mostofsky, SH, Lasker, AG, Singer, HS, Denckla, MB and Zee, DS (2001) Oculomotor abnormalities in boys with Tourette syndrome with and without ADHD. J. Am. Acad. Child. Adolesc. Psychiatry, 40: 1464-1472. Neufeld, MY, Berger, Y, Chapman, J and Korczyn, AD (1990) Routine and quantitative EEG analysis in Gilles de la Tourette's syndrome. Neurology, 40: 1837-1839. Nishida, H, Shinbo, Y and Motomura, H (1994) (A study of pathogenesis and symptoms of Tourette's syndrome - mainly on the importance of startle reflex through Latah reaction). Seishin Shinkeigaku Zasshi, 96: 26-47. Oades, RD, Dittmann-Balcar, A, Schepker, R, Eggers, C and Zerbin, D (1996) Auditory event-related potentials (ERPs) and mismatch negativity (MMN) in healthy children and those with attention-deficit or Tourette/tic symptoms. Biol. Psychol., 43: 163-185. Obeso, JA, Rothwell, JC and Marsden, CD (1981) Simple tics in Gilles de la Tourette's syndrome are not prefaced by a normal premovement potential. J. Neurol. Neurosurg. Psychiatry, 44: 735-738. Peterson, BS, Skudlarski, P, Anderson, AW, Zhang, H, Gatenby, JC, Lacadie, CM et al. (1998) A functional magnetic resonance imaging study of tic suppression in Tourette syndrome. Arch. Gen. Psychiatry, 55: 326333. Sachdev, PS, Chee, KY and Aniss, AM (1997) The audiogenic startle reflex in Tourette's syndrome. Biol. Psychiatry, 41: 796-803. Sawle, GV, Lees, AJ, Hymas, NF, Brooks, DJ and Frackowiak, RS (1993) The metabolic effects of limbic leucotomy in Gilles de la Tourette syndrome. J. Neurol. Neurosurg. Psychiatry, 56: 1016-1019. Serrien, OJ, Nirkko, AC, Loher, TJ, Lovblad, KO, Burgunder, JM and Wiesendanger, M (2002) Movement control of manipulative tasks in patients with Gilles de la Tourette syndrome. Brain, 125: 290-300. Shucard, DW, Benedict, RH, Tekok-Kilic, A and Lichter, DG (1997) Slowed reaction time during a continuous performance test in children with Tourette's syndrome. Neuropsychology, II: 147-155. Singer, HS and Walkup, JT (1991) Tourette syndrome and other tic disorders. Diagnosis, pathophysiology, and treatment. Medicine (Baltimore), 70: 15-32.

558 Smith, SJ and Lees, AJ (1989) Abnormalities of the blink reflex in Gilles de la Tourette syndrome. J. Neurol. Neurosurg. Psychiatry, 52: 895-898. Stamenkovic, M, Schindler, SD, Aschauer, HN, De Zwaan, M, Willinger, D, Resinger, E et al. (2000) Effective open-label treatment of Tourette's disorder with olanzapine. Int. Clin. Psychopharmacol., 15: 2328. Stell, R, Thickbroom, GW and Mastaglia, FL (1995) The audiogenic startle response in Tourette's syndrome. Mov. Disord., 10: 723-730. Stem, E, Silbersweig, DA, Chee, KY, Holmes, A, Robertson, MM, Trimble, M et al. (2000) A functional neuroanatomy of tics in Tourette syndrome. Arch. Gen. Psychiatry, 57: 741-748. Straube, A, Mennicken, JB, Riedel, M, Eggert, T and Muller, N (1997) Saccades in Gilles de la Tourette's syndrome. Mov. Disord., 12: 536-546. Swerdlow, NR, Karban, B, Ploum, Y, Sharp, R, Geyer, MA . and Eastvold, A (2001) Tactile prepuff inhibition of startle in children with Tourette's syndrome: in search of an "fMRI-friendly" startle paradigm. BioI. Psychiatry, 50: 578-585. Terada, K, Ikeda, A, Van Ness, PC, Nagarnine, T, Kaji, R, Kimura, J et al. (1995) Presence of Bereitschaftspoten-

M.HALLETT

tial preceding psychogenic myoclonus: clinical application of jerk-locked back averaging. J. Neurol. Neurosurg. Psychiatry, 58: 745-747. Tijssen, MA, Brown, P, Morris, HR, Lees, A (1999) Late onset startle induced tics. J. Neurol. Neurosurg. Psychiatry, 67: 782-784. Tulen, JH, Azzolini, M, de Vries, JA, Groeneveld, WH, Passchier, J and Van De Wetering, BJ (1999) Quantitative study of spontaneous eye blinks and eye tics in Gilles de la Tourette's syndrome. J. Neurol. Neurosurg. Psychiatry, 67: 80D-802. Van Woerkom, TC, Roos, RA and Van Dijk, JG (1994) Altered attentional processing of background stimuli in Gilles de la Tourette syndrome: a study in auditory event-related potentials evoked in an oddball paradigm. Acta. Neurol. Scand,90: 116-123. Voderholzer, D, Muller, N, Haag, C, Riemann, D and Straube, A (1997) Periodic limb movements during sleep are a frequent finding in patients with Gilles de la Tourette's syndrome. J. Neurol., 244: 521-526. Ziemann, D, Paulus, W and Rothenberger, A (1997) Decreased motor inhibition in Tourette's disorder: evidence from transcranial magnetic stimulation. Am. J. Psychiatry, 154: 1277-1284.

Movement Disorders Handbook of Clinical Neurophysiology, Vol. I

M. Hallett (Ed.) © 2003 Elsevier B.Y. All rights reserved

559 CHAPTER 34

Electrophysiological investigations in cranial hyperkinetic syndromes M. Aramideb'v", J. Valls-Sole" and B.W. Ongerboer de Visser b 'Department of Neurology/Clinical Neurophysiology, Medical Center Alkmaar, 1800 AM Alkmaar; The Netherlands h Department of Neurology/Clinical Neurophysiology Unit, Academic Medical Center, Amsterdam, The Netherlands c Unitat d' EMG, Servei de Neurologia, Hospital Clinic, Barcelona, Spain

34.1. Introduction In this chapter, we will focus on electrophysiological assessments of patients with the so-called cranial hyperkinetic syndromes. The term cranial hyperkinetic syndrome is introduced to indicate those disorders that are characterized by involuntary contractions of the muscles innervated by the cranial nerves. As it is impossible to discuss all types of cranial hyperkinetic syndromes in a chapter of a handbook, hemifacial spasm, the post facial palsy synkinesia, blepharospasm and its related eyelid movement disorders, such as involuntary levator inhibition and orbicularis oculi motor persistence, are selected for presentation in this chapter. Some other facial movement abnormalities, such as hemimasticatory spasm, facial action myoclonus and facial myokymia will be mentioned briefly.

34.2. Hemifacial spasm Hemifacial spasm (HFS) is characterized by paroxysmal involuntary bursts of initially progressive, irregular, clonic or tonic contractions of the facial muscles on one side. HFS is idiopathic in most instances, but a comprehensive medical history and a series of tests are needed to rule out intra- or extraaxial brain stem pathology. Several types of posterior fossa tumors have also been reported in association with hemifacial spasm (Digre and Corbett, 1988).

* Correspondence to: Dr. Majid Aramideh, MD, PhD, Department of Neurology/Clinical Neurophysiology, Medical Center Alkmaar, P.O. Box 501, 1800 AM Alkmaar, The Netherlands. E-mail address:[email protected]

Although there is no general agreement, the most commonly reported cause of the so-called "primary" HFS is compression by a vascular malformation or an artery, often a cerebellar artery, impinging on the facial nerve at its exit from the pons (Digre and Corbett, 1988). The condition is called primary to distinguish this type of HFS from HFS due to tumors in the brain stem. Standard needle EMG recording shows neither denervation potentials nor abnormal motor unit potentials, but paroxysmal bursts of highfrequency discharges (up to 300 Hz) of motor unit action potentials. The firing pattern varies, with short bursts of one or a few motor unit potentials, lasting from 10 to 100 ms, or prolonged spasms of several synchronized motor units, lasting even some seconds. Simultaneous recordings from an upper and a lower facial muscle often show this paroxysmal activity to be synchronous in the two muscles (Fig. 1). If, by chance, spontaneous activity is scarce or absent at the moment of recording, it may be induced by two to three minutes of hyperventilation, which presumably causes respiratory alkalosis that decreases the calcium level, that in tum triggers ectopic excitation (Nielsen, 1984). Facial stimulation in HFS causes a normal CMAP that is sometimes followed, immediately or after a pause of electrical silence, by after-activity. This is highly variable, and is far better examined by needle-recordings. It may consist of a single motor unit potential or a doublet, or trains of potentials lasting 50 ms or more. In any event, the inter-spike frequency is very high, as it is for the spontaneous activity. Electrical stimuli applied to the supraorbital nerve may also activate small collateral axons of the facial nerve terminals innervating the orbicularis oculi muscle. This feature has been used by Montero et a1.

M. ARAMIDEH ET AL.

560 A

Right OOculi

1100/lV 200ms

Rig.tO~

B Right OOculi

RightOOris

Fig. 1. Surface recording of the EMG activity in the orbicularis oculi and orbicularis oris in a patient with postparalytic facial syndrome and hemifacial synkinesis (A), and a patient with idiopathic hemifacial spasm (B). Note the synkinetic bursts of EMG in the orbicularis oris with blink in A, and the absence of spread of the EMG activity to the lower facial muscles in the first blink in B.

(1996) to demonstrate the presence of axon reflexes in lesions of the facial nerve. Electrical stimuli are applied over the supraorbital nerve, and responses are recorded from the ipsilateral orbicularis oculi and orbicularis oris muscles. In these conditions, and RI response can be expected in both muscles in patients with hemifacial spasm as well as in those with aberrant regeneration. However, it is also possible that the responses recorded in ipsilateral muscles are not true reflex Rl responses but axon reflex responses conveyed antidromically by impulses traveling through fibers of the facial nerve up to the point of the lesion, and then back to facial muscles innervated by other branches. Montero et al. (1996) applied step by step stimuli, as in the inching

technique, from the supraorbital zone towards the malar zone and following the zygomatic branch of the facial nerve. The striking observation was that the 'Rl' responses decreased their latency when the stimulating electrode was moved from the supraorbital point to a malar point. With this maneuver the length of the axon reflex arc is shorter and, consequently, there is a decrease in response latency. If the response was due to a trigemino-facial reflex, induced by supraorbital nerve stimulation, the Rl response latency should increase after malar point stimulation because the length of the sensory nerve, in this case, would be longer. Selective stimulation of the mandibular branch of the facial nerve sometimes evokes a response in the

561

ELECTROPHYSIOLOGICAL INVESTIGATIONS IN CRANIAL HYPERKINETIC SYNDROMES L NVII. 2nd branch stimulus (*)

r·.lV'w""-:--"">---

L NVII,3rd branch stimulus (*) ~ Orb Oculi m

.. --------1 Mentalis m

L L---"---_

~

--'

Fig. 2. Ephaptic transmission in a patient with hemifacial spasm. Left panel: the evoked potentials in the left orbicularis oculi muscle (upper three traces) and in the left mentalis muscle (lower three traces) via the lateral spreading after the stimulation of the second branch of the left facial nerve. Right panel: the same recording after the stimulation of the third branch of the left facial nerve (horizontal bar: 5 ms, vertical bar: 0.2 mV).

orbicularis oculi muscle after about 9 ms, whereas stimulation of the zygomatic branch may conversely evoke a delayed response in the lower face muscles (Fig. 2). This has been interpreted as an expression of the ephaptic transmission, also known as lateral spreading, taking place at the facial nerve root entry zone, in the posterior fossa, where nerve damage can be caused by a compressing arterial vessel (Nielsen, 1984). Ectopic generation of discharges due to local demyelination of the facial nerve root could be responsible for the spasms. However, facial motoneuronal hyperexcitability has also been suggested, which may arise from a 'kindling' effect triggered

by the peripheral nerve lesion (Meller and Jannetta, 1984; Eekhof et al., 2000). In any instance, the contribution of hyperexcitability of facial nerve motoneurons and trigemino-facial intemeurons has been demonstrated in a proportion of patients, by studying the blink reflex excitability recovery curve (Valls-Sole and Tolosa, 1989; Eekhof et al., 1996). Regardless of the presence of paroxysmal spontaneous activity, blink reflex studies are very useful in the evaluation of HFS. In recordings from the orbicularis oculi muscle, Rl and R2 are sometimes slightly delayed on the ipsilateral side, although not all investigators agree with this finding. Stimulation

orbicularis oculi

orbicularis oculi

orbicularis oris

orbicularis oris

r

~

IO.2mV

10ma

Fig. 3. Example of synchronic early and late reflex responses in the orbicularis oculi and oris muscles on the right (r) side, evoked by stimulation of the right supraorbital nerve (r*) in a patient with chronic Bell's palsy. Stimulation of the left supraorbital nerve (1*), on the unaffected side, elicits normal reflex responses in the left (I) orbicularis oculi muscle. Responses on the right side are slightly delayed.

562

M. ARAMIDEH ET AL.

orbicularis oculi

.--...... 10.2 mV 10ms

orbicularis oris

Fig. 4. Example of repetitive discharges in synchronic reflex responses in the left (1) orbicularis oculi and oris muscles after stimulation of the left supraorbital nerve (1*) in a patient with idiopathic hemifacial spasms on the left side. Responses were picked up by the surface electrodes and two responses were recorded below each other.

of the supraorbital nerve often evokes anomalous responses that resemble the two blink reflex components in lower face muscles, such as the orbicularis oris or mentalis (Fig. 3). This observation is most useful for the diagnosis (Auger, 1975). In HFS, the latency of the Rl-like response in lower face muscles is similar to the latency of the orbicularis oculi Rl response. Stimulation of facial nerve branches, as well as stimulation of the supraorbital nerve, may also evoke high-frequency discharges that are synchronous in upper and lower facial muscles (Fig. 4). 34.3. Post facial palsy synkinesia Peripheral facial nerve palsy is, with an annual incidence of 35 per 100,000, the most frequent encountered cranial nerve disorder (Katusic et al., 1986; Devriese et aI., 1990). It is common to observe an increased blinking rate in patients with facial palsy. Pastor et aI. (1998) reported enhanced blink reflex excitability recovery

curve in two patients examined within the first month after initiation of a Bell's palsy. This finding was later confirmed by Syed et al. (1999). This observation, together with the reported occurrence of contralateral blepharospasm in some patients with peripheral facial palsy (Baker et al., 1997), suggests that these patients may have a compensatory hyperactivity on their non-paralyzed side. It also indicates that plastic changes may take place in the central nervous system in an attempt to compensate for the eyelid weakness. Facial synkinesia occurs in most patients that have experienced significant axonal degeneration (Kimura et aI., 1975). The most likely cause is abnormal regeneration due to excessive branching of axons at the site of the lesion or ephaptic transmission (Montserrat and Benito, 1988). However, some evidence indicates changes in the facial nerve nucleus as well, because lesions involving just one facial branch may give rise to synkinesis in all hemifacial muscles (Martinelli et aI., 1992). Eekhof et aI. (2000) recorded ephaptic transmission in 50% of their patients, who developed facial synkinesia after Bell's palsy and suggested that there may be some alteration in the facial nucleus excitability. To investigate the contribution of the facial motoneuronal hyperexcitability in the generation of synkinetic movements, Cossu et aI. (1999) examined 18 patients with facial palsy at the time of expected onset of regeneration, when there were only one or a few motor units activated in the hemifacial muscles. Stimuli applied to ipsilateral or contralateral supraorbital nerves or even to the median nerve induced responses in the orbicularis oris formed by the same motor units that were firing tonically or under voluntary command (Fig. 5). These findings imply that synkinesia and abnormal firing of motor units occur already at onset of the regeneration process, and that at the same time there appears to be an enhanced excitability of facial nerve motoneurons. An increased excitability of the facial nerve reflex responses and myokymic discharges have been reported to be still present in 23 patients examined between 1.5 and 16 years after Bell's palsy (Valls-Sole et aI., 1992). Five of these patients experienced not just hemifacial synkinesia, but a severe discomfort, consisting of muscular pain, tension, and spasms induced by common maneuvers such as eating, smiling, kissing, or singing. This "post-paralytic facial syndrome" is one of the most

ELECTROPHYSIOLOGICAL INVESTIGATIONS IN CRANIAL HYPERKINETIC SYNDROMES

563

* = Activity related to spontaneous blinking Myokymic discharges ....= Activity time-locked to breathing

0=

......................................................., o:~ ..~~.~

.

Fig. 5. Needle EMG recording from the orbicularis oris in a patient with post-paralytic facial syndrome, eight months after presentation of the paralysis, showing: I: Tonic firing of small motor units at a frequency of about 16 Hz (white squares), 2: Rhythmic activation of a larger amplitude motor unit coinciding with inhalation of respiratory cycle (arrow heads), and 3: Large amplitude short duration bursts synchronous with spontaneous blinking (asterisks). Calibration is I s and 0.1 m V.

deteriorating consequences of severe facial nerve lesions, which fortunately responds well to injections with botulinum toxin (Boroojerdi et al., 1998). The "post-paralytic" HFS, though infrequent, should be differentiated from the "primary" HFS, as well as from post facial palsy synkinesia. In all three conditions, synkinesia between upper and lower facial muscles is clinically evident. Similarly, synkinetic responses are often evoked by selective stimulation of facial nerve branches or the supraorbital nerve. Simultaneous recording of the ongoing activity of orbicularis oculi and orbicularis oris muscles may show that in post facial palsy synkinesia every burst of activity induced in the orbicularis oculi muscle, i.e. in spontaneous or reflex blinks, spreads to the orbicularis oris muscle. On the contrary, in HFS, spread occurs in most but not all occasions.

34.4. Blepharospasm Blepharospasm is a focal dystonia of the eyelids (Marsden, 1976), characterized by tremulous, phasic and/or clonic discharges in the orbicularis oculi muscle (Fig. 6) (Berardelli et al., 1985; Aramideh et al., 1994; Aramideh et al., 1996). The antagonistic activity between the orbicularis oculi and the levator palpebrae muscles is disturbed in some patients

(Fig. 7) and a minority of patients also have involuntary levator palpebrae inhibition (vide infra). Although in some patients with blepharospasm the R2 response is prolonged (Berardelli et al., 1985), the reflex features are usually normal (Eekhof et al., 1996,2001; Deuschl and Goddemeier, 1998). Patients with blepharospasm have an enhanced brain stem interneuronal excitability, as investigated with the recovery curves of R2, similar to that found in patients with parkinsonism (Berardelli et aI., 1985; Tolosa et al., 1988; Cruccu et aI., 1991; Aramideh et al 1995a) (Fig. 8). However, the pathophysiology underlying such abnormality may be different for the two disorders. The exact pathophysiology of dystonia is still unknown (see Berardelli et al., 1998 for a review). In a group of 33 patients with involuntary eyelid closure based on EMG patterns, Aramideh et al. (1995a) showed that recovery of R2 was enhanced in ail patients with pure blepharospasm, but it was normal in all patients with pure involuntary levator palpebrae inhibition and in 75% of patients with a combination of both disorders. Nakashima et al. (1990), Pauletti et al. (1993) and Eekhof et al. (1996) found that the abnormalities of the recovery curve are also present in patients with cervical and generalized dystonia, even in those without blepharospasm. In contrast, normal excitability recovery curves are usually seen in patients

564

M. ARAMIDEH ET AL.

A

A LP

00

LP

~MiIltM~~.'~ lOOms

00

B

LP

LP 00

00

200 ms

0.2 mV

c 200 ms LP

B

00 \0,05 mV .

LP

-.....

I

200 ms

Fig. 6. Different types of dystonic acnvrnes of the orbicularis oculi muscle (00) in a patient with blepharospasm. A: tremulous discharges, while the eyes are kept closed. B: tonic activity causing forceful closure of the eyelids, accompanied by inhibition of the levator palpebrae muscle (LP) activity. C: phasic discharges in the 00 muscle accompanied by inhibition of LP activity.

with focal dystonia limited to the upper limb (Tolosa et al., 1988; Nakashima et aI., 1990). The recovery curves of the SP2 component of the masseter inhibitory reflex may be enhanced in dystonic patients, with or without facial dystonia (Cruccu et aI., 1991). It is interesting that the recovery of SP2 was also facilitated in dystonic patients not exhibiting jaw closing dystonia (Pauletti et aI., 1993), as was the recovery of the R2 component of the blink reflex, which was facilitated in patients without blepharospasm (Nakashima et aI., 1990). In the examination of sternocleidomastoid muscle inhibitory reflex in patients with spasmodic torticollis, Nakashima et aI. (1989) found a decrease in the duration and depth of EMG suppression. Carella et aI. (1994) found similar abnormalities, but in

00 0.2mV 200 ms

Fig. 7. Alternating dystonic activities of the levator palpebrae (LP) and the orbicularis oculi (00) muscles, recorded in a patient with blepharospasm. (A) The start of 00 involuntary muscle activity is not accompanied by immediate inhibition of LP, and vice versa. There are, thus, some periods of co-contraction due to affected reciprocal inhibition (upper two traces). On the command to close the eyes, there is an increase in duration of LP dystonic activities (lower two traces). (B) EMG recording several months after the last botulinum toxin injection into the 00 muscle shows a considerable increase of 00 dystonic muscle activities. The disturbed reciprocal inhibition is also more pronounced.

patients with blepharospasm without torticollis. These findings indicate that the spinal and brain stem interneuronal dysfunction is more widespread than the clinical signs and symptoms would suggest.

565

ELECTROPHYSIOLOGICAL iNVESTIGATIONS IN CRANIAL HYPERKINETIC SYNDROMES

interval (sec)

condition

test

'~"""""'~~~"""'~~~~-'-'-'-'-'A

recovery curve blink reflex R2 175,------------------------, 0.3

c

150

0.5 1

3 10 10 ms/div. condition

test

B 50

0.3 25

0.5

1 0.01

3 10

0.02

0.1

0.3

1.0 3.0 10.0 delay 52-81 (log sec)

20.0

10 ms/div.

Fig. 8. Recovery curve of the R2 response of the blink reflex. Rectified and averaged (6 x ) EMG responses are presented at intervals from 220 ms to 10 s between conditioning and test stimuli, in a control subject (A) and in a patient with blepharospasm (B). C showscompleterecovery curvesfor both subjects. The patientB showssignificantly less suppression at intervals smallerthan I s (Sl = conditioning stimulus; S2= test stimulus). Dystonia may arise from an abnormal processing of sensory inputs (Hallett, 1995), leading to an excessive activity in premotor circuits. GomezWong et al. (1998) found that prepulse inhibition of the blink reflex is preserved in patients with blepharospasm, who still benefit transiently from sensory tricks (geste antagonistique), while it is abnormally reduced in those who do not experience any relief from sensory tricks (Fig. 9).

34.4.1. 1nvoluntary levator papebrae inhibition Patients with involuntary levator palpebrae inhibition, also known as apraxia of eyelid opening, have difficulty in initiating the act of lid opening on command (Goldstein and Cogan, 1965; Boghen, 1997). Many of these patients also exhibit an inability to keep the eyelids open for a longer period of time. Synchronous needle EMG recordings from the levator palpebrae and the orbicularis oculi

muscles (Fig. 10) reveals involuntary inhibition of the levator palpebrae muscle activity causing inability to keep the eyelids open or to reopen them after involuntary closure of the lids (Aramideh et al., 1994). While the eyelids are open, the levator palpebrae may exhibit episodes of involuntary inhibition, without contraction of the orbicularis oculi, causing the eyelids to drop. Apraxia of eyelid opening, also known as involuntary levator palpebrae inhibition, may accompany blepharospasm (see focal dystonias) and is encountered more frequently in patients with extrapyramidal disorders.

34.4.2. Orbicularis oculi motor persistence Involuntary levator palpebrae inhibition should be differentiated from "motor persistence of the orbicularis oculi" muscle (Aramideh et aI., 1995b, 2001). Following voluntary closure of the eyelids on

566

M. ARAMIDEH ET AL.

A

LP

LP

00 ......, .... do' , • •

A'

be"''''

II..... . "....... '.....

Interval between prepulse and blink reflex (ms)

Fig. 9. Somatosensory prepulse inhibition in a healthy subject (A) and in patients with blepharospasm. The traces are rectified responses, recorded from the orbicularis oculi to single supraorbital nerve stimulation in the upper (control) traces, and the same stimulus (arrows), preceded by a weak electrical stimulus in the 3rd finger (thin vertical bars) in the lower (test) traces. C shows the results of examining a group of patients with or without sensory trick (ST), compared to a control group. The asterisks mark the intervals, at which there were statistically significant differences between control subjects and patients without sensory trick.

command (Fig. 11), the patients with the latter abnormality are also unable to open the lids on command. However, needle EMG recording shows that these patients are unable to suppress the activity of the orbicularis oculi muscle (Aramideh et al., 1995b, 2001). It is almost impossible to differentiate blepharospasm from involuntary levator palpebrae inhibition or motor persistence of orbicularis oculi based on the clinical observations alone.

34.5. Other facial movement disorders 34.5.1. Hemimasticatory spasm Hemimasticatory spasm is a rare condition, often associated with hemifacial atrophy (Kaufman, 1980; Kim et al., 2000), and is characterized by painful

10.2mv 200 ms

Fig. 10. EMG recording in a patient with involuntary levator palpebrae inhibition (apraxia of eyelid opening) and blepharospasm. (A) The patient shows involuntary inhibition periods (lIPs) of the levator palpebrae (LP) muscle, resulting in drooping of the eyelids and periods of suppression of LP activity resulting in inability to open the eyelids voluntarily on the command 'open eyes'. (B) Dense burst of phasic discharges during spasms of the orbicularis oculi muscle (00) are accompanied by inhibition of LP activity.

involuntary contractions of the masseter or temporal muscles on one side. The involuntary movements are paroxysmal and may appear as brief twitches or long-lasting spasms, as in hemifacial spasm. Needle electromyography of masticatory muscles shows typical high-frequency discharges of synchronized potentials, as in hemifacial spasm (Auger et al., 1992). The jaw-jerk is absent on the affected side. Study of the masseter inhibitory reflex during the spasm shows an efferent block, i.e. SPI and SP2 are markedly diminished or absent in the affected muscle, regardless of the side of stimulation. The

ELECTROPHYSIOLOGICAL INVESTIGATIONS IN CRANIAL HYPERKINETIC SYNDROMES

t

"open eyes"

567

slowing of conduction in motor nerve fibers (Cruccu et al., 1994). 34.5.2. Facial action myoclonus

LP

Abnormal facial movements can be observed in patients with multiple system atrophy, Wilson's disease or Huntington's chorea. In patients with olivopontocerebellar atrophy (OPCA), voluntary or automatic activation of facial muscles, when showing the teeth or smiling, may induce a rhythmic involuntary movement of the cheeks, known as facial action myoclonus (Lou et al., 1994). The study of brain stem reflexes in these patients revealed multiple scattered abnormalities in different circuits, suggesting possibly a multi-focal loss of neurons, with secondary changes in excitability (Valls-Sole et al., 1994).

00

LP

00

34.5.3. Facial myokymia

LP

00 10.5mV 200 ms

Involuntary twitches of portions of the orbicularis oculi muscle is common in normal individuals. These fasciculations generally affect only the lower eyelids. Orbicularis oculi muscle myokymia may be part of facial myokymia with different causes. Facial myokymia can appear in various facial muscles in patients with multiple sclerosis, Guillain-Barre syndrome, multiple system atrophy, intrinsic brain stem tumors such as pontine gliomas or extra-axial neoplasms (May and Galetta, 1990). The needle EMG shows motor units, which are of normal shape and duration and are grouped with a regular rhythmic discharge about every 100-200 ms. but are not synchronized throughout the muscle (Sivak et al., 1993). It should, however, also be mentioned that myokymic discharges may be seen using EMG recordings even in those patients with no apparent myokymia (Gutmann et al., 2001).

Fig. 11. Orbicularis oculi motor persistence. EMG recording showing that after voluntary closure of the eyelid and upon the command to "open eyes", patient is unable to suppress the contraction of the orbicularis oculi. After about I s, the density of the bursts of action potentials in the orbicularis gradually begins to diminish and is accompanied by an increase in intensity and frequency of the discharges in the levator palpebrae (LP). The eyelid opens when the orbicularis muscle activity is almost completely inhibited (at open arrow).

References

silent periods are absent probably because the motor potentials are ectopically generated along the nerve, and cannot be suppressed by the reflex inhibitory input on the motoneurons (Cruccu et al., 1991). Hemimasticatory spasm is probably secondary to a purely motor trigeminal neuropathy (Thompson et al., 1986), as indicated by the finding of a focal

Aramideh, M, Ongerboer de Visser, BW, Devriese, PP, Bour, LJ and Speelman, JD (1994) Electromyographic features of levator palpebrae superioris and orbicularis oculi muscles in blepharospasm. Brain, 117: 27-38. Aramideh, M, Eekhof, JLA, Bour, LJ, Koelman, JHTM, Speelman, JD and Ongerboer de Visser, BW (1995a) Electromyography and blink reflex recovery in involuntary eyelid closure: A comparative study (published erratum appears in J. Neurol. Neurosurg. Psychiatry,

568 59: 662). J. Neurol. Neurosurg. Psychiatry, 58: 692-698. Aramideh, M, Ongerboer de Visser, BW, Koelman, JHTM and Speelman, JD (1995b) Motor persistence of orbicularis oculi muscle in eyelid-opening disorders. Neurology, 45: 897-902. Aramideh, M, Ongerboer de Visser, BW, Holstege, G, Majoie, CBLM and Speelman, JD (1996) Blepharospasm in association with a lower pontine lesion. Neurology, 46: 476-478. Aramideh, M, Koelman, JHTM, Speelman, JD and Ongerboer de Visser, BW (2001) Eyelid movement disorders and electromyography. Lancet, 357: 805-806. Auger, RG (1975) Hemifacial spasm: Clinical and electrophysiologic observations. Neurology, 25: 989-993. Auger, RG, Litchy, WJ, Cascino, TL and Ahlskog, JE (1992) Hemimasticatory spasm: Clinical and electrophysiologic observations. Neurology, 42: 2263-2266. Baker, RS, Sun, WS, Hasan, SA, Rouholiman, BS, Chuke, JC, Cowen, DE and Porter, JD (1997) Maladaptive neural compensatory mechanisms in Bells' palsyinduced blepharospasm. Neurology, 49: 223-229. Berardelli, A, Rothwell, JC, Day, BL and Marsden, CD (1985) Pathophysiology of blepharospasm and oromandibular dystonia. Brain, 108: 593-609. Berardelli, A, Rothwell, JC, Hallett, M, Thompson, PD, Manfredi, M and Marsden, CD (1998) Pathophysiology of primary dystonia. Brain, 121: 1195-1212. Boghen, D (1997) Apraxia of lid opening: A review. Neurology, 48: 1491-1503. Boroojerdi, B, Ferbert, A, Schwarz, M, Herath, H and Noth, J (1998) Botulinum toxin treatment of synkinesia and hyperiacrimation after facial palsy. J. Neurol. Neurosurg. Psychiatry, 65: 111-114. Carella, F, Ciano, C, Musicco, M and Scaioli, V (1994) Exteroceptive reflexes in dystonia: a study of the recovery cycle of the R2 component of the blink reflex and of the exteroceptive suppression of the contracting sternocleidomastoid muscle in blepharospasm and torticollis. Mov. Disord., 9: 183-187. Cossu, G, Valls-Sole, J, Valldeoriola, F, Munoz, E, Benitez, P and Aguilar, F (1999) Reflex excitability of facial motoneurons at onset of muscle reinnervation after facial nerve palsy. Muscle Nerve, 22: 614-620. Cruccu, G, Pauletti, G, Agostino, R, Berardelli, A and Manfredi, M (1991) Masseter inhibitory reflex in movement disorders. Huntington's chorea, Parkinson's disease, dystonia, and unilateral masticatory spasm. Electroencephalogr. Clin. Neurophysiol., 81: 24-30. Cruccu, G, Inghilleri, M, Berardelli, A, Pauletti, G, Casali, C, Coratti, P, Frisardi, G, Thompson, PD and Manfredi, M (1994) Pathophysiology of hemimasticatory spasm. J. Neurol. Neurosurg. Psychiatry, 57: 43-50.

M. ARAMIDEH ET AL.

Deuschl, G and Goddemeier, C (1998) Spontaneous and reflex activity of facial muscles in dystonia, Parkinson's disease and in normal subjects. J. Neurol. Neurosurg. Psychiatry, 64: 320-324. Devriese, PP, Schumacher, T, Scheide, A, De Jongh, RH and Houtkooper, JM (1990) Incidence, prognosis and recovery of Bell's palsy. A survey of about 1000 patients. Clin. Otolaryngol., 15: 15-27. Digre, K and Corbett, JJ (1988) Hemifacial spasm: Differential diagnosis, mechanism, and treatment. Adv. Neurol.,49: 151-176. Eekhof, JLA, Aramideh, M, Bour, LJ, Hilgevoord, AAJ, Speelman, JD and Ongerboer de Visser, BW (1996) Blink reflex recovery curves in blepharospasm, torticollis spasmodica, and hemifacial spasm. Muscle Nerve, 19: 10-15. Eekhof, JLA, Aramideh, M, Speelman, JD, Devriese, PP and Ongerboer de Visser, BW (2000) Blink reflexes and lateral spreading in patients with synkinesia after Bell's palsy and in hemifacial spasm. Eur. Neurol., 43: 141-146. Eekhof, JLA, Aramideh, M, Speelman, JD and Ongerboer de Visser, BW (2001) Orbicularis oculi and orbicularis oris blink reflexes in blepharospasm and torticollis spasmodica during spasm-free intervals. Eur. Neurol. (in press). Goldstein, JE and Cogan, DG (1965) Apraxia of lid opening. Arch. Ophthalmol., 73: 155-159. Gomez-Wong, E, Marti, MJ, Tolosa, E and Valls-Sole, J (1998) Sensory modulation of the blink reflex in patients with blepharospasm. Arch. Neurol., 55: 1233-1237. Gutmann, L, Libell, D and Gutmann, L (2001) When is myokymia neuromyotonia? Muscle Nerve, 24: 151-153. Hallett, M (1995) Is dystonia a sensory disorder? Ann. Neurol., 38: 139-140. Kaufman, MD (1980) Masticatory spasm in facial hemiatrophy. Ann. Neurol., 7: 585-587. Katusic, SK, Beard, CM, Wiederholt, WC, Bergstralh, EJ and Kurland, LT (1986) Incidence, clinical features and prognosis in Bell's palsy, Rochester, Minnesota. Ann. Neurol., 20: 622-627. Kim, HJ, Jeon, BS and Lee, KW (2001) Hemimasticatory spasm associated with localized scleroderma and facial hemitrophy. Arch. Neurol., 57: 576-580. Kimura, J, Rodnitzky, RL and Okawara, S (1975) Electrophysiological analysis of aberrant regeneration after facial nerve paralysis. Neurology, 25: 989-993. Lou, JS, Valls-SoM, J, Toro, C and Hallett, M (1994) Facial action myoclonus in patients with olivopontocerebellar atrophy. Mov. Disord., 9: 223-226. Marsden, CD (1976) Blepharospasm-oromandibular dystonia syndrome (Brueghel's syndrome): a variant of

ELECTROPHYSIOLOGICAL INVESTIGATIONS IN CRANIAL HYPERKINETIC SYNDROMES

adult-onset torsion dystonia? J. Neurol. Neurosurg. Psychiatry, 39: 1204-1209. Martinelli, P, Giuliani, Sand Ippoliti, M (1992) Hemifacial spasm due to peripheral injury of facial nerve. A nuclear syndrome. Mov. Disord., 7: 181-184. May, M and Galetta, S (1990) The facial nerve and related disorders of the face. In: JS Glaser (Ed.), Neuro-ophthalmology (2nd ed.). Philadelphia: Lippincott, pp. 239-277. Meller, AR and Jannetta, PJ (1984) On the origin of synkinesis in hemifacial spasm. Results of intracranial recording. J. Neurosurg., 61: 569-576. Montero, J, Serra, J and Montserrat, L (1996) Axon reflexes or ephaptic responses simulating blink reflex Rl after XII-VII anastomosis. Muscle Nerve, 19: 848852. Montserrat, L and Benito, M (1988) Facial synkinesis and aberrant regeneration of facial nerve. In: J Jankovic and E Tolosa (Eds.) Advances in Neurology (Vol. 49). Raven Press, New York, pp. 211-224. Nakashima, K, Thompson, PD, Rothwell, JC, Day, BL, Stell, R and Marsden, CD (1989) An exteroceptive reflex in the sternocleidomastoid muscle produced by electrical stimulation of the supraorbital nerve in normal subjects and patients with spasmodic torticollis. Neurology, 39: 1354-1358. Nakashima, K, Rothwell, JC, Thompson, PD, Day, BL, Berardelli, A, Agostino, R, Artieda, J, Papas, SM, Obeso, JA and Marsden, CD (1990) The blink reflex in patients with idiopathic torsion dystonia. Arch. Neurol., 47: 413-416. Nielsen, VK (1984) Pathophysiology of hemifacial spasm: I. Ephpatic transmission and ectopic excitation. Neurology, 34: 418-426.

569

Pastor, P, Munoz, E, Valldeoriola, F et al. (1998) Enhanced blink rate and involuntary contralateral eye closure in patients with Bell's palsy. Muscle Nerve, 21: 1596. Pauletti, G, Berardelli, A, Cruccu, G, Agostino, Rand Manfredi, M (1993) Blink reflex and masseter inhibitory reflex in patients with dystonia. Mov. Disord., 8: 495-500. Sivak, M, Ochoa, J and Fernandez, J (1993) Positive manifestations of nerve fiber dysfunction: Clinical electrophysiologic and pathologic correlates. In: WF Brown and CF Bolton (Eds.), Clinical Electromyography (3rd ed). Boston: Butterworth-Heinemann, pp. 128-134. Syed, NA, Delgado, A, Sandbrink, F, Schulman, AE, Hallett, M and Floeter, MK (1999) Blink reflex recovery in facial weakness. An electrophysiological study of adaptive changes. Neurology, 52: 834-838. Thompson, PD, Obeso, JA, Delgado, G, Gallego, J and Marsden, CD (1986) Focal dystonia of the jaw and the differential diagnosis of unilateral jaw and masticatory spasm. J. Neurol. Neurosurg. Psychiatry, 49: 651-656. Tolosa, E, Montserrat, L and Bayes, A (1988) Blink reflex studies in focal dystonias. Enhanced excitability of brain stem interneurons in cranial dystonia and spasmodic torticollis. Mov. Disord., 3: 61-69. ValIs-Soh~, J and Tolosa, ES (1989) Blink reflex excitability cycle in hemifacial spasm. Neurology, 39: 1061-1066. ValIs-Soh~, J, Tolosa, ES and Pujol, M (1992) Myokymic discharges and enhanced facial nerve reflex responses after recovery from idiopathic facial palsy. Muscle Nerve, 15: 37-42. Valls-Sole, J, Lou, JS and Hallett, M (1994) Brain stem reflexes in patients with olivopontocerebellar atrophy. Muscle Nerve, 17: 1439-1448.

Movement Disorders Handbook of Clinical Neurophysiology, Vol. I M. Hallett (Ed.) © 2003 Elsevier B. V. All rights reserved

571 CHAPTER 35

Choreas, athetosis, dyskinesias, hemiballismus Alfredo Berardelli'> and Antonio Curta" "Dipartimento di Scienze Neurologiche, Universita di Roma "La Sapienza", Viale Universita 30,00185 Rome, Italy jo lstituto Neurologico Mediterraneo "Neuromed", Via Atinense 18, 86077 Pozzilli, IS, Italy

Chorea and choreoathetosis, ballism, and levodopainduced dyskinesias are involuntary movements that differ in rhythm, speed, duration, pattern, induction, and suppressibility. They may affect restricted regions or involve the entire body, and have various causes. Chorea refers to involuntary, abnormal movements that are unpredictable in timing, direction and distribution and seem to flow from one body part to another. Ballism consists of very large amplitude choreic movements occurring in the proximal part of the limbs and causing flinging and flailing limb movements. Athetosis describes a class of slow, writhing, continuous involuntary movements sometimes associated with sustained contractions (athetosis blending with dystonia), sometimes with faster choreic movements (athetosis blending with chorea, or choreoathetosis). Levodopa-induced dyskinesias (LIDs) refer to the unwanted, abnormal and excessive movements observed in patients with Parkinson's disease (PD) and associated with levodopa therapy.

35.1. Huntington's disease Huntington's disease (HD) is an autosomaldominant disease characterized by movement disorders and cognitive and behavioral changes. The pathologic feature of HD is the degeneration of neurons in the caudate, the putamen and the globus pallidus. In the early stages of the disease there is a selective loss of striatal GABNenkephalin neurons projecting to the globus pallidus externus (GPe) (the "indirect pathway"), accompanied in the later stages

* Correspondence to: Dr. A. Berardelli, Dipartimento di Scienze Neurologiche, Universita di Roma "La Sapienza", Viale Universita 30, 00185 Rome, Italy. E-mail address:[email protected] Fax: +39-06-4991 4302.

by a loss of striatal efferents projecting to the globus pallidus internus (GPi) (the "direct pathway"). The current functional model of HD suggests that in the early stages of the disease the indirect circuit to the GPi is preferentially affected (Reiner et aI., 1988; Albin 1992; Hedreen and Folstein, 1995). Some researchers, however, found no evidence of a preferential loss but disclosed parallel reductions of both pathways (Turjanski et aI., 1995; Ginovart et aI., 1997).

35.1.1. Involuntary movements EMG recordings of involuntary movements show bursts of EMG activity of variable duration with a random pattern of muscle activation and variable combination including co-contraction (Thompson et aI., 1988). As well as choreic movements, patients with HD may also manifest dystonic and myoclonic jerks (Thompson et aI., 1994).

35.1.2. Spinal cord function Presynaptic inhibition can be studied with the technique of reciprocal inhibition between agonist and antagonists muscles. With this method presynaptic inhibition reaches a maximum at the conditioning-test interval of 20 ms. In patients with HD we found reduced presynaptic inhibition in the arm muscles (Priori et aI., 2000). The most likely explanation for this abnormality is an abnormal descending control exerted by the basal ganglia on the spinal cord, possibly through the brain stem reticular formation. In the same patients, the Hreflex recovery cycle studied with paired electrical shocks also showed an abnormally prominent facilitation. This increased excitability could be induced by the decreased presynaptic inhibition of the large muscle afferents.

572 35.1.3. Brainstemfunction

Patients with HD may also have abnormalities of the "blink reflex", chiefly a prolonged latency and a greater habituation of the R2 response than normal (Caraceni et aI., 1976; Bollen et aI., 1986; Agostino et aI., 1988). Some of these abnormalities correlate with the distribution and severity of choreic movements in the face (Agostino et aI., 1988). They reflect reduced excitability of polysynaptic networks within the brainstem and indicate a depression of the blink reflex in HD. 35.1.4. Somatosensory and long-latency reflex abnormalities

A commonly reported abnormality in HD is a reduction in the size of the SEPs recorded at cortical level without changes at subcortical levels (Ehle et aI., 1984; Noth et aI., 1984; Abbruzzese et aI., 1990; Topper et aI., 1993). This abnormality is also present in asymptomatic gene carriers and in rigid forms of the disease, suggesting that the SEP abnormalities result from the disease and not from the choreic movements. A possible explanation is that the hyperactivity of the reticular nucleus of the thalamus reduces the relay of somatosensory inputs from the ventral posterior thalamic nuclear complex to the cortex. Some patients with HD also have reduced or absent long-latency stretch reflexes in hand and arm muscles (Noth et aI., 1983, 1985; Thompson et aI., 1988). These abnormalities are not present in other choreic disorders, implying that they are related to the underlying pathology of HD rather than to the chorea itself. 35.1.5. Cortical function

35.1.5.1. Movement-related potentials Recent research has shown significantly reduced amplitude of movement-related cortical potentials in patients with HD, regardless of whether the movement was internally or externally cued (Johnson et aI., 2001). External cues had no effects on the premovement cortical activity. These results have been interpreted as a problem in movement preparation in HD and reflect impaired activity in the SMA. 35.1.5.2. Execution of voluntary movements Intense research investigating disordered movement in HD has focused on the kinematic variables

A. BERARDELLI AND A. CURRA

operating during movement preparation and execution. Early evidence in patients with HD showed prolonged reaction times for simple and complex finger movements (Hefter et aI., 1987; Girotti et aI., 1988; Jahanshahi et aI., 1993). More recent evidence that patients with HD have the same delay in the reaction time during simple and sequential motor acts indicates that impaired motor preparation in HD probably depends on kinematic variables other than the number of sub-movements (Curra et aI., 2000). Patients with HD also perform simple voluntary movements slowly (Thompson et aI., 1988). This bradykinetic dysfunction is present from the early stages of the disease. Detailed EMG studies have helped to explain the defective motor activation responsible for movement slowness in HD. In patients performing wrist flexion movements the EMG recording shows an abnormal pattern with prolonged bursts from the agonist and antagonist muscles. The duration of EMG bursts is variable and the sequence of agonist-antagonist contraction is often absent. In addition, movements are more variable in velocity and accuracy (Thompson et aI., 1988). Patients are also slow in executing repetitive finger movements (Garnett et aI., 1984; Hefter et aI., 1987). Patients with HD are also bradykinetic in performing complex movements such as sequential hand squeezing-elbow flexing (Thompson et aI., 1988). In this task, patients are abnormally slow in executing each movement and in switching from one movement to the next. Both problems also affect their performance of sequential tracking arm movements. But unlike patients with Parkinson's disease, patients with HD do not have the accompanying progressive slowing with sequence completion (Agostino et aI., 1992). Patients with HD also have difficulty in adjusting movement execution to target size and movement length (Georgiou et aI., 1997). This applies also to free arm movements (Curra et aI.,2000). Motor performance also depends crucially on how movements are cued. In two studies, Bradshaw et al. (1992) and Georgiou et aI. (1995) showed that HD patients are too reliant on external cues to sequence motor programs effectively. Investigating the kinematic variables of rapid sequential free arm movements executed with different types of sub-movement cueing, Curra et aI. (2000) studied how the same motor sequence was performed when subjects

CHOREAS, ATHETOSIS, DYSKINESIAS, HEMIBALLISMUS

573

had to initiate movements in response to consecutive visual go signals (externally-triggered condition) or when they could start the movement at will (selfinitiated condition). HD patients performed both tasks slowly. Although both groups executed externally-triggered tasks faster than self-initiated tasks the time needed to perform the two tasks differed less in patients than in controls, suggesting that HD impairs internal more than external cueing mechanisms. Two further studies showed that auditory cues improved neither the performance of bi-manual co-ordination (Johnson et aI., 2000) nor gait (Churchyard et al., 2000). A recent study showed that patients with HD were slow across all cue types especially in the absence of external visual cues. They also derived no benefit from an external auditory timing cue in uni-manual upper limb movements (Johnson et aI., 2001, Ph.D. thesis). Studies using neuroimaging techniques to investigate brain activation during paced and triggered movements in patients with HD have helped to identify the sites of major impairment. During paced joystick arm movements in freely chosen directions patients had reduced activation of contralateral primary motor, medial premotor areas, bilateral parietal and bilateral prefrontal areas (Weeks et aI., 1997). During externally-triggered finger oppositions they showed impaired activity of the rostral supplementary motor area, anterior cingulate and premotor cortex, and increased activation of bilateral parietal areas (Bartenstein et aI., 1997). Bradykinesia in HD appears to be related to underactivity of primary and non-primary motor areas. Bradykinesia is one of the cardinal manifestations of disordered movement in HD. The slowness impairs simple and complex movements no matter how they are cued. Patients with HD therefore have a generalized movement execution deficit. HD disturbs external as well as internal cueing mechanisms. Like other movement disorders originating from basal ganglia abnormalities, however, it predominantly affects the mechanisms controlling the internal cueing of sub-movements (Berardelli et aI., 1999).

the studies confirmed that patients with HD have normal cortico-motoneuron connections and normal corticospinal conduction (Thompson et aI., 1986). Recordings of the cortical silent period (SP) and the responses to paired-pulse TMS make it possible to investigate the excitability of cortical motor areas (lnghilleri et aI., 1993; Kujirai et aI., 1993). In an early study we found that the duration of the SP is increased in patients with HD (Priori et aI., 1994). In addition, SP duration had a greater variance in patients than in controls, a finding later confirmed by other investigators (Tegenthoff et aI., 1996). We recently showed that abnormalities of the cortical SP are disclosed only by measuring the SP from unselected traces (Modugno et aI., 2001). Collecting consecutive trials therefore seems a useful method for detecting abnormalities in the duration of the SP that other methods of collection might leave undetected (Fig. 1). The prolonged SP could originate from the pathophysiological mechanisms underlying HD. According to widely accepted models of basal ganglia function (Wichmann and DeLong, 1996), the loss of striatal neurons in HD results in excessive thalamo-cortical facilitation and increased activity of cortical inhibitory interneurons (Priori et aI., 1994). An alternative hypothesis is based on the assumption that the restart of voluntary EMG activity after the EMG inhibition evoked by a magnetic stimulus depends on the time needed by cortico-cortical pathways to resume their drive over corticospinal cells after the cortical shock (Modugno et aI., 2001). In normal subjects engaged in reaction time tasks, the voluntary response can be delayed by delivering a magnetic stimulus over the sensorimotor cortex just before the movement (Hallett et aI., 1991; Berardelli et aI., 1994). The prolonged SPs in patients with HD could therefore indicate an excessive delay in restarting a motor task that has been interrupted. A delay in the process of restarting a voluntary movement after TMS also correlates with the finding of a delayed reaction time during simple and complex movements in HD (Thompson et aI., 1988; Curra et aI., 2000) and with evidence of reduced activation of primary, non-primary motor areas, and parietal cortex (Bartenstein et aI., 1997; Weeks et aI., 1997). One study conducted with the paired shock technique reported reduced cortico-cortical inhibition at short interstimulus intervals but an increased facilitation at longer intervals (Abbruzzese et aI.,

35.1.5.3. Stimulation of cortical motor areas Corticospinal function has been examined in HD using the technique of transcranial electrical and magnetic stimulation (TMS) of the motor cortex. All

574

A. BERARDELLI AND A.

A

cuRRA

B

Fig. 1. Raw data from ten consecutive unselected recordings of the cortical silent period in a patient with Huntington's disease (A) and a normal subject (B). Note the variable duration of the cortical silent period in the patient and the constant duration in the control. Horizontal calibration=200 ms, vertical calibration=500 11Y. 1997). Conversely, others have reported normal cortico-cortical inhibition and facilitation in patients with lID (Hanajima et al., 1996; Priori et al., 2000). Studying cortico-cortical inhibition at short and long interstimulus intervals at rest and during muscle contraction, Priori et al. (2000) found that patients had practically normal cortical motor excitability under all conditions.

35.2. Other choreic syndromes and athetosis In Sydenham's chorea the choreic movements are associated with synchronous EMG activity (Hallett and Kaufman, 1981). In chorea secondary to vascular lesions in the basal ganglia, EMG studies show rhythmic bursting as well as random bursts similar to those observed in tardive dyskinesia (Bathien et al., 1984). Athetoid movements show slowly changing tonic-type EMG bursts associated with long contractions, whereas chorea is characterized by shorter and more random burst patterns (Bathien et al., 1981, 1984; Yanagisawa and Nezu, 1987; Hashimoto et aI., 1994). In one patient with paroxysmal kinesiogenic choreoathetosis, the EMG recording showed synchronous bursts that had a stereotyped pattern of onset similar to that described in a Jacksonian march (Hayashi et al., 1997).

In contrast to lID, in other choreic syndromes neurophysiological findings are generally normal. In patients with symptomatic chorea the long-latency reflexes and the Hoffmann reflex have normal amplitude and duration (Deuschl et al., 1989). The major changes in the cortical SEP components observed in HD are not seen in Sydenham's chorea (Gledhill et al., 1990), and a cortical potential preceding a choreic movement can be recorded in patients with choreoacanthocytosis but not in those with HD (Shibasaki et al., 1982). Overall, these findings suggest that whereas the cortical activity needed to produce movement is abnormal in HD, in other choreas it is more preserved. Studies of rapid elbow flexion movements in patients with athetosis showed a triphasic EMG pattern with bursts of long duration, often with synchronous bursts in agonist and antagonist muscles. Movements were slow and unusually variable and the first agonist burst was prolonged (Hallett and Alvarez, 1983). These neurophysiological findings closely resemble those in patients with dystonia.

35.3. Levodopa-induced dyskinesias Together with motor fluctuations the occurrence of dyskinesias is one of the major problems in the

CHOREAS, ATHETOSIS, DYSKINESIAS, HEMIBALLISMUS

575

long-term management of patients with Parkinson's disease (PO). A community-based population study of PO patients concluded that dyskinesias are strongly related to the duration of levodopa treatment (Schrag and Quinn, 2000). However, studies performed in squirrel monkeys demonstrated that dyskinesias can also be induced by short periods of levodopa treatment and in the absence of nigrostriatal damage (Togasaki et al., 2001). Patients manifest an array of motor symptoms that also differ in their time of onset after levodopa intake (Marconi et al., 1994; Fahn, 2000). Some patients first experience these motor symptoms near the peak of clinical benefit (peak-dose dyskinesias), when levodopa or dopamine agonists reach brain concentrations sufficient to overactivate the dopamine receptors in the striatum, especially the putamen. Peak-dose dyskinesias can involve almost the whole body or a single region alone, usually that most prominently affected by the disease. Movements are predominantly choreiform, but they can also be dystonic, myoclonic or ballic. Levodopa-induced dyskinesias (LIDs) may also become manifest at the onset of the clinical effect, or at both the onset and offset (diphasic dyskinesia), i.e. when the brain dopamine concentration is rising or falling. In most affected persons they typically manifest as stereotyped movements predominantly affecting the legs. Off-period dystonia typically present as fixed dystonic posture and develop when the dopamine concentration in the brain is low. Few investigations have been performed on the clinical physiology of LIDs (Hallett, 2000) and most of the neurophysiological data is based on intraoperative studies in patients undergoing functional neurosurgery for advanced PD. In PD an abnormal output from the basal ganglia generates error signals. Recording from single GPi cells showed hyperactivity in patients off-medication and reduced rates after dopaminergic activation by apomorphine injection. Conversely, GPe cells showed low firing rates in the off-medication period, and increased rates when patients were on-medication (Hutchison et al., 1994, 1997; Beric et aI., 1996; Stefani et al., 1997; Merello et al., 1999). It was initiaIly suggested that dyskinesias were due to reduced firing of GPi neurons resulting in the loss of its inhibitory function on the thalamus and consequently in the increase of thalamo-cortical drive (Wichmann and Delong, 1996). The reduction in the GPi firing rate and the

concurrent increased firing rate in GPe time-locked with apomorphine-induced dyskinesias suggests that the indirect pathway might have a predominant role in the development of LIDs. Lesions in the GPe do not alleviate LIDs (Blanchet et al., 1994), however, suggesting that changes in the indirect pathway alone are not enough to generate LIDs. A recent neurophysiological study suggested that stimulation of the ventral pallidum controls dyskinesias by activating the large GPe axons responsible for inhibiting GPi neurons (Wu et al., 2001). Not only abnormal firing rates but also changes in the quality of signaling - i.e. the pattern, synchronization, and somatosensory responsiveness - are responsible for the emergence of dyskinesias. Changes in the quality of signaling in the GPi neurons provoke corresponding changes in the activity of the thalamus that interfere with thalamocortical signal transmission. Disruption of the normal spatio-temporal pattern of cortical neuronal activity leads to disordered cortical output and altered motor control (Obeso et al., 2000). This may explain why the removal of this abnormal activity by thalamotomy abates LIDs. But it fails to explain why LIDs respond also to pallidotomy - a procedure expected to reduce inhibitory control over the thalamo-cortical pathway even further. PalIidotomy presumably either directly or indirectly - removes the abnormal pattern of thalamic neuronal activity. Microrecordings performed during interventions for STN chronic stimulation provided other information on the pathophysiology of LIDs. In patients with LIDs, most STN neurons have a high mean firing rate and an irregular firing pattern (Hutchison et al., 1998). Krack et aI. (1999) correlated the firing rate and pattern of STN neurons with different types of dyskinesias. Off-period dystonia was associated with cellular hyperactivity. Diphasic dystonia reflected an altered pattern of STN activity with alternating periods of hyperactivity and decreased activity in cell responsible for specific dyskinetic muscles. Peak-dose dyskinesias were accompanied by STN hypoactivity (Krack et al., 1999). Chronic high-frequency stimulation of the STN abates LIDs through different mechanisms. In the short term, by altering activity in the STN it reduces the overactivity in the GPi and the hypoactivity in the GPe, thus stabilizing basal ganglia output activity. In the long term, chronic STN stimulation ameliorates LIDs by lowering the daily intake of

576 levodopa and dopamine agonists (Fraix et aI., 2000).

A. BERARDELLI AND A. CURM

levels are necessary to reverse these abnormalities, because they disappear during peak-dose dyskinesias, but do not disappear with optimal treatment.

35.3.1. Electromyographicfeatures ofLiDs On surface EMG studies, LIDs typically consist of synchronous or asynchronous irregular antagonist muscle bursts that last up to 500-800 ms (Yanagisawa and Nezu, 1987). This feature clearly differentiates choreic movements in LIDs from choreic movements in HD, which cause shortduration EMG bursts (about 30 ms). The most typical LID pattern is that of chorea associated with synchronous EMG activity, a pattern closely resembling that observed in Sydenham's chorea (Hallett and Kaufman, 1981). When LIDs manifest as repetitive alternating movements the EMG pattern consists of asynchronous, alternating, and pseudo-rhythmic bursts at about 1 Hz (Luquin et al., 1992). An early investigation by Yanagisawa and Nezu (1987) showed that in patients with advanced PD the dyskinesias can be detected in one arm while tremor is detected in the other and - even more remarkably - these two abnormal movements can alternate second-by-second in the same limb.

35.3.2. Blinking and the blink reflex in patients with LiDs Blinking is a function subserved by brainstem circuitry sensitive to changes in brain dopamine concentrations. Spontaneous blinking is reduced in patients with PD and increases with dopamine replacement (Karson et al., 1983; Agostino et aI., 1987; Deuschl and Goddemeier, 1998). In patients with LIDs, the blink rate is faster than in those who are on but without dyskinesias (Karson et aI., 1983). Excitability of the brainstem circuitry is commonly studied by the blink reflex recovery curve. In parkinsonian patients the inhibition of the R2 component is reduced during the OFF state and increases during the ON state (Agostino et aI., 1987). R2 inhibition is more intense in dyskinetic than in non-dyskinetic patients (Iriarte et aI., 1989). In conclusion, in off-therapy parkinsonian patients the deficiency of dopamine leads to decreased spontaneous blinking but increased blink reflex excitability. Excessively high dopamine plasma

35.3.3. Other neurophysiological investigations of LiDs Involuntary movements and tardive oro-facial dyskinesias have been recorded with Doppler ultrasound detectors (McCleland et al., 1987). This method failed to show relationships between whole body ultrasound measures and Abnormal Involuntary Movement Scale (AIMS) ratings (Bartzokis et aI., 1989). A better correlation with several, but not all, subjective rating scales for dyskinesia was found by studying involuntary movements in tardive dyskinesias with other and more complex methods. A balloon with a pneumatic transducer placed in the mouth measured buccal and lingual movements in tardive dyskinesia (Chien et aI., 1977), and a pencilsized pressure transducer between the third and fourth fingers recorded hand movements (Denney and Casey, 1975). More recently another approach was used in the attempt to develop an objective ambulatory measurement of LIDs. Patients with advanced PD underwent a standardized protocol consisting of l-min simultaneous multi-channel accelerometry and videorecordings during rest, talking, stress, and four activities of daily life (Hoff et aI., 2001). The severity of LIDs was assessed also with a modified AIMS to determine convergent validity, reproducibility, and responsiveness between the objective measures and the subjective score. Although objective measures of LIDs proved reliable and responsive, they failed to distinguish LIDs from voluntary movements, indicating that they are useful for examining LIDs in resting patients or during motor activities of daily life by deriving signals from body segments not normally involved, i.e. trunk and legs. No physiological study has yet investigated changes in cortical excitability in patients with LIDs. One reason is that the rapidly changing motor activity that makes it very difficult to fix a relatively stable baseline, either at rest or during constant motor activation. In these patients, the changes in cortical function are presumably time-locked with the changes in their motor state. Hence they cannot be identified after dyskinesias have been abated.

577

CHOREAS, ATHETOSIS, DYSKlNESIAS, HEMIBALLISMUS

35.4. Ballism Ballism affects proximal muscles of the arms and legs resulting in uncontrolled, violent, flinging or throwing actions. Ballism often occurs in association with athetosis, chorea, and dystonia. In the general population this condition is very rare (Shannon et aI., 1998), whereas in the pre-levodopa era of stereotactic surgery its occurrence in the clinical series varied from 0.2 to 10% (Guridi and Obeso, 2001). The first to propose a relation between hemiballism (ballism restricted to one side of the body) and the subthalamic nucleus was Purdon Martin, in the early decades of the last century (Martin, 1927). Acute ballism can develop - but does not always do so - in patients who have suffered small vascular lesions of the STN. Even though the lesion persists, most patients spontaneously recover after weeks or months. Understanding the mechanisms of this recovery might shed light on how dyskinesias develop and how to treat them. One possibility is that patients who recover have lesions that cause only limited damage to the critical areas of motor circuitry. Another is that the lesion not only involves the STN but also affects an antidyskinetic pathway. Among animals lesioned in the STN those in whom no hemiballism developed had larger lesions that involved portions of the GPi or the pallidofugal projections, or both structures (Carpenter and Strominger, 1966). In a series of patients who underwent STN lesioning for PD, none of those who received a correctly placed lesion suffered intractable hemiballism (Alvarez et aI., 2001; Guridi and Obeso, 2001). Because the classic model of basal ganglia functioning predicts that after an STN lesion the GPi/SNpr reduce their firing rate thereby releasing the thalamus to produce dyskinesias, patients who had a good outcome the lesion may have involved the white matter around the STN, whereas in patients in whom hemiballism developed the lesion was located within the nucleus. One of these patients suffered remarkably severe lesion-induced dyskinesias that further pallidotomy abolished (Alvarez et aI., 2001; Guridi and Obeso, 2001). Even more remarkably this indicates the GPi as a key structure for the expression of hemiballism. Micro-recordings in one patient with hemiballism showed that GPi cells have low discharge rates and irregular bursts with intermittent pauses, a pattern similar to that observed in patients with dystonia or

PD on-medication. But low-frequency modulation and pauses in the spike trains were more pronounced in the patient with hemiballism than in patients with dystonia, and pauses were more frequent than in patients with PD, either on- or off-medication. Moreover - unlike patients with PD - in hemiballism GPe and GPi neurons have similar mean rates and patterns of discharge, and especially in the GPe the firing rates closely resembled those observed in the dystonic patients at rest (Suarez et al., 1997; Vitek et aI., 1999). Finally in hemiballism, the somatosensory responsiveness of neurons in the GPi is reduced, whereas in PD it is increased (Vitek et aI., 1999). Changes in the degree of synchronization of GPi neurons also contribute to the development of hemiballism, as suggested by the strong correlation of neuronal to electromyographic (EMG) activity in both flexor and extensor muscles of the ballistic limb in a patient with hemiballism (Vitek et al., 1999). Among the very rare EMG recordings of ballism is the case of a young woman with acute demyelinating polyneuropathy in whom involuntary flinging movements affecting the face and the four limbs developed. Surface EMG of the involuntary movements showed 1.5-2 Hz rhythmic grouping discharges (Odaka et aI., 1999). 35.5. Conclusions Chorea, athetosis, ballism, and levodopa-induced dyskinesias are a variety of different movement disorders released involuntarily by the motor cortex due to a dysfunction of the basal ganglia circuitry. They result from a complex mechanism that combines changes in signaling features - i.e. the pattern, synchronization, discharge rates and somatosensory responsiveness - of neurons in the direct and indirect pathways of the pallido-thalamo-cortical circuit. But in patients with movement disorders the aforementioned changes have been demonstrated not only in the pallidal (VLo), but also in the cerebellar receiving area, i.e. the ventralis posterior lateralis pars oralis (VPLo) of the motor thalamus, an area that is not directly connected with the GPi. This location/anatomical distinction suggests that additional pathways contribute to the changes in neuronal activity that occur in the VPLo, for example the pathway from the GPe to the reticularis

578

A. BERARDELLI AND A. CURM

MOTOR CIRCUIT IN PD WITH LEVOPODA·INDUCED DYSKINESIA CURRENT MODEL

PROPOSED MODEL

CORTEX

•

PPN

Inhibitory

~

Excitatory

•

Altered firing pattern

Fig. 2. Proposed model of basal ganglia functioning in LIDs in Parkinson's disease (Obeso et al., 2000 with permission). A revision of the classical model that - in addition to the previously suggested disinhibition of the thalamo-cortical projections - considers also: (1) abnormal pattern of discharges through the GPe-STN-GPi circuit that is translated to the cortex; (2) a relative hyperactivity of GPe-STN; (3) increased activity of the cortico-striatal connections. GPi: Globus Pallidus pars interna; GPe: Globus Pallidus pars externa; SNr: substantia nigra pars reticularis; SNc substantia nigra pars compacta; PPN: pedunculopontine nucleus; VL: ventralis lateralis; CM: centro-medianum; Pf: parafascisularis nucleus; DA: dopamine. nucleus of the thalamus (Hazrati and Parent, 1991), and the pathway from the pedunculopontine nucleus to the motor thalamus (Steriade et aI., 1988). The recent models of hyperkinetic disorders incorporating changes in pattern and increased synchronization of neuronal activity in the basal ganglia and thalamus postulate that a striatal underactivity in the indirect pathway plays a significant role in the development of LID, whereas a loss of excitatory input from the STN nucleus to the GPi is postulated to underlie the development of hemiballism. The connections to and from the pedunculopontine nucleus and the midbrain extrapyramidal area, and from the GPe to the reticularis nucleus of the thalamus, also contribute to the changes in neuronal activity that produce these hyperkinetic movement disorders (Fig. 2). In conclusion, the relative changes in neuronal activity in these pathways, together with the relative changes in the signal

features of neurons in each of these pathways determines the phenotypic expression of the involuntary movement.

References Abbruzzese, G, Dall'Agata, D, Morena, M, Reni, Land Favale, E (1990) Abnormalities of parietal and prerolandic somatosensory evoked potentials in Huntington's disease. Electroencephalogr. Clin. Neurophysiol., 77: 340-346. Abbruzzese, G, Buccolieri,A, Marchese, R, Trompetto, C, Mandich, P and Schieppati, M (1997) Intracortical inhibition and facilitation are abnormal in Huntington's disease: a paired magnetic stimulation study. Neurosci. Lett., 228: 87-90. Agostino, R, Berardelli, A, Cruccu, G, Stocchi, F and Manfredi, M (1987) Corneal and blink reflexes in Parkinson's disease with "on-off' fluctuations. Mov. Disord., 2: 227-235.

CHOREAS, ATHETOSIS, DYSKINESIAS, HEMIBALLISMUS

579

Agostino, R, Berardelli, A, Cruccu, G, Pauletti, G, Stocchi, F and Manfredi, M (1988) Correlation between facial involuntary movements and abnormalities of blink and corneal reflexes in Huntington's chorea. Mov. Disord., 3: 281-289. Agostino, R, Berardelli, A, Formica, A, Accornero, Nand Manfredi, M (1992) Sequential arm movements in patients with Parkinson's disease, Huntington's disease and dystonia. Brain, 115: 1481-1495. Albin, RL, Reiner, A, Anderson, KD, Dure, LS, Handelin, B, Balfour, R et al. (1992) Preferential loss of striatoexternal pallidal projection neurons in presymptomatic Huntington's disease. Ann. Neurol., 31: 425-430. Alvarez, L, Macias, R, Guridi, J, Lopez, G, Alvarez, E, Maragoto, C, Teijeiro, J, Torres, A, Pavon, N, Rodriguez-Oraz, MC, Ochoa, L, Hetherington, H, Juncos, J, DeLong, MR and Obeso, JA (2001) Dorsal subthalamotomy for Parkinson's disease. Mov. Disord., 16: 72-78. Bartenstein, P, Weindl, A, Spiegel, S, Boecker, H, Wenzel, R, Ceballos-Baumann, AO, Minoshima, S and Conrad, B (1997) Central motor processing in Huntington's disease. A PET study. Brain, 120: 1553-1567. Bartzokis, G, Wirshing, WC, Hill, MA, Cummings, JL, Altshuler, L and May, PR (1989) Comparison of electromechanical measures and observer ratings of tardive dyskinesia. Psychiatry Res., 27: 193-198. Bathien, N, Rondot, P and Koutlidis, RM (1981) Electrophysiological and pharmacological analysis of L-DOPA-induced dyskinesia and tardive dyskinesia. J. Physiol. (Paris), 77 : 131-141. Bathien, N, Koutlidis, RM and Rondot, P (1984) EMG patterns in abnormal involuntary movements induced by neuroleptics. J. Neurol. Neurosurg. Psychiatry, 47: 1002-1008. Berardelli, A, Inghilleri, M, Polidori, L, Priori, A, Mercuri, B and Manfredi, M (1994) Effects of transcranial magnetic stimulation on single and sequential arm movements. Exp. Brain Res., 98: 501-506. Berardelli, A, Noth, J, Thompson, PD, Bollen, EL, Curta, A, Deuschl, G, Van Dijk, JG, Topper, R, Schwarz, M and Roos, RA (1999) Pathophysiology of chorea and bradykinesia in Huntington's disease. Mov. Disord., 14: 398-403. Beric, A, Sterio, D, Dogali, M, Fazzini, E, Eidelberg, D and Kolodny, E (1996) Characteristics of pallidal neuronal discharges in Parkinson's disease patients. Adv. Neurol., 69: 123-128. Blanchet, PJ, Boucher, R and Bedard, PJ (1994) Excitotoxic lateral pallidotomy does not relieve LDOPA-induced dyskinesia in MPTP parkinsonian monkeys. Brain Res., 650: 32-39. Bollen, E, Arts, RJHM, Van der Velde, EA and Buruma, OJS (1986) Brainstem reflexes and brainstem auditory

evoked responses in Huntington's chorea. 1. Neurol. Neurosurg. Psychiatry, 49: 313-315. Bradshaw, JL, Phillips, JG, Dennis, C, Mattingley, JB, Andrewes, D, Chiu, E, Pierson, JM and Bradshaw, JA (1992) Initiation and execution of movement sequences in those suffering from and at-risk of developing Huntington's disease. J. Clin. Exp. Neuropsychol., 14: 179-192. Caraceni, T, Avanzini, G, Spreafico, R, Negri, S, Broggi, G and Girotti, F (1976) Study of the excitability of the blink reflex in Huntington's chorea. Eur. Neurol., 14: 465-472. Carpenter, ME and Strominger, NL (1966) Corticostriate encephalitis and paraballism in the monkey. Evidence concerning distinctive functions of the globus pallidus. Arch. Neurol., 14: 241-253. Chien, CP, Jung, K, Ross-Townsend, A and Stearns, B (1977) The measurement of persistent dyskinesia by piezoelectric recording and clinical rating scales. Psychophannacol. Bull., 13: 34-36. Churchyard, AJ, Morris, ME, Georgiou, N, Chiu, E, Cooper, R and Iansek, R (2001) Gait dysfunction in Huntington's disease: parkinsonism and a disorder of timing. Implications for movement rehabilitation. Adv. Neurol., 87: 375-385. Curra, A, Agostino, R, Galizia, P, Fittipaldi, F, Manfredi, M and Berardelli, A (2000) Sub-movement cueing and motor sequence execution in patients with Huntington's disease. Clin. Neurophysiol., 111: 1184-1190. Denney, D and Casey, D (1975) An objective method for measuring dyskinetic movements in tardive dyskinesia. Electroencephalogr: CUn. Neurophysiol., 38: 645-646. Deuschl, G and Goddemeier, C (1998) Spontaneous and reflex activity of facial muscles in dystonia, Parkinson's disease, and in normal subjects. 1. Neurol. Neurosurg. Psychiatry, 64: 320-324. Deuschl, G, Liicking, CH and Schenck, E (1989) Hand muscle reflexes following electrical stimulation in choreatic movement disorders. J. Neurol. Neurosurg. Psychiatry, 52: 755-762. Ehle, AL, Stewart, RM, Lellelid, NA and Leventhal, NA (1984) Evoked potentials in Huntington's disease. A comparative and longitudinal study. Arch. Neurol., 41: 379-382. Fahn, S (2000) The spectrum of Levodopa-Induced Dyskinesias. Ann. Neurol., 47: (Suppl. 1): S2-S11. Fraix, V, Pollak, P, Van Blercom, N, Xie, J, Krack, P, Koudsie, A and Benabid, AL (2000) Effect of subthalamic nucleus stimulation on levodopa-induced dyskinesia in Parkinson's disease. Neurology, 55: 1921-1923. Garnett, ES, Firnau, G, Nahmias, C, Carbotte, Rand Bartolucci, G (1984) Reduced striatal glucose consumption and prolonged reaction time are early

580 features in Huntington's disease. J. Neural. Sci., 65: 231-237. Georgiou, N, Bradshaw, JL, Philips, JG, Chiou, E and Bradshaw, JA (1995) Reliance on advance information and movement sequencing in Huntington's disease. Mov. Disord., 10: 477-481. Georgiou, N, Phillips, JG, Bradshaw, JL, Cunnington, R and Chiu, E (1997) Impairments of movement kinematics in patients with Huntington's disease: a comparison with and without a concurrent task. Mov: Disord., 12: 386-396. Ginovart, N, Lundin, A, Farde, L, Hallidin, C, Backman, L, Swahn, C-G et al. (1997) PET study of the pre- and post-synaptic dopaminergic markers for the neurodegenerative process in Huntington's disease. Brain, 120: 503-514. Girotti, F, Marano, R, Soliveri , P, Geminiani, G and Scigliano, G (1988) Relationship between motor and cognitive disorders in Huntington's disease. 1. Neural., 235: 454-457. Gledhill, RF and Thompson, PD (1990) Standard neurodiagnostic tests in Sydenham's chorea. J. Neural. Neurosurg. Psychiatry, 53: 534-535. Guridi, J and Obeso, JA (2001) The subthalamic nucleus, hemiballism and Parkinson's disease: reappraisal of a neurosurgical dogma. Brain, 124: 5-19. Hallett, M (2000) Clinical Physiology of DOPA-induced dyskinesia. Ann. Neurol., 47: SI47-S153. Hallett, M and Alvarez, N (1983) Attempted rapid elbow movements in patients with athetosis. J. Neural. Neurosurg. Psychiatry, 46: 1273-1279. Hallett, M and Kaufman, C (1981) Physiological observations in Sydenham's chorea. J. Neural. Neurosurg. Psychiatry, 44: 829-832. Hallett, M, Cohen, LG and Bierner, SM (1991) Studies of sensory and motor cortex physiology: with observations on akinesia in Parkinson's disease. Electraencephalogn Clin. Neurophysiol., 43 (Supp!.): 76-85. Hamada, I (1998) Direct and indirect pathways: why do pallidotomy and subthalamic stimulation work? In: Proceedings International Basal Ganglia Society (6th Triennial Meeting), Brewster, MA. Hanajima, R, Hugawa, Y, Terao, Y, Ugata, K and Kanazawa, I (1996) Ipsilateral cortico-cortical inhibition of the motor cortex in various neurological disorders. 1. Neurol. Sci., 140: 109-116. Hashimoto, T and Yanagisawa, N (1994) A comparison of the regularity of involuntary muscle contractions in vascular chorea with that in Huntington's chorea, hemiballism and parkinsonian tremor. J. Neural. Sci., 125: 87-94. Hayashi, R, Hanyu, N, Yahikozawa, Hand Yanagisawa, N (1997) Ictal muscle discharge pattern and SPECT in

A. BERARDELLI AND A.

cuRRA

paroxysmal kinesigenic choreoathetosis. Electromyogr. Clin. Neurophysiol., 37, 89-94. Hazrati, L-N and Parent, A (1991) Projection from the external pallidum to the reticular thalamic nucleus in the squirrel monkey. Brain, 550: 142-146. Hedreen, JC and Folstein, SE (1995) Early loss of neostriatal striosome neurons in Huntington's disease. J. Neuropathol. Exp. Neurol., 54 : 105-120. Hefter, H, Homberg, V, Lange, HW and Freund, HJ (1987) Impairment of rapid movement in Huntington's disease. Brain, 110: 585-612. Hoff, JI, Van den Plas, AA, Wagemans, EA and Van Hilten, 11 (2001) Accelerometric assessment of levodopa-induced dyskinesias in Parkinson's disease. Mov. Disord., 16: 58-61. Hutchison, WD, Lozano, AM, Davis, KD, Saint-Cyr, JA, Lang, AE and Dostrovsky, 10 (1994) Differential neuronal activity in segments of filobus pallidus in Parkinson's disease patients. NeuroReport, 5: 15331537. Hutchison, WD, Levy, R, Dostrovsky, JO, Lozano, AM and Lang, AE (1997) Effects of apomorphine on globus pallidus neurons in parkinsonian patients. Ann. Neurol., 42: 767-775. Hutchison, WD, Allan, RJ, Opitz, H, Levy, R, Dostrovsky, 10, Lang, AE and Lozano, AM (1998) Neurophysiological identification of the subthalamic nucleus in surgery for Parkinson's disease. Ann. Neurol., 44: 622-628. Inghilleri, M, Berardelli, A, Cruccu, G and Manfredi, M (1993) Silent period evoked by transcranial magnetic stimulation of the human motor cortex and cervicomedullary junction. J. Physiol., 466: 521-534. Iriarte, LM, Chacon, J, Madrazo, J. Chaparro, P and Vadillo, J (1989) Blink reflex in dyskinetic and nondyskinetic patients with Parkinson's disease. Eur. Neurol., 29: 67-70. Jahanshahi, M, Brown, RG and Marsden, CD (1993) A comparative study of simple and choice reaction time in Parkinson's, Huntington's and cerebellar disease. J. Neural. Neurosurg. Psychiatry, 56: 1169-1177. Johnson KA (2001) Movement preparation and execution in Huntington's disease and Parkinson's disease. Ph.D. Thesis. Monash University, Clayton, Australia. Johnson, KA, Bennett, JE, Georgiou, N, Bradshaw, JL, Chiu, E, Cunnington, R et a!. (2000) Bimanual coordination in Huntington's disease. Exp. Brain Res., 134: 483-489. Johnson, KA, Cunnington, R, Iansek, R, Bradshaw, JL, Georgiou, Nand Chiu, E (2001) Movement-related potentials in Huntington's disease: movement preparation and execution. Exp. Brain Res.,138: 492-499. Kaneoke, Y, Vitek, JL, Ashe, MR et a!. (1993) Altered somatosensory response properties of neurons in the

CHOREAS, ATHETOSIS, DYSKINESIAS, HEMIBALLISMUS

581

cerebellar receiving area of the motor thalamus in MPTP treated parkinsonian monkeys. Soc. Neurosci., 19: 1585 (Abstract). Karson, CN (1983) Spontaneous eye-blink rates and dopaminergic systems. Brain, 106: 643-653. Krack, P, Pollak, P, Limousin, P, Benazzouz, A, Deuschl, G and Benabid, AL (1999) From off-period dystonia to chorea: the clinical spectrum of varying subthalamic nucleus activity. Brain, 122: 1133-1146. Kujirai, T, Cararnia, MD, Rothwell, JC, Day, BL, Thompson, PD, Ferbert, A, Wroe, S, Asselmann, P and Marsden, CD (1993) Corticocortical inhibition in human motor cortex. J. Physiol., 471: 501-519. Luquin, MR, Scipioni, 0, Vaamonde, J et al. (1992) Levodopa-induced dyskinesias in Parkinson's disease: clinical and pharmacological classification. Mov. Disord., 7: 117-124. Marconi, R, Lefebvre-Caparros, D, Bonnet, AM, Vidailhet, M, Dubois, B and Agid, Y (1994) Levodopa induced dyskinesias in Parkinson's disease phenomenology and pathophysiology. Mov. Disord., 9: 2-12. Martin, JP (1927) Hemichorea resulting from a local lesion of the brain (the syndrome of the body of Luys). Brain, 50: 637-651. McCleland, HA, Fairbairn, AF, McDonald, M, Nicholas, TRM and Cox, GA (1987) The evaluation of an ultrasound detector (UD) in the measurement of oro-facial dyskinesia. Int. Clin. Psychopharmacol., 2: 159-164. Merello, M, Lees, AJ, Balej, J, Cammarota, A and Leiguarda, R (1999) GPi firing rate modification during beginning-of-dose motor deterioration following acute administration of apomorphine. Mov. Disord., 14: 481-483. Modugno, N, Curta, A, Giovannelli, M, Priori, A, Squitieri, F, Ruggieri, S, Manfredi, M and Berardelli, A (2001) The prolonged cortical silent period in patients with Huntington's disease. Clin. Neurophysiol., 112: 1470-1474. Noth, J, Friedemann, HH, Podoll, K and Lange, HW (1983) Absence of long latency reflexes to imposed finger displacements in patients with Huntington's disease. Neurosci. Lett., 35: 97-100. Noth, J, Engel, L, Friedemann, HH and Lange, HW (1984) Evoked potentials in patients with Huntington's disease and their offspring. 1. Somatosensory evoked potentials. Electroencephalogr. Clin. Neurophysiol., 32: 134-141. Noth, J, Podoll, K and Friedemann, HH (1985) Long loop reflexes in small hand muscles studied in normal subjects and in patients with Huntington's disease. Brain, 108 : 65-80. Obeso, JA, Rodriguez-Oroz, MC, Rodriguez, M, DeLong, MR and Olanow, CW (2000) Pathophysiology of

levodopa-induced dyskinesias in Parkinson's disease: problems with the current model. Ann. Neurol., 47: S22-32. Odaka, M, Yuki, N and Hirata, K (1999) Bilateral ballism in a patient with overlapping Fisher's and GuillainBarre syndromes. J. Neurol. Neurosurg. Psychiatry, 67: 206-208. Priori, A, Berardelli, A, Inghilleri, M, Polidori, L and Manfredi, M (1994) Electromyographic silent period after transcranial brain stimulation in Huntington's disease. Mov. Disord., 9: 178-182. Priori, A, Polidori, L, Rona, S, Manfredi, M and Berardelli, A (2000) Spinal and cortical inhibition in Huntington's disease. Mov. Disord., 15: 938-946. Reiner, A, Albin, RL, Anderson, KD, D' Amato, CJ, Penney, JB and Young, AB (1988) Differential loss of striatal projection neurons in Huntington's disease. Proc. Natl. Acad. Sci., 85: 5733-5737. Schrag, A and Quinn, N (2000) Dyskinesias and motor fluctuations in Parkinson's disease. A Communitybased study. Brain, 123: 2297-2305. Shannon, KR (1998) Ballism. In: J Jankovic and E Tolosa (Eds.), Parkinson's Disease and Movement Disorders (3rd ed.). Baltimore, Williams and Wilkins, pp. 365375. Shibasaki, H, Sakai, T, Nishimura, H, Sato, Y, Goto, I and Kuroiwa, Y (1982) Involuntary movements in ChoreaAcanthocytosis: a comparison with Huntington's chorea. Ann. Neurol., 12: 311-314. Stefani, A, Stanzione, P, Bassi, A, Mazzone, P, Vangelista, T and Bernardi, G (1997) Effects of increasing doses of apomorphine during stereotaxic neurosurgery in Parkinson's disease: clinical score and internal globus pallidus activity. 1. Neural. Trans., 104: 895-904. Steriade, M, Pare, D, Parent, A and Smith, Y (1988) Projections of cholinergic and non-cholinergic neurons of the brainstem core to relay and associational thalamic nuclei in the cat and macaque monkey. Neuroscience, 25: 47-67. Suarez, Jl, Verhagen Metman, L, Reich, SG, Dougherty, PM, Hallett, M and Lenz, FA (1997) Pallidotomy for hemiballism: efficacy and characteristics of neuronal activity. Ann. Neurol., 42: 807-811. Tegenthoff, M, Vorgerd, M, Juskowiak, F, Roos, V and Malin, JP (1996) Postexcitatory inhibition and double train stimulation in Huntington's disease. Electroenceph. Clin. Neurophysiol., 101: 298-303. Thompson, PD, Dick, JP, Day, BL, Rothwell, JC, Berardelli, A, Kachi, T and Marsden, CD (1986) Electrophysiology of the corticomotoneuron pathways in patients with movement disorders. Mov. Disord., 1: 113-117. Thompson, PD, Berardelli, A, Rothwell, JC, Day, BL, Dick, JP, Benecke, R and Marsden, CD (1988) The

582 coexistence of bradykinesia and chorea in Huntington's disease and its implications for theories of basal ganglia control of movement. Brain, Ill: 223-244. Thompson, PD, Bhatia, KP, Brown, P, Davis, MB, Pires, M, Quinn, NP, Luthert, P, Honovar, M, O'Brien, MD and Marsden, CD (1994) Cortical myoclonus in Huntington's disease. Mov. Disord., 9: 633-641. Togasaki, DM, Tan, L, Protell, P, Di Monte, DA, Quik, M and Langston, JW (2001) Levodopa induces dyskinesias in normal squirrel monkeys. Ann. Neurol., 50: 254-257. Topper, R, Schwarz, M, Podoll, K, Domges, F and Noth, J (1993) Absence of frontal somatosensory evoked potentials in Huntington's disease. Brain, 116: 87101. Turjanski, N, Weeks, R, Dolan, R, Harding, AE and Brooks, DJ (1995) Striatal D 1 and D2 receptor binding in patients with Huntington's disease and other choreas: a PET study. Brain, 118: 689-696.

A. BERARDELLI AND A. CURRA

Vitek, JL, Chockkan, V, Zhang, JY, Kaneoke, Y, Evatt, M, DeLong, MR, Triche, S, Mewes, K, Hashimoto, T and Bakay, RA (1999) Neuronal activity in the basal ganglia in patients with generalized dystonia and hemiballismus. Ann. Neurol., 46: 22-35. Weeks, RA, Ceballos-Baumann, A, Piccini, P, Boecker, H, Harding, AE and Brooks, DJ (1997) Cortical control of movement in Huntington's disease. A PET activation study. Brain, 120: 1569-1578. Wichmann, T and DeLong, MR (1996) Functional and pathophysiological models of the basal ganglia. Curro Opin. Neurobiol., 6: 751-758. Wu, YR, Levy, R, Ashby, P, Tasker, RR and Dostrovsky, JO (2001) Does stimulation of the GPi control dyskinesia by activating inhibitory axons? Mov. Disord., 16: 208-216. Yanagisawa, Nand Nezu, A (1987) Pathophysiology of involuntary movements in Parkinson's disease. Eur. Neurol., 26: 30-40.

Movement Disorders Handbook of Clinical Neurophysiology, Vol. I M. Hallett (Ed.) © 2003 Elsevier B.Y. All rights reserved

583 CHAPTER 36

Restless legs syndrome and periodic limb movements Wayne Hening* Johns Hopkins Center for Restless Legs Syndrome. 5th Floor, Room 5B71C. Asthma & Allergy Center, 5501 Hopkins Bayview Circle. Baltimore. MD 21224. USA

36.1. Introduction Restless legs syndrome (RLS) and periodic limb movements (PLM) (for recent reviews, see Hening et al., 1999b; Montplaisir et al., 2000; Allen and Earley, 2001; Hening, 2002; Chokroverty et al., 2003) have been linked since the the 1960s when Lugaresi and his colleagues discovered that most patients with RLS suffer from periodic limb movements in sleep (PLMS) (Lugaresi, 1965, 1986). While highly associated, however, they do not appear to be merely alternate forms of one central neural process. Either can exist in a person without the other. Evidence suggests that RLS is only the most common condition associated with PLM (Hening et al., 1999b). 36.1.1. Definition and history

As recently defined by a consensus conference at the NIH held in conjunction with the International RLS Study group (IRLSSG), RLS is a clinical condition in which cognitively intact adult patients complain of four key symptoms (Table 1) (Allen et al., 2003). In brief, these are: (1) an urge to move, associated with abnormal sensation; (2) induction or aggravation by rest; (3) relief with activity; and (4) intensification at later phases of the normal day/night cycle (evening and night). While these symptoms are necessary to define RLS, there are mimics such as leg cramps or some polyneuropathic discomforts that can also satisfy these criteria and must be separately distinguished from RLS. RLS may have been first described by Thomas Willis in 1672 (Willis, 1672). After scattered case

* Correspondence to: Wayne Hening, M.D., Ph.D., 321 Avenue C (I-F), New York, NY 10009-1643, USA. E-mail address:[email protected] Tel.: 212-477-7527; fax: 410-550-8078.

reports and small case series in ensuing centuries, the condition was given its name and described at length by Karl Ekbom (Ekbom, 1945, 1950, 1960). The only major feature of the condition that he did not appreciate was the association with PLM, established by Lugaresi in the 1960s (Lugaresi et al., 1965, 1986). In recent years, additional studies have provided more information about the range of the clinical condition, its epidemiology, an initial genetic linkage, an array of therapeutic options, and some tantalizing but indefinite clues as to pathophysiology (Hening, 2002). These studies have indicated that there is a true circadian fluctuation of symptoms (Hening et al., 1999c; Trenkwalder et al., 1999a), that the arms as well as the legs can be involved (Michaud et al., 2000) (and less commonly other body parts), that the disease is a chronic condition with both slowly progressive idiopathic and more rapidly developing secondary forms (Allen and Earley, 2000), particularly associated with iron deficiency (O'Keeffe et al., 1994; Sun et al., 1998), uremia (Winkelman et al., 1996; Hui et al., 2002), and pregnancy (Goodman et al., 1988; Lee et al., 2001). Epidemiological studies have shown that, at least in populations derived from the North and West of Europe, symptoms of RLS are quite common (Lavigne and Montplaisir, 1994; Phillips et al., 2000; Rothdach et al., 2000; Ulfberg et al., 2001a, b) and that there is likely family aggregation (Winkelmann et al., 2000). Segregation analyses of families suggest that there may be a variety of means of familial aggregation, but that those families in which individuals experience an early onset (mean onset before age 30) show a pattern of familial distribution most consistent with a dominant inheritance (Winkelmann et al., 2002). Two linkages have been reported, one uses a recessive model of inheritance (Desautels et al., 2001) and the other, a dominant model (Bonati et al., 2003), but it seems likely that

584

W. HENING

Table 1 Clinical features of the restless legs syndrome. Diagnostic features I. 2. 3. 4.

An urge to move the legs, usually accompanied or caused by uncomfortable and unpleasant sensations in the legs. The urge to move or unpleasant sensations begin or worsen during periods of rest or inactivity such as lying or sitting. The urge to move or unpleasant sensations are partially or totally relieved by movement, such as walking or stretching. The urge to move or unpleasant sensations are worse in the evening or night than during the day or only occur in the evening or night.

Supportive clinical features 1. Positive family history 2. Positive response to dopaminergic therapy 3. Presence of periodic limb movements (during wakefulness or sleep) Associated features of RLS I. Variable clinical course, but typically chronic and often progressive. 2. Physical examination normal in idiopathic/familial forms. 3. Sleep disturbance is a common complaint in more affected patients. Diagnostic features are those mandatory for a definite clinical diagnosis. Supportive clinical features are those which may increase the probability of a diagnosis in doubtful cases, such as is common in children. Associated features are typical, but do not contribute to diagnosis. Modified from Allen, 2003.

others will be located with a different inheritance model. Clinical trials have now established that dopaminergic agents are almost universally successful as therapy (Hening et aI., 2004), but that opioids, anticonvulsants, and benzodiazepines are also useful (Hening et aI., 1999). Resolution of the underlying cause of secondary RLS, such as iron repletion (Nordlander, 1953; O'Keeffe et aI., 1994), kidney transplantation (Yasuda et aI., 1986), or delivery (Lee et aI., 2001), usually cures the disorder. Based on therapeutic response and evidence of deficient iron storage in RLS patients, it is now suspected that both iron and dopamine dysfunctions play a role in the pathogenesis of RLS (Earley et aI., 2000a). PLM are repetitive limb movements that usually occur during sleep (PLMS) (Atlas Task Force of the American Sleep Disorders Association, 1993), but can also occur while individuals are awake (PLM while awake, PLMW), especially in patients with RLS (Hening et aI., 1999b). PLM are defined by their occurrence in a series (4 or more) of similar movements with a wide range of periods (5 to 90 s, by the latest official description) (Atlas Task Force

of the American Sleep Disorders Association, 1993) and a duration of 0.5 to 5 s. It has recently been proposed that PLMW can be longer lasting (up to 10 s), perhaps due to a voluntary prolongation of an initially involuntary movement (Michaud et aI., 2001). The movements involve the legs in almost every case, but in RLS patients, especially those severely affected, the arms may also be involved. PLM were first described among a series of disorders called nocturnal myoclonus (Symonds, 1953). This name was also adopted by Lugaresi and colleagues when they first associated these movements with RLS (Lugaresi et aI., 1968; Lugaresi, 1986). However, most PLM are not myoclonic in speed and it was later decided to adopt a terminology emphasizing their tendency to occur repetitively in a fairly regular manner (Coleman et al., 1980). Since their description by Lugaresi's group, additional clinical information about them has been gathered, some suggestive localization of the primary oscillator has been found, and epidemiological studies have shown that they are extremely common, especially in the elderly: more than half of whom

RESTLESS LEGS SYNDROME AND PERIODIC LIMB MOVEMENTS

may have more than 5 per hour of sleep (PLM index, PLMI, >5) (Ancoli-Israel et al., 1985; Mosko et al., 1988), a frequency which has been suggested to be clinically significant. It is currently controversial whether they cause a sleep disorder by themselves (which would be called PLM disorder or PLMD) (Mendelson, 1996; Nicolas et al., 1998; Montplaisir et al., 2000). They appear to have a basis in the spinal cord, but in intact individuals, may be associated with altered function of the brainstem (red nucleus and pontine reticular formation). Their treatment is essentially the same as RLS, although the intensity of therapy is usually less (Chesson et al., 1999). Recent studies have suggested they may be associated with enhanced spinal reflexes and it has been argued that they are most manifest in disorders of the dopamine system (Montplaisir et al., 2000). 36.1.2. Role ofclinical neurophysiology and related techniques in the investigation of RLS and PLM

In contrast to RLS, PLM can usually only be established by laboratory evaluation. They are most commonly a laboratory finding rather than a distinct clinical complaint. This results in somewhat different approaches to the two disorders and, as indicated above, while they overlap, they remain distinct conditions. While PLM are so common, at least in the elderly (Ancoli-Israel et al., 1985; Mosko et al., 1988), to suggest they may even be a normal variant, RLS appears to be a common, but distinct disorder that involves some specific pathophysiological factors, very likely to be both genetic and environmental (Hening, 2002). Those factors which provoke RLS, however, do clearly accentuate the underlying tendency to develop PLM. Many studies, moreover, tend to conflate the two disorders, so that many studies of PLM have not distinguished between patients with RLS and those without or have relied on the PLM in RLS patients to study various aspects of PLM. 36.1.2.1. Clinical investigations Clinical neurophysiologic techniques, usually polysomnography, are involved in the basic diagnosis of PLM, but not RLS (Chesson et al., 1997). However, a variety of techniques can be used in the characterization of the disorder, determination of its severity, and assessment of therapeutic impact.

585

Another related use of these techniques is in establishing potential secondary causes of either condition, but especially RLS. 36.1.2.2. Research investigations Clinical neurophysiologic techniques have also been used to elucidate the pathophysiology of RLS and PLM. These approaches have attempted to determine the locus of abnormality in the two conditions and to identify specific neural systems whose dysfunction may contribute to them.

36.2. Clinical investigations 36.2.1. Diagnosis

Diagnosis of PLM, usually PLMS, requires the establishment of a minimum number of leg movements within a time interval, usually the PLMI (number of PLMS per hour) (Diagnostic Classification Steering Committee of the American Sleep Disorders Association, 1997). This typically involves a sleep study or polysomnography (PSG) as originally practiced by Lugaresi and colleagues. EMG profiles, either simple or complex (Fig. 1), must match a predetermined level relative to biocalibration EMG levels and have a duration within the set limits (Table 2) (Atlas Task Force of the American Sleep Disorder Association, 1993; Michaud et al., 200 I). In addition, each successive PLM must occur within 4 to 90 s of the onset of the previous one. Surface EMG electrodes are usually placed on one, both, or linked anterior tibialis muscles which are usually involved in the movements (Chokroverty, 1999). PLMS temporally associated with overt sleep-related respiratory disturbances (SRD) are usually not counted, since arousals and associated movements can be generated by the SRD and, based on EMG alone, cannot be distinguished from independent PLM (Atlas Task Force of the American Sleep Disorder Association, 1993). As a result, respiratory monitoring, including airflow and respiratory effort, are a component of studies determining the presence of PLMS. As a general rule, a PSG is done with EEG, chin muscle tone, EKG, and EOG monitoring, since the diagnosis of PLM is not apparent prior to study and the PSG is usually undertaken to differentiate between a range of possible disorders (see Chapter 10). A more recent technique, actigraphic monitoring uses self-contained miniaturized accelerometers

586

...

W. HENING

'"

..... .1..11..1

.'

~

~

.

A

,I

v

J

'"

"I'" n ,mall I..

'1

..,

'.,.

v

!Y"

J '~

, r 11 I

SaO 9,7

,

d.l

.

~

.J. ,., •

liV

'1

.~

f

,

d

J

'1' I,j,

l'

..

,

.....

" .I.

T'

~I

'-"

'

'l

H.

V' 'I

I""

.,.

"

J. L I,,,,,,

... 'r

I

,-

. ,'.

,h, ' IlL .11.. ,

r {' Ir

v

....

, ,oJ ".1

1'1

/1

'V"

. ..

i.

.

?l ".

j

r l'

,

y

1 1

T

' ....,

'

.LL ,L.L l

.. .. '''''

,I

,j,

, r

ll.mc

..

,.A ~J,.

•. " I'"

u.

In~"

,,~

T

..J ,>1,

Ilt.iJ.JluJ

..... '1/'

I

l'

,

r"r

..

,

'1"

,l,

r'"

,.

'1'

"f, TI>' .... 0

1

II v v

'v

v

1"

T

1""

S, X 94.

Tae lop h

)

Fig. 1. Series of periodic limb movements in a sleeping patient. These occur almost exclusively in the left leg. Burst at arrow shows several initial high amplitude brief components. Middle burst in record is prolonged, consistent with an arousal leading to voluntary prolongation of movement. After this burst, there is an altered EEG rhythm and EMG activity spreading to chin and right leg, as well as altered respiratory rhythm. (Chin EMG has respiratory artifacts through tracing.) The bursts recur in a nearly periodic fashion. Top four traces - EEG from vertex (top two traces) and occiput (third and fourth traces) referenced to the opposite ear. Fifth and sixth traces, left and right EOG (electrooculograms). Seventh trace, chin EMG. Eighth trace, EKG. Ninth and tenth traces, left and right tibialis anterior EMGs. Eleventh trace, oral air flow. Twelve and thirteenth traces, thoracic and abdominal respiratory effort. Bottom trace, sound recording. Trace superimposed on abdominal effort is a displaced oximeter tracing indicating oxygen saturation. Thick vertical lines indicate 16 s divisions.

(Chapter 13) (Tryon, 1991) with central processing and memory. These measure movement in terms of limb acceleration and are generally placed on the ankle or foot. They are clearly not identical to the standard EMG monitoring, since they record movement and not electrical potentials. While early models of these instruments could only provide summed activity counts for certain bin sizes (usually minutes), more recent models have additional computational resources that allow them to count movements to determine whether they meet the duration and period criteria for PLM. Some recent studies suggest that they fairly accurately reflect the PLMI or total PLM count in individuals (Kazenwa-

del et al., 1995; Gorny et al., 1998). It is also controversial whether the important phenomena is the electrical potential or the actual movement and it has been argued that EMG can overcount PLM. The advantage of actigraphy is that it is a technique which is much less costly than PSG, can be conducted in any environment, and is rather easily used on a long-term basis. The disadvantage is that actigraphy provides less information about the context of movement (Allen, 2003). It does not do a particularly good job of discriminating sleep from wake and it cannot detect those movements associated with SRD. Actigraphy is also non-specific and cannot tell which muscles are generating a move-

RESTLESS LEGS SYNDROME AND PERIODIC LIMB MOVEMENTS

Table 2 Operational definition of periodic limb movements (Atlas Task Force of the American Sleep Disorders Association, 1993). Period

4 to 90 s

Duration of EMG burst

0.5 to 5 s*

Number of bursts in sequence

Minimumof4

Amplitude of EMG burst

25% of biocalibration

Period is measured from onset to onset To be counted, movements must occur in a sequence of the required length

* During the suggested immobilization test when patients are awake, a duration of up to lOs has been proposed as acceptable (Michaud et aI., 2001). ment. Some recent instruments do include light (Ancoli-Israel et al., 1997; Matsumoto et al., 1998; Jean-Louis et al., 2001) or position detectors (Gorny et al., 2001) which can provide some state information. They can indicate when lights are out and the subjects are likely in bed or when they are in a reclining position when PLM are more likely to occur. While RLS is a purely clinical diagnosis, in uncertain or complicated cases PSG may be performed to support the diagnosis. As a result, the presence of PLM is indicated as a supportive feature of the diagnosis of RLS (Table 1). This is especially true in children, in whom the clinical diagnosis of RLS is less clearcut. A diagnosis of probable RLS in children can be made with the presence of PLM and other clinical features, such as a positive family history of RLS (Allen, 2003). Of course, having an objective diagnostic test would be attractive for RLS. Unfortunately, attempts to use PLMS or PLMW counts or indices to diagnose RLS have resulted in only modest levels of specificity and sensitivity (Michaud et al., 2002b). Such counts have been examined not only during a PSG, but also in a provocative test called the suggested immobilization test (SIT) (Montplaisir et al., 1998; Inoue et al., 2002; Michaud et al., 2002a). In this test, the subject is asked to sit quietly, typically for an hour, while leg movements (usually determined by EMG) and subjective complaints are monitored. Most patients with RLS will experience greater sensory discomfort

587

than normals and will also have a fair number of movements during this period which fulfill criteria for PLMW (Michaud et al., 2002a). A combination of increased discomfort during the SIT and PLMW during PSG using optimal thresholds has been reported to provide a better discrimination from controls than any single measure (sensitivity 82%; specificity 100%) (Michaud et al., 2002b). 36.2.2. Characterization of the disorders 36.2.2.1. PLM and sleep/wake state The PSG has been used to determine the modulation of PLM by specific sleep states (Pollmacher et al., 1993; Nicolas et al., 1999) or activities (Dzvonik et al., 1986). Sleep state is determined by an examination of EEG, EOG, and chin EMG (Chapter 10). The period, duration, and amplitude of movements has been studied as well as the extent of time during which movements occur. The general finding is that period is decreased in wake, then increases through the non-rapid eye movement (NREM) stages of sleep from light (stage 1) to deep (stages 3 and 4; slow wave sleep, SWS). The period is more regular during sleep than wake. Both duration and amplitude are reduced in slow wave sleep and rapid eye movement (REM) sleep (Pollmacher et al., 1993). Two temporal patterns of PLM prevalence have been noted: a decrescendo pattern with most movements early during the sleep period and a more even pattern (Culpepper et al., 1992). The decrescendo pattern is associated with RLS (Trenkwalder et al., 1999a), but is also found in many individuals without RLS. Some patients with PLMS, such as those with narcolepsy (Godbout et al., 1988) or REM behavior disorder (RBD) (Lapierre and Montplaisir, 1992), tend to have more movements during REM sleep. These narcoleptic or RBD patients with REM period PLM are presumably manifesting the REM-related motor dyscontrol that characterizes those conditions, allowing a greater emergence of the PLM despite REM inhibition. 36.2.2.2. PLM and the cyclic alternating pattern (CAP)

The PSG has also been used to examine the microstructure of sleep. Transient changes in the complex of EEG frequencies can indicate a brief lightening of sleep. If these persist for 3 s, they are called arousals (Sleep Disorders Atlas Task Force of

588 the American Sleep Disorders Association, 1992). Arousals can also continue into full wakefulness. It has been suggested that PLMS can cause arousals, but one problem with this association is that PLM often follow, rather than precede, an arousal (Montplaisir et aI., 1996). However, PLM occurring during the lighter stages of sleep, NREM 1 and 2, have been reported more likely to be associated with arousals (Pollmacher et al., 1993). Another form of microstructure is the cyclic alternating pattern (CAP) (Terzano et aI., 1985). This consists of repetitive cycles of more activated (A phase) and less activated (B phase) EEG rhythms. By definition, these cycles last between 4 and 120 s, almost exactly the same period as defined for PLM (Terzano et aI., 2001). It has been found that PLMS tend to be associated with the A phase of the CAP (Droste et aI., 1996; Parrino et aI., 1996). The significance of CAP remains to be further defined, but it is taken as a manifestation of unstable or dynamically changing sleep, A phases with predominant slow (delta) activity (AI) are associated with deepening sleep while those with predominant faster (alpha) activity (A3) are associated with lightening sleep (Terzano et aI., 2000). CAP and PLM also appear to be related to cyclic alternation of autonomic arousal and quiescence, as indicated by the increase in EKG-established heart rate that both accompanies the A phase and PLM (Winkelman, 1999; Ferini-Strambi et aI., 2000; Ferri et aI., 2000). A phases with a higher portion of alpha (8-13 hertz EEG frequencies) are considered to be more arousing and are also tightly associated with PSG arousals (Parrino et aI., 2001). The link of PLM to these other cyclic phenomena also suggests that they are part of an extensive system of rhythmically modulating internal states (Lugaresi et aI., 1972; Droste et al., 1996; Sforza et aI., 2002). The exact source and significance of these modulating states remains to be further clarified.

36.2.2.3. RLS and circadian rhythm In RLS, EMG studies (usually with PSG) have been used to establish the circadian rhythm of associated PLM (Trenkwalder et aI., 1999a). During wake, the movements are monitored by repeated SITs, while during sleep, the standard PSG is used to establish frequency of movements. These studies have established that RLS patients are most likely to have PLM in the period from 23:00 to 3:00, during

W. HENING

the first half of the usual sleep period, and least likely between 9:00 and 14:00, the time after normal waking (Fig. 2). Similarly, if the SIT is modified to allow the patient to move when symptomatic, restless, voluntary motor activity can be determined by actigraphy (Hening et aI., 1999c). The voluntary movements have a similar circadian profile, as do subjective complaints of RLS patients. Studies with ambulatory rectal core temperature recordings have shown that RLS is worst on the falling phase of the core temperature cycle, but that the timing of the circadian cycle is relatively normal (Hening et aI., 1999c). It has been suggested that RLS may be activated by some process that varies with circadian time, such as blood levels of iron or dopamine (Hening et aI., 1999c; Trenkwalder et aI., I999a).

36.2.3. Assessment of severity and therapeutic impact 36.2.3.1. Assessment of PLM severity For PLM, the first measure of severity is a simple count or index of movements, however determined (Diagnostic Classification Steering Committee of the American Sleep Disorders Association, 1997). More is worse. However, it has been suggested, although never confirmed, that another measure is the degree to which the movements cause or are associated with arousals or awakenings. The PLM-arousal index (PLMAI) determined by PSG is therefore often reported as an additional measure of severity. In many cases, however, it has been shown that there is little clinical significance to the number of PLM or even the PLMAI (Mendelson, 1996; Nicolas et al., 1998; Montplaisir et aI., 2000). 36.2.3.2. Assessment of RLS severity 36.2.3.2.1. PSG measures and MSLT. PLM measures, determined from PSG, as well as indications of sleep quality are used as objective assessments of RLS (Hening et aI., 1999b; Hening et aI., 2004). The start and stop times of sleep and its different NREM and REM states are determined. 1\vo important measures are the sleep efficiency (percent of time in bed during which the person is asleep) and the sleep latency (time from lights out to sleep) (Chapter 10). The fraction of SWS, considered more restorative, and the number of arousals are also used as measures of severity. Both PLM measures and PSG sleep measures are frequently

589

RESTLESS LEGS SYNDROME AND PERIODIC LIMB MOVEMENTS

Circadian Plot

100

12:00

24:00

Time

Fig. 2. Circadian plot of RLS and PLM. Normalized values of the RLS measures are plotted against clock time from 8:00 through 24:00 to 8:00. Hours represent beginning of a SIT period or beginning of an hour of enumerating PLMS. PLMS - periodic limb movements in sleep. Hourly averages from the first study from 23:00 to 7:00 are plotted as percentage of the maximum (94.9 per hour in the period from 24:00 to 1:00). Mean value for 8 subjects and two nights PSG. PLMW periodic limb movements while awake; ACT - activity counts (measure of motor restlessness); LEG - subjective leg discomfort. All these are plotted as percentage from minimum (0%) to maximum (100%). PLMW is from first study (Trenkwalder et aI., I999a); ACT and LEG from second (Hening et aI., 1999c). PLMW - minimum of 13.8/hour (9:00), maximum of 91.5/hour (2:00). ACT - minimum of 124.9 counts (12:00), maximum of 447.8 (3:00). SBJ - minimum of 2.96 (on 0 to 10 scale averaged over five determinations per one hour mSIT) at 9:00; maximum of 6.04 (at 12:00). The values of PLMW, ACT, and SBJ were taken by averaging daytime SITs (or mSITs) from the first two days of study; values from hours between 11:00 and 6:00 were taken from the night of sleep deprivation.

used as outcome measures for therapeutic studies in RLS (Hening et al., 1999a). The MSLT, a standard methodology for assessing such sleep disorders as narcolepsy or sleep disturbed breathing (SDB) has been little used in RLS and what results have been supported suggest that the MSLT does not adequately measure severity of RLS nor respond clearly to changes in clinical status.

36.2.3.2.2. Actigraphy and alternate methodologies. Actigraphy can also be used to assess RLS (Allen et aI., 1992; Earley et al., 1999; Hening et aI., 1999a, 2004). It has been used to measure PLM during therapeutic trials using instruments that can

detect and count PLM. It can also be used to examine activity profiles, which are elevated during the evening and night hours in patients with RLS (Hening et aI., 1999c). This is a new technology (American Sleep Disorders Association, 1995; Sadeh et aI., 1995). Its advantages include its ambulatory use, its relatively low capital costs, its ability to be employed for extended recording, and its potential to provide additional information if combined with recording of ambient light, body posture, or measures of sleep state such as respiration or eye movement. Its current disadvantage is the lack of good validating studies and the relatively non-specific measure of movement that it provides.

590

Future developments should make this a more attractive and economic means of assessing PLM and RLS. In addition to these more standard methodologies, a number of whole body recording techniques using either bed motion (Salmi and Leinonen, 1986; Kaartinen et al., 1996; Polo-Kantola et al., 2001) or temperature (Lu et al., 1999; Tamura et al., 1999) have been developed that can record sleep and, it has been reported, enumerate PLM as well as respiration. Developed in regional centers (Finland and Japan, respectively), they have not yet had widespread exposure or confirmation of usefulness outside their originating areas. 36.2.3.3. Assessment of therapeutic impact 36.2.3.3.1. Assessment of therapeutic impact in PLMD. In evaluating PLMD, it is important to establish both that the PLM have been reduced and that the objective manifestations of the associated sleep complaint have been improved. This generally requires a PSG for insomnia complaints (to show, in particular, increased sleep efficiency, reduced wake after sleep onset, reduced arousals and awakenings, and increased amounts of deeper sleep stages) and an MSLT for EDS complaints. Some of these measures could be accomplished by actigraphy or other alternate recording techniques (see previous section). Objective techniques can also be combined with subjective measures related to the patients' complaints. In recent years, there have been fewer studies of PLM, perhaps because of the concern that PLM may not be of clinical significance (Mendelson, 1996; Nicolas et al., 1998; Montplaisir et al., 2000). 36.2.3.3.2. Assessment of therapeutic impact in RLS. In general, there has been a tendency in the last couple of years to shift from objective to subjective measures for evaluation of therapeutic efficacy. Since not all RLS patients have either PLMS or marked sleep disturbance, subjective measures have also or even preferentially been used to measure treatment effect. In large part, this is due to the cost of the usual objective measures (PSG sleep studies and related PSG techniques). The recent literature, however, reflects a mix in which objective techniques (PSG, actigraphy) have been combined with subjective measures (Hening et al., 1999a, 2004). A new validated rating scale, the

W. HENING

IRLSSG rating scale (IRLS), is the first of what may be a series of standardized subjective ratings (The International Restless Legs Syndrome Study Group, 2003). How subjective and objective ratings interrelate remains to be studied, although most therapeutic studies using a combination of subjective and objective measures have found parallel improvement in both (Hening et al., 1999a, 2004). At least one recent multi-center study used new technologies for storing and transmitting PSG data (converting various proprietary files into European data format and transferring to a central scoring station by internet) in order to standardize the evaluation of this procedure when used at different locations (Penzel et al., 2002).

36.3. Research investigations 36.3.1. Establishing a locus for PLM

The report that patients with clinically complete spinal cord lesions still experience PLM has led to the suggestion that there is a generator for PLM within the spinal cord (de Mello et al., 1995, 1996, 1999). Similar to studies of rhythmic behavior like locomotion (Calancie et al., 1994), particularly in other species, studies in these patients find motor activity that resembles PLM in patients with severed descending pathways to the spinal cord. This does not indicate that there are no descending influences in the intact patient; the sleep/wake state distribution of PLM in the spinal cord transected (SCT) patients is quite different from that of intact patients. Nor does it even indicate that all of the circuitry for the complete generator is located within the spinal cord, since it is not clear that the movements of PLM in the SCT patients is exactly the same as in intact patients. In general, most studies looking for a generator have either not discriminated between "pure" PLM patients and those with PLM as a component of RLS or have studied all RLS patients. As a result, the findings may, in many cases, only be applicable to the RLS patients. 36.3.1.1. Polymyography Several studies have used polymography to characterize the distribution and spread of muscle activity in PLM. This technique has been useful in determining the spread of excitation in such cases as

RESTLESS LEGS SYNDROME AND PERIODIC LIMB MOVEMENTS

propriospinal myoclonus (Brown et al., 1991, 1994; Chokroverty et al., 1992; Fouillet et al., 1995), presumably mediated by slowly conducting cord propriospinal fibers. The results of this technique when applied to PLM in RLS, however, have been somewhat inconclusive. One consistent finding has been that the movements vary considerably in timing and distribution between patients and even within patients. This supports an earlier video-analysis which came to the same conclusion (Walters et al., 1984). A German group found that, in some patients, the progression of muscle activation, the latency between EMG onset in different muscles, the consistency of pattern, and the possibility of recruitment both rostrally and caudally in the cord during specific movements supported a propriospinal mechanism for these movements (Trenkwalder et al., 1996). This would support a localized spinal generator for the PLM. Other polymyographic studies have found that PLM in different limbs can occur asynchronously with different periods, suggesting that there may be multiple oscillators acting (Hening et al., 1989; Gupta et al., 1996). Meanwhile, an Italian group could not replicate the finding that individual patients exhibited consistent patterns of spread or progression consistent with a propriospinal propagation (Provini et al., 2001). They conclude that it is more likely that the muscles recruited in any given moment are those which are at the lowest threshold for excitation and that the movements are driven by some more generalized excitation. They also reported that they could not support an earlier suggestion that the movements are relatively isomorphic with the Babinski reflex (Smith, 1985, 1987) or a flexion withdrawal reflex (Bara-Jimenez et al., 2000). The most parsimonious explanation for these findings is that there are distributed, limb specific oscillators within the spinal cord. This would correspond to findings in other species that there may be oscillators for rhythmic behavior that consist of locally competent oscillators for each limb or even muscle group that are recruited into an overall pattern by a synchronizing mechanism (Grillner, 1985; Grillner and Dubuc, 1988). Similar issues have arisen with the documentation that Parkinsonian tremor can have a different frequency in each limb and sometimes even in parts of limbs (Hening et al., 1987; Hunker and Abbs, 1990), although the dominant hypothesis has been that these different

591

rhythms arise from distinct brainstem or basal ganglia oscillators (Bergman et al., 1994; Hutchison et al., 1997; Rothwell, 1998). One study which examined both idiopathic and uremic patients with RLS found that there was no discernible difference between the movements of the two groups of patients (Trenkwalder et al., 1996). 36.3.1.2. Studies of reflex responses and evoked potentials Both monosynaptic and polysynaptic reflexes have been studied in PLM and RLS. These reflexes have been known to have both spinal and brainstem bases. Simple reflexes (H-reflex) and evoked potentials have generally been normal (Bucher et al., 1996). The reports related to polysynaptic reflexes have given equivocal results in different studies. The blink reflex has been reported to be normal in some studies (Bucher et al., 1996), but abnormal in others (Wechsler et al., 1986; Briellmann et al., 1996). A problem of these studies has been that they have generally examined patients who were asymptomatic and who were studied during the day. Any abnormality in PLM or RLS may be state dependent and may require a more difficult investigation during symptomatic periods, especially in the evening or night. One recent study used the flexor withdrawal reflex to examine the difference between RLS patients and controls and also examined patients in different states, including when some of them were symptomatic with RLS (Bara-Jimenez et al., 2000). The RLS patients showed a response that was more easily elicited and widely distributed than controls. In addition, the RLS patient response showed enhancement of some components during sleep and otherwise revealed a different state dependence than that of the normal controls. While this result may support a spinal cord basis for the reflex abnormality, it could still be due to descending influences modulating intrinsic cord circuitry. This result requires further confirmation, but it seems more likely to provide an insight into altered excitability levels in RLS patients at least. The report also indicated that there were similarities between the patients' recorded PLM and the EMG patterns of the flexor withdrawal reflexes. This may indicate some shared circuitry. Studies of standard evoked potentials (SEP, YEP, BAEP) in RLS and PLM patients has not revealed

592 any abnormalities, though relatively few studies have been undertaken (Mosko and Nudleman, 1986; Provini et aI., 2001).

36.3.2. Establishing a locus for RLS In looking for the locus for RLS, there is the disadvantage that the measure of symptoms is subjective. However, this may also encourage studies to be done in a symptomatic state to examine the impact of immediate interventions. Structural examinations such as MRI have been normal in RLS (Bucher et aI., 1996) as have functional studies looking at regional brain activity at rest through PET scanning (Trenkwalder et aI., 1999b).

36.3.2.1. Functional imaging studies In one study, RLS patients were studied with fMRI in the symptomatic state with control states (asymptomatic rest; simulated PLM) (Bucher et aI., 1997). This study found that development of sensory symptoms (leg discomfort) led to activation of sensory tract nodes (thalamus, sensorimotor cortex) as well as the ipsilateral cerebellum. Addition of spontaneous PLM produced additional activation of the red nucleus and pontine reticular formation. Simulated movements did not activate the red nucleus and reticular formation, but activated normal motor system structures such as the basal ganglia and motor cortex. This study supports the idea that PLM do not originate from cortical structures or use normal voluntary motor pathways. However, it reveals less about distinctive pathways of sensory symptoms and it is not possible to conclude that any of the activated structures need to be considered as the generators for PLM or symptoms. In one small study of a parent and child affected by RLS, the authors found using SPECT to measure regional blood flow semi-quantitatively that increased painful sensory symptoms were associated with increased anterior cingulate blood flow in the two patients, combined with increased flow to the thalamus and decreased flow to the caudate (San Pedro et aI., 1998). These findings, however, are consistent with the kinds of responses generally found in pain conditions and so may not reflect any distinctive cerebral involvement in RLS. It is unclear whether the responses are specific for painful forms of RLS, which make up about one third of patients.

W. HENING

36.3.2.2. Transcranial magnetic stimulation The basic motor evoked responses to transcranial magnetic stimulation (TMS) have shown normal thresholds and conduction velocities in RLS (Entezari-Taber et aI., 1999; Tergau et aI., 1999; Provini et aI., 2001). This suggests that the motor pathways are generally intact. However, it has been noted that a different phenomena, the inhibitory period after stimulation, is reduced in RLS (Entezari-Taher et al., 1999). Because the peripheral silent period after direct motor neuron stimulation is normal while the TMS silent period is prolonged, the authors conclude that the inhibitory potential is reduced at a supraspinal leveL In another study, the Goettingen group examined the facilitatory and inhibitory responses to paired TMS pulses (Tergau et aI., 1999) (Chapter 8). While they did not find a prolonged cortical silent period, they did find that the RLS patients showed decreased inhibitory responses in both hand and foot muscles, suggesting a general deficit in corticocortical inhibition. Based on the current understanding of the modulation of cortical excitability, this suggests that there is some sub-cortical alteration in inhibitory influences. However, there was increased corticocortical facilitation only in the hand muscles. The authors suggest that the lack of facilitation in the foot muscles may be due to some covert activation. In doing this study, subjects were examined in the late afternoon when RLS patients may become symptomatic; in fact, roughly half of the RLS patients were symptomatic during the test. These studies do not specify where between cortex and spinal cord there is an imbalance in neural tone or what brings it about. However, the finding that dopamine blockers can produce similar phenomena (Ziemann et aI., 1997) supports the dopamine theory of RLS causation. 36.3.3. Examining the contribution to specific neural systems to RLS Early studies found that a variety of different neuroactive substances could improve RLS symptoms (Walters and Hening, 1987). This led to proposals that the endogenous opiate and dopamine systems contribute to the pathogenesis of RLS. These have subsequently been investigated with pharmacologically monitored interventions and imaging studies to determine whether abnormal

RESTLESS LEGS SYNDROME AND PERIODIC LIMB MOVEMENTS

responses or number of binding sites for these symptoms support the diagnosis. More recently, it has been suggested, again based initially on therapeutic response (O'Keeffe et al., 1993, 1994), that iron deficiency might contribute to RLS (Sun et al., 1998). RLS patients have been suspected of having abnormalities of iron transport, storage, and elimination, leading to low brain levels (Earley et al., 2000). 36.3.3.1. Pharmacological probes of neural systems Several studies have examined the ability of opioid and dopamine blockers to activate symptoms or an abnormal biochemical response in RLS patients. Patient response was measured during SITs or sleep studies. In two early studies, it was found that naloxone could activate symptoms in opioid dependent patients and pimozide could activate symptoms in dopamine treated patients (Walters et al., 1986; Montplaisir et al., 1991). In the second study, using a single patient, blocking the dopamine system relieved both symptoms and PLM, but blocking the opiate system only relieved leg discomfort (Montplaisir et al., 1991). A more recent study examined patients who were not medication dependent and compared them to controls (Winkelmann et al., 2001). Tests were run in the afternoon and both PLM and honnonal responses were measured. While giving the originally asymptomatic patients naloxone or metoclopramide increased symptoms, the changes did not reach statistical significance. This may have been due to the small number of subjects resulting in inadequate power. A similar finding was found in the increase of prolactin in the patients after metoclopramide administration. One problem with this study was that it was done in the afternoon when patients were not symptomatic. It was reported in one earlier study that even opioid treated patients did not show a response during periods of the day when they were expected to be asymptomatic (Walters et al., 1986). While somewhat inconclusive, these results do suggest the involvement of the opiate and dopamine systems in RLS, if only in those patients already treated with the medications. The differential effect of blocking the two systems has never been replicated, but does suggest that the dopamine system is more directly involved in generating symptoms. The studies are somewhat complicated by considerations

593

of the normal circadian rhythm of RLS (Hening et al., 1999c; Trenkwalder et a1., 1999a), which may require symptomatic patients to be studied, since there may be a dynamic and functional abnormality rather than a static one. 36.3.3.2. Imaging studies of neurotransmitter systems Both SPECT and PET studies have been performed to determine the level of pre- and post-synaptic dopamine markers in RLS and/or PLMD. These studies have been inconsistent in outcome, either showing normal levels of these markers (Trenkwalder et a1., 1999a; Eisensehr et a1., 2001; Michaud et al., 2002c; Tribl et a1., 2002) or modest decrements (10-15%) (Staedt et a1., 1995; Turjanski et al., 1999; Ruottinen et a1., 2000; Michaud et a1., 2002c) in the striatum. While the picture is quite mixed, there is perhaps slightly more evidence for deficits in the putamen and postsynaptically. All studies have been done during the morning or afternoon, when patients are not fully symptomatic. Not all have had well defined control populations or considered whether patients were under treatment with dopaminergic agents. Those studies which examined treated patients did not find any difference between treated and untreated patients. Taken as a whole, these studies do point to some modest decrement of dopamine markers in the basal ganglia. However, the significance of this finding for the pathogenesis of RLS is unclear and without studies during the symptomatic period or dynamic studies to examine all aspects of dopamine synthesis, release, and uptake, the actual functional abnormality of the dopamine system in RLS is yet to be discerned. In addition, there is little evidence that the striatonigral pathway is abnormal in RLS. Most RLS patients do not have signs of Parkinsonism and their motor (Alberts et al., 2001) and olfactory (Adler et a1., 1998) functions, which are impaired in Parkinson's disease, appear to be intact. Various measures of magnetic resonance can be used to highlight specific atomic species. To measure iron, R2', representing the relaxation resulting from field inhomogeneities that can be reversed by an 1800 pulse provides the most specific measure of regional brain iron. This is because they will not be contaminated by other factors affecting relaxation rates such as brain water content (Gelman et a1., 1999). In one initial report, iron was decreased in the

w.

594

basal ganglia, the substantia nigra more notably than the putamen, compared to age-matched controls (Allen et aI., 2001). MRI measured iron stores were negatively related to severity of RLS. The possibility of an iron deficiency is further supported by low ferritin values in a subset of RLS patients (O'Keeffe et aI., 1993; Sun et aI., 1998), decreased CSF ferritin (Earley et aI., 2000b), and an autopsy study confirming the reduction of substantia nigra iron (Connor et aI., 2003). 36.3.3.3. Autonomic function testing The period appearance of PLM has been associated not only with sleep microstructure (arousals and CAP, see previous material), but also with known cyclical activities in the autonomic nervous system (Lugaresi et aI., 1972; Droste et aI., 1996). While earlier studies merely analogized the PLM rhythm to known autonomic rhythm (Lugaresi et al., 1972), more recent studies have actually established that PLM are associated with autonomic arousal (primarily measured by increased heart rate) (Winkelman, 1999; Sforza et aI., 2002). This is true even when the arousal does not extend to evident cortical EEG rhythm changes (Winkelman, 1999; Sforza et aI., 2002). The changes, however, are not purely reactive, but show a time-course which suggests that autonomic changes develop prior to onset of EMG activity and actual movement (Sforza et aI., 2002). In addition, one study found that, after treatment, periodic PLM associated with arousals were substantially decreased, but that periodic arousals persisted largely unchanged (Montplaisir et aI., 1996). This finding suggests that PLM are but one of a number of discrete manifestations of a complex periodicity that involves many physiologic domains.

36.4. Summary and conclusion This review shows that a variety of electrodiagnostic techniques have been important for diagnosis and assessment, especially in PLMD but also in RLS, while these techniques have provided important leads towards the location of a lesion or the systems involved in pathogenesis in these conditions. In clinical assessment, the trend and modem conditions favor a gradual shift from laboratorybased PSG studies to more ambulatory, perhaps actigraphically based studies. As actigraphic tech-

HENING

nology develops and gains further capacities, this is likely to be an increasingly attractive option for examining PLM and motor restlessness in these conditions. It is clear that some form of validated objective measure will be a necessity for therapeutic trials, if only as a secondary outcome point, since placebo arms of RLS trials have found strong placebo effects in relation to subjective outcome measures (Telstad et aI., 1984; Trenkwalder et aI., 2001). The use of clinical electrodiagnostic tools for investigating the pathophysiology of RLS has not reached the same level of successful application. Most results have been suggestive, but inconclusive. A major problem of many studies has been the lack of a careful dissociation of PLM alone from RLS and the failure to evaluate patients at times when they are symptomatic. Recent studies do show a recognition of these issues. Future studies should be more appropriately structured. Finding involved genes, which may be in the offing (Desautels et al., 2001), could help by more clearly directing studies at certain neural pathways or markers. Animal models (Ondo et aI., 2000; Baier et aI., 2002) may also help provide opportunities to apply additional neurophysiological techniques. PLM and RLS may tum out to be very intriguing phenomena. PLM may be involved in a periodic modulation of physiological cycles whose ramifications have not yet been fully uncovered and which may be important for a wide range of different processes. RLS, in tum, may be a relative form of disorder, one in which the structural elements are relatively unaffected, but the functional activities of key neural systems are altered in a state-dependent manner. Solving the underlying abnormality may require the use of modified or new tools. The final explanation may tell us much about the way in which the motor system is normally modulated through different states, particularly those involved in rest/activity alteration and circadian rhythms.

References Adler, CR, Gwinn, KA and Newman, S (1998) Olfactory function in restless legssyndrome. Mov. Disord., 13(3): 563-565.

Alberts, JL, Adler, CR, Saling, M and Stelmach, GE (2001) Prehension patterns in restless legs syndrome patients. Parkinsonism Relat. Disord., 7(2): 143-148.

RESTLESS LEGS SYNDROME AND PERIODIC LIMB MOVEMENTS

Allen, RP (2003) Ambulatory activity monitoring. In: S Chokroverty, WA Hening and AS Walters (Eds.), Sleep and Movement Disorders. Philadelphia: ButterworthHeineman, pp. 144-152. Allen, RP and Earley, CJ (2000) Defining the phenotype of the restless legs syndrome (RLS) using age-ofsymptom-onset. Sleep Med., 1(1): 11-19. Allen, RP and Earley, CJ (2001) Restless legs syndrome: a review of clinical and pathophysiologic features. 1. Clin. Neurophysiol., 18(2): 128-147. Allen, RP, Kaplan, PW, Buchholz, DW, Earley, CJ and Walters, JK (1992) Accuracy of a physical activity monitor (PAM) worn on the ankle for assessment of treatment response for periodic limb movements in sleep. Sleep Res., 21: 329. Allen, RP, Barker, PB, Wehrl, F, Song, HK and Earley, CJ (200I) MRI measurement of brain iron in patients with restless legs syndrome. Neurology, 56(2): 263-265. Allen, RP, Hening, WA, Montplaisir, J, Picchietti, D, Trenkwalder, C, Walters, AS et al. (2003) Restless legs syndrome: diagnostic criteria, special considerations, and epidemiology: A report from The RLS Diagnosis and Epidemiology workshop at the National Institute of Health. Sleep Med., 4(2): 101-119. American Sleep Disorders Association (1995) Practice parameters for the use of actigraphy in the clinical assessment of sleep disorders. Sleep, 18(4): 285-287. Ancoli-Israel, S, Kripke, DF, Mason, Wand Kaplan, OJ (1985) Sleep apnea and periodic movements in an aging sample. J. Gerontol., 40(4): 419-425. Ancoli-Israel, S, Klauber, MR, Jones, DW, Kripke, DF, Martin, J, Mason, W et al. (1997) Variations in circadian rhythms of activity, sleep, and light exposure related to dementia in nursing-home patients. Sleep, 20(1): 18-23. Atlas Task Force of the American Sleep Disorders Association (1992) EEG arousals: scoring rules and examples. Sleep, 15(2): 173-184. Atlas Task Force of the American Sleep Disorders Association (1993) Recording and scoring leg movements. Sleep, 16: 748-759. Baier, PC, Winkelmann, J, Hohne, A, Lancel, M and Trenkwalder, C (2002) Assessment of spontaneously occurring periodic limb movements in sleep in the rat. J. Neurol. Sci., 198(1-2): 71-77. Bara-Jimenez, W, Aksu, M, Graham, B, Sato, Sand Hallett, M (2000) Periodic limb movements in sleep: state-dependent excitability of the spinal flexor reflex. Neurology, 54(8): 1609-1616. Bergman, H, Wichmann, T, Karmon, B and DeLong, MR (1994) The primate subthalamic nucleus. II. Neuronal activity in the MPTP model of parkinsonism. J. Neurophysiol., 72(2): 507-520.

595

Bonati, MT, Ferini-Strambi, L, Aridon, P, Oldani, A, Zucconi, M and Casari, G (2003) Autosomal dominant restless legs syndrome maps on chomrosome 14q. Brain, 126(6): 1485-1492. Briellmann, RS, RosIer, KM and Hess, CW (1996) Blink reflex excitability is abnormal in patients with periodic leg movements in sleep. Mov. Disord., 11(6): 710714. Brown, P, Thompson, PD, Rothwell, JC, Day, BL and Marsden, CD (1991) Axial myoclonus of propriospinal origin. Brain, 114(1A): 197-214. Brown, P, Rothwell, JC, Thompson, PD and Marsden, CD (1994) Propriospinal myoclonus: evidence for spinal "pattern" generators in humans. Mov. Disord., 9(5): 571-576. Bucher, SF, Trenkwalder, C and Oertel, WH (1996) Reflex studies and MRI in the restless legs syndrome. Acta. Neurol. Scand., 94(2): 145-150. Bucher, SF, Seelos, KC, Oertel, WH, Reiser, M and Trenkwalder, C (1997) Cerebral generators involved in the pathogenesis of the restless legs syndrome. Ann. Neurol., 41(5): 639-645. Calancie, B, Needham-Shropshire, B, Jacobs, P,Willer, K, Zych, G and Green, BA (1994) Involuntary stepping after chronic spinal cord injury: Evidence for a central rhythm generator for locomotion in man. Brain, 117: 1143-1159. Chesson, AL, Jr, Ferber, RA, Fry, JM, Grigg-Damberger, M, Hartse, KM, Hurwitz, TD et al. (1997) The indications for polysomnography and related procedures. Sleep, 20(6): 423-487. Chesson, AL, Jr, Wise, M, Davila, D, Johnson, S, Littner, M, Anderson, WM et al. (1999) Practice parameters for the treatment of restless legs syndrome and periodic limb movement disorder. An American academy of sleep medicine report. Standards of practice committee of the American Academy of sleep medicine. Sleep, 22(7): 961-968. Chokroverty, S (1999) Sleep disorders medicine: basic science, technical considerations, and clinical aspects (2nd ed.). Boston: Butterworth-Heinemann. Chokroverty, S, Walters, A, Zimmerman, T and Picone, M (1992) Propriospinal myoclonus: a neurophysiologic analysis. Neurology, 42(8): 1591-1595. Chokroverty, S, Hening, WA and Walters, AS (Eds.) (2003) Sleep and movement disorders. Philadelphia: Butterworth-Heinemann. Coleman, RM, Pollak, CP and Weitzman, ED (1980) Periodic movements in sleep (nocturnal myoclonus): relation to sleep disorders. Ann. Neural., 8: 416-421. Connor, JR, Boyer, PJ, Menzies, SL, Dellinger, B, Allen, RP, Ondo, WG and Earley, CJ (2003) Neuropathological examination suggests impaired brain iron

596 acquisition in restless legs syndrom. Neurology, 61(2): 278-279. Culpepper, W, Badia, P and Shaffer, J (1992) Time-ofnight patterns in PLMS activity. Sleep, 15: 306-311. De Mello, MT, Da Silva, AC and Tufik, S (1995) Sleep study after acute physical activity in spinal cord injury. Sleep Res., 24A: 391. De Mello, MT, Lauro, FA, Da Silva, AC and Tufik, S (1996) Incidence of periodic leg movements and of the restless legs syndrome during sleep following acute physical activity in spinal cord injury subjects. Spinal Cord, 34(5): 294-296. De Mello, MT, Poyares, DL and Tufik, S (1999) Treatment of periodic leg movements with a dopaminergic agonist in subjects with total spinal cord lesions. Spinal Cord, 37(9): 634-637. Desautels, A, Turecki, G, Montplaisir, J, Sequeira, A, Verner, A and Rouleau, GA (200I) Identification of a major susceptibility locus for restless legs syndrome on chromosome 12q. Am. J. Hum. Genet., 69(6): 12661270. Diagnostic Classification Steering Committee of the American Sleep Disorders Association (MJ Thorpy (Chair.j) (1997) The international classification of sleep disorders: diagnostic and coding manual. Rochester, Minnesota: American Sleep Disorders Association. Droste, DW, Krauss, JK, Hagedorn, G and Kaps, M (1996) Periodic leg movements are part of the B-wave rhythm and the cyclic alternating pattern. Acta. Neurol. Scand., 94(5): 347-352. Dzvonik, ML, Kripke, DF, Klauber, M and Ancoli-Israel, S (1986) Body position changes and periodic movements in sleep. Sleep, 9(4): 484-491. Earley, CJ, Yaffee, JP and Allen, RP (1999) Randomized, double-blind, placebo-controlled trial of pergolide in restless legs syndrome. Neurology, 51(6): 1599-1602. Earley, CJ, Allen, RP, Beard, JL and Connor, JR (2000a) Insight into the pathophysiology of restless legs syndrome. J. Neurosci. Res., 62(5): 623-628. Earley, CJ, Connor, JR, Beard, JL, Malecki, EA, Epstein, DK and Allen, RP (2000b) Abnormalities in CSF concentrations of ferritin and transferrin in restless legs syndrome. Neurology, 54(8): 1698-1700. Eisensehr, I, Wetter, TC, Linke, R, Noachtar, S, Von Lindeiner, H, Gildehaus, FJ et a!. Normal IPT and IBZM SPECT in drug-naive and levodopa-treated idiopathic restless legs syndrome. Neurology, 57(7): 1307-1309. Ekbom, KA (1945) Restless legs: a clinical study. Acta. Med. Scand., 158 (Supp!.): 1-122. Ekbom, KA (1950) Restless legs. A report of 70 new cases. Acta. Med. Scand., 246 (Supp!.): 64-68.

W. HENING

Ekbom, KA (1960) Restless legs syndrome. Neurology, 10: 868-873. Entezari-Taher, M, Singleton, JR, Jones, CR, Meekins, G, Petajan, JH and Smith, AG (1999) Changes in excitability of motor cortical circuitry in primary restless legs syndrome. Neurology, 53(6): 1201-1205. Ferini-Strambi, L, Bianchi, A, Zucconi, M, Oldani, A, Castronovo, V and Smirne, S (2000) The impact of cyclic alternating pattern on heart rate variability during sleep in healthy young adults. Clin. Neurophysiol., 111(I): 99-101. Ferri, R, Parrino, L, Smerieri, A, Terzano, MG, Elia, M, Musumeci, SA et a!. (2000) Cyclic alternating pattern and spectral analysis of heart rate variability during normal sleep. J. Sleep Res., 9(1): 13-18. Fouillet, N, Wiart, L, Arne, P,Alaoui, P, Petit, Hand Barat, M (1995) Propriospinal myoclonus in tetraplegic patients: clinical, electrophysiological and therapeutic aspects. Paraplegia, 33(11): 678-681. Gelman, N, Gorell, JM, Barker, PB, Savage, RM, Spickler, EM, Windham, JP et a!. (1999) MR imaging of human brain at 3.0 T: preliminary report on transverse relaxation rates and relation to estimated iron content. Radiology, 210(3): 759-767. Godbout, R, Montplaisir, J, Poirier, G and Bedard, MA (1988) Distinctive electrographic manifestations of periodic leg movements during sleep in narcoleptic vs. insomniac patients. Sleep Res., 17: 182. Goodman, JD, Brodie, C and Ayida, GA (1988) Restless legs syndrome in pregnancy. Brit. Med. J., 297: 1101-1102. Gorny, S, Allen, R, Krausman, D and Earley, C (1998) Accuracy of the PAM-RL system for automated detection of periodic leg movements (Abstract). Sleep, 21 (Supp!.): S183. Gorny, SW, Allen, RP, Krausman, DT and Cammarata, J (2001) Initial demonstration of the accuracy and utility of an ambulatory, three-dimensional body position monitor with normals, sleepwalkers and restless legs patients. Sleep Med., 2(2): 135-143. Grillner, S (1985) Neurobiological bases of rhythmic motor acts in vertebrates. Science, 228: 143-149. Grillner, S and Dubuc, R (1988) Control of locomotion in vertebrates: spinal and supraspinal mechanisms. Adv. Neurol., 47: 425--453. Gupta, P, Hening, W, Rahman, K, Walters, A and Chokroverty, S (1996) Periodic limb movements (PLMs) in a patient with multiple sclerosis and central sleep apnea: independent right and left leg movement periods suggest lateralized PLM oscillators (abstract). Sleep Res., 25: 417. Hening, W, Saines, N and Lovelace, R (1987) Is parkinsonian tremor generated by autonomous seg-

RESTLESS LEGS SYNDROME AND PERIODIC LIMB MOVEMENTS

mental oscillators (abstract)? Neurology, 37 (Supp!. 1): 362. Hening, WA (2002) Restless legs syndrome: a sensorimotor disorder of sleep/wake motor regulation. Curro Neurol. Neurosci. Rep., 2(2): 186-196. Hening, WA, Walters, AS, Chokroverty, S and Truong, D (1989) Are there dual oscillators producing dyskinesias of the arms and legs in the restless legs syndrome (RLS) (abstract)? Muscle Nerve, 12: 751. Hening, WA, Allen, R, Earley, C, Kushida, C, Picchietti, D and Silber, M (l999a) The treatment of restless legs syndrome and periodic limb movement disorder. an American academy of sleep medicine review. Sleep, 22(7): 970-999. Hening, WA, Allen, R, Walters, AS and Chokroverty, S (I 999b ) Motor functions and dysfunctions of sleep. In: S Chokroverty (Ed.), Sleep Disorders Medicine (2nd ed.). Boston: Butterworth-Heinemann, pp. 441-507. Hening, WA, Walters, AS, Wagner, M, Rosen, R, Chen, V, Kim, S et a!. (I 999c ) Circadian rhythm of motor restlessness and sensory symptoms in the idiopathic restless legs syndrome. Sleep, 22(7): 901-912. Hening, WA, Allen, RP, Earley, CJ, Picchietti, DL and Silber, M (2004) An update on the doparninergic treatment of restless legs syndrome and periodic limb movement disorder: an American academy of sleep medicine interim review. Sleep, 25: in press. Hui, D, Wong, T, Li, T, Ko, F, Choy, D, Szeto, C et a!. (2002) Prevalence of sleep disturbances in Chinese patients with End Stage Renal Failure on maintenance hemodialysis. Med. Sci. Monit., 8(5): CR331-336. Hunker, CJ and Abbs, JH (1990) Uniform frequency of parkinsonian resting tremor in the lips, jaw, tongue, and index finger. Mov. Disord., 5(1): 71-77. Hutchison, WD, Lozano, AM, Tasker, RR, Lang, AE and Dostrovsky, JO. Identification and characterization of neurons with tremor-frequency activity in human globus pallidus. Exp. Brain Res., 113(3): 557-563. Inoue, Y, Nanba, K, Honda, Y, Takahashi, Y and Arai, H (2002) Subjective sleep quality and suggested immobilization test in restless leg syndrome and periodic limb movement disorder. Psychiatry Clin. Neurosci., 56(3): 293-294. Jean-Louis, G, Kripke, DF, Mason, WJ, Elliott, JA and Youngstedt, SD (2001) Sleep estimation from wrist movement quantified by different actigraphic modalities. 1. Neurosci. Meth., 105(2): 185-191. Kaartinen, J, Erkinjutti, M and Rauhala, E (1996) Automatic SCSB analysis of motor and autonomic nervous functions compared with sleep stages. NeuroReport, 7(5): 1102-1106. Kazenwadel, J, Pollmacher, T, Trenkwalder, C, Oertel, WH, Kohnen, R, Kunzel, M et a!' (1995) New

597

actigraphic assessment method for periodic leg movements (PLM). Sleep, 18(8): 689-697. Lapierre, 0 and Montplaisir, J (1992) Polysomnographic features of REM sleep behavior disorder: development of a scoring method (see comments). Neurology, 42(7): 1371-1374. Lavigne, GJ and Montplaisir, JY (1994) Restless legs syndrome and sleep bruxism: prevalence and association among Canadians. Sleep, 17(8): 739-743. Lee, KA, Zaffk:e, ME and Baratte-Beebe, K (2001) Restless legs syndrome and sleep disturbance during pregnancy: the role of folate and iron. J. Women's Health Gend.-Based Med., 10(4): 335-341. Lu, L, Tamura, T and Togawa, T (1999) Detection of body movements during sleep by monitoring of bed temperature. Physiol. Meas., 20(2): 137-148. Lugaresi, E, Tassinari, CA, Coccagna, G and Ambrosetto, C (1965) Particularities cliniques et polygraphiques du syndrome d'impatience des membres inferieurs. Rev. Neurol., 113: 545-555. Lugaresi, E, Coccagna, G, Berti Ceroni, G and Ambrosetto, C (1968) Restless legs syndrome and nocturnal myoclonus. In: H Gastaut, E Lugaresi, G Berti Ceroni and G Coccagna (Eds.), The Abnormalities of Sleep in Man. Bologna: Aulo Gaggi Editore, pp. 285-294. Lugaresi, E, Coccagna, G, Montovani, M and Lebrun, R (1972) Some periodic phenomena arising during drowsiness and sleep in man. Electroencephalogr. Clin. Neurophysiol., 32: 701-705. Lugaresi, E, Cirignotta, F, Coccagna, G and Montagna, P (1986) Nocturnal myoclonus and restless legs syndrome. Adv. Neurol., 43: 295-307. Matsumoto, M, Miyagishi, T, Sack, RL, Hughes, RJ, Blood, ML and Lewy, AJ (1998) Evaluation of the Actillume wrist actigraphy monitor in the detection of sleeping and waking. Psychiatry Clin. Neurosci., 52(2): 160-161. Mendelson, WB (1996) Are periodic leg movements associated with clinical sleep disturbance? (see comments). Sleep, 19(3): 219-223. Michaud, M, Chabli, A, Lavigne, G and Montplaisir, J (2000) Arm restlessness in patients with restless legs syndrome. Mov. Disord., 15(2): 289-293. Michaud, M, Poirier, G, Lavigne, G and Montplaisir, J (2001) Restless Legs Syndrome: scoring criteria for leg movements recorded during the suggested immobilization test. Sleep Med., 2(4): 317-321. Michaud, M, Lavigne, G, Desautels, A, Poirier, G and Montplaisir, J (2002a) Effects of immobility on sensory and motor symptoms of restless legs syndrome. Mov. Disord., 17(1): 112-115. Michaud, M, Paquet, J, Lavigne, G, Desautels, A and Montplaisir, J (2002b) Sleep laboratory diagnosis of restless legs syndrome. Eur. Neurol., 48(2): 108-113.

598 Michaud, M, Soucy, JP, Chabli, A, Lavigne, G and Montplaisir, J (2002c) SPECT imaging of striatal preand postsynaptic dopaminergic status in restless legs syndrome with periodic leg movements in sleep. J. Neurol., 249(2): 164-170. Montplaisir, J, Lorrain, D and Godbout, R (1991) Restless legs syndrome and periodic leg movements in sleep: the primary role of dopaminergic mechanism. Eur. Neurol., 31(1): 41-43. Montplaisir, J, Boucher, S, Gosselin, A, Poirier, G and Lavigne, G (1996) Persistence of repetitive EEG arousals (K-alpha complexes) in RLS patients treated with L-DOPA. Sleep, 19(3): 196-199. Montplaisir, J, Boucher, S, Nicolas, A, Lesperance, P, Gosselin, A, Rompre, P et al. (1998) Immobilization tests and periodic leg movements in sleep for the diagnosis of restless leg syndrome. Mov. Disord., 13(2): 324-329. Montplaisir, J, Nicolas, A, Godbout, R and Walters, A (2000) Restless legs syndrome and periodic limb movement disorder. In: MH Kryger, T Roth and WC Dement (Eds.), Principles and Practice of Sleep Medicine (3rd ed.). Philadelphia: WB Saunders, pp. 742-752. Montplaisir, J, Michaud, M, Denesle, R and Gosselin, A (2001) Periodic leg movements are not more prevalent in insomnia or hypersomnia but are specifically associated with sleep disorders involving a dopaminergic impairment. Sleep Med., 1(2): 163-167. Mosko, SS and Nudleman, KL (1986) Somatosensory and brainstem auditory evoked responses in sleep-related periodic leg movements. Sleep, 9(3): 399-404. Mosko, SS, Dickel, MJ, Paul, T, LaTour, T, Dhillon, S, Ghanim, A et al. (1988) Sleep apnea and sleep-related periodic leg movements in community resident seniors. 1. Am. Geriatr. Soc., 36(6): 502-508. Nicolas, A, Lesperance, P and Montplaisir, J (1998) Is excessive daytime sleepiness with periodic leg movements during sleep a specific diagnostic category? Eur. Neurol., 40(1): 22-26. Nicolas, A, Michaud, M, Lavigne, G and Montplaisir, J (1999) The influence of sex, age and sleep/wake state on characteristics of periodic leg movements in restless legs syndrome patients. Clin. Neurophysiol., 110(7): 1168-1174. Nordlander, NB (1953) Therapy in restless legs. Acta. Med. Scand., 145: 453-457. O'Keeffe, ST, Noel, J and Lavan, IN (1993) Restless legs syndrome in the elderly. Postgrad. Med. J., 69(815): 701-703. O'Keeffe, ST, Gavin, K and Lavan, IN (1994) Iron status and restless legs syndrome in the elderly. Age Ageing, 23(3); 200-203.

W. HENING

Ondo, WG, He, Y, Rajasekaran, S and Le, WD (2000) Clinical correlates of 6-hydroxydopamine injections into All dopaminergic neurons in rats: a possible model for restless legs syndrome. Mov. Disord., 15(1): 154-158. Parrino, L, Boselli, M, Buccino, GP, Spaggiari, MC, Di Giovanni, G and Terzano, MG (1996) The cyclic alternating pattern plays a gate-control on periodic limb movements during non-rapid eye movement sleep. J. Clin. Neurophysiol., 13(4): 314-323. Parrino, L, Smerieri, A, Rossi, M and Terzano, MG (2001) Relationship of slow and rapid EEG components of CAP to ASDA arousals in normal sleep. Sleep, 24(8): 881-885. Penzel, T, Brandenburg, D, Peter, JH, Otto, R, Hundemer, HP, Lledo, A et al. (2002) A new design of a polysomnography-based multi-center treatment study for the restless legs syndrome. Clin. Neurophysiol., 113(4): 571-578. Phillips, B, Young, T, Finn, L, Asher, K, Hening, WA and Purvis, C (2000) Epidemiology of Restless Legs Symptoms in Adults. Arch. Intern. Med., 160(14): 2137-2141. Pollmacher, T and Schulz, H (1993) Periodic leg movements (PLM): their relationship to sleep stages. Sleep, 16(6): 572-577. Polo-Kantola, P, Rauhala, E, Erkkola, R, Irjala, K and Polo, 0 (2001) Estrogen replacement therapy and nocturnal periodic limb movements: a randomized controlled trial. Obstet. Gynecol., 97(4): 548-554. Provini, F, Vetrugno, R, Meletti, S, Plazzi, G, Solieri, L, Lugaresi, E et al. (200 I) Motor pattern of periodic limb movements during sleep. Neurology, 57(2): 300-304. Rothdach, AJ, Trenkwalder, C, Haberstock, J, Keil, D and Berger, K (2000) Prevalence and risk factors of RLS in an elderly population: the MEMO study. Memory and Morbidity in Augsburg Elderly. Neurology, 54(5): 1064-1068. Rothwell, JC (1998) Physiology and anatomy of possible oscillators in the central nervous system. Mov. Disord., 13 (Suppl. 3): 24-28. Ruottinen, HM, Partinen, M, Hublin, C, Bergman, J, Haaparanta, M, Solin, 0 et al. (2000) An FDOPA PET study in patients with periodic limb movement disorder and restless legs syndrome. Neurology, 54(2): 502504. Sadeh, A, Hauri, PJ, Kripke, DF and Lavie, P (1995) The role of actigraphy in the evaluation of sleep disorders. Sleep, 18(4): 288-302. Salmi, T and Leinonen, L (1986) Automatic analysis of sleep records with static charge sensitive bed. Electroencephalogr. Clin. Neurophysiol., 64: 84-87. San Pedro, EC, Mountz, JM, Mountz, JD, Liu, HG, Katholi, CR and Deutsch, G (1998) Familial painful

RESTLESS LEGS SYNDROME AND PERIODIC LIMB MOVEMENTS

restless legs syndrome correlates with pain dependent variation of blood flow to the caudate, thalamus, and anterior cingulate gyrus. J. Rheumatol., 25(11): 2270-2275. Sforza, E, Juony, C and Ibanez, V (2002) Time-dependent variation in cerebral and autonomic activity during periodic leg movements in sleep: implications for arousal mechanisms. Clin. Neurophysiol., 113(6): 883891. Smith, RC (1985) Relationship of periodic movements in sleep (nocturnal myoclonus) and the Babinski sign. Sleep, 8: 239-243. Smith, RC (1987) Comparison of Babinski-like response in periodic movements in sleep (Nocturnal Myoclonus). BioI. Psychiatry, 22: 1271-1273. Staedt, J, Stoppe, G, Kogler, A, Riemann, H, Hajak, G, Munz, DL et a!. (1995) Nocturnal myoclonus syndrome (periodic movements in sleep) related to central dopamine D2-receptor alteration. Eur. Arch. Psychiatry Clin. Neurosci., 245(1): 8-10. Sun, ER, Chen, CA, Ho, G, Earley, CJ and Allen, RP (1998) Iron and the restless legs syndrome. Sleep, 21(4): 371-377. Symonds, CP (1953) Nocturnal myoclonus. J. Neurol. Neurosurg. Psychiatry, 16: 166-171. Tamura, T, Miyasako, S, Ogawa, M, Togawa, T and Fujimoto, T (1999) Assessment of bed temperature monitoring for detecting body movement during sleep: comparison with simultaneous video image recording and actigraphy. Med. Eng. Phys., 21(1): 1-8. Telstad, W, Serensen, 0, Larsen, S, Lillevold, PE, Stensrud, P and Nyberg-Hansen, R (1984) Treatment of the restless legs syndrome with carbamazepine: a double blind study. Br. Med J., 288: 444-446. Tergau, F, Wischer, S and Paulus, W (1999) Motor system excitability in patients with restless legs syndrome. Neurology, 52(5): 1060-1063. Terzano, MG, Mancia, D, Salati, MR, Costani, G, Decembrino, A and Parrino, L (1985) The cyclic alternating pattern as a physiological component of normal NREM sleep. Sleep, 8: 137-145. Terzano, MG, Parrino, L, Boselli, M, Smerieri, A and Spaggiari, MC (2000) CAP components and EEG synchronization in the first 3 sleep cycles. Clin. Neurophysiol., 111(2): 283-290. Terzano, MG, Parrino, L, Sherieri, A, Chervin, R, Chokroverty, S, Guilleminault, C et a!. (2001) Atlas, rules, and recording techniques for the scoring of cyclic alternating pattern (CAP) in human sleep. Sleep Med., 2: 537-555. The International Restless Legs Syndrome Study Group (2003) Validation of the international restless legs syndrome study group rating scale for restless legs syndrome. Sleep Med., 4: 121-132.

599

Trenkwalder, C, Bucher, SF and Oertel, WH (1996) Electrophysiological pattern of involuntary limb movements in the restless legs syndrome. Muscle Nerve, 19(2): 155-162. Trenkwalder, C, Hening, WA, Walters, AS, Campbell, SS, Rahman, K and Chokroverty, S (1999a) Circadian rhythm of periodic limb movements and sensory symptoms of restless legs syndrome. Mov. Disord., 14(1): 102-110. Trenkwalder, C, Walters, AS, Hening, WA, Chokroverty, S, Antonini, A, Dhawan, V et a!. (1999b) Positron emission tomographic studies in restless legs syndrome. Mov. Disord., 14(1): 141-145. Trenkwalder, C, Brandenburg, U, Hunderner, H-P, Lledo, A, Quail, D, Swieca, J et a!. (2001) A randomized longterm placebo-controlled multicenter trial of pergolide in the treatment of RLS - the PEARLS-Study (abstract). Neurology, 56 (Supp!. 3): A5. Tribl, GG, Asenbaum, S, Klosch, G, Mayer, K, Bonelli, RM, Auff, E et al. (2002) Normal IPT and IBZM SPECT in drug naive and levodopa-treated idiopathic restless legs syndrome. Neurology, 59(4): 649-650. Tryon, WW (1991) Activity measurement in psychology and medicine. New York: Plenum. Turjanski, N, Lees, AJ and Brooks, DJ. Striatal dopaminergic function in restless legs syndrome: 18F-DOPA and 1lC-raclopride PET studies. Neurology, 52(5): 932-937. Ulfberg, J, Nystrom, B, Carter, N and Edling, C (200Ia) Restless legs syndrome among working-aged women. Eur. Neurol., 46(1): 17-19. Ulfberg, J, Nystrom, B, Carter, N and Edling, C (2001b) Prevalence of restless legs syndrome among men aged 18 to 64 years: an association with somatic disease and neuropsychiatric symptoms. Mov. Disord., 16(6): 1159-1163. Walters, AS and Hening, W (1987) Review of the clinical presentation and neuropharmacology of restless legs syndrome. Clin. Neuropharmacol., 10: 225-237, Erratum 482. Walters, A, Hening, W, Kavey, N, Gidro-Frank, S, Cote, L and Fahn, S (1984) Restless Legs Syndrome: a pleomorphic sensorimotor disorder (abstract). Neurology, 34 (Supp!. I): 129. Walters, A, Hening, W, Cote, Land Fahn, S (1986) Dominantly inherited restless legs with myoclonus and periodic movements of sleep: a syndrome related to the endogenous opiates? Adv. Neurol., 43(10): 309-319. Wechsler, LR, Stakes, JW, Shahani, BT and Busis, NA (1986) Periodic leg movements of sleep (nocturnal myoclonus): an electrophysiological study. Ann. Neurol., 19: 168-173.

600 Willis, T (1672) De animae brutorum. London: Wells and Scott. Winkelman, JW (1999) The evoked heart rate response to periodic leg movements of sleep. Sleep, 22(5): 575580. Winkelman, JW, Chertow, GM and Lazarus, JM (1996) Restless legs syndrome in end-stage renal disease. Am. J. Kidney Dis., 28(3): 372-378. Winkelmann, J, Wetter, TC, Collado-Seidel, V, Gasser, T, Dichgans, M, Yassouridis, A, et al. (2000) Clinical characteristics and frequency of the hereditary restless legs syndrome in a population of 300 patients. Sleep, 23(5): 597-602. Winkelmann, J, Schadrack, J, Wetter, TC, Zieglgansberger, Wand Trenkwalder, C (2001) Opioid and dopamine antagonist drug challenges in untreated

W. HENING

restless legs syndrome patients. Sleep Med., 2(1): 57-61. Winkelmann, J, Muller-Myhsok, B, Wittchen, HU, Hock, B, Prager, M, Pfister, H et al. (2002) Complex segregation analysis of restless legs syndrome provides evidence for an autosomal dominant mode of inheritance in early age at onset families. Ann. Neurol., 52(3): 297-302. Yasuda, T, Nishimura, A, Katsuki, Y and Tsuji, Y (1986) Restless legs syndrome treated successfully by kidney transplantation - A case report. Clin. Transplants, 138. Ziemann, U, Tergau, F, Bruns, D, Baudewig, J and Paulus, W (1997) Changes in human motor cortex excitability induced by doparninergic and anti-doparninergic drugs. Electroencephalogr. Clin. Neurophysiol., 105: 43Q-437.

Movement Disorders Handbook of Clinical Neurophysiology, Vol. 1 M. Hallett (Ed.) © 2003 Elsevier B.V. All rights reserved

601 CHAPTER 37

Hemiparesis Paolo Maria Rossinia,b,c,* and Flavia Pauri'" a Cattedra di Neurologia, Universita Campus Bio-Medico, Rome, Italy AFaR. Dip. Neuroscienze. Ospedale Fatebenefratelli Isola Tiberina, Rome. Italy c IRCCS Centro S. Giovanni di Dio, Fatebenefratelli, Brescia, Italy d Dipartimento di Neurologia e Otorinolaringoiatria, Universita La Sapienza, Rome. Italy b

Stroke damages a region of the brain and alters the function of areas adjacent to or distant from the lesion through neuronal networks. The most popular term for these remote effects of stroke is "diaschisis", which implies that acute neuronal failure in the ischemic area induces remote modulatory effects on neuronal function, blood flow, metabolism and cortical excitability of unaffected areas of the same hemisphere via cortico-cortical connections and of the contralateral unaffected hemisphere (UH) via transcallosal fibers (Andrews, 1991; Pappata et al., 1993; Conti and Manzoni, 1994). Examples of such a mechanism are the metabolic enhancement followed by progressive depression of the UH, as documented by PET and SPECT in 'acute' and 'subacute' post-stroke epochs (Baron et al., 1981; Powers and Reichle, 1987). A large number of studies focused on neuronal reorganization following stroke deal particularly with the brain sensorimotor areas devoted to hand control; this is because of the large hand representation which makes it easier to disentangle even minor changes in the homuncular topography, and the importance of hand control for the quality of everyday life. The recovery of hand sensorimotor function after a cerebrovascular lesion can be ascribed to several mechanisms, including short-term phenomena that resolve in few days like reabsorption of perilesional edema, and recovery from the functional block of still-living neurons and fibers, but also to the site and extension of the damaged area, interindividual

* Correspondence to: Prof. Paolo Maria Rossini, Direzione Scientifica AFaR, Associazione Fatebenefratelli per la Ricerca, Lungotevere degli Anguillara 12, 00186 Rome, Italy. E-mail address:[email protected] Tel.: +39066837300; fax: +39066837360.

variability of the territory of the MCA perfusion, amount and extent of collaterals and neuronal pools receiving blood supply from adjacent arteries, multiple representations of the same muscles in separate clusters of cortical motoneurons in primary motor cortex (MI), and presence and amount of ipsilateral corticospinal fibers (Rossini, 2001; Rossini and Liepert, 2002). It is, indeed, a common experience that a slow, but consistent recovery of the neurological deficits takes place in the weeks and months following a stroke (Twitchell, 1951). Such a recovery can greatly vary in a range spanning from little to great functional impact even in patients with apparently identical clinical pictures in the early stroke stages. Neurophysiological techniques - including advanced EEG, magnetoencephalography (MEG), transcranial magnetic stimulation (TMS) - are able to provide an excellent time discrimination of the examined function down to fractions of milliseconds, distinguish excitatory and inhibitory phenomena and provide an indirect measure of the strength of the contributing brain generators in a given circuitry. The EEG provides useful information about the localization of acute cerebral ischemia, but recording densities of 64 channels or higher are required for accurate spatial characterization of focal stroke-related EEG changes. To determine loss of spatial and clinical information resulting from spatial undersampling, Luu et al. (2001) subsampled data obtained from a 128-channel recording montage into 64-, 32-, and, 19-channel arrays. They found that well-defined topography of stroke-related EEG changes is dependent on adequate spatial sampling density and that accurate description was achieved only with the 64- and 128-channel EEG. In fact, if the number of recording channels decreases to 32 or lower, the

602

distribution of the scalp EEG spectra is distorted, leading to a mislocalization of the affected region. MEG and TMS have shown a high percentage of subjects for whom sensorimotor hand somatotopy was reorganized following stroke in the affected hemisphere (see Rossini and Pauri, 2000; Rossini and Liepert, 2002 for reviews). This observation is possible since many studies demonstrated that in healthy persons most of the neurophysiological characteristics of the hand sensorimotor primary areas are symmetrical in the two hemispheres (Rossini et aI., 1994; Cicinelli et aI., 1997; Wikstrom et aI., 1997). This is different from the posterior parietal areas devoted to sensory integration and to uni- vs. bimanual sensory perception which react quite asymmetrically to TMS of each hemisphere (Oliveri et aI., 1999, 2001). For TMS the measures which have been found to be symmetrical on the two hemispheres include: threshold of excitability during rest, latency of motor evoked potentials (MEPs), central conduction time (CCT), number and location of the excitable scalp sites with respect to anatomical landmarks (Cicinelli et aI., 1997), center of gravity of the motor maps, and recovery curves to paired stimuli reflecting intracortical facilitatory (ICF) and inhibitory (ICI) mechanisms (Kujiraj et aI., 1993; Shimizu et aI., 1999; Cicinelli et aI., 2000). Regarding the sensory area the following were found to be symmetrical: locations, depth, latency and strength of the dipoles responsible for somatosensory responses from SI after median nerve, or little and thumb fingers stimulation, as well as the extent of the hand representation measured as the euclidean distance between the 5th and thumb finger baricenters of the equivalent current dipoles (ECO) both with respect to skull and brain (0. or e-shaped knob of the central sulcus) landmarks (Pizzella et al., 1999). It is worth remembering that the morphology of brain responses to peripheral stimuli reflects the cerebral circuitry that shapes the final response in time (peak latency), firing rate and number of synchronously firing neurons (peak amplitude) and excitatory/inhibitory net effect (peak polarity). Absolute waveforms are extremely variable across different subjects, specifically when wave polarity and shape are considered. However, intrasubject interhemispheric differences are extremely low with a correlation coefficient which - in the healthy - is close to 1 (=perfect identity; Tecchio et al., 2000). The spatial sensory coordinates of the finger ECOs

P.M. ROSSINI AND F. PAURI

are consistent with the known sensory homunculus somatotopy with little finger more medial and posterior, thumb more lateral and anterior and median nerve in between; moreover, they are strictly adjacent to the knob of the central sulcus containing the hand motor area (Yousry et aI., 1997; Pizzella et aI., 1999). The hand extension in the healthy ranges between 12 and 17 mm. The hand motor output area shows an average of 3.5 excitable sites during TMS on each hemisphere (Fig. 1) within a 2 em spaced grid (Cicinelli et aI., 1997). All these neurophysiological measures are stable when repeatedly measured over time in the same healthy subject (Cicinelli et aI., 1997; Tecchio et al., 1998). Moreover, intersubject variability of their interhemispheric differences vary relatively little, differently from absolute values (Rossini et aI., 1994; Wikstrom et aI., 1997). Therefore, it is easily understandable that a monohemispheric stroke affecting the hand sensorimotor areas, when followed by any 'plastic' reorganization will remarkably affect many of these neurophysiological measures, in particular their interhemispheric symmetry. An increasing number of studies has been devoted to patients with monohemispheric stroke, with the aim of evaluating the long- and short-term effects on cortical sensorimotor organization and their interhemispheric differences. Such an approach aims to provide important tools to elucidate the problem, whether clinical recovery of sensory and motor

• •

•

• •

•

•

•

•

• •

•

•• • ••• •

0-20% 21- 50% 51-80% 81-100%

•

Fig. 1. Somatotopic representation of ADM muscle in the healthy population, showing a symmetrical cortical arrangement in both right and left hemisphere. (From: P. Cicinelli et al. (1997) Post-stroke reorganization of brain motor output to the hand: a 2-4 month follow-up with focal magnetic transcranial stimulation. Electroencephalogr. Clin. Neurophysiol., 105: 438-450, with permission.)

HEMIPARESIS

functions is based upon the re-establishment of previously damaged - but not destroyed - corticospinal connections or upon "plastic" rearrangements of cortical somatotopy, in which previously functionally silent or differently operating cortico-spinal fibers (activated in newly-established cortico-cortical connections) are taking over the lost functions. When damage to a functional system is partial, a within-system recovery is possible, whereas after complete destruction, substitution by functionally related systems remains the only alternative (Seitz and Freund, 1997). The importance of the pyramidal fiber contingent for motor function - mainly upper limb and hand - is well known; experimental and clinical studies have shown that approximately onefifth of the pyramidal fibers are sufficient to ensure restitution of fractionated hand finger movement. Hence, a within-pyramidal system reorganization when possible - is a major candidate for functional recovery of motor control of upper limb and hand (Jane et al., 1968; Lawrence and Kuypers, 1968a, b; Warabi et al., 1990; Cicinelli et al., 1997; Traversa et al., 1997, 1998; Rossini et al., 1998a; Tecchio et al., 1998). TMS has been used as a predictor of outcome after stroke. However, results are not entirely consistent. For example, Timmerhuis et al. (1996) found that only MEPs measured in the acute stage had predictive value. In contrast, Arac et al. (1994) reported that irrespective of presence or absence of MEPs in the acute stage, patients had the same motor function 3-6 months later. Catano et al. (1996) suggested that motor threshold determination 30 days after stroke had the best correlation with outcome. Others (Di Lazzaro et al., 1999) found the same result at eight hours after stroke. In most studies TMS was performed within the first days (e.g. Heald et al., 1993; Pereon et al., 1995; Escudero et al., 1998; Pennisi et al., 1999; Yang et al., 1999) or even hours (Di Lazzaro et al., 1999) after stroke. Results obtained by TMS were compared to overall functional outcome, measured by Barthel Index or Rankin scale, or to motor recovery. Outcome evaluation was usually performed 2 to 12 months after stroke. Some studies repeated TMS to compare electrophysiological follow-up with clinical progress (Pereon et al., 1995; Catano et al., 1996; Rapisarda et al., 1996; Timmerhuis et al., 1996; D'Olhaberriague et al., 1997; Hendricks et al., 1997; Rossini et al., 1998b; Cruz Martinez et al.,

603

1999; Pennisi et al., 1999). Patients were often subdivided into groups according to the TMS results. Patients with recordable MEPs were usually compared with those without MEPs. Other investigators defined patient groups according to their clinical symptoms (D'Olhaberriague et al., 1997). In some studies patients were not grouped a priori but statistical analysis was performed to discriminate the most important variables for predicting final outcome (e.g. Feys et al., 2000). Patient populations were not homogeneous; in some studies the number of severely affected patients with poor recovery was particularly high, and in most studies cortical and subcortical infarctions as well as ischemic and and hemorrhagic lesions were mixed. The above-mentioned different methods probably contribute substantially to the different results that are encountered. Palmer et al. (1992) were unable to stimulate ipsilateral pathways (iMEPs) in stroke patients and concluded that ipsilateral corticospinal connections do not contribute to motor recovery. In other studies with stroke patients iMEPs could either be elicited by TMS over the non-lesioned hemisphere (Caramia et al., 1996, 2000; Turton et al., 1996; Hendricks et al., 1997; Netz et al., 1997; Trompetto et al., 2000) or over the affected hemisphere (Fries et al., 1991; Trompetto et al., 2000; Alagona et al., 2001). In these patients iMEPs were usually obtained by stimulating anteriorly and medially to the primary motor cortex. This probably indicates that corticospinal pathways originating from premotor and supplementary motor areas were activated. Those patients with iMEPs when stimulating the nonaffected hemisphere still had a variable outcome: some authors (Caramia et al., 1996,2000; Trompetto et al., 2000) found a correlation between iMEPs and motor recovery while others described an association between the occurrence of iMEPS and a poor outcome (Turton et al., 1996; Netz et al., 1997, Hendricks et al., 1997). An association between iMEPs produced by stimulation of the affected hemisphere (AH) and bimanual dexterity 6-months after stroke was found suggesting the existence of a hyperexcitability of premotor areas in the AH (Alagona et al., 2001). It remains unclear if clinical, pathophysiological or methodological differences are responsible for these varying observations. The study of intracortical inhibition and facilitation (ICIIICF) to paired stimuli (conditioning/test;

604

Kujirai et al., 1993) in patients with hemiplegia early after the event has shown a decrease of ICI two weeks after stroke (Liepert et aI., 2000); this may indicate a downregulation of GABA activity (Ziemann et aI., 1996; Chen et al., 1998). Two main mechanisms could be responsible: damage of transcallosal fibers which could lead to a loss of physiological interhemispheric inhibitory modulation (Ferbert et aI., 1992; Boroojerdi et al., 1996; Liepert et aI., 2001) or an enhanced use of the unaffected arm in all daily activities as ICI is modified in a task- and use-dependent manner (Liepert et al., 1998). The idea of disinhibition in the UH is also supported by the observation of MEPs of larger amplitudes (Trompetto et al., 2000) and of significantly less ICI, particularly in patients with a cortical stroke (Shimizu et al., 2002). Moreover, patients with cortical lesions showed no transcallosal inhibition in the active, unaffected hand muscles during TMS of the affected motor cortex, while this phenomenon was present in patients with subcortical lesions (Shimizu et al., 2002). Another measure, that is proven to be helpful in stroke evaluation is the Silent Period (SP; Fuhr et al., 1991; Cantello et aI., 1992; Roick et al., 1993; Triggs et aI., 1993). SP duration shows a high interindividual variability, but intraindividual interhemispheric differences are small (Haug et al., 1992; Cicinelli et al., 1997; Fritz et aI., 1997). Early after stroke the SP generally is asymmetrically prolonged on the paretic side (Haug et al., 1992; Kukowski and Haug, 1992; Braune and Fritz, 1995; Liepert et aI., 1995; Faig and Busse, 1996; Classen et al., 1997; Ahonen et al., 1998). Small lesions within the primary sensorimotor cortex are associated with a reduction of SP duration whereas patients with subcortical infarcts or lesions in premotor, parietal and temporal areas show SP prolongations on the affected side. Patients with motor disturbances that resembled motor neglect have particularly prolonged SPs (Von Giesen et aI., 1994; Classen et aI., 1997). The development of spasticity during clinical follow-up is often associated with a shortening of SP duration (Liepert et al., 1995; Catano et al., 1997a, b; Cruz Martinez et aI., 1998), thus reflecting a decreased activity of inhibitory circuits as well as an increased intraspinal excitability as testified by an exaggerated H-Reflex (Marque et al., 2001). In this respect, quantitative evaluation of muscle tone in post-stroke patients can be obtained by correlating

P.M. ROSSINI AND F. PADRI

biomechanical indices with conventional clinical scales and neurophysiological measures, therefore providing characterization of passive and neural components of muscle tone. For instance, mechanical stretches of the wrist flexor muscles can be produced by means of a torque motor at constant speed. Data have been correlated with the Ashworth scale for spasticity and the Medical Research Council score for residual muscle strength. The neurophysiological measures are usually the Hoffmann reflex latency, Hm./M m• x ratio, stretch reflex threshold speed (SRTS), stretch reflex (SR) latency and area, passive (lSI) and total (TSI) stiffness indices. Hm./M m• x ratio, SR area, lSI and TSI values are significantly higher in patients, while SRTS are significantly lower. TSI, SRTS and SR area are highly correlated to the Ashworth score. EMGbiomechanical procedures allow an objective evaluation of changes in muscle tone in post-stroke patients, providing easily measurable, quantitative indices of muscle stiffness. The linear distribution of these measures is particularly indicated for monitoring changes induced by treatment (Pisano et al., 2000). A facilitation of the transmission in the interneuronal pathway coactivated by group I and group II afferents, probably resulting from a change in their descending control in spastic hemiplegic patients was found. An impaired modulation of quadriceps tendon jerk reflex during spastic gait, studied at 16 phases of the step cycle was found. In healthy subjects the size of the reflexes was profoundly modulated and was generally depressed throughout the step cycle. In patients with cerebral lesion there was less depression of the reflex size and a reduced reflex modulation on the affected side compared with healthy subjects. On the 'unaffected' side of these patients reflex modulation was similar to that of healthy subjects, but the reflex size during gait was not significantly different from standing control values. The unilateral preservation of fiber tracts in hemiparesis could be responsible for this behavior (Faist et al., 1999). For complex motor behaviors such as locomotion, afferent input related to load and hip-joint position probably has an important role in the proprioceptive contribution to the activation pattern of the leg muscles. A substantial part of corticospinal excitation to upper limb motoneurons is mediated through cervical premotoneurons located rostral to moto-

605

HEMIPARESIS

neurons. The indirect component of the corticospinal command passing through the propriospinal relay may be updated by the extensive convergence at this level of afferent inputs (both excitatory and inhibitory) from the moving limb. Propriospinal neurons are potently inhibited by feedback inhibitory interneurons facilitated from the motor cortex. Movement disorders such as spasticity involve the defective use of afferent input in combination with secondary compensatory processes. This has implications for therapy, which should be directed to take advantage of the plasticity of the CNS (see Dietz, 2002 for a review).

37.1. Long-term post-stroke neural reorganization The analysis of TMS during the recovery stage after stroke often shows enlargement of motor cortical representation from that in the acute stage,

a)

c)

b)

Barthel Index

Rivermead

Canadian Neurological Scale

90

9.0

80

85

70

a

shift of the motor hot-spot on the affected hemisphere, enhanced activation of secondary motor areas and anomalous motor representations for the paretic hands (Traversa et al., 1997, 1998; Byrnes et al., 1999; Cramer et al., 2000; Marshall et al., 2000). When following-up the characteristics of motor output of stroke patients, recovery slopes of the various neurophysiological and clinical measures can be determined (Fig. 2) and correlated with each other. In particular, the excitability threshold progressively decreases during the follow-up only in the AH, while MEP amplitudes and latencies tend toward normal, more in the resting state than when there is a background contraction (Figs. 3 and 4). The steepest parts of the recovery slopes were concentrated in the 40 to 80 days following stroke (Traversa et al., 2000). Patients with cortical (C) and subcortical (S) stroke were investigated via MEG recordings (Ros-

12

8.0

~

60

7.5

50

40

TO

T1

T2

T3

T4

TO

T1

T2

T3

Hand Motor Score

T2

T3

T4

Ashworth scale

1.1

1.1

1.0

1.0

.9

.9 e

.8

a .8

.7

.7

.6

.6

.5 .4,---,,~

T1

e)

d)

~

TO

T4

__

.5

TO

T2

T4

Fig. 2. Follow up of clinical and functional scales in stroke patients. TO: 5 weeks after the stroke, Tl: 15 days after TO, T2: 42 days from TO, T3: 90 days from TO,T4: 120 days from TO. (From: Traversa et al. (2000) Neurophysiological follow-up of motor cortical output in stroke patients. Clini. Neurophysiol., 111: 1695-1703, with permission.)

606

eo

P.M. ROSSINIAND F. PAVRI

Excitability threshold

70

'# 60

50

e-

40

10

E9

T1

E9

12

E9~ 13

T4

•

AH

0

UH

Fig. 3. Follow-up of ADM muscle excitability threshold from the Affected (AH) and Unaffected (UH) Hemisphere. TO-T4 as in Fig. 2. (From: Traversa et al. (2000). Neurophysiological follow-up of motor cortical output in stroke patients. Clin. Neurophysiol., 111: 1695-1703, with permission.)

sini et al., 2001b): a group with cortical (C) and a group with subcortical (S) stroke. In all but one somatosensory evoked fields (SEFs) were recordable from the AH, and both the M20 and M30 components were reliably identified from all the stimulated nerve and digits. In the remaining case, the entire SEF was missing on the AH during finger stimulation. Peak latencies were always normal on the UH. The mean percentage of latency alterations on the AH was 23% after an S lesion, and 13% after a C lesion. In 78% of patients with reliable SEFs from the AH there was an excessive interhemispheric asymmetry of signal strength at least for one stimulation site. In 25% of ECD pairs from homologous cortical areas there were asymmetries exceeding the normative limits, with similar incidence in C and S patients. All ECD pairs with excessive asymmetries had the strongest ECD in the AH after a C lesion, whereas this was seen in only a quarter of cases following an S lesion. When considering ECD strength in relation with the clinical outcome, this was more frequently abnormal in patients with better clinical recovery; it is worth noting that all ECDs that were much stronger in AH respect to UH belonged to this group. An excessive asymmetry of waveshape was found in 60% of cases, and was more frequent with S than C lesions; this finding was also combined with a clearcut spatial shift of the ECD toward an area outside the

normative limits. Abnormal- wave shapes were observed in 75 and 50% of patients with and without good clinical recovery, respectively. Thumb and little finger ECDs were always positioned according to the classical homunculus topography. When present, the hand area enlargement was explained by a medial shift of the little finger ECD. Therefore, cortical reorganization as reflected by enlargement of the "hand area" took place at the expense of the forearm more than the face. All ECDs were localized outside the area of structural lesion, as evaluated by MEGIMRI integration. When measuring the extension of the sensory "hand area", the mean percentage of altered measures was 20% of Sand 13% of C patients. The mean "hand area" length in the abnormal cases was 39 ± 7 mm in the S group and 27 ± 7 mm in the C patients (vs. 16± 5 mm in healthy controls); this was not significantly related with clinical outcome. Such observations were confirmed by comparing regional blood flow through PET and MEG; in seven patients with either unilateral internal carotid artery or middle cerebral artery occlusion and minor neural impairments, there was a reduction of 'early' SEFs and an augmentation of the later responses of the neurons in the primary sensory area. This showed correlation with the severity of cortical ischemia (Bundo et al., 2002). In a longitudinal study, Feydy et al. (2002) assessed the changes in cortical activation across three fMRI sessions after stroke and related them to the extent of the lesion and degree of motor recovery. They found that ipsilateral Ml and its corticospinal connections alone are not sufficient for recovery of normal hand functions. However, ipsilateral Ml may contribute to recovery via its interhemispheric or corticoreticular connections. Movement-related slow cortical potentials and event-related desynchronization of alpha (alphaERD) and beta (beta-ERD) EEG rhythms after self-paced voluntary finger movements have been studied in ischemic supratentorial stroke patients and age-matched control subjects during movement preparation and actual performance. The multimodal EEG analysis revealed specific impairment of movement-related electrical activity of the brain. The readiness potential of paretic subjects was centered more anteriorly and laterally; they also had increased beta-ERD at left lateral frontal areas, during movements, as well as reduced alpha-ERD and beta-ERD during both movement preparation and actual per-

607

HEMIPARESIS

50

a) MEP's amplitude at rest

21

c) MEP's latency at rest

2

40

20

~H 101()h-------------

TO

400

T1

T2

T3

T4

0

UH

TO

T1

T2

T3

T4

d) MEP's latency during contracti n

b) MEP's amplitude during contr ction

300

~

~ 1

200

1

6

~H 0101-.-----_--_--_

TO

T1

T2

T3

T4

0

UH

EI

1

EI

EI

e:J

~H 0

TO

T1

T2

T3

T4

UH

Fig. 4. Mean MEP amplitude at rest (a) and during contraction (b) and mean MEP amplitude at rest (c) and during contraction (d) as a function of time and hemisphere (AH, affected, UH, unaffected) TD-T4 as in previous figures. (From Traversa et al. (2000) Neurophysiological follow-up of motor cortical output in stroke patients. Clin. Neurophysiol., III: 1695-1703, with permission.)

formance. Patients with ideomotor apraxia showed more lateralized frontal movement-related slow cortical potentials during both movement preparation and performance, and reduced left parietal beta-ERD during movement preparation. It was concluded that: (I) disturbed motor efference is associated with an increased need for excitatory drive of pyramidal cells in motor and premotor areas or an attempt to drive movements through projections from these areas to brainstem motor systems during movement preparation; (2) an undisturbed somatosensory afference might contribute to the release of relevant cortical areas from their 'idling' state when movements are prepared and performed; and (3) apraxic patients have a relative lack of activity of the mesial frontal motor system and the left parietal cortex, which is believed to be part of a network subserving ideomotor praxis (Platz et al.,

2000). Furthermore intersubject differences in movement-related electric brain activity and activation of sensorimotor areas during movement preparation and deactivation of other cortical areas during movement execution seem to be factors that predict a favorable outcome. Platz et al. (2002) found that electric brain activity during movement preparation explained the variance of motor improvement scores completely. Electric brain activity during movement as well as baseline motor performance accounted each for 50% of the variance of motor improvement scores. Functional connection between the motor cortex and the muscles can also be measured by EEG-EMG coherence. EEG-EMG of the hand, forearm, and arm muscles recorded during three tonic contraction tasks in chronic subcortical stroke patients revealed that EEG-EMG coherence was localized over the

608

P.M. ROSSINIAND F. PAURI

muscles but not for the biceps muscle. The different effects of the lesion on the proximal and distal muscles appear to be associated with the strength of the corticospinal pathway, and functional connections appear to come mainly from contralateral cortical motor area (Mirna et al., 2001).

37.2. Conclusions From all the reported findings it can be inferred that primary sensorimotor and adjacent brain areas following a vascular insult undergo remarkable neuroplastic changes, mainly within the AH but also in the UH. Matching results with the clinical outcome demonstrated that patients with sensory and/or motor recovery of hand function showed

Fig. 5. Male 62 years old, ischemic small lesion in the region of the right basal ganglia; no sensory deficits and partial recovery of motor weakness. MEGIMRI integration shows the asymmetrical location of ECDs activated by fifth finger's stimulation at the two successive MEG examinations: respectively UH (b) and AH (c) at 1 month - black circle - and 24 months - black square. Such an interhemispheric asymmetry increased at 24 months. In (d), the projection of thumb and little finger ECDs (at 24 months) on the same illustrative axial section for UH and AH, note the frontal shift of both ECDs and the medial displacement of the little finger ECD, determining the "hand extension" enlargement. (From Rossini et al. (l998b) On the reorganizationof sensory hand areas after mono-hemispheric lesion: a functional (MEG)/anatomical (MRI) integrative study. Brain Res., 782: 153-166, with permission.) contralateral sensorimotor area in all circumstances. No significant coherence was seen at the ipsilateral side. EEG-EMG coherence was significantly smaller on the affected side for the hand and forearm

MEDIAN NERVE

Fig. 6. Ischemic stroke involving the parietal lobe: partial motor recovery. Note the amplitude increase expecially for the M30 component, of the response after the median nerve stimulationin the affected hemisphere (AH), respect to the unaffected one (UH). Waveshape analysis in this case gave a similarity between the two hemisphere of 0.71 (out of the normality range). (From Rossini et al. (2001) Interhemispheric differences of sensory hand areas after monohemispheric stroke: MEG/MRI integrative study. Neuroimage, 14: 474-485, with permission.)

609

HEMIPARESIS

"U nbalancing" A.H.

U.H.

G.V. 62 yrs male

T1

T1

IIIIVL

ll11VL

20ms

20

T2

lOS

T2

,mvL

,mvL 20ms

20 rns

"Balancing" A.A. 30 yrs

A.H.

U.H.

female

T1

T1

,mvL

llmvL 20ms

20m,

T2

T2

,mvL 20ms

,mvL 20 illS

Fig. 7. "Unbalancing" phenomenon of hemispheric excitability: Hemiplegic patient with a left subcortical lesion. No evident clinical recovery between Tl (about 2 months after the stroke) and T2 (8 weeks after Tl). MEPs from the AH were absent both in Tl and T2, whereas MEPs from the UH were increasingly larger than normal both in Tl and T2. "Balancing" right hemiparetic patient, with a cortical lesion. Clinical improvement between Tl and T2. Note that MEP amplitude decreases from UH, and increases from AH during the follow-up. (From Traversa R et aI. (1998) Follow-up of interhemispheric differences of motor evoked potentials from the 'affected' and 'unaffected' hemispheres in human stroke. Brain Res., 803: 1-8, with permission.)

610

more interhemispheric asymmetry than patients with no recovery. This possibly reflects the cortical reorganization of the hand sensorimotor area as a mechanism underlying the functional recovery, and suggests that the reorganization of neural circuits for primary sensorimotor control of the paretic hand favor functional recovery. When the damage affects the core of the sensorimotor areas, reorganization is absolutely crucial for functional recovery; meanwhile, reorganization outside the usual boundaries never reaches the same functional efficiency of the original circuits. Interhemispheric asymmetries were more frequently encountered in the 'subcortical' than in the 'cortical' type of lesions (Fig. 5; Rossini et al., 1998b). It is noteworthy that larger than normal amplitudes of somatosensory evoked responses with an asymmetrical waveshape were found frequently from the AH, especially following a cortical lesion (Fig. 6; Wikstrom et al., 2000; Rossini et al., 2001). A somewhat similar pattern of abnormality was seen in patients showing larger than normal MEPs during TMS of the unaffected hemisphere; this was usually combined with absent response from the AH and was a bad prognostic index. In fact, whenever improvement at follow-up was seen, a progressive balancing between the two hemispheres was observed; that is an increment of the response amplitude from AH and a decrement of the 'giant' response from the UH (Fig. 7; Traversa et al., 1998). Such an asymmetrical and unbalanced hyperreactivity of the AH and UH have been ascribed to transient or permanent loss of transcallosal fiber modulation; whether and to which extent this has to do with clinical recovery, plastic reorganization and post-stroke epilepsy remains to be elucidated (Glassmann, 1974; Glassmann and Malamut, 1976; Buchkremer-Ratzmann et al., 1996; Traversa et al., 1998; Wikstrom et al., 2000; Rossini et al., 2001; Shimizu et al., 2002). As summarized, functional reorganization of the motor output following an hemispheric stroke has been repeatedly reported, often with an excessive asymmetry of the sensorimotor hand topography between the AH and the UH. "Map migration" is usually on the medio-lateral axis, but an anteroposterior shift of up to several mms of the map's center of gravity has also been observed. This has been particularly true in patients in the chronic stages of brain lesions and represented a dynamic phoenomenon during follow-up.

P.M. ROSSINI AND F. PAURI

Besides research activity, in everyday clinical practice 'early' (in the acute stage, within the first 2-3 days from stroke) analysis of hemispheric responsiveness examined via TMS and MEG (or high-resolution EEG) provide useful clinical information: absent responses are a bad indicator for long-term outcome (both for survival and recovery of lost functions), while present responses suggest a significant possibility for recovery. Follow-up of interhemispheric differences in sensorimotor maps of the hand is useful in later, subacute stages, since it can provide information on remodeling of the damaged circuits within or adjacent to the lesion, as well as on the type of rehabilitation procedure which is best for the patient (e.g. constraint-therapy when hyperexcitable and progressively enlarging maps are found on the unaffected hemisphere). Reflexology, tone and gait analysis are extremely useful in subacute and chronic stages for the evaluation of spasticity and for planning best rehabilitative and pharmacological treatment. The relationship between the amount of clinical recovery after stroke and the information gathered via neurophysiological techniques in the early (hours or days) stages aimed to find a prognostic tool of clinical evolution in the first days of the lesion needs extensive studies in large and homogeneous samples. Moreover, additional investigations are needed in following up brain excitability changes in the subacute and chronic stages (weeks, months) and in correlating them with the clinical outcome. Stroke is so disabling in its sequelae, and the way we can cure it is still so limited, that efforts aiming to understand it (and possibly modulate it via drug and rehabilitation therapy) are certainly worthwhile.

References Ahonen, JP, Jehkonen, M, Dastidar, P, Molnar, G and Hakkinen, V (1998) Cortical silent period evoked by transcranial magnetic stimulation in ischemic stroke. Electroencephalogr. Clin. Neurophysiol., 109: 224229. Alagona, G, Delvaux,V, Gerard, P,De Pasqua, V, Pennisi, G, Delwaide, PJ, Nicoletti, F and Maertens de Noordhout, A (2001) Ipsilateral motor responses to focal transcranial magnetic stimulation in healthy subjects and acute-stroke patients. Stroke, 32: 1304-1309. Andrews, RJ (1991) Transhemispheric diaschisis. Stroke, 22: 943-949.

HEMIPARESIS

Arac, N, Sagduyu, A, Binai, Sand Ertekin, C (1994) Prognostic value of transcranial magnetic stimulation in acute stroke. Stroke, 25: 2183-2186. Baron, lC, Bousser, MG, Comar, D, Soussaline, F and Castaigne, P (1981) Noninvasive tomographic study of cerebral blood flow and oxygen metabolism in vivo. Potentials, limitations, and clinical applications in cerebral ischemic disorders. Eur. Neurol., 20(3): 273-284. Boroojerdi, B, Diefenbach, K and Ferbert, A (1996) Transcallosal inhibition in cortical and subcortical cerebral vascular lesions. J. Neurol. Sci., 144: 160170. Braune, HI and Fritz, C (1995) Transcranial magnetic stimulation-evoked inhibition of voluntary muscle activity (silent period) is impaired in patients with ischemic hemispheric lesion. Stroke, 26: 550-553. Buchkremer-Ratzmann, I, August, M, Hagemann, G and Witte, OW (1996) Electrophysiological transcortical diaschisis after acute photothrombosis in rat brain. Stroke, 27: 1005-1111. Bundo, M, Inao, S, Nakamura, A, Kato, T, Ito, K, Tadokoro, M, Kabeya, R, Sugimoto, T, Kajita, Y and Yoshida, 1 (2002) Changes of neural activity correlate with the severity of cortical ischemia in patients with unilateral major cerebral artery occlusion. Stroke, 33(1): 61-66. Byrnes, ML, Thickbroom, GW, Phillips, BA, Wilson, SA and Mastaglia, FL (1999) Physiological studies of the corticomotor projection to the hand after subcortical stroke. Clin. Neurophysiol., 110: 487--498. Cantello, R, Gianelli, M, Civardi, C and Mutani, R (1992) Magnetic brain stimulation: The silent period after the motor evoked potential. Neurology, 42: 1951-1959. Caramia, MD, lani, C and Bernardi, G (1996) Cerebral plasticity after stroke as revealed by ipsilateral responses to magnetic stimulation. NeuroReport, 7: 1756-1760. Caramia, MD, Palmieri, MG, Giacomini, P, lani, C, Dally, Land Silvestrini, M (2000) Ipsilateral activation of the unaffected motor cortex in patients with hemiparetic stroke. Clin. Neurophysiol., 111: 1990-1996. Catano, A, Houa, M, Caroyer, 1M, Ducame, H and Noel, P (1996) Magnetic transcranial stimulation in acute stroke: early excitation threshold and functional prognosis. Electroencephalogr. Clin. Neurophysiol., 101: 233-239. Catano, A, Houa, M and Noel, P (l997a) Magnetic transcranial stimulation: dissociation of excitatory and inhibitory mechanisms in acute strokes. Electroencephalogr. Clin. Neurophysiol., 105: 29-36. Catano, A, Houa, M and Noel, P (l997b) Magnetic transcranial stimulation: Clinical interest of the silent

611 period in acute and chronic stages of stroke. Electroencephalogr. Clin. Neurophysiol., 105: 290-296. Chen, R, Corwell, B, Yaseen, Z, Hallett, M and Cohen, LG (1998) Mechanisms of cortical reorganization in lowerlimb amputees. J. Neurosci., 18: 3443-3450. Cicinelli, P, Traversa, R, Bassi, A, Scivoletto, G and Rossini, PM (1997) Interhemispheric differences of hand muscle representation in human motor cortex. Muscle Nerve, 20: 535-542. Cicinelli, P, Traversa, R, Oliveri, M, Palmieri, MG, Filippi, MM, Pasqualetti, P and Rossini, PM (2000) Intracortical excitatory and inhibitory phenomena to paired transcranial magnetic stimulation in helathy human subjects: differences between the right and left hemisphere. Neurosci. Lett., 288: 171-174. Classen, 1, Schnitzler, A, Binkowski, F, Werhahn, KJ, Kim, Y, Kessler, KR and Benecke, R (1997) The motor syndrome associated with exaggerated inhibition within the primary motor cortex of patients with hemiparetic stroke. Brain, 120: 605-619. Conti, F and Manzoni, T (1994) The neurotransmitters and postsynaptic actions of callosally projecting neurons. Behav. Brain Res., 64: 37-53. Cramer, SC, Moore, CI, Finklestein, SP and Rosen, BR (2000) A pilot study of somatotopic mapping after cortical infarct. Stroke, 31: 668-671. Cruz Martinez, A, Munoz, 1 and Palacios, F (1998) The muscle inhibitory period by transcranial magnetic stimulation. Study in stroke patients. Electromyogr. Clin. Neurophysiol., 38: 189-192. Cruz Martinez, A, Tejada, 1 and Dietz Tejedor, E (1999) Motor hand recovery after stroke. Prognostic yield of early transcranial magnetic stimulation. Electromyogr. Clin. Neurophysiol., 39: 405--410. Di Lazzaro, V, OIiviero, A, Profice, P, Satumo, E, Pilato, F and Tonali, P (1999) Motor cortex excitability changes within 8 hours after ischaemic stroke may predict the functional outcome. Eur. J. Emerg. Med., 6: 119-121. Dietz, V (2002) Proprioception and locomotor disorders. Nat. Rev. Neurosci., 3: 781-790. D'Olhaberriague, L, Espadaler Gamissans, 1M, Marrugat, J, Valls, A, Oliveras Ley, C and Seoane, JL (1997) Transcranial magnetic stimulation as a prognostic toolin stroke. J. Neurol. Sci., 147: 73-80. Escudero, IV, Sancho, 1, Bautista, D, Escudero, M and Lopez-Trigo, 1 (1998) Prognostic value of motor evoked potential obtained by transcranial magnetic brain stimulation in motor function recovery in patients with acute ischemic stroke. Stroke, 29: 1854-1859. Faig, 1 and Busse, 0 (1996) Silent period evoked by transcranial magnetic stimulation in unilateral thalamic infarcts. J. Neurol. Sci., 142: 85-92. Faist, M, Ertel, M, Berger, Wand Dietz, V (1999) Impaired modulation of quadriceps tendon jerk reflex

612 during spastic gait: differences between spinal and cerebral lesions. Brain, 122 :567-579. Ferbert, A, Vielhaber, S, Meincke, U and Buchner, H (1992) Transcranial magnetic stimulation in pontine infarction: correlation to degree of paresis. J. Neurol. Neurosurg. Psychiatry, 55: 294-299. Feydy, A, Carlier, R, Roby-Brami, A, Bussel, B, Cazalis, F, Pierot, L, Burnod, Y and Maier, MA (2002) Longitudinal Study of Motor Recovery After Stroke. Recruitment and Focusing of Brain Activation. Stroke, 33: 1610-1617. Feys, H, Van Hees, J, Bruyninckx, F, Mercelis, R and De Weerdt, W. Value of somatosensory and motor evoked potentials in predicting arm recovery after a stroke. J. Neurol. Neurosurg. Psychiatry, 68: 323-331. Fries, W, Daneck, A and Witt, TN (1991) Motor responses after transcranial electrical stimulation of cerebral hemispheres with degenerated pyramidal tract. Ann. Neurol., 29: 646-649. Fritz, C, Braune, HJ, Pylatiuk, C and Pohl, M (1997) Silent period following transcranial magnetic stimulation: a study of intra- and inter-examiner reliability. Electroencephalogr. Clin. Neurophysiol., 105: 235240. Fuhr, P, Agostino, R and Hallett, M (1991) Spinal motor neuron excitability during the silent period after cortical stimulation. Electroencephalogr. CUn. Neurophysiol., 81: 257-262. Glassman, RB (1974) Functional recovery after lesions of the nervous system. IV. Structural correlates of recovery in adult subjects. Equipotentiality and sensorimotor function. Neurosci. Res. Program Bull., 12: 246-249. Glassman, RB and Malamut, BL (1976) Recovery from electroencephalographic slowing and reduced evoked potentials after somatosensory cortical damage in cats. Behav. Bioi., 17: 333-354. Hallett, M (2000) Transcranial magnetic stimulation and the human brain. Nature, 406: 147-150. Hari, R, Karhu, J, Hamalainen, M, Knuutila, J, Salonen, 0, Sams, M and Vilkman, V (1993) Functional organization of the human first and second somatosensory cortices: a neuromagnetic study. Eur. J. Neurosci., 5: 724-734. Haug, BA, Schonle, PW, Knobloch, C and Kohne, M (1992) Silent period measurement revives as a valuable diagnostic tool with transcranial magnetic stimulation. Electroencephalogr. CUn. Neurophysiol., 85: 158-160. Heald, A, Bates, D, Cartlidge, NE, French, 1M and Miller, S (1993) Longitudinal study of central motor conduction time following stroke. 2. Central motor conduction measured within 72 h after stroke as a predictor of functional outcome at 12 months. Brain, 116: 1371-1385.

P.M. ROSSINIAND F. PAURI

Hendricks, HT, Hageman, G and Van Limbeek, J (1997) Prediction of recovery from upper extremity paralysis after stroke by measuring evoked potentials. Scand. J. Rehabil. Med., 29: 155-159. Jane, JA, Yashon, D, Becker, DP, Beatty, R and Sugar, 0 (1968) The effect of destruction of the corticospinal tract in the human cerebral peduncle upon motor function and involuntary movements. Report of II cases. J. Neurosurg., 29: 581-585. Kujirai, T, Caramia, MD, Rothwell, JC, Day, BL, Thompson, PD, Ferbert, A, Wroe, S, Asselman, P and Marsden, CD (1993) Corticocortical inhibition in human motor cortex. J. Physiol., 471: 501-519. Kukowski, B and Haug, B (1992) Quantitative evaluation of the silent period, evoked by transcranial magnetic stimulation during sustained muscle contraction, in normal man and in patients with stroke. Electromyogr. CUn. Neurophysiol., 32: 373-378. Lawrence, DG and Kuypers, HG (1978a) The functional organization of the motor system in the monkey. I. The effects of bilateral pyramidal lesions. Brain, 91: 1-14. Lawrence, DG and Kuypers, HG (1968b) The functional organization of the motor system in the monkey. IT. The effects of lesions of the descending brain-stem pathways. Brain, 91: 15-36. Liepert, J, Tegenthoff, M and Malin, J-P (1995) Changes of postexcitatory inhibition after transcranial magnetic stimulation in the course of hemiparesis. Neurol. Psych. Brain Res., 4: 1-6. Liepert, J, Miltner, WHR, Bauder, H, Sommer, M, Dettmers, C, Taub, E and Weiller, C (1998) Motor cortex plasticity during constraint-induced movement therapy in stroke patients. Neurosci. Lett., 250, 5-8. Liepert, J, Storch, P, Fritsch, A and WeiHer, C (2000) Motor cortex disinhibition in acute stroke. Clin. Neurophysiol., Ill: 671-676. Liepert, J, Uhde, I, GrM, S, Leidner, 0 and Weiller, C (2001) Motor cortex plasticity during forced use therapy in stroke patients. J. Neurol., 248: 315-321. Luu, P, Tucker, DM, Englander, R, Lockfeld, A, Lutsep, H and Oken, B (2001) Localizing acute stroke-related EEG changes: assessing the effects of spatial undersampling. J. Clin. Neurophysiol., 18: 302-317. Marque, P, Simonetta-Moreau, M, Maupas, E and Roques, CF (2001) Facilitation of transmission in heteronymous group IT pathways in spastic hemiplegic patients. J. Neurol. Neurosurg. Psychiatry, 70: 36-42. Marshall, RS, Perera, GM, Lazar, RM et al. (2000) Evolution of cortical activation during recovery from corticospinal tract infarction. Stroke, 31: 656-661. Mima, T, Toma, K, Koshy, B and Hallett, M (2001) Coherence between cortical and muscular activities after subcortical stroke. Stroke, 32: 2597-2601.

HEMIPARESIS

Nakasato, N, Levesque, MF, Barth, DS, Baumgartner, C, Rogers, RL and Sutherling, WW (1994) Comparisons of MEG, EEG, and ECoG source localization in neocortical partial epilepsy in humans. Electroencephalogr. Clin. Neurophysiol., 91: 171-178. Netz, J, Lammers, T anhd Homberg, V (1997) Reorganization of motor output in the non-affected hemisphere after stroke. Brain, 120: 1579-1586. Oliveri, M, Rossini, PM, Pasqualetti, P, Traversa, R, Cicinelli, P, Palmieri, MG, Tomaiolo, F and Caltagirone, C (1999) Interhemispheric asymmetries in the perception of unimanual and bimanual cutaneous stimuli. A study using transcranial magnetic stimulation. Brain, 122: 1721-1729. Oliveri, M, Turriziani, P, Carlesimo, GA, Koch, G, Tomaiuolo, F, Panella, M and Caltagirone, C (2001) Parieto-frontal interactions in visual-object and visualspatial working memory: evidence from transcranial magnetic stimulation. Cereb. Cortex, II: 606-618. Palmer, E, Ashby, P and Hajek, VE (1992) Ipsilateral fast corticospinal pathways do not account for recovery in stroke. Ann. Neurol., 32: 519-525. Pennisi, G, Rapisarda, G, Bella, R, Calabrese, V,Maertens de Noordhout, A and Delwaide, PJ (1999) Absence of response to early transcranial magnetic stimulation in ischemic stroke patients: prognostic value for hand motor recovery. Stroke, 30: 2666-2670. Pappata, S, Fiorelli, M, Rommel, T et al. (1993) PET study of changes in local brain haemodynamics and oxygen metabolism after unilateral middle cerebral artery occlusion in baboons. J. Cerebr. Blood Flow Metab., 14: 892-902. Pereon, Y,Aubertin, P and Guihenuc, P (1995) Prognostic significance of electrophysiological investigations in stroke patients: somatosensory and motor evoked potentials and sympathetic skin response. Neurophysiol. Clin., 25: 146-157. Pisano, F, Miscio, G, Del Conte, C, Pianca, D, Candeloro, E and Colombo, R (2000) Clin. Neurophysiol., Ill: 1015-1022. Pizzella, V, Tecchio, F, Romani, GL and Rossini, PM (1999) Functional localization of the sensory hand area with respect to the motor central gyrus knob. NeuroReport, 16: 3809-3814. Platz, T, Kim, IH, Pintschovius, H, Winter, T, Kieselbach, A, Villringer, K, Kurth, R and Mauritz, KH (2000) Multimodal EEG analysis in man suggests impairmentspecific changes in movement-related electric brain activity after stroke. Brain, 12: 2475-2490. Platz, T, Kim, IH, Engel, D, Kieselbach, A and Mauritz, KH (2002) Brain activation pattern as assessed with multi-modal EEG analysis predict motor recovery among stroke patients with mild arm paresis who

613 receive the Arm Ability Training. Restor. Neurol. Neurosci., 20: 21-35. Powers, WJ and Rechle, ME (Eds.) (1987) Cerebrovascular diseases. Proceedings of the 15th Research Conference. New-York, Raven Press, pp. 273-281. Rapisarda, G, Bastings, E, Maertens de Noordhout, A, Pennisi, G and Delwaide, PJ (1996) Can motor recovery in stroke patients be predicted by early transcranial magnetic stimulation? Stroke, 27: 21912196. Roick, H, Von Giesen, HJ and Benecke, R (1993) On the origin of the postexcitatory inhibition seen after transcranial magnetic brain stimulation in awake human subjects. Exp. Brain Res., 94: 489-498. Rossini, PM (2001) Brain redundancy: responsivity or plasticity? Ann. Neurol., 48: 128-129. Rossini, PM and Pauri, F (2000) Neuromagnetic integrated methods tracking human brain mechanisms of sensorimotor areas 'plastic' reorganisation. Brain Res. Rev., 33: 131-154. Rossini, PM and Rossi, S (1998) Clinical applications of motor evoked potentials. Electroncephalogr. Clin. Neurophysiol., 106: 180-194. Rossini, PM, Narici, L, Martino, G, Pasquarelli, A, Peresson, M, Pizzella, V, Tecchio, F and Romani, GL (1994) Analysis of interhemispheric asymmetries of somatosensory evoked magnetic fields to right and left median nerve stimulation. Elecotrencephalogr. Clin. Neurophysiol., 91: 476-482. Rossini, PM, Caltagirone, C, Castriota-Scandenberg, A, Cicinelli, P, Del Gratta, C, Demartin, M, Pizzella, V, Traversa, R and Romani, GL (1998a) Hand motor cortical areas reorganization in stroke: a study with fMRI, MEG and TCS maps. NeuroReport, 9: 21412146. Rossini, PM, Tecchio, F, Pizzella, V,Lupoi, D, Cassetta, E, Pasqualetti, P, Romani, GL and Orlacchio, A (1998b) On the reoganizationnn of sensory hand areas after monohemispheric lesion: a functional (MEG)/anatomical (MRI) integrative study. Brain Res., 782: 153-166. Rossini, PM, Tecchio, F, Pizzella, V, Lupoi, D, Cassetta, E and Pasqualetti, P (200Ib) Interhemispheric differences of sensory hand areas after monohemispheric stroke: MEGIMRI integrative study. Neurolmage, 14: 474-485. Seitz, RJ and Freund, HJ (1997) Plasticity of the human motor cortex. Adv. Neurol., 73: 321-333. Shimizu, T, Oliveri, M, Filippi, MM, Palmieri, MG, Pasqualetti, P and Rossini, PM (1999) Effect of paired transcranial magnetic stimulation on the cortical silent period. Brain Res., 834: 74-82. Shimizu, T, Hosaki, A, Hino, T, Sato, M, Komori, T, Hirai, S and Rossini, PM (2002) Motor cortical dishinibition

614 in the unaffected hemisphere after unilateral cortical stroke. Brain, 125: 1896-1907. Tecchio, F, Rossini, PM, Pizzella, V, Cassetta, E, Pasqualetti, P and Romani, GL (1998) A neuromagnetic normative data set for hemispheric sensory hand cortical representations and their interhemispheric differences. Brain Res. Protocols, 2: 306-314. Tecchio, F, Pasqualetti, P, Pizzella, V, Romani, G and Rossini, PM (2000) Morphology of somatosensory evoked fields: interhemispheric similarity as a parameter for physiological and pathological neural connectivity. Neurosci. Lett., 287: 203-206. Timmerhuis, TP, Hageman, G, Oosterloo, SJ and Rozeboom, AR (1996) The prognostic value of cortical magnetic stimulation in acute middle cerebral artery infarction compared to other parameters. Clin. Neurol. Neurosurg., 98: 231-236. Traversa, R, Cicinelli, P, Bassi, A, Rossini, PM and Bernardi, G (1997) Mapping of motor cortical reorganization after stroke. Stroke, 28: 110-117 . Traversa, R, Cicinelli, P, Pasqualetti, P, Filippi, MM and Rossini, PM (1998) Follow-up of interhemispheric differences of motor evoked potentials from the 'affected' and 'unaffected' hemispheres in human stroke. Brain Res., 803: 1-8. Traversa, R, Cicinelli, P, Oliveri, M, Palmieri, MG, Filippi, MM, Pasqualetti, P and Rossini, PM (2000) Neurophysiological follow-up of motor output in stroke patients. Clin. Neurophysiol., Ill: 1695-1703. Triggs, WJ, Cros, D, Macdonell, RAL, Chiappa, KH, Fang, J and Day, BJ (1993) Cortical and spinal motor excitability during the transcranial magnetic stimulation silent period in humans. Brain Res., 628: 39-48. Trompetto, C, Assini, A, Buccolieri, A, Marchese, Rand Abbruzzese, G (2000) Motor recovery following stroke: a transcranial magnetic stimulation study. Clin. Neurophysiol., 111: 1860-1867.

P.M. ROSSINIAND F. PAURI

Turton, A, Wroe, S, Trepte, N, Fraser, C and Lemon, RN (1996) Contralateral and ipsilateral EMG responses to transcranial magnetic stimulation during recovery of arm and hand function after stroke. Electroencephalogr. Clin. Neurophysiol., 101: 316-328. Twitchell, TE (1951) The restoration of motor function following hemiplegia in man. Brain, 74: 443-480. Yang, C, Dunbabin, D and Kilpatrick, D (1999) Correlation between functional and electrophysiological recovery in acute ischemic stroke. Stroke, 30: 2126-2130. Von Giesen, H-J, Roick, H and Benecke, R (1994) Inhibitory actions of motor cortex following unilateral brain lesions as studied by magnetic brain stimulation. Exp. Brain Res., 99: 84-96. Warabi, T, Inoue, K, Noda, H and Murakami, S. Recovery of voluntary movement in hemiplegic patients. Correlation with degenerative shrinkage of the cerebral peduncles in CT images. Brain, 113: 177-189. Wikstrom, H, Roine, RO, Salonen, 0, Aronen, HJ, Virtanen, J, Ilmoniemi, RJ and Huttunen, J (1997) Somatosensory evoked magnetic fields to median nerve stimulation: interhemispheric differences in a normal population. Electroencephalogr. Clin. Neurophysiol., 104: 480-487. Wikstrom, H, Roine, RO, Aronen, HJ et al. (2000) Specific changes in somatosensory evoked magnetic fields during recovery from sensorimotor stroke. Ann. Neurol., 47: 353-360. Yousry,TA, Schmid, DD, Alkadhi, H, Schmidt, D, Peraud, A, Buettner, A and Winkler, P (1997) Localisation of the motor hand area to a knob on the precentral gyrus: a new landmark. Brain, 120: 141-157. Ziemann, D, Lonnecker, S, Steinhoff, BJ and Paulus, W (1996) The effect of lorazepam on the motor cortical excitability in man. Exp. Brain Res., 109: 127-135.

Movement Disorders Handbook of Clinical Neurophysiology, Vol. I M. Hallett (Ed.) © 2003 Elsevier B. V. All rights reserved

615 CHAPTER 38

Movement disorders: spasticity Reiner Benecke* Department of Neurology, University of Rostock, Gehlsheimer Strasse 20, D-18147 Rostock, Germany

38.1. Introduction and terminology Spasticity is not a neurological disease but a motor disorder or syndrome characterized by a velocity-dependent increase of muscle tone sometimes combined with the clasp knife phenomenon and exaggerated monosynaptic and oligosynaptic stretch reflexes and polysynaptic tonic reflexes. Clinically, the increase of stretch reflexes is often combined with reflex irradiation, i.e. the application of a phasic stretch by a reflex hammer and the mechanical effects of the primarily contracting muscle lead to excitation of muscle spindles and reflexes in distant muscles. Especially in triceps surae and quadriceps, muscle clonus may arise after sudden stretch of these muscles. Most clinicians also regard Babinski's reflex as a representative symptom of spasticity (Fig. 1). Spasticity is not synonymous with either the upper motor neuron syndrome (UMS) or the spastic syndrome, as it is also called. Here spasticity is only a constitutive part of that clinical picture (Table 1). In spastic syndromes or VMSs other clinical signs occur, such as pareses, disturbed coordination, increased flexor reflexes, autonomic hyperreflexia, dystonia and spasms. Spasms in flexor and extensor muscles usually combined with pain often occur in VMS. They appear either spontaneously or in response to sensory stimuli. Frequently they are tonic in nature, occasionally interspersed with myoclonic bursts. There is much confusion about the term spasms and a generally acceptable definition has not been given so far. In pathophysiological terms spasms can be * Correspondence to: Reiner Benecke, Department of Neurology, University of Rostock, Gehlsheirner Strasse 20, D-18147 Rostock, Germany. E-mail address:[email protected] Tel.: 0049-381-4949510; fax: 0049-381-4949512.

regarded as an expression of exaggerated nociceptive reflexes induced either by enhanced excitation of receptors or by a disturbed supraspinal control at the spinal interneuronal level. The spasms occurring in UMSs are certainly mainly the result of a disturbed supraspinal control. Traumatic, inflammatory and degenerative lesions not only of the skin but also of numerous other organs are able to induce enhanced activity in group II, III and IV afferent fibers (Fig. 2). These fiber types are the typical sources of nociceptive flexor reflexes. For example, painful lesions of joints lead to spasms of neighboring muscles, which stabilize the joint in a protective manner, thereby preventing extensive excursion of the limb which could exacerbate the joint pain. The same mechanisms seem to be active in degenerative changes of the spine with resulting back pain in spasms of the paraspinal muscles. Table 1 Clinical signs of the spastic syndrome (UMS). Positive symptoms • velocity-dependent increase of muscle tone with facultative clasp knife phenomenon* • increased deep tendon jerks and clonus* • Babinski's reflex* • increased flexor reflexes and mass reflexes • autonomic hyperreflexia (detrusor-sphincter dyssynergia) • spasms • dystonia (spastic dystonia or dystonic spasticity) • subtraction-paresis Negative symptoms • • • •

paresis or plegia impaired dexterity disturbed coordination early fatigue

* Signs of pure

spasticity.

616

R. BENECKE

Fig. I. Signs and symptoms of spasticity. (a) A velocity dependent increase of muscle tone can be detected by passive stretch of the affected muscle (M. triceps brachii). (b) Exaggerated tendon reflexes are induced by enhanced excitability of o-motoneurons. (c) Illustration of reflex irradiation (upper part) and Babinski's reflex (lower part). In case of reflex irradiation, application of a phasic stretch by a reflex hammer and the mechanical effects of the primarily contracting muscle lead to excitation of muscle spindles and reflexes in distant muscles due to lowered excitation threshold.

In most patients, detection of spastic muscle tone is relatively easy clinically. However, spastic muscle tone may be so severe that passive movements of the patient may be difficult to perform. In these cases, it may be difficult to distinguish between spastic muscle tone and fixed contractures or severe dystonia. In rare cases, it may also be difficult to differentiate spastic muscle tone from the stiffness found in the stiff-person syndrome. Additionally, difficulty arises because spastic muscle tone and dystonia may be present simultaneously in the same limb. Spastic muscle tone can be found more readily in certain muscle groups (adductors and flexors in the upper limb; adductors and extensors in the lower limb). In some textbooks the motor syndrome of weakness, spasticity and hyperreflexia is still called pyramidal tract syndrome. Use of this term is not correct because a spastic syndrome does not occur

after a selective damage to the corticospinal (pyramidal) tract. In humans, the acute effects of disruption of the corticospinal system by pyramidotomy are pronounced in the beginning but are followed by a remarkable degree of recovery. Immediately after selective destruction, the limbs become hypotonic and severely weak, the arms more than the legs. After several days, however, subjects are able to stand unsupported, to walk and to perform proximal arm movements. Movements of the trunk and proximal portion of the limbs appear normal. Distal limb movements recover more slowly. Initially, the hand is held in an abnormal position and handling of objects is impaired. Eventually, there is a nearly full recovery of the use of limbs, including the hand, with the exception of a permanent loss of independent movements of the fingers. Similar results are associated with section of corticospinal tracts at the level of the cerebral peduncle. With pyramidal tract

617

MOVEMENT DISORDERS: SPASTICITY

DESCENDING PATHWAYS DORSAL COLUNM TRACT SPINOTHALAMIC TRACT

III,N

PROPRIOSPINAL

Fig. 2. Schematic illustration of receptors mediating nociceptive reflexes and descending control of this pathway. Traumatic, inflammatory and degenerative lesions not only of the skin but also of numerous other organs are able to induce and enhance activity in group II, III and IV afferent fibers.

lesions there is a dramatic effect also in muscle tone and in deep tendon reflexes that are reduced or absent in the contralateral extremities. Within days to weeks, however, muscle tone and reflexes show considerable recovery and usually return to normal. The Babinski sign, however, may be present and persist. Thus, it must be emphasized that spasticity is not seen after pure corticospinal tract lesions. In the discussion of clinical aspects and treatment of spasticity, spasms and VMSs, three major problems arise. The first problem facing a clinician is to diagnose and correctly interpret the symptoms observed. The second clinical relevant problem is to delineate which disease in an individual patient causing spasticity or an VMS is present. The third problem is to decide which causal therapy if available is to be applied and how symptomatic therapy of spasticity and the other features of the VMS can be used in the individual patient. Except

some forms of genetically determined spastic paraplegias, spasticity usually does not occur separately; in nearly all diseases spasticity is embedded in an

VMS. 38.2. Pathophysiological aspects Pathoanatomically, spasticity is ascribed to lesions of an interindividually variable set of descending fiber tracts that influence bulbar and spinal reflex activities. The pyramidal tract mayor may not belong to the systems affected, however, as has already been mentioned. Obviously, although the other features of a spastic syndrome or VMS are caused by different lesions, it might well be that the negative symptoms such as paresis, loss of dexterity and coordination problems are mainly the result of a dysfunction of the corticospinal tract, especially as far as the distal hand muscles are concerned. Pareses

618

and coordination problems in proximal muscles, however, may also be induced by lesions of the rubrospinal, reticulospinal and vestibulospinal tracts. Affections of descending fiber tracts can either result from a direct lesion of the tract with axonal degeneration or severe conduction block or by secondary functional affection. The latter aspect has to be taken into account especially in cerebral lesions of motor cortices, subcortical white matter, internal capsule and mesencephalon, which all can disturb activity of target neurons in the bulbar reticular formation (Brown, 1994). This corticobulbar system has also been shown to be heavily modulated by the premotor cortex, paramedian cerebellar cortex, and fastigial nucleus. Inhibitory influences from the bulbar reticular formation are conducted in the spinal cord by the dorsal reticulospinal tract in the dorsal half of the lateral funiculus, in close relationship with the lateral corticospinal tract. The dorsal reticulospinal tract transmits inhibitory signals to circuits for spinal stretch reflexes as well as flexor reflexes. Thus, in addition to enhanced stretch reflexes, flexor reflexes are released with damage to this system. The descending fiber tracts which influence the bulbar reticular formation in some parts of the brain lie in close vicinity to the corticospinal tract, and for this reason some researchers use the term parapyrarnidal fiber tracts. In this context it should also be mentioned that the corticospinal tract itself and the rubrospinal tract send collaterals to the reticular formation, and thereby also influence functionally the activity of the reticulospinal tracts. At first sight, it could be expected that motor syndromes induced by lesions of descending motor fiber tracts would be extremely variable due to the interindividual spectrum of fiber tract affections. However, this is not the case. We know from daily clinical practice that lesions, which can be extremely well defined in site and extension by means of modem imaging techniques along the entire neuraxis from cortex to upper lumbal spinal cord all lead to the stereotyped clinical picture of spasticity or spastic syndrome even though flexor and extensor spasms are a typical feature of spinal spasticity. It is, however, clear that the site of lesion governs the topographic pattern of spastic syndromes, e.g. in cerebral lesions a spastic hemisyndrome can be observed whereas in upper cervical lesions a spastic

R. BENECKE

tetrasyndrome occurs. Furthermore, due to the somatotopic organization of motor cortices, subcortical white matter and internal capsule, spastic syndromes can occur separately in one limb or even only in proximal or distal muscle groups. Nevertheless, when we detect, for example, spasticity in elbow flexors, this can be induced by a cortical, subcortical, capsular, mesencephalic, brain stem or spinal cord lesion, irrespective of which descending fiber tract or spectrum of tract lesions is present. The mystery of the development of a stereotyped spastic syndrome can only be explained if one postulates a stereotyped neuroplastic mechanism on spinal segmental levels which is exerted whenever denervation and loss of supraspinal control of spinal interneurons and motoneurons occur. Mainly two fundamental mechanisms for the genesis of spasticity have been discussed in the past: (1) the imbalance; and (2) the sprouting theory. It has been suggested that intraspinal sprouting by dorsal root afferents from muscle and skin receptors caudal to a lesion of descending motor tracts on the various segmental levels could occur. The formation of new synaptic contacts on partially denervated interneurons and motoneurons was thought to be responsible for the increased reflex activity following spinal cord lesions (McCouch et al., 1958; Goldberger and Murray 1988). The putative targets for such sprouting are the interneurons located in the intermediate gray matter of the spinal cord, as they receive a substantial amount of converging input from supraspinal descending and segmental afferent fibers. However, the use of modem markers for sprouting of primary afferents in this region could not confirm this theory, even using the same animal model of low thoracic cord hemisection (Nacimiento et al., 1993, 1995). Other studies have attributed the enhanced excitability of o-motoneurons in spastic syndromes to sprouting of serotonergic fibers originating from the undamaged contralateral serotonergic system and crossing the midline at segmental levels (Saruhashi et al., 1996). The serotonergic system exerts an excitatory influence on a-motoneurons. Serotonergic reinnervation could thus lead to the recovery from the depression of motoneurons following spinal hemisection, but whether hyperexcitability could also result from this serotonergic reinnervation is questionable. Restoration of motor function and rise of spasticity could also be explained by another observation, namely

MOVEMENT DISORDERS: SPASTICITY

that pre-existing silent descending pathways crossing at the segmental level may be reinforced following supraspinal unilateral lesions of the descending motor tracts (Thallmair et al., 1998). However, recovery from spinal shock after complete spinal transsection is difficult to explain by these mechanisms as long as sprouting of any fiber system crossing in the scar of the spinal lesions has never been observed under experimental and clinical conditions. The presently favored mechanism for the rise of spasticity after lesions of descending motor tracts is an imbalance between the activity of excitatory and inhibitory intemeurons mediating excitatory and inhibitory control of the spinal reflexes. It has been shown that both the inhibitory and excitatory intemeurons get influences from descending motor tracts. Modem morphological techniques applied to animal experiments helped to elucidate cellular mechanisms in the spinal cord which are involved in the pathophysiology of spasticity (Nacimiento et al., 1997). Changes in motor function following low thoracic spinal cord hemisection in cats and rats revealed important components which also occur in human spasticity. On the surface of motoneurons distal to the lesion, structural changes were observed predominantly in inhibitory axosomatic synaptic boutons. This suggests a lesion induced imbalance in synaptic input, with a disturbance of inhibitory synapses and subsequent dominance of excitatory input from unlesioned descending fiber tracts. Such a mechanism may contribute to the well-known increased excitability of a-motoneurons caudal to interruption of supraspinal descending pathways. It is not clear so far which neurotransmitters and postsynaptic receptors are involved in this synaptic plasticity and to what extent the various transmitter systems contribute to the development of spasticity. The imbalance of excitatory and inhibitory synapses after partial lesion of descending fiber systems could be observed both at spinal intemeurons and motoneurons (Nacimiento et al., 1997).

38.3. The motor disability in spasticity and the VMS An obvious candidate mechanism underlying weakness, as one of the principal negative signs of the VMS, is deficient motoneuronal activation,

619

resulting from diminished corticomotoneuronal drive as a consequence of interruption of the corticomotoneuronal fiber tracts. Another mechanism of weakness is the increased spastic muscle tone in the antagonist which is stretched when the agonist is voluntarily contracted. This type of weakness is called subtraction-paresis. Spastic muscle tone has two components: enhanced stretch reflex activity and altered viscoelastic properties of muscle fibers. The mechanical resistance to a sudden torque was found to be decreased in voluntarily activated spastic muscles (Ibrahim et al., 1993). This decrease in mechanical resistance in the active state was entirely due to a massive reduction in reflexive electromyographic activity, since the ratio of electromyographic activity to torque was found to be increased even in active muscles of patients with spasticity (Ibrahim et al., 1993). This finding probably indicates that increased viscoelasticity may not be particularly relevant to the motor deficit, at least in circumstances in which reflexive electromyographic activity is elicited. A controversy has arisen whether the increase in stretch resistance in spasticity, which undoubtedly exists under experimental conditions (Thilmann et al., 1991), affects the natural movements of spastic patients. Biomechanical recordings and EMG analyses of patients with spastic gait disorders have revealed that in a large number of patients, no pathological EMG activity was detectable in the triceps surae during its lengthening in the swing phase (Dietz et al., 1981). The characteristic impairment of active dorsiflexion of the .spastic foot was thought to result from other factors, such as changes in mechanical properties of spastic muscle fibers or changes in the viscoelastic elements. A similar finding was obtained in patients with spastic arm paresis performing voluntary movements at the elbow (Fellows et al., 1994). The result of these behavioral studies has clearly demonstrated that enhanced phasic and tonic stretch reflexes are not necessarily responsible for the functional impairment of natural movements in spasticity. It should be kept in mind, however, that patients recruited for these studies were patients with weak spasticity, who were only slightly impaired during walking. In severe spasticity, in which pathological flexor reflexes can be elucidated by stimulation of the skin, enhanced activity of spinal intemeurons and amotoneurons of course is responsible for the

620

R.BENECKE

symptoms of spasucity, but these patients are normally unable to walk.

38.4. Measurements of spasticity There are many approaches to measure spasticity. On the one hand there are indirect measures (scales), on the other hand there are many direct measures, in which biomechanical and neurophysiological phenomena are quantified. In the majority of papers dealing with measurements of spasticity representative muscles or muscle groups acting at one joint have been analyzed. Some measurements concentrate on EMG recordings, others include biomechanical aspects. The most representative measurements of spasticity are certainly analyses of complex natural movements such as gait, bicycling

and reaching movements of the arm and hand (Benecke, 1987) (Fig. 3). Scales are the most common clinical approach to the routine measurement of levels of spasticity. The first spasticity scale was proposed for use in multiple sclerosis by Ashworth (1964) (Table 2). In 1987 Bohannon and Smith (1987) proposed a modification (Table 3), which they suggested would improve its usefulness. In summary, these scales are limited by reliability and difficulties with some definitions. One simple test is the Wartenberg Pendulum Test (Wartenberg, 1951) in which the patients lay supine or can sit with the knee flexed. The clinician then extends the knee and releases the foot, allowing the leg to swing. In a normal healthy joint, it will swing about six times, but in the presence of spasticity the number of swings is reduced in accordance with the

B

R·

IT

I

.....

A[it::i] II'LI

.$ ,..,_.

,----, 1s [1 mV

B~

cCJ::iJ pedal position

6

120

5

100

X x"".>/,/

c

0

-6

lox M.rect. fern. . . . ,0 -'1

-4

2

d

e

8

20

x

eo 60 40

,

9

24

28

20

min

32

... .. -.. ..... ....

]1mV

b

1"-

c ...

d

'. .' ...

'

e

9

'is'

10mg Diozepom

Fig. 3. (A) Quantitative measurements of spasticity during bicycling using quotient R. Upper part: Mean angular range (black) of four routinely examined muscles and standard deviation (white) presented as concentric arcs. The measurements were performed at 30 cycles/min rotation and 4 kpm work load. Lower part (I): representative EMG recordings of right rectus femoris muscle in a normal subject (A) and in two patients with a spastic paraparesis (B, C). IT: Corresponding average (eight rotations) of EMG in a normal subject and in two patients with severe spastic paraparesis. The horizontal axis shows the position of the pedal with the origin corresponding to the lowest position of the right pedal. Computerevaluated EMG-integral (II) between 1200 and 300 0 (mean values of muscle recruitment evaluated in normal subjects; white arc) and between 300 0 and 120° (10; black area), where in normal subjects the muscle is resting. R=II/lO. Note flattening of activity and enhanced duration of muscle recruitment in Band C. R is inversely related to severity of spastic paraparesis. (B) Quantitative measurements of drug action on the EMG patterns during bicycling. The diagram shows the R-values (.) and the IO-values (x) prior to and after intravenous administration of 10 mg diazepam. On the right side representative original registrations from the right rectus femoris muscle are illustrated prior to (a,b) and after administration of the drug (c-g) (modified from Benecke 1987).

621

MOVEMENT DISORDERS: SPASTICITY

Table 2 Original Ashworth scale. en

Grade

Description

o

No increase in tone

I

Slight increase in tone giving a "catch" when the limb is moved in flexion or extension

2

More marked increase in tone, but limb easily flexed

3

Considerable increase in tone, passive movement difficult

4

Limb rigid in flexion or extension

velocity dependence. Leslie et al. (1992) investigated the use of the relaxation index and its relationship to the original Ashworth-Scale (Fig. 4). They found a good correlation and showed that the Pendulum Test appeared to be a more sensitive measure than the Ashworth Scale. More recently, Fowler et al. (2000) used the same test in cerebral palsy patients. Although the Wartenberg Test appears to be a valid technique for measuring spasticity at the knee, it is difficult to apply at other joints and it may not be applicable in cases of more severe spasticity. Numerous mechanical methods measure torqueangle relationships at spastic joints during passive flexion and extension. Controlled stretches can be delivered by devices containing torque motors. The perturbation can be a single step or more complex, Table 3 Modified Ashworth scale. Grade

Description

o

No increase in muscle tone

I

Slight increase in tone - a catch and release at the end of the range of motion

I+

Slight increase in tone - catch, followed by minimal resistance in remainder of range

2

More marked increase in muscle tone through most of range

3

Considerable increase in tone, passive movement difficult

4

Affected parts rigid in flexion or extension

c:

.~

Vl

.....o

Al

Q)

"0

:J

:t=

"is.

E


Relaxation Index

=~

Time Fig. 4. Definition of Relaxation Index using the Wartenberg test.

such as a sinusoid. The mechanical response of the limb is measured. Objective techniques are an improvement over bedside testing and theoretically may have a role in differentiating spastic muscle tone from other tone disorders, although the distinction is usually made clinically. Additionally, these techniques might provide a means of monitoring muscle tone in longitudinal studies or studies aimed at studying the effects of a therapeutic intervention. However, it is important to bear in mind that the clinical value depends on the value of the measures assessed by them. Diagnostic nerve or motor point blocks may help to differentiate between the contributions of neural and intrinsic mechanical factors to spastic muscle tone. Quantitative differentiation of the intrinsic and neural components of spastic muscle tone can also be performed noninvasively. A simple and semiquantitative assessment is provided by multichannel surface electromyography (Pullman et al., 2000), but the most accurate assessment is provided by a combination of electromyographic and biomechanical techniques. Certain indices of stiffness are derived from torque-position relationships at various angular velocities. An intrinsic stiffness index corresponds to an angular velocity in situations in which no reflexive electromyographic activity is noted. Conversely, the total stiffness index obtained at higher angular velocities would reflect both intrinsic and neuronal contributions to stiffness. The intrinsic stiffness index differs between female and male subjects, most probably owing to the sexdependent difference in arm muscle mass. The total stiffness index is significantly higher in subjects who exhibit a stretch reflex than in those who do not have

622 a stretch reflex even at high angular velocities. The total stiffness index has been validated in stroke patients by demonstrating that it correlates well with conventional clinical scales of spasticity. Conversely, correlations with conventional neurophysiological measures, such as H-reflex latency, Hma,/Mmax ratio, stretch reflex latency, and area were poor (Pisano et al., 2000). Similar set-ups, such as those employed by Pisano and co-workers, may be used for the study of the lower limbs (Dietz and Berger, 1983). Reflex studies employing electrical nerve stimulation are generally not needed for the proper clinical management of a patient's spasticity. They might, however, be used in a scientific context. Once it has been established that spastic muscle tone has a significant neuronal component, there are a variety of elaborated techniques available for determining which of the interneurons involved in the stretchreflex circuitry function abnormally. o-motoneuronal excitability can be assessed with the use of F-waves or the H-reflex. An inventory of techniques available for testing human spinal reflexes is provided by Hallett et al. (1994), Kimura et al. (1994), Capaday (1997), Pierrot-Deseilligny (1997), Burke et al. (1999). These tests characterize reciprocal inhibition, presynaptic inhibition, synergistic type Ia facilitation, and type Ib inhibition. The main principle underlying the methods is that the H-reflex is used as a test reflex, and the effect of conditioning stimulation to various nerves is evaluated. The conditioning effect is determined as a difference in the sizes of the conditioned and unconditioned reflexes. The conditioning effect is expressed as a percentage of the ~ax or of the unconditioned test reflex (Kimura et al., 1994). While these tests are generally performed in the resting patient, the basic principles of deficient human motor control are best determined by studying freely moving subjects during motor tasks that are as natural as possible, for reasons previously outlined. Weakness is the leading factor that determines disability in the vast majority of patients with the VMS. In a given patient, it may be useful to quantity and visualize the functional consequences of a lesion of descending fiber tracts. Transcranial magnetic stimulation may be used to assess the integrity of the corticospinal tract (Benecke, 1993). Needle recordings may be used both to visualize insufficient recruitment of motor units and in the assessment of a reduction in the maximal upper frequency to which

R.BENECKE

motor units can be driven when the patient performs a maximal isometric contraction. Disturbances of the muscle activation pattern may be visualized by recording surface electromyographic activity during the execution of common and natural motor tasks. If these recordings are used in a control setting, such as reaching for a target, or bicycling, they may reveal abnormal coactivation of agonist and antagonist muscles, and temporarily extended muscular activity (Benecke, 1987). These measures may then be used to control for the effect of therapeutic interventions, such as physiotherapy. Deficits in the control of limb dynamics may also be visualized and quantitated by using optoelectronic systems (Beer et al., 2000) or computer-operated robot arms. 38.5. Diseases with spasticity or an VMS

Spasticity and the VMS are the most frequent motor abnormalities seen in neurological patients. Lesions of descending motor pathways which are followed by spasticity or VMS can affect the motor cortices, the subcortical gray matter including the internal capsule, the mesencephalon, the medulla oblongata or the spinal cord down to the segmental level L3, lesions below the L3 level are normally not followed by spasticity. The pathophysiological process leading to a lesion of descending motor tracts is irrelevant, it may be a hemorrhage, an ischemia, a tumor, a metabolic disorder, an encephalitis, a myelitis, an encephalomyelitis or a systemic or multisystemic degenerative process. Spasticity or an upper motor neuron lesion can be the only or an initial sign of a disease, it can be a key symptom of a disease or it may be a rare facultative or unimportant clinical sign in an individual patient. The first step in the differential diagnosis in cases with spasticity or an VMS is to decide whether spasticity or an VMS is the only motor problem or whether other clinical signs are present in addition. The most important diseases presenting with isolated spasticity, an isolated VMS and diseases in which spasticity or an VMS among other clinical symptoms are of diagnostic relevance are listed in Tables 4-6. 38.6. Treatment of spasticity

The stretch-induced muscle overactivity seen in spasticity results in stiffness and spasms, to which

623

MOVEMENT DISORDERS: SPASTICITY

Table 4

Table 6

Diseases with isolated spasticity.

Diseases with a dominant UMS combined with other clinical signs.

Hereditary spastic paraplegia (pure forms) Spastic type of amyotrophic lateral sclerosis Spastic type of combined degeneration of the spinal cord in vitamin B 12 deficiency Cervical myelopathy Thoracic myelopathy Primary lateral sclerosis Spastic type of Friedreich's disease Adrenomyeloneuropathy Encephalomyelitis disseminata

Table 5 Isolated UMSs. Hereditary spastic paraplegia (pure forms) Spastic type of amyotrophic lateral sclerosis Spastic type of combined degeneration of the spinal cord in vitamin B I 2 deficiency Cervical myelopathy Thoracic myelopathy Primary lateral sclerosis Spastic type of Friedreich's disease Adrenomyeloneuropathy Encephalomyelitis disseminata Basilar impression Lacunar ischemic infarct Ischemic MCA infarct Ischemic ACA infarct MCA=middle cerebral artery; ACA=anterior cerebral artery.

there is both a neurogenic and a biomechanical component. Spasticity does not always cause functional disability and can assist in motor performance and enabling a patient to stand when their limb weakness would not otherwise allow it. When spasticity causes functional disability, therapeutic interventions aim to decrease muscle overactivity, to reduce pain and to prevent secondary complications such as muscle contractures and various connective tissue alterations. One of the central aims of physical therapy is muscle lengthening. However, there is currently no published evidence concerning the effect of muscle shortening on disability and there remains much debate over this issue. The evidence concerning the benefits of muscle lengthening and stretching has to be viewed with some circumspec-

Hereditary spastic paraplegia (complex forms) Amyotrophic lateral sclerosis Friedreich's disease Adrenomyeloneuropathy Metachromatic leucodystrophy Mitochondrial encephalomyopathy Autosomal dominant cerebellar ataxias Hereditary motor sensory neuropathy type V Basilar impression Cervical myelopathy Thoracic myelopathy Tropical spastic paraplegia Encephalomyelitis disseminata Central pontine myelinolysis Syringomyelia Ischemic MCA infarct Ischemic ACA infarct Lacunar ischemic infarct Syndrome of the anterior spinal artery Weber syndrome Milliard-Gubler syndrome Jackson syndrome

tion. Physical treatment is commonly based on logic evidence rather than on evidenced-based medicine. The medical treatment of established spasticity is principally of symptomatic nature, since spasticity is in the majority of cases a consequence of irreversible central nervous system damage. In some diseases, however, a causal therapy is available which enables improvement of spasticity. This is the case, e.g. in combined degeneration of the spinal cord in vitamin B 12 deficiency, spinal cord compressions by disc prolapses or meningiomas. As spasticity presents with polymorphic clinical pictures resulting in various types and degrees of discomfort and dysfunction, the aim of the symptomatic treatment also varies. In general, it includes relief of discomfort, facilitation of patient's care, improvement of functional capacity, and protection of the long-term consequences of unevenly distributed excessive muscle tone. Theoretically, the main aim of an antispastic drug is to reduce increased muscle tone, given the definition of spasticity in strictly physiologic terms as a velocity-dependent increase in tonic stretch reflexes. However, spasticity is often asso-

624

ciated with other symptoms including brisk tendon reflexes, clonus, flexor, extensor, or adductor spasm which can be painful, loss of dexterity, and pareses, all of which constitute the VMS or spastic syndrome. Given the general profile of action, antispastic drugs can be expected to be more efficacious in relieving subjective discomfort due to positive symptoms, avoiding secondary complications due to increased muscle tone such as contractures, and facilitating nursing care and physiotherapy, whereas the other strategic aim of medical treatments, e.g. improvement of functional status, can be achieved only in a limited number of patients. If spasticity is focal, local botulinum toxin injections are the treatment of first choice. If the pattern of spasticity is regional and mainly affecting the legs, then intrathecal baclofen or widespread nerve blockade by phenol would be the treatment of choice. When spasticity is mild to moderate and occurs in all four limbs, then oral antispastic drug treatment seems to be the appropriate choice. The indications and limitations of the available treatments are discussed in the following.

R.BENECKE

survey of the most important antispastic drugs and indicates the suggested sites of action. Baclofen is a GABA B receptor agonist and was one of the first specific antispastic agents. The net physiologic effect of this drug is a suppression of the release of excitatory neural transmitters from primary afferent terminals. Baclofen is believed to exert its antispastic action directly at the spinal level by depressing monosynaptic and polysynaptic transmission in the spinal cord. In clinical trials it appears to be effective in spinal and cerebral spasticity due to various pathological conditions, and some 75% of patients reported a decrease of variable magnitude in muscle tone. Its usefulness in cerebral spasticity, however, is considered to be substantially less than that in spinal spasticity. The frequent adverse effects are tiredness, muscle weakness, hypotonia, and sedation, whereas nausea, vomiting, dizziness,

Ia

38.6.1. Oral antispastic therapy

There are four substances which successfully emerged from clinical trials and have become established major antispastic drugs. These include baclofen, dantrolene, benzodiazepines, and tizanidine. In addition to the major antispastic drugs, there is a second group of agents including those substances which exert a certain antispastic action but are not primarily used for this purpose. These include phenothiazines, adrenergic blocking agents, steroids, phenytoin, tetrahydrocannabinol, and morphine. In fact, all drugs which have a centrally mediated analgesic effect probably also have an antispastic action. The site and mode of action of antispastic drugs are predominantly discussed in the light of the enhanced segmental reflexes and spastic syndromes. Some drugs lower the sensitivity of peripheral sensory receptors, others reduce the release of excitatory transmitters at the presynaptic terminals of afferent nerve fibers or diminish the facilitatory effects of descending motor fibers on presynaptic terminals, interneurons and motoneurons. Furthermore, dantrolene interferes with the contractile mechanism of the skeletal muscle. Figure 5 gives a

Danlrolene (Botulinum toxin)

I Tizanidine

Fig. 5. Baclofen acts by stimulating GABAB receptors and the net physiological effect, is a suppression of excitatory neurotransmitter release and an enhancement of presynaptic inhibition. Benzodiazepines (e.g. diazepam, tetrazepam) exert a number of pharmacodynamic effects, one of which is muscle relaxation. The mechanism of antispastic action is to enhance the efficiency of GABAergic transmission by facilitating or potentiating the postsynaptic effects of GABA, which results in an increased chloride conductance and membrane hyperpolarization, thereby increasing the presynaptic inhibition of Ia afferents at axons of excitatory interneurons in the spinal cord. Tizanidine is a centrally active a-2 adrenergic receptor agonist. It drastically reduces the tonic, polysynaptic component of the stretch reflex, whereas the phasic, monosynaptic component remains virtually unchanged.

625

MOVEMENT DISORDERS: SPASTICITY

euphoria, depression, and confusion are less frequent side effects. In most of the clinical studies, very little attention was paid to functional changes, and there is no evidence for a positive effect on functional activities such as gait, ambulation or daily living activities. Tizanidine is a centrally active a-2 adrenergic receptor agonist. In in-vitro studies it shows very little affinity to o-I receptors and no appreciable affinity to the other receptor sites. It was found to inhibit the release of excitatory amino acid aspartate in spinal cord slices, to cause postsynaptic reduction in the effectiveness of excitatory transmitters and to markedly decrease the firing rate of adrenergic locus ceruleus neurons in the rat, which was reversed by the a-2 antagonist yohimbine. It is assumed that tizanidine exerts its antispastic actions through one or all of these mechanisms. Tizanidine also has an antinociceptive effect which is brought about by inhibition of the synaptic transmission of nociceptive stimuli in the spinal pathways through an a-2 adrenergic receptor-mediated action. In in-vivo experiments, tizanidine shows a potent and selective major relaxant action. It drastically reduces the tonic, polysynaptic component of the stretch reflex, whereas the phasic, monosynaptic component remains virtually unchanged. Side effects are possible and include sedation, dizziness and dry mouth, but most reports point out that weakness is not a great problem, compared with diazepam or baclofen (Groves et al., 1998). From the literature, the indication for its use are mainly in spasticity due to spinal cord lesions. It has been particularly used in encephalomyelitis disseminata patients (Nance et aI., 1997; Stien et aI., 1987). As compared to baclofen, tizanidine might have some better action also in spasticity of cerebral origin. Again, there is little information about the possible functional changes resulting from tizanidine treatment. Benzodiazepines as (e.g. diazepam, tetrazepam) GABA A receptor agonists exert a number of pharmacodynamic effects, one of which is muscle relaxation. The mechanism of antispastic action is to enhance the efficiency of GABAergic transmission by facilitating or potentiating the postsynaptic effects of GABA, which results in an increased chloride conductance and membrane hyperpolarization, thereby increasing the presynaptic inhibition of la afferents in the spinal cord. Somnolence, fatigue, muscular weakness, dizziness,

vertigo, ataxia, habituation, addiction, and physical dependence are among its drawbacks. The indications for the use of benzodiazepines are mainly in the treatment of spasticity of spinal origin, particularly incomplete spinal cord injuries for patients with multiple sclerosis. Use in patients with spasticity of cerebral origin, traumatic brain injuries, cerebral palsy, vascular accidents, is less well documented in the literature. Dantrolene acts directly at muscle fibers. It decreases the amount of Ca release from the endoplasmic reticulum, thereby reducing the effectiveness of excitation-contraction coupling and thus reducing the contraction of skeletal muscles. Although it does depress the central nervous system, this does not seem to be the mechanism of its antispastic action. It effectively reduces increased muscle tone and spasms both in spinal and cerebral spasticity. General weakness, diarrhea, occasional liver toxicity, which can be fatal, and to a lesser extent drowsiness and dizziness limit its use. Average doses and maximal doses of major antispastic drugs can be taken from Table 7.

38.6.2. Local injection of neurolytic agents Local intramuscular injection of phenol or alcohol for treatment of spasticity is an approach known for decades. These agents induce a focal chemodenervation. The mechanism by which this is achieved, is partly by denaturation, but also by nonselective tissue destruction, which includes nerve coagulation and muscle necrosis. In using this approach, there are two possible strategies. One is motor nerve block, which can be achieved by low volume perineural injection, this leads to a complete blockade of the nerve and complete reduction of muscle tone. The other approach is motor point block, which is Table 7 Average doses and maximal doses of major antispastic drugs. Drugs

Average dose/day

Maximal dose/day

Baclofen

3x 10 mg

up to 150 mg

Tizanidine

3x 4mg

24mg

Tetrazepam

3x25 mg

400mg

Dantrolene

3x50 mg

400mg

626 much more difficult to apply, because it is targeted to small motor nerve branches, consequently multiple intramuscular injections are necessary. The indications for this approach are mainly focal spasticity and targets are mainly the proximal muscles and particularly the proximal muscles in the lower limb. The clinical effects are reduction of muscle tone and clonus without improvement of strength or involuntary contraction. One possible advantage is the long duration of the effect. Alcohol and phenol injections are not totally safe procedures as there are significant side effects. Patients may complain of dysesthesia, causalgia or even neuralgic pain. In addition, definite tissue damage occurs, that can lead to edema or venous thrombosis. There are no controlled studies which show data on functional parameters such as gait, or daily living activities. 38.6.3. Intrathecal therapy with bacIofen The rationale for the intrathecal use of baclofen was to avoid the systemic side effects, which can be dose limiting, through direct administration of the drug in the vicinity of the spinal cord, the site at which baclofen exerts its antispastic action. Chronic, intractable spasticity contributes significantly to the disability of patients especially with multiple sclerosis or spinal cord injury. Severe spasticity often causes pain, interrupts sleep, and interferes with function. Baclofen can be delivered safely and accurately by a programmable drug pump which permits flexibility with dosing and thus accomodates the individual lifestyle of each patient. Careful patient selection and screening are required. Patients must satisfactorily respond to a trial dose of 25-100 ug applied directly via a lumbar puncture cannula. Patients who meet this screening criteria may be elected to have the drug pump then implanted to facilitate continuous delivery. During the postoperative period, all antispastic medications are gradually discontinued. The intrathecal dose of baclofen is titrated according to the patient's degree of spasticity and level of function. Adverse effects associated with intrathecal baclofen include drowsiness, weakness in the lower extremities, dizziness, seizures, headache, and nausea. The majority of patients are treated with daily doses between 100-400 ug, Intrathecal baclofen therapy is mainly used in patients with severe spasticity of lower limbs; spasticity in the upper extremities is difficult

R. BENECKE

to treat because higher doses are necessary and by diffusion of the drug can induce side effects at brain stem. The same problem exists when the catheter is placed more cranially at the cervical spinal cord instead of the common placement at a midthoracic level. 38.6.4. Intramuscular injection ofbotulinum toxin type A Currently botulinum toxin is used in a wide variety of conditions that are characterized by focal muscle hyperactivity. Botulinum toxin takes effect gradually after several days to weeks and causes clinically detectable weakness and muscle atrophy for 2-4 months in most situations, sometimes even longer. The effects wear off gradually, the degree and the duration of muscle weakness are dose-dependent. The antispastic effect of botulinum toxin injected to certain muscle groups is also completely reversible after 3-4 months. Therefore, repeat injections are often necessary lifelong. Botulinum toxin works by causing chemical denervation of intrafusal and extrafusal muscle fibers by blocking the release of acetylcholine. The most important antispastic effect of botulinum toxin is exerted by weakening the extrafusal muscle fibers, denervation also of intrafusal muscle fibers forming the muscle spindle apparatus may also playa role by reducing Ia afferent fiber activity and thereby decreasing activity of spinal ce-motoneurons. Local botulinum toxin A injections are especially indicated in spastic syndromes in which increased muscle tone, spasms and pain are focused on a certain group of muscles. The total dose is limited to about 500 units of Botox" and 2000 units of Dysport", respectively. These maximal doses are not sufficient to treat muscles at all four extremities in a severe tetraspastic patient. Commonly used total dosages of Botox" and Dysport" are, however, helpful to treat distal or proximal muscles of one arm or one leg. Higher doses cannot be applied because of considerable systemic side effects such as general weakness and dysphagia. When considering the literature on the use of botulinum toxin in spasticity two findings are remarkably constant, irrespective of whether obtained from open or controlled studies. The resistance to passive movement was invariabily reduced in injected muscles, whichever scale was used to

MOVEMENT DISORDERS: SPASTICITY

measure resistance to passive movement. The second consistant finding has been the difficulty of demonstrating functional improvement from botulinum toxin injections in spastic syndromes. Whenever treating a patient with a spastic syndrome it is important to define the goals of treatment clearly. Botulinum toxin use may not aim to produce longterm gains in active function, the focus may be more on improvement in comfort and passive function, on pain relief, prevention of contractures, easing carer burden and cosmesis. References Ashworth, B (1964) Preliminary trials of carisoprodol in multiple sclerosis. Practitioner, 192: 540-542. Beer, RF, Dewald, IP and Rymer, WZ (2000) Deficits in the coordination of multijoint arm movements in patients with hemiparesis: evidence for disturbed control of limb dynamics. Exp. Brain Res., 131: 305-319. Benecke, R (1987) Spasticity/spasms. Clinical aspects and treatment. In: R Benecke, B Conrad and CD Marsden (Eds.), Motor Disturbances I. Academic Press, London, pp. 169-177. Benecke, R (1993) The role of the corticospinal tract in spasticity studied by magnetic brain stimulation. In: AF Thilmann, DJ Burke, WZ Rymer, AF Thilmann, DJ Burke and WZ Rymer (Eds.), Spasticity: Mechanisms and Management. Springer Verlag, Berlin, Heidelberg, New York. Bohannon, RW and Smith, MB (1987) Inter-rater reliability of a modified Ashworth scale of muscle spasticity. Phys. Ther., 67: 206-207. Brown, P (1994) Pathophysiology of spasticity. J. Neurol. Neurosurg. Psychiatry, 57: 773-777. Burke, D, Hallett, M, Fuhr, P and Pierrot-Deseilligny, E (1999) H-reflexes from the tibial and median nerves. The International Federation of Clinical Neurophysiology. Electroencephalogr. Clin. Neurophysiol., 52 (Suppl.): 259-262. Capaday, C (1997) Neurophysiological methods for studies of the motor system in freely moving human subjects. J. Neurosci. Meth., 74: 201-218. Dietz, V and Berger, W (1983) Normal and impaired regulation of muscle stiffness in gait: a new hypothesis about muscle hypertonia. Exp. Neurol., 79: 680-687. Dietz, V, Quintern, 1 and Berger, W (1981) Electrophysiological studies of gait in spasticity and rigidity: evidence that altered mechanical properties of muscle contribute to hypertonia. Brain, 104: 431-449.

627 Fellows, SI, Kaus, C and Thilmann, AF (1994) Voluntary movement at the elbow in spastic hemiparesis. Ann. Neurol., 94: 397-407. Fowler, EG, Nwigwe, AI and Ho, TW (2000) Sensitivity of the pendulum test for assessing spasticity in persons with cerebral palsy. Dev. Med. Child Neurol., 42: 182-189. Goldberger, ME and Murray, M (1973) Patterns of sprouting and implications for recovery of function. In: SG Waxman (Ed.), Advances in Neurology. Raven Press, New York, pp. 361-385. Groves, L, Shellenberger, MK and Davis, CS (1998) Tizanidine treatment of spasticity: a meta-analysis of controlled, double-blind, comparative studies with baclofen and diazepam. Adv. Ther., 15: 241-251. Hallett, M, Berardelli, A, Delwaide, P, Freund, HI, Kimura, 1, LUcking, C et al. (1994) Central EMG and tests of motor control. Report of an IFCN committee. Electroencephalogr. Clin. Neurophysiol., 90: 404-432. Ibrahim, IK, Berger, W, Trippel, M and Dietz, V (1993) Stretch-induced electromyographic activity and torque in spastic elbow muscles. Differential modulation of reflex activity in passive and active motor tasks. Brain, 116: 971-989. Kimura, 1, Daube, 1, Burke, D, Hallett, M, Cruccu, G, Ongerboer de Visser, BW et al. (1994) Human reflexes and late responses. Report of an IFCN committee. Electroencephalogr. Clin. Neurophysiol., 90: 393-403. Leslie, GC, Muir, C, Part, NI and Roberts, RC (1992) A comparison of the assessment of spasticity by the Wartenberg Pendulum test and the Ashworth grading scale in patients with multiple sclerosis. Clin. Rehab., 6: 41-48. McCouch, GP, Austin, GM and Liu, CY (1957) Sprouting as a cause of spasticity. J. Neurophysiol., 21: 205-216. Nacimiento, W, Mautes, A, Toepper, R et al. (1993) B-50 (GAP-43) in the spinal cord caudal to hemisection: Indication for lack of intraspinal sprouting in dorsal root axons. J. Neurosci. Res., 35: 603-617. Nacimiento, W, Sappok, T, Brook, GA et al. (1995) B-50 (GAP-43) in the rat spinal cord caudal to hemisection: lack of intraspinal sprouting by dorsal root axons. Neurosci. Lett., 194: 13-16. Nacimiento, W, Brook, GA and Noth, 1 (1997) Lesioninduced neuronal reorganization in the spinal cord: morphological aspects. In: HI Freund, BA Sabel and OW Witte (Eds.), Brain Plasticity. Advances in Neurology (Vol. 73). Lippincott-Raven, Philadelphia, pp.37-59. Nance, PW, Sheremata, WA, Lynch, SG, Vollmer, T et al. (1997) Relationship of the antispasticity effect of tizanidine to plasma concentration in patients with multiple sclerosis. Arch. Neurol., 54: 731-736.

628 Pierrot-Deseilligny, E (1997) Assessing changes in presynaptic inhibition of Ia afferents during movement in humans. 1. Neurosci. Meth., 74: 189-199. Pisano, F, Miscio, G, Del Conte, C, Pianca, D, Candeloro, E and Colombo, R (2000) Quantitative measures of spasticity in post-stroke patients. Clin. Neurophysiol., 111: 1015-1022. Pullman, SL, Goodin, DS, Marquinez, AI, Tabbal, S and Rubin, M (2000) Clinical utility of surface EMG: report of the therapeutics and technology assessment subcommittee of the American Academy of Neurology. Neurology, 55: 171-177. Saruhashi, Y,Wise, Y and Perkins, R (1996) The recovery of 5-HT immunoreactivity in lumbosacral spinal cord and locomotor function after thoracic hemisection. Exp. Neurol., 139: 203-213.

R. BENECKE

Stien, R, Nordal, HI, Oftedal, SI and Slettebo, M (1987) The treatment of spasticity in multiple sclerosis: a double-blind clinical trial of a new anti-spastic drug tizanidine compared with baclofen. Acta. Neurol. Scand.,75: 190-194. Thallmair, M, Metz, GS, Z'Graggen, WI, Raineteau, 0, Kartje, GL and Schwab, ME (1998) Neurite growth inhibitors restrict plasticity and functional recovery following corticospinal tract lesions. Nat. Neurosci., 2: 124-131. Thilmann, AF, Fellows, SI and Garms, E (1991) The mechanism of spastic muscle hypertonus: variation in reflex gain over the time course of spasticity. Brain, 114: 233-244. Wartenberg, R (1951) Pendulousness of the legs as a diagnostic test. Neurology, 1: 18-24.

Movement Disorders Handbook of Clinical Neurophysiology, Vol. 1 M. Hallett (Ed.) © 2003 Elsevier B. V. All rights reserved

629 CHAPfER39

Psychogenic movement disorders M. Hayes" and P.O. Thompson?" b

a Department of Neurology, Concord Repatriation Hospital, Sydney, NSW, Australia Department of Neurology and University Department ofMedicine, Royal Adelaide Hospital and University ofAdelaide, Adelaide, SA 5000, Australia

39.1. Introduction The term "psychogenic movement disorder" (PMD) refers to a movement disorder that does not have a recognized pathophysiological or "organic" basis. These disorders constitute a spectrum of behaviors ranging from the unconscious manifestations of hysteria, conversion reactions or somatization to the conscious actions of malingering. From a clinical standpoint, there are two main problems with the conventional view of hysteria (Ron, 2001). First, there is the conceptual leap of accepting that unconscious psychodynamic forces, perhaps as part of a dissociative state, in response to a physical stress or psychological conflict, are shaping this behavior and deciding whether these are present in a particular individual. Second, and of practical significance, is the difficulty in distinguishing between feigned (malingering) or unconscious (conversion disorder) motives. Recent investigation with functional imaging of the brain in hemiparesis due to a conversion disorder has shed light on this difficult question (Vuilleumier et al., 2001). The demonstration that different physiological processes may be responsible for feigning and conversion serves the dual purpose of giving credibility to the diagnosis and avoids reliance on a diagnosis of exclusion. The extent to which this approach can be applied to "positive" neurological phenomena such as PMDs as opposed to paresis remains to be explored. Electrophysiological investigation is useful in identi-

fying and characterising reflex responses and certain involuntary movements, thereby distinguishing them from PMDs. However, to date, these techniques cannot distinguish between voluntary movements and those caused by a conversion disorder, that are by definition unconscious, and therefore involuntary. The behavior of patients presenting with a PMD usually mimics a recognizable entity such as tremor, myoclonus, dystonia or hemiparesis. The majority of patients with PMD believe the movements have a physical basis and the clinical presentation can be so convincing that distinction from an organic movement disorder is extremely difficult (Fahn, 1994). Since the range of organic movement disorders is extensive, virtually any psychogenic presentation has the potential to cause diagnostic confusion. The task of the physician is to make an accurate diagnosis thereby allowing appropriate management. The misdiagnosis of a physical disorder as psychogenic is particularly feared by physicians. Uncertainty in diagnosis and a reluctance to apply the label of "PMD" frequently result in extensive investigation and as a result, the diagnosis of PMD often remains one of exclusion. Nevertheless, clues from the clinical history, the physical examination and in some cases, investigation, alert the physician to the possibility of a psychogenic cause and assist in arriving at this diagnosis. 39.2. Clinical aspects

39.2.1. History

* Correspondence to: Prof. P.D. Thompson, Department

of Neurology and University Department of Medicine, Royal Adelaide Hospital and University of Adelaide, Adelaide, SA 5000, Australia.

E-mail address:[email protected] Tel.: +61 (8)82225502; fax: +61 (8)8223 3870.

The medical history is likely to offer many clues. The onset of the symptoms in PMD is frequently abrupt. This sudden onset invariably raises the possibility of a vascular event despite the absence of appropriate risk factors or radiological evidence of

630

vascular disease or its consequences. The abrupt onset may coincide with, or relate to, a minor injury or emotional disturbance, but this is not always the case. Occasionally, a PMD appears with delayed onset, but unlike delayed-onset dystonia and spasticity following cerebral hypoxia (Hawker and Lang, 1990) or trauma (Scott and Jankovic, 1996) there is no clinical or radiological evidence of the major pathology that follows these insults. Perhaps the most vital clue in the history is the variability of symptoms. There is often a striking disparity between the severity of symptoms while under observation during the consultation and the disability inferred or reported from the patient's account of daily activities. Frequently, marked fluctuations in the severity of symptoms will be described, including periods of complete normality. Although a number of rare disorders such as episodic ataxias (Gordon, 1998), paroxysmal dyskinesias (Bressman et al., 1988) and DOPA-responsive dystonia (Segawa et al., 1976) are characterized by marked symptom fluctuation, this remains an unusual pattern in most organic disorders. Indeed, nearly two thirds of cases presenting with paroxysmal nonkinesigenic dyskinesias are psychogenic in origin (Bressman et aI., 1988). Mild degrees of somatization are common (Lipowski, 1988; Mace and Trimble, 1991) but a history of chronic multiple, wide-ranging physical symptoms in a young person as defined by Somatization Disorder (DSM-IV, 1994) or Briquet's syndrome affects only a small minority of those with PMD. Evidence of anxiety or depression should always be sought since these are the commonest psychiatric disturbances accompanying PMD (Crimlisk et al., 1998), though many patients deny mood disorder. It is important to obtain independent confirmation of the history whenever possible. This may reveal previous episodes of psychogenic disturbance, perhaps many years earlier that had been forgotten or omitted from the history. At the extremes of age PMD is rare and it is prudent to avoid this diagnosis in the elderly without a pre-existing history. Anxiety is a common cause of mild embellishment of physical signs in the elderly and often responds to simple physical therapy and reassurance. Organic and psychogenic disorders are by no means mutually exclusive as illustrated by the common association of pseudo-seizures with epilepsy. Diagnosis can be

M. HAYES AND P.O. THOMPSON

contentious in such circumstances often requiring extensive investigation. Sometimes this can only be resolved in hindsight following sudden and complete cure by suggestion or placebo. Although there appears to be a gender difference with a preponderance of young women with PMDs this is probably culturally-based and it is notable that the largest epidemic of this disorder occurred in young men during the, 1914-18 war. The traditional view is that a PMD will resolve only when the prospect or need, of either primary or secondary gain disappears. Indeed, in the absence of identifiable gain psychiatrists are often unwilling to accept a diagnosis of psychogenic disorder. Although the history may reveal an obvious benefit from the changed behavior it is not unusual for this to remain obscure. It should also be remembered that physical disease is not devoid of these dynamics.

39.2.2. Physical examination PMDs can be viewed as a form of non-verbal communication or body language in which the movements are often dramatic yet the meaning remains elusive. The resulting movement disorders, both hypokinetic and hyperkinetic are typically modelled on previous personal or family experience. As with all movement disorders, the physician attempts to categorize the movements based initially on observation. Frequently, when assessing PMDs, the initial clinical impression is that the movements simply do not look organic although this view is often difficult to justify. The movements may be variable, fluctuate in distribution and appearance. Combinations of shaking, spasms and jerking simulating tremor, dystonia and myoclonus, along with gait disorder, represent the commonest manifestations of PMDs. One of the cornerstones of diagnosis in PMD is the absence of relevant objective neurological signs on examination. A marked discrepancy between apparent disability and physical signs should also raise the possibility of psychogenic factors. There are some caveats such as dystonias and conditions such as isolated gait ignition failure (Nutt et al., 1993) or midline cerebellar lesions (Adams, 1976) in which signs may not be elicited routinely. The second key feature is variability in the physical signs. For example, marked fluctuations in the severity of movements and inconsistencies in

PSYCHOGENIC MOVEMENT DISORDERS

appearance may seem discordant for organic disorders. Although a degree of fluctuation and variability are common in organic neurological disorders, they are rarely orchestrated by the examiner. In contrast, the movements of a PMD may not be evident during the history but emerge during physical examination. Movements may disappear when the patient is unaware they are under observation. This variability can be manipulated by various forms of distraction such as asking the patient with limb tremor to perform a repetitive or rhythmic task with the contralateral hand. This maneuver typically leads to diminution or cessation of the tremor. Repetitive movements of the contralateral hand may also entrain a psychogenic or voluntary tremor to the same frequency. Another variation on this theme is to ask the patient to perform a novel task such as walking backwards when a psychogenic gait disturbance is suspected. This usually results in excessively slow progress or an evolving gait pattern as the patient attempts to adjust to the new task. It is important to remember however that a dystonic gait may also improve following such an instruction. These techniques are useful when a PMD is continuous or present most of the time. Conversely, the opposite strategy of suggestion may elicit or enhance a movement disorder by drawing attention to it when the movement is paroxysmal or infrequent. Less reliable clinical clues are the presence of 'give-way' weakness, a positive Hoover's sign suggesting inadequate effort when testing muscle strength and apparent sensory loss in a nonanatomical pattern. Each may be seen in mild degrees of somatization and anxiety (Gould and Miller, 1986). There is often an 'overflow' effect in PMDs whereby other abnormalities, including a range of different movements or other embellishments emerge as the examination proceeds. An abnormal gait and a variety of speech abnormalities are common in this context. Several distinctive features including exaggerated effort, extreme fatigue, and marked slowness of movement, which are characteristically disproportionate to any apparent weakness, provide further clues to a PMD. These are most commonly seen in hypokinetic PMDs such as hemiparesis or gait disturbance. A staggering or lurching gait, usually without falls, is physically demanding and requires excellent balance to maintain equilibrium, arguing

631 against a significant disturbance of motor control. Similarly, tight-rope walking with prominent rescue responses, marked sway with Romberg's test, convulsive shaking (often seen when asked to stand on one leg) and exaggerated genuflexion with each step, all suggest problems with balance and walking but paradoxically attest to the contrary. As Lempert et al. (1991) emphasized, patients with physical disease tend to minimize energy expenditure and the display of excessive energy or maintenance of an uneconomic posture can be a clue to a PMD. For example, camptocormia (literally bent trunk), in which excessive trunkal flexion leads to forward displacement of the center of gravity, was originally associated with PMDs in fit young men. It should be emphasized that a bizarre gait is not pathognomonic of a PMD, since an unusual gait pattern is common in Sydenham's chorea, neuroacanthocytosis and Huntington's disease. However, diagnostic confusion is rare in such cases because of the clinical context in which the movements emerge. When there is additional weakness, the pattern of weakness in a PMD does not conform to recognized anatomical patterns such as predominant weakness of distal muscles and upper limb extensors and lower limb flexors in hemiparesis due to corticospinal tract lesions. The characteristic postures of upper limb pronation and foot inversion in corticospinal tract lesions are often not evident in psychogenic weakness and there are no associated reflex changes. In psychogenic dystonia, the onset of the abnormal posture is often acute with rigidity and presents as a static or fixed dystonic posture in contrast to the typical presentation of primary dystonia with an action dystonia (Fahn and Williams, 1988). In addition to the sudden onset, remission and variability, psychogenic tremor may persist in unusual combinations during rest, posture and movement and vary in frequency particularly with distraction or entrainment (Deuschl et al., 1998; Koller et al., 1989). Co-activation of antagonist muscles is a common mechanism in psychogenic limb tremor and can be detected as an increase in muscle tone during passive rotation of the wrist. The tremor disappears during complete relaxation though this may be difficult to achieve (Deuschl et aI., 1998). Finally, the affect and physical appearance of patients are important sources of further clues to a PMD but this remains a highly subjective assessment. Anxious and emotional patients with organic

632

M. HAYES AND P.D. THOMPSON

disorders may embellish the clinical picture in a confounding way whereas depressed or stoic patients may be misconstrued as having 'la belle indifference' .

analysis of myoclonus and deciding whether it is of cortical, brainstem or spinal origin.

39.3. Investigation

Tremor is a common manifestation of a PMD and in some series is the commonest PMD (Factor et al., 1995). The arm is most commonly involved in psychogenic tremor and is accompanied by lower limb tremor in 28% and generalized tremor in 20% of cases (Deuschl et al., 1998). Psychogenic tremor is more restricted in distribution than parkinsonian or essential tremor (O'Suilleabhain and Matsumoto, 1998). The rhythmic sinusoidal movements of psychogenic and organic tremors are produced by reciprocal alternating bursts of muscle activity in antagonist muscle pairs such as the forearm flexors and extensors. Alternating tremors generally have a frequency of less than 6-8 Hz but there is considerable overlap between different organic and psychogenic tremors within and below this frequency range. For example, the frequency of parkinsonian tremor was 2.5-6.8 Hz, essential tremor 4.3-6.1 Hz and psychogenic tremor 4.5-7.1 Hz in the study of 0' Suilleabhain and Matsumoto (1998). Normal subjects mimicking tremor typically perform rhythmic oscillatory movements at a frequency of 5.5-6.5 Hz (Britton et al., 1993). Tremor frequency alone is a useful discriminating factor only for frequencies greater than 11 Hz, such as 16 Hz orthostatic tremor, which cannot be imitated and therefore are always organic. In an individual patient, the frequency of a psychogenic tremor is more variable than organic tremors. Psychogenic tremors can be entrained by a variety of maneuvers such as finger or toe tapping, or repetitive movements with another body part in response to an external rhythmic signal from a metronome. Entrainment is detected by the tremor adopting a new frequency similar to, or identical with, that of the repetitive movement of the remote body part. Counting or mental arithmetic may cause a psychogenic tremor decompose into an intermittent variable jerky movement, losing the previous rhythmicity or abate completely. Similarly, the instruction to execute purposeful or skilled movements with the tremulous limb in PMD may provoke

39.3.1. Neurophysiological investigation

Neurophysiological investigation provides an objective method of examining the electromyographic characteristics of the movement disorder, the cerebral activity preceding movement or involved in movement preparation and the reflex basis of the movements (Brown and Thompson, 2001). The information derived from these tests can be compared with normal motor behavior and reflex function. These techniques are best suited to the analysis of simple stereotyped movements such as jerks, muscle spasms and tremor that occur reasonably frequently. Where the movements are more complex these methods are of less value because of the many technical difficulties, including movement artefact, limitations in the number of recording sites and the capacity for analysing vast amounts of electromyographic and electroencephalographic data. The diagnostic value of neurophysiological tests in a suspected PMD is maximized by selecting a test that addresses a specific physiological mechanism relevant to the movement. For example, looking for stereotyped cerebral activity preceding a jerk, or measuring the latency between a stimulus and a stimulus induced jerk. The absence of predicted abnormal reflex behavior may also provide further clues. Simple electromyographic recordings may reveal the characteristic motor unit and muscle fiber discharges of myokymia in cramps due to neuromyotonia or the hypersynchronous short bursts of motor unit activity in cortical myoclonus. However, since most movement disorders are driven by central nervous system activity, the pattern of motor unit discharge and recruitment will usually not distinguish between a psychogenic and an organic source. Accordingly, it will be necessary to record activity from a number of muscle groups simultaneously and examine the characteristics of the muscle bursts and their relationship to each other. For example, the identification of stereotyped patterns of muscle activation is particularly useful in the

39.3.2. Tremor

PSYCHOGENIC MOVEMENT DISORDERS

an excessively slow and deliberate performance with marked changes in tremor frequency. There may be complete disruption of the tremor or the tremor may become intermittent. In contrast, organic tremors are usually accentuated by these maneuvers. Faster tremors (> 8 Hz), such as physiological tremor, are of smaller amplitude and generally caused by cocontraction between antagonist muscle pairs. Electromyography shows cocontraction between antagonist muscles with grouping of muscle discharges at frequencies around 8-10 Hz rather than discrete, alternating tremor bursts. Cocontraction, detected at the onset of a tremor as an increase in muscle tone, which subsides along with the tremor during relaxation has been emphasized as an important diagnostic feature of psychogenic tremor (Deuschl et al., 1998). Persistent cocontraction of antagonist muscles enhances physiological tremor in normal subjects, particularly when the muscles fatigue and motor unit recruitment becomes synchronized (Young and Hagbarth, 1980). Synchronous motor unit discharges produce additional jerky movements superimposed upon the physiological tremor. Such movements are particularly pronounced during slow voluntary movements and may interfere with the performance of fine manual tasks (Young and Hagbarth, 1980). This mechanism may contribute to faster (8-10 Hz) psychogenic tremors where cocontraction of muscle is readily detected by manipulation of the limb during attemptl;d rest or voluntary activation of the upper limb. It will also be evident on electromyographic recordings. Loading the limb with weights increases the amplitude of cocontracting psychogenic tremor, though this may also be seen in physiological tremor (Deuschl et al., 1998). Attempts to reset tremor by peripheral nerve stimulation or mechanical perturbation of the limb do not distinguish between essential, parkinsonian or voluntary mimicked tremors (Britton et al., 1992; Britton et al., 1993).

39.3.3. Jerks Neurophysiological methods are particularly useful in the analysis of jerks and distinguishing myoclonus and tics from psychogenic jerks. The duration of muscle bursts, pattern of muscle recruitment, cerebral activity preceding jerks and the

633

characteristics of reflex responses yield important clues to the origin of jerks. Myoclonus is typically caused by brief bursts of muscle activity, 25-50 ms in duration, as in cortical and reticular reflex myoclonus. The brief bursts of muscle activity are recorded simultaneously in antagonist muscle pairs. Myoclonus of brainstem or spinal origin usually recruits a number of muscle groups, often bilaterally, in a specific order or pattern of activation..Bursts of muscle activity in hyperekplexia, the auditory startle response and spinal myoclonus are generally longer than 100 ms. Jerks in dystonia (which are often stereotyped and restricted to a few muscle groups) and chorea (in which jerky movements flow randomly from one body part to another) may last up to 500 ms. The electromyographic pattern of psychogenic jerks is often organized in a triphasic sequence of agonist-antagonist-agonist muscle activation as occurs during voluntary ballistic (rapid) movements (Thompson et al., 1992). The duration of bursts of voluntary muscle activity in a ballistic movement is generally within the range of 75-150 ms, depending on the amplitude and speed of movement. When doubt persists about the origin of jerks, backaveraged electroencephalographic activity (EEG) preceding the jerk (jerk-locked averaging) may provide evidence of a cortical origin. In cortical myoclonus, a cortical spike or sharp wave is detected in the EEG preceding the muscle jerk. The interval between cortical discharge and myoclonus is approximately, 20 ms for hand muscles and 40 ms muscles of the foot, corresponding to the time taken for conduction from motor cortex to muscle. This technique is particularly useful when the routine surface EEG contains apparent discharges that may be caused by movement artefact. Backaveraging will demonstrate whether the discharges are coincident with or precede the limb jerking. Not all cases of cortical myoclonus have recordable cortical discharges preceding myoclonus. The finding of enlarged cortical somatosensory evoked potentials also provides a strong argument in favor of a cortical origin for myoclonus. Voluntary movement is preceded by a slow negative movement related cortical potential (MRCP) recorded in the EEG during the final 1-2 s before movement. Simple motor tics (Obeso et al., 1981) and complex stimulus induced tics (Tijssen et al., 1999) are not preceded by a premovement

634

potential, whereas voluntary movements that mimic the tics are preceded by a MRCP (Obeso et al., 1981). These observations suggest the cerebral preparation for movement and execution of motor commands for voluntary movements and tics utilize different pathways and this technique may be useful in distinguishing between voluntary and involuntary movements (Obeso et al., 1981; Terada et al., 1995). The MRCP is thought to be generated by an internal signal to move during self-paced movements. However, when normal subjects perform movements in response to external cues, this cerebral wave is absent or abbreviated (Papa et al., 1991). A similar short or abbreviated negative cerebral potential may also be seen during the final 100 ms before simple motor tics (Karp et al., 1996) and myoclonus in myoclonic dystonia (Quinn et al., 1988). The distribution and appearance of this waveform was similar to the late (NS') component of the normal MRCP (Quinn et al., 1988; Karp et al., 1996). These observations suggest such movements may utilize some of the neural circuitry involved in the preparation for voluntary movement. Alternatively, an internal trigger or cue that "releases" movements or tics, as in Tourette's syndrome, may attenuate or abolish a premovement potential without necessarily implying a motor control mechanism that differs from voluntary movement. Conversely, repetitive voluntary movements, cued in a particular manner, might not be preceded by a MRCP. Accordingly, the absence of a premovement potential does not prove a movement is involuntary (Karp et al., 1996). Measuring the latency to onset of a stimulus induced response in a reflex jerk is particularly useful in distinguishing between reflex myoclonus and a voluntary reaction or stimulus "contingent" jerks. Reflex latencies greater than 100 ms, variability in response latency and the pattern of resulting movement are suggestive of a voluntary reaction to the stimulus, particularly if the EMG pattern consists of an organized sequence of muscle activation as in a ballistic movement (Thompson et al., 1992). The reflex latencies for cortical myoclonus, brainstem myoclonus and voluntary reaction times to various stimuli in different muscles are shown in Table 1. As with spontaneous jerks, the pattern of "reflex" muscle activation also provides important diagnostic information. Complex responses organized into

M. HAYES AND P.O. THOMPSON

triphasic bursts, as in a voluntary ballistic movement, are suggestive of a voluntary response. Voluntary responses are also more variable in the extent and distribution of muscle recruitment in contrast to the simple stereotyped brief bursts of muscle activity in myoclonus. The normal auditory startle response consists of a blink accompanied by a brisk jerk of the neck and upper body in response to a sudden or unexpected noise. An unexpected visual stimulus or tap to the body may also lead to a similar "jump" or startle response. The response habituates rapidly with repetition of the stimulus. In contrast, an exaggerated auditory startle response or hyperekplexia does not habituate and stereotyped bilateral synchronous upper body jerks persist after repeated auditory stimuli and also in response to taps to the face, forehead, nose and upper body. A voluntary auditory startle response is readily identified by the exaggerated and variable response unlike the simple stereotyped startle response. Recording the latency and pattern of muscle responses after auditory stimulation may be needed to confirm a voluntary reaction to sound. The normal auditory startle response habituates after two or three trials and is followed at a variable interval by any voluntary reactions to the stimulus. Accordingly, recording the startle response on a sweep time of one second or so will ensure any late responses after the early auditory reflex response are captured. Finally, it is particularly important to observe the patient during the test procedure and correlate the observed stimulus induced jerk with the electrophysiological result to confirm the psychogenic jerk follows a normal, habituating auditory startle.

39.3.4. Spasms

In PMD, jerking is often associated with spasms some of which may be prolonged. The differential diagnosis of spasms encompasses many conditions (Thompson, 1993). Dystonia and the stiff-man syndrome are often misdiagnosed as PMD (and vice versa) and neurophysiological tests may be useful in this differential diagnosis. The definition of dystonia emphasizes the sustained muscle spasms that twist body parts into abnormal postures, upon which tremor and myoclonus may be superimposed. The importance of the postural mobility in dystonia is often neglected. In

635

PSYCHOGENIC MOVEMENT DISORDERS

contrast, psychogenic dystonia is characterized by fixed postures with reduced movement of the affected limb. A recent review of primary dystonia summarized the associated electrophysiological findings (Berardelli et aI., 1998) (see also Chapter XX). Surface electromyographic recordings may demonstrate prolonged spasms of muscle activity during movement, cocontraction of muscles and overflow of muscle activity to remote muscle groups not normally involved in the movement under study. A reduction in reciprocal inhibition between forearm muscles, particularly at latencies corresponding to presynaptic inhibition has been demonstrated in arm dystonia and those with dystonia at remote sites such as cranial dystonia and torticollis. The utility of this test as a diagnostic procedure, which is performed with the subject at

rest, has not been evaluated but provides an objective assessment of central nervous interneuronal function. The analysis of patterns of motor unit synchronization in dystonic muscle contractions may be another method of distinguishing between voluntary concontraction and dystonic muscle activity. In a recent study, coherence between sternomastoid and splenius capitis muscle activity at 4-7 Hz was both specific and sensitive for cervical dystonia (Tijssen et aI., 2000). Spasm of lumbar paraspinal muscles commonly follows lumbar disc or radicular injury and is often associated with antalgic postures and a restricted range of lumbar movement. In such cases, the question of a PMD or rare conditions such as the stiff man syndrome may be raised. The latter is

Table I Reflex latencies (ms) recorded in A: cortical reflex myoclonus, B: brainstern auditory startle reflexes, hyperekplexia and reflex responses to supraorbital nerve stimulation and taps to the face and head and C: voluntary reaction times in different muscles in response to various stimuli in normal subjects. A: Cortical reflexes stimulate hand stimulate foot B: Brain stem reflexes

Record Orbicularis oculi Masseter Sternomastoid Biceps Tibialis anterior

Cortical reflex myoclonus

Corticobasal degeneration

51.1±3.3 (APB) 85-95 (AHB)

43.1±3.3 (APB) 75-85 (AHB)

Auditory startle

Electrical stimulation supraorbital nerve

Vertex, jaw tap

Normal

Hyperekplexia

Normal

Hyperekplexia

Normal

Hyperekplexia

37 (R2) 59 58 70

32-47

32-36 (R2)

-40

14-19 (Rl) 8-10 (jaw jerk) 15-20

14-17 15-20

-74 -91 -132

31-85 60-120

C: Voluntary reaction times to various stimuli in sternomastoid, biceps brachii and tibialis anterior muscles in normal subjects. The subjects were requested to make a generalized flexion movement in response to the different stimuli. Values are mean and range from 10-20 trials for each stimulus in eight normal subjects.

Auditory Tap knee Tap radius Electrical stimulation median nerve Mean (all stimuli)

Sternomastoid

Biceps

Tibialis anterior

115 ( 81-140) 122 ( 90-153) 113 ( 89-138) 137 (104-176) 122

110 (76-138) 116 (93-146) 116 (83-169) 118 (97-141) 115

155(127-180) 153 (125-192) 148 (122-181) 163 (133-189) 155

636

characterized by several reflex abnormalities that can be readily detected with simple neurophysiological tests. First, continuous muscle activity in lumbar paraspinal muscles persists during standing and when lying supine (but disappears during sleep). This activity gives rise to the exaggerated lumbar lordosis in the stiff man syndrome. In addition, exteroceptive reflexes elicited by peripheral nerve stimulation are enhanced, particularly if repetitive stimulation is used to facilitate temporal and spatial summation in interneuronal spinal pathways. These reflexes, elicited clinically by cutaneous stimulation, consist of a brief series of phasic responses followed by a prolonged tonic phase of muscle activity that may be bilateral and recruit other muscle groups. This pattern of reflex muscle activity is a cardinal feature of the stiff man syndrome and is not seen in normal subjects or other causes of muscle spasm (Meinck et aI., 1984). 39.3.5. Functional imaging

These studies involve small numbers of patients and differing methodology but are conceptually similar in that they invoke established concepts of inhibitory inputs acting on common pathways to primary motor cortex. Patients with conversion disorders manifested by unilateral paralysis acted as their own controls allowing comparison of activation processes in both hemispheres. Marshall et al. (1997) discussed the pre-existing evidence for anterior cingulate and orbito-frontal cortex inhibiting spontaneous movement and reported a regional blood flow (rCBF) PET study in which the paradigm involved 'preparation to move' as well as 'attempted movement' in a patient with conversion disorder. Both preparation and attempted movement of the paralyzed leg lead to activation of movement-related areas including dorsolateral pre-frontal cortex (DLPFC) arguing against feigning but no activation of the contralateral pre-motor or primary sensorimotor cortex was evident. However, the contralateral anterior cingulate and orbito-frontal cortex were the only areas differentially activated in a comparison of attempting to move the paralyzed relative to the normal leg. The authors suggested that the anterior cingulate and orbito-frontal cortex combine to inhibit movement of the paralyzed leg by functional disconnection of pre-motor from primary motor cortex. It is of interest that the anterior cingulate is

M. HAYES AND P.D. THOMPSON

also thought to have some role in the integration of motor control with both cognition and emotion (Devinsky et aI., 1995). Spence et al. (2000) proposed that the DLPFC is the site of dysfunction in volitional disorders and compared PET rCBF in subjects with unilateral psychogenic weakness and controls feigning weakness. Left pre-frontal hypofunction was evident in patients with conversion disorder when they attempted to move the affected limb irrespective of symptom lateralization. This finding is interpreted as a disorder of volition and the authors observed that this is consistent with pre-Freudian views of hysteria as a disorder of the will. Vuilleumier et al. (2001) used SPECT to measure rCBF in seven patients with unilateral psychogenic weakness. The authors incorporated two different methodological techniques. They assessed brain functional activation using passive vibratory stimulation of both hands to avoid the ambiguity and uncertainty of what constitutes an attempt to move a paralyzed limb. Second, they compared brain activation following recovery from the conversion disorder, with the patients acting as their own controls. The study showed a decrease in rCBF in thalamus and basal ganglia contralateral to the weak limbs in every subject with resolution of this asymmetry following recovery. The authors postulated this may reflect a functional disorder in the striato-thalamocortical circuits that fonn the basis for both learned and intentional motor activity. Parkinson's disease is the best understood clinical example of dysfunction in this system and both freezing and start hesitation are examples of disruption to volitional acts. This circuitry incorporates reciprocal connections to limbic cortex allowing for the possibility of behavioral modulation by affective areas. The authors also drew comparison with the phenomenon of motor neglect and anosognosia that are both thought to be mediated by striatothalarnic dysfunction. Although each of these studies report regional blood flow hypofunction at different levels of the motor system contralateral to the deficit, this may, in part, reflect the different methodological paradigms. The common thread is that functional asymmetries have been demonstrated in these studies and these changes may reflect underlying neural correlates of what could tum out to be a truly 'functional' disorder.

637

PSYCHOGENIC MOVEMENT DISORDERS

39.4. Treatment

References

It is difficult to discuss rational therapy for a disorder in which both etiology and underlying mechanisms remain uncertain. At some stage it is necessary to discuss the diagnosis with the patient and it is reasonable to assert that, by definition, there is not a physical basis to the disorder. This positive aspect needs to be emphasized and is usually received with a combination of relief and perplexity. It should be emphasized this does not mean that there is 'nothing wrong' since this is clearly untrue. Attempting to resolve this apparent paradox is a core issue when dealing with these disorders and requires dexterity, sensitivity and skilful communication on the part of the physician. The role of the physician is arguably even more pivotal in PMD than in conventional neurology since the diagnosis is perceived as less tangible. The primary task of the physician is to establish a firm diagnosis of PMD since no one else can reliably do this. Equally important, the patient must have enough confidence in the physician to accept the diagnosis and avoid excessive, potentially harmful and unrewarding medical investigations. Despite the absence of physical disease the physician must avoid being seen as dismissive of the problem. The distinction between malingering and hysteria should be clarified with both patient and family so as to avoid misconceptions. If patients are able to accept, or at least consider, the possibility that the psyche is a factor in their presentation, this tends to be a good prognostic factor. However, it may raise the question "do you think I'm crazy?" which requires further reassurance. Although anxiety and depression are common antecedents of PMDs (Crimlisk et aI., 1998), more florid psychiatric diagnoses are rare. It is helpful to explain that various forms of somatization or mindbody interactions are common in medical practice and most patients are familiar with entities such as tension headache, migraine, irritable bowel or skin conditions that can fluctuate with mood. Illustrations such as these increase acceptance of more unusual or extreme disorders as encountered in PMD. Those patients whose symptoms do not resolve rapidly will usually require some form of physical therapy or rehabilitation that should take place in the context of judicious psychotherapy.

Adams, RD (1976) Nutritional cerebellar degeneration. In: PJ Vinken and GW Bruyn (Eds.), Handbook ofClinical Neurology (Vol. 28). North-Holland, Amsterdam, pp.271-283. Berardelli, A, Rothwell, JC, Hallett, M, Thompson, PO, Manfredi, M and Marsden, CD (1998) The pathophysiology of primary dystonia. Brain, 121: 1195-1212. Bressman, SB, Fahn, S and Burke, RE (1988) Paroxysmal non-kinesigenic dystonia. Adv. Neurol., 50: 403-413. Britton, TC, Thompson, PO, Day, BL, Rothwell, JC, Findley, LJ and Marsden, CD (1992) Resetting of postural tremors at the wrist with mechanical stretches in Parkinson's disease, essential tremor and normal subjects mimicking tremor. Ann. Neurol., 31: 507514. Britton, TC, Thompson, PO, Day, BL, Rothwell, JC, Findley, LJ and Marsden, CD (1993) Modulation of postural tremors at the wrist by supramaximal electrical median nerve shocks in essential tremor, Parkinson's disease and normal subjects mimicking tremor. J. Neurol. Neurosurg. Psychiatry, 56: 10851089. Crimlisk, HL, Bhatia, K, Cope, H, David, A, Marsden, CD and Ron, MA (1998) Slater revisited: six year follow up study of patients with medically unexplained motor symptoms. BMJ, 316: 582-586. Deuschl, G, Seifert, G, Heinen, F, Illert, M and Lucking, CH (1992) Reciprocal inhibition of forearm flexor muscles in spasmodic torticollis. J. Neurol. Sci., 113: 85-90. Deuschl, G, Koster, B, Lucking, CH and Scheidt, C (1998) Diagnostic and pathophysiological aspects of psychogenic tremors. Mov. Disord., 13: 294-302. Devinsky, 0, Morrell, MJ and Vogt, BA (1995) Contributions of anterior cingulate cortex to behaviour. Brain, 118: 279-306. Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) (1994) (4th ed.). American Psychiatric Association, Washington, D.C.. Factor, SA, Podskalny, GO and Molho, ES (1995) Psychogenic movement disorders: frequency, clinical profile and characteristics. J. Neural. Neurosurg. Psychiatry, 59: 406-412. Fahn, S (1994) Psychogenic movement disorders. In: CD Marsden and S Fahn (Eds.), Movement Disorders 3, Butterworth Heinemann, Oxford, pp. 359-372. Fahn, S and Williams, DT (1988) Psychogenic dystonia. Adv. Neurol., 50: 431-455. Farmer, SF, Sheean, GL, Mayston, MJ et al. (1998) Abnormal motor unit synchronisation of antagonist muscles underlies pathological co-contraction in upper limb dystonia. Brain, 121: 801-814.

638 Gironell, A, Lopez-Villegas, D, Barbonoj, M and Kulisevsky, J (1997) Psychogenic tremor: clinical, electrophysiologic and psychopathologic assessment. Neurologia, 12: 293-299. Gordon, N (1998) Episodic ataxia and channelopathies. Brain Dev., 20: 9-13. Gould, R and Miller, BL (1986) The validity of hysterical signs and symptoms. J. Nerv. Ment. Dis., 174: 593597. Hawker, K and Lang, AE (1990) Hypoxic-ischaemic damage of the basal ganglia. Case reports and review of the literature. Mov. Disord., 5: 219-224. Karp, BI, Porter, S, Toro, C and Hallett, M (1996) Simple motor tics may be preceded by a premotor potential. J. Neurol. Neurosurg. Psychiatry, 61: 103-106. Koller, W, Lang, A, Vetere-Overfield, B, Findley, L, Cleeves, L, Factor, S, Singer, C and Weiner, W (1989) Psychogenic tremors. Neurology, 39: 1094-1099. Lempert, T, Brandt, T, Dieterich, M and Huppert, D (1991) How to identify psychogenic disorders of stance and gait. Video study in 37 patients. J. Neurol., 238: 140-146. Lipowski, ZJ (1988) Somatization: the concept and its clinical application. Am. J. Psychiatry, 145: 1358-1368. Mace, CJ and Trimble, MR (1991) "Hysteria": "functional" or "psychogenic"? A survey of British neurologists preferences. JR Soc. Med., 84: 471-475. Marshall, JC, Halligan, PW, Fink, GR, Wade, DT and Frackowiak, RSJ (1997) The functional anatomy of a hysterical paralysis. Cognition, 64: BI-B8. Meinck, H-M, Ricker, K and Conrad, B (1984) The stiffman syndrome: new pathophysiological aspects from abnormal exteroceptive reflexes and the response to clomipramine, clonidine, and tizanidine. J. Neurol. Neurosurg. Psychiatry, 47: 280-287. Monday, K and Jankovic, J (1993) Psychogenic myoclonus. Neurology, 43: 349-352. Nakashima, K, Rothwell, JC, Day, BL, Thompson, PD, Shannon, K and Marsden, CD (1989) Reciprocal inhibition between forearm muscles in patients with writer's cramp and other occupational cramps, symptomatic hemidystonia and hemiparesis due to stroke. Brain, 112: 681-697. Nutt, JG, Marsden, CD and Thompson, PD (1993) Human walking and higher level gait disorders, particularly in the elderly. Neurology, 43: 268-279. Obeso, JA, Rothwell, JC and Marsden, CD (1981) Simple tics in Gilles de la Tourette's syndrome are not prefaced by a normal premovement EEG potential. J. Neurol. Neurosurg. Psychiatry, 44: 735-738.

M. HAYES AND P.D. THOMPSON

O'Suilleabhain, PE and Matsumoto, JY (1998) Timefrequency analysis of tremors. Brain, 121: 2127-2134. Panizza, M, Hallett, M and Nilsson, J (1989) Reciprocal inhibition in patients with hand cramps. Neurology, 39: 85-89. Panizza, M, Lelli, S, Nilsson, J and Hallett, M (1990) Hreflex recovery curve an reciprocal inhibition of H-reflex in different kinds of dystonia. Neurology, 40: 824-828. Papa, SM, Artieda, J and Obeso, JA (1991) Cortical activity predecding self-initiated and externally triggered voluntary movement. Mov. Disord., 6: 217-224. Peters, M (1977) Simultaneous performance of two motor activities: the factor of timing. Neuropsychologia, 15: 461-465. Quinn, NP, Rothwell, JC, Thompson, PD and Marsden, CD (1988) Hereditary myoclonic dystonia, hereditary torsion dystonia and hereditary essential myoclonus: an area of confusion. Adv. Neurol., 50: 391-401. Ranawaya, R, Riley, D and Lang, A (1990) Psychogenic dyskinesias in patients with organic movement disorders. Mov. Disord., 5: 127-133. Ron, MA (2001) Explaining the unexplained: understanding hysteria. Brain, 124: 1065-1066. Scott, BL and Jankovic, L (1996) Delayed-onset progressive movement disorders after static brain injury. Neurology, 46: 68-74. Segawa, M, Hosaka, A, Miyagawa, F, Nomura, Y and Imai, H (1976) Hereditary progressive dystonia with marked diurnal fluctuation. Adv. Neurol., 14: 215-233. Sherrington, CS (1910) In: TC Allbut and HD Rolleston (Eds.), Tremor, "Tendon Phenomenon," and Spasm in System of Medicine. London: MacMillan, pp.290304. Shibasaki, Hand Kuroiwa, Y (1975) Electroencephalographic correlates of myoclonus. Electroencephalogr. Clin. Neurophysiol., 39: 455-463. Shibasaki, H, Sakai, T, Nishimura et al. (1982) Involuntary movements in chorea-acanthocytosis: a comparison with Huntington's chorea. Ann. Neurol., 12: 311-314. Spence, SA, Crirnlisk, HL, Cope, H, Ron, MA and Grasby, PM (2000) Discrete neurophysiological correlates in prefrontal cortex during hysterical and feigned disorder of movement. Lancet, 355: 1243-1244. Terada, K, Ikeda, A, Van Ness, PC, Nagamine, T, Kaji, R, Kimura, J and Shibasaki, H (1995) Presence of Bereitschaftspotential preceding psychogenic myoclonus: clinical application of jerk-locked averaging. J. Neurol. Neurosurg. Psychiatry, 58: 745-747. Tijssen, MAJ, Brown, P, Morris, H and Lees, AJ (1999) Acquired startle-induced tics. J. Neurol. Neurosurg. Psychiatry, 67: 782-784.

PSYCHOGENIC MOVEMENT DISORDERS

Thompson, PO (1994) Stiff people. In: CD Marsden and S Fahn (Eds.), Movement Disorders 3. Oxford. Butterworth-Heinemann, pp. 373-405. Thompson, PO, Colebatch, JG, Rothwell, JC, Brown, P, Day, BL, Obeso, JA and Marsden, CD (1992) Voluntary stimulus-sensitive jerks and jumps mimicking myoclonus or pathological startle syndromes. Mov. Disord., 7: 257-262.

639 Vuilleumier, P, Chicherio, C, Assai, F, Schwartz, S, Siosman, 0 and Landis, T (2001) Functional neuroanatomical correlates of hysterical sensorimotor loss. Brain, 124: 1077-1090. Williams, OT, Ford, Band Fahn, S (1995) Phenomenology and psychopathology related to psychogenic movement disorders. Adv. Neurol., 65: 231-257.

Movement Disorders Handbook of Clinical Neurophysiology, Vol. 1 M. Hallett (Ed.) © 2003 Elsevier B.Y. All rights reserved

641 CHAPTER 40

Other gait disorders Lewis Sudarsky* Department of Neurology, Brigham and Women's Hospital and Harvard Medical School, Boston MA 02115, USA

Gait disorders are a common manifestation of neurologic disease, particularly in the elderly, Reliable estimates of prevalence are difficult to obtain, as there are no standard diagnostic criteria. In a study from Durham, North Carolina, 15% of volunteers over 60 were found to exhibit some abnormality of gait on neurologic exam (Newman et al., 1960). In the East Boston Neighborhood Health Study, a degree of shuffling or difficulty with turns was noted in 15% of the population aged 67-74, 29% of those 75-84, and 49% of the population 85 and above (Odenheimer et al., 1994). Gait disorders are particularly important in the elderly because they compromise independence and contribute to the risk of falls and injury (Tinetti, 1988; Rubenstein, 1997). Many older people limit their activity because of concerns about ambulation and fear of falling (Gill et aI., 2001). 40.1. Anatomy and physiology of locomotion For walking to proceed in a steady fashion, there are two tasks which the nervous system must simultaneously address: locomotion (stepping) and balance. Spinal pattern generators and higher centers organize the stepping, specifying the timing, the advance position, the loading and unloading of the limbs. Dynamic balance must be maintained during the gait cycle to prevent instability and falls. The neural pathways which manage these functions are briefly reviewed below. In quadrupedal animals, stepping is organized by spinal pattern generators in the cervical and lumbar cord (Grillner, 1981). Dogs and cats with high spinal

* Correspondence to: Dr. L. Sudarsky, Department of Neurology ASBl-2, Brigham and Women's Hospital, 75 Francis St., Boston, MA 02115, USA. E-mail address: [email protected] Tel.: 617 732-5405; fax: 617 732-6083.

transection can achieve a crude pattern of locomotion on the treadmill, provided their balance is supported. This "fictive locomotion" can be stimulated with levodopa or clonidine, and occurs independent of sensory feedback (Forssberg and Grillner, 1973). It is difficult to produce sustained spinal locomotion in primates, unless descending pathways in at least one ventral quadrant of the spinal cord are preserved (Eidelberg et al., 1981). In the past few years, efforts have been made to train patients with incomplete spinal injuries to utilize spinal stepping to recover elements of walking (Dietz, 1995). Primate bipedal locomotion depends on higher command and control centers in the brainstem and forebrain (Orlovsky, 1972). Eidelberg et al. (1981) demonstrated a form of patterned locomotion with stimulation of the mesencephalic locomotor region, an area near the nucleus cuneiformis, below the superior cerebellar peduncle. This area includes the cholinergic neurons of the pedunculopontine nucleus (PPN), which is a good candidate for a locomotor center. The PPN receives afferent connections from the basal ganglia, cerebellum and motor cortex, and projects to the brain stem reticular nuclei (GarciaRill, 1986). The fastigial nucleus in the cerebellum can also induce expression of locomotor programs from the brainstem and spinal cord (Mori et al., 2000, 2001). Cerebral control is ultimately required to initiate locomotion, provide direction and purpose, and maintain locomotor drive (Mori, 1997; Armstrong, 1988). Stepping commands are passed along descending pathways in the ventral spinal cord, including the reticulospinal and vestibulospinal tracts (Lawrence and Kuypers, 1968; Armstrong, 1988). The anatomic basis of human postural control is less well-defined, but physiologic studies provide some understanding of the mechanisms involved. Long latency postural responses can be recorded from the anterior compartment muscles in the legs

642

beginning more than 100 ms after a perturbation of the support surface (Nashner, 1980). The vestibular nucleus and midline cerebellum participate in this task, and imbalance is evident if these systems are impaired. While rhythmic stepping proceeds independent of sensory feedback, sensory information is needed for maintenance of balance during locomotion in a real world environment. Good quality sensory afferent information is particularly important when walking across uneven surfaces or in poorly-lit areas. Sensory information from the vestibular system, the visual system, proprioception and information on plantar touch are all utilized. Cervical mechanoreceptors and muscle spindle afferents may also playa role (Roll et aI., 1992). A variety of methods exist for measurement of postural responses in human subjects (Chapter 19). 40.2. Classification of gait disorders

Because of the complexity of neural systems which support postural control and locomotion, there are many problems which can arise. Consequently, the gait disorders encountered in clinical practice are heterogeneous, and sometimes multifactorial. Neurologic disorders can overlap with arthritic and antalgic gaits. Sometimes failing gaits appear fundamentally similar even when they are mechanistically different. The disorder of gait we observe clinically is the product of a physiological abnormality and a compensatory response. Some of the salient features such as widened stance and increased double support time are non-specific, and represent biomechanical adaptations to improve stability and efficiency. The classification and diagnosis of gait disorders based on observational gait analysis is thus a difficult challange. Neurologic disorders of gait can be classified based on: (1) common clinical syndromes; (2) etiologic factors; or (3) underlying physiologic mechanisms. Each system has its proponents and its limitations. Nutt (1998) has proposed a classification of gait disorders into eight groups based on clinical characteristics: cautious, weak, stiff, ataxic, veering, freezing, toppling, and bizarre. The advantage of this syndrome based approach is that it lends itself to observational analysis, and is easily understood and applied. It tends to cluster together cases which share similar physiologic mechanism, though many complex cases are not easily handled.

L. SUDARSKY

Classification of gait disorders based on etiology is practical (Sudarsky, 2001). Etiologic diagnosis is the first step in therapeutic intervention. A neurologic evaluation of 50 patients over 65 with an undiagnosed neurologic disorder of gait was successful at identifying the principal etiologic factor in 85% of cases (Sudarsky and Ronthal, 1983). The categories of disease encountered are summarized in Table 1. Sometimes an etiology is not known and cannot be established, or the disorder is multifactorial. This system is also not easily applied to older patients with higher level gait disorders, and patients with more advanced disability. A physiologic approach to the problem was proposed by Nutt, Marsden and Thompson (1993). The lowest level gait disorders included those due to arthritis, neuromuscular disease, and sensory loss. Middle level gait disorders include hemiplegic gait, cerebellar gait, and the extrapyramidal syndromes. The highest level gait disorders were the major focus of this review, and were subdivided further into five categories: cautious gait, subcortical disequilibrium, frontal disequilibrium, isolated gait ignition failure (i.e. pure freezing gait), and frontal gait. The major difficulty with this approach is that the criteria for the five highest level gait disorders are somewhat arbitrary and are not always uniformly applied. The categories do not correlate well with etiologic diagnosis. A patient with progressive supranuclear palsy, for example, might begin with a subcortical disequilibrium or freezing gait, and progress to a frontal gait over time. Efforts have been made to incorporate the strengths of each system into a single framework (Jankovic et al., 2001). 40.3. Specific gait disorders and clinical syndromes

This discussion will focus on some of the commonly observed neurologic disorders of gait as clinical syndromes. The characteristic gait abnormalities and associated physiology are briefly reviewed. Gait disorders due to pain (antalgic gait), limb deformity and arthritis are not included. Developmental disorders are also outside the scope of this discussion.

40.3.1. The cautious gait The term "cautious gait" appears in several clinically-based classifications. This gait, partie-

643

OTHER GAIT DISORDERS Table 1 1980-2

1990-4

Total

Percent

12

20

16.7

Myelopathy

8

Parkinsonism

5

9

14

11.7

Hydrocephalus

2

6

8

6.7

Multiple infarcts

8

10

18

15.0

Cerebellar degeneration

4

4

8

6.7

Sensory deficits

9

13

22

18.3

Toxic/metabolic

3

0

3

2.5

Psychogenic

1

3

4

3.3

Other

3

3

6

5.0

Unknown cause

7

10

17

14.2

50

70

120

Total

100

Classification of gait disorder in 120 patients, according to etiologic cause. Initial fifty patients are from Sudarsky and Ronthal (1983). Seventy additional patients are from Sudarsky (1997). Reproduced from Masdeu L, Sudarsky L, and Wolfson L (1997) Gait Disorders of Aging: Falls and Therapeutic Strategies, Lippincott-Raven, with permission.

ularly common in older people, is characterized by reduced stride length, widened base, and a shortened swing phase, with preservation of rhythmic stepping. Posture is guarded and slightly stooped, "as if walking on a slippery surface" (Murray et aI., 1969; Nutt et aI., 1993). This gait pattern is, in essence, an adaptation to perceived imbalance or postural threat. The cautious gait was the most frequently observed higher level gait disorder in the series of Nutt et al. (1993). While this disturbance in locomotion is common, the presentation is entirely non-specific. In some cases, there may be balance problems which threaten stability. In patients with cautious gait, defensive adaptations are the salient feature, and these adaptations may mask an underlying disturbance in locomotor physiology. Features characteristic of the underlying disorder may emerge as the patient is followed over time. Despite the narrow intent, the designation "cautious gait" has sometimes been expanded to include a variety of psychogenic gait disorders dominated by anxiety and fear of falling (Sudarsky and Tideiksaar, 1997). The anxious, phobic gait is considered separately along with psychogenic gait disorders below; the use of the term cautious gait should be consistent with the original intent.

40.3.2. Stiff-legged gait

In patients with cerebrovascular disease, cerebral palsy, demyelinating disease, and various spinal disorders, a stiff-legged gait is the most common syndrome. Examples include hemiplegic gait, paraparesis of spinal origin, cerebral diplegia. The common denominator in these conditions is an element of spasticity in the lower limbs. There is usually a degree of weakness as well. A different kind of stiff legged gait is observed in patients with autoimmune stiff man syndrome. Physiologic mechanisms responsible for spastic, stiff-legged gait have been characterized (Dietz, 1997). Reduced knee flexion during the swing phase of gait is often an early sign (Kerrigan et aI., 2001). The spastic gait does not have a unitary mechanism, however, and the details are clinically relevant. Seven patterns of lower limb dysfunction have been described by the group at Moss Rehabilitation Hospital in Philadelphia, all of which impact on gait: (1) the flexed hip; (2) the flexed knee; (3) the stiff knee; (4) equinovarus; (5) the valgus foot; (6) hyperextension of the great toe (the striatal toe); and (7) hip adduction (scissoring). The distinction can be drawn with the aid of dynamic EMG or with

644

kinematic gait lab studies (Wissel et al., 1999; Mayer et al., 2001). Oral spasticity medications are often used in patients with stiff legged gait. While there are no randomized clinical trials in which measures of gait are the primary endpoint, patients report subjective benefits and other manifestations of spasticity are improved. Intrathecal baclofen, administered through an implanted pump, has been used for treatment failures (Ochs, 2001). Intervention with botulinum toxin, nerve block or surgery can be helpful, provided that the intervention is appropriately targeted. A different stiff-legged gait is observed in patients with auto-immune stiff man syndrome. The disorder is related to anti-GAD antibodies, and a deficiency of GABA mediated synaptic inhibition at a brain and spinal level (Levy et al., 1999). There is spasm with hyperlordosis of the lumbar spine, and the patients walk as if the limbs and trunk were fused in a solid block. This "Frankenstein gait" improves with treatments that release muscle overactivity, including high dose benzodiazepines, intrathecal baclofen, and immunotherapy. There have been several reports of success with plasmapharesis and with intravenous immunoglobulin (lVIg) in stiff man syndrome (Karlson et al., 1994; Souza-Lima et al., 2000). 40.3.3. Freezing gait

Gait freezing is a distinctive phenomenon, first observed in the 19th century in patients with Parkinson's disease. Motor blocks (arrests of movement) are usually brief, often occurring with gait initiation or when the patient turns to change direction. There may be a series of quick, ineffective stepping movements with side to side shifting of weight, but no forward engagement ("slipping clutch syndrome"). Locomotor movements are then normal or near normal once the patient is in motion, but freezing may return as the patient turns or attempts to navigate the door threshhold. There is variability, in that the same doorway may be a problem on one occasion, but not the next. Freezing is often overcome with the aid of sensory cues, particularly visual cues. Freezing of gait was reported in 26% of Parkinson's patients at the 14 month endpoint of the DATATOP study (Giladi et al., 2001), and in 25% of patients at 3--4 years in the ropinerole 056 study (Rascol et al., 2000). It is more common with

L. SUDARSKY

advancing disease, and is correlated with dysarthria, axial rigidity and postural instability. Pahapill and Lozano (2000) speculate about the role of cell loss in the pedunculopontine nucleus in the gait freezing of Parkinson's disease. Experience is variable with pallidotomy and DBS surgical procedures, which may improve gait, but do not consistently help gait freezing (Stolze et al., 2001a). Gait freezing is not limited to Parkinson's disease, and also occurs in patients with related neurodegenerative diseases such as progressive supranuclear palsy, multiple system atrophy, corticobasal degeneration, and diffuse Lewy body disease. A syndrome of pure akinesia has been described, with dysarthria, gait freezing, postural instability, and a poor response to levodopa. Some patients with pure akinesia exhibit the neuropathology of progressive supranuclear palsy at post mortem (Riley et al., 1994). Primary progressive freezing gait ("pure gait ignition failure") may be a similar disorder with restricted expression (Achiron et al., 1993; Atchison et al., 1993). Imaging studies in patients with primary progressive freezing gait do not show evidence of frontal hypometabolism (Fabre et al., 1998) or dopamine cell loss (Jennings et al., 2000). The other major cause of this phenomenon is cerebrovascular disease, particularly small vessel disease involving the basal ganglia and deep white matter. Freezing gait in this context is sometimes described as lower body Parkinsonism, though it is not a dopamine deficiency disorder, or as gait apraxia, though it is not a true apraxia (Fitzgerald and Jankovic, 1989). Thompson and Marsden (1987) characterized the gait disorder in twelve patients with subcortical arteriosclerotic encephalopathy (Binswanger's disease), based on observation of patients' gait and chair rise. Start and tum hesitation are salient features; festination was not observed. Yanagisawa et al. captured episodes of gait freezing in the laboratory with surface EMG and analysis of ground reaction forces (1991). Cocontraction was observed in antagonist muscles in the legs, which interferes with phasic activation of locomotor movements and restrains forward motion. Elble et al. studied the physiology of gait initiation in five patients with vascular Parkinsonism and freezing gait (1996). The basic architecture of the first step was preserved in these patients, though steps were irregular and sometimes aborted. Postural

645

OTHER GAIT DISORDERS

shifts necessary to initiate forward movement were not effectively generated. Gait laboratory studies of patients with subcortical arteriosclerotic encephalopathy reveal variability of timing and size of stepping (Ebersbach et al., 1999). In these patients, there are compensations for imbalance and some ataxic features, in addition to short steps and freezing gait. 40.3.4. Frontal gait disorder The term frontal gait describes the combination of locomotor disorder with a degree of disequilibrium, related to disease of the forebrain. The spectrum ranges from a predominantly hesitant, freezing gait to a primarily ataxic disturbance of gait and balance. The best studied example is the gait disorder of hydrocephalus. The classic description is of a patient "stuck to the floor", who just can't seem to get the feet moving forward. In hydrocephalus, there is a disturbance of frontal subcortical circuits, due to stretching and edema of periventricular white matter pathways. Kinematic studies reveal decreased stride length, increased sway, and reduced foot floor clearance (Knutsson and Lying-Tunell, 1985; Stolze et al., 2001b). Surface EMG demonstrates co-contraction of antagonist muscles throughout the gait cycle; a failure of phasic activation of the leg muscles during stepping is also observed in patients with severe Parkinson's disease and gait freezing (Sudarsky and Simon, 1987). Stride length generally improves after removal of 30 cc of CSF, which is the basis of a commonly-used predictive test for neurosurgical intervention (Stolze et al., 2000). The gait disorder of patients with cerebrovascular white matter disease, described above, may have a similar mechanism. The periventricular white matter is frequently impacted by ischemic change in these patients, and the locomotor disorder is similar. These patients exhibit a substantial degree of imbalance with disease progression (Thompson and Marsden, 1987). The distinction between a freezing gait and frontal gait in these patients comes down to how much balance impairment is evident. 40.3.5. Dystonic gait Dystonia produces unusual and sometimes bizarre disorders of gait. Dystonia can be focal, restricted to the lower limb, or generalized. The commonest focal disorder is dystonic inversion at the ankle, sometimes associated with extension of the great toe.

Dystonic toe flexion also occurs, as do more complex proximal lower limb dystonias. Focal dystonia is generally amenable to botulinum toxin injection, provided that the active muscles can be successfully identified and targeted. Dynamic EMG or kinematic studies can sometimes be of help. In patients with generalized dystonia, there may be torsion of the trunk, and the disturbance of lower limb function is usually bilateral. The two principal gaits (walking and running) may be differentially affected. For many patients, dystonic gait is not apparent when walking backwards. Dystonic flexion ofthe thoracolumbar spine (camptocormia) has been identified in patients with Parkinson's disease and related disorders, many of whom have kyphoscoliosis caused by tone abnormalities from their neurologic disease (Nieves et al., 2001). 40.3.6. Cerebellar gait Cerebellar ataxia produces imbalance and a distinct locomotor disorder. The gait is typically slow and halting, with a widened base of support. Stepping is irregular, which results in a lurching quality as the upper body segments struggle to maintain alignment. The patient may stumble or veer off to the side. Truncal instability is more pronounced when attempting to walk on a narrow base, or tandem, heel to toe. Patients also exhibit imbalance when they tum or too quickly change direction. Control of gait and balance is localized within the cerebellum to the midline structures, including the efferent pathways of the interposed and fastigial nucleus (Lechtenberg and Gilman, 1978). Patients with alcoholic cerebellar degeneration, which affects primarily the midline vermis, often have disproportionate gait ataxia. Laboratory gait analysis in cerebellar patients has confirmed the features described above. Palliyath et al. (1998) noted a reduced velocity and stride length in ten patients with cerebellar degeneration. All components of the gait cycle showed increased variability; stepping was irregular in timing, force, direction and amplitude. Irregularity of step generation can be considered one of the defining features of cerebellar gait. Interestingly, the gait was not widebased in this study, though other labs have confirmed a wide base of support in cerebellar patients (Cueman-Hudson and Krebs, 2000). Other characteristic features include reduced dorsiflexion of the

646

ankle at the onset of the swing phase, which might predispose to tripping. Analysis of moments about the knee and ankle joints reveals poor intralimb coordination. Kinetic studies of ground reaction forces demonstrate a reduced peak at weight acceptance, reflecting a difficulty planting the foot soundly. Roughly 60% of patients with cerebellar degeneration have a form of hereditary ataxia. There has been great progress over the last ten years in understanding the hereditary ataxias, many of which can be identified through DNA diagnostic testing. The cost of genetic testing in ataxia patients is roughly comparable to the cost of MRI, and the information is very helpful for prognosis and genetic counseling. The neurodegenerative ataxias differ in their effect on cerebellar circuit anatomy. Physiologic studies and therapeutic clinical trials are more informative when they are based on homogeneous, genetically-defined subgroups. 40.3.7. Sensory ataxia

Locomotor ataxia was described in the 19th century in patients with tabetic neurosyphilis. There is lateral path deviation, but a regular stride. Some patients are visually dependent in their walking, and do poorly in the dark. While there is a healthy redundancy of sensory information to provide feedback for postural control during locomotion, bipeds are particularly vulnerable to a loss of proprioception. Causes of a predominantly ataxic neuropathy in clinical practice include vincristine and cisplatin chemotherapy, cobalamin deficiency, paraproteinemia, and subacute sensory neuropathy, which may be paraneoplastic (Sabin, 1997). Patients with peripheral neuropathy affecting large fibers are substantially more likely to fall (Richardson et al., 1992). Several studies have examined the impact of proprioceptive deficits on postural control and gait. Horak (2001) has demonstrated that moderate neuropathy results in increased body sway during stance, and displacement of the center of mass while walking. In order to adjust to the excess motion of the center of mass, the patient alters lateral step placement. Somatosensory information derived from fingertip contact with a stable object (haptic information) can provide a degree of balance compensation. Lajoie et al. (1996) reported kinematic

L. SUDARSKY

studies on a patient with advanced sensory neuropathy, who had absent proprioception in the limbs by clinical measures. The focus of the discussion was on the compensatory strategies employed by such patients to facilitate locomotion in the absence of somatosensory feedback. This patient walked with a widened base of support and reduced stride, biomechanical adaptations to improve stability. A forward tilt was observed, with flexion of the neck such that the patient could monitor the performance visually. EMG demonstrated co-activation of the vastus lateralis and medial hamstring during weight acceptance, effectively bracing the leg during the stance phase. Although the gait appears more mechanical, this strategy effectively decreases the number of degrees of freedom and simplifies the task of postural control. 40.3.8. Myopathy and weakness

When the salient feature of the gait disorder is weakness, a significant degree of weakness should be apparent on muscle resistance testing. Distal weakness produces a foot drop: the foot slaps against the ground and there is difficulty with plantar flexion at the ankle during push off. There is usually a compensatory increase in step height to maintain foot floor clearance and avoid tripping. Some patients with proximal weakness walk with exaggerated pelvic rotation and a waddling gait. In essence, the pelvis cannot be held level during stance, but tilts towards the swing leg. The patient compensates by leaning toward the support leg. Weakness of the hip girdle muscles is apparent when the patient is asked to stand from a chair. EMG will provide confirmation of the underlying neuromuscular disease. Patients with degenerative lumbar spinal stenosis sometimes exhibit flexed posture and weakness of the legs with progression as they walk. These patients may have normal strength and reflexes when they are examined in a seated position. Spine imaging with CT or MRI demonstrate critical narrowing of the spinal canal. 40.3.9. Psychogenic gait disorder

Functional disorders of gait were described by Charcot and others in the medical literature from the 19th century. A variety of psychogenic gait disorders were observed among combatants in World War I,

647

OTHER GAIT DISORDERS

including astasia-abasia, camptocormia, and malingering. Psychogenic disorders of gait occur at all ages, though extra caution should be used in making this diagnosis in the elderly. Older people often exhibit a combination of physical disability and an exaggerated compensatory response, particularly if there is apprehension and fear of falling. Lempert et al. (1991) characterized the recognition features which help identify psychogenic disorders of gait from a review of video in thirty-seven cases. The abnormalities are often distinctive, and psychiatric history is often informative. Extreme slow motion may occur, as if the patient were walking through a viscous substance. Uneconomical postures with wastage of muscular effort are another classic example. Dramatic fluctuations may occur over minutes, which is unusual for patients with neurologic disease. Observation and distraction may activate and/or suppress the findings. Hayes et al. (1999) elaborated on these criteria in a video atlas, noting some diagnostic pitfalls. Prognosis is often good for those patients in whom the history is recent and the decompensation is acute. Dramatic cures are often possible (Keane, 1989). When psychogenic disorders of gait persist for six months or longer and become part of a established pattern of dependence and disability, it is often difficult to restore function.

40.4. Use of clinical neurophysiologic testing The failing gait is a product of an underlying physiologic disturbance and a compensatory res-

ponse. Consequently many failing gaits look fundamentally similar, and it is often difficult to make a diagnosis from observation. Neurophysiologic testing may be helpful in the deconstruction of gait disorders for purposes of diagnosis. In patients with stiff-legged gait, laboratory gait analysis may be used to help target an intervention. Gait laboratory studies can also be used as an endpoint or measure of efficacy in clinical trials. Table 2 reviews some of the testing used for diagnosis and for therapeutic targeting. Patients with sensory deficits presenting with cautious gait or sensory ataxia may require nerve conduction studies and/or laboratory tests of vestibular function. This testing is also appropriate for the patient presenting with recurrent falls. Patients with skeletal muscle weakness often require EMG and more sophisticated testing to diagnose myopathy and neuromuscular disease. Laboratory gait analysis has been used for decades to measure the success of orthopedic interventions and cerebral palsy surgery. As these techniques grow in their descriptive sophistication, and as the technology becomes more widely available, neurologists will also utilize the laboratory for clinical investigation. Recent publications demonstrate the usefullness of laboratory gait and balance testing to document the effects of deep brain stimulation for Parkinson's disease (Stolze et al., 2001a). Baezner et al. (2001) recently published a randomized prospective trial using gait analysis to measure improvements in a group of patients with

Table 2 Use of clinical neurophysiologic testing in gait disorders. Syndrome

Testing used

Cautious gait

EMG, nerve conduction tests posturography

Stiff legged gait

gait analysis to target surgery, nerve block, or botox

Freezing gait, frontal gait

(MRI and imaging tests), kinematic testing as outcome measures in clinical investigation

Dystonic gait

dynamic EMG or kinematic studies to target botox

Cerebellar ataxia

(MRI, imaging, genetic testing)

Sensory ataxia

EMG, nerve conduction tests ENG and oculographic recordings posturography

Myopathy and weakness

EMG, motor unit analysis, single fiber EMG

648

"lower body Parkinsonism" (frontal gait) treated with amantadine.

References Achiron, A, Ziv, I, Goren, M, Goldberg, H, Zoldan, Y, Sroka, Hand Melamud, E (1993) Primary progressive freezing gait. Mov. Disord., 8: 293-297. Armstrong, DM (1988) The supraspinal control of mammalian locomotion. J. Physiol., 405: 1-37. Atchison, PR, Thompson, PD, Frackowiak, RSJ and Marsden, CD (1993) The syndrome of gait ignition failure. Mov. Disord., 8: 285-292. Baezner, H, Oster, M, Henning, 0, Cohen, S and Hennerici, MG (2001) Amantadine increases gait steadiness in frontal gait disorder of subcortical vascular encephalopathy: a double-blind randomized placebo-controlled trial based on quantitative gait analysis. Cerebrovas. Dis., 11: 235-244. Cueman-Hudson, C and Krebs, DE (2000) Frontal plane dynamic stability and coordination in subjects with cerebellar degeneration. Exp. Brain Res., 132: 103113. Dietz, V (1997) Neurophysiology of gait disorders: present and future applications. EEG Clin. Neurophysiol., 103: 333-355. Dietz, V, Colombo, DM, Jensen, DM and Baumgartner, L (1995) Locomotor capacity of spinal cord in paraplegic patients. Ann. Neurol., 37: 574-582. Eidelberg, E, Walden, JG and Nguyen, LH (1981) Locomotor control in Macaque monkeys. Brain, 104: 647-663. Elble, RJ, Cousins, R, Leffler, K and Hughes, L (1996) Gait initiation by patients with lower-half parkinsonism. Brain, 119: 1705-1716. Fabre, N, Brefel, C, Sabatini, U, Celsis, P, Montastruc, JL, Chollet, F and Rascol, 0 (1998) Normal frontal perfusion in patients with frozen gait. Mov. Disord., 13: 677-683. Fitzgerald, PM and Jankovic, J (1989) Lower body parkinsonism: evidence for a vascular etiology. Mov. Disord., 4: 249-260. Forssberg, Hand Grillner, S (1973) The locomotion of the acute spinal cat injected with c1onidine. Brain Res., 50: 184-186. Fuh, JL, Lin, KN, Wang, SJ, Ju, TH, Chang, Rand Liu, HC (1992) Neurologic disease presenting with gait impairment in the elderly. In: M Woollacott and F Horak (Ed.), Posture and Gait: Control Mechanisms (Vol. II). University of Oregon Books, pp. 71-74. Garcia-Rill, E (1986) The basal ganglia and the locomotor regions. Brain Res. Rev., 11: 47-63.

L. SUDARSKY

Giladi, N, McDermott, MP, Fahn, S, Przedborski, S, Jankovic, J, Stem, M and Tanner, C (2001) Freezing of gait in PD. Neurology, 56: 1712-1721. Gill, TM, Desai, MM, Gahbauer, EA, Holford, TR and Williams, CS (2001) Restricted activity among community-living older persons: incidence, precipitants, and health care utilization. Ann. Int. Med., 135: 313321. Grillner, S (1981) Control of locomotion in bipeds, tetrapods and fish. In: VB Brooks (Ed.), Handbook of Physiology, the Nervous System, Motor Control (Section I, Vol. II, Part 2). American Physiologic Society, Bethesda, pp. 1179-1236. Hayes, MW, Graham, S, Heldorf, P, de Moore, G and Morris, JGL (1999) A video review of the diagnosis of psychogenic gait: appendix and commentary. Mov. Disord., 14: 914-921. Horak, F (200 I) Postural ataxia related to somatosensory loss. Adv. Neurol., 87: 111-117. Jennings, DL, Factor, SA, Molho, ES, Friedman, JH, Innis, RB, Seibyl, JP and Marek, KL (2000) Primary progressive freezing gait disorder: a neurimaging study to evaluate dopamine transporter density. Neurology, 54 (Suppl. 3): A191. Karlson, EW, Sudarsky, L, Ruderman, E, Pierson, S, Scott, M and Helfgot, SM (1994) Treatment of stiff man syndrome with intravenous immune globulin. Arth. Rheum., 37: 915-918. Keane, JR (1989) Hysterical gait disorders: 60 cases. Neurology, 39: 586-589. Kerber, KA, Enrietto, JA, Jacobson, BA and Baloh, RW (1998) Disequilibrium in older people. Neurology, 51: 574-580. Kerrigan, DC, Karvosky, ME, and Riley, PO (2001) Spastic paretic stiff-legged gait: joint kinetics. Am. J. Phys. Med. Rehabil., 80: 244-249. Knutsson, E and Lying-Tunnel, U (1985) Gait apraxia in normal-pressure hydrocephalus. Neurology, 35: 155160. Lajoie, Y, Teasdale, N, Cole, JD, Burnett, M, Bard, C, Fleury, M, Forget, R, Paillard, J and Lamarre, Y (1996) Gait of a deafferented subject without large myelinated sensory fibers below the neck. Neurology, 47: 109115. Lawrence, DG and Kuypers, HGJM (1968) The functional organization of the motor system in the monkey (Part 2). Brain, 91: 15-36. Lechtenberg, R and Gilman, S (1978) Localization of function in the cerebellum. Neurology, 28: 376. Lempert, T, Brandt, T, Dieterich, M and Huppert, D (1991) How to identify psychogenic disorders of stance and gait. J. Neurol., 238: 140-146. Levy, LM, Dalakas, MC and Floeter, MK (1999) The stiffperson syndrome: an autoimmune disorder affecting

OTHER GAIT DISORDERS

neurotransmission of GABA. Ann. Int. Med., 131: 522-530. Mayer, NH, Esquenazi, A and Keenan, MA (200 I) Patterns of upper motoneuron dysfunction in the lower limb. Adv. Neurol., 87: 311-320. Mori, S, Matsui, T, Mori, F, Nakajima, K and Matsuyama, K (2000) Instigation and control of treadmill locomotion in high decerebrate cats by stimulation of the hook bundle of Russell in the cerebellum. Can. J. Physiol. Pharmacol., 78: 945-957. Mori, S, Matsuyama, K, Mori, F and Nakajima, K (2001) Supraspinal sites that induce locomotion in the vertebrate central nervous system. Adv. Neurol., 87: 25--40. Nashner, LM (1980) Balance adjustments of humans perturbed while walking. Neurophysiol., 44: 650-664. Newman, G, Dovermuehle, RH and Busse, EW (1960) Alterations in neurologic status with age. J. Am. Geriatr. Soc., 8: 915-917. Nieves, AF, Miyasaki, 1M and Lang, AE (2001) Acute onset dystonic camptocormia caused by lenticular lesions. Mov. Disord., 16: 177-180. Nutt, JG (1998) Gait and balance disorders, a syndrome approach. In: J Jankovic and E Tolosa (Ed.), Parkinson's Disease and Movement Disorders (3rd ed.). Williams and Wilkins, Baltimore. Nutt, JG, Marsden, CD and Thompson, PD (1993) Human walking and higher-level gait disorders, particularly in the elderly. Neurology, 43: 268-279. Ochs, G (1999) Oral medications and intrathecal baclofen for spastic gait. International symposium on Gait Disorders, Prague. Ochs, G, Struppler, A, Meyerson, BA, Linderoth, B, Gybels, J, Gardner, BP, Teddy, P, Jamous, A and Weinmann, P (1989) Intrathecal baclofen for long-term treatment of spasticity: a multi-center study. J. Neurol. Neurosurg. Psychiatry, 52: 933-939. Odenheimer, G, Funkenstein, HH, Beckett, L, Chown, M, Pilgrim, D, Evans, D and Albert, M (1994) Comparison of neurologic changes in successfully aging persons vs. the total aging population. Arch. Neurol., 51: 573580. Orlovsky, GN (1972) The effect of different descending systems on flexor and extensor activity during locomotion. Brain Res., 40: 359-371. Pahapill, PA and Lozano, AM (2000) The pedunculopontine nucleus and Parkinson's disease. Brain, 123: 1767-1783. Palliyath, S, Hallett, M, Thomas, SL and Lebiedowska, MK (1998) Gait in patients with cerebellar ataxia. Mov. Disord., 13: 958-964. Rascol, 0, Brooks, DJ, Korczyn, AD, De Deyn, PP, Clarke, CE and Lang, AE (2000) Five year study of the incidence of dyskinesia in patients with early Parkin-

649 son's disease who were treated with ropinerole or levodopa. N. Eng. J. Med., 342: 1484-1491. Richardson, JK, Ching, C and Hurvitz, EA (1992) The relationship between electromyographically documented peripheral neuropathy and falls. J. Am. Geriatr. Soc.,40: 1008-1012. Rubenstein, L and Johnson, KR (1997) Interventions to reduce the multifactorial risks for falling. In: J Masdeu, L Sudarsky and L Wolfson (Eds.), Gait Disorders of Aging: Falls and Therapeutic Strategies. LipincottRaven, pp. 309-328. Riley, DE, Fogt, N and Leigh, RJ (1994) The syndrome of "pure akinesia" and its relationship to progressive supranuclear palsy. Neurology, 44: 1025-1029. Sabin, TD (1997) Peripheral neuropathy: disorders of proprioception. In: J Masdeu, L Sudarsky and L Wolfson (Ed.), Gait Disorders of Aging: Falls and Therapeutic Strategies. Lippincott-Raven, pp. 273282. Souza-Lima, CFL, Ferraz, HB, Braz, CA, Arraujo, AM and Manzano, GM (2000) Marked improvement in a stiff-limb patient treated with intravenous immunoglobulin, Mov. Disord., 15: 358-359. Stolze, H, Kuhtz-Buschbeck, JP, Drucke, H, Johnk, F, Diercks, C, Palmie, S, Medhorn, HM, Illert, M and Deuschl, G (2000) Gait analysis in idiopathic normal pressure hydrocephalus - which parameters respond to the CSF tap test. Clin. Neurophysiol., Ill: 16781686. Stolze, H, Klebe, S, Poepping, M, Lorenz, D, Herzog, J, Hamel, W, Schrader, B, Raethjen, J, Wenzelburger, R, Mehdorn, HM, Deuschl, G and Krack, P (2001a) Effects of bilateral subthalamic nucleus stimulation on parkinsonian gait. Neurology, 57: 144-146. Stolze, H, Kuhtz-Buschbeck, JP, Drucke, H, Johnk, K, Illert, M and Deuschl, G (2001b) Comparative analysis of the gait disorder of normal pressure hydrocephalus and Parkinson's disease. J. Neurol. Neurosurg. Psychiatry, 70: 289-297. Sudarsky, L (1997) Clinical approach to gait disorders of aging. In: J Masdeu, L Sudarsky and L Wolfson (Eds.), Gait Disorders of Aging: Falls and Therapeutic Strategies. Lippincott-Raven, pp. 147-158. Sudarsky, L (2001) Gait disorders: prevalence, morbidity, and etiology. Adv. Neurol., 87: 111-117. Sudarsky, Land Ronthal, M (1983) Gait disorders among elderly patients: a survey study of 50 patients. Arch. Neurol., 40: 740-743. Sudarsky, L and Simon, S (1987) Gait disorder in late-life hydrocephalus. Arch. Neurol., 44: 263-267. Sudarsky, L and Tideiksaar, R (1997) The cautious gait, fear of falling, and psychogenic gait disorders. In: J

650 Masdeu, L Sudarsky and L Wolfson (Eds.), Gait Disorders of Aging: Falls and Therapeutic Strategies. Lippincott-Raven, pp. 283-296. Thompson, PD and Marsden, CD (1987) Gait disorder of subcortical arteriosclerotic encephalopathy: Binswanger's disease. Mov. Disord., 2: 1-8. Tinetti, ME, Speechley, M and Ginter, SF (1988) Risk factors for falls among elderly persons living in the community. N. Eng. J. Med., 319: 1701-1707. Whipple, RH (1997) Improving balance in older adults: identifying the significant training stimuli. In: J Masdeu, L Sudarsky and L Wolfson (Eds.), Gait

L. SUDARSKY

Disorders of Aging: Falls and Therapeutic Strategies. Lippincott-Raven, pp. 355-380. Wissel, J, Muller, J, Baldauf, A, Ung, SC, Ndayisaba, J, Stockl, B, Frischhut, B, Haberfellner, H and Poewe, W (1999) Gait analysis to assess the effects of botulinum toxin A treatment in cerebral palsy. Eur. J. Neurol., 6(4): 63-68. Yanagisawa, N, Ueno, E and Takarni, M (1991) Frozen gait of Parkinson's disease and vascualar Parkinsonism. In: M Shimura, S Brillner and VR Edgerton (Eds.), Neurobiological Basis of Human Locomotion. Japan Scientific Societies Press, Tokyo, pp. 291-304.

Movement Disorders Handbook of Clinical Neurophysiology. Vol. I M. Hallett (Ed.) © 2003 Elsevier B. V. All rights reserved

651 CHAPTER 41

Focal injection therapy Jean-Michel Gracies":" and David M. Simpson" b

u Department of Neurology, Mount Sinai Medical Center, New York. NY 10029, USA Clinical Neurophysiology Laboratories, Mount Sinai Medical Center, New York, NY 10029, USA

41.1. Introduction - rationale for the use of local treatment We address the treatment of all types of disabling muscle overactivity, i.e. the inability to relax a muscle in situations when it should be at rest. In most central nervous system (CNS) pathology associated with muscle overactivity (spasticity, rigidity, tremor, symptomatic dystonia), muscle overactivity is not equally distributed throughout all muscles in the body, but is particularly severe in some muscles. There is often imbalance between mildly hyperactive agonists and severely hyperactive antagonists (Denny-Brown, 1966; Tardieu et al., 1979). The paired agonist-antagonist tends to produce torque oriented primarily in the direction of the more overactive muscle. This imbalance between agonist and antagonist is often a key factor in the impairment of purposeful functional movements. In such cases, the use of general (oral) or regional (intrathecal) therapies is not likely to improve active function, since these administration routes lead to indiscriminate reduction of motor neuron excitability and recruitment in both agonists and antagonists. However, local treatments allow selective weakening of targeted hyperactive muscles. Before the emergence of botulinum toxin (BTX), the types of compounds used to provide local muscle relaxation included local anesthetics (Le. lidocaine and congeners) with a fully reversible action of short duration, and alcohols, chiefly ethyl alcohol (ethanol) and phenyl alcohol (phenol), with a longer duration of action. This chapter reviews these

* Correspondence to: Dr. Jean-Michel Gracies, M.D., Ph.D., Department of Neurology, The Mount Sinai Medical Center, One Gustave L. Levy Place. Annenberg 2/Box 1052, New York, NY 10029-6574, USA. E-mail address: [email protected] Tel.: (212) 241-5607; fax: (212) 987-7363.

classical chemical local treatments and BTX from several perspectives: history of the introduction of each compound, pharmacological properties, histological effects, physiological action, risks and potential adverse effects, and clinical effects and indications. Controlled studies evaluating traditional neurolytic therapy vs botulinum toxin are scarce. We present a theoretical comparison of these treatments in the final section. Finally we describe our recommended injection technique when using any of these agents, which is the exploratory stimulation technique. This is the proper technique to distinguish a muscle among its synergistic neighbors, and is also the only technique that permits intramuscular localization, for example when targeting motor points or areas dense in endplates. The various agents available for focal injection therapy and their respective properties have also been reviewed in two previous publications (Grades and Simpson, 2000; Grades et aI.,2002a). 41.2. Historical aspects of focal injection therapy 41.2.1. Local anesthetic agents

In 1919, Liljestrand and Magnus reported that intramuscular injection of procaine reduced triceps rigidity in a decerebrate cat. This observation went relatively unnoticed at the time, when most of the scientific community viewed the response of a muscle to its own stretch as an intrinsic muscle property and not as a reflex mediated through neurons. This concept was soon to be corrected, following Liddell and Sherrington's studies (1924). Interest for Liljestrand and Magnus observations and the prospect of therapeutical applications emerged 40 years later, after a second major step in the understanding of stretch reflexes: the elucidation of the role of the recently discovered gamma motoneurons (Kuffler et al., 1951).

652

Procaine, the first local anesthetic agent to be synthetized in 1905, was commonly used up to the 1970s (Matthews and Rushworth, 1957; Khalili and Benton, 1966; Burke and Lance, 1973; Franz and Perry, 1974) but was then replaced by lidocaine, etidocaine and bupivacaine as preferred local anesthetics for relaxation of overactive muscles or in physiological research (Murdoch-Ritchie and Greene, 1980; Hogan et aI., 1993; Kaji et al., 1995; Kiernan et aI., 1997). 41.2.2. Chemical neurolysis by alcohol and phenol 41.2.2.1. Alcohol Ethyl alcohol ("alcohol") was the first alcohol compound to be studied experimentally on nerve cells (May, 1912; Gordon, 1914) and used for neuromuscular block (Tardieu G et aI., 1962; Tardieu C et aI., 1964), and the only one assessed in controlled protocols for this indication (Tardieu C et aI., 1964; Tardieu G et al., 1968). Despite a better safety record than phenol (see below), alcohol has not been used as extensively for the treatment of muscle overactivity (Glenn, 1990). Neurologists used local injections of alcohol for sympathectomy (lumbar paravertebral injections) and for the treatment of pain (trigeminal neuralgia, intractable carcinoma with paraganglionic and plexic injections) prior to its use in spasticity (Murdoch-Ritchie, 1985). This new indication began with injections of 35-45% ethyl alcohol (alcohol), epidurally (Tardieu G et aI., 1962), intramuscularly (Tardieu G et aI., 1964), perineurally (Hariga et al., 1964; Hariga, 1966), and then intrathecally (Bruno, 1975). 41.2.2.2. Phenol Phenol (benzyl alcohol, or carbolic acid) is the major oxidized metabolite of benzene, human leukemogen and ubiquitous environmental pollutant, and widely used as a disinfectant and antiseptic. Cell damaging properties of phenol were first exploited in antispasticity treatment with intrathecal administration of 2-5% phenol (Kelly and Gauthier-Smith, 1959; Nathan, 1959; Kjellberg et aI., 1961), then followed use in perineural (Khalili et al., 1964) and intramuscular injections (Halpern and Meelhuysen, 1966; Awad, 1972a, b; Delateur, 1972). Since then, phenol use has been more often reported in adult brain injury patients (Garland et aI., 1982; Moore et aI., 1991; Koman et aI., 1996; Kirazli et al., 1998; On

J-M. ORACIES AND D.M. SIMPSON

et aI., 1999) than in children with cerebral palsy (Spira, 1971; Yadav et al., 1994). Glenn has defined chemical neurolysis as "a nerve block that impairs nerve conduction by means of destruction of a portion of the nerve" (Glenn, 1990). As with local anesthetic blocks, early experience with chemical neurolysis was encouraged by the speculation that alcohol and phenol would selectively affect small diameter fibers (Kjellberg et al., 1961; Hariga, 1966; Khalili and Benton, 1966). However, this was later disproven (Nathan et al., 1965). While Tardieu performed controlled studies of intramuscular chemical neurolysis with alcohol (Tardieu C et aI., 1964; Tardieu G et al., 1968), the literature on chemical neurolysis with phenol contains mostly anecdotal reports. These have indicated relief of muscle overactivity following these treatments, with effects lasting from months to years (Glenn, 1990; Koman et aI., 1996). Intrathecal neurolysis is now rarely used, and the main types of injections currently used are perineural (Yadav et aI., 1994), i.e. close to a nerve trunk, also called "mixed sensorimotor nerve blocks", and intramuscular at the motor point, called "motor nerve blocks" (Glenn, 1990) or "neuromuscular blocks" (Koman et al., 1996). Perineural blocks are easier and more effective than intramuscular injections, and may be considered for proximal muscles or when several muscles are to be injected in the same nerve territory. For example, perineural injections can be used for treating overactivity in muscle groups that are less accessible to direct injection, such as iliopsoas, quadratus lumborum, or paraspinals (Hariga et aI., 1964). On the other hand, perineural injections carry a higher risk of adverse effects, such as painful dysesthesia post injection (see below). 41.2.3. Botulinum toxin

In the late nineteenth century, Van Ermengem discoverered the Clostridium botulinum bacteria, and first postulated its action in the central nervous system (Van Ermengem, 1897). Peripheral action of a neurotropic protein called botulinum toxin (BTX) was then described (Dickson et aI., 1923; Edmunds and Long, 1923; Guyton and McDonald, 1947; Burgen et aI., 1949; Brooks, 1954). It is now known that the anerobic gram positive bacterium Clostridium botulinum, and certain strains of the bacterial species Clostridium butyricum, Clostridium baratii

653

FOCAL INJECTION THERAPY

and Clostridium argentinense produce seven serotypes (A to G) of BTX (Hatheway, 1989). Almost all the serotypes block acetylcholine (Ach) release from peripheral motor nerves resulting in flaccid paralysis, but only types A, B, E and F have been linked to cases of food-borne botulism in humans (Simpson LL, 1981). Type A (BTX-A) causes the most severe illness and is the best characterized BTX. BTX-A was the first serotype to be used via injection for the treatment of human disease in the 1970s, for strabismus following several years of monkey experiments (Scott et aI., 1973; Scott, 1980). In 1989, the U.S. Food and Drug Administration (FDA) approved BTX-A for three human disorders, strabismus, blepharospasm, and postparetic hemifacial spasm in patients 12 years of age and older. BTX-A and botulinum toxin type B (BTX-B) were later approved for spasmodic torticollis in 2000. Approval of BTX for cerebral palsy and adult spasticity has been granted in several countries, although not yet in the United States. Despite these restricted authorizations, BTX is used off-label in many other conditions involving overactivity of muscle or of parasympathetically innervated glands. 41.3. Pharmacology of the agents available for focal injection therapy

41.3.1. Local anesthetics

Local anesthetics are drugs that block nerve conduction when applied locally to nerve tissue, and whose action is reversible, without causing structural damage to nerve fibers or cells when used in appropriate concentrations (Murdoch-Ritchie and Greene, 1980). Local anesthetics act on peripheral and central neurons (Bohbot et aI., 1996; Boehnke and Rasmusson, 2001). The mechanism of the conduction block involves a partial prevention of the large transient increase in sodium permeability that is produced by membrane depolarization, especially at the nodal regions (Baker, 2000). Increases in potassium conductance, and blocks of abnormal impulses arising from non-inactivating sodium channels, have also been reported (Lin and Rydqvist, 1999; Khodorova et aI., 2001). These channel effects seem to be mediated by changes in the lipid phase of the membrane (Lin and Rydqvist, 1999). Therefore, a local anesthetic in contact with a nerve trunk

causes both sensory loss and motor paralysis in the innervated area (Flanagan et aI., 1997). Since ionic mechanisms of excitability are similar in nerve and muscle, these agents also act on all types of muscular tissue. When a local anesthetic is injected close to a peripheral nerve, its effect begins within minutes (3 minutes for lidocaine, 15 minutes for bupivacaine). The delay of onset is greater when there is a need for diffusion of the agent from its site of injection to its site of action, for example, when a nerve plexus is blocked (the delay of onset of the effect of lidocaine injected about the brachial plexus reaches 10 min; Simon et aI., 1999). The duration of action of a local anesthetic depends primarily on the lipid solubility and the protein binding affinity of the compound. It is also proportional to the time during which the anesthetic is in actual contact with nervous tissues, and inversely related to the regional blood flow at the injection site (Karatassas et aI., 1993). Thus, procedures that keep the drug at the nerve by reducing local blood flow prolong the period of anesthesia. They also help reduce its systemic toxicity by slowing its absorption into the circulation. Therefore in clinical practice, solutions of local anesthetics often contain vasoconstrictors: epinephrine or a suitable synthetic congener, norepinephrine or phenylephrine (Murdoch-Ritchie and Greene, 1980). However, because of the possibly intense vasoconstriction, epinepluine-containing solutions should not be injected into tissues supplied by end-arteries, such as fingers and toes. Clonidine also prolongs the duration of block, whether it is administered locally with the lidocaine, or orally before the lidocaine block administration (Dobrydnjov and Samarutel, 1999). 41.3.2. Alcohol and phenol

At concentrations higher than 10%, alcohol is a hypobaric compound that immediately and nonselectively denatures proteins and injures cells by precipitating and dehydrating protoplasm (MurdochRitchie, 1985). Like alcohol, phenol denatures protein, causing tissue necrosis. As with local anesthetics and alcohol, the destructive effects of phenol are non-selective across fiber types, and correlate with the concentration of phenol applied (Nathan et aI., 1965).

654

Metabolism and toxicity distinguish phenol from alcohol. Ninety to 98% of ethyl alcohol that enters the body is normally oxidized. Following oral or intravenous administration in mice, phenol is metabolized into phenol sulfate, phenol glucuronide, and hydroquinone glucuronide (Kenyon et al., 1995). Benzene causes leukemia and aplastic anemia in humans. Its oxidative metabolites, phenol and hydroquinone, reproduce its hematotoxicity and genotoxicity, cause DNA and chromosomal damage present in leukemia, and alter hematopoiesis and clonal selection (Chen and Eastmond, 1995; McDonald et al., 2001; see section side effects and precautions). 41.3.3. Botulinum toxin

The various serotypes of BTX are structurally and functionally similar (Mochida et al., 1990; Tonello et al., 1996). They are high molecular weight protein complexes consisting of the neurotoxin and additional nontoxic proteins that function to protect the toxin molecule (Callaway et al., 2001). BTX-A is produced in crystalline form from an anerobic culture. It is a 900 kDa protein composed of two molecules of neurotoxin (ca. 150 kDa), noncovalently bound to nontoxic proteins (gelatin or albumin), that play an important role in the stability and effective toxicity of the toxic unit (Schantz and Johnson, 1992). The ultrastructural mechanism of BTX action in muscle has been defined mainly from studies with BTX-A, although serotypes B, Cl , D, F, G have also been studied (Tonge, 1974; Kauffman et al., 1985; Sanchez-Prieto et al., 1987; Hallett et al., 1994; Coffield et al., 1997; Foran et al., 2003). All serotypes inhibit the release of acetylcholine and other neurotransmitters, but do not alter either neurotransmitter synthesis or storage (Simpson LL, 1981). The neurotoxin molecules are di-chain proteins, in which the heavy chain binds to two distinct acceptors on nerve terminal membranes (Simpson LL, 1981; Black and Dolly, 1986a; Montecucco and Schiavo, 1995; Daniels-Holgate and Dolly, 1996; Foran et al., 2003). This ectoacceptor-toxin interaction is mediated by a conformation of the heavy chain that can be maintained only through its association with the light chain (Daniels-Holgate and Dolly, 1996). The ensuing uptake of BTX is

J-M. ORAClES AND D.M. SIMPSON

energy and temperature-dependent and is accelerated by nerve stimulation, a treatment that also shortens the time course of the toxin-induced neuroparalysis (Dolly et al., 1984; Black and Dolly, 1986b; DasGupta, 1994). The light chain is a zinc-dependent endoprotease that cleaves synaptic proteins implicated in docking and fusion of vesicles (SNARE proteins). Specifically, BTX-A, Cl, and E selectively proteolyze SNAP-25 (BTX-Cl also cleaves syntaxin1), while BTX-B, D, F, and G cleave synaptobrevin (Foran et al., 2003). These enzymatic actions lead to blocking of transmission at all cholinergic synapses in the nervous system, and other types of synapses such as norepinephrinergic synapses in the peripheral nervous system (SanchezPrieto et al., 1987; Coffield et al., 1997; Morris et al., 2002). The main physiological consequence in muscle is inhibition of spontaneous and evoked quantal neurotransmitter acetylcholine release at the ex (alpha) motoneuron ending (Black and Dolly, 1986a, b; Molgo et al., 1989; Maselli et al., 1992; Callaway et al., 2001). This results in reduced muscle activity, whatever the stimulus, including voluntary activation or electrical nerve stimulation. Research on laboratory animal preparations indicates that there is absence of blockade when the tissue is incubated with the toxin at low temperature (< lO°C). The toxin binds to the tissue but does not cause paralysis because receptor-mediated endocytosis is arrested at low temperature, and thus toxin is not internalized (Coffield et al., 1997). At 33°C, nearly complete decline of miniature endplate potentials occurs within 40 minutes to a few hours after BTX injection (Kauffman et al., 1985), the process being slower at lower temperatures (Kao et al., 1976; Humeau et al., 2000). The lag time in these experimental tissue preparations could account for internalization and then onset of blockade (Coffield et al., 1997). However, it is still unclear why the onset of clinical effect following BTX injection in patients is typically delayed to 24 to 72 hours. This may be related in part to continuing spontaneous non-quantal release from the presynaptic supply of acetylcholine, until its exhaustion (Stanley and Drachman, 1983; Dolly et al., 1987), although some studies have shown that the non-quantal release is also decreased after BTX exposure (Dolezal et al., 1983). Motor recovery occurs following motor nerve terminal outgrowth (sprouting), formation of new endplates (Yee and Pestronk, 1987; Sellin et al.,

FOCAL INJECTION THERAPY

1996; Lee et a1., 1999), and the ultimate regeneration of the originally blocked endplate (de Paiva et a1., 1999). Both the intracellular stability of the catalytically active light chain of each serotype and the speed of replenishment of each of the toxin substrates contribute to the variable durations of effect among serotypes (Adler et a1., 2001; Foran et a1., 2003). The reference method of dose standardization for BTX-A has been the mouse assay, as established by Schantz and colleagues (Schantz and Johnson, 1990). In this assay, one unit (U) is defined as the amount of BTX-A that kills 50% of a group of 18 to 20 gram female Swiss-Webster mice after intraperitoneal injection (the LDso) . BTX-A is produced in North America, distributed under the name Botox" (Allergan Inc, Irvine, CA), and also in Europe and Japan. The formulation available in Europe, Dysport" (Ipsen, France), is prepared using a different method of purification involving enzymes and various exchangers (Schantz and Johnson, 1992). BTX-B is also manufactured in the USA (Myobloc", Elan, USA). Its potency is lower than BTX-A. In spasmodic torticollis, the number of mouse units of BTX-B required to achieve clinical response is over ten times higher than the corresponding doses of BTX-A (Tsui et a1., 1995; Lewet a1., 1997). BTX-F is manufactured in Japan (Mezaki et a1., 1995, 1999). 41.4. Histological effects 41.4.1. Local anesthetics

There are typically no significant histological changes after intramuscular injection of local anesthetics. However, animal experiments suggest that local anesthetics carry a low risk of neural toxicity (myelin and axon destruction) after perineural injection, which is proportional to the concentration used and to the conduction blocking potency of the compound (Kalichmann et a1., 1993; Kanai et al., 2000). 41.4.2. Alcohol and phenol 41.4.2.1. Perineural injections Alcohol: Absolute alcohol causes neuronal degeneration in animals, with extensive fibrosis and partial regeneration (May, 1912). At lower concentrations, axonal destruction was variable. Using 80% alcohol, Gordon reported various degrees of neuronal degen-

655

eration and surrounding fibrosis (Gordon, 1914). With 35% alcohol, Tardieu and colleagues observed mostly myelin lesions in small fibers three weeks post injection with no axonal damage, and endplate cholinesterase activity was reduced in the endplates of the muscle spindles only (not in the extrafusal fibers). These data initially suggested a selective effect on small diameter gamma motor neurons of 35% alcohol (Tardieu C et a1., 1964). However, these conclusions were later disputed, as non-selective effects on the responses of fibers of various diameters were reported after injection of 10 to 50% alcohol (Fisher et al., 1970; Taylor and Woolsey, 1976). Phenol: After administration of 2% aqueous phenol, the main effect around the nerve is damage to the microcirculation, including sludging, oscillation, plasma skimming and inverted flow (Okazaki, 1993b). This may lead to occlusion of small blood vessels and fibrosis in the injected area, and might account for long-term effects (Mooney et a1., 1969; Burkel and McPhee, 1970; Okazaki, 1993b). At 5% in saline, coagulation of peripheral nerves at the site of injection occurs one hour following injection, with the axons in the center of the nerve less affected when phenol is dripped onto the nerve. Axonal degeneration is evident within the injected nerve two days later. By two weeks, the innervated muscles atrophy to about 50% of control (Bodine-Fowler et a1., 1996). Wallerian degeneration occurs in the weeks following injection. Reinnervation with the first axonal sprouts appears between two and four weeks following the nerve block, and the sprouts become myelinated by four weeks, even after injection of 7% phenol (Westerlund et al., 2001). However, at five months, the muscle is still partially atrophied and consists of more fast fibers than control (Bodine-Fowler et a1., 1996). Eventually, there is regrowth of most axons, including gamma efferent axons (Burkel and McPhee, 1970; Wolf and English, 2000). When phenol is applied directly to an exposed peripheral nerve, damage to surrounding soft tissue may be limited by neutralization with alcohol (Koman et al., 1996). 41.4.2.2. Intramuscular injections Alcohol at concentrations above 20% produces local dose-dependent coagulation necrosis, followed by granulation tissue formation and subsequent fibrosis (Carpenter and Seitz, 1980; Brazeau and

656

Fung, 1989, 1990). From 20 to 40%, alcohol is still myotoxic in animals with creatine kinase release inhibited by adjunction of dibucaine, a potent local anesthetic (Brazeau and Fung, 1989, 1990). The destructive effect of alcohol on tissue has led to the use of absolute alcohol as treatment for benign and malignant tumors by topical injection (Burgener and Steinmetz, 1987; Papini et al., 1995), or for sclerotherapy of venous malformations (de Lorimier, 1995). Phenol has similar effects. Intramuscular injection causes local muscle necrosis and an associated inflammatory reaction of the fascia and subfascial tissues, within days of the procedure (Halpern, 1977). This reaction is intensified by two weeks, and gradually resolves. There is neurogenic atrophy of the muscle by two months, and collateral reinnervation and regeneration of muscle fibers. Return to normal of the muscle at three months only occurs with concentrations of aqueous phenol lower than 3% (Halpern, 1977). 41.4.3. Botulinum toxin

Intramuscular BTX-A injections do not cause any irreversible destruction to muscle or motor nerve terminals (Duchen, 1970, 1972; Pamphlett, 1989; Angaut-Petit et al., 1990; Harris et aI., 1991; Borodic et al., 1993; Hassan et al., 1994). The histochemical changes in muscle are transient. Plastic changes may occur in nerve and at neuromuscular junctions, and include the formation of new preterminal axon sprouts with multiple projections into myoneural junctions (Duchen, 1970; Angaut-Petit et al., 1990). 41.4.3.1. Changes in the peripheral nerve No motoneuronal toxicity has been shown with any BTX serotype, even after peri- or intraneural injections (Lu et al., 1998; Gomez-Ramirez et al., 1999; Eleopra et al., 2002). The changes in nerve following intramuscular BTX injection are reversible axotomy-like changes. Terminal and nodal sprouting begins within 48 hours of injection, and correlates with the formation of new neuromuscular junctions on the adjacent muscle surface after several weeks (Duchen, 1972; Pamphlett, 1989; Harris et al., 1991; Hassan et al., 1994). The new junctions are partly responsible for the reversibility of BTX clinical effect (de Paiva et al., 1999).

J-M. ORACIES AND D.M. SIMPSON

41.4.3.2. Changes in the muscle There is no fibrosis, necrosis, inflammation or persistent muscle atrophy following BTX injection (Harris et aI., 1991; Hassan et aI., 1995). Soon after BTX-A injection, a reversible denervation-like hypertrophy occurs (Spencer and McNeer, 1987). Later, there is transient muscle atrophy that progresses over several weeks and then resolves over several months (Duchen, 1970; Hassan et al., 1995). Fast-twitch muscle fibers (e.g. gastrocnemius) are functionally denervated longer than slow-twitch muscle (Duchen, 1970). 41.4.3.3. Changes at the neuromuscular junction Acetylcholine receptors may have an inducing role on sprouting, and their proliferation on the postsynaptic membrane after injection correlates with the amount of nerve sprouting (pestronk and Drachman, 1978; Schantz and Johnson, 1992). The blocked junction is also ultimately repaired (Bambrick and Gordon, 1987; de Paiva et aI., 1999). At least some of the new neuromuscular junctions are eliminated following muscle recovery (Hassan et al., 1994).

41.5. Physiological action 41.5.1. Local anesthetics

The physiological effects of local anesthetics vary according to the type and activity of the nerve fibers blocked. 41.5.1.1. Differential effects according to fiber type It was thought since Gasser and Erlanger that local anesthetics block impulses more readily in small than in large nerve fibers (Gasser and Erlanger, 1929). Muscle proprioceptive afferent and muscle efferent fibers, which are of the same diameter, were equally sensitive to procaine but the smaller gamma fibers supplying the muscle spindles were more rapidly blocked by the local anesthetic (Matthews and Rushworth, 1957). Franz and Perry suggested that this could be due to the shorter internodal distance in smaller axons, since more nodes were immediately accessible to the local anesthetic, whereas larger axons require more time for the anesthetic to diffuse to an equivalent number 'of

657

FOCAL INJECTION THERAPY

nodes (Franz and Perry, 1974). The same might account for the slower recovery of small axons at reversal of the process (Murdoch-Ritchie and Greene, 1980; Bouaziz et a1., 1998). This faster block of the gamma fibers caused confusion in the literature of the sixties with several studies relying on the assumption of exclusive gamma block with local anesthetics (Kjellberg et a1., 1961; Hariga, 1966; Khalili and Benton, 1966; Burke and Lance, 1973). Similar confusion appeared regarding the properties of alcohol and phenol (Kjellberg et a1., 1961; Hariga, 1966; Khalili and Benton, 1966; see below). More recent clinical studies confirm that all fiber types are sensitive to local anesthetics (Tay et a1., 1997), and some show an even higher sensitivity of large fibers with lidocaine (Huang et al., 1997; Itoh and Noda, 1998). 41.5.1.2. Differential effects according to pattern of impulse transmission The frequency and pattern of nerve impulse transmission also determine the types of nerve that are primarily affected, probably by varying the duration of the open states of sodium channels. For instance, etidocaine appears to block somatic motor nerves more than somatic sensory fibers (Bromage et al., 1974), while the opposite is true for tonicaine, a derivate of lidocaine (Gerner et a1., 2000). Indeed, the firing frequency of motor neurons rarely surpasses 10Hz. In contrast, the frequency of muscle spindle afferents is about 15 to 20 Hz within intermediate ranges of static muscle stretch, and can increase to more than 40 Hz during dynamic stretch or contraction, as measured in microneurographic experiments in normals (Vallbo, 1974), and in hemiplegic patients (Wilson et a1., 1999). In the central nervous system (CNS), lidocaine has been reported to preferentially block neurons that discharge at high frequencies (Manabe et a1., 1998). 41.5.1.3. Differential effects according to recent firing history The degree of block produced by a given concentration of local anesthetic depends on how much and how recently the nerve has been stimulated (Murdoch-Ritchie and Greene, 1980). Thus, a resting or normoactive nerve is less sensitive to a local anesthetic than one that has been recently and repetitively stimulated (Sotgiu et a1., 1992). There-

fore, a local anesthetic may have a greater efficacy on the overactive muscles that are target of therapy than on normally active muscles that it may have reached by diffusion. A similar phenomenon has been suggested for BTX (Hesse et a1., 1995). 41.5.1.4. Effect at the neuromuscular junction Local anesthetics also affect transmission at the neuromuscular junction (Harvey, 1939; de long, 1977) by unclear mechanisms (Ruff, 1977). To our knowledge, the question of whether this effect is greater or weaker than the conduction block on muscle afferent and efferents has not been examined. If greater, one may speculate that the intramuscular injection of a threshold dose of local anesthetic into endplate areas (see below) might be used as a shortterm mimic of BTX, and might allow the clinician to anticipate the effects of BTX injections. 41.5.2. Alcohol and phenol 41.5.2.1. Alcohol At low concentrations (5 to 10%), alcohol acts as a local anesthetic by decreasing sodium and potassium conductance. Tardieu and colleagues reported reduction of stretch reflexes but no weakness after injection of 35% alcohol into the posterior tibial nerve in cats (Tardieu C et a1., 1964, Tardieu G et a1., 1968). While stretch reflexes are consistently reduced regardless of the concentration, weakness has been reported with concentrations above 45-50% only (May, 1912; Gordon, 1914). Thus Labat, using 48% and 95% alcohol in dogs, provoked temporary paralysis with both concentrations, lasting usually less than two months (Labat, 1933). 41.5.2.2. Phenol Phenol has only local anesthetic properties with perineural injections of concentrations up to 2% in water, potentially faster acting than 2% lidocaine (Wood, 1978; Okazaki, 1993a). This rapid effect contributes to the transient anesthesia and weakness commonly seen after nerve blocks, and can be confounding if the assessment is made too early after the injection. Studies of the action potentials obtained by electrical stimulation after injection of 2% phenol around a nerve show a depression with a

658 biphasic time course (Okazaki, 1993b). Clinically, phenol concentrations lower than 3% generally give poor results (Glass et al., 1968). Hence, the dilution used most usually in clinical indications is 3 to 6%. Above 3%, there is almost immediate and then monophasic, constant denervation in EMG studies in humans, the conduction block being concentration and volume-dependent between 3% to 5% concentration (Fusfeld, 1968; Wood, 1978; Okazaki, 1993a). Maximal tension of the innervated muscle is still only 74% of control five months after injection of 5% phenol around a nerve (Bodine-Fowler et al., 1996). A similar discrepancy between large effects on reflexes and relatively small effects on muscle contractions of descending origin has been observed with phenol (Fisher et aI., 1970, 1971; Khali1i, 1984). Animal studies show that 7% phenol block results in complete denervation of muscle spindles, followed by rapid sensory reinnervation, and that reinnervation by gamma motor neurons is either incomplete or significantly delayed (Wolf and English, 2000). Thus, when correlating histological and physiological effects of phenol injections, it appears that significant effect depends on the use of a concentration higher than 3%, with histological destruction of nerve. 41.5.3. Botulinum toxin

In addition to the block at the cholinergic neuromuscular junction, the physiological mechanisms of BTX action after intramuscular injection may involve the following factors: 41.5.3.1. Central nervous system (CNS) effects There is evidence for central effects after intramuscular injection of BTX. 41.5.3.1.1. Retrograde axonal transport. In vivo animal studies have shown that BTX can enter the CNS following peripheral administration (Boroff and Chen, 1975). Histoautoradiography data have shown that, after intramuscular injection in animals, BTX or its fragments are transferred to the ventral roots and to the adjacent spinal cord by retrograde axonal transport, and spread partially to the contralateral cord and ventral roots (Wiegand et al., 1976). Direct stimulation of the injected muscle

J-M. ORAClES AND D.M. SIMPSON

increases the retrograde transport to the ipsilateral spinal cord half segments (Wiegand et al., 1976). 41.5.3.1.2. Action on central synapses. Once in the central nervous system, BTX or its fragments may affect a number of synaptic types. It exerts an inhibitory effect on cholinergic release in rat cerebrocortical synaptosomes (Dolly et al., 1982), as well as in non-cholinergic cerebellar neurons (Foran et al., 2003). BTX applied directly to the spinal cord blocks recurrent Renshaw inhibition, which is known to be mediated by the cholinergic synapses between recurrent axon collaterals of alpha-motoneurons and Renshaw cells (Hagenah et al., 1977; Hagenah, 1979). However, it has not been demonstrated that the mechanism of this block is indeed an inhibition of cholinergic release from the collateral (Renshaw cells are themselves glycinergic and GABAergic; Cullheim and Kellerth, 1981). 41.5.3.1.3. Central effects in animals. Significant alterations in central synaptic transmission, reflexes and motoneuronal firing pattern have been observed after high-dose intramuscular injections or in experimental botulism; they probably are related to retrograde axonal transport and central spread (Mikhailov, 1958; Wiegand et aI., 1974; Hagenah et al., 1977; Mikhailov and Barashkov, 1977; Wiegand and Wellhoner, 1977; Pastor et aI., 1997). 41.5.3.1.4. Central effects in humans. In humans, evidence also suggests that intramuscular BTX injection may have central effects. These effects may include decreased Renshaw inhibition and increased presynaptic inhibition exerted on the motoneurons supplying the injected muscle, and reduction of their overall excitability. - Decreased Renshaw inhibition has been supported by at least three observations in humans: (i) the persistence of H-reflexes in weakened muscles in patients with botulism, even when stimulus amplitude was increased to the point where the reflex is inhibited in normals (Tyler, 1963); (ii) the increase of the F-wave to M-wave ratio in the human extensor digitorum brevis after BTX injection (Hamjian and Walker, 1994); (iii) reciprocal decrease in spastic co-contraction, found both in the injected agonist and its antagonist in patients with spasticity (Gracies et aI., 2001, 2002b).

659

FOCAL INJECTION THERAPY

Indeed, decreased Renshaw inhibition releases reciprocal Ia inhibition from the injected agonist to the antagonist during movement (in previous studies, changes in reciprocal Ia inhibition were measured at rest only in previous studies; Priori et al., 1995; Modugno et al., 1998). - Increased presynaptic inhibition directed toward the motoneurons supplying the injected muscle has been shown after intramuscular injections of BTX, while Ia reciprocal inhibition at rest, i.e. in the absence of gamma motoneuronal activity or significant ongoing stretch, seems unchanged (Priori et al., 1995; Modugno et al., 1998). - Reduced motoneuronal excitability in the motoneurons supplying the injected muscle and in distant motoneurons is suggested by several arguments: (i) Transcranial magnetic stimulation studies demonstrate increased central conduction time to the injected muscle (but not to its antagonist or to the same muscle contralaterally) two weeks after intramuscular BTX injection (Pauri et al., 1999,2000). (ii) Prolongation of the latencies and reduction of the persistence of Fwaves in segments remote from the injection site are observed one week after treatment (Wohlfarth et al., 2001). Widespread reduction of motoneuronal excitability after focal BTX injections, as suggested by the F-wave changes (Wohlfarth et al., 2001), may have particular value in any form of symptomatic multifocal dystonia. In patients with spastic paralysis for instance, there is increased amplitude and higher persistence of Fwaves (Bischoff et al., 1992; Hultbom and Nielsen, 1995, 1996). Such distant hypoexcitability effects of BTX injection may contribute to reducing distant synergistic contractions ("overflow"), as has been reported in some cases of non-spastic dystonic syndromes (Gelb et al., 1991; Valls-Sole et al., 1991). Central reflex alterations have also been reported after blepharospasm injections (Behari and Raju, 1996) although other blepharospasm or cervical dystonia studies have been negative in that respect (Valls-Sole et al., 1994; Girlanda et al., 1996). These central effects do not seem to include changes in conduction along axons (Guyton and McDonald, 1947; Wohlfarth et al., 2001) or in ion channels (e.g. sodium, fast and slow potassium, calcium) at the presynaptic membrane (Mallart et aI., 1989).

41.5.3.2. Action on most active terminals BTX uptake is enhanced in nerve terminals that are most active, whether hyperactivity is induced by nerve stimulation or increased voluntary activity (Hughes and Whaler, 1962; Eleopra et al., 1997; Chen et al., 1999). This should allow the weakening effect to be focused in those muscle fibers that are pathologically overactive. However, it has been difficult to assess the clinical significance of this finding. Hesse's studies in hemiparetic patients show greater improvement following BTX injection when performing periodic stimulation of the injected muscle and its antagonist for three 30-minute sessions a day during the three days following injection (Hesse et al., 1995, 1998). However, electrical stimulation by itself improves tone and function in patients wth spasticity (Pandyan et al., 1997; Chae et al., 1998; Hesse et al., 1998). 41.5.3.3. Potential intrafusal action Acetylcholine is also a neurotransmitter in intrafusal muscle fibers, suggesting that the mechanism of action of BTX may also relate to blockade of intrafusal fibers. However, the animal data suggesting this hypothesis is limited (Manni et al., 1989; Filippi et al., 1993; Rosales et al., 1996). The human data available is also scarce. In hemiplegic patients assessed before and after BTX injection into biceps brachii, our results suggest that spasticity in the elbow flexors is more readily affected by BTX than contractions of descending origin such as in cocontraction or in maximal voluntary effort (Gracies et al., 2001). Most intrafusal fibers are encapsulated in spindles and it is not known how well the 900Kd BTX molecule could penetrate the capsule. In the context of muscle disuse due to extrafusal dysfunction after injection of BTX-A, atrophy of intrafusal fibers is not necessarily proof that spindle function has been blocked (Rosales et al., 1996). In patients with spastic paralysis assessed before and after BTX injection into overactive muscles, we could not find changes in spindle thixotropy (dependence of stretch responses on the immediate history of contraction and length changes) that would be suggestive of intrafusal block (Gracies et al., unpublished data). 41.5.3.4. Spread of BTX action - To neighboring muscles: Animal models demonstrate that BTX-A readily spreads across muscle

660 fascia (Shaari et al., 1991; George et al., 1992). In humans, weakening and endplate dysfunction have also been shown in muscles adjacent to injected muscles or to non-muscular injection sites, such as in hyperhidrosis (Girlanda et al., 1996; Saadia et al., 2001; Swartling et al., 2001). The potential for spread of BTX action may vary between serotypes: spread of BTX-B to noninjected muscles in the same limb segment seems less than with BTX-A (Callaway et al., 2002). - To remote muscles: Subtle abnormalities in endplate function have been consistently observed in non-injected muscles remote from injection sites, particularly with single fiber EMG studies (Wiegand and Wellhoner, 1974; Sanders DB et al., 1986; Lange et al., 1987, 1991; Olney et al., 1988; Girlanda et al., 1992). However, maximal compound muscle action potential is unchanged in muscles distant from the injected muscles (Wohlfarth et al., 2001). Therefore, remote endplate dysfunction after local BTX injection may be due to general stimulation of terminal sprouting, without involving significant presynaptic inhibition of acetylcholine release in remote muscles, i.e. spread of BTX into the systemic circulation (Olney et al., 1988). 41.5.4. Greater effect on reflex than on descending contractions

With any of the blocking agents available, alcohol, phenol, BTX, or local anesthetics, strength is preserved relative to stretch reflexes i.e. reflex responses are blocked more readily and completely than direct muscle responses to descending command or peripheral nerve stimulation. The greater effects on reflexes may be understood as the result of a synergistic block of reflex activity by interrupting both efferents (alpha and gamma motor neurons) and afferents from the muscle spindle (Fisher et al., 1970, 1971). This may be true even when only interrupting the alpha and gamma efferents, since gamma efferents act functionally as afferents, enhancing the messages from spindles. Hence partial curarization, which affects only the neuromuscular junction and not afferents, reduces spasticity to a greater degree than voluntary strength (Burman, 1939). The same phenomenon has been suggested with BTX injections in the upper limb of hemiplegic patients (Gracies et al., 2001).

J-M. ORACIES AND D.M. SIMPSON

41.6. Adverse effects, precautions and contraindications 41.6.1. Local anesthetics

The risks described below require that resuscitation equipment and personnel trained in the management of acute cardiopulmonary emergencies are immediately available when local anesthetic is infiltrated into tissue. 41.6.1.1. Toxic systemic effects If local anesthetics inadvertently enter the systemic circulation, they may interfere with the function of all organs where impulse conduction or transmission occurs. Central nervous system effects may begin with central stimulation, including restlessness, tremor and convulsions, which may respond to benzodiazepines. The mechanism responsible for these effects is not clear; both reduced garnma-aminobutyric acid (GABA)-mediated inhibition and direct stimulation of cortical cells have been reported (Ueyama et al., 1999; Ye et al., 1999). However, in cases of severe overdose, stimulation effects are followed by CNS depression, and death can occur due to respiratory failure (Murdoch-Ritchie and Greene, 1980). In cases of slight overdose and mild diffusion into systemic circulation, patients may experience temporary symptoms such as dizziness, fuzziness, blurred vision and auditory impairment (Shiomi et al., 1997). These mild systemic symptoms may last about one hour following lidocaine overdose. The cardiovascular system may be affected at higher systemic concentrations than those affecting the CNS. Local anesthetics injected systemically have a spasmolytic action on smooth muscles, and most cause arteriolar dilatation (though cocaine does not), likely due to decrease in sympathetic nerve efferent activity (Murdoch-Ritchie and Greene, 1980). Therefore, a decrease in arterial pressure and an inhibition of cardiovascular reflexes are possible, as shown with lidocaine in animal experiments (Karatassas et al., 1993; Ness, 2000). Lidocaine has been associated with sudden cardiac collapse after injection at high doses in humans (100 mg intravenously plus 300 mg intramuscularly; Berntsen and Rasmussen, 1992). In the heart, local anesthetics act primarily in the myocardium, in decreasing excitability, conduction rate and force of contraction, similarly to quinidine. For these reasons, lidocaine is

FOCAL INJECTION THERAPY

commonly used to prevent ventricular fibrillation at the initial phase of an acute myocardial infarction (Berntsen and Rasmussen, 1992). 41.6.1.2. Hypersensitivity This rare but potentially major adverse reaction may result in a mild allergic rash to a fatal anaphylactic reaction (deJong, 1977). It is especially encountered with local anesthetics of the ester type (cocaine, procaine, tetracaine). 41.6.1.3. Fall or joint injury Following local anesthetic block in the lower limb, particularly in para- or hemiparetic patients, there can be significant change in ambulation and transfers. Patients may have become used to the joint position and tension from their overactive muscles. When overactivity is suddenly removed, there is a risk of joint sprain that mandates careful monitoring by the clinician. 41.6.1.4. Precautions and contraindications - Known hypersensitivity, especially with local anesthetics of the ester type (cocaine, procaine, tetracaine), represents an absolute contraindication because of the risk of fatal anaphylactic reaction (deJong, 1977). - Liver insufficiency: Since the metabolism of local anesthetics occurs mainly in the liver (especially for lidocaine), the extensive use of a local anesthetic in patients with severe hepatic dysfunction should be avoided (Murdoch-Ritchie and Greene, 1980).

661

the pain experienced during the procedure (Hariga et al., 1964; Kaji et al., 1995). 41.6.2.2. Chronic dysesthesia and pain This is a common and troublesome adverse effect of perineural injections in particular. The incidence of dysesthesia reported after perineural phenol blocks has varied from 2 to 32% (Helweg-Larsen and Jacobsen, 1969; Petrillo et aI., 1980; Glenn, 1990; Yadav et al., 1994). The incidence may even be higher in reality since many cases may not be reported, and the same is true for alcohol blocks (Truscelli, personal communication). Dysesthesias are usually reported from a few days to about two weeks after the procedure, and are generally experienced as burning paresthesiae, exacerbated by light tactile stimulation (Petrillo et aI., 1980). The typical duration is several weeks but chronic dysesthesia has been reported (Helweg-Larsen and Jacobsen, 1969). The mechanism is not clear, although it may involve abnormal regrowth of sensory axons. A uniformly applied compressive garment such as a sock, glove, elastic wrap, or Lycra sleeve (Gracies et aI., 1997,2000) may minimize the effects of other superficial stimulation and decrease edema if present. Other therapeutic methods have been to reblock the nerve with alcohol or phenol (Petrillo et aI., 1980), to perform surgical neurolysis (Braun et aI., 1973), or to use systemic analgesic treatment (DeLateur, 1972).

Reports of adverse effects are less common with alcohol than phenol injections.

41.6.2.3. Sensory loss or permanent peripheral nerve palsy Sensory loss is common in the first hours or days following perineural phenol but this usually resolves (Petrillo et aI., 1980). Permanent peripheral nerve palsy has been reported after phenol injection in the obturator and peroneal nerves (Glenn, 1990; Koman et aI., 1996). Permanent functional loss of sensation is a rare occurrence (Glenn, 1990).

41.6.2.1. Pain at injection Ethyl alcohol and phenol injected intramuscularly cause burning pain (Glenn, 1990), such that some have suggested conscious sedation or general anesthesia, particularly in children (Koman et al., 1996). Application of ethyl chloride spray, or lidocaine spray or cream (Ernla®) over the skin insertion site, or injection of lidocaine or other local anesthetic into the injection site prior to alcohol injection, decrease

41.6.2.4. Vascular complications Cases of phlebitis have been reported with alcohol (0' Hanlan et aI., 1969), and with phenol neurolysis (Macek, 1983). Mechanisms may involve necrosis of the intima of arteries and veins and thrombotic occlusion of small vessels. This makes the routine precaution of aspirating before injecting very important. Other potential mechanisms include trauma to the vein or loss of muscle pumping action, leaving

41.6.2. Alcohol and phenol

662

the extremity more susceptible to stagnation of venous blood. Local pain and swelling may be present for a few days and mimic deep venous thrombosis, especially when it occurs in the calf (Halpern and Meelhuysen, 1966; Garland et al., 1984). Spinal ischemia following deep plexus injections has been reported with alcohol (Wong and Brown, 1995) and with phenol (Holland and Youssef, 1978). 41.6.2.5. Muscle necrosis and skin lesions Painful muscle necrosis has been a rare but difficult problem in children when using alcohol concentrated beyond 75% (Truscelli D, personal communication). Skin irritation may be secondary to superficial injection. Local hyperemia (redness) may occur after injection, but this lasts usually less than 36 hours (Carpenter and Seitz, 1980; Carpenter, 1983). Skin slough and torpid ulcerations have been reported with alcohol (Somma-Mauvais et al., 1992) and with phenol injections (Glenn, 1990). Phenol and alcohol are bacteriocidal at the concentrations used for neurolysis, and local infection at the site of phenol injection has been rarely reported (Felsenthal, 1974a; Koman et al., 1996). After intramuscular phenol injection, induration with tender nodules has been reported one to three weeks post injection (Easton et al., 1979; Sun, 1990). 41.6.2.6. Systemic effects Acute: Ninety to 98% of ethyl alcohol that enters the body is completely oxidized, and patients may exhibit signs and symptoms of acute ethanol intoxication in the immediate post injection period (Koman et al., 1996). Severe overdosage has not been reported in association with phenol neurolysis. However, the possibility of general side effects warrants caution in avoiding accidental intravascular injection. Overdose with phenol causes tremor, convulsions, central nervous system depression, cardiac dysrrythmias and cardiovascular collapse (Glenn, 1990; Morrison et al., 1991; Itoh, 1995). However, the amount of phenol routinely used for nerve blocks is usually well below the lethal range, which begins at 8.5 gr (Wood, 1978). A 10 ml injection of 5% phenol contains 0.5 gr of phenol. To remain within safe limits, no more than 1 gr per day should be injected (Glenn, 1990). Chronic: To our knowledge, no retrospective or prospective studies have been reported in humans, in

J-M. GRACIES AND D.M. SIMPSON

particular in children, evaluating the genotoxic and myelotoxic risks of repeat phenol injections (McDonald et al., 2001). However, methods for detection and quantification of phenol in plasma have been developed to increase safety, both in environmental and industrial use and for children given phenol injections (Harrison et al., 1991; Morrison et a!., 1991). 41.6.2.7. Risks associated with injection sites other than intramuscular and perineural Subarachnoid administration of phenol was often used from the 1950s to the 1970s (Maher, 1955; Nathan, 1959; Cain, 1965; Nathan et al., 1965; Fagerberg and Hook, 1970; Greitz et al., 1972; Browne and Catton, 1975). However, whether with phenol or alcohol, this technique carries grave risks of arachnoiditis or spinal necrosis (Pedersen and Reske-Nielsen, 1965; Stefanko and Zebrowski, 1968; Hughes, 1970; Holland and Youssef, 1978; Morgan and Steller, 1994; Hetherington and Dooley, 2000) and is now generally reserved as an antispastic means for severe tetra- and paraplegic patients (Scott et al., 1985; Iwatsubo et al., 1994; Asensi et al., 1999). It continues to be used in analgesic indications for pain related to intractable cancer or multiple sclerosis (Dahm et al., 1998). Epidural alcohol and phenol injections have been used to target otherwise inaccessible proximal muscles (i.e. iliopsoas, quadratus lumborum), or lumbar or sacral paraspinal muscles (Tardieu G et al., 1968, 1971; Loubser, 1995). However, epidural 3% and 6% phenol may spread centrally, and may cause damage to posterior nerve roots, and direct spinal cord injury (Stefanko and Zebrowski, 1968; Katz et a!., 1995). Finally, the intrinsic complications of any epidural intervention must be considered, especially the possibility of spinal subdural hematoma (Tekkok et al., 1996). Perineural in lumbosacral paravertebral regions: This injection site has been used for muscles that are difficult to access directly, such as the psoas major, because of its size, the quadratus lumborum, because of its proximity to the peritoneal cavity and the kidney, and the paraspinal lumbar or sacral muscles (Tardieu G et al., 1962). To target these muscles, injection can be made at the paravertebral level using fluoroscopy, ultrasound or CT guidance (Meelhuysen et al., 1968; Koyama et al., 1992). Lumbosacral paravertebral blocks with phenol have the risk of

663

FOCAL INJECTION THERAPY

accidental intrathecal injection via the root sleeves, and this could also cause cauda equina or spinal cord injury (O'Rahilly, 1986).

an emergency by contacting the manufacturer of Botox", Allergan (800) 433 8871 or (714) 246 5954.

41.6.3. Botulinum toxin

41.6.3.3. Contraindications The only absolute contraindication to BTX is known hypersensitivity to any component of the formulation. The following are relative contraindications and require caution:

Excellent tolerance overall is the greatest advantage of BTX when compared to other focal injection therapies. The most common side effects are local, minor and of short duration.

41.6.3.1. Local adverse effects - Hematoma at the injection site may result from puncture of small blood vessels during the injection. This risk exists for any injection of blocking agent when the patient is anticoagulated, particularly when the INR (International Normalized Ratio) is greater than 3.5 (Parziale et al., 1988). Patients receiving anticoagulant therapy should be treated with caution depending on the intensity of anticoagulation, the depth of the muscle to be injected and the consequences of local bleeding. - A skin rash resembling neurodermatitis is a rare complication that may occur after repeated injections (LeWitt and Trosch, 1997).

41.6.3.2. Regional or systemic adverse effects Rare cases akin to plexopathy have been reported after use ofBTX-A (Glanzman et al., 1990; Sampaio et al., 1993). Cases of dysphagia, dry mouth and ptosis have been reported more often with BTX-B than BTX-A after use for spasmodic torticollis (Racette et al., 2002). Systemic effects or effects remote from the injection site are highly uncommon. The LD50 of Botox" in monkeys after intravenous or intramuscular administration is estimated to be 40 U per kg of body weight (Herrero et al., 1967; Scott and Suzuki, 1988). The LD50 in humans has been extrapolated from these data in monkeys and is approximately 3500 U for an adult 70 kg human male (Scott, 1981), which is 6 to 7 times the ceiling dose currently recommended (The WE MOVE Spasticity Study Group, 2002). Should overdosage, accidental injection or oral ingestion occur, the patient should be monitored for several days for signs and symptoms of systemic weakness or muscle paralysis. Additional information may be obtained in

- Pregnancy: "category B2". The neurotoxin molecular weight of 150 kDa makes passive transport across the placental barrier unlikely. There are reported cases of botulism during the second and third trimester of pregnancy in otherwise healthy women, in which the neonate had no evidence of botulism (St. Clair et al., 1975; Robin et al., 1996). However, an active transport mechanism has not been excluded, and safety for use in pregnancy and lactation is not demonstrated, as animal reproduction studies have not been conducted with BTX injection. It is thus recommended that therapy be withheld from pregnant and nursing women or women of childbearing age unless use of a suitable contraceptive is confirmed. - Coagulopathy or concomitant anticoagulant treatment with International Normalized Ratio (INR) greater than 3.5. - Pre-existing disorder of the neuromuscular junction, e.g. myasthenia gravis (Borodic, 1998). - Use of aminoglycosides and other drugs interfering with neuromuscular transmission. 41.7. Modalities of use, clinical effects and indications

41.7.1. Local anesthetics 41.7.1.1. Modalities of use of the different local anesthetics There are a large number of synthetic local anesthetics. They differ by their delay and duration of action, anesthetic potency and associated risks (Murdoch-Ritchie and Greene, 1980). Local anesthetics may be divided into three categories by their duration of action: short (20 to 45 minutes) such as procaine, intermediate (1 to 4 hours) such as

664

lidocaine and prilocaine, and long (several hours) such as tetracaine, etidocaine and bupivacaine. The duration of action increases with the amount of drug injected, as does the risk of systemic toxic reactions. Therefore, duration of action is more safely prolonged by the addition of epinephrine (see above). As noted above, lidocaine, etidocaine, and bupivacaine are now generally preferred over procaine. Lidocaine produces more prompt, intense, longlasting and extensive anesthesia than does an equal amount of procaine (Kalichman and Calcutt, 1992). Unlike procaine, which is an ester, lidocaine is an aminoethylarnide and, therefore, is less likely to provoke hypersensitivity reactions (see above). Lidocaine is available as lidocaine hydrochloride (Lignocaine", Xylocaine", others) in dilutions from 0.5% to 4%, with or without epinephrine (l :200000; 5 I-L/rnl), which can double the duration of effect (Murdoch-Ritchie and Greene. 1980). Dilutions of lidocaine used for infiltrations and blocks are usually between 0.5% and 2% (Kaji et al., 1995; Kiernan et al., 1997). Lidocaine 2% demonstrates faster onset and longer duration of action than lidocaine 1% in ulnar nerve motor blocks (Atanassoff et al., 1998). When used without epinephrine, up to 5 rug/kg of 0.5-2% lidocaine solution can be used for nerve block or infiltration anesthesia (Murdoch-Ritchie and Greene, 1980). We keep the injected volume of lidocaine 2% below 10-12cc in adult patients to avoid mild toxicity effects (dizziness, blurred vision, see above). Many specialists now favor Etidocaine (Duranest") and bupivacaine hydrochloride (Marcaine") for tests of muscle relaxation, for their longer duration of action than lidocaine. Bupivacaine is also more potent than lidocaine, and can be used in amounts up to 3 mg/kg of 0.25 to 0.75% solution (0.6 ml/kg for a 0.5% solution; Murdoch-Ritchie and Greene, 1980). The long duration of action of bupivacaine may allow thorough assessment during the anesthetic period. Etidocaine is a long-acting derivative of lidocaine that is favored by some clinicians for its propensity to block motor fibers more than sensory fibers (Bromage et al., 1974). Its effects last 2-3 times as long as lidocaine, with about the same induction time. Etidocaine is available in 0.5% and 1.0% solution with or without epinephrine, and in 1.5% solution with epinephrine (l :200,000; 5 I-Lg/ml). The maximal dose is 6 mg/kg without epinephrine and 8 mg/kg with epinephrine.

J-M. ORAClES AND D.M. SIMPSON

41.7.1.2. Technical issues for use in patients with muscle overactivity 41.7.1.2.1. Intramuscular use. The maximum effect of an intramuscular block may be obtained when the drug is injected within the target muscle in the vicinity of the neuromuscular junctions (Koman et al., 1996), which is consistent with the known sensitivity of neuromuscular junctions to local anesthetics (De long, 1977; Ruff, 1977). Intramuscular blocks may be more painful than nerve blocks at proximal branches or sensorimotor trunks (Glenn, 1990). Also, large volumes of anesthetic increase the likelihood of spread to nearby muscles or other structures (nerve trunks) that one may not wish to affect. However, intramuscular injection blocks motor more than sensory activity and thus may predict more accurately the subsequent action of an agent such as BTX. 41.7.1.2.2. Perineural use: nerve block. A mixed peripheral nerve consists of individual nerve fascicles surrounded by epineurium. The vascular supply is usually centrally located. When a local anesthetic is injected near a peripheral nerve, it diffuses from the outer surface toward the core down its concentration gradient, blocking first the nerve fibers located in the outer mantle of the trunk. These fibers usually innervate more proximal structures than those situated near the core. The duration of their blockade is also longer than for central fibers, because the vascular uptake of the anesthetic usually occurs primarily in the core of the mixed nerve. 41.7.1.3. Indications in muscle overactivity: diagnostic procedures and therapeutic tests Intramuscular local anesthesia (lidocaine, etidocaine, bupivacaine) has long been used as a diagnostic tool (Carpenter, 1983). The short duration of effect of local anesthetic blocks makes them useful as temporary diagnostic tests, before providing long lasting treatment (Wassef, 1993). A local anesthetic block can help answer a number of questions, not only about therapeutic indications, but also about the mechanisms involved in the functional impairment. 41.7.1.3.1. Prediction offunctional changes with long-term therapy. The primary question addressed by a short-term block is whether a long-term block of the same muscle or nerve might be functionally

665

FOCAL INJECTION THERAPY

useful. In order to answer this question, objective means of functional assessment must be designed before the block is performed.

41.7.1.3.2. Understanding of impairment mechanisms. To evaluate the mechanisms of impairment using short-term block, we recommend that the block be powerful enough to significantly weaken the muscle injected (as opposed to merely reduce reflex reactions to stretch). The experience with alcohol and phenol injections (see below) and with BTX suggests that, regardless of the blocking agent, significant weakening of the injected muscle may be required to improve antagonist function in patients with spastic paralysis (Gracies et al., 2001, 2002b). The issues that a local anesthetic block may then help address include: (i) What are the relative roles of muscle overactivity and of contracture in the impairment of function? By temporarily paralyzing a muscle without lengthening it, the block may provide an answer to this question. (ii) Which muscles contribute to the pathologic posturing? Glenn provides a clear illustration of this question with the example of foot inversion during gait in hemiplegia (Glenn, 1990). In some cases of foot inversion during swing phase of gait, the deformity may result from the inability of the tibialis anterior to overcome overactive or contractured plantar flexors, so that all of its contractile force moves the foot towards inversion rather than dorsiflexion. In such cases, blocking the pure plantarflexors (soleus and gastrocnemii) may confirm this hypothesis by reducing the varus deformity if the resistance opposed by the pure plantarflexors is due to overactivity and not shortening. In other cases, inversion during swing phase is primarily due to overactivity of the tibialis posterior, which may be confirmed by blocking this muscle. (iii) In cases of spastic hypertonia, what are the contributions of spasticity and contracture in the resistance to passive stretch? This is a more technical issue relating to the interpretation of the clinical examination. It is overlooked by any clinical scale that only assesses resistance to passive movement in a semi-quantitative fashion (e.g. "mild", "more marked", "consider-

able"; Ashworth, 1964) instead of assessing its type (e.g. the Tardieu Scale: existence of a catch-and-release, or of a clonus, fatigable or not, which are specifically produced by muscle contraction in reaction to stretch, and not by soft tissue resistance; Gracies et al., 2000).

41.7.1.3.3. Preparation to casting. Local anesthetic blocks have been used as preparation for casting (plaster or fiberglass) treatment of contractures (Glenn, 1990). Placing the extremity in a cast soon after the block stretches the relaxed muscle, which may inhibit the return of overactivity and help improve the efficacy and tolerability of the cast. However, forceful overactivity may later return within the cast so that the angle of the cast may be adjusted at only 50% of the gain in passive range of motion obtained from the block (Elovic E, personal communication). 41.7.1.3.4. Analgesic. Lidocaine has been used as an analgesic for intramuscular injection procedures when mixed with the medication to be injected. It reduces the pain associated with intramuscular injection of antibiotics (Schichor et al., 1994), of the anesthetic propofol (Eriksson, 1995), or of alcohol (Tardieu G et al., 1964; Hasegawa et al., 1990; Kaji et al., 1995), and with steroid injection around nerves or epidurally (Martin et al., 1994; Hong et al., 1996). However, binding between the injected molecule and the local anesthetic agent may require injecting the drug in greater amounts (Eriksson, 1995). The possibility of such interference should be considered when mixing lidocaine and alcohol (Kaji et al., 1995). 41.7.2. Chemical neurolysis with alcohol and phenol 41.7.2.1. Modalities of use: dilutions Alcohol: The dilution range most commonly reported in intramuscular treatment of spastic overactivity with alcohol injections has been 35% to 60% (Tardieu G et al., 1962; O'Hanlan et al, 1969; Pelissier et al., 1993; Kong and Chua, 2002). In our experience in adults, these dilutions have proved insufficient to achieve long duration of effect, and we routinely use absolute (98%) dehydrated alcohol

666 (Faulding Pharmaceuticals Co, or Taylor Pharmaceuticals) for motor point injections. Phenol: Selected peripheral nerves can be injected with 4% to 6% aqueous phenol percutaneously or at higher concentrations under direct vision (Keenan et al., 1987; Yadav et al., 1994). Glycerin may be added to render the phenol more soluble in aqueous solutions (Koman et al., 1996).

41.7.2.2. Clinical results and indications 41.7.2.2.1. Spastic overactivity. Alcohol: In the early 1960s, Guy Tardieu and colleagues pioneered alcohol (45%) injections directly into muscle in children with cerebral palsy (Tardieu G et al., 1962; Hariga et al., 1964; Tardieu C et al., 1964; Tardieu G et al., 1964). After motor point injection, they reported reduction of spasticity in most cases without change in voluntary strength. The duration of effect lasted from 6 months to 3 years (Hariga et al., 1964; Tardieu C et al., 1964; Tardieu G et aI., 1964; Tardieu G et al., 1968; Cockin et aI., 1971). Instead of targeted motor point injections, a technique consisting of injecting large quantities of 45% alcohol (between 10 and 40 ml according to the muscle) into multiple locations within the target muscles of spastic patients also resulted in significant spasticity reduction without loss of voluntary motor power or sensation (O'Hanlan et al., 1969). Carpenter and Seitz popularized this technique under the name "intramuscular alcohol wash" and performed the procedure under general anesthesia because of the local pain during injection (Carpenter and Seitz, 1980; Carpenter, 1983). These authors obtained their best results with gastrocnemius muscle injections, but reported a short duration of effect, from one to six weeks (Carpenter and Seitz, 1980). Chua and Kong recently revived the use of perineural and intramuscular alcohol in spastic paralysis. They reported experience with 50%-100% alcohol injections into the musculocutaneous, sciatic, obturator and tibial nerves, and 50% into the forearm flexors in spasticity (Kong and Chua, 1999a, b, 2002; Chua and Kong, 2000, 2001). Increased range of passive motion and reduced tone were reported for at least six months in all cases. Functional changes were assessed only qualitatively in these open label studies. Phenol: Phenol has been used in perineural more often than in motor point sites (Griffith and Mel-

J-M. GRACIES AND D.M. SIMPSON

ampy, 1977; Garland et aI., 1984; Morrison et al., 1989) and the literature on the clinical efficacy of phenol in muscle overactivity consists chiefly of anecdotal reports (Katz et al., 1967; Khalili, 1984). Focal blocks permit global release of overactivity in the limb injected (Mooney et al., 1969). In contrast to the precise physiological and histological data available (Bodine-Fowler et aI., 1996; Westerlund et al., 2001) the reported duration of clinical effect of phenol injections has been as variable as with alcohol. It has ranged from 10 days to 36 months (Katz et al., 1967; Easton et al., 1979; Petrillo et al., 1980; Khalili, 1984; Petrillo and Knoploch, 1988). To our knowledge there has been only one double blind evaluation of the effects of phenol in spasticity (Kirazli et aI., 1998). This randomized study compared 3 m1 of 5% phenol injected perineurally about the tibial nerve in the popliteal fossa with 400 units of BTX-A injected intramuscularly to treat overactive calf muscles in chronic stroke patients. Three months post treatment, there was no difference in efficacy between the two techniques. Complications such as common peroneal nerve palsy occurred with perineural phenol only, such that intramuscular BTX-A was deemed safer than perineural phenol. There has not been a comparison of these two agents using intramuscular injection technique for both. Intrathecal injections: This treatment has the potential to convert spastic to flaccid paraplegia (Scott et al., 1985). There are few reports using alcohol (Bruno, 1975; Chabal et aI., 1989; Loubser, 1990; Asensi et al., 1999), or phenol (Kelly and Gauthier-Smith, 1959; Nathan, 1959; Fagerberg and Hook, 1970; Browne and Catton, 1975; Iwatsubo et aI., 1994; Jarrett et al., 2002). However, complications have been frequently reported, including bladder, bowel and sexual dysfunction, as well as sensory loss or painful dysesthesiae in the lower limbs, and thrombosis of spinal arteries (Hughes, 1970; Jarrett et al., 2002). We do not advocate this site of injection unless as a last resort solution in very severely affected patients - for instance who already have evident bladder and bowel dysfunction before injection - when other treatments are impossible or not suitable (Asensi et aI., 1999).

41.7.2.2.2. Non-spastic overactivity. Alcohol has recently been used in association with 0.5% to 1% lidocaine, in repeated injections as local treatment of upper limb dystonia (10% dilution; Kaji et al., 1995)

667

FOCAL INJECTION THERAPY

or spasmodic torticollis (Poemnyi et al., 1976; Hasegawa et al., 1990; Mezaki et al., 2000). Its lesional effects on muscular tissue have also been exploited to lesion the "re-entry circuits" characteristic of Wolff Parkinson White Syndrome (Sun et al., 1987; Khao et al., 1990; Creswell et al., 1994). Phenol: Halpern and Meelhuysen efficiently treated "extrapyramidal" rigidity of sternomastoid and leg adductors in two patients with Parkinson's disease with 3-5% phenol injections (Halpern and Meelhuysen, 1966). The concentrations used (higher than 3%) may have been sufficient to also block descending outputs to the muscle rather than just the reflexes from afferents within the muscle. Similarly, phenol has been used in spasmodic torticollis (Poemnyi et al., 1976; Massey, 1995; Danisi et al., 2000).

41.7.2.2.3. Other indications. Exploitation of the tissue destructive effects of phenol and alcohol have included subtrigonal injections in hyperactive bladder (Bennani, 1994), sclerotherapy in hemorroids (Santos et al., 1993) or esophageal varices (Supe et aI., 1994), chemonucleolysis in intradiscal injections as an alternative to surgical treatment for lumbar disc herniation (Chiba, 1993), retrobulbar injection for pain relief in cases of blind painful eyes (Birch et al., 1993), sympathectomy in limb ischemia (Greenstein et al., 1994), proximal gastric vagotomy in refractory ulcers (Neuberger et al., 1994), and caudal epidural injection in hyperhydrosis in patients with cervical cord injury (Yamauchi et al., 1993). 41.7.2.2.4. Duration of effect. Overall, various factors may influence the duration of effect of chemical neurolysis, but none has been studied in a controlled fashion: - Concentration and volume used for injection (Mooney et al., 1969; Sung et al., 2001). - Site of the block: intramuscular, peripheral nerve, paravertebral, "intramuscular targeting endplates" (DeLateur, 1972). - Treatment variables after the block; for example, effective stretch after nerve block may further control spastic overactivity (Feldman, 1990). - The presence of selective motor control in the muscles supplied by the nerve treated prior to the block may be associated with a longer duration of effect (Braun et al., 1973).

- Number of prior injections: while Awad suggested that the effect might become definitive after three or four phenol injections (Awad, 1972a), others have not reported duration change after repeat injections (Helweg-Larsen and Jacobsen, 1969).

41.7.3. Botulinum toxin 41.7.3.1. Modalities of use 41.7.3.1.1. Dosing. We limit here our discussion to the two serotypes currently approved for focal injection therapy, types A and B. BTX-A Two approved clinical preparations of BTX-A (Botox", Allergan; Dysport", Ipsen) have been compared based on their potency in animals. These comparisons have yielded highly variable results, with conversion factors ranging from I: I (equivalence) to 5:1 (i.e. 1 unit of Botox" equivalent to 5 units of Dysport") in animals' studies (Pearce et al., 1994; Wohlfarth et al., 1997). For facial injections in patients, a single-blind study found that a 4: 1 potency ratio (l unit Botox®=4 units Dysport") was a satisfactory estimate (Sampaio et al., 1997a). Apparent potency differences between the two toxins may simply be related to the different dilutions used for each (Bigalke et al., 2001). In addition, the mouse lethality bioassay classically used to evaluate the potency of a BTX preparation is characterized by an inherent lack of sensitivity and large inter-laboratory variability (Pearce et al., 1994; Mclellan et al., 1996). Regional denervation assays may have better performance for this purpose (Pearce et al., 1995; Sesardic et al., 1996). While guidelines based on clinical experience have been proposed (The WE MOVE Spasticity Study Group, 2002), BTX-A dosages for each muscle are not accurately established and may vary depending upon the size of the muscle injected, the severity of muscle overactivity and the body weight of the patient. Even though Botox" doses of up to 900 U per session have been safely injected (Jayasooriya et al., 2002), a panel of experienced clinicians have recommended to stay within a ceiling total dose per injection session, currently estimated at 600 U in adults, and 12-15U/kg in children (The WE MOVE Spasticity Study Group, 2002). We are not aware of such consensus decision on a ceiling

668 dose regarding Dysport", but the highest doses reported as safe in controlled trials have been 1500 V in adults (Bakheit et al., 2000; Hyman et al., 2000) and 30 Vlkg in children (Baker et al., 2002). These constraints should not influence the dose used in one particular muscle, which should remain compatible with clinical efficacy. However, the ceiling dose often limits the number of muscles that can be injected in one session.

BTX-B The doses of BTX-B (Myobloc" in the VSA, Neurobloc" in the rest of the world, Elan) are not as well established as the drug's use is recent, particularly for limb indications. Preclinical acute toxicity studies of BTX-B in monkeys detected no systemic muscle weakness in doses up to 960 U/kg (Meyer et al., 1999). Thus, it appears safe to use doses at or below 300 Vlkg (21,000 V in a 70 kg patient) in initial human investigations (Moberg-Wolff and Walke, 2002). Current safe and well-tolerated doses of up to 20,000 V of Myobloc" have been used in open-label clinical trials in limb injections following brain damage (Lew et al., 1997,2000; Cullis et al., 2001; Jayasooriya et al., 2002; Moberg-Wolff and Walke, 2002; Royal and Jenson, 2002; Brashear et aI.,2003). 41.7.3.1.2. Preparation. Botox" (Allergan) and Dysport" (Ipsen) are prepared in lyophilized forms of BTX-A, while Myobloc" (Elan) is prepared as a solution of BTX-B (5000 Uzml), Botox" is supplied sterile in glass vials, each containing 100 units of BTX-A, 0.5 mg of human albumin, 0.9 mg of sodium chloride and no preservative, at neutral pH. Dysport" is available in vials of 500 units (50 ng) of BTX-A hemagglutinin complex, 0.125 mg of human albumin and 2.5 mg of lactose. BTX-B (Myobloc", Neurobloc", Elan) is supplied as a solution in sterile glass vials containing 0.5, 1 or 2 ml of BTX-B 5000Vlml in 0.05% human serum albumin, 0.01 M sodium succinate, 0.1 M sodium chloride at pHS. 41.7.3.1.3. Storage. Careful handling of purified toxin is important for maintenance of stability. BTXA is readily denatured by heat at temperatures above 40°C, or in atmospheres of nitrogen or carbon dioxide. When BTX-A is stored at 4°C in the form of the last crystallization, toxicity is retained for 10 or

J-M. GRACIES AND D.M. SIMPSON

more years (Schantz and Johnson, 1992). However, the commercially available product is vacuum-dried and sterile, and Allergan recommends storing it in a freezer at or below -5°C. BTX-B must not be frozen and is to be stored under refrigeration at 2-8°C. 41.7.3.1.4. Dilution and administration. BTX-A must be diluted in sterile, non-preserved saline prior to use for injection. Injection of the diluent into the vial must avoid bubbling or agitation that may denature the toxin. Reconstituted BTX-A should be stored in a refrigerator (2° to 8°C) and may then retain its potency even after two weeks following reconstitution (Sloop et al., 1997). Animal studies have demonstrated that the degree of paralysis is related to the injected volume (Shaari and Sanders, 1993). Dilutions of BTX for human use have varied considerably (Gracies and Simpson, 2000). We conducted a double-blind protocol comparing the efficacy of two dilutions of Botox", 100 Vlml and 20 Vlml, in the biceps brachii in spastic hemiparetic patients. When injecting biceps brachii with BTX-A in spasticity, a high-volume/ low-potency dilution (20 Vlml, i.e. 5 ml per vial of Botox") provides significantly greater neuromuscular block and spasticity reduction than a low-volume/high-potency dilution (l00 Vlml, i.e. 1 ml per vial) (Gracies et al., 2002d). For injection in large muscles, such as those in the lower limb or in upper limb muscles above the elbow, we recommend using high dilution (e.g. 20 Uzml), in particular when the primary goal is to improve passive function. BTX-B can be injected as drawn or further diluted with normal saline. If diluted the preparation must be used within 4 hours, as it does not contain any preservative. 41.7.3.1.5. Antibody formation. It is the principle of protective immunity that any foreign protein administered into the body causes the formation of antibodies. The important question is whether the normal process of immunogenicity is likely to have significant impact on the clinical efficacy of BTX treatment. Several arguments point to a negative response, and to a marginal significance of this issue for clinical use. Antibodies to BTX may be of two types: nonneutralizing (the most common) and neutralizing. By definition, only neutralizing antibodies reduce therapeutic efficacy. Non-neutralizing antibodies

FOCAL INJECTION THERAPY

bind on any member of the packaging proteins or any portion of the neurotoxin that does not disrupt its function. Neutralizing antibodies bind on any portion of the protein that disrupts its entry into or function within neurons. Neutralizing antibodies exert effects by inhibiting cellular receptor binding by the toxin at the heavy chain binding domain (Dertzbaugh and West, 1996). Antibodies against the active enzymatic site of the light chain are less likely to be of significance or even present, because the light chain is protected until it gets into the neuron. Various tests have been developed to assess the presence of antibodies. The in vivo mouse protection bioassay (commercially available through the Northwest Pacific Laboratories, Berkeley, CA) tests whether increasing dilutions of a patient's serum protects experimental mice from lethal test doses of BTX-A (Hatheway and Dang, 1994). Several in vitro assays have been developed, including an enzymelinked immunosorbent assay (ELISA, Notermans et al., 1978; Dezfulian and Bartlett, 1984), a spherelinked immunodiagnostic assay (SLIDA, Siatkowski et al., 1993), and a radioimmuno-precipitation assay (Palace et al., 1998; Hanna et al., 1999). However, despite its questionable sensitivity (Borodic et al., 1996; Hanna and Jankovic, 1998; Hanna et al., 1999), only the mouse bioassay appears clinically relevant, since it reflects "neutralizing" antibodies. The mouse bioassay was the only assay that had 100% specificity and positive predictive value in two retrospective studies (Hanna and Jankovic, 1998; Hanna et al., 1999). Other in vitro tests may be useful for detecting neutralizing antibodies, using mouse nerve-muscle preparation such as the Mouse Diaphragm Assay (Goschel et al., 1997; Dressler et al., 2001, 2002). Investigators have also proposed clinical assays in patients, to determine whether clinical resistance to BTX injections can be ascribed to neutralizing antibodies. These assays include test injections in sudomotor glands (Birklein et al., 2002), in selected forehead muscles (Hanna et al., 1999), or in other readily accessible indicator muscles, such the extensor digitorum brevis (Kessler and Benecke, 1997). However, targeting thin muscles such as the frontalis with a non-guided injection may not ensure a reproducible amount of neuromuscular block. Repeated exposure to BTX may not impart longterm immunity, since recurrent episodes of botulism caused by the same serotype have been reported

669 (Beller and Middaugh, 1990). However, the immune response to parenteral presentation of the antigen may not follow this pattern. In laboratory workers who repeatedly receive toxoid injections, BTX-A antibodies were slow to develop and began rising steeply after the fourth year of immunization (Siegel, 1988). The threshold dose or injection frequency of BTX required to trigger significant antibody formation in humans is not known and may vary among individuals. Clinical resistance is a complex problem that may be ascribed to multiple mechanisms before invoking antibodies. It is not well correlated with the presence of neutralizing antibodies, even measured with a sensitive test such as the Mouse Diaphragm Assay (Dressler et al., 2001, 2002). The following arguments need to be taken into account when evaluating the clinical relevance of the issue of BTX antibodies. - The repeated suggestions that increasing dose or frequency of injections may lead to increased antibody presence and to clinical resistance, and all the ensuing recommendations to "keep the dose of BTX as low as possible" are mostly based on retrospective studies (Tsui et al., 1988; Hambleton et al., 1992; Zuber et al., 1993; Greene et al., 1994, Hanna and Jankovic, 1998; Hanna et al., 1999; Rollnik et al., 2001). However, retrospective studies are inappropriate to evaluate this issue. In such series it is possible that "clinical resistance" to injections was partly due to inappropriate choice of target muscles in difficult cases. "Clinical resistance" may then have led to increased injection frequency and doses, and not the reverse. - "Clinical resistance" or the so-called "secondary non-response" has been raised more frequently by specialists treating non-spastic dystonic disorders than by those treating spastic conditions, even though the latter typically use much higher doses. It is noteworthy that none of the dystonia studies employed the stimulation technique for muscle targeting (Tsui et al., 1988; Hambleton et al., 1992; Zuber et al., 1993; Greene et al., 1994; Hanna and Jankovic, 1998; Hanna et al., 1999). Therefore, a second potential cause for "clinical resistance" may have been uncontrolled targeting accuracy, even in appropriately selected muscles (Hanna et al., 1999).

670

- The presence of antibodies at a given point in time does not predict resistance to later injections, as antibody titers spontaneously vary in time, and many patients, initially positive with the mouse protection bioassay, are found negative when retested after several months (Sankhla et al., 1998; Hanna et al., 1999; Dressler and Bigalke, 2002). This suggests lack of reproducibility of bioassays, or waning of the anamnestic immunologic response to BTX-A. - Whether type A or B, the quantity of BTX administered at each treatment is of the order of nanograms, i.e. a minute quantity. Reducing this quantity at each injection, as has often been recommended (Greene et al., 1994; Rollnik et al., 2001; Tintner and Jankovic, 2001) does not change its order of magnitude, and there is no prospective evidence that these recommendations have any impact on BTX immunogenicity, or most importantly on the efficacy of further treatments. To the contrary, there is recent evidence that increasing the dose of BTX injections may overcome existing neutralizing antibodies and maintain normal BTX efficacy without increasing antibody titers (Dressler et al., 2002).

In brief, there are no definitive answers concerning the incidence of neutralizing antibodies, the risk factors for developing them, or more importantly their clinical relevance, i.e. their correlation with clinical efficacy of BTX injections. Most studies investigating "clinical resistance to BTX" have been retrospective, and none has used the stimulation technique to ensure accurate muscle targeting. A significant part of the "secondary non-response" reported in retrospective studies may have been due to factors other than neutralizing antibodies. Only one prospective study has been done in the patient population that requires the highest doses (patients wih spasticity), and showed development of neutralizing antibodies in only 1% of cases (Turkel et al., 2002). Guidelines imposing maximal single dose or minimal time interval between injections for strict immunologic purposes may lead to underdosing patient treatments, while they have not been validated with rigorous prospective studies.

41.7.3.1.6. Antidotes. In case of accidental overdose, trivalent equine antitoxin, which reduces botulism mortality if administered early, is available

J-M. ORACIES AND D.M. SIMPSON

from the Centers for Disease Control and Prevention (CDC) through an emergency distribution system (Shapiro et al., 1998). Injection of antitoxin at a specific time following BTX injection partially prevents BTX effects (Scott, 1988). However, the equine antitoxin may lead to undesirable antiserum reactions (Hibbs et al., 1996). One may also use human immunoglobulin G pooled from immunized human volunteers, which is a potential treatment for infant botulism (Frankovich and Arnon, 1991). Within four hours after BTX treatment for blepharospasm, injection of human botulinum immune globulin (IG, 3.2 x 10-3 international units (ill) per unit of Botox") into the levator palpebrae superioris is effective in limiting or avoiding ptosis (Scott, 1997). Other antidotes are under evaluation for potential use. Physiological means may reduce the action of BTX-A, such as increasing the intracellular calcium concentration with ionophore treatment (Ashton and Dolly, 1991), or using the potassium-channel blocker aminopyridine (Siegel et al., 1986; Gansel et al., 1987; Adler et al., 1996, 2000; Dock et al., 2002). Other compounds awaiting trial include anticholinesterase drugs (e.g. edrophonium; Stahl et al., 1998), enalapril (Captopril") for its anti-zinc effect (Bhattacharyya and Sugiyama, 1989), arninoquinolines, such as chloroquine, and other metalloprotease inhibitors (Sheridan et al., 1997; Adler et al., 1998).

41.7.3.2. Clinical results and indications BTX blocks all cholinergic synapses and has been used as local therapy for both its antinicotinic effects in muscle overactivity, and its antimuscarlnic effects in autonomic disorders. 41.7.3.2.1. Spastic muscle overactivity. For maximal efficacy, treatment with BTX in this type of musle overactivity must be associated with a stretch program for the injected muscles and, when possible, a training or electrical stimulation program on the antagonist (Gracies, 2001; Hesse et al., 2002). Upper limb There is considerable evidence that injection of BTX into muscles with spastic overactivity reduces resistance to passive movement in joints supplied by the injected muscles and improves resting posture (Das and Park, 1989a, b; Hesse et al., 1992; Konstanzer et al., 1993; Wall et al., 1993; Dunne

FOCAL INJECTION THERAPY

et al., 1995; Simpson et al., 1996; Yablon et al., 1996; Corry et al., 1997; Sampaio et aI., 1997b; Bakheit et al., 2000, 2001; Bhakta et aI., 2000; Friedman et aI., 2000; Hurvitz et al., 2000; Lagalla et aI., 2000; Richardson et aI., 2000; Rodriquez et aI., 2000; Smith et al., 2000). However, it has been difficult to demonstrate functional improvement, except for activities accomplished for the patient by the caregiver (i.e. "passive function"; Mayer et aI., 2001; Sheean, 2001; Brashear et aI., 2002). The demonstration of improvement in active function (activities that the patient can do alone using the paretic limb) represents the most difficult challenge in clinical research on focal injection therapy and spasticity. Meeting this challenge might set apart focal injection therapy from the systemic antispastic treatments (oral and intrathecal), still used as primary therapy by most clinicians (Gracies et al., 2002c). Lower limb Resistance to passive movement in joints supplied by injected muscles is reduced, whether in open or controlled studies (Dengler et aI., 1992; Hesse et al., 1994, 1996; Dunne et aI., 1995; Burbaud et al., 1996). Few studies assessed gait velocity in adult hemiplegia (Hesse et al., 1994, 1996; Burbaud et aI., 1996; Kirazli et al., 1998), and reported increases by 15-17% after calf muscle injection in double blind protocols (Burbaud et aI., 1996; Kirazli et al., 1998). Means to enhance the efficacy of BTX include splinting of the injected muscle to provide continuous stretch (Reiter et aI., 1998), or chronic stimulation of the injected muscle and its antagonist (Hesse et al., 1995, 1998). In cerebral palsy (CP), BTX therapy results in functional improvement, and potential delaying of tendon-lengthening surgery (Koman et al., 1993, 1994, 1996; Cosgrove et aI., 1994; Corry et aI., 1998; Baker et aI., 2002; Polak et aI., 2002; Reddihough et al., 2002). In multiple sclerosis (MS), BTX-A or BTX-B in overactive adductors may benefit nursing care and improve comfort for sitting in a wheelchair (Snow et aI., 1990; Borg-Stein et aI., 1993; Kerty and Stien, 1997; Hyman et al., 2000; Oechsner, 2002). 41.7.3.2.2. Non-spastic muscle overactivity. - Parkinsonian syndromes: Examples of use of BTX in extrapyramidal overactivity include treat-

671

ment of focal "off dystonia" in Parkinson's disease (Pacchetti et aI., 1995; Limousin et aI., 1997; Tsui, 2003), treatment of gait freezing by calf muscle injection (Giladi and Honigman, 1997; Giladi et aI., 2001), and relief of rigidity in progressive supranuclear palsy (Polo and Jabbari, 1994). - Tremor: Open label studies have suggested usefulness of BTX in parkinsonian and essential tremor, whether in hand (Trosch and Pullman, 1994; Modugno et al., 1998; Pacchetti et aI., 2000), head (Finsterer et aI., 1996; Wissel et aI., 1997) or vocal cords (injection into the thyroarytenoid muscle; Hertegard et al., 2000; Warrick et al., 2000a, b). Controlled trials for hand tremor have involved injecting equal doses into agonists and antagonists contributing to the tremor (Pahwa et al., 1995; Henderson et al., 1996; Jankovic et al., 1996; Brin et aI., 2001). In these trials, the risk of wrist drop and weakness of grip mitigated against functional improvement (Pahwa et al., 1995; Henderson et al., 1996; Jankovic et al., 1996; Pacchetti et aI., 2000). An important factor to consider is the localization of the main contributors to the tremor. Domzal suggested lower doses of BTX for the forearm extensors than flexors (Domzal, 1998), and we routinely obtain functional improvement with injection into one agonist only (flexors or pronators; unpublished). Occasionally injection must be done into shoulder muscles if these are the main drivers of the tremor (Slawek et aI., 2000). - Blepharospasm and hemifacial spasm: Numerous open reports (Frueh et al., 1984; Savino et al., 1985; Dutton and Buckley, 1988; Kalra and Magoon, 1990) and controlled trials (Yoshimura et al., 1992; Mezaki et al., 1995, 1999; Nussgens et aI., 1997; Sampaio et al., 1997a) have shown the efficacy of BTX injections in hemifacial spasm and blepharospasm, with success rates exceeding 90%. Clinical improvement negatively correlates with past psychiatric history (Burbaud et al., 1995). The most common complications are local and transient. They include ptosis (Dutton and Buckley, 1988; Nussgens et al., 1997), symptomatic dry eye (Dutton and Buckley, 1988), weakness of eye closure, tearing (Savino et aI., 1985; Kalra and Magoon, 1990), photophobia, altered facial expression, facial droop (Kalra and Magoon, .1990), facial bruising (Kalra and

672

Magoon, 1990), subtle visual blurring (Kalra and Magoon, 1990) and diplopia (Dutton and Buckley, 1988). The frequency of these complications does not increase with the number of injections (Mauriello and Aljian, 1991) but does correlate with the dose injected (Yoshimura et aI., 1992). "Apraxia of lid opening", which may occur in parkinsonian syndromes such as progressive supranuclear palsy, has been successfully treated with BTX (Krack and Marion, 1994; Lepore et al., 1995; Piccione et aI., 1997; Forget et aI., 2002). BTX has also been used to increase facial symmetry contralateral to treated hemifacial spasm (Clark and Berris, 1989), to treat postparalytic synkinesis following facial nerve regrowth or hypoglossal-facial anastomosis (Boroojerdi et aI., 1998), and to induce protective ptosis in patients with corneal ulcers or prophylactically in those with severe peripheral facial palsy with injections in the levator palpebrae superioris (Adams et aI., 1987). - Sphincter disorders: BTX may be used to treat several conditions characterized by sphincter hyperactivity. A recent controlled double blind study has shown that injections of BTX into the external urethral sphincter using a transperineal technique were more efficient than lidocaine 0.5% in improving detrusor-sphincter dyssynergia due to spinal cord injury (De Seze et aI., 2002), which confirmed previous reports on the role of BTX in this indication (Dykstra et al., 1988; Dykstra and Sidi, 1990; Schurch et aI., 1990, 1997). Injections of BTX into the external urethral sphincter have also been successfully used in the treatment of acute urinary retention post pelvic surgery and other forms of voiding dysfunction (Phelan et aI., 2001; Smith CP et aI., 2002). However, this treatment may cause stress incontinence in women with chronic urinary retention (Fowler et aI., 1992). BTX-A or BTX-B injection into the external anal sphincter may be helpful for patients with anismus (Hallan et aI., 1988), chronic anal fissure (Jost, 1997,2001; Maria et al., 1998; Brisinda et aI., 1999), and persistent constipation after surgery for Hirschprung disease (Langer and Birnbaum, 1997; Minkes and Langer, 2000). Potential issues are a transient fecal incontinence or anal hematoma that may occur after this

J-M. GRACIES AND D.M. SIMPSON

treatment (Jost, 1997; Madalinski et aI., 2002), and late recurrence of chronic anal fissure when the effect of BTX wears off (Minguez et al., 2002). BTX may be injected into the cricopharyngea1 muscle for patients with dysphagia due to overactivity of the upper esophageal sphincter (Schneider et aI., 1994), and into the lower esophageal sphincter in mild to moderate achalasia (Pasricha et aI., 1995; Neubrand et aI., 2002), although pneumatic dilatation has been reported to be a more effective and less costly treatment, particularly in moderate to severe achalasia (Vaezi et aI., 1999; Ghoshal et al., 2002; Neubrand et al., 2002). BTX injections may also be used diagnostically and therapeutically in Oddi sphincter dysfunction (Pasricha et aI., 1994; Muehldorfer et aI., 1997; Wehrmann et al., 2000) and are under evaluation in the treatment of infantile hypertrophic pyloric stenosis (Albanese et al., 1995; Heinen et aI., 1999). - Spasmodic torticollis: BTX is FDA approved and now considered the standard medical treatment for patients with spasmodic torticollis since placebo- and trihexyphenidyl-controlled studies have demonstrated the efficacy of BTX-A and BTX-B (Tsui et aI., 1986; Brans et aI., 1996; Lew et al., 1997, 2000). However, proper selection of the involved muscles and choice of dose is critical in achieving a favorable outcome (Munchau et aI., 2001). Specific physical therapy to train the antagonists of the muscles responsible for the pathological postures may also be helpful, as are psychosocial interventions (Soulayrol et al., 1993; Scheidt et aI., 1995, 1998; Gundel et aI., 2001; Sharma and Gupta, 2002). Transient side effects of BTX injection include swallowing difficulties and neck weakness (Berardelli et aI., 1990). The hypothesis that autonomic side effects, including dryness of mouth, accommodation difficulties and conjunctival irritation, may be more common with BTX-B than BTX-A, is under evaluation (Dressler and Benecke, 2003). - Focal limb dystonia: injection of BTX into selected hand and forearm muscles was effective in a placebo-controlled double blind study of focal hand dystonia (Cole et aI., 1995). Injections may be started in a few of the involved muscles, chosen as being the likely "trigger muscles", and associated with retraining of deficient motor skills

673

FOCAL INJECTION THERAPY

(Marion, 1999). However, BTX treatment restores near-normal movement only in a minority of mildly affected patients (Wissel et al., 1996). - Ophthalmologic indications: Since Scott's pioneering work (Scott et al., 1973), retrobulbar injections of BTX-A into eye muscles have become a valuable adjunct to surgery in the treatment of strabismus, or acquired or congenital nystagmus when it causes oscillopsia or decreased visual acuity, particularly after 18 months of age (Repka et al., 1994; Carruthers, 1995; Lennerstrand et al., 1998; Ruiz et al., 2000). Potential adverse effects are due to excessive spread of BTX and include diplopia, keratitis and ptosis (Tomsak et al., 1995; Lennerstrand et al., 1998). - Miscellaneous indications: BTX has been used for pre- or postoperative management of orthopedic surgery, to facilitate immobilization, by injection into muscles surrounding the target joint (Traynelis et al., 1992; Racette et al., 1998; Umstadt, 2002; Kayikcioglu et al., 2003). BTX has been successfully used in inherited benign crampfasciculation syndrome at doses that did not significantly affect muscle strength (Bertolasi et al., 1997). Oromandibular and lingual dystonia, and abductor and adductor spasmodic dysphonia are commonly managed with BTX injection (Brin et al., 1987; Blitzer et al., 1989). BTX is also used for cosmetic purpose, including reduction of facial wrinkles and glabellar lines (Keen et al., 1994; Ascher et al., 1995), with the latter indication recently approved in the United States by the FDA.

41.7.3.2.3. Autonomic disorders. The anticholinergic effect of BTX is clinically apparent following local injection near secretory glands (Ekstrom et al., 1977). These properties are now used in the treatment of various hypersecretory disorders, including Frey's syndrome (gustatory sweating; Laskawi et al., 1998), axillar or palmar hyperhidrosis (Schnider et al., 1997; Saadia et al., 2001; Naumann and Lowe, 2002), hyperlacrimation (Boroojerdi et al., 1998), and drooling (Jongerius et al., 2003). In these indications, the duration of effect of BTX may be considerably longer than its action at the neuromuscular junction (Laskawi et al., 1998; Wo1lina et al., 2002).

41.7.3.2.4. Pain disorders. After two controlled studies suggesting a positive effect of BTX in low back pain (Foster et al., 2001) and migraine prevention (Silberstein et al., 2000), BTX has been used for various specific pain syndromes such as the piriformis syndrome, myofascial pain syndrome, chronic neck and low back pain, and fibromya1gia (Fishman et al., 2002; Sheean, 2002). However, the double blind controlled evidence for the use of BTX in pain conditions is still weak (Sheean, 2002). Indeed, double blind placebo-controlled studies have been negative for tension-type headaches (Schmitt et al., 2001; Rollnik and Dengler, 2002), orofacial pain (Nixdorf et al., 2002), and myofascial pain syndrome (Wheeler et al., 1998). In addition a placebocontrolled study has shown that there is no direct peripheral antinociceptive effect of BTX-A after sub-cutaneous injection in humans (Blersch et al., 2002). BTX in intra ganglionic injections as a sympatholytic remains to be evaluated for syndromes ascribed to sympathetically maintained pain (Childers et al., 2002; Kim et al., 2002).

41.7.3.2.5. Delay and duration of action. The latency period from injection to onset of clinical improvement is two to six days. After BTX injection in human extensor digitorum brevis, the peaks of maximum compound action potential (CMAP) and mean rectified voltage (MRV) reduction are reached in six days, and both CMAP and MRV are still only two thirds of their control preinjection value 100 days after injection (Hamjian and Walker, 1994). In most non-spastic conditions, BTX-A therapy relieves undesired muscle contraction for an average duration of three to four months, after which repeat injection is required. In spastic conditions the effect is often longer (Corry et al., 1998). The action of BTX for autonomic disorders appears more prolonged than its action at the neuromuscular junction (Schnider et al., 1997; Laskawi et al., 1998). 41.8. Injection by electrical stimulation technique

For both intramuscular and perineural injections, and whichever blocking agent is used, we recommend using the exploratory stimulation technique (Gunduz et al., 1992; Wassef, 1993; Geenen et al.,

674

1996; Gracies and Simpson, 1999, 2000), as is traditionally used for nerve blocks in rehabilitation and anesthesiology (Paqueron et a1., 1999; Chua and Kong, 2001). The same needle that will inject the drug is used to transmit repetitive monopolar stimulation of the targeted nerve or muscle, in order to adjust its position. The tip of the needle is directed as close as possible to the nerve trunk by searching for the minimal stimulation able to elicit the appropriate muscle twitch or paresthesia. Similarly, the needle will lie selectively in the targeted muscle when the largest bulk of twitch, as assessed clinically, is obtained in isolation with the minimal possible stimulation. For the most precise localization among muscles, the stimulation area must be as small as possible. Therefore, pulse widths should be as short as possible (0.05 to 0.2 ms) and needles of small caliber (27-30G) should be used. These needles are commonly used for BTX injections, as opposed to the 22G needles often used for nerve blocks in anesthesia, and are usually rigid enough to penetrate deep limb muscles. The stimulation technique may be slightly more time consuming than other techniques (e.g. "clinical, visually guided", or "EMG guided" i.e. guided by the acoustic feedback of the muscle activity detected from the needle tip), but it is the only technique that can determine with certainty which muscle the needle tip lies in (Geenen et a1., 1996). Children with cerebral palsy may represent an exception: in this population injection into superficial muscles may be accomplished by visually guided injection, as it is much faster than the stimulation technique and does not require sedation. Otherwise, deep conscious sedation or general anesthesia have been recommended for perineural injections, in particular when exposing the target peripheral nerve surgically (Griffith and Melampy, 1977; Easton et a1., 1979; Koman et a1., 1996). Specific intramuscular targeting Motor point targeting: When using alcohol, phenol or local anesthetics, an intramuscular block destroying distal nerve branches, may allow easier titration of the effect than a more proximal block of the whole trunk, although the intramuscular procedure may be more painful (Griffith and Melampy, 1977). The identification of small motor nerve branches can be facilitated by the use of atlases or charts that depict the usual location of motor points

J-M. GRACIES AND D.M. SIMPSON

within a given muscle (Walthard and Tchicaloff, 1971; Warfel, 1985). Anatomic guides may help the practitioner in nerve localization (Felsenthal, 1974b; Labib and Gans, 1984; O'Rahilly, 1986). Endplate targeting: To optimize the efficacy of intramuscular injections, particularly with BTX, it is possible to target endplates within a muscle (Gracies et a1., 2002d). DeLateur proposed a technique of injection as close as possible to endplate areas (DeLateur, 1972). Motor endplates usually do not occur at random in muscles, but cluster at characteristic areas (the "innervation band"), since the endplate generally lies near the midpoint of any given muscle fiber (Lapicque, 1931; Zack, 1971). However, there are exceptions, such as numerous innervation bands scattered throughout human sartorius, gracilis and gastrocnemius muscles (Zack, 1971; Aquilonius et a1., 1984; Sanders et a1., 1998). DeLateur used a hollow Teflonf-coated injection needle as an electromyographic exploring electrode to find characteristic electrical potentials in the muscle at rest, signifying the immediate proximity of endplates (DeLateur, 1972). These include the characteristic "endplate ripple" or "endplate noise" (Jones et a1., 1955; Buchthal and Rosenfalck, 1966; Wiederholt, 1970), a low voltage increase in irregularity of the baseline of about 10 to 40 I..l.V, and monophasic spike discharges, or diphasic with negative onset, entirely or almost entirely negative in sign, with a random pattern of discharge (as opposed to fibrillation potentials; Buchthal and Rosenfalck, 1966). Buchthal and Rosenfalk showed that when the concentric needle was displaced slightly from the area of endplate ripple, the discrete monophasic negative spikes were reached (Buchthal and Rosenfalck, 1966). Accordingly, when the monophasic negative spikes are found, the exploring electrode is close to the endplate zone. While this electromyographic exploring technique of endplate targeting seems attractive, we have not found it highly practical, since it may be tedious and contact of the needle with endplate zones can be painful for the patient, even with slow movements of the exploring electrode. Ethyl chloride in spray, or skin wheals of xylocaine can be used over each of the approximate cutaneous sites corresponding to endplate areas. However, these techniques may only reduce cutaneous nociceptive activity, without preventing pain sensation from deep nociceptors around the endplate.

675

FOCAL INJECTION THERAPY

Open nerve blocks: These may be performed to ensure that only motor branches are being blocked, but this involves anesthesia, and an incision that might temporarily restrict the use of the extremity (Keenan et aI., 1987). Sedation may be used if the adult patient is likely to have difficulty tolerating the procedure. In young children or uncooperative, brain injured patients, general anesthesia may be the only way to ensure a proper completion of the block (Helweg-Larsen and Jacobsen, 1969; Griffith and Melampy, 1977; Easton et aI., 1979). Targeting techniques according to anatomical location For the user of focal injection therapy, the literature on chemical neurolysis provides many examples of the use of the stimulation technique. The practical value of this literature may lie more in the targeting techniques described for various anatomical sites, than in the details of the clinical results obtained from these open label trials. We have listed useful references for injection techniques into various anatomical sites in Table 1.

41.9. Conclusions The use of focal injection therapy for patients with disorders characterized by muscle overactivity, autonomic dysfunction or pain is still in its early stages, particularly for its most recently introduced agent botulinum toxin. Whichever blocking agent is under consideration, we recommend using the exploratory stimulation technique, whether a nerve or a muscle is targeted. Functional benefit from the block may depend on significant weakening in the targeted muscle. When long lasting blocking treatment is considered, a temporary block with a local anesthetic may be useful to assess the mechanisms of functional impairment and to help predict what improvement may be expected. In the last three decades, alcohol and phenol blocks have only rarely been evaluated in a controlled fashion, as opposed to botulinum toxin. For these alcohol compounds there is a need for placebocontrolled studies to reach firm conclusions on their safety and efficacy. The number of adverse effects

Table 1 Site of block

Lower limb Paravertebral hip flexor Obturator nerve Sciatic nerve Vastus latera1is motor point Posterior tibial nerve

Upperlimb Brachial plexus Subscapularisnerve Pectoralis major motor point Musculocutaneous nerve Brachioradialis motor point Forearm flexors motor points Median nerve Ulnar motor branch by surgical access

References Mee1huysen et aI., 1968;Awad, 1972b; Koyama et aI., 1992 Pelissier et aI., 1993;Wassef, 1993;Yadav et aI., 1994; Kong and Chua, 1999b; Viel et aI., 2002 Pelissier et aI., 1993; Chua and Kong, 2000 Albert et aI., 2002 Garland and Rhoades, 1978;Garland et aI., 1982; Petrillo and Moore and Anderson, 1991;Wichers, 1991;Yadav et al., 1994; Kirazli et al., 1998; Chua and Kong,2001

Katz et aI., 1966; Keenan, 1988 Chironna and Hecht, 1990; Hecht, 1992 Botte and Keenan, 1988; Keenan, 1988; Pelissier et al., 1993; Garland and Rhoades, 1978;Garland et al., 1982; Wainapel, 1988; Keenan et aI., 1990; Kong and Chua, 1999a Keenan et aI., 1990 Garland et al., 1984; Kiwerski, 1984; Kong and Chua, 2002 Kiwerski, 1984; Keenanet aI., 1987; Keenan, 1988; Pelissier et al., 1993 Kiwerski, 1984; Keenan et aI., 1987; Keenan, 1988

676

reported with phenol is greater than with alcohol, as is the number of publications reporting phenol use. Whether the benzyl core of phenol carries a significant myelotoxic and genotoxic risk after repeat injection, especially in children, remains to be evaluated.

Comparison between BTX and alcohol compounds There has not been a comparison of these two agents using a similar intramuscular technique. In comparison with BTX, alcohol and phenol have advantages in their early onset of action and perhaps longer duration of effect, low cost, absence of antigenicity, and better stability and practicality of preparation. However, their lack of selectivity on motor function, tissue destructive effect, propensity to cause pain during injection, adverse effects such as chronic painful dysesthesia, local muscle transformations, and vascular reactions may favor BTX use. In current practice, clinicians often use both types of treatment in combination. Alcohol and phenol may be injected perineurally to block large muscles, for which the effective BTX dose would approach or exceed the ceiling dose. BTX is often reserved for injection into smaller and more distal muscles that can be targeted selectively. Respective indications may also be based on the severity and prognosis of the disorder, and the goals of treatment. The absence of histological destruction after repeated BTX injections and the specific action on efferent fibers might make this the preferable agent where there is hope of recovery of active function in the injected limb. Because of their chronic histological effects and the destruction of sensory fibers, alcohol or phenol may prove more appropriate than BTX in cases where treating muscle overactivity is performed primarily for hygiene and comfort, i.e. in patients with severe deficits and poor functional prognosis, where preservation of intact sensory perception is not critical. Finally, BTX is the only focal injection therapy that has potential in autonomic disorders or pain conditions. In disorders with muscle overactivity, pharmacoeconomic considerations mandate that controlled studies comparing neurolytic agents and BTX be carried out in specific patient populations to determine the appropriate indications for each. In the near future, focal injection therapy will be increasingly evaluated against systemic therapy for muscle overactivity, autonomic or pain disorders. In

J-M. GRACIES AND D.M. SIMPSON

contrast with oral treatments, the use of focal injection therapy requires skill, and the quality of the results partially depends on the injector's technique and talent. This might appear as a disadvantage in the current context, in which only a minority of physicians has expertise with the injection techniques. We would predict, however, that its rationale, safety and efficacy will make this emerging treatment strategy a staple of 21st century medicine, and that medical education will be adapted accordingly.

References Adams, GG, Kirkness, CM and Lee, JP (1987) Botulinum toxin A induced protective ptosis. Eye, I: 603-608. Adler, M, Macdonald, DA, Sellin, LC and Parker, GW (1996) Effect of 3,4-diaminopyridine on rat extensor digitorum longus muscle paralyzed by local injection of botulinum neurotoxin. Toxicon, 34: 237-249. Adler, M, Nicholson, JD, Comille, F and Hackley, BE, Jr (1998) Efficacy of a novel metalloprotease inhibitor on botulinum neurotoxin B activity. FEBS Lett., 429(3): 234-238. Published erratum in FEBS Lett., 437: 164. Adler, M, Capacio, Band Deshpande, SS (2000) Antagonism of botulinum toxin A-mediated muscle paralysis by 3, 4-diaminopyridine delivered via osmotic minipumps. Toxicon., 38: 1381-1388. Adler, M, Keller, JE, Sheridan, RE and Deshpande, SS (2001) Persistence of botulinum neurotoxin A demonstrated by sequential administration of serotypes A and E in rat EDL muscle. Toxicon, 39: 233-243. Albanese, A, Bentivoglio, AR, Cassetta, E, Viggiano, A, Maria, G and Gui, D (1995) Review article: the use of botulinum toxin in the alimentary tract. Aliment. Pharmacol. Ther., 9: 599-604. Albert, TA, Yelnik, A, Bonan, I, Lebreton, F and Bussel, B (2002) Effectiveness of femoral nerve selective block in patients with spasticity: preliminary results. Arch. Phys. Med. Rehabil., 83: 692-696. Angaut-Petit, D, Molgo, J, Comella, JX, Faille, L and Tabti, N (1990) Terminal sprouting in mouse neuromuscular junctions poisoned with botulinum type A toxin: morphological and electrophysiological features. Neuroscience, 37: 799-808. Aquilonius, SM, Askmark, H, Gillberg, PG, Nandedkar, S, Olsson, Y and Stalberg, E (1984) Topographical localization of motor endplates in cryosections of whole human muscles. Muscle Nerve, 7: 287-293. Ascher, B, Klap, P, Marion, MH and Chanteloub, F (1995) Botulinum toxin in the treatment of frontoglabellar and periorbital wrinkles. An initial study. Ann. Chir. Plast. Esthet., 40: 67-76. Asensi, V, Asensi, JM, Carton, JA, Maradona, JA, Ona, M and Arechaga, C (1999) Successful intrathecal ethanol

FOCAL INJECTION THERAPY

block for intractable spasticity of AIDS-related progressive multifocal leukoencephalopathy. Spinal Cord, 37: 450--452. Ashton, AC and Dolly, JO (1991) Microtubule-dissociating drugs and A23187 reveal differences in the inhibition of synaptosomal transmitter release by botulinum neurotoxins types A and Br. J. Neurochem., 56: 827-835. Ashworth, B (1964) Preliminary trial of carisoprodol in multiple sclerosis. Practitioner, 192: 540-542. Atanassoff, PG, Kelly, DJ, Ayoub, CM and Broil, SJ (1998) Electromyographic assessment of ulnar nerve motor block induced by lidocaine. J. Clin. Anesth., 10: 641-645. Awad, EA (1972a) Intramuscular neurolysis for stroke. Minn. Med., 8: 711-713. Awad, EA (l972b) Phenol block for control of hip flexor and adductor spasticity. Arch. Phys. Med. Rehabil., 53: 554-557. Baker, MD (2000) Selective block of late Na( +) current by local anaesthetics in rat large sensory neurons. Br. 1. Pharmacol., 129: 1617-1626. Baker, R, Jasinski, M, Maciag-Tymecka, 1, MichalowskaMrozek, J, Bonikowski, M, Carr, L, Macl.ean, J, Lin, JP, Lynch, B, Theologis, T, Wendorff, J, Eunson, P and Cosgrove, A (2002) Botulinum toxin treatment of spasticity in diplegic cerebral palsy: a randomized, double-blind, placebo-controlled, dose-ranging study. Dev. Med. Child Neurol., 44: 666-675. Bakheit, AM, Thilmann, AF, Ward, AB, Poewe, W, Wissel, J, Muller, J, Benecke, R, Collin, C, Muller, F, Ward, CD and Neumann, C (2000) A randomized, doubleblind, placebo-controlled, dose-ranging study to compare the efficacy and safety of three doses of botulinum toxin type A (Dysport) with placebo in upper limb spasticity after stroke. Stroke, 31: 24022406. Bakheit, AM, Pittock, S, Moore, AP, Wurker, M, Otto, S, Erbguth, F and Coxon, L (2001) A randomized, double-blind, placebo-controlled study of the efficacy and safety of botulinum toxin type A in upper limb spasticity in patients with stroke. Eur. J. Neural., 8: 559-565. Bambrick, L and Gordon, T (1987) Acetylcholine receptors and sodium channels in denervated and botulinum-toxin-treated adult rat muscle. J. Physiol. (Lond.), 382: 69-86. Behari, M and Raju, GB (1996) Electrophysiological studies in patients with blepharospasm before and after botulinum toxin A therapy. 1. Neurol. Sci., 135: 7477. Beller, M and Middaugh, JP (1990) Repeated type E botulism in an Alaskan Eskimo. N. EnglJ Med., 322: 855.

677

Bennani, S (1994) Evaluation of sub-trigonal injections in the treatment of the hyperactive bladder. Ann. Urol. Paris, 28: 13-19. Berardelli, A, Carta, A, Stocchi, F, Formica, A, Agnoli, A and Manfredi, M (1990) Botulinum A toxin injection in patients with blepharospasm, torticollis and hemifacial spasm. Ital. J. Neural. Sci., 11: 589-593. Berntsen, RF and Rasmussen, K (1992) Lidocaine to prevent ventricular fibrillation in the prehospital phase of suspected acute myocardial infarction: the NorthNorwegian Lidocaine Intervention Trial. Am. Heart J., 124: 1478-1483. Bertolasi, L, Priori, A, Tomelleri, G, Bongiovanni, LG, Fincati, E, Simonati, A, De Grandis, D and Rizzuto, N (1997) Botulinum toxin treatment of muscle cramps: a clinical and neurophysiological study. Ann. Neurol., 41: 181-186. Bhakta, BB, Cozens, JA, Chamberlain, MA and Bamford, JM (2000) Impact of botulinum toxin type A on disability and carer burden due to arm spasticity after stroke: a randomized double blind placebo controlled trial. J. Neurol. Neurasurg. Psychiatry, 69: 217-221. Bhattacharyya, SD and Sugiyama, H (1989) Inactivation of botulinum and tetanus toxins by chelators. Infect. Immun., 57: 3053-3057. Bigalke, H, Wohlfarth, K, Irmer, A and Dengler, R (2001) Botulinum A toxin: Dysport improvement of biological availability. Exp. Neural., 168: 162-170. Birch, M, Strong, N, Brittain, P and Sandford-Smith, J (1993) Retrobulbar phenol injection in blind painful eyes. Ann. Ophthalmol., 25: 267-270. Birklein, F, Walther, D, Bigalke, H, Winterholler, M and Erbguth, F (2002) Sudomotor testing predicts the presence of neutralizing botulinum A toxin antibodies. Ann. Neurol., 52: 68-73. Bischoff, C, Schoenle, PW and Conrad, B (1992) Increased F-wave duration in patients with spasticity. Electramyogr. Clin. Neurophysiol., 32: 449-453. Black, JD and Dolly, JO (I 986a) Interaction of 125I-labeled botulinum neurotoxins with nerve terminals. I. Ultrastructural autoradiographic localization and quantitation of distinct membrane acceptors for types A and B on motor nerves. J. Cell. Biol., 103: 521-534. Black, JD and Dolly, JO (I 986b) Interaction of 125I-labeled botulinum neurotoxins with nerve terminals. II. Autoradiographic evidence for its uptake into motor nerves by acceptor-mediated endocytosis. J. Cell. Biol., 103: 535-544. Blersch, W, Schulte-Mattler, WJ, Przywara, S, May, A, Bigalke, H and Wohlfarth, K (2002) Botulinum toxin A and the cutaneous nociception in humans: A prospective, double-blind, placebo-controlled, randomized study. J. Neurol. Sci., 205: 59-63.

678 Blitzer, A, Brin, ME Greene, PE and Fahn, S (1989) Botulinum toxin injection for the treatment of oromandibular dystonia. Ann. Otol. Rhinol. Laryngol., 98: 93-97. Bodine-Fowler, SC, Allsing, Sand Botte, MT (1996) Time course of muscle atrophy and recovery following a phenol-induced nerve block. Muscle Nerve, 19: 497-504. Boehnke, SE and Rasmusson, DD (2001) Time course and effective spread of lidocaine and tetrodotoxin delivered via microdialysis: an electrophysiological study in cerebral cortex. J. Neurosci. Meth., 105: 133-141. Bohbot, V, Otahal, P, Liu, Z, Nadel, L and Bures, J (1996) Electroconvulsive shock and lidocaine reveal rapid consolidation of spatial working memory in the water maze. Proc. Natl. Acad. Sci. USA, 93: 4016-4019. Boroojerdi, B, Ferbert, A, Schwarz, M, Herath, Hand Noth, J (1998) Botulinum toxin treatment of synkinesia and hyperlacrimation after facial palsy. J. Neurol. Neurosurg. Psychiatry, 65: 111-114. Borg-Stein, J, Pine, ZM, Miller, JR and Brin, MF (1993) Botulinum toxin for the treatment of spasticity in multiple sclerosis. New observations. Am. J. Phys. Med. Rehabil., 72: 364-368. Borodic, G (1998) Myasthenic crisis after botulinum toxin. Lancet, 352: 1832. Borodic, GE, Pearce, LB, Smith, KL, Phelan, M and Ferrante, R (1993) Botulinum B toxin as an alternative to botulinum A toxin: a histologic study. Ophthal. Plast. Reconstr. Surg., 9: 182-190. Borodic, G, Johnson, E, Goodnough, M and Schantz, E (1996) Botulinum toxin therapy, immunologic resistance, and problems with available materials. Neurology, 46: 26-29. Boroff, DA and Chen, GS (1975) On the question of permeability of the blood-brain barrier to botulinum toxin. Int. Arch. Allergy Appl. Immunol., 48: 495-504. Botte, MJ and Keenan, MA (1988) Percutaneous phenol blocks of the pectoralis major muscle to treat spastic deformities. J. Hand. Surg. (Am), 13: 147-149. Bouaziz, H, Narchi, P, Mercier, FJ, Khoury, A, Poirier, T and Benhamou, D (1998) The use of a selective axillary nerve block for outpatient hand surgery. Anesth. Analg., 86: 746-748. Brans, JW, Lindeboom, R, Snoek, JW, Zwarts, MJ, Van Weerden, TW, Brunt, ER, Van Hilten, 11, Van der Kamp, W, Prins, MH and Speelman, JD (1996) Botulinum toxin vs. trihexyphenidyl in cervical dystonia: a prospective, randomized, double-blind controlled trial. Neurology, 46: 1066-1072. Brashear, A, Gordon, MF, Elovic, E, Kassicieh, VD, Marciniak, C, Do, M, Lee, CH, Jenkins, S and Turkel, C (2002) Botox Post-Stroke Spasticity Study Group. Intramuscular injection of botulinum toxin for the

J-M. ORAClES AND D.M. SIMPSON

treatment of wrist and finger spasticity after a stroke. N. Eng. J. Med., 347: 395-400. Brashear, A, McAfee, AL, Kuhn, ER and Ambrosius, WT (2003) Treatment with botulinum toxin type B for upper-limb spasticity. Arch. Phys. Med. Rehabil., 84: 103-107. Braun, RM, Hoffer, MM, Mooney, V, McKeever, J and Roper, B (1973) Phenol nerve block in the treatment of acquired spastic hemiplegia in the upper limb. J. Bone Joint Surg., 55: 580-585. Brazeau, GA and Fung, HL (1989) Use of an in vitro model for the assessment of muscle damage from intramuscular injections: in vitro-in vivo correlation and predictability with mixed solvent systems. Pharm. Res., 6: 766-771. Brazeau, GA and Fung, HL (1990) Mechanisms of creatine kinase release from isolated rat skeletal muscles damaged by propylene glycol and ethanol. J. Pharm. Sci., 79: 393-397. Brin, MF, Fahn, S, Moskowitz, C, Friedman, A, Shale, HM, Greene, PE, Blitzer, A, List, T, Lange, D, Lovelace, RE et al. (1987) Localized injections of botulinum toxin for the treatment of focal dystonia and hemifacial spasm. Mov. Disord., 2: 237-254. Brin, MF, Lyons, KE, Doucette, J, Adler, CH, Caviness, IN, Comella, CL, Dubinsky, RM, Friedman, JH, Manyam, BV, Matsumoto, JY, Pullman, SL, Rajput, AH, Sethi, KD, Tanner, C and Koller, WC (2001) A randomized, double masked, controlled trial of botulinum toxin type A in essential hand tremor. Neurology, 56: 1523-1528. Brisinda, G, Maria, G, Bentivoglio, AR, Cassetta, E, Gui, D and Albanese, A (1999) A comparison of injections of botulinum toxin and topical nitroglycerin ointment for the treatment of chronic anal fissure. N. Eng. J. Med., 341: 65-69. Bromage, PR, Datta, S and Dunford, LA (1974) An evaluation of etidocaine in epidural analgesia for obstetrics. Can. Anaesth. Soc. J., 21: 535-545. Brooks, VB (1954) The action of botulinum toxin on motor nerve filaments. J. Physiol., 123: 501-515. Browne, RA and Catton, DV (1975) The use of intrathecal phenol for muscle spasms in multiple sclerosis. A description of two cases. Can. Anaesth. Soc. J., 22: 208-218. Bruno, G (1975) Intrathecal alcohol block-experiences on 41 cases. Paraplegia, 12: 305-306. Buchthal, F and Rosenfalck, P (1966) Spontaneous electrical activity of human muscle. Electroencephalogr. Clin. Neurophysiol., 20: 321-336. Burbaud, P,Arnaud, A, Burbaud, A and Lagueny, A (1995) Clinical and psychological factors influencing the efficacy of botulinum toxin in the treatment of

679

FOCAL INJECTION THERAPY

hemifacial spasm and blepharospasm. J. Fr. Ophthalmol., 18: 751-757. Burbaud, P,Wiart, L, Dubos, JL, Gaujard, E, Debelleix, X, Joseph, PA, Mazaux, JM, Bioulac, B, Barat, M and Lagueny, A (1996) A randomized, double blind, placebo controlled trial of botulinum toxin in the treatment of spastic foot in hemiparetic patients. J. Neurol. Neurosurg. Psychiatry, 61: 265-269. Burgen, ASV, Dickens, F and Zatman, LJ (1949) The action of botulinum toxin on the neuromuscular junction. J. Physiol., 109: 10-24. Burgener, FA and Steinmetz, SO (1987) Treatment of experimental adenocarcinomas by percutaneous intratumoral injection of absolute ethanol. Invest. Radiol., 22: 472-478. Burke, 0 and Lance, JW (1973) Pathophysiology of spasticity. Studies of the reflex effects of primary and secondary spindle endings in spasticity. In: JE Desmedt (Ed.), New Developments in Electromyography and Clinical Neurophysiology (Vol. 3). Karger, Basel, pp. 475-495. Burkel, WE and McPhee, M (1970) Effect of phenol injection into peripheral nerve of rat: electron microscope studies. Arch. Phys. Med. Rehabil., 51: 391397. Burman, MS (1939) Therapeutic use of curare and erythroidine hypochloride for spastic and dystonic states. Arch. Neurol. Psychiatry, 41: 307-327. Cain, HD (1965) Subarachnoid phenol block in the treatment of pain and spasticity. Paraplegia, 3: 152160. Callaway, JE, Arezzo, JC and Grethlein, AJ (2001) Botulinum toxin type B: an overview of its biochemistry and preclinical pharmacology. Semin. Cutan. Med. Surg., 20: 127-136. Callaway, JE, Arezzo, JC and Grethlein, AJ (2002) Botulinum toxin type B: an overview of its biochemistry and preclinical pharmacology. Dis. Mon., 48: 367-383. Carpenter, EB (1983) Role of nerve blocks in the foot and ankle in cerebral palsy: therapeutic and diagnostic. Foot Ankle, 4: 164-166. Carpenter, EB and Seitz, DG (1980) Intramuscular alcohol as an aid in the management of spastic cerebral palsy. Dev. Med. Child Neurol., 22: 497-501. Carruthers, J (1995) The treatment of congenital nystagmus with Botox. J. Pediatr. Ophthalmol. Strabismus, 32: 306-308. Chabal, C, Jacobson, L and White, J (1989) Electrical localization of spinal roots for the treatment of spasticity by intrathecal alcohol injection. Anesth. Analg., 68: 527-529. Chae, J, Bethous, F, Bohinc, T, Dobos, L, Davis, T and Friedl, A (1998) Neuromuscular stimulation for upper

extremity motor and functional recovery in acute hemiplegia. Stroke, 29: 975-979. Chen, H and Eastmond, DA (1995) Synergistic increase in chromosomal breakage within the euchromatin induced by an interaction of the benzene metabolites phenol and hydroquinone in mice. Carcinogenesis, 16: 1963-1969. Chen, R, Karp, BI, Goldstein, SR, Bara-Jimenez, W, Yaseen, Z and Hallett, M (1999) Effect of muscle activity immediately after botulinum toxin injection for writer's cramp. Mov. Disord., 14: 307-312. Chiba, K (1993) An experimental study on the pathological changes of the intervertebral disc and its surrounding tissues after intradiscal injection of various chemical substances (the first report). Nippon Seikeigeka Gakkai Zasshi., 67: 1055-1069. Childers, MK, Wilson, OJ, Gnatz, SM, Conway, RR and Sherman, AK (2002) Botulinum toxin type A use in piriformis muscle syndrome: a pilot study. Am. J. Phys. Med. Rehabil., 81: 751-759. Chironna, RL and Hecht, JS (1990) Subscapularis motor point block for the painful hemiplegic shoulder. Arch. Phys. Med. Rehabil., 71: 428-429. Chua, KS and Kong, KH (2000) Alcohol neurolysis of the sciatic nerve in the treatment of hemiplegic knee flexor spasticity: clinical outcomes. Arch. Phys. Med. Rehabil., 81: 1432-1435. Chua, KS and Kong, KH (200 1) Clinical and functional outcome after alcohol neurolysis of the tibial nerve for ankle-foot spasticity. Brain Inj., 15: 733-739. Clark, RP and Berris, CE (1989) Botulinum Toxin: a treatment for facial asymmetry caused by facial nerve paralysis. Plast. Reconstr. Surg., 84(2): 353-355. Cockin, J, Hamilton, EA, Nicholds, PJR and Price, DA (1971) Preliminarv report on the treatment of spasticity with 45% Ethyl Alcohol injection into the muscles. Br. J. Clin. Prac., 25: 73-75. Coffield, JA, Bakry, N, Zhang, RD, Carlson, J, Gomella, LG and Simpson, LL (1997) In vitro characterization of botulinum toxin types A, C and 0 action on human tissues: combined electrophysiologic, pharmacologic and molecular biologic approaches. 1. Pharmacol. Exp. The~,280:

1489-1498.

Cole, R, Hallett, M and Cohen, LG (1995) Double-blind trial of botulinum toxin for treatment of focal hand dystonia. Mov. Disord., 10: 466-471. Corry, IS, Cosgrove, AP, Walsh, EG, McClean, 0 and Graham, HK (1997) Botulinum toxin A in the hemiplegic upper limb: a double-blind trial. Dev. Med. Child Neurol., 39: 185-193. Corry, IS, Cosgrove, AP, Duffy, CM, McNeill, S, Taylor, TC and Graham, HK (1998) Botulinum toxin A compared with stretching casts in the treatment of

680 spastic equinus: a randomized prospective trial. J. Pediatr. Orthop., 18: 304-311. Cosgrove, AP, Corry, IS and Graham, HK (1994) Botulinum toxin in the management of the lower limb in cerebral palsy. Dev. Med. Child Neurol., 36: 386-396. Creswell, LL, Rosenbloom, M, Pirolo, JS, Saffitz, JE and Cox, JL (1994) Potential ablation of accessory atrioventricular pathways: injection of alcohol into the atrioventricular groove. Ann. Thorac. Surg., 57: 203207. Cullheim, Sand Kellerth, 10 (1981) Two kinds of recurrent inhibition of cat spinal alpha-motoneurons as differentiated pharmacologically. J. Physiol., 312: 209224. Cullis, PA (2001) Safety and efficacy of botulinum toxin type B: an open-label, dose escalation study. J. Neurol. Sci., (Suppl. 1): 187. Dahm, PO, Nitescu, PV, Appelgren, LK and Curelaru, I (1998) Long-term intrathecal (i.t.) infusion of bupivacaine relieved intractable pain and spasticity in a patient with multiple sclerosis. Eur. J. Pain, 2: 81-85. Daniels-Holgate, PU and Dolly, 10 (1996) Productive and non-productive binding of botulinum neurotoxin A to motor nerve endings are distinguished by its heavy chain. J. Neurosci. Res., 44: 263-271. Danisi, F, Gracies, JM, Wortzel, SJ and Brin, M (2000) Intramuscular 6% phenol for patients with cervical dystonia resistant to Botulinum toxin type A. Mov. Disord., 15: P162. Das, TK and Park, DM (1989a) Effect of treatment with botulinum toxin on spasticity. Postgrad. Med. J., 65: 208-210. Das, TK and Park, DM (1989b) Botulinum toxin in treating spasticity. Br. J. Clin. Pract., 43: 401-403. DasGupta, BR (1994) Structures of botulinum neurotoxin, its functional domains, and perspectives on the crystalline type A toxin. In: J Jankovic and M Hallett (Eds.), Therapy with Botulinum Toxin. NewYork: Marcel Dekker, pp. 15-39. De Jong, RH (1977) Local anesthetics. Charles C Thomas, Pub, Springfield, Ill. DeLateur, BJ (1972) A new technique of intramuscular phenol neurolysis. Arch. Phys. Med. Rehabil., 53: 179-185. De Lorimier, AA (1995) Sclerotherapy for venous malformations. J. Pediatr. Surg., 30: 188-193. Dengler, R, Neyer, U, Wohlfahrt, K, Bettig, U and Janzik, HH (1992) Local botulinum toxin in the treatment of spastic drop foot. J. Neurol., 239: 375-378. Denny-Brown, D (1966) The cerebral control of movement. Liverpool University Press, Liverpool pp. 124-143, 171-184. De Paiva, A, Meunier, FA, Molgo, J, Aoki, KR and Dolly, JO (1999) Functional repair of motor endplates after

J-M. ORAClES AND D.M. SIMPSON

botulinum neurotoxin type A poisoning: biphasic switch of synaptic activity between nerve sprouts and their parent terminals. Proc. Natl. Acad. Sci. USA, 96: 3200-3205. Dertzbaugh, MT and West, MW (1996) Mapping of protective and cross-reactive domains of the type A neurotoxin of Clostridium botulinum. Vaccine, 14: 1538-1544. De Seze, M, Petit, H, Gallien, P, de Seze, MP, Joseph, PA, Mazaux, 1M and Barat, M (2002) Botulinum a toxin and detrusor sphincter dyssynergia: a double-blind lidocaine-controlled study in 13 patients with spinal cord disease. Eur. Urol., 42: 56-62. Dezfulian, M and Bartlett, JG (1984) Detection of Clostridium botulinum type A toxin by enzyme-linked immunosorbent assay with antibodies produced in immunologically tolerant animals. 1. Clin. Microbiol., 19: 645-648. Dickson, EC and Shevky, E (1923) Botulism. Studies on the manner in which the toxin of Clostridium Botulinum acts on the body. II. The effect on the voluntary nervous system. J. Exp. Med., 38: 327-346. Dobrydnjov, I and Samarutel, J (1999) Enhancement of intrathecal lidocaine by addition of local and systemic clonidine. Acta. Anaesthesiol. Scand., 43: 556-562. Dock, M, Ben Ali, A, Karras, A, Misset, B, GarrousteOrgeas, M, Deletie, E, Goldstein, F and Carlet, J (2002) Treatment of severe botulism with 3,4-diaminopyridine. Presse Med., 31: 601-602. Dolezal, V, Vyskocil, F and Tucek, S (1983) Decrease of the spontaneous non-quantal release of acetylcholine from the phrenic nerve in botulinum-poisoned rat diaphragm. Pflugers Arch., 397: 319-322. Dolly, 10, Williams, RS, Black, JD, Tse, CK, Hambleton, P and Melling, J (1982) Localization of sites for 1251_ labeled botulinum neurotoxin at murine neuromuscular junction and its binding to rat brain synaptosomes. Toxicon,20: 141-148. Dolly, 10, Black, J, Williams, RS and Melling, J (1984) Acceptors for botulinum neurotoxin reside on motor nerve terminals and mediate its internalization. Nature, 307: 457-460. Dolly, 10, Lande, S and Wray, DW (1987) The effects of in vitro application of purified botulinum neurotoxin at mouse motor nerve terminals. J. Physiol., 386: 475-484. Dornzal, TM (1998) Botulinum toxin in the treatment of tremor. Neurol. Neurochir. Pol., 32 (Suppl. 1): 51-56. Dressler, D, Dimberger, G, Bhatia, KP, Irmer, A, Quinn, NP, Bigalke, H and Marsden, CD (2000) Botulinum toxin antibody testing: comparison between the mouse protection assay and the mouse lethality assay. Mov. Disord., 15: 973-976.

681

FOCAL INJECTION THERAPY

Dressler, D and Dirnberger, G (2001) Botulinum toxin antibody testing: comparison between the immunoprecipitation assay and the mouse diaphragm assay. Eur. Neurol., 45: 257-260. Dressler, D and Bigalke, H (2002) Botulinum toxin antibody type A titres after cessation of botulinum toxin therapy. Mov. Disord., 17: 170-173. Dressler, D, Munchau, A, Bhatia, KP, Quinn, NP and Bigalke, H (2002) Antibody-induced botulinum toxin therapy failure: can it be overcome by increased botulinum toxin doses? Eur. Neurol., 47: 118-121. Dressler, D and Benecke, R (2003) Autonomic side effects of botulinum toxin type B treatment of cervical Dystonia and hyperhidrosis. Eur. Neurol., 49: 34-38. Duchen, LW (1970) Changes in motor innervation and cholinesterase localization induced by botulinum toxin in skeletal muscle of the mouse: differences between fast and slow muscles. 1. Neurol. Neurosurg. Psychiatry, 33: 40-54. Duchen, LW (1972) Motor nerve growth induced by botulinum toxin as a regenerative phenomenon. Proc. R. Soc. Med., 65: 196-197. Dunne, JW, Heye, N and Dunne, S (1995) Treatment of chronic limb spasticity with botulinum toxin A. l. Neurol. Neurosurg. Psychiatry, 58: 232-235. Dutton, JJ and Buckley, EG (1988) Long-term results and complications of botulinum A toxin in the treatment of blepharospasm. Ophthalmology, 95: 1529-1534. Dykstra, DD, Sidi, AA, Scott, AB, Pagel, 1M and Goldish, GD (1988) Effects of botulinum A toxin on detrusorsphincter dyssynergia in spinal cord injury patients. l. Urol., 139: 919-922. Dykstra, DD and Sidi, AA (1990) Treatment of detrusorsphincter dyssynergia with botulinum A toxin: a double-blind study. Arch. Phys. Med. Rehabil., 71: 24-26. Easton, JKM, Ozel, T and Halpern, D (1979) Intramuscular neurolysis for spasticity in children. Arch. Phys. Med. Rehabil., 60: 155-158. Edmunds, CV and Long, PH (1923) Contribution to the pathologic physiology of botulism. lAMA, 81: 542547. Ekstrom, J, Kemplay, SK, Garrett, JR and Duchen, LW (1977) Effect of botulinum toxin on the choline acetyltransferase activity in salivary glands of cats. Experientia, 33: 1458-1460. Eleopra, R, Tugnoli, V and De Grandis, D (1997) The variability in the clinical effect induced by botulinum toxin type A: the role of muscle activity in humans. Mov. Disord., 12: 89-94. Eleopra, R, Tugnoli, Y, Quatrale, R, Gastaldo, E, Rossetto, 0, De Grandis, D and Montecucco, C (2002) Botulinum neurotoxin serotypes A and C do not affect motor units survival in humans: an electrophysiological study

by motor units counting. Clin. Neurophysiol., 113: 1258-1264. Eriksson, M (1995) Prilocaine reduces injection pain caused by Propofol. Acta. Anaesthesiol. Scand., 39: 210-213. Fagerberg, G and Hook, (1970) Multiple sclerosis. 8. Intrathecal phenol block in spasticity. Lakartidningen, 67: 3315-3317. Feldman, PA (1990) Upper extremity casting and splinting. In: Glenn & Whyte (Eds), The Practical Management of Spasticity in Children and Adults. Lea & Febiger, Philadelphia-London, Felsenthal, G (l974a) Pharmacology of phenol in peripheral nerve blocks: a review. Arch. Phys. Med. Rehab., 55: 13-16. Felsenthal, G (l974b) Nerve blocks in the lower extremities: anatomic considerations. Arch. Phys. Med. Rehabil., 55: 504-507. Filippi, GM, Errico, P, Santarelli, R, Bagolini, Band Manni, E (1993) Botulinum A toxin effects on rat jaw muscle spindles. Acta. Otolaryngologica (Stockholm), 113: 400-404. Finsterer, J, Muellbacher, Wand Mamoli, B (1996) Yes/ yes head tremor without appendicular tremor after bilateral cerebellar infarction. l. Neurol. Sci., 139: 242-245. Fisher, E et al. (1970) Evoked nerve conduction after nerve block by chemical means. Am. l. Phys. Med., 49: 333-347. Fisher, E et al. (1971) Recovery of nerve conduction after nerve block by phenol. Am. l. Phys. Med., 50: 230234. Fishman, LM, Anderson, C and Rosner, B (2002) BOTOX and physical therapy in the treatment of piriformis syndrome. Am. l. Phys. Med. Rehabil., 81: 936-942. Flanagan, MT, Walker, FO and Butterworth, J (1997) Failure of meperidine to anesthetize human median nerve. A blinded comparison with lidocaine and saline. Reg. Anesth., 22: 73-79. Foran, PG, Mohammed, N, Lisk, GO, Nagwaney, S, Lawrence, GW, Johnson, E, Smith, L, Aoki, KR and Dolly, JO (2003) Evaluation of the therapeutic usefulness of botulinum neurotoxin B, CI, E, and F compared with the long lasting type A. Basis for distinct durations of inhibition of exocytosis in central neurons. 1. BioI. Chern., 278: 1363-1371. Forget, R, Tozlovanu, V, Iancu, A and Boghen, D (2002) Botulinum toxin improves lid opening delays in blepharospasm-associated apraxia of lid opening. Neurology, 58: 1843-1846. Foster, L, Clapp, L, Erickson, M and Jabbari, B (2001) Botulinum toxin A and chronic low back pain: a randomized, double-blind study. Neurology, 56: 12901293.

°

682 Fowler, CJ, Betts, CD, Christmas, TJ, Swash, M and Fowler, CG (1992) Botulinum toxin in the treatment of chronic urinary retention in women. Br. J. Urol., 70: 387-389. Frankovich, TL and Arnon, SS (1991) Clinical trial of botulism immune globin for infant botulism. West J. Med., 154: 103. Franz, DN and Perry, RS (1974) Mechanisms for differential blocks among single myelinated and nonmyelinated axons by procaine. J. Physiol. (Lond.), 236: 193-210. Friedman, A, Diamond, M, Johnston, MV and Daffner, C (2000) Effects of botulinum toxin A on upper limb spasticity in children with cerebral palsy. Am. J. Phys. Med. Rehabil., 79: 53-59, quiz 75-76. Frueh, BR, Felt, DP, Wojno, TH and Musch, DC (1984) Treatment of blepharospasm with botulinum toxin. A preliminary report. Arch. Ophthalmol., 102(10): 1464-1468. Fusfeld, RD (1968) Electromyographic findings after phenol blocks. Arch. Phys. Med. Rehabil., 49: 217220. Gansel, M, Penner, R and Dreyer, F (1987) Distinct sites of action of clostridial neurotoxins revealed by doublepoisoning of mouse motor terminals. Pflugers Arch. Eur. J. Physiol., 409: 404--416. Garland, DE and Rhoades, ME (1978) Orthopedic management of brain-injured adults. Part II. Clin. Orthop., 131: 111-122. Garland, DE, Lucie, RS and Waters, RL (1982) Current uses of open phenol block for adult acquired spasticity. Clin. Orthop., 165: 217-222. Garland, DE, Lilling, M and Keenan, MA (1984) Percutaneous phenol blocks to motor points of spastic forearm muscles in head-injured adults. Arch. Phys. Med. Rehabil., 65: 243-245. Gasser, HS and Erlanger, J (1929) The role of fiber size in the establishment of a nerve block by pressure or cocaine. Am. J. Physiol., 88: 581-591. Geenen, C, Consky, E and Ashby, P (1996) Localizing muscles for botulinum toxin treatment of focal hand dystonia. Can. J. Neurol. Sci., 23: 194-197. Gelb, DJ, Yoshimura, DM, Olney, RK, Lowenstein, DH and Arninoff, MJ (1991) Change in pattern of muscle activity following botulinum toxin injections for torticollis. Ann. Neurol., 29: 370-376. George, EF, Zimbler, M, Wu, BL, Biller, HF and Sanders, I (1992) Quantitative mapping of the effect of botulinum toxin injections in the thyroarytenoid muscle. Ann. 0101. Rhinol. Laryngol., 101: 888-892. Gerner, P, Nakamura, T, Quan, CF, Anthony, DC and Wang, GK (2000) Spinal tonicaine: potency and differential blockade of sensory and motor functions. Anesthesiology, 92: 1350-1360.

J-M. ORACIES AND D.M. SIMPSON

Ghoshal, UC, Aggarwal, R, Kumar, S and Naik, SR (2002) Pneumatic dilation vs. intrasphincteric botulinum toxin injection in the treatment of achalasia cardia in India: an economic analysis. Ind. J. Gastroenterol., 21: 193-196. Giladi, N and Honigman, S (1997) Botulinum toxin injections to one leg alleviate freezing of gait in a patient with Parkinson's disease. Mov. Disord., 12: 1085-1086. Giladi, N, Gurevich, T, Shabtai, H, Paleacu, D and Simon, ES (2001) The effect of botulinum toxin injections to the calf muscles on freezing of gait in parkinsonism: a pilot study. J. Neurol., 248: 572-576. Girlanda, P, Quartarone, A, Sinicropi, S, Nicolosi, C and Messina, C (1996) Unilateral injection of botulinum toxin in blepharospasm: single fiber electromyography and blink reflex study. Mov. Disord., 11: 27-31. Girlanda, P, Vita, G, Nicolosi, C, Milone, S and Messina, C (1992) Botulinum toxin therapy: distant effects on neuromuscular transmission and autonomic nervous system. J. Neurol. Neurosurg. Psychiatry, 55: 844845. Glanzman, RL, Gelb, DJ, Drury, I, Bromberg, MB and Truong, DD (1990) Brachial plexopathy after botulinum toxin injections. Neurology, 40: 1143. Glass, A, Cain, HD, Liebgold, H and Mead, S (1968) Electromyographic and evoked potential responses after phenol blocks of peripheral nerves. Arch. Phys. Med. Rehabil., 49: 455--459. Glenn, MB (1990) Nerve blocks. In: Glenn and Whyte (Eds.), The Practical Management of Spasticity in Children and Adults. Lea and Febiger, PhiladelphiaLondon,pp.227-258. Gomez-Ramirez, AM, Villegas-Perez, MP, Miralles de Imperial, J, Salvador-Silva, M and Vidal-Sanz, M (1999) Effects of intramuscular injection of botulinum toxin and doxorubicin on the survival of abducens motoneurons. Invest. Ophthalmol. Vis. Sci., 40: 414424. Gordon, A (1914) Experimental study of intraneural injections of alcohol. J. Nerv. Ment. Dis., 41: 81-95. Goschel, H, Wohlfarth, K, Frevert, J, Dengler, R and Bigalke, H (1997) Botulinum A toxin therapy: neutralizing and nonneutralizing antibodies - therapeutic consequences. Exp. Neurol., 147: 96-102. Gracies, JM (2001) Pathophysiology of impairment in spasticity, and use of stretch as a treatment of spastic hypertonia. Phys. Med. Rehabil. Clin. N. Am., 12: 747-768. Gracies, JM and Simpson, D (1999) Neuromuscular blockers. Phys. Med. Rehabil. Clin. N. Am., 10: 357-383, viii. Gracies, JM and Simpson, D (2000) Therapy with Botulinum toxin. The Neurologist, 6: 98-115.

FOCAL INJECTION THERAPY

Grades, JM, Marosszeky, JE, Renton, R, Sandanam, J, Gandevia, SC and Burke, D (2000) Short-term effects of dynamic Lycra splints on upper limb in hemiplegic patients. Arch. Phys. Med. Rehabil., 81: 1547-1555. Gracies, 1M, Weisz, DJ, Yang, BY, Flanagan, S and Simpson D (2001) Evidence for increased antagonist strength and movement speed following botulinum toxin injections in spasticity. Neurology, 56: A3. Grades, JM, Elovic, E, McGuire, J and Simpson, D (2002a) Traditional pharmacologic treatments of spasticity - Part 1. Local Treatments. In: NH Mayer and DM Simpson (Eds.), Spasticity: Etiology, Evaluation, and the Role of Botulinum toxin. New York: WE MOVE, pp. 44-64. Grades, JM, Weisz, DJ, Yang, BY, Flanagan, Sand Simpson, D (2002b) Spastic cocontraction and movement speed: effects of botulinum toxin type A (BTX-A) injection into an agonist. Ann. Neurol., 52 (Suppl. 1): S89. Gracies, JM, Elovic, E, McGuire, J, Nance, P and Simpson, D (2002c) Traditional pharmacologic treatments of spasticity - Part II Systemic Treatments. In: NH Mayer and DM Simpson (Eds.), Spasticity: Etiology, Evaluation, Management and the Role of Botulinum toxin. New York: WE MOVE, pp. 65-93. Grades, JM, Weisz, DJ, Yang, BY, Flanagan, S and Simpson, D (2002d) Impact of botulinum toxin type A (BTX-A) dilution and endplate targeting technique in upper limb spasticity. Ann. Neurol., 52(3) (Suppl. 1): S87. Greene, P, Fahn, S and Diamond, B (1994) Development of resistance to botulinum toxin type A in patients with torticollis. Mov. Disord., 9: 213-217. Greenstein, D, Brown, TF and Kester, RC (1994) Assessment of chemical lumbar sympathectomy in critical limb ischemia using thermal imaging. Int. J. Clin. Monit. Comput., 11: 31-34. Greitz, T, Hannerz, J and Lindblom, U (1972) Treatment of spasticity with intrathecal block with phenol under myelographic controls. Acta. Neurol. Scand., 51: 437438. Griffith, ER and Melampy, CN (1977) General anesthesia used in phenol intramuscular neurolysis in young children with spasticity. Arch. Phys. Med. Rehabil., 58: 154-157. Gundel, H, Wolf, A, Xidara, V, Busch, Rand CeballosBaumann, AO (2001) Social phobia in spasmodic torticollis. J. Neurol. Neurosurg. Psychiatry, 71: 499-504. Comment in: J. Neurol. Neurosurg. Psychiatry (2002), 73: 461, discussion 461-462. Gunduz, S, Kalyon, TA, Dursun, H, Mohur, H and Bilgic, F (1992) Peripheral nerve block with phenol to treat spasticity in spinal cord injured patients. Paraplegia, 30: 808-811.

683 Guyton, AC and McDonald, MA (1947) Physiology of Botulinus toxin. Arch. Neurol. Psychiatry (Chicago), 57: 578-592. Hagenah, R, Benecke, R and Wiegand, H (1977) Effects of type A botulinum toxin on the cholinergic transmission at spinal Renshaw cells and on the inhibitory action at Ia inhibitory intemeurons. Naunyn. Schrniedebergs Arch. Pharmacol., 299: 267-272. Hagenah, R (1979) Neuropharmacologic and neurotoxic studies on the Renshaw mechanism. Fortschr. Med., 97: 1840-1844. Hallan, RI, Williams, NS, Melling, J, Waldron, DJ, Womack, NR and Morrison, IF (1988) Treatment of anismus in intractable constipation with botulinum A toxin. Lancet, 2: 714-717. Hallett, M, Glocker, FX and Deuschl, G (1994) Mechanism of action of botulinum toxin. Ann. Neurol., 36: 449-450. Halpern, D (1977) Histologic studies in animals after intramuscular neurolysis with phenol. Arch. Phys. Med. Rehabil., 58: 438-443. Halpern, D and Meelhuysen, FE (1966) Phenol motor point block in the management of muscular hypertonia. Arch. Phys. Med. Rehab., 47: 659-664. Hambleton, P, Cohen, HE, Palmer, BJ and Melling, J (1992) Antitoxins and botulinum toxin treatment. Brit. Med. J., 304: 959-960. Hamjian, JA and Walker, FO (1994) Serial neurophysiological studies of intramuscular botulinum-A toxin in humans. Muscle Nerve, 17: 1385-1392. Hanna, PA and Jankovic, J (1998) Mouse bioassay vs. Western blot assay for botulinum toxin antibodies: correlation with clinical response. Neurology, 50: 1624-1629. Hanna, PA, Jankovic, J and Vincent, A (1999) Comparison of mouse bioassay and immunoprecipitation assay for botulinum toxin antibodies. J. Neurol. Neurosurg. Psychiatry, 66: 612-616. Hariga, J, Tardieu, G, Tardieu, C and Gagnard, L (1964) Effets de l' application de l' alcool dilue sur Ie nerf. Confrontation de l'etude dynamographique et de l' etude histologique chez Ie chat decerebre, Revue Neurol. (Paris), 111: 472-474. Hariga, J (1966) Influences sur la motricite de la suppression des effecteurs gamma par alcoolisation des nerfs peripheriques. Acta. Neurol. Belg., 66: 607-611. Harris, CP, Alderson, K, Nebeker, J, Holds, JB and Anderson, RL (1991) Histologic features of human orbicularis oculi treated with botulinum A toxin. Arch. Ophthalrnol., 109: 393-395. Harrison, LM, Morrison, IE and Fennessey, PV (1991) Microtechnique for quantifying phenol in plasma by gas chromatography-mass spectrometry. Clin. Chern., 37: 1739-1742.

684 Harvey, AM (1939) The actions of procaine on neuromuscular transmission. Bull. Johns Hopkins Hosp., 65: 223-238. Hasegawa, 0, Nagatomo, H and Suzuki, Y (1990) Local alcoholisation treatment of spasmodic torticollis. Rinsho Shinkeigaku, 30: 718-722. Hassan, SM, Jennekens, FGI, Wienecke, G and Veldman, H (1994) Elimination of superfluous neuromuscular junctions in rat calf muscles recovering from botulinum toxin-induced paralysis. Muscle Nerve, 17: 623631. Hassan, SM, Jennekens, FGI and Veldman, H (1995) Botulinum toxin-induced myopathy in the rat. Brain, 118: 533-545. Hatheway, CL (1989) Bacterial sources of Clostridial neurotoxins. In: LL Simpson (Ed.), Botulinum Neurotoxin and Tetanus Toxin. San Diego, Academic Press Inc., pp. 3-24. Hatheway, CH and Dang, C (1994) Immunogenicity of the neurotoxins of Clostridium Botulinum. In: J Jankovic and M Hallett (Eds.), Therapy with Botulinum Toxin. New York: Marcel Dekker, pp. 93-107. Hecht, JS (1992) Subscapular nerve block in the painful hemiplegic shoulder. Arch. Phys. Med. Rehabil., 73: 1036--1039. Heinen, F, Mall, V, Ruckauer, KD, Greiner, P, Sontheimer, J, Brandis, M and Korinthenberg, R (1999) Lack of response to botulinum toxin A in patients with hypertrophic pyloric stenosis. Eur. J. Pediatr., 158: 436. Helweg-Larsen, J and Jacobsen, E (1969) Treatment of spasticity in cerebral palsy by means of phenol nerve block of peripheral nerves. Dan. Med. Bull., 16: 2025. Henderson, JM, Ghika, JA, Van Melle, G, Haller, E and Einstein, R (1996) Botulinum toxin A in non-dystonic tremors. Eur. Neurol., 36: 29-35. Herrero, BA, Ecklung, AE, Streett, CS, Ford, DF and King, JK (1967) Experimental botulism in monkeys a clinical pathological study. Exp. Mol. Pathol., 6: 84-95. Hertegard, S, Granqvist, Sand Lindestad, PA (2000) Botulinum toxin injections for essential voice tremor. Ann. Otol. Rhinol. Laryngol., 109: 204-209. Hetherington, NJ and Dooley, MJ (2000) Potential for patient harm from intrathecal administration of preserved solutions. Med. J. Aust., 173: 141-143. Hesse, S, Friedrich, H, Domasch, C and Mauritz, KH (1992) Botulinum toxin therapy for upper limb flexor spasticity: preliminary results. J. Rehab. Sci., 5: 98101. Hesse, S, Lucke, D, Malezic, M, Bertelt, C, Friedrich, H, Gregoric, M and Mauritz, KH (1994) Botulinum toxin treatment for lower limb extensor spasticity in chronic

J-M. GRACIES AND D.M. SIMPSON

hemiparetic patients. J. Neurol. Neurosurg. Psychiatry, 57: 1321-1324. Hesse, S, Jahnke, MT, Luecke, D and Mauritz, KH (1995) Short-term electrical stimulation enhances the effectiveness of botulinum toxin in the treatment of lower limb spasticity in hemiparetic patients. Neurosci. Lett., 201: 37-40. Hesse, S, Krajnik, J, Luecke, PT, Jahnke, MT, Gregoric, M and Mauritz, KH (1996) Ankle muscle activity before and after botulinum toxin therapy for lower limb extensor spasticity in chronic hemiparetic patients. Stroke, 27: 455-460. Hesse, S, Reiter, F, Konrad, M and Jahnke, MT (1998) Botulinum toxin type A and short-term electrical stimulation in the treatment of upper limb flexor spasticity after stroke: a randomized, double-blind, placebo-controlled trial. Clin. Rehabil., 12: 381-388. Hesse, S, Werner, C, Bardeleben, A and Brandl-Hesse, B (2002) Management of upper and lower limb spasticity in neuro-rehabilitation, Acta. Neurochir., Suppl., 79: 117-122. Hibbs, RG, Weber, IT, Corwin, A, AlIos, BM, Abd el Rehim, MS, Sharkawy, SE, Sam, JE and McKee, KT, Jr (1996) Experience with the use of an investigational F(ab')2 heptavalent botulism immune globulin of equine origin during an outbreak of type E botulism in Egypt. Clin. Infect. Dis., 23: 337-340. Hogan, QH, Stadnicka, A, Stekiel, TA, Bosnjak, ZJ and Kampine, JP (1993) Effects of epidural and systemic lidocaine on sympathetic activity and mesenteric circulation in rabbits. Anesthesiology, 79: 1250-1260. Holland, AJC and Youssef, M (1978) A complication of subarachnoid phenol blockade. Anesthesia, 34: 260262. Hong, CZ, Long, HA, Kanakamedala, RV,Chang, YM and Yates, L (1996) Splinting and local steroid injection for the treatment of ulnar neuropathy at the elbow: clinical and electrophysiological evaluation. Arch. Phys. Med. Rehabil., 77: 573-577. Huang, JH, Thalhammer, JG, Raymond, SA and Strichartz, GR (1997) Susceptibility to lidocaine of impulses in different somatosensory afferent fibers of rat sciatic nerve. J. Phannacol. Exp. Ther., 282: 802-811. Hughes, R and Whaler, BC (1962) Influence of nerve ending activity and of drugs on the rate of paralysis of rat diaphragm preparations by Clostridium botulinum type A toxin. J. Physiol., 160: 221-223. Hughes, JT (1970) Thrombosis of the posterior spinal arteries. A complication of an intrathecal injection of phenol. Neurology, 20: 659-664. Hultborn, H and Nielsen, JB (1995) H-reflexes and Fresponses are not equally sensitive to changes in

FOCAL INJECTION THERAPY

motoneuronal excitability. Muscle Nerve, 18: 1471-1474. Hultborn, H and Nielsen, JB (1996) Comments: methodological problems of comparing F responses and H-reflexes. Muscle Nerve, 19: 1347-1348. Humeau, Y, Doussau, F, Grant, NJ and Poulain, B (2000) How botulinum and tetanus neurotoxins block neurotransmitter release. Biochimie, 82: 427-446. Hurvitz, EA, Conti, GE, Flansburg, EL and Brown, SH (2000) Motor control testing of upper limb function after botulinum toxin injection: a case study. Arch. Phys. Med. Rehabil., 81: 1408-1415. Hyman, N, Barnes, M, Bhakta, B, Cozens, A, Bakheit, M, Kreczy-Kleedorfer, B, Poewe, W, Wissel, J, Bain, P, Glickman, S, Sayer, A, Richardson, A and Dott, C (2000) Botulinum toxin (Dysport) treatment of hip adductor spasticity in multiple sclerosis: a prospective, randomized, double blind, placebo controlled, dose ranging study. J. Neurol. Neurosurg. Psychiatry, 68: 707-712. Itoh, M (1995) The role of brain acetylcholine in phenol-induced tremor in mice. Arch. Oral. Biol., 40: 365-372. Itoh, S and Noda, K (1998) An e1ectrophysiological and histological study on the neurotoxicity of lidocaine in excized rabbit cervical vagus nerve. Masui, 47: 806-816. Iwatsubo, E, Okada, E, Takehara, T, Tamada, K and Akatsu, T (1994) Selective intrathecal phenol block to improve activities of daily living in patients with spastic quadriplegia. A preliminary report. Paraplegia, 32: 489-492. Jankovic, J, Schwartz, K, Clemence, W, Aswad, A and Mordaunt, J (1996) A randomized, double-blind, placebo-controlled study to evaluate botulinum toxin type A in essential hand tremor. Mov. Disord., 11: 250-256. Jarrett, L, Nandi, P and Thompson, AJ (2002) Managing severe lower limb spasticity in multiple sclerosis: does intrathecal phenol have a role? J. Neurol. Neurosurg. Psychiatry, 73: 705-709. Jayasooriya, SM, Francisco, G and Healy, W (2002) Early experience with the use of botulinum toxin type B for the treatment of upper limb spastic hypertonia. Arch. Phys. Med. Rehabil., 83: 1677. Jones, RV, Jr, Lambert, EH and Sayre, GP (1955) Source of type of insertion activity in electromyography with evaluation of histologic method of localization. Arch. Phys. Med., 36: 301-310. Jongerius, PH, Joosten, F, Hoogen, FJ, Gabreels, FJ and Rotteveel, 11 (2003) The treatment of drooling by ultrasound-guided intraglandular injections of botulinum toxin type A into the salivary glands. Laryngoscope, 113: 107-111.

685 Jost, WH (1997) One hundred cases of anal fissure treated with botulin toxin: early and long-term results. Dis. Colon Rectum, 40: 1029-1032. Jost, WH (2001) Botulinum toxin type B in the treatment of anal fissures: first preliminary results. Dis. Colon Rectum, 44: 1721-1722. Kaji, R, Rothwell, JC, Katayama, M, Ikeda, T, Kubori, T, Kohara, N, Mazaki, T, Shibasaki, H and Kimura, J (1995) Tonic vibration reflex and muscle afferent block in writer's cramp. Ann. Neurol., 38: 155-162. Kalichman, MW and Calcutt, NA (1992) Local anestheticinduced conduction block and nerve fiber injury in streptozotocin-diabetic rats. Anesthesiology, 77: 941947. Kalichmann, MW, Moorhouse, DF, Powell, HC and Myers, RR (1993) Relative neural toxicity of local anesthetics. J. Neuropathol. Exp. Neurol., 52: 234240. Kalra, HK and Magoon, EH (1990) Side effects of the use of botulinum toxin for treatment of benign essential blepharospasm and hemifacial spasm. Ophthalmic Surg., 21: 335-338. Kanai, Y, Katsuki, Hand Takasaki, M (2000) Lidocaine disrupts axonal membrane of rat sciatic nerve in vitro. Anesth. Analg., 91: 944-948. Kao, I, Drachman, DB and Price, DL (1976) Botulinum toxin: mechanism of presynaptic blockade. Science, 193: 1256-1258. Karatassas, A, Morris, RG and Slavotinek, AH (1993) The relationship between regional blood flow and absorption of lignocaine. Aust. N2. J. Surg., 63: 766-771. Katz, J, Knott, LK and Feldman, OJ (1966) Brachial plexus infiltration with dilute phenol solution in the management of upper extremity spasticity. Anesthesiology, 27: 87-88. Katz, J, Knott, LW and Feldman, DJ (1967) Peripheral nerve injections with phenol in the management of spastic patients. Arch. Phys. Med. Rehabil., 48: 97-99. Katz, JA, Sehlhorst, S and Blisard, KS (1995) Histopathological changes in primate spinal cord after single and repeated epidural phenol administration. Reg. Anesth., 20: 283-290. Kauffman, JA, Way, JF, Jr, Siegel, LS and Sellin, LC (1985) Comparison of the action of types A and F botulinum toxin at the rat neuromuscular junction. Toxicol. Appl. Pharmacol., 79: 211-217. Kayikcioglu, A, Erk, Y, Mavili, E, Vargel, I and Ozgur, F (2003) Botulinum toxin in the treatment of zygomatic fractures. Plast. Reconstr. Surg., 111: 341-346. Keen, M, Blitzer, A, Aviv, J, Binder, W, Prystowsky, J, Smith, Hand Brin, M (1994) Botulinum toxin A for hyperkinetic facial lines: results of a double-blind, placebo-controlled study. Plast. Reconstr. Surg., 94: 94-99.

686 Keenan, MA (1988) Management of the spastic upper extremity in the neurologically impaired adult. Clin. Orthop., 233: 116-125. Keenan, MA, Todderud, EP, Henderson, Rand Botte, M (1987) Management of intrinsic spasticity in the hand with phenol injection or neurectomy of the motor branch of the ulnar nerve. J. Hand. Surg. (Am), 12: 734-739. Keenan, MA, Tomas, ES, Stone, L and Gersten, LM (1990) Percutaneous phenol block of the musculocutaneous nerve to control elbow flexor spasticity. J. Hand. Surg. (Am), 15: 340-346. Kelly, RE and Gauthier-Smith, PC (1959) Intrathecal phenol in the treatment of reflex spasms and spasticity. Lancet, ii: 1102-1105. Kenyon, EM, Seeley, ME, Janszen, D and Medinsky, MA (1995) Dose-, route- and sex-dependent urinary excretion of phenol metabolites in B6C3Fl mice. J. Toxicol. Environ. Health, 44: 219-233. Kerty, E and Stien, R (1997) Treatment of spasticity with botulinum toxin. Tidsskr Nor Laegeforen, 117: 20222024. Kessler, KR and Benecke, R (1997) The EBD test - a clinical test for the detection of antibodies to botulinum toxin type A. Mov. Disord., 12: 95-99. Khalili, AA (1984) Physiatric management of spasticity by phenol nerve and motor point block. In: AP Ruskin (Ed.), Current Therapy in Physiatry. Philadelphia, WB Saunders. Khalili, AA and Benton, JG (1966) A physiological approach to the evaluation and the management of spasticity with procaine and phenol nerve block. Clin. Orthop., 47: 97-104. Khalili, AA, Harmel, MH, Forster, S and Benton, JG (1964) Management of spasticity by selective peripheral nerve block with dilute phenol solutions in clinical rehabilitation. Arch. Phys. Med. Rehab., 45: 513-519. Khao, ShY, Dan, ChV and Dzhi, ChV (1990) Treament of Wolff Parkinson White syndrome by local myocardial injection of absolute alcohol. Grud-Serdechnososudistaia Khir, 4: 63-64. Khodorova, A, Meissner, K, Leeson, S and Strichartz, GR (2001) Lidocaine selectively blocks abnormal impulses arising from noninactivating Na channels. Muscle Nerve, 24: 634-647. Kiernan, MC, Hales, JP, Gracies, JM, Mogyoros, I and Burke, D (1997) Paraesthesiae induced by prolonged high frequency stimulation of human cutaneous afferents. J. Physiol. (Lond.), 501: 461-471. Kim, HJ, Seo, K, Yum, KW, Oh, YS, Yoon, TG and Yoon, SM (2002) Effects of botulinum toxin type A on the superior cervical ganglia in rabbits. Auton. Neurosci., 102: 8-12.

J-M. GRACIES AND D.M. SIMPSON

Kirazli, Y, On, AY, Kismali, Band Aksit, R (1998) Comparison of phenol block and botulinus toxin type A in the treatment of spastic foot after stroke: a randomized, double-blind trial. Am. J. Phys. Med. Rehabil., 77: 510-515. Kiwerski, J (1984) New possibilities of improving the function of the hand of patients with spastic hemiplegia. Int. J. Rehabil. Res., 7: 293-298. Kjellberg, RN et al. (1961) Gait improvement in parkinsonian patients by gamma motor neuron suppression. Trans. Am. Neural. Assoc., 86: 126-130. Koman, LA, Mooney, JF, III, Smith, BP, Goodman, A and Mulvaney, T (1993) Management of cerebral palsy with botulinum-A toxin: preliminary investigation. J. Ped. Orthop., 13: 489-495. Koman, LA, Mooney, JF, III, Smith, BP, Goodman, A and Mulvaney, T (1994) Management of spasticity in cerebral palsy with botulinum-A toxin: report of a preliminary, randomized, double-blind trial. J. Ped. Orthop., 14: 299-303. Koman, LA, Mooney, JF and Patersen Smith, B (1996) Neuromuscular blockade in the management of cerebral palsy. J. Child Neurol., II: S23-S28. Kong, KH and Chua, KS (1999a) Neurolysis of the musculocutaneous nerve with alcohol to treat poststroke elbow flexor spasticity. Arch. Phys. Med. Rehabil.,80: 1234-1236. Kong, KH and Chua, KS (1999b) Outcome of obturator nerve block with alcohol for the treatment of hip adductor spasticity. Int. J. Rehabil. Res., 22: 327-329. Kong, KH and Chua, KS (2002) Intramuscular neurolysis with alcohol to treat post-stroke finger flexor spasticity. Clin. Rehabil., 16: 378-381. Konstanzer, A, Ceballos-Baumann, AD, Dressnandt, J and Conrad, B (1993) Lokale Injektionsbehandlung mit Botulinum-Toxin A bei schwerer Arm- und Beinspastik. Nervenarzi, 64: 517-523. Koyama, H, Murakami, K, Suzuki, T and Suzaki, K (1992) Phenol block for hip flexor muscle spasticity under ultrasonic monitoring. Arch. Phys. Med. Rehabil., 73: 1040-1043. Krack, P and Marion, MH ( 1994) "Apraxia of lid opening," a focal eyelid dystonia: clinical study of 32 patients. Mov. Disord., 9: 610-615. Kuffler, SW, Hunt, CC and Quilliam, JP (1951) Function of medullated small-nerve fibers in mammalian ventral roots: efferent muscle spindle innervation. J. Neurophysiol., 14: 29-54. Labat, G (1933) The action of alcohol on the living nerve: experimental and clinical considerations. Anesth. Analg. (Curr. Res.), 12: 190-196. Labib, KB and Gans, BM (1984) Chemical neurolysis: technique and anatomical consideration (videotape).

FOCAL INJECTION THERAPY

Boston, Dept of Rehabilitation Medicine, New England Medical Center. Lagalla, G, Danni, M, Reiter, F, Ceravolo, MG and Provinciali, L (2000) Post-stroke spasticity management with repeated botulinum toxin injections in the upper limb. Am. l. Phys. Med. Rehabil., 79: 377-384, quiz 391-394. Lange, DJ, Brin, MF, Warner, CL, Fahn, S and Lovelace, RE (1987) Distant effects of local injection of botulinum toxin. Muscle Nerve, 10: 552-555. Erratum in: Muscle Nerve (1988), II: 520. Lange, DJ, Rubin, M, Greene, PE, Kand, UJ, Moskowitz, CB, Brin, MF, Lovelace, RE and Fahn, S (1991) Distant effects of locally injected botulinum toxin: a double-blind study of single fiber EMG changes. Muscle Nerve, 41: 672--675. Comment in: Muscle Nerve (1993), 16: 677. Langer, JC and Birnbaum, E (1997) Preliminary experience with intrasphincteric botulinum toxin for persistent constipation after pull-through for Hirschsprung's disease. J. Pediatr. Surg., 32: 1059-1061, discussion 1061-1062. Lapicque, L (1931) Has the muscular substance a longer chronaxie than the nervous substance? l. Physiol. (Lond.), 73: 189-214. Laskawi, R, Drobik, C and Schonebeck, C (1998) Up-todate report of botulinum toxin type A treatment in patients with gustatory sweating (Frey's syndrome). Laryngoscope, 108: 381-384. Lee, RE, Tartell, PB, Karmody, CS and Hunter, DD (1999) Association of adhesive macromolecules with terminal sprouts at the neuromuscular junction after botulinum treatment. Otolaryngol. Head Neck Surg., 120: 255261. Lennerstrand, G, Nordbo, GA, Tian, S, Eriksson-Derouet, B and Ali, T (1998) Treatment of strabismus and nystagmus with botulinum toxin type A. An evaluation of effects and complications. Acta. Ophthalmol. Scand., 76: 27. Lepore, V, Defazio, G, Acquistapace, D, Melpignano, C, Pomes, L, Lamberti, P, Livrea, P and Ferrari, E (1995) Botulinum A toxin for the so-called apraxia of lid opening. Mov. Disord., 10: 525-526. Lew, MF, Adornato, BT, Duane, DD, Dykstra, DD, Factor, SA, Massey, JM et al. (1997) Botulinum toxin type B (BotB): a double-blind placebo-controlled, safety and efficacy study in cervical dystonia. Neurology, 49: 701-707. Lew, MF, Brashear, A and Factor, S (2000) The safety and efficacy of botulinum toxin type B in the treatment of patients with cervical dystonia: summary of three controlled clinical trials. Neurology, 55: S29-35.

687 LeWitt, PA and Trosch, RM (1997) Idiosyncratic adverse reactions to intramuscular botulinum toxin type A injection. Mov. Disord., 12: 1064-1067. Liddell, EGT and Sherrington, CS (1924) Reflexes in response to stretch (myotatic reflexes). Proc. Royal Soc. (Series B), 96: 212-242. Liljestrand, G and Magnus, R (1919) Uber die Wirkung des Novocaine auf den normalen und den tetanusstarren Skelettmuskel und iiber die Entstehung der Muskelstaue beim Wundstarrkampf. Pfiiigers Arch. Ges. Physiol., 176: 168-208. Limousin, P, Memin, B and Pollak, P (1997) Treatment of dystonia occurring in parkinsonian syndromes by botulinum toxin. Eur. Neurol., 37: 66-67. Lin, JH and Rydqvist, B (1999) The mechanotransduction of the crayfish stretch receptor neuron can be differentially activated or inactivated by local anaesthetics. Acta. Physiol. Scand., 166: 65-74. Loubser, PG (1990) Intrathecal alcohol injection guided by electrical localization of spinal roots. Anesth. Analg., 70: 119-121. Comment on: Anesth. Analg. (1989), 68: 527-529. Loubser, PG (1995) Epidural phenol administration for iliopsoas hypertonicity. Anesth. Analg., 80: 639-640. Lu, L, Atchabahian, A, Mackinnon, SE and Hunter, DA (1998) Nerve injection injury with botulinum toxin. Plast. Reconstr. Surg., 101: 1875-1880. Macek, C (1983) Medical news: Venous thrombosis results from some phenol injections. lAMA, 249: 1807. Madalinski, MH, Slawek, J, Duzynski, W, Zbytek, B, Jagiello, K, Adrich, Z and Kryszewski, A (2002) Side effects of botulinum toxin injection for benign anal disorders. Eur. J. Gastroenterol. Hepatol., 14: 853-856. Maher, RM (1955) Relief of pain in incurable cancer. Lancet, 268: 18-22. Mallart, A, Molgo, J, Angaut-Petit, D and Thesleff, S (1989) Is the internal calcium regulation altered in type A botulinum toxin-poisoned motor endings? Brain Res.,479: 167-171. Manabe, Y, Saito, T and Saito, H (1998) Effects of lidocaine on salicylate-induced discharge of auditory neurons in the inferior colliculus of the guinea pig. Nippon libiinkoka Gakkai Kaiho, 101: 807-813. Manni, E, Bagolini, B, Pettorossi, VE and Errico, P (1989) Effect of Botulinum Toxin on extraocular muscle proprioception. Documenta Ophtalmologica, 72: 189198. Maria, G, Cassetta, E, Gui, D, Brisinda, G, Bentivoglio, AR and Albanese, A (1998) A comparison of botulinum toxin and saline for the treatment of chronic anal fissure. N. Eng. l. Med., 338: 217-220.

688 Marion, MH (1999) Treatment of dystonias. Presse Med., 28: 312-315. Martin, C, Thomachot, L and Albanese, J (1994) Clinical pharmacokinetics of cefotetan. Clin. Pharmacokinet., 26: 248-258. Maselli, RA, Burnett, ME and Tonsgard, JH (1992) In vitro microelectrode study of neuromuscular transmission in a case of botulism. Muscle Nerve, IS: 273276. Massey, JM (1995) Treatment of spasmodic torticollis with intramuscular phenol injection (letter). J. Neurol. Neurosurg. Psychiatry, 58: 258-259. Matthews, PBC and Rushworth, G (1957) The relative sensitivity of muscle nerve fibers to procaine. J. Physiol. (Lond.), 135: 263-269. Mauriello, JA and Aljian, J (1991) Natural history of treatment of facial dyskinesias with botulinum toxin: a study of 50 consecutive patients over seven years. Br. J. Ophthalmol., 75: 737-739. May, 0 (1912) The functional and histological effects of intraneural and intraganglionic injections of alcohol. Br. Med. J. (Aug 31): 465--470. Mayer, NH, Esquenazi, A and Keenan, MA (2001) Patterns of upper motoneuron dysfunction in the lower limb. Adv. Neurol., 87: 311-319. McDonald, TA, Holland, NT, Skibola, C, Duramad, P and Smith, MT (2001) Hypothesis: phenol and hydroquinone derived mainly from diet and gastrointestinal flora activity are causal factors in leukemia. Leukemia, IS: 10-20. Mclellan, K, Das, RE, Ekong, TA and Sesardic, D (1996) Therapeutic botulinum type A toxin: factors affecting potency. Toxicon, 34: 975-985. Meelhuysen, FE, Halpern, D and Quast, J (1968) Treatment of flexor spasticity of hip by paravertebral lumbar spinal nerve block. Arch. Phys. Med. Rehabil., 49: 36--41. Meyer, KE, Caputo, FA, Gasper, CA et al. (1999) A comparative systemic toxicity study of Elan's botulinum toxin type B study, evaluated under the name Neurobloc®, in adult and juvenile cynomolgus monkeys. International Conference 1999; Basic therapeutic aspects of botulinus and tetanus toxins (Nov 15-18). Orlando, FL. Mezaki, T, Kaji, R, Kohara, N, Fujii, H, Katayama, M, Shimizu, T, Kimura, J and Brin, MF (1995) Comparison of therapeutic efficacies of type A and F botulinum toxins for blepharospasm: a double-blind, controlled study. Neurology, 45: 506-508. Mezaki, T, Kaji, R, Brin, MF, Hirota-Katayama, M, Kubori, T, Shimizu, T and Kimura, J (1999) Combined use of type A and F botulinum toxins for blepharospasm: a double-blind controlled trial. Mov. Disord., 14: 1017-1020.

J-M. GRACIES AND D.M. SIMPSON

Mezaki, T, Matsumoto, S, Sakamoto, T, Mizutani, K and Kaji, R (2000) (Cervical echomyography in cervical dystonia and its application to the monitoring for muscle afferent block (MAB». Rinsho Shinkeigaku, 40: 689-693. Mikhailov, VV (1958) Electrophysiological analysis of disturbances in reflex activity of the spinal cord in experimental botulism. Bull. Exp. BioI. Med., 46: 1188-1192. Mikhailov, VV and Barashkov, GN (1977) Mechanism of disorders of inhibition of electrogenesis in spinal alpha-motor neurons in experimental local botulin poisoning. Bull. Exp. BioI. Med., 83: 651-654. Minguez, M, Herreros, B, Espi, A, Garcia-Granerom, E, Sanchiz, V, Mora, F, Lledo, Sand Benages, A (2002) Long-term follow-up (42 months) of chronic anal fissure after healing with botulinum toxin. Gastroenterology, 123: 1l2-117. Minkes, RK and Langer, JC (2000) A prospective study of botulinum toxin for internal anal sphincter hypertonicity in children with Hirschsprung's disease. J. Pediatr. Surg., 35: 1733-1736. Moberg-Wolff, E and Walke, K (2002) Use of botulinum toxin type B in the pediatric population. Arch. Phys. Med. Rehabil., 83: 1673. Mochida, S, Poulain, B, Eisel, U, Binz, T, Kurazono, H, Niemann, H and Tauc, L (1990) Exogenous mRNA encoding tetanus or botulinum neurotoxins expressed in Aplysia neurons. Proc. Natl. Acad. Sci. USA, 87: 7844-7848 Modugno, N, Priori, A, Berardelli, A, Vacca, L, Mercuri, B and Manfredi, M (1998) Botulinum toxin restores presynaptic inhibition of group Ia afferents in patients with essential tremor. Muscle Nerve, 21: 1701-1705. Molgo, J, Siegel, LS, Tabti, N and Thesleff, S (1989) A study of synchronization of quantal transmitter release from mammalian motor endings by the use of botulinal toxins type A and D. J. Physiol. (Lond.), 411: 195-205. Montecucco, C and Schiavo, G (1995) Structure and function of tetanus and botulinum neurotoxins. Q. Rev. Biophys., 28: 423--472. Mooney, V, Frykman, G and McLamb, J (1969) Current status of intraneural phenol injections. Clin. Orthop., 63: 122-131. Moore, TJ and Anderson, RB (1991) The use of open phenol blocks to the motor branches of the tibial nerve in adult acquired spasticity. Foot Ankle, ll: 219-221. Morgan, RJ and Steller, PH (1994) Acute paraplegia following intrathecal phenol block in the presence of occult epidural malignancy. Anesthesia, 49: 142-144. Morris, JL, Jobling, P and Gibbins, IL (2002) Botulinum neurotoxin A attenuates release of norepinephrine but

FOCAL INJECTION THERAPY

not NPY from vasoconstrictor neurons. Am. J. Physiol. Heart Circ. Physiol., 283: H2627-2635. Morrison, JE, Jr., Hertzberg, DL, Gourley, SM and Matthews, DJ (1989) Motor point blocks in children. A technique to relieve spasticity using phenol injections. AORN J., 49: 1346-1347, 1349-1351, 1354. Review. Morrison, JE, Jr, Matthews, D, Washington, R, Fennessey, PV and Harrison, LM (1991) Phenol motor point blocks in children: plasma concentrations and cardiac dysrhythmias. Anesthesiology, 75: 359-362. Muehldorfer, SM, Hahn, EG and Ell, C (1997) Botulinum toxin injection as a diagnostic tool for verification of sphincter of Oddi dysfunction causing recurrent pancreatitis. Endoscopy, 29: 120-124. Munchau, A, Filipovic, SR, Oester-Barkey, A, Quinn, NP, Rothwell, JC and Bhatia, KP (2001) Spontaneously changing muscular activation pattern in patients with cervical dystonia. Mov. Disord., 16: 1091-1097. Murdoch-Ritchie, J and Greene, NM (1980) Local anesthetics. In: LS Goodman & A Gilman (Eds.), The Pharmacological Basis of Therapeutics (6th ed.). Macmillan, pp. 300-320. Murdoch-Ritchie, J (1985) The aliphatic alcohols. In: AG Gilman, LS Goodman, TW Rall and F Murad (Eds.),The Pharmacologic Basis of Therapeutics (7th ed.). New York. Macmillan. Nathan, PW (1959) Intrathecal phenol to relieve spasticity in paraplegia. Lancet, ii: 1099-1102. Nathan, PW, Sears, TA and Smith, M (1965) Effects of phenol solutions on the nerve roots of the cat: an electrophysiological and histological study. J. Neurol. Sci., 2: 7-29. Naumann, MK, Lowe, NJ et al. (2002) Effect of botulinum toxin type A on quality of life measures in patients with excessive axillary sweating: a randomized controlled trial. Br. J. Dermatol., 147: 1218-1226. Ness, TJ (2000) Intravenous lidocaine inhibits visceral nociceptive reflexes and spinal neurons in the rat. Anesthesiology, 92: 1685-1691. Neuberger, TJ, Wittgen, CM, Schneider, TA, II, Andrus, CH, Panneton, WM and Kaminski, DL (1994) Evaluation of alternative proximal gastric vagotomy techniques after a 9-month interval in a rat model. Gastrointest. Endosc., 40: 316-320. Neubrand, M, Scheurlen, C, Schepke, M and Sauerbruch, T (2002) Long-term results and prognostic factors in the treatment of achalasia with botulinum toxin. Endoscopy, 34: 519-523. Comment in: Endoscopy (2002),34: 580-581. Nixdorf, DR, Heo, G and Major, PW (2002) Randomized controlled trial of botulinum toxin A for chronic myogenous orofacial pain. Pain, 99: 465-473.

689 Notermans, S, Dufrenne, J and Van Schothorst, MM (1978) Enzyme-linked immunosorbent assay for detection of Clostridium Botulinum Neurotoxin Type A. Jap. J. Med. Sci. BioI., 31: 81-85. Nussgens, Z and Roggenkamper, P (1997) Comparison of two botulinum-toxin preparations in the treatment of essential blepharospasm. Graefes Arch. Clin. Exp. Ophthalmol., 235: 197-199. Oechsner, M (2002) Treatment of hip adductor spasticity with botulinum toxin type B. Nervenarzt, 73: 11791182. O'Hanlan, JT, Galford, HR and Bosley, J (1969) The use of 45% alcohol to control spasticity. Virginia Med. Monthly, 96: 429-436. Okazaki, A (1993a) The effects of two and five percent aqueous phenol on the cat tibial nerve in situ I. the local anesthetic-like effect of aqueous phenol. Masui, 42: 726-732. Okazaki, A (1993b) The effects of two and five percent aqueous phenol on the cat tibial nerve in situ II. Effect on the circulation of the tibial nerve. Masui, 42: 819-825. Olney, RK, Aminoff, MJ, Gelb, DJ and Lowenstein, DH (1988). Neuromuscular effects distant from the site of botulinum neurotoxin injection. Neurology, 38: 17801783. On, AY, Kirazli, Y, Kismali, Band Aksit, R (1999) Mechanisms of action of phenol block and botulinus toxin Type A in relieving spasticity: electrophysiologic investigation and follow-up. Am. J. Phys. Med. Rehabil., 78: 344-349. O'Rahilly, R (1986) Gardner-Gray-O'Rahilly anatomy: a regional study of human structure (5th ed.). Philadelphia, WB Saunders. Pacchetti, C, Albani, G, Martignoni, E, Godi, L, Alfonsi, E and Nappi, G (1995) "Off' painful dystonia in Parkinson's disease treated with botulinum toxin. Mov. Disord., 10: 333-336. Pacchetti, C, Mancini, F, Bulgheroni, M, Zangaglia, R, Cristina, S, Sandrini, G and Nappi, G (2000) Botulinum toxin treatment for functional disability induced by essential tremor. Neurol. Sci., 21: 349-353. Pahwa, R, Busenbark, K, Swanson-Hyland, EF, Dubinsky, RM, Hubble, JP, Gray, C and Koller, WC (1995) Botulinum toxin treatment of essential head tremor. Neurology, 45: 822-824. Palace, J, Nairne, A, Hyman, N, Doherty, TV and Vincent, A (1998) A radioimmuno-precipitation assay for antibodies to botulinum A. Neurology, 50: 1463-1466. Pamphlett, R (1989) Early terminal and nodal sprouting of motor axons after botulinum toxin. J. Neural. Sci., 92: 181-192.

690 Pandyan, AD, Granat, MH and Stott, DJ (1997) Effects of electrical stimulation on flexion contractures in the hemiplegic wrist. Clin. Rehab., II: 123-130. Papini, E, Pacella, CM and Verde, G (1995) Percutaneous ethanol injection (PEl): what is its role in the treatment of benign thyroid nodules? Thyroid, 5: 147-150. Paqueron, X, Bouaziz, H, Macalou, D, Labaille, T, Merle, M, Laxenaire, MC and Benhamou, D (1999) The lateral approach to the sciatic nerve at the popliteal fossa: one or two injections? Anesth. Analg., 89: 1221-1225. Parziale, JR, Marino, AR and Herndon, JH (1988) Diagnostic peripheral nerve block resulting in compartment syndrome. Case report. Am. J. Phys. Med. Rehabil., 67: 82-84. Pasricha, PJ, Miskovsky, EP and Kalloo, AN (1994) Intrasphincteric injection of botulinum toxin for suspected sphincter of Oddi dysfunction. Gut, 35: 13191321. Pasricha, PJ, Ravich, WJ, Hendrix, TR, Sostre, S, Jones, B and Kalloo, AN (1995) Intrasphincteric botulinum toxin for the treatment of achalasia. N. Eng. J. Med., 332: 774-778. Erratum in: N. Eng. J. Med., 333: 75. Pastor, AM, Moreno-Lopez, B, De La Cruz, RR and Delgado-Garcia, JM (1997) Effects of botulinum neurotoxin type A on abducens motoneurons in the cat: ultrastructural and synaptic alterations. Neuroscience, 81: 457-478. Pauri, F, Boffa, L, Cassetta, E, Pasqualetti, P and Rossini, PM (1999) Botulinum toxin type-A treatment in spasticity increases the central conduction time to brain stimulation. Electroencephalogr. CUn. Neurophysiol., 51 (Suppl): 250--259. Pauri, F, Boffa, L, Cassetta, E, Pasqualetti, P and Rossini, PM (2000) Botulinum toxin type-A treatment in spastic paraparesis: a neurophysiological study. J. Neurol. Sci., 181: 89-97. Pearce, LB, Borodic, GE, First, ER and MacCallum, RD (1994) Measurement of botulinum toxin activity: evaluation of the lethality assay. Toxicol. Appl. Pharmacol., 128: 69-77. Pearce, LB, Borodic, GE, Johnson, EA, First, ER and MacCallum, R (1995) The median paralysis unit: a more pharmacologically relevant unit of biologic activity for botulinum toxin. Toxicon, 33: 217-227. Pedersen, E and Reske-Nielsen, E (1965) Neuropathology of subarachnoid phenol-glycerin. Acta. Neuropathol. (Berl.), 5: 112-116. Pelissier, J, Viel, E, Enjalbert, M, Kotzki, N and Eledjam, JJ (1993) Chemical neurolysis using alcohol (alcoholization) in the treatment of spasticity in the hemiplegic. Cah. Anesthesiol., 41: 139-143.

J-M. GRACIES AND D.M. SIMPSON

Pestronk, A and Drachman, DB (1978) Motor nerve sprouting and acetylcholine receptors. Science, 199: 1223-1225. Petrillo, CR, Chu, DS and Davis, SW (1980) Phenol block of the tibial nerve in the hemiplegic patient. Orthopedics, 3: 871-874. Petrillo, CR and Knoploch, S (1988) Phenol block of the tibial nerve for spasticity: a long-term follow-up study. Int. Disabil. Stud., 10: 97-100. Phelan, MW, Franks, M, Somogyi, GT, Yokoyama, T, Fraser, MO, Lavelle, JP, Yoshimura, N and Chancellor, MB (2001) Botulinum toxin urethral sphincter injection to restore bladder emptying in men and women with voiding dysfunction. J. Urol., 165: 1107-1110. Piccione, F, Mancini, E, Tonin, P and Bizzarini, M (1997) Botulinum toxin treatment of apraxia of eyelid opening in progressive supranuclear palsy: report of two cases. Arch. Phys. Med. Rehabil., 78: 525-529. Poemnyi, FA, Barsukova, MD and Gutorova, IuY. Treatment of spastic torticollis with phenol-glycerin and alcohol-novocaine blockade. Zh. Nevropatol. Psikhiatr. Im S S. Korsakova, 76: 1326-1330. Polak, F, Morton, R, Ward, C, Wallace, WA, Doderlein, L and Siebel, A (2002) Double-blind comparison study of two doses of botulinum toxin A injected into calf muscles in children with hemiplegic cerebral palsy. Dev. Med. Child Neurol., 44: 551-555. Polo, KB and Jabbari, B (1994) Botulinum toxin-A improves the rigidity of progressive supranuclear palsy. Ann. Neurol., 35: 237-239. Priori, A, Berardelli, A, Mercuri, B and Manfredi, M (1995) Physiological effects produced by botulinum toxin treatment of upper limb dystonia. Changes in reciprocal inhibition between forearm muscles. Brain, 118: 801-807. Racette, BA, Lauryssen, C and Perlmutter, JS (1998) Preoperative treatment with botulinum toxin to facilitate cervical fusion in dystonic cerebral palsy. Report of two cases. J. Neurosurg., 88: 328-330. Racette, BA, Lopate, G, Good, L, Sagitto, S and Perlmutter, JS (2002) Ptosis as a remote effect of therapeutic botulinum toxin B injection. Neurology, 59: 1445-1447. Reddihough, DS, King, JA, Coleman, OJ, Fosang, A, McCoy, AT, Thomason, P and Graham, HK (2002) Functional outcome of botulinum toxin A injections to the lower limbs in cerebral palsy. Dev. Med. Child Neurol., 44: 820--827. Reiter, F, Danni, M, Lagalla, G, Ceravolo, G and Provinciali, L (1998) Low-dose botulinum toxin with ankle taping for the treatment of spastic equinovarus foot after stroke. Arch. Phys. Med. Rehabil., 79: 532-535.

FOCAL INJECTION THERAPY Repka, MX, Savino, PJ and Reinecke, RD (1994) Treatment of acquired nystagmus with botulinum neurotoxin A. Arch. Ophthalmol., 112: 1320-1324. Richardson, D, Sheean, G, Werring, D, Desai, M, Edwards, S, Greenwood, R and Thompson, A (2000) Evaluating the role of botulinum toxin in the management of focal hypertonia in adults. J. Neurol. Neurosurg. Psychiatry, 69: 499-506. Robin, L, Herman, D and Redett, R (1996) Botulism in a pregnant woman. N. Eng. J. Med., 335: 823-824. Rodriquez, AA, McGinn, M and Chappell, R (2000) Botulinum toxin injection of spastic finger flexors in hemiplegic patients. Am. J. Phys. Med. Rehabil., 79: 44-47. Rollnik, JD, Wohlfarth, K, Dengler, Rand Bigalke, H (2001) Neutralizing botulinum toxin type a antibodies: clinical observations in patients with cervical dystonia. Neurol. Clin. Neurophysiol., 2001: 2--4. Rollnik, JD and Dengler, R (2002) Botulinum toxin (DYSPORT) in tension-type headaches. Acta Neurochir., 79 (Supp!.): 123-126. Rosales, RL, Arirnura, K, Takenaga, Sand Osame, M (1996) Extrafusal and intrafusal muscle effects in experimental botulinum toxin-A injection. Muscle Nerve, 19: 488--496. Royal, M and Jenson, M (2002) Botulinum toxin type Bin the treatment of refractory cervicothoracic myofascial pain. Arch. Phys. Med. Rehabil., 83: 1670. Ruff, RL (1977) A quantitative analysis of local anaesthetic alteration of miniature end-plate currents and end-plate current fluctuations. J. Physiol. (Lond.), 264: 89-124. Ruiz, MF, Moreno, M, Sanchez-Garrido, CM and Rodriguez, JM (2000) Botulinum treatment of infantile esotropia with abduction nystagmus. 1. Pediatr. Ophthalmol. Strabismus, 37: 196-205. Saadia, D, Voustianiouk, A, Wang, AK and Kaufmann, H (2001) Botulinum toxin type A in primary palmar hyperhidrosis: randomized, single-blind, two-dose study. Neurology, 57: 2095-2099. Sampaio, C, Castro-Caldas, A, Sales-Luis, ML, Alves, M, Pinto, Land Apolinario, P (1993) Brachial plexopathy after botulinum toxin administration for cervical dystonia. J. Neurol. Neurosurg. Psychiatry, 56: 220-221. Sampaio, C, Ferreira, n. Simoes, F, Rosas, MJ, Magalhaes, M, Correia, AP, Bastos-Lima, A, Martins, Rand Castro-Caldas, A (1997a) DYSBOT: a single-blind, randomized parallel study to determine whether any differences can be detected in the efficacy and tolerability of two formulations of botulinum toxin type A - Dysport and Botox - assuming a ratio of 4:1. Mov. Disord., 12: 1013-1018. Sampaio, C, Ferreira, 11,Pinto, AA, Crespo, M, Ferro, 1M and Castro-Caldas, A (1997b) Botulinum toxin type A

691 for the treatment of arm and hand spasticity in stroke patients. Clin. Rehabil., 11: 3-7. Sanchez-Prieto, J, Sihra, TS, Evans, D, Ashton, A, Dolly, JO and Nicholls, DG (1987) Botulinum toxin A blocks glutamate exocytosis from guinea-pig cerebral cortical synaptosomes. Eur. J. Biochem., 165: 675-681. Sanders, DB, Massey, EW and Buckley, EG (1986) Botulinum toxin for blepharospasm: single fiber EMG studies. Neurology, 36: 545-547. Sanders, I, Mu, L, Amirali, A et a!. (1998) Motor endplate mapping of the human biceps brachii muscle. Ann. Neurol., 44: 501. Sankhla, C, Jankovic, J and Duane, D (1998) Variability of the immunologic and clinical response in dystonic patients immunoresistant to botulinum toxin injections. Mov. Disord., 13: 150-154. Santos, G, Novell, JR, Khoury, G, Winslet, MC and Lewis, AA (1993) Long-term results of large-dose, single session phenol injection sclerotherapy for hemorroids. Dis. Colon Rectum, 36: 958-961. Savino, PJ, Sergott, RC, Bosley, TM and Schatz, NJ (1985) Hemifacial spasm treated with botulinum A toxin injection. Arch. Ophthalmol., 103: 1305-1306. Schantz, EJ and Johnson, EA (1990) Dose standardization of botulinum toxin. Lancet, 335: 421. Schantz, EJ and Johnson, EA (1992) Properties and use of botulinum toxin and other microbial neurotoxins in medicine. Microbiol. Rev., 56: 80-99. Scheidt, CE, Rayki, 0, Heinen, F and Nickel, T (1995) Subgroups of torticollis spasmodicus from the psychosomatic viewpoint. Results of a cluster analysis of 144 cases. German Study Group of Dystonia Research. Nervenarzt, 66: 422--429. Scheidt, CE, Rayki, 0, Nickel, T, Heinen, F, Wissel, J, Poewe, W, Benecke, R, Arnold, G, Oertel, W, Dengler, Rand Deuschl, G (1998) Psychosomatic aspects of idiopathic spasmodic torticollis. Results of a multicenter study. Psychother. Psychosom. Med. Psychol., 48: 1-12. Schichor, A, Bernstein, B, Weinermann, H, Fitzgerald, J, Yordan, E and Schechter, N (1994) Lidocaine as a diluent for ceftriaxone in the treatment of gonorrhea. Does it reduce the pain of injection? Arch. Pediatn Adolesc. Med., 148: 72-75. Schmitt, WJ, Slowey, E, Fravi, N, Weber, Sand Burgunder, JM (2001) Effect of botulinum toxin A injections in the treatment of chronic tension-type headache: a double-blind, placebo-controlled tria!. Headache, 41: 658-664. Schneider, I, Thumfart, WF, Pototschnig, C and Eckel, HE (1994) Treatment of dysfunction of the cricopharyngeal muscle with botulinum A toxin: introduction of a new, noninvasive method. Ann. Otol. Rhinol. Laryngol., 103: 31-35.

692 Schnider, P, Binder, M, Auff, E, Kittler, H, Berger, T and Wolff, K (1997) Double-blind trial of botulinum A toxin for the treatment of focal hyperhidrosis of the palms. Br. J. Dennatol., 136: 548-552. Schurch, B, Hauri, D, Largo, M, Kreienbuhl, B, Meyer, E and Rossier, AB (1990) Effects of Botulinum A Toxin on the periurethral striated sphincter of the neurogenic bladder. Preliminary study. J. Urol. (Paris), 96: 375380. Schurch, B, Hodler, J and Rodic, B (1997) Botulinum A toxin as a treatment of detrusor-sphincter dyssynergia in patients with spinal cord injury: MRl controlled transperineal injections. J. Neurol. Neurosurg. Psychiatry, 63: 474-476. Scott, AB (1980) Botulinum toxin injection into extraocular muscles as an alternative to strabismus surgery. Ophthalmology, 87: 1044-1049. Scott, AB (1981) Botulinum toxin injection of eye muscles to correct strabismus. Trans. Am. Ophthalmol. Soc., 79: 734-770. Scott, AB (1988) Antitoxin reverses botulinum side effects. Eye, 2: 29-32. Scott, AB (1997) Preventing ptosis after botulinum treatment. Ophthal. Plast. Reconstr. Surg., 13: 81-83. Scott, AB and Suzuki, D (1988) Systemic toxicity of botulinum toxin by intramuscular injection in the monkey. Mov. Disord., 3: 333-335. Scott, AB, Rosenbaum, A and Collins, CC (1973) Pharmacologic weakening of extraocular muscles. Invest. Ophthalmol., 12: 924-927. Scott, BA, Weinstein, Z, Chiteman, R and Pulliam, MW (1985) Intrathecal phenol and glycerin in metrizamide for treatment of intractable spasms in paraplegia. Case report. J. Neurosurg., 63: 125-127. Sellin, LC, Molgo, J, Tornquist, K, Hansson, Band Thesleff, S (1996) On the possible origin of giant or slow-rising miniature end-plate potentials at the neuromuscular junction. Pfliigers Arch., 431: 325-334. Sesardic, D, McLellan, K, Ekong, TA and Das, RG (1996) Refinement and validation of an alternative bioassay for potency testing of therapeutic botulinum type A toxin. Pharmacol. Toxicol., 78: 283-288. Shaari, CM, George, E, Wu, BL, Biller, HF and Sanders, I (1991) Quantifying the spread of botulinum toxin through muscle fascia. Laryngoscope, 101: 960-964. Shaari, CM and Sanders, I (1993) Quantifying how location and dose of botulinum toxin injections affect muscle paralysis. Muscle Nerve, 16: 964-969. Shapiro, RL, Hatheway, C and Swerdlow, DL (1998) Botulism in the United States: a clinical and epidemiologic review. Ann. Intern. Med., 129: 221-228. Sharma, P and Gupta, N (2002) Social phobia in spasmodic torticollis: some conceptual issues. J. Neurol. Neurosurg. Psychiatry, 73: 461; author reply

J-M. ORAClES AND D.M. SIMPSON

461-462. Comment on: J. Neurol. Neurosurg. Psychiatry (2001), 71: 499-504. Sheean, GL (2001) Botulinum treatment of spasticity: why is it so difficult to show a functional benefit? Curr. Opin. Neurol., 14: 771-776. Sheean, GL (2002) Botulinum toxin for the treatment of musculoskeletal pain and spasm. Curr. Pain Headache Rep., 6: 460-469. Sheridan, RE, Deshpande, SS, Nicholson, JD and Adler, M (1997) Structural features of arninoquinolines necessary for antagonist activity against botulinum neurotoxin. Toxicon., 35: 1439-1451. Shiomi, Y, Funabiki, K, Naito, Y, Fujiki, N and Tsuji, J (1997) The effect of intravenous lidocaine injection on hearing thresholds. Auris Nasus Larynx, 24: 351-356. Siatkowski, RM, Tyutyunikov, A, Biglan, AW, Scalise, D, Genovese, C, Raikow, RB, Kennerdell, JS and Feuer, WJ (1993) Serum antibody production to botulinum A toxin. Ophthalmology, 100: 1861-1866. Siegel, LS, Johnson-Vinegar, AD and Sellin, LC (1986) Effect of 3,4-diarninopyridine on the survival of mice injected with botulinum neurotoxin type A, B, E, or F. Toxicol. Appl. Phannacol., 84: 255-263. Siegel, LS (1988) Human immune response to botulinum pentavalent (ABCDE) toxoid determined by a neutralization test and by an enzyme-linked immunosorbent assay. J. Clin. Microbiol., 26: 2351-2356. Silberstein, S, Mathew, N, Saper, J and Jenkins, S (2000) Botulinum toxin type A as a migraine preventive treatment. For the BOTOX Migraine Clinical Research Group. Headache, 40: 445-450. Simon, MA, Vree, TB, Gielen, MJ, Booij, LH and Lagerwerf, AJ (1999) Similar motor block effects with different disposition kinetics between lidocaine and ( + or -) articaine in patients undergoing axillary brachial plexus block during day case surgery. Int. J. Clin. Phannacol. Ther., 37: 598-607. Simpson, D, Alexander, DN, O'Brien, CF, Tagliati, M, Aswad, AS, Leon, JM, Gibson, J, Mordaunt, JM and Monaghan, EP (1996) Botulinum toxin type A in the treatment of upper extremity spasticity: a randomized, double-blind, placebo-controlled trial. Neurology, 46: 1306-1310. Simpson, LL (1981) The origin, structure, and pharmacologic activity of botulinum toxin. Phannacol. Rev., 33: 155-188. Slawek, J, Szymkiewicz-Rogowska, A and Friedman, A (2000) Rubral tremor of Holmes, rare case of pathological tremor: case report. Neurol. Neurochir. Pol., 34: 775-782. Sloop, RR, Cole, BA and Escutin, RO (1997) Reconstituted botulinum toxin type A does not lose potency in

FOCAL INJECTION THERAPY

humans if it is refrozen or refrigerated for two weeks before use. Neurology, 48: 249-253. Smith, SJ, Ellis, E, White, S and Moore, AP (2000) A double-blind placebo-controlled study of botulinum toxin in upper limb spasticity after stroke or head injury. Clin. Rehabil., 14: 5-13. Smith, CP, O'Leary, M, Erickson, J, Somogyi, GT and Chancellor, MB (2002) Botulinum toxin urethral sphincter injection resolves urinary retention after pubovaginal sling operation. Int. Urogynecol. J. Pelvic Floor Dysfunct., 13: 185-186. Snow, BJ, Tsui, JKC, Bhatt, MH, Varelas, M, Hashimoto, SA and Calne, DB (1990) Treatment of spasticity with botulinum toxin: a double blind study. Ann. Neurol., 28: 512-515. Somma-Mauvais, H, Poujet, J, Gueriot-Milandre, C, Grob, JJ, Bonerandi, JJ and Gastaut, JL (1992) Trigeminal neurotrophic ulcer and vascular disorders of the brain stem. A clinico-electrophysiological study. Rev. Neural. (Paris), 148: 708-712. Sotgiu, ML, Lacerenza, M and Marchettini, P (1992) Effect of systemic lidocaine on dorsal hom neuron hyperactivity following chronic peripheral nerve injury in rats. Somatosens. Mot. Res., 9: 227-233. Soulayrol, S, Caperan, A, Penot-Ragon, C, Beaulieu, JP and Gastaut, JL (1993) Treatments by local injections of botulinum toxin in neurology. Indications and results. Presse Med., 22: 957-963. Spencer, RF and McNeer, KW (1987) Botulinum Toxin paralysis of adult monkey extraocular muscle. Structural alterations in orbital, singly innervated muscle fibers. Arch. Ophthalmol., 105: 1703-1711. Spira, R (1971) Management of spasticity in cerebral palsied children by peripheral nerve block with phenol. Dev. Med. Child Neurol., 13: 164-173. Stahl, JS, Averbuch-Heller, L, Remler, BF and Leigh, RJ (1998) Clinical evidence of extraocular muscle fibertype specificity of botulinum toxin. Neurology, 51: 1093-1099. Stanley, EF and Drachman, DB (1983) Botulinum toxin blocks quantal but not non-quantal release of Ach at the neuromuscular junction. Brain Res., 261: 172-175. St Clair, EH, DiLiberti, JH and O'Brien, ML (1975) Letter: Observations of an infant born to a mother with botulism. J. Pediatr., 87: 658. Stefanko, S and Zebrowski, S (1968) Histological changes in the nerve roots and spinal cord after intrathecal administration of phenol for relief of spasticity. Pol. Med. J., 7: 1204-1208. Stefanko, S and Zebrowski, S (1968) Histopathological picture of neural roots and spinal cord after injection of phenol into the spinal canal to remove spasticity. Neural. Neurochir. Pol., 2: 19-23.

693 Sun, CF, Li, ZJ, Wu, LH, Xu, LT, Wu, N, Fan, W, Jin, L and Fang, Q (1987) Successful surgical treatment of three cases of type A Wolff Parkinson White syndrome using intramyocardial injection of absolute alcohol. Proc. Chin. Acad. Med. Sci. Peking Union Med. csu.. 2: 165-167. Sung, DH, Han, TR, Park, WH, Je Bang, H, Kim, 1M, Chung, SH and Woo, EJ (2001) Phenol block of peripheral nerve conduction: Titrating for optimum effect. Arch. Phys. Med. Rehabil., 82: 671-676. Supe, AN, Mathur, SK and Borwankar, SS (1994) Esophageal endoscopic sclerotherapy in children using 3% aqueous phenol. Ind. J. Gastroenterol., 13: 1-4. Swartling, C, Farnstrand, C, Abt, G, Stalberg, E and Naver, H (2001) Side-effects of intradermal injections of botulinum A toxin in the treatment of palmar hyperhidrosis: a neurophysiological study. Eur. J. Neurol., 8: 451-456. Tardieu, C, Tardieu, G, Hariga, J, Gagnard, L and Velin, J (1964) Fondement experimental d' une therapeutique des raideurs d' origine cerebrale (effets de l' alcoolisation menagee du nerf moteur sur Ie reflexe d' etirement de l'animal decerebre). Arch. Fr. Pediatn, 21: 5-23. Tardieu, C, Tardieu, G, Colbeau-Justin, P, Huet de la Tour, E and Lespargot, A (1979) Trophic muscle regulation in children with congenital cerebral lesions. J. Neural. Sci., 42: 357-364. Tardieu, G, Tardieu, C, Hariga, J, Gagnard, Land Joly, C (1962) Action des infiltrations d'alcool dilue sur diverses raideurs liees aux lesions cerebrates. Bull. Mem. Soc. MM. Hop. Paris, 113: 7-12. Tardieu, G, Hariga, J, Tardieu, C, Gagnard, Land Velin, J (1964) Traitement de la spasticite par infiltration d'alcool dilue aux points moteurs, ou par injection epidurale. Rev. Neurol., 110: 563-566. Tardieu, G, Tardieu, C, Hariga, J and Gagnard, L (1968) Treatment of spasticity by injection of dilute alcohol at the motor point or by epidural route. Clinical extension of an experiment on the decerebrate cat. Dev. Med. Child Neurol., 10: 555-568. Tardieu, G, Tardieu, C and Hariga, J (1971) Infiltrations with 45 degrees ethyl alcohol of the motor points, the roots by the epidural route, or of the posterior tibial nerve. Their indications and contra-indications in the various forms of spasticity (experience of 10 years) Rev. Neural. (Paris), 125: 63-68. Tay, B, Wallace, MS and Irving, G (1997) Quantitative assessment of differential sensory blockade after lumbar epidural lidocaine. Anesth. Analg., 84: 10711075. Taylor, JJ and Woolsey, RM (1976) Dilute ethyl alcohol: effect on the sciatic nerve of the mouse. Arch. Phys. Med. Rehabil., 57: 233-237.

694 Tekkok, IH, Carter, DH and Brinker, R (1996) Spinal subdural haematoma as a complication of immediate epidural blood patch. Can. 1. Anaesth., 43: 306-309. The WE MOVE Spasticity Study Group (2002) Dosing, administration, and a treatment algorithm for use of botulinum toxin type A for adult-onset muscle overactivity in patient with an upper motorneuron lesion. In: NH Mayer and DM Simpson (Eds.), Spasticity: Etiology, Evaluation, and the Role of Botulinum toxin. New York: WE MOVE, pp. 154-165. Tintner, R and Jankovic, J (2001) Focal dystonia: the role of botulinum toxin. Curr. Neurol. Neurosci. Rep., 1: 337-345. Tomsak, RL, Remler, BF, Averbuch-Heller, L, Chandran, M and Leigh, RJ (1995) Unsatisfactory treatment of acquired nystagmus with retrobulbar injection of botulinum toxin. Am. J. Ophthalmol., 119: 489-496. Tonello, F, Morante, S, Rossetto, 0, Schiavo, G and Montecucco, C (1996) Tetanus and botulism neurotoxins: a novel group of zinc endopeptidases. Adv. Exp. Med. BioI., 389: 251-260. Tonge, DA (1974) Chronic effects of botulinum toxin on neuromuscular transmission and sensitivity to acetylcholine in slow and fast skeletal muscles of the mouse. J. Physiol. (Lond.), 241: 127-139. Traynelis, VC, Ryken, T, Rodnitzky, RL and Menezes, AH (1992) Botulinum toxin enhancement of postoperative immobilization in patients with cervical dystonia. Technical note. J. Neurosurg., 77: 808-809. Trosch, RM and Pullman, SL (1994) Botulinum toxin A injections for the treatment of hand tremors. Mov. Disord., 9: 601-609. Tsui, JK (2003) Treatment of dystonia in Parkinson's disease. Adv. Neurol., 91: 361-364. Tsui, JKC, Eisen, A, Stoessl, AJ, Calne, S and Calne, DB (1986) Double-blind study of botulinum toxin in spasmodic torticollis. Lancet, 2: 245-247. Tsui, JKC, Wong, NLM, Wong, E and Calne, DB (1988) Production of circulating antibodies to botulinum-A toxin in patients receiving repeated injections for dystonia. Ann. Neurol., 23: 181. Tsui, JKC, Hayward, M, Mak, EKM and Schulzer, M (1995) Botulinum toxin type B in the treatment of cervical dystonia: a pilot study. Neurology, 45: 2109-2110. Turkel, C, Dru, R, Daggett, Sand Brin, M (2002) Neutralizing antibody formation is rare following repeated injections of a low protein formulation of botulinum toxin type A (BTX-A) in patients with poststroke spasticity. Neurology, 58: A316. Tyler, HR (1963) Physiological observations in human botulism. Arch. Neurol., 9: 661-670. Umstadt, HE (2002) (Botulinum toxin in oromaxillofacial surgery). Mund. Kiefer Gesichtschir., 6: 249-260.

J-M. GRACIES AND D.M. SIMPSON

Ueyama, T, Maru, E, Shimada, M, Suzuki, H and Ikeda, M (1999) (Lidocaine increases the granule cell excitability in hippocampal dentate area instead of affecting GABAergic inhibition). Masui, 48: 739-746. Vaezi, MF, Richter, JE, Wilcox, CM, Schroeder, PL, Birgisson, S, Slaughter, RL, Koehler, RE and Baker, ME (1999) Botulinum toxin vs. pneumatic dilatation in the treatment of achalasia: a randomized trial. Gut, 44: 231-239. Vallbo, AB (1974) Afferent discharge from human muscle spindles in non-contracting muscles. Steady state impulse frequency as a function of joint angle. Acta. Physiol. Scand., 90: 303-318. Valls-Sole, J, Tolosa, ES and Ribera, G (1991) Neurophysiological observations on the effects of botulinum toxin treatment in patients with dystonic blepharospasm. J. Neurol. Neurosurg. Psychiatry, 54: 310-313. Valls-Sole, J, Tolosa, ES, Marti, MJ and Allam, N (1994) Treatment with botulinum toxin injections does not change brainstem interneuronal excitability in patients with cervical dystonia. Clin. Neurophannacol., 17: 229-235. Van Ermengem, E (1897) Uber einem neuen anaeroben Bacillus und seine Beziehungen zum Botulismus. Z. Hyg. lnfektionskrankh., 26: 1-56 (English translation: A new anaerobic bacillus and its relation to botulism. Rev. lnf. Dis., 1: 701-719). Viel, EJ, Perennou, D, Ripart, J, Pelissier, J and Eledjam, 11 (2002) Neurolytic blockade of the obturator nerve for intractable spasticity of adductor thigh muscles. Eur. J. Pain, 6: 97-104. Wainapel, SF, Haigney, D and Labib, K (1984) Spastic hemiplegia in a quadriplegic patient: treatment with phenol nerve block. Arch. Phys. Med. Rehabil., 65: 786-787. Wall, SA, Chait, LA, Temlett, JA, Perkins, B, Hillen, G and Becker, P (1993) Botulinum A chemodenervation: a new modality in cerebral palsied hands. Br. J. Plast. Surg., 46: 703-706. Walthard, KM and Tchicaloff, M (1971) Motor points. In: S Licht (Ed.), Electrodiagnosis and Electromyography (3rd ed.). New Haven, Licht. Warfel, JH (1985) The extremities: muscles and motor points (5th ed.). Lea & Febiger, Philadelphia. Warrick, P, Dromey, C, Irish, J and Durkin, L (2000a) The treatment of essential voice tremor with botulinum toxin A: a longitudinal case report. J. Voice, 14: 410421. Warrick, P, Dromey, C, Irish, JC, Durkin, L, Pakiam, A and Lang, A (2000b) Botulinum toxin for essential tremor of the voice with multiple anatomical sites of tremor: a crossover design study of unilateral vs. bilateral injection. Laryngoscope, 110: 1366-1374.

FOCAL INJECTION THERAPY

Wassef, MR (1993) Interadductor approach to obturator nerve blockade for spastic conditions of adductor thigh muscles. Reg. Anesth., 18: 13-17. Wehrmann, T, Schmitt, TH, Arndt, A, Lembcke, B, Caspary, WF and Seifert, H (2000) Endoscopic injection of botulinum toxin in patients with recurrent acute pancreatitis due to pancreatic sphincter of Oddi dysfunction. Aliment Pharmacol. Ther., 14: 14691477. Westerlund, T, Vuorinen, V and Roytta, M (2001) Same axonal regeneration rate after different endoneurial response to intraneural glycerol and phenol injection. Acta. Neuropatho!. (Ber!.), 102: 41-54. Wheeler, AH, Goolkasian, P and Gretz, SS (1998) A randomized, double-blind, prospective pilot study of botulinum toxin injection for refractory, unilateral, cervicothoracic, paraspinal, myofascial pain syndrome. Spine, 23: 1662-1666, discussion 1667. Wichers, MJ (1991) Partial blocking of the tibial nerve with phenol as treatment of gait disorders due to pes equinus in central paralysis. Ned. Tijdschr. Geneeskd., 135: 752-754. Wiederholt, WC (1970) "End-plate noise" in electromyography. Neurology, 20: 214-224. Wiegand, Hand Wellhoner, HH (1974) Proceedings: Type A botulinum toxin in cats: neural ascent and action on spinal cord reflexes. Naunyn Schmiedebergs Arch. Phannacol., 282: R106. Wiegand, H and Wellhoner, HH (1977) The action of botulinum A neurotoxin on the inhibition by antidromic stimulation of the lumbar monosynaptic reflex. Naunyn Schmiedebergs Arch. Pharmacol., 298: 235238. Wiegand, H, Erdmann, G and Wellhoner, HH (1976) '25I-labeled botulinum A neurotoxin: pharmacokinetics in cats after neuromuscular injection. Naunyn-Schmiedebergs Arch. Phannakol., 292: 161-165. Wilson, LR, Gandevia, SC, Inglis, JT, Gracies, JM and Burke, D (1999) Muscle spindle activity in the affected upper limb after a unilateral stroke. Brain, 122: 20792088. Wissel, J, Kabus, C, Wenzel, R, Klepsch, S, Schwarz, U, Nebe, A, Schelosky, L, Scholz, U and Poewe, W (1996) Botulinum toxin in writer's cramp: objective response evaluation in 31 patients. J. Neurol. Neurosurg. Psychiatry, 61: 172-175. Wissel, J, Masuhr, F, Schelosky, L, Ebersbach, G and Poewe, W (1997) Quantitative assessment of botulinum toxin treatment in 43 patients with head tremor. Mov. Disord., 12: 722-726.

695 Wohlfarth, K, Goschel, H, Frevert, J, Dengler, Rand Bigalke, H (1997) Botulinum A toxins: units vs. units. Naunyn Schmiedebergs Arch. Phannacol., 355: 335340. Wohlfarth, K, Schubert, M, Rothe, B, Elek, J and Dengler, R (2001) Remote F-wave changes after local botulinum toxin application. Clin. Neurophysiol., 112: 636640. Wolf, JH and English, AW (2000) Muscle spindle reinnervation following phenol block. Cells Tissues Organs, 166: 325-329. Wollina, U, Karamfilov, T and Konrad, H (2002) Highdose botulinum toxin type A therapy for axillary hyperhidrosis markedly prolongs the relapse-free interval. J. Am. Acad. Dennato!., 46: 536-540. Wong, GY and Brown, DL (1995) Transient paraplegia following alcohol celiac plexus block. Reg. Anesth., 20: 352-355. Wood, KE (1978) The use of phenol as a neurolytic agent: a review. Pain, 5: 205-229. Yablon, SA, Agana, BT, Ivanhoe, CB and Boake, C (1996) Botulinum toxin in severe upper extremity spasticity among patients with traumatic brain injury: an openlabeled trial. Neurology, 47: 939-944. Yadav, SL, Singh, U, Dureja, GP, Singh, KK and Chaturvedi, S (1994) Phenol block in the management of spastic cerebral palsy. Ind. J. Pediatr., 61: 249-255. Yamauchi, Y, Kojoh, H, Nagaro, T, Miyasaki, H, Kimura, Sand Arai, T (1993) Treatment of hyperhydrosis with caudal epidural alcohol block in a patient with cervical cord injury. Masui, 42: 606-610. Yee, WC and Pestronk, A (1987) Mechanisms of postsynaptic plasticity: remodeling of the junctional acetylcholine receptor cluster induced by motor nerve terminal outgrowth. J. Neurosci., 7: 2019-2024. Ye, JH, Ren, J, Kmjevic, K, Liu, PL and McArdle, JJ (1999) Cocaine and lidocaine have additive inhibitory effects on the GABAA current of acutely dissociated hippocampal pyramidal neurons. Brain Res., 821: 2632. Yoshimura, DM, Aminoff, MJ, Tami, TA and Scott, AB (1992) Treatment of hemifacial spasm with botulinum toxin. Muscle Nerve, 15: 1045-1049. Zack, SI (1971) The Motor Endplate. Philadelphia, WB Saunders. Zuber, M, Sebald, M, Bathien, N, de Recondo, J and Rondot, P (1993) Botulinum antibodies in dystonic patients treated with type A botulinum toxin: frequency and significance. Neurology, 43: 1715-1718.

Movement Disorders Handbook of Clinical Neurophysiology, Vol. I M. Hallett (Ed.) © 2003 Elsevier B. V. All rights reserved

697 CHAPTER 42

Deep brain stimulation in Parkinson's disease: technique and prospective, facts and comments Alim Louis Benabid*, Stephan Chabardes, Abdelhamid Benazzouz, Adnan Koudsie, Valerie Fraix, Paul Krack, Claire Ardouin and Pierre Pollak INSERM U318, Joseph Fourier University, and Albert Michal/on University Hospital, 38043 Grenoble, France

42.1. Introduction

42.1.1. Rationale High frequency stimulation (HFS) is based on the empirical observation that intraoperative stimulation induces, at high frequency but not at low frequency, an inhibition of the neural structure which is stimulated. This has been observed and verified in the thalamic target, the pallidum and also in the subthalamic nucleus. Clinical evidence coming from various groups has confirmed our first observations that HFS can mimic the effect obtained by the usual surgical lesioning of the target, usually achieved by electrocoagulation (and in the past by injection of alcohol or wax, or by implantation of radioactive seeds, particularly 90 Yttrium). Thalamic stimulation has proven to be very efficient on essentially one symptom, the tremor. Pallidal stimulation is able to influence the three main symptoms of Parkinson's disease (PD), but is specifically efficient on the so-called dyskinesias, abnormal involuntary movements induced after several years of levodopa treatment at relatively high doses. The subthalamic nucleus (STN), which was introduced as a target in 1993, very quickly appeared to be the best target and is able to control all the cardinal symptoms of PD, namely rigidity, akinesia, and tremor. For this reason this method has become currently the major surgical alternative for advanced

* Correspondence to: Dr. A.L. Benabid, INSERM 318, Joseph Fourier University, Centre Hospitalier Universitaire, Pavilion B, BP 217, 38043 Grenoble Cedex 9, France. E-mail address:[email protected] or [email protected] Fax: +33(4) 6765619.

stages of PD. Controlled clinical trials still need to be performed to define the safety of the method, the extent of the indications, and of the beneficial effects for the patient. The mechanism of action of deep brain stimulation (DBS), whatever the target is, is still a matter of debate. Is it an inhibition? Is it the excitation of an indirect inhibitory part of the network? Is it due to a jamming or a perturbation of the neural message, which finally is not properly transmitted, and if this message is pathological, then does this perturbation lead to the alleviation of the symptoms?

42.1.2. Objective The purpose of DBS of STN is to provide a surgical alternative to medical treatment in advanced stages of PD. This goal was formerly achieved by lesioning of various targets, mainly the thalamus and then the pallidum. However these methods, although they were effective, were plagued with a high rate of complications, particularly when surgery had to be done bilaterally, which happens quite often, at this stage. Therefore, research has focused on new methods which could provide flexibility, reversibility, adaptability and still a rather high rate of success and efficient effects for the patient similar to those observed during ablative functional neurosurgery, but without these complications. Neuronal grafts have led to clinical trials during the last three decades following a large body of experimental evidence. However, although the approach is extremely interesting, the clinical result cannot honestly be proposed to patients as an equal opportunity and alternative within a panel of surgical solutions. It might be possible that the technology derived from stem cells might produce new cell lines easily available, compatible with many patients, and

698 also able to differentiate according to the needs represented by the target they are implanted in. Recent evaluation of a series of patients implanted in the striatum using fetal grafts have failed to demonstrate a very significant beneficial effect, but on the other hand, have revealed that a large proportion of those grafted patients present dyskinesias, highly comparable to those induced by levodopa treatment, which they were supposed to alleviate. Additionally, the intensity of the beneficial effect is still rather poor and not really usable. Brain stimulation offers a wide field of application methods which can not only be adapted to various targets for PD but also for other pathologies, such as dystonia, epilepsy, obsessive-compulsive disorders and possibly in malignant obesity.

42.2. Methods 42.2.1. Indications and patient selections 42.2.1.1. Indications aim at the best motor benefit DOPA response is the best predictor of the success ofDBS: Among the clinical criteria which allow selection of the patients before surgery, it has been observed that DOPA response parallels significantly the final outcome (Charles et al., 2002), provided that the electrodes are inserted in the right place. Univariate analysis demonstrated that improvement from levodopa, as measured by change in the UPDRS-III score, correlated positively with postoperative improvement from stimulation (r=0.58, p
A.L. BENABID ET AL.

response to levodopa) x 3.9 - (freezing response to levodopa) x 2.4. This was simplified by retaining only the three most powerful factors. Predicted postoperative improvement relative to preoperative UPDRS-III off score=34-agexO.29+(rigidity response to levodopa) x 1.3+ (pull test response to levodopa) x 3.8. Other signs, if assessed intraoperatively, might also predict outcome. The pull test may be a strong predictor based on its ability to identify patients with minimal lesions outside the dopaminergic nigrostriatal system. As a rule, the UPDRS III score at the best on, after DOPA challenge (200 milligrams Levodopa) provides a good picture of the level of improvement which can be achieved after bilateral STN stimulation (Hilker R, et al., 2002) This allows either selection or rejection of patients according to their response to Levodopa (Krack et al., 2000), or to be able to inform, as precisely as possible, the patients, the family and the doctors of what could be the postoperative outcome. This is of primary importance as the correct expectation from the patient of the benefit of surgery, must be clearly stated before the common decision is made. This is an important part of the informed consent as patients may expect surgery to eradicate the disease and bring them a new life. The differences between the expectations and the final outcome may be responsible, at least in part, of some postoperative depression states which have been observed. This correlation was observed in a previous series of 30 patients (Limousin, 1998) and has been confirmed in more recent evaluation at longer follow-ups. Particularly, this is true for the main triad symptoms, tremor, rigidity, bradykinesia, and provide a fairly good prognosis of the improvement of global functions, particularly gait. Conversely, nonDOPA-sensitive symptoms and symptoms outside of the idiopathic PD profile, such as dysautonomic symptoms, are poorly or not improved. However, the improvement of stature, of freezings, and dystonic postures, are more difficult to predict, given the fact that their response to Levodopa may exist but may be less visible and wrongly considered at the preoperative stage as non-DOPA responsive. Particular points must be made about speech and hypophonia which respond poorly to STN stimulation, although they are improved, but not as efficiently as other symptoms by levodopa or doparninergic agonists. This discrepancy with DOPA sensitive symptoms

699

DEEP BRAIN STIMULATION IN PARKINSON'S DISEASE

and the response to STN stimulation raises questions - it might be due to a possible "symptomatotopy" within STN. During the operative stage, the clinical exploration correlated to electrophysiology (stimulation and recording) concerns principally the motor symptoms and motor performances, and concentrates also on side effects. We have not, so far, designed a definite strategy to evaluate intraoperatively the effect of stimulation on speech. Then, this might lead to a systematic missing of the speech target if there is a speech target different from the somato-motor area in the STN. It might be also possible that, from the evolutionary point of view, the subthalamic nucleus which is an early appearing substructure, might not be involved in the control of the motricity of speech which is a function which appeared later. Therefore, by essence, STN stimulation would not be able to improve speech and this might explain why addition of dopaminergic drugs in post-operative stage improves the hypophonia. Conversely, the important decrease in drug dosage allowed by the beneficial effects of surgery might be particularly responsible for the majoration of hypophonia after surgery, in patients who had it already before. 42.2.1.2. Indications must avoid neuropsychological failure The occurrence of depression, although usually temporary, is observed in about 20% of the cases and may be sufficiently intense to impair the quality of life of the patient and of his caregiver. The selection criteria so far eliminate patients with a past history of mental illness, severe depression, suicidal attempts or psychosis and hallucinations. However, special attention should be paid preoperatively to detect patients in whom surgery will increase the depressive tendency or would induce it. At least, a careful neuropsychological selection should be done, and if necessary psychotherapeutic support should be provided or suggested to the patients during the waiting time before surgery. Expectations of the patients should be carefully investigated, the prediction of the outcome regarding every symptom and every deficit of the patients should be clearly stated from the statistical observation currently available, in order to make them perfectly aware of what can be expected and to avoid postoperative disappointment. An important point which appears is the profound change in daily life which is induced

by STN stimulation. In the best cases, the majority of symptoms are significantly alleviated and the drug dosage is also decreased. As some patients have stated, "surgery gives birth to adults" and creates situations which are totally novel (Lagrange et aI., 2002). A patient who spent 10 years embedded in his disease surrounded by care givers, withdrawn from his family, social, professional world, is suddenly thrown in a totally new situation which he did not, or at least wrongly, envision. The limitations which made him disabled, invalid and unable to keep normal social relationships or professional activity are removed, but this doesn't mean that they can return to these activities, as the social environment may have changed, friends, sometimes the spouse, are gone. The patient's family has aged by the same amount of time as the duration of the disease. Similarly, the return to professional activity is most of the time difficult, if not impossible. As some patients express it, they are taken back by 10 years in their disease history, but they have not recuperated 10 years of life or of age. Moreover, due to the clinical improvement of the patients, they do not need the same support from caregivers or attention from the family. In other words, they totally lose their secondary benefits that are often significant. All this might be evaluated preoperatively and, at that time, explained to the patient. In the postoperative situation, this has to be attended to. The patient as well as the family may need psychotherapeutic support. This could be considered as a drawback of this surgery. In absolute terms, this is actually a drawback or a side effect. But this should not be used to compare STN stimulation to other methods and rank it below them. This is a new situation due to the fact that STN stimulation provides a much better improvement to the patient than other methods. 42.2.2. Electrode implantation Targeting is a prominent factor of success (Aziz et aI., 2001; Yokoyama et aI., 2001; Zhu et aI., 2002) There is a clear-cut correlation between the accuracy of placement of the electrode and the beneficial effects for the patient. Therefore all means must be used to achieve this goal with the best possible accuracy, using electrophysiology associated with intraoperative clinical evaluation, various neuroradiological methods including ventriculography,

700

magnetic resonance imaging (MRI), CT scan, navigation methods and software. Ventriculography was the basis for functional neurosurgery procedures until the advent of more modern and computer-based imaging modalities. The regression of its use is due to the necessity of a dedicated stereotactic room to perform ventriculography. A majority of teams use :a conventional surgical operating theatre. There is also the risk which is moderate of inducing bleeding due to the penetration of a cannulla into the ventricle through the brain. Intolerance to the positive contrast media has been also reported. Finally there is a debate about the possibility of brain shift due to the opening of the skull, introduction of air and positive contrast. This actually can be ruled out by careful methodology, which would avoid the leakage of cerebrospinal fluid. Brain shift could also happen when the surgical procedure and implantation of the electrode are done using a large opening of the skull, through which a significant amount of cerebrospinal fluid is lost. On the other hand, ventriculography allows the visualization of precise landmarks used in most atlases, such as the anterior and posterior commissures, midline and the floor of the lateral ventricle. There might be a magnification coefficient due to the parallax of the X-rays, that is maximal when one uses a C-arm fluoroscope and can be minimized when the patient is placed in a dedicated set up with long distance (3.5 meters) and biorthogonal X-ray tubes. This magnification coefficient varies from one side to the other side of the brain and this variation is also minimized by the long distance set-up. These arguments must be carefully discussed when one compares different methodologies. CT scan and magnetic resonance imaging (MRI) modalities have the advantage of being non invasive and can most of the time be performed outside of the operating room. The digital nature of the images helps in utilizing navigation softwares. This allows the matching of these different modalities to produce a combined picture of the brain. CT scan is reputed to have no spatial deformation due to the nature of the linear propagation of the X-rays and the absence of distortion of the beams. However the spatial resolution and moreover the density resolution are limited as compared to MRI and do not allow the direct visualization of some subcortical structures which are used currently as stereotactic targets.

A.L. BENABID ET AL.

The MRI on the other hand has a high density resolution, allowing very precise anatomical images of slices of the brain and the difference in densities which depends strongly on the pulse sequences that are used to visualize subcortical targets such as the internal Pallidum (GPi) and the subthalamic nucleus. The thalamus is visible as a structure delimited by the ventricle around it and the internal capsule laterally and although the quality of the imaging and the resolution is improving, it is not yet possible to discern subdivisions in the thalamus. The main drawback of MRI is the extreme sensitivity to differences in magnetic susceptibility induced by the body of the patient and of some particular constituents inducing inhomogeneities of the magnetic field. This has, as a general consequence, an imprecise spatial localization which is not easy to determine and varies from one patient to another. Therefore, this deformation inherent to the MRI principle makes it difficult to reach a mm or submm precision, particularly when it is used for postoperative control of electrode position (Saint-Cyr et al., 2002). This can be improved by using currently available methods of matching the MRI and the CT images into one common database and compensating the MRI distortions by the fusion with non-distorted CT images. This is only true in theory as actually most of the fusion software just achieves a least square error correspondence between the two sets of images, and it is rare that currently the software really reformating the image to correct the MRI distortion. There is no doubt that this will be achieved in the future, but at the moment, the problem of MRI deformations and localization insufficiencies coming from this modality are persistent and yet unsolved. However, the MRI allows a pretargeting that in most cases is considered by the users as acceptable, and it has the interesting capability to check the brain structures all along the track. The ventricle, and, at the cortical surface, the vessels can be avoided to decrease the risk of bleeding. 42.2.3. Associated surgical methods Electrode insertion is necessarily a stereotactic method. The anatomical targeting on the direct visualization of the target or on statistical coordinates does not have enough precisions. The final location of the chronic electrode requires additional

701

DEEP BRAIN STIMULATION IN PARKINSON'S DISEASE

functional identification, that is provided by electrophysiology, including microrecording of specific spontaneous and evoked neuronal activities (such as movement-related neurons of the STN). (Abosch et al., 2002). The most important data gathered during this functional step is the recognition of the sites where symptoms are alleviated by acute stimulation at the same parameters that will be used in the chronic situation. The beneficial effects as well as the side effects recorded during these steps allow the optimal placement of the electrode. The four contacts can be separately tested and activated using the programmable stimulator which is implanted in the subclavicular area.

42.2.4. Evaluation ofpatients and follow-ups Patients are followed-up after surgery according to the same criteria and using the same clinical tools as before surgery. In a first step, the follow-up will achieve the proper setting of the parameters of the stimulator. During VIM stimulation, the parameters can be set within a few hours and then readjusted within a few days, to block the tremor. Tuning of STN patients is more difficult to achieve, the drug dosage must be progressively reduced, simultaneously the voltage of the stimulation must be increased slowly, in order to avoid induction of dyskinesias. This is usually achieved over a period of a few weeks, and, depending on the teams, can be either done during the immediate postoperative period or undertaken later after the patient has returned at home for several weeks. This depends essentially on the characteristics of the local health systems and of the medical culture surrounding the medical and surgical teams. It has been now clearly established that two main factors are predominant in the postoperative setting, which are the contact and the stimulation parameters. This requires attention and time, but can be achieved following a rather simple procedure. There are not 64 combinations, as it may have been stated. Selection of the contacts: the monopolar situation is efficient. The easiest way to program the stimulation of a patient is to set the case positive and to scan the four contacts of the electrode, using 60 micros pulse width and 130 hertz frequency. It is therefore rather easy and fast to determine for each contact the average value of improvement (as compared to the off-medication off-stimulation situation), and the

main side effects induced by each contact. This usually leaves one of the four contacts as the best candidate. The knowledge of the distance of each contact to the theoretical functional target, as presented in this paper, may speed up this process by suggesting which would be the best contact as it must be either the nearest (or the immediate adjacent) contact (Fig. 1). It is useless to fiddle with pulse width and frequency as using the basic setting of 60 micros and 130 hertz will not allow missing any good results. On the contrary, changing them will be useful only when the optimal results are close to side effects (which means that the electrode is not optimally placed). Bipolar stimulation is used only to minimize side effects, always due to non-optimal electrode placement. Patients are then followed up at three months, one year, and every year, depending on the interest the team has in the long term evaluation of their results. It has been stated that DBS surgery is more time consuming for the doctors. This statement does not take into account that in other situations, such as medical treatment, most of the work is done by the neurologist close to the patient or by his general practitioner, both of whom at the moment are not yet trained to take care of stimulation parameters and tuning. In ablative surgery, there is no tuning. This is a benefit when the surgical result is satisfactory. When the result is not optimal and side effects and complications have occurred, this a drawback of ablative surgery, but cannot be considered as a drawback of stimulation. Actually, this is part of its positive aspect as side effects and complications are often voltage dependent and reversible. For the neurologists, who are taking care of the patient before and after surgery, it is clear that the follow-up is made easier by the stability of the effect which does not need from the patient to be calling his doctor as often as usual.

42.3. Results 42.3.1. Location of electrodes Electrodes form a cluster which projects onto the STN silhouette in atlases. Precision of localization cannot allow determination of what subset of neurons of STN are involved. Our data do not allow support or contradiction of the hypothesis that the functional target is the somato-motor part supposed

702

A.L. BENABID ET AL.

Fig. I. Euclidian distance of electrode contacts to the theoretical target. The theoretical target (TT), or theoretical functional target (TFf), is defined as the point at which coordinates are the average of the coordinates of the "best contacts" (on the basis of the clinical benefit obtained by stimulation of these contacts) in the whole patient population. The Euclidian distance is the spatial distance between two points (center of the contact and TT) in Cartesian coordinates. The table in the upper left comer shows the distances of the four contacts (named 0, 1,2,3) of the right and left electrodes. The distances are expressed in Stereotactic Units (SU, where the antero-posterior distance and the vertical distance are normalized vs. the bi-commissural length and the height of the Thalamus, respectively), or in mms vs. the.TFT of the general population (average TFf) or the TFf recalculated for each patient (Individual TFf). For each side, the shortest distance is enlightened with bold letter and in yellow, and correspond to the suggested contacts in the window below, as compared to the actual clinically based choice. One may see that in the present case the best contacts were the predicted contacts (closest to the target). The upper right comer diagram shows the graph of the distance (ordinates) vs. the contact number (abscissa), according to the various measures (Dark blue: SU, Red: average TFf. Pale blue: Individual TFf). The lower diagrams represent the schematic lateral (left diagram, posterior and anterior commissures at 0 and 12 respectively, top of the thalamus at 8) and antero-posterior (right diagram, laterality in mm from the midline) views of the electrode, vs. the average theoretical target (yellow-green dot and the Standard Deviation rectangles). Contacts 0 are the lower, contacts 3 the upper.

to be in the upper, lateral and anterior part of the STN. However our data are not coherent with the hypothesis that the functional target is the ZI, or the Forel's fields. Measurement for the electrode coordinates is done on the ventriculographic teleradiological images, in

a defined geometry and are corrected from magnification although the tube to film distance (3.5 m) reduces it to a minimal value of 1.05 (5%). The resulting coordinates are: STN coordinates: (200 bilaterally implanted patients) antero-posterior: 5.22±0.92 1/12° of

DEEP BRAIN STIMULATION IN PARKINSON'S DISEASE

AC-PC length, extremes: 2.88 to 7.08. Vertical: -1.18±0.67 1/8° of HT, extremes: -3.29 to 0.19. Laterality (corrected from X-ray magnification) 11.64± 1.15 mm from midline, extremes 9 to 15.2). Vim coordinates: antero-posterior: 3.53±0.91 1/12° of AC-PC length, extremes: 1.43-5.98. Vertical: 1.15± 1.18 1/8° of HT, extremes: -2.37-+4.26. Laterality 15.36± 1.61 mm from midline, extremes 12.27-19.22). To take care of the third ventricle (V3) changes in width (which affects directly the position of Vim which is lateral to V3 and not of STN which is below V3), one may use the "rule of Tasker" where the laterality L for Vim is L= 11.5 mm+(V3 width)l2. GPi coordinates: antero-posterior: 8.4± 1.2 1/12° of AC-PC length, extremes: 6.6-9.9. Vertical:0.7±0.8 1/8° of HT, extremes:-1.7-+0.3. Laterality 19.1±2.9mm from midline, extremes 16.0-23.2).

42.3.2. Beneficial effects Since the beginning of HFS DBS in Vim in 1987 (Benabid et a\., 1987) and in STN in 1993 (Limousin et a\., 1995a, 1998) the three major targets for PD surgical treatment have been evaluated by several teams (Limousin-Dowsey et a\., 1999). Vim-HFS is effective on parkinsonian rest tremor and on the associated unilateral pain (Benabid et a\., 1987, 1991, 1996). There was almost no change in bradykinesia or in any other symptom of Parkinson's disease. Immediately after surgery, a rnicrothalamotomy-like effect is responsible for transitory tremor suppression for a few days (22 (20.5%) out of 107 patients, 23 (15%) out of 153 electrodes). A very good result, (scores 3 and 4 on a 5 point scale) was obtained in 88% of cases with PD, 68% of cases with essential tremor (ET), and only in 18% of cases of tremor related to other causes. Resting tremor is better controlled than action tremor, distal limb tremor better than proximal or axial tremor, and upper limb better than lower limb tremor. In all cases the effect is strictly coincident with the stimulation, without significant delay at onset nor at cessation of stimulation. Thirty-nine out of 80 PD patients (48.7%) had their levadopa dosage decreased by 20% at 3 months follow-up. They were only 12 (15%) at the last follow-up, due to progression of the disease. Caparros-Lefebvre et al have reported much better results in five patients with levodopa induced interdose, choreic dyskinesias and ballistic dyski-

703

nesias which were clearly improved by thalamic stimulation. However, these patients were possibly implanted into the Centrum medianum-Parafascicularis complex (CM-Pf) either (Caparros-Lefebvre et a\., 1999a, b). Pallidal DBS is a safe and effective procedure for treatment of advanced PD (Kumar et a\., 2000) Bilateral DBS results in an improvement of about 34% in the activity of daily life (ADL) score during the off period and of about 35% to 40% improvement of the motor score for the off period. In addition, there are significant improvements in patients' symptoms during the on period and in onoff motor fluctuations. Compared to STN stimulation, the levodopa induced dyskinesias are directly diminished even when the drug dosage was not decreased. Actually, GPi stimulation does not allow a significant decrease in drug dosage (+ 10% at 6 months, -15% at 36 months) and may even have opposite effects depending on the sublocation within the nucleus. (Krack et aI., 1997b, 1998a, c) This method seems to be surpassed by the effects of STN stimulation which has progressively replaced it during the last few years (Krack et aI., 1998d; The Deep-Brain Stimulation for Parkinson's Disease Study Group, 2001). Subthalamus (i) Symptoms of the triad in PD: STN-HFS reduces rigidity, akinesia, and tremor (Krack et aI., 1997c, b. Limousin et aI., 1995a, b, 1998). At 12-month follow-up, bilateral STN stimulation greatly improves motor symptoms (UPDRS III -55%) and activities of daily living (UPDRS II --45%, Schwab and England Scale + 142%) in off-drug condition as well as dyskinesias in on-drug condition (-90%). Dopaminergic treatment (levodopa equivalent dose) was decreased by 50%. The continuous follow-up of these patients (48) shows an increasing improvement of about 60% of all symptoms evaluated on the corresponding scales, and a decrease in drug dosage of about 30 to 100% (mean 50%), which was responsible for the disappearance of levodopa induced dyskinesias (LIDs). Patients who had LIDs before surgery may exhibit similar complications when STN stimulation is on while they have their regular drug regimen. This diminishes with time, as long as drug dosages are

704

progressively decreased (57). Tremor is abolished by STN-HFS in a way comparable to Vim-HFS. As a general rule, it may be said that STN HFS provides the patient with a permanent level of improvement equal to his best status during on-periods, and that all DOPAsensitive parkinsonian symptoms were similarly improved. (ii) The improvement is visible also on midline symptoms: gait (Faist et al., 2001; Xie et al., 200 I; Ferrarin et al., 2002), arising, and stability. Speech (Gentil et al., 1999a, b, c. 2000, 200 I) is also improved although not as well as the others. Off-dystonia is strongly improved. When the stimulator is turned on, dystonia disappears within seconds, and reappears as quickly, when the stimulator is turned off again. It may even be efficient after a prior pallidotomy (Mogilner et aI., 2002). (iii) Dyskinesias: The improvement of the levodopa induced dyskinesias (LIDs) following STN stimulation is due to the strong improvement of akinesia and rigidity (Limousin et al., 1995a, b, 1998; Krack et al., 1997a, 1999; Fraix et al., 2000,2001), allowing a decrease by about 55% of the amount of drugs in those patients (Benabid et aI., 2000a; Fraix et aI., 2001; Kleiner-Fisman et aI., 2002; Moro et aI., 2002; Wenzelburger et al., 2002; Thobois et al., 2003). About 10% of our patients are drug free now. Actually, by over-stimulating the patient higher than needed to control the parkinsonian symptoms, one can induce dyskinesias (Levy et al., 2002), which could be described as a ballismus or choreoballic dyskinesias (Krack et aI., 1999). With time Levodopa challenges induce decreasing LIDs. Similarly, STN stimulation induced dyskinesias are more difficult to induce with time, that raises the question of post-synaptic desensitization as observed also during apomorphine long-term administration (Broussolle et al., 1992; de Saint Victor et al., 1992). (iv) Hypophonia: It must not be considered as a complication of surgery as it may respond to increased doses of levodopa or even to stimulation. However, in these patients, the reduction in voice volume may be disabling as the patient is sometimes barely understandable. The current hypothesis to try to explain this phenomenon

A.L. BENABID ET AL.

could be due to a somatotopic organization of the STN. The current functional method of targeting is actually based on the rigidity assessed by the passive mobilization of the wrist, which is a good indicator for the limbs but probably not for orofacial activity. It might be then that if the midline functions, such as the voice, are located in a different part of STN, we might consistently miss it according to the method employed. Therefore, the patient during chronic stimulation, being strongly improved as far as rigidity and akinesia of the limbs are concerned, is deprived from L-DOPA which is no longer useful and also participate to the induction of dyskinesias. As a consequence it might be that the patient, as far as the hypophonia is concerned, is not medically treated because of the significant reduction of drug doses and is not surgically treated because of the non-placement of the electrode into the STN area corresponding to voice control. This must be confirmed but it pushes us to develop intra-operative methods of voice exploration which would allow us to target better this symptom. (v) Cognitive functions modified by STN HFS and quality of life (QoL)? Neuropsychological testing has not shown any change. Long term bilateral subthalamic or pallidal stimulation for Parkinson's disease affects neither memory nor executive functions in consecutive series of patients (Ardouin et al., 1999; Jahanshahi et al., 2000; Pillon et al., 2000, 2002). Patients were only mildly depressed before surgery (Beck Depression Inventory 10.45 + 6.6) and there was a very mild but significant improvement of mood after surgery (Beck Depression Inventory 8.5 +4.1). No significant differences were found in on-medication condition for motor score, activities of daily living or mentation and behavior as assessed by UPDRS I. We had had 5 transient psychiatric complications (I mania, I paranoia, 3 depressions, including one suicide and two suicide attempts). The PDQL (Parkinson's disease Quality of Life) total score (maximum 185) improved from 90.3 to 129, parkinsonian symptoms (maximum 70) from 33.2 to 49.1, systemic symptoms (maximum 35) from 17.3 to 23.1, emotional functioning (maximum 45) from 24.2 to 31.2, social functioning (maximum

705

DEEP BRAIN STIMULATION IN PARKINSON'S DISEASE

35) from 15.7 to 25.6. The improvement of the score of the UPDRS III with bilateral STN stimulation is significantly correlated with the improvement of all aspects of the health-related QoL including emotional and social functioning. (Lagrange et al., Neurology, 2002). Although QoL improved, it did not normalize, but rather improved to the level of a large population of PD patients with less advanced disease. This finding is not surprising as bilateral STN stimulation only improves the motor symptoms in off-drug condition and dyskinesias in on-drug condition. On-period symptoms as well as cognitive symptoms show little or no improvement. Decrease of PD patient social isolation is the real success of STN stimulation. Therefore, in selected, highly levodopa-sensitive patients, suffering from motor complications of dopaminergic treatment, and without severe depression, the motor complications seem to be the main determinant of QoL (Just et al., 2002; Martinez-Martin et al., 2002). Side effects related to surgery, stimulation or changes in medication are likely to influence QoL. However, the overall study tends to confirm that the benefit of motor improvement largely prevails over the impact of side effects on QoL and that it is worth taking the relatively small risk, and operating on patients before they have reached a too low level of QoL.

42.3.3. Side effects: How to take advantage of them? Over stimulation may also be responsible for spreading of the current to immediately adjacent structures and for induction of side effects related to these abnormally involved structures. This could be strong depressive stages, or, on the contrary, irrepressible laughter, as we have observed in one patient. Actually, the type of the side effects is informative about the location of the electrodes, and may be also about the pathogenesis of symptoms or of function of surrounding structures. Obesity is a demonstrative example, as, in general, the patient's weight tends to increase by an average of 5 to 10 kgs. The cause of this effect is probably multifactorial, including improvement of the patient, change in metabolic rate parallel to dopamine effects, involvement of the zona incerta more than of the hypothalamus. However, similar findings have been reported in patients with GPi lesion or stimulation (Lang et al., 1997). The mechanisms

might be different, or the same if two different spots are respectively involved in a same network.

42.3.4. Complications: how to avoid them? 42.3.4.1. Related to procedure Functional neurosurgery is bound to create no further disability than the ones due to the disease. Procedures should be developed in order to decrease the rate of complication. Confusion: in about 20% of the cases, patients exhibit a confusional state which may last from few hours to several weeks and always resolves without sequellae. This happens more often in STN cases than in the patients implanted in other targets (Woods et al., 2002). This might be due to the fact that the patient selection is different, the STN cases being generally more severe and older than the Vim, purely trernoric, patients. This might be due also to the structures traversed during the approach of the target. Although more analysis of these cases must be performed, it could be possible that the bilateral exploring track, using five tubes, may lesion the caudate nuclei, laterally to the ventricle, and induce a temporary neurocognitive disturbance. Hemorrhages: all stereotactic procedures expose the patients to risk of bleeding, by hitting a vessel at the cortical surface, in the Fl-F2 or the mesial sulci where branches of the pericallosal artery are, along the ventricular wall surface, or in the parenchyma. Post-operative MRI or CT scans may show the presence of blood signal along the track, most often asymptomatic (15.6%) but sometimes responsible for permanent deficits (2.1 %). Preoperative MRI with Gadolinium contrast enhancement provides images of the vasculature allowing during the preplanning to see the vessels and to plan to avoid them. 42.3.4.2. Related to hardware DBS, as ventricular shunting, spinal osteosynthesis, cranial prosthesis, may produce complications related to the presence of foreign material in the body. The current materials (metal, silicone plastics, etc.) are well biocompatible and very rarely induce allergenic rejections. Most of the time, infection is due to intraoperative surgical contamination. This happens immediately and needs removal of the material, antibiotic treatment and reinsertion of the material removed after three months of total healing.

706 This did not happen in our series of 349 patients (137 Vim, 12 GPi, 200 STN, 616 sides). On the contrary, late secondary infections can happen several months or years later, due to skin erosions, induced by the bulk of the implanted parts, particularly at the level of the connector between the electrode and the extensions, under the scalp. This risk can be decreased or prevented when the material is inserted against the skull below the galea or the epicranium, and skin incisions are not crossing the wires or the implanted material. Nevertheless, this happened in 3.5% and needed removal of part of the material in half of the cases. Interestingly, in two cases, the infection reached the subcutaneous scalp and only needed cutting the leads at the level of the external skull, without removal of the intracranial part, which was replaced three months later. 42.3.4.3. Related to target Side effects are related to the location of the electrode vs. the target and its surrounding structures. Because of the low morbidity of DBS, lesions of these structures are rarely induced but wrong or non-optimal position of the electrode may induce side effects which may be limiting factors in the optimal tuning of the stimulation and therefore constitute a sort of complication. The best way to avoid them is to pay the highest attention to optimal positioning of the electrodes. Vim has two major neighbors, the somatosensory thalamus VPL and the internal capsule IC. VPL induces intensity-related paresthesias if the electrode is too posterior. This is why we finally targeted Vim on its anterior border to decrease the current spread to the posterior boundary of Vim which is adjacent to VPL. IC induces muscular contractions by current spreading to the pyramidal fibers. Intraoperative microstimulation may predict it and allow placement of the electrode more medially. GPi is anatomically close to the IC and to the optic tract which runs below and might induce visual flashes. This can be also predicted during surgery but usually selection of the immediate upper contact avoids this side effect. STN is a wonderful target but is inconveniently placed. The IC, including fibers of the cortico nuclear tract (inducing facial contractions and conjugated ocular deviations) is lateral, the lemniscus medialis (inducing paresthesias) is posterior, the fibers of the third nerve (inducing monocular deviations) are inferior. Fortunately, when the electrode is in the core of

A.L. BENABID ET AL.

STN, the thresholds of side effects are always higher than those of the beneficial effects. However, STN has specific side effects when its stimulation is too strong, dyskinesias are induced which are of the same nature as the hemiballism secondary to STN hemorrhagic lesions. Stimulation parameters have to be progressively increased at the same time that dopamine treatment is decreased as their effects are synergistic (Krack et al., 1997a). 42.3.4.4. Related to surgery in general Functional neurosurgery must not submit the patient to risks higher than the disability created by the disease. As movement disorders are not lethal diseases, surgery must be contraindicated if patients have major risks such as heart disease, blood coagulation problems, kidney or liver insufficiencies, in general health condition not compatible with major procedures, particularly because of the long duration of the implanting session (Lopiano et aI., 2002). 42.4. Discussion

42.4.1. Faster or better? "Faster and better" should be preferred, or at least "Faster" could be acceptable as long as the best results obtained with other (longer?) methods are still preserved. 42.4.1.1. Optimize precision Optimization of precision is the key goal to be reached as it is directly linked to the quality of the beneficial effect. A best compromise has to be reached between: (i) Accumulation of tools and methods to improve precision (multi modal imaging, extensive intraoperative stimulation and microrecording, number of electrodes, time spent during these explorations, maneuvers and tests performed during surgery); and (ii) risks specific to each step, costs related to techniques used, time spent, patient fatigue, etc. 42.4.1.2. Optimize speed Speed, as usual in neurosurgery, has to be as fast as possible for obvious reasons, including duration of surgery for the patient, time related complications, cost of human resources, availability of facilities in the OR. But speed is not an objective

707

DEEP BRAIN STIMULATION IN PARKINSON'S DISEASE

"per se", it is justified as long as it does not hurt the patient and as long the final result is maintained,

42.4.1.3. Preserve quality Quality of the result should be the only goal of the surgery. It cannot be sacrificed to time or to money. If choice has to be made between two methods, the one which provides the best results is the one to be preferred, if two methods offer similar outcomes, then the fastest must be chosen. The result provided to a patient will stay for years and cannot be jeopardized for the sake of a few hours spared in the OR. Questions are raised about how much the global burden of a procedure can be increased for research reason. The answer is in the agreement of the patient, in the context of controlled studies under IRB approval, which itself depends on how valid and useful are the scientific goals aimed at by the intraoperative scientific approach.

the last decades makes us oblivious of the reasons why this surgery disappeared, besides the advent of levodopa. These reasons are exactly the opposite of those which make the success of Deep Brain HFS. permanent against reversible side effects, restricted unilateral against usual bilateral procedures, rather high against rather low morbidity, invariability of effects (bad or good) against adaptability (of side effects or of decreasing beneficial effects). Same causes provide same effects: the rare comparative studies (Schuurman et al., 2000) have demonstrated better results of Vim DBS as compared to Vim thalamotomy for PD tremor. More frequent reports trying lesions in place of DBS are confirming that the expected complications are indeed reobserved (Parkin et al., 2001; Vilela Filho et al., 2001; Doshi et al., 2002). Why should we lose the benefit of flexibility?

42.4.2. Cost

42.4.4. Mechanism of STN stimulation

The equipment is expensive, and has to be replaced. This is true for every kind of prosthetic material, including pacemakers, orthopedic plates and screws, stents and vascular synthetic sleeves. The cost has to be justified first by the benefit provided by its use, which should be better than with competiting methods. This is the case for DBS in movement disorders: (i) The beneficial effect saves drugs and then money. (ii) Additional benefits come from saving care givers, equipments, helpers. (iii) The decrease in suffering is not quantitatively assessable but is extremely valuable and must be taken into account (Benabid and Pollak, 2000).

Although this is not the main subject of this chapter, some points must be outlined.

42.4.3. Stimulation or lesion? History teaches lessons which are usually forgotten. The success of DBS in movement disorders is responsible for the spreading of this method, as visible through the number of operated patients, of published scientific papers and on the increasing percentage of posters dealing with the subject in specialized meetings. However, the cost and the time spent to achieve an optimal result raises concerns at various levels (consequences on neurosurgical private practice, duration of hospitalization, availability of the hardware in developing countries) The long period of silence through which functional neurosurgery of movement disorders went along during

42.4.4.1. Highfrequency is the key What has made DBS available for movement disorder treatment, and possibly for other indications, is the observation that effects similar, but reversibly, to those obtained by ablative methods were obtained when frequency was higher than 100 hertz, up to 2500 hertz. Although precise studies of the curve frequency-response have still to be done, the frequency cut-off threshold for the appearance of a significant effect may vary slightly from one target to another, but high frequency within the same range is a common denominator. The other parameters, such as the pulse width and the waveform, seem to be less critical, although square pulses are probably not the best waveforms, but they are easy to produce, to understand and to analyze. Other waveforms must be evaluated which could produce better results. 42.4.4.2. What is the real functional target? Currently, three different nuclei are used as stereotactic targets: Vim in the Thalamus, GPi in the Pallidum and STN in the subthalamic area. For each of them, one must know which part of the nucleus is particularly responsible of the effect, or, in other words, what are the specific functions of the subcomponents of each nucleus. Vim is obviously

708

the efficient part of the motor thalamus but the role of Voa and Vop are not ruled out. Similarly, in GPi, the most efficient part is postero-Iateral and inferior, but opposite effects may be observed at very short distances (Krack et aI., I997b, 1998a, c). STN in rodents and non-human primates (and possibly by extension in human) has three subparts: somatomotor, associative and limbic, which might be functionally different, but clear evidence is still lacking. Recent data would tend to suggest that the upper lateral and anterior part of STN might be where the antiparkinsonian effects are achieved as a rule. However, because of this rather high localization of the putative functional target, some teams have reported data suggesting that the adjacent zona incerta or the Forel's fields might be involved in the process. Our data do not support this hypothesis but the clear demonstration of the real target is made difficult by the lack of precision of the imaging methods, for reasons developed above. Autopsy material which will necessarily be available within the next years will help solving this important question. Similarly, as suggested by the current concepts of the organization of the basal ganglia (Alexander and Crutcher, 1990), Substantia nigra pars reticulata (SNr) could be a potential target as it seems to have a similar role as has GPi. It seems, however, that there is no controversy on the subject, as no beneficial clinical effect has been observed in patients in whom lower contacts situated below STN in SNr were stimulated, allowing to consider so far that, functionally speaking, "SNr is not GPi". However, within the STN area, the location determinates the results. Measurement for each patient of the distance of each contact of the electrode to the theoretical target provides predictability of 70% for the closest contact and of 85% for the closest or its next neighbor. 42.4.4.2.1. What are the cellular structures involved by HFS? This is currently the subject of a hot debate among the laboratories working on this problem. The observation that excitation, which usually corresponds to stimulation, might end up in a phenomenon mimicking inhibition is difficult to reconcile with the classical neurophysiologic concepts. One must take into account the effects of electrical pulses on each component of a neuron: Cell bodies? Axons? Neurons as isolated units or as

A.L. BENABID ET AL.

parts of networks? The effects of HFS are qualitatively comparable to those of lesions in all targets so far stimulated. Even time courses are similar: the improvement of dystonia after surgery of the pallidum is delayed, progressive, and continuing, for lesions as well as for HFS. This is not true for fibers: when HFS is delivered in an axonal structure, such as the pyramidal tract, the optic tract, the lemniscus medialis, the fibers of the third cranial nerve, the observed effects are consistently those induced by stimulation, such as, respectively, motor contractions, visual flashes, paresthesias, ocular deviations. 42.4.4.3. Experimental data They are too controversial and quickly changing to be summarized in this chapter (Benazzouz et aI., 1995, 2000a; Gao et al., 1996, 1999; Benazzouz and Hallett, 2000;. Ni et al., 2000). It can be expected that this problem will be solved within the next two years, demonstrating one or several mechanisms involved, a possible cellular mechanism in Parkinson's disease such as silencing of subthalamic neurons by high-frequency stimulation (MagarinosAscone et al., 2002), or composite effects of excitation and/or inhibition (Vitek et al., 2002), or for instance, regulation of the timing and pattern of action potential generation by GABA-A IPSPs (BeVan et aI., 2002). 42.4.5. Specific issues 42.4.5.1. Neuroprotection hypothesis The hypothesis that neuroprotection or plasticity could be induced by long term HFS of STN is suggested by experimental data showing that a subthalamic nucleus lesion in rats prevents dopaminergic nigral neuron degeneration after striatal 6-0HDA injection (Piallat et aI., 1996, 1999), that high-frequency stimulation of the subthalamic nucleus selectively reverses dopamine denervationinduced cellular defects in the output structures of the basal ganglia in the rat (Salin et al., 2002), and that 'short-term plasticity shapes the response to simulated normal and parkinsonian input patterns in the globus pallidum (Hanson et al., 2002 ). Clinical data on long term followed-up patients are either suggestive or at least in agreement with this hypothesis, but there is no clear evidence so far. PET and SPECT studies are to be done.

709

DEEP BRAIN STIMULATION IN PARKINSON'S DISEASE

42.4.5.2. Depression and suicide, Mood and Behavioral changes This is typically a multifactorial problem. The reaction of the patient to the newly created situation coming from the beneficial effects of HFS depend on pre-existing depression, social context, drug regimen changes (Doshi PK et aI., 2002). This should be compared to similar situations where changes are dramatic and suicides are observed, such as epilepsy surgery or cardiac surgery. STN is a very efficient target but is situated within a very rich and eloquent neighborhood. Spreading of the current to the surrounding structures, within a few mms, can therefore be responsible for a large spectrum of effects, which are currently being reported, such as transient mania with hypersexuality after high frequency stimulation of STN (Romito et al., 2002), mirthful laughter (Krack et al., 2001), aggressive behavior induced by intraoperative stimulation in the triangle of Sano (Bejjani et aI., 2002), and behavioral disorders (Houeto et al., 2002). 42.4.5.3. New indications outside movement disorders Due to the non disease-specific effects of HFS, it was legitimate to take advantage of it and to try to apply this new surgical method to other fields of functional neurosurgery. This is being currently made in the following cases. Epilepsy: antiepileptic effect of high-frequency stimulation of the subthalamic nucleus has been shown in cases of medically intractable epilepsy, such as those caused by focal dysplasia in eloquent areas, with a significant follow-up (Benabid et al., 2002). OCD: in order to substitute HFS for lesion in already practiced fields, results have been reported of successful HFS of the anterior limb of the internal capsule in severe cases of OCD. Chronic electrical stimulation instead of bilateral capsulotomy was done in four selected patients with long-standing treatment-resistant obsessive-compulsive disorder. In three of them beneficial effects were observed (Nuttin et al., 1999). The question of the best target is again raised and the Nucleus Accumbens is currently evaluated. Recent reports draw attention on the STN where PD patients have had their OCD symptoms relieved as well as the concomitant PD symptoms for which they had been implanted

(Mallet et aI., 2002). This effect could be mediated by the involvement of the limbic part of STN and can find support in experimental PET study data showing that STN stimulation affects striato-anterior cingulate cortex circuits in a response conflict task (Schroeder et aI., 2002).

42.5. Conclusion HFS applied to DBS has created a new situation for functional neurosurgery. Besides the large number of human applications to various neurological diseases, this triggers exciting research, opening new avenues. Concerning Parkinson's disease, STN is the best PD target, improves all motor symptoms and has replaced Vim for the treatment of tremor, and GPi for the treatment of dyskinesias as STN allows the decrease of drug dosage. STN has demonstrated stability of the benefit and has not revealed long term side effects. Earlier indications should be considered on this basis: why should we wait for the patient to lose autonomy, job, spouse and friends, dignity, and to be bed ridden. Long term stimulation might be neuroprotective but this has to be ruled out within the next years. Concerning the health care problem raised by the cost of the method, studies are being done proving that it is even money saving. Further technical developments can be expected which will improve the usage of the method: stimulators should be of smaller size, flatter, with shorter wires. Multiple electrodes could be used to better cover the targets. Bioenergetic batteries could be designed, teletuning and telechecking will help the management of patients living far away from the medical centers. Molecular approaches and electroporation transfer gene methods (Luo et aI., 2002) might introduce revolutionary perspectives for further therapeutical improvements (Benabid et al., 2000b).

References Abosch, A et al. (2002) Movement-related neurons of the subthalamic nucleus in patients with Parkinson disease. J. Neurosurg., 97: II 67-1172. Alexander, GE and Crutcher, MD (1990) Functional architecture of basal ganglia circuits: neural substrates of parallel processing. Trends Neurosci., 13: 266-271. Ardouin, C, Pillon, B, Peiffer, E, Bejjani, P, LimousinDowsey, P, Damier, P, Amulf, I, Benabid, AL, Agid, Y and Pollak, P (1999) Bilateral subthalamic or pallidal stimulation for Parkinson's disease affects neither

710 memory nor executive functions, a consecutive series of 62 patients. Ann. Neurol., 46: 217-223. Aziz, TZ et al. (2001) Targeting the subthalamic nucleus. Stereotact. Funct. Neurosurg., 77: 87-90. Bejjani, BP et al. (2002) Aggressive behavior induced by intraoperative stimulation in the triangle of Sano. Neurology, 59: 1425-1427. Benabid, AL and Pollak, P (2000) Evaluation des cofits de la neurochirurgie fonctionnelle des mouvements anormaux par la stimulation des noyaux gris centraux, en particulier du noyau sub-thalamique (NST). Rev. Neurol. (Paris), 156 (Suppl. 2): 270-273. Benabid, AL, Pollak, P, Louveau, A, Henry, S and de Rougemont, J (1987) Combined (thalamotomy and stimulation) stereotactic surgery of the VIM thalamic nucleus for bilateral Parkinson disease. Appl. Neurophysiol., 50: 344-346. Benabid, AL, Pollak, P, Gervason, C, Hoffmann, D, Gao, DM, Hommel, M, Perret, JE and de Rougemont, J (1991) Long-term suppression of tremor by chronic stimulation of the ventral intermediate thalamic nucleus. Lancet, 337: 403-406. Benabid, AL, Pollak, P, Gao, D, Hoffmann, D, Limousin, P, Gay, E, Payen, I and Benazzouz, A (1996) Chronic electrical stimulation of the ventralis intermedius nucleus of the thalamus as a treatment of movement disorders. J. Neurosurg., 84: 203-214. Benabid, AL, Benazzouz, A, Limousin, P, Koudsie, A, Krack, P, Piallat, B and Pollak, P (2000a) Dyskinesias and the subthalamic nucleus. Ann. Neurol., 47 (Suppl. 1): SI89-192. Benabid, AL, Koudsie, A, Pollak, P, Kahane, P, Chabardes, S, Hirsch, E, Marescaux, C and Benazzouz, A (2000b) Future prospects of brain stimulation. Neurol. Res., 22: 237-246. Benabid, AL et aI. (2002) Antiepileptic effect of highfrequency stimulation of the subthalamic nucleus (corpus luysi) in a case of medically intractable epilepsy caused by focal dysplasia, a 30-month followup, technical case report. Neurosurgery, 50: 13851391. Benazzouz, A, Piallat, B, Pollak, P and Benabid, AL (1995) Responses of substantia nigra pars reticulata and globus pallidus complex to high frequency stimulation of the subthalamic nucleus in rats, electrophysiological data. Neurosci. Lett., 189: 77-80. Benazzouz, A, Gao, DM, Ni, ZG, Piallat, B, BoualiBenazzouz, R and Benabid, AL (2000a) Effect of high-frequency stimulation of the subthalamic nucleus on the neuronal activities of the substantia nigra pars reticulata and ventrolateral nucleus of the thalamus in the rat. Neuroscience, 99: 289-295.

A.L. BENABID ET AL.

Benazzouz, A and Hallett, M (2000b) Mechanism of action of deep brain stimulation. Neurology, 55 (12 SuppI6): S13-16. Bevan, MD et al. (2002) Regulation of the timing and pattern of action potential generation in rat subthalamic neurons in vitro by GABA-A IPSPs. J. Neurophysiol., 87(3): 1348-1362. Broussolle, E, Marion, MH and Pollak, P (1992) Continuous subcutaneous apomorphine as replacement for levodopa in severe parkinsonian patients after surgery. Lancet, 340(8823): 859-860. Caparros-Lefebvre, D, Pollak, P, Fellin, MP, Blond, S and Benabid, AL (1999a) Effet de la stimulation thalamique sur les dyskinesies induites par la L-DOPA Evaluation d'un cible nouvelle, Ie center median parafasciculaire. Rev. Neurol. (Paris), 155: 543-550. Caparros-Lefebvre, D, Blond, S, Feltin, MP, Pollak, P and Benabid, AL (1999b) Improvement of levodopa induced dyskinesias by thalamic deep brain stimulation is related to slight variation in electrode placement, possible involvement of the center median and parafascicularis complex. J. Neurol. Neurosurg. Psychiatry, 67: 308-314. Charles, PD, Van Blercom, N, Krack, P, Lee, SL, Xie, J, Besson, G, Benabid, AL and Pollak, P (2002) Predictors of effective bilateral subthalamic nucleus stimulation for PD. Neurology, 59: 932-934. De Saint Victor, JF, Pollak, P, Gervason, CL and Perret, J (1992) Levodopa-induced diphasic dyskinesias improved by subcutaneous apomorphine. Mov. Disord., 7: 283-284. Doshi, PK et al. (2002a) Depression leading to attempted suicide after bilateral subthalamic nucleus stimulation for Parkinson's disease. Mov. Disord., 17(5): 10841085. Doshi, PK et al. (2002b) Hemiballism during subthalamic nucleus lesioning. Mov. Disord., 17(4): 848-849. Faist, M, Xie, J, Kurz, D, Berger, W, Maurer, C, Pollak, P and LUcking,CH (2001) Effect of bilateral subthalamic nucleus stimulation on gait in Parkinson's disease. Brain, 124(8): 1590-1600. Ferrarin, M et al. (2002) Quantitative analysis of gait in Parkinson's disease, a pilot study on the effects of bilateral sub-thalamic stimulation. Gait Posture, 16(2): 135-148. Fraix, V, Pollak, P, Van Blercom, N, Xie, J, Krack, P, Koudsie, A and Benabid, AL (2000) Effect of subthalamic nucleus stimulation on levodopa-induced dyskinesia in Parkinson's disease. Neurology, 55(12): 1921-1923. Fraix, V, Pollak"P, Van Blercom, N, Xie, J, Krack, P, Koudsie, A and Benabid, AL (2001) Effect of subthalamic nucleus stimulation on levodopa-induced

DEEP BRAIN STIMULATION IN PARKINSON'S DISEASE

dyskinesia in Parkinson's disease. Neurology, 57(10 Suppl 3): S60-S62. Gao, DM, Hoffmann, D and Benabid, AL (1996) Simultaneous recording of spontaneous activities and nociceptive responses from neurons in the pars compacta of substantia nigra and in the lateral habenula. Eur. J. Neurosci., 8(7): 1474-1478. Gao, DM, Benazzouz, A, Piallat, B, Bressand, K, Ilinsky, lA, Kultas-Ilinsky, K and Benabid, AL (1999) Highfrequency stimulation of the subthalamic nucleus suppresses experimental resting tremor in the monkey. Neuroscience, 88(1): 201-212. Gentil, M, Garcia-Ruiz, P, Pollak, P and Benabid, AL (1999a) Effect of stimulation of the subthalamic nucleus on oral control of patients with parkinsonism. J. Neurol. Neurosurg. Psychiatry, 67: 329-333. Gentil, M, Perrin, S, Toumier, CL and Pollak, P (1999b) Lip, tongue and forefinger force control in Parkinson's disease. Clin. Ling. Phon., 13: 45-54. Gentil, M, Toumier, CL, Pollak, P and Benabid, AL (1999c) Effect of bilateral subthalamic nucleus stimulation and dopatherapy on oral control in Parkinson's Disease. Eur. Neurol., 42: 136--140. Gentil, M, Garcia-Ruiz, P, Pollak, P and Benabid, AL (2000) Effect of bilateral deep-brain stimulation on oral control of patients with parkinsonism. Eur. Neurol.,44: 147-152. Gentil, M, Chauvin, P, Pinto, S, Pollak, P and Benabid, AL (2001) Effect of bilateral stimulation of the subthalamic nucleus on parkinsonian voice. Brain Lang., 78: 233-240. Hanson, JE et a!. (2002) Short-term plasticity shapes the response to simulated normal and parkinsonian input patterns in the globus pallidus. J. Neurosci., 22(12): 5164-5172. Hilker, R et a!. (2002) Deep brain stimulation of the subthalamic nucleus vs. levodopa challenge in Parkinson's disease, measuring the on- and off-conditions with FDG-PET. J. Neural. Transm., 109(10): 12571264. Houeto, JL et a!. (2002) Behavioural disorders, Parkinson's disease and subthalamic stimulation. J. Neurol. Neurosurg. Psychiatry, 72(6): 701-707. Jahanshahi, M, Ardouin, CM, Brown, RG, Rothwell, JC, Obeso, J, Albanese, A, Rodriguez-Oroz, MC, Moro, E, Benabid, AL, Pollak, P and Limousin-Dowsey, P (2000) The impact of deep brain stimulation on executive function in Parkinson's disease. Brain, 123(6): 1142-1154. Just, H et a!. (2002) Health-related quality of life in patients with advanced Parkinson's disease treated with deep brain stimulation of the subthalamic nuclei. Mov. Disord., 17(3): 539-545.

711 Kleiner-Fisman, G et a!. (2002) Subthalamic DBS replaces levodopa in Parkinson's disease. Neurology, 59(8): 1293-1294. Krack, P, Limousin, P, Benabid, AL and Pollak, P (1997a) Chronic stimulation of subthalamic nucleus improves levodopa-induced dyskinesias in Parkinson's disease. Lancet, 350: 1676. Krack, P, Pollak, P, Limousin, P and Benabid, AL (1997b) Levodopa-inhibiting effect of pallidal surgery. Ann. Neurol.,42: 129-130. Krack, P, Pollak, P, Limousin, P, Benazzouz, A and Benabid, AL (1997c) Stimulation of subthalamic nucleus alleviates tremor in Parkinson's disease. Lancet, 350: 1675. Krack, P, Pollak, P, Limousin, P, Hoffmann, D, Benazzouz, A and Benabid, AL (1998a) Inhibition of levodopa effects by internal pallidal stimulation. Mov. Disord., 13: 648-652. Krack, P, Benazzouz, A, Pollak, P, Limousin, P, Piallat, B, Hoffmann, D, Xie, J and Benabid, AL (1998b) Treatment of tremor in Parkinson's disease by subthalamic nucleus stimulation. Mov. Disord., 13: 907914. Krack, P, Pollak, P, Limousin, P, Hoffmann, D, Benazzouz, A, Le Bas, JF, Koudsie, A and Benabid, AL (1998c) Opposite motor effects of pallidal stimulation in Parkinson's disease. Ann. Neurol., 43: 180-192. Krack, P, Pollak, P, Limousin, P, Hoffmann, D, Xie, J, Benazzouz, A and Benabid, AL (1998d) Subthalamic nucleus or internal pallidal stimulation in young onset Parkinson's disease. Brain, 451-457. Krack, P, Pollak, P, Limousin, P, Benazzouz, A, Deuschl, G and Benabid, AL (1999) From off-period dystonia to peak-dose chorea. The clinical spectrum of varying subthalamic nucleus activity. Brain, 122: 1133-1146. Krack, P, Limousin-Dowsey, P, Benabid, AL, Acarin, N, Benazzouz, A, Kunig, G, Leenders, KL, Obeso, JA and Pollak, P (2000) Ineffective subthalamic nucleus stimulation in levodopa-resistant postischemic parkinsonism. Neurology, 54: 2182-2184. Krack, P, Kumar, R, Ardouin, C, Dowsey, P, McVicker, JM, Benabid, AL and Pollak, P (200 1) Mirthful laughter induced by subthalamic nucleus stimulation. Mov. Disord., 16: 867-875. Kumar, R, Lang, AE, Rodriguez-Oroz, MC, Lozano, AM, Limousin, P, Pollak, P, Benabid, AL, Guridi, J, Ramos, E, Van der Linden, C, Vandewalle, A, Caemaert, J, Lannoo, E, Van den Abbeele, D, Vingerhoets, G, Wolters, M and Obeso, JA (2000) Deep brain stimulation of the globus pallidus pars intema in advanced Parkinson's disease. Neurology, 55 (Supp!. 6): S3439.

712 Lagrange, E et al. (2002) Bilateral subthalamic nucleus stimulation improves health-related quality of life in PD. Neurology, 59(12): 1976-1978. Lang, AB, Lozano, A, Tasker, R, Duff, J, Saint-Cyr, J and Trepanier, L (1997) Neuropsychological and behavioural changes and weight gain after medial pallidotomy. Ann. Neurol., 4: 834-836. Levy, R et al. (2002) Simultaneous repetitive movements following pallidotomy or subthalamic deep brain stimulation in patients with Parkinson's disease. Exp. Brain Res., 147(3): 322-331. Limousin, P, Pollak, P, Benazzouz, A et al. (1995a) Effect on parkinsonian signs and symptoms of bilateral subthalamic nucleus stimulation. Lancet, 345, 91-95. Limousin, P, Pollak, P, Benazzouz, A et al. (1995b) Bilateral subthalamic nucleus stimulation for severe Parkinson's disease. Mov. Disord., 10: 672-674. Limousin, P, Krack, P, Pollak, P, Benazzouz, A, Ardouin, C, Hoffmann, D and Benabid, AL (1998) Electrical stimulation of the subthalamic nucleus in advanced Parkinson's disease. N. Eng. J. Med., 339: 1105-1111. Limousin-Dowsey, P, Pollak, P, Van Blercom, N, Krack, P, Benazzouz, A and Benabid, AL (1999) Thalamic, subthalamic nucleus and internal pallidum stimulation in Parkinson's disease. J. Neurol., 246 (Suppl. 2): III 42-II/45. Lopiano, L et al. (2002) Deep brain stimulation of the subthalamic nucleus in PD, an analysis of the exclusion causes. J. Neural. Sci., 195(2): 167-170. Luo, J et al. (2002) Subthalamic GAD gene therapy in a Parkinson's disease rat model. Science, 298(5592): 425-429. Magarinos-Ascone, C et al. (2002) High-frequency stimulation of the subthalamic nucleus silences subthalamic neurons, a possible cellular mechanism in Parkinson's disease. Neuroscience, 115(4): 1109-1117. Mallet, L, Mesnage, V, Houeto, JL, Pelissolo, A, Yelnik, J, Behar, C, Gargiulo, M, Welter, ML, Bonnet, AM, Pillon, B, Cornu, P, Dormont, D, Pidoux, B, Allilaire, JF and Agid, Y (2002) Compulsions, Parkinson's disease, and stimulation. Lancet, 360: 1302-1304. Martinez-Martin, P et al. (2002) Bilateral subthalamic nucleus stimulation and quality of life in advanced Parkinson's disease. Mov. Disord., 17: 372-377. Mogilner, AY et al. (2002) Subthalamic nucleus stimulation in patients with a prior pallidotomy. J. Neurosurg., 96(4): 660-665. Moro, E et al. (2002) Response to levodopa in parkinsonian patients with bilateral subthalamic nucleus stimulation. Brain, 125(11): 2408-2417. Ni, ZG, Gao, DM, Benabid, AL and Benazzouz, A (2000) Unilateral lesion of the nigrostriatal pathway induces a transient decrease of firing rate with no change in the

A.L. BENABID ET AL.

firing pattern of neurons of the parafascicularis nucleus in the rat. Neuroscience, 101(4): 993-999. Nuttin, B, Cosyns, P, Demeulemeester, H, Gybels, J and Meyerson, B (1999) Electrical stimulation in anterior limbs of internal capsules in patients with obsessivecompulsive disorder. Lancet, 354: 1526. Parkin, S et al. (2001) Lesioning the subthalamic nucleus in the treatment of Parkinson's disease. Stereotact. Funct. Neurosurg., 77(1-4): 68-72. Piallat, B, Benazzouz, A and Benabid, AL (1996) Subthalamic nucleus lesion in rats prevents dopaminergic nigral neuron degeneration after striatal 6-0HDA injection, behavioural and immuno-histochemical studies. Eur. J. Neurosci., 8(7): 1408-1414. Piallat, B, Benazzouz, A and Benabid, AL (1999) Neuroprotective effect of chronic inactivation of the subthalamic nucleus in a rat model of Parkinson's disease. J. Neural. Transm. 55 (Suppl.): 71-77. Pillon, B (2002) Neuropsychological assessment for management of patients with deep brain stimulation. Mov. Disord., 17 (Suppl. 3): SlI6-122. Pillon, B, Ardouin, C, Damier, P, Krack, P, Houeto, JL, Klinger, H, Bonnet, AM, Pollak, P, Benabid, AL and Agid, Y (2000) Neuropsychological changes between "off' and "on" STN or GPi stimulation in Parkinson's disease. Neurology, 55: 411-418. Romito, LM et al. (2002) Transient mania with hypersexuality after surgery for high frequency stimulation of the subthalamic nucleus in Parkinson's disease. Mov. Disord., 17: 1371-1374. Saint-Cyr, JA et al. (2002) Localization of clinically effective stimulating electrodes in the human subthalamic nucleus on magnetic resonance imaging. J. Neurosurg., 97(5): 1152-1166. Salin, P et al. (2002) High-frequency stimulation of the subthalamic nucleus selectively reverses dopamine denervation-induced cellular defects in the output structures of the basal ganglia in the rat. J. Neurosci., 22(12): 5137-5148. Schroeder, U et al. (2002) Subthalamic nucleus stimulation affects striato-anterior cingulate cortex circuit in a response conflict task, a PET study. Brain, 125: 1995-2004. Schuurman, PR, Bosch, DA, Bossuyt, PM, Bonsel, GJ, Van Someren, EJ, de Bie, RM, Merkus, MP and Speelman, JD (2000) A comparison of continuous thalamic stimulation and thalamotomy for suppression of severe tremor. N. Eng. J. Med., 342: 461-468. The Deep-Brain Stimulation for Parkinson's Disease Study Group (2001) Deep brain stimulation of the subthalamic nucleus or the pars interna of the globus

DEEP BRAIN STIMULATION IN PARKINSON'S DISEASE

pallidus in Parkinson's disease. N. Eng. J. Med., 345(13): 956-963. Thobois, S et al. (2003) The timing of antiparkinsonian treatment reduction after subthalamic nucleus stimulation. Eur. Neurol., 49(1): 59-63. Vitek, JL (2002) Mechanisms of deep brain stimulation, excitation or inhibition. Mov. Disord., 17 (Suppl. 3): S69-72. Vilela Filho, 0 et al. (2001) Stereotactic subthalamic nucleus lesioning for the treatment of Parkinson's disease. Stereotact. Funct. Neurosurg., 77: 79-86. Wenzelburger, R et al. (2002) Dyskinesias and grip control in Parkinson's disease are normalized by chronic stimulation of the subthalamic nucleus. Ann. Neurol., 52(2): 240-243.

713 Woods, SP et al. (2002) Neuropsychological sequellaes of subthalamic nucleus deep brain stimulation in Parkinson's disease, a critical review. Neuropsychol. Rev., 12(2): 111-126. Xie, J, Krack, P, Benabid, AL and Pollak, P (2001) Effect of bilateral subthalamic nucleus stimulation on parkinsonian gait. J. Neurol., 248(12): 1068-1072. Yokoyama, T et al. (2001) The optimal stimulation site for chronic stimulation of the subthalamic nucleus in Parkinson's disease. Stereotact. Funet. Neurosurg., 77(1--4): 61-67. Zhu, XL et al. (2002) Magnetic resonance imaging-based morphometry and landmark correlation of basal ganglia nuclei. Acta Neurochir. (Wien.), 144(10): 959969.

Movement Disorders Handbook of Clinical Neurophysiology, Vol. 1 M. Hallett (Ed.) © 2003 Elsevier B.Y. All rights reserved

717

CHAPTER 43

Research studies in normal subjects and patients: current and future J.C. Rothwell* Sobell Department, Institute of Neurology, Queen Square, London WCIN 3BG, UK

43.1. Introduction Clinical neurophysiology was born in the first half of the 20th century following the development of tools to study EMG and EEG by pioneers such as Adrian and Matthews. Advances in manufacture of stable valve amplifiers led to its increasing popularity and after the second world war, application of the techniques in clinical settings became routine. The development of averaging by Dawson in the 1950s and the arrival of cheap computing power in the 1970s brought the technique the popularity that it still has today. For most of its past, clinical neurophysiology has been a descriptive science devoted primarily to investigating the physiology of the normal and damaged central nervous system. In the field of movement disorders, this has involved using reflexes and nerve conduction studies to test connectivity and explore nervous pathways. In addition, investigators have used EEG and MEG to quantify patterns of brain activation in different tasks. This type of work in the 1970s and 1980s was highly successful in categorising movement disorders and revealing some of their underlying pathophysiology. In dystonia, for example, such a wide range of abnormalities in spinal, brainstem and cortical physiology have been described that it is difficult to remember that some of the focal dystonias were once regarded as psychogenic in origin. So what is in store for the 21st century? In many respects, the basic techniques of clinical neurophysiology, that is, EEG and EMG have been the same since they were first introduced. Many of the

* Correspondence to: I.e. Rothwell, Sobell Department, Institute of Neurology, Queen Square, London WCIN 3BG, UK. E-mail address: e-mail: [email protected]

advances in the past 50 years have depended on increased computing power to analyze the information that those signals contain. Thus, the advent of averaging produced an enormous expansion in the study of evoked potentials, and more recently, frequency analysis of EEG and EMG signals has brought with it the development of coherence techniques that allow us to investigate the relation between electrical signals in different parts of the CNS. In the years ahead there will undoubtedly be further advances in extracting information from these signals, particularly the EEG. It also seems probable that clinical neurophysiology will begin to interact more closely with functional brain imaging. The latter has been the domain of cognitive neuroscience and has been less influential in clinical neurophysiology partly because of its cost and partly because of the lack of temporal resolution. However, the combined use of clinical neurophysiological tools with functional brain imaging may well be a very powerful way forward in the 21st century. But will the introduction of new methodologies change the way in which clinical neurophysiology is used? Unless a spectacular new development occurs, clinical neurophysiology is probably going to become less and less important in diagnosing movement disorders. Genetic typing and anatomical brain imaging will take precedence in these areas. Neurophysiologists may still continue to provide useful objective measures of the clinical response to new therapies, but this function will remain in the domain of specialized centers. So is the future bleak? I do not think so. In fact, rather than becoming sidelined in its relationship to movement disorders I think the 21st century may be the time when clinical neurophysiology becomes of age. The reason for this is that many recent studies have begun to move away from a purely descriptive

J.e. ROTHWELL

718

approach and are using neurophysiological methods to probe and manipulate CNS plasticity. In other words, the future may be much more interventionist than the past. Clinical neurophysiologists will be employed to probe how systems respond to changes in genetic makeup, how they react to injury, and how application of neurophysiological methods may be able to change the underlying physiology and lead to development of new therapies. For example, I imagine that rather than describing the change in reflex pathways that occurs in patients with spasticity, much more attention in the future will be directed towards asking why these changes occur. Neurophysiologists will be exploring the nature of the signals that normally prevent this pattern of adaptation from occurring in the healthy animal, and identifying the factors that cause some of the maladaptive changes that we see. If the cause of these changes can be found, then it may be possible to use that knowledge to try to reverse them, perhaps by providing other inputs from peripheral sources or by strengthening other neural pathways. In other words clinical neurophysiologists will not only probe the mechanism of disease but also try to provide potential therapeutic methods of reversing or ameliorating these changes. There are three main areas which I would highlight as of great potential importance in the 21st century. 43.2. Combining functional imaging (fMRII PET) with clinical neurophysiological tools (EEG, MEG, TMS) There are three ways in which the high spatial resolution of functional imaging can be combined with the temporal resolution of electrophysiological methods: (1) by using the techniques separately in parallel sets of experiments; (2) using both methods simultaneously; or (3) sequentially, as in TMS studies, where the brain is imaged after a period of rTMS in order to follow the time course of restoration of normal function.

43.2.1. Parallel experiments It is already routine in many centers to use the anatomical information provided by MRI images to constrain models of EEGIMEG sources (George et al., 1995). In the future, tMRI images of activated areas will be used more frequently to constrain

sources further to sites that are likely to have activity associated with a task. The same approach can be used to compare patterns of connectivity revealed with statistical modeling of the tMRI signal (Buchel and Friston, 2001) and the connectivity deduced from studies of EEG coherence at different scalp sites. It is also becoming common to design parallel imaging and TMS experiments to ask whether the activation seen in tMRIlPET is critical for task performance. In such studies, TMS is used as a "virtual lesion" (Walsh and Cowey, 2000) to test whether sites activated in the tMRI are contributing essentially to task performance. Single or multiple TMS pulses are used to interfere with the brain at a particular time during the task. The argument is that if performance of the task is reduced, areas of the brain that are stimulated under the coil are likely to contribute to the function being studied (Walsh and Cowey, 2000). Clearly, with the recognition that TMS can produce effects at a distance this type of argument is not watertight and control experiments with stimulation at different sites on the scalp are needed to address this issue (Jahanshahi and Rothwell, 2000). The elegance of this approach has recently been demonstrated in the field of movement disorders in a paper by Johansen-Berg and colleagues (JohansenBerg et al., 2002). These authors used tMRI to show that monohemispheric stroke patients had additional activity in the premotor cortex opposite to the damaged hemisphere when they tried to use the weak hand. In order to test whether this extra activation on the undamaged side was contributing to hand function, they applied TMS over the premotor cortex and showed that they could disrupt hand function much more readily in patients than in healthy age-matched controls. The implication was that this premotor activity was contributing to the patients' recovered clinical function.

43.2.2. Simultaneous experiments It is impossible to use MEG at the same time as PET or tMRI, but although technologically difficult, some groups have reported success at interfacing EEG with tMRI and TMS with tMRI/PET. Thus, recording EEG while epileptic patients are undergoing tMRI can help identify the source of a spike focus (Krakow et al., 2000). However the quality of

RESEARCH STUDIES IN NORMAL SUBJECTS AND PATIENTS: CURRENT AND FUTURE

the EEG records obtained in an MRI environment is not yet as high as in a routine EEG laboratory and this has made it difficult to study the much smaller signals recorded from the scalp of healthy subjects. TMS pulses have also been administered in PET (Paus et al., 1997) and fMRI (Bohning et al., 1998; Baudewig et al., 2001). The combination promises to allow us to use TMS pulses to activate one area of brain and image its effects at distant connected sites. This type of functional connectivity may well be influenced by the excitability of the connections at the time the TMS pulse is given so that it may be influenced by task or pathology. If so, it may prove a very useful tool to probe physiological connectivity in the intact human brain.

43.2.3. Sequential experiments A novel way of interfacing TMS and fMRI is to image the after effects of a period of rTMS. Since these may last for 30 min or more depending on the parameters of stimulation, it is possible to apply rTMS outside the scanner and then image the recovery of function over the next hour or so. Potentially such an approach would be capable of identifying how the pattern of task related activation changed over that period. Thus, if rTMS produced a mild lesion-like effect, then other structures may compensate by increasing their activation so that performance would be relatively unaffected. As the interval after rTMS increased, this compensation would decline and would be evident in the BOLD signal as a time-related change in the pattern of activation. Such work could not only give insight into the way areas of the brain cooperate to perform a task, but also provide an acute model for reorganization of the brain after injury.

43.3. Clinical neurophysiology and genetics Knowing a patient's genetic deficit is only the beginning of a very long road to understanding the final clinical presentation. Secondary adaptations may conceal a deficit, whereas secondary insults may reveal the symptoms. In the future, clinical neurophysiology will be a useful tool to help distinguish these factors and also indicate those which are important in any particular condition. As an example, consider the mutation of the DYTl gene that can cause severe generalized dystonia. The mutation is exactly the same in all

719

individuals who carry it, yet some individuals are completely normal whereas others suffer from severe generalized dystonia. Why does this difference occur? Do some individuals manage to compensate for an underlying genetic deficit by plastic changes in their nervous system? Alternatively do some patients suffer a second insult which reveals the deficit? Clinical neurophysiology may go some way towards answering these questions. Eidelberg and colleagues (Eidelberg et al., 1998) have shown that there is indeed an underlying deficit even in non-clinically manifesting gene carriers. They used principal component analysis of the pattern of resting brain metabolism to show that individuals who carry the DYTl mutation, whether or not they are clinically affected, have the same abnormal pattern as non-genetically characterized patients with dystonia. The implication is that the gene itself carries some pathophysiological consequences. Additional neurophysiological studies could help to identify these consequences in the range of pathways known to function abnormally in dystonia. It may be, for example, that non-manifesting patients will not have the full range of physiological deficits as are seen in manifesting patients. Clinical symptoms may depend on the presence of an extra deficit(s) that may be linked to a second genetic change or to an environmental insult. Another recent example of genetic studies provides the indication of how neurophysiological techniques may help probe the consequences of gene deficits. Nielsen and colleagues investigated patients with hyperekplexia who had a genetic deficit in the gene for the glycine receptor (Nielsen et al., 2002). This meant that patients had no disynaptic reciprocal inhibition between the antagonist muscles in the calf. Nevertheless, these patients walked quite well, and Nielsen and colleagues suggested that one reason for this might be adaptation of circuits within the cerebral cortex which maintain a reciprocal inhibitory pattern between those muscles, at least when used in volitional movement.

43.4. Clinical neurophysiology and CNS plasticity Physiotherapy is a traditional way of influencing CNS plasticity. However physiotherapists have been dogged for years, first by the problem of proving the

720

efficacy of their treatments, and second by lack of real knowledge about the mechanisms involved. It seems likely that one of the major advances in clinical neurophysiology in the 21st century will be to unlock solutions to these problems. In a highly influential study of acquired focal dystonia, Byl et aI. (1996) showed that excessive training of a hand movement in monkeys produced reorganization of the hand areas of the primary somatosensory cortex. The suggestion was that this could in some lead to problems in sensory-motor integration, and that the selection of a motor program or its execution would become disorganized. Similar changes in the somatosensory cortex have been found in patients with focal hand dystonia using SEP or MEG measures (Bara et aI., 1998; Meunier et aI., 2001) and have lead Byl and colleagues to propose a method of retraining somatosensory organization using specific sensory discrimination tasks in individual patients (Byl and McKenzie, 2000). These tasks are designed to restore the normal organization of the cortex, and thereby perhaps restore motor function. In this example, clinical neurophysiology was used in its classic mode to describe the plastic changes that occur following an insult and was then extended to help search for a possible method of reversing the effect that had been identified. The same approach is presently being used in many other conditions. After stroke, many imaging, EEG and TMS studies have shown that recovery of function depends upon additional activity in non-damaged areas of the brain (Hamdy, 1998; Johansen-Berg, 2002; Staudt, 2002). These may be in the peri-infarct region in the affected hemisphere, or in sites in the unaffected hemisphere. However, I suspect that in the future not only will this descriptive approach be used, but a more interventionist approach will be seen in which basic knowledge about the scientific basis of brain plasticity is used to focus physiological techniques on promoting synaptic plasticity. 43.4.1. New types of intervention

Recent studies give some indication of the approaches that might be used. There is a large and growing body of knowledge about the basic mechanisms of synaptic plasticity. Even at this time, several groups working in humans have made use of

J.C. ROTHWELL

the fact that short term changes in sensory input can influence the motor system for hours or more afterwards. For example, Hamdy et aI. (Hamdy et aI., 1998; Fraser et al., 2002) have shown that stimulation of the pharynx can change the excitability of the cortical representation of swallowing muscles. They have also shown that this can improve swallowing in patients in whom the corticobulbar pathways are compromized with dysphagia. Ridding and colleagues (McKay et aI., 2002) have shown that long periods of somatosensory stimulation given on repeated days increase the excitability of corticospinal projections. They have suggested that a similar technique might be used to improve walking in patients with stroke. Muellbacher et aI. (2002) have anesthetized the shoulder area of patients with hemiplegia to improve motor function in the hand. Treadmill walking training has been used by Dietz et al. (1994) and others to try to restore excitability to the spinal pattern generators involved in walking. It seems likely that a major area of research in the future will focus on providing the optimal inputs to retrain damaged networks in patients with movement disorders. A second major neurophysiological method that is currently being used by many centers to produce plasticity in the human motor system is transcranial brain stimulation. At the present time, transcranial magnetic stimulation is the method of choice, but in the future this may be supplemented by electrical DC stimulation which has recently been (re)introduced by Paulus and colleagues (Nitsche and Paulus, 2000). The advantage of these methods is that it allows the neurophysiologist to stimulate central neurons directly rather than relying on external inputs from sensory stimulation or antidromic inputs from motor stimulation. An example of the efficacy of direct central stimulation was provided recently by Classen and colleagues with their technique of paired associative plasticity in the motor cortex (Stefan et al., 2000). They found that if they stimulated the motor cortex at the same time as arrival of sensory input from the periphery, then they could increase cortical excitability as measured with single pulse TMS. The timing of the paired inputs was crucial; if the sensory input arrived too late or too soon, then facilitation was no longer present. It is quite possible to envisage a similar technique being used elsewhere in the brain, either with a different form of sensory input or perhaps even with pairs of

RESEARCH STUDIES IN NORMAL SUBJECTS AND PATIENTS: CURRENT AND FUTURE

magnetic stimulation given to two different connected areas of cortex that converge on a common target. An area of increasing interest in clinical neurophysiology is the possibility of using repetitive TMS to produce long term plastic changes in the brain. Repetitive stimulation is thought to be able to produce changes in circuitry that may be similar to long term depression and long term potentiation at synaptic connections in animal experiments. These after-effects of rTMS may last for minutes, hours or even days. In future, it is probable that other methods will be used to control the effects that rTMS produces, for example. Ziemann et aI. (1998) have shown that the effect of rTMS on the motor cortex can be modulated by removing afferent input with anesthesia. Siebner and colleagues have shown that the effect of rTMS may change in disease (Siebner et aI., 1999). Thus, careful combination of rTMS intensities and current orientations coupled with attempts to influence excitability of specific circuits with afferent input or perhaps with task performance may greatly facilitate the effects. At one time, it was thought that the effects of rTMS were confined to the site of stimulation, limiting the amount of brain that could be targeted for therapeutic use. Physiological and imaging studies have now shown very clearly that this is not the case and that rTMS to one area of the brain produces effects in connected sites at a distance. Effectively, it appears that rTMS could be used to target specific disordered circuits within the brain rather than a single area of brain (Siebner and Rothwell,2003). Therapeutic trials are already taking place with rTMS. Limited success has been described in treatment of depression (Loo et al., 2001; Dannon et aI., 2002), but less success has been seen in treatment of movement disorders where changes in clinical parameters have been small and short lasting (Mally and Stone, 1999; Siebner et aI., 1999; George et al., 2001). However, in the future, as knowledge of mechanisms improves, there may be much greater possibility of obtaining larger clinical effects, particularly if rTMS is applied in conjunction with other interventions or even in conjunction with neuromodulatory drugs. DC stimulation produces effects on the brain by polarising neurons under the anode and cathode (Nitsche and Paulus, 2000). There is an immediate

721

effect on the excitability of those neurons, with anodal stimulation usually increasing the excitability and cathodal stimulation decreasing it. More interestingly, if DC stimulation is applied for several minutes (for example 10 minutes), then there are lasting after-effects on excitability (Nitsche and Paulus, 2001). In the 1960s and early 1970s some authors reported using DC stimulation for several minutes or hours as a potential ways of treating depression. However, this was of questionable success. With new knowledge about the effects of DC stimulation, it may well be that it can be used successfully to target circuits in the brain, and perhaps to interact with other methods of producing neural plasticity. A final example of this new approach of clinical neurophysiology and plasticity concerns deep brain stimulation (DBS). DBS is a useful treatment for late stage Parkinson's disease, dystonia and tremor, yet we are far from understanding exactly how it works. Numerous studies have now indicated that the initial explanations of its action in Parkinson's disease may be incomplete. For example, the 1990 model of under activity in the direct pathway and over-activity in the indirect pathway producing excess inhibitory output of the pallidum in patients with Parkinson's disease is now regarded as incomplete (Marsden and Obeso, 1994; Mink, 1996; Vitek, 2002). This model predicted for example that pallidal lesions or functional inactivation with DBS would ameliorate parkinsonian symptoms by normalising the pattern of pallidal output. However, those models did not predict the outstanding success of pallidotomy in Parkinson's disease, which was in preventing the occurrence of drug induced dyskinesia. If anything, the classic model of basal ganglia function would predict that lesions to the pallidum would increase the chances of producing dyskinesias (Wichmann and DeLong, 1993). Because of these discrepancies, it is now proposed that lesions or DBS of basal ganglia structures reduce parkinsonian symptoms because they prevent an abnormal basal ganglia output from interfering with function in other brain structures (Vitek, 2002). However, this explanation itself implies that the brain can function reasonably well in the absence of a meaningful basal ganglia input (Berardelli et aI., 2001). If this is the case, then some other structures must be contributing to functions normally performed by the basal ganglia. The aim of clinical neurophysiology will be to

722

indicate what these structures may be and what functions they perform in the modified circuitry of patients with deep brain stimulation. Even more intriguing is to understand how DBS alleviates dystonia. Several centers have indicated this can be an extraordinarily successful intervention when chronic stimulation is applied bilaterally to the globus pallidus (Krack and Vercueil, 2001). However, the time course of its clinical effect is often measured over weeks rather than minutes or hours after turning the stimulator on or off. The implication is that some long term process of adaptation is occurring in response to the change in basal ganglia signal. What sort of changes are these? What part of the brain do they affect? What is the underlying mechanism that informs one brain area of the state of a connected site at a distance? Such questions can now be tackled with clinical neurophysiological techniques and will help us understand much more clearly the success of deep brain stimulation as a clinical therapy. 43.5. Conclusion

Clinical neurophysiology at the start of the 21st century is approaching a new era where research will begin to encompass functional imaging techniques and where conventional electrical methods will move from describing the behavior of the CNS to a more interventionist approach in which we will use them to modulate neural excitability and connectivity in healthy and diseased brain. The increased understanding of mechanisms of adaptation in neural circuits will be of obvious importance in both the evaluation and development of new therapeutic techniques. References Bara, JW, Catalan, MJ, Hallett, M and Gerloff, C (1998) Abnormal somatosensory homunculus in dystonia of the hand. Ann. Neurol., 44: 828-831. Baudewig, J, Siebner, HR, Bestmann, S et al. (2001) Functional MRI of cortical activations induced by transcranial magnetic stimulation (TMS). NeuroReport, 12: 3543-3548. Berardelli, A, Rothwell, JC, Thompson, PD and Hallett, M (2001) Pathophysiology of bradykinesia in Parkinson's disease. Brain, 124: 2131-2146. Bohning, DE, Shastri, A, Nahas, Z et al. (1998) Echoplanar BOLD fMRI of brain activation induced by

J.e. ROTHWELL concurrent transcranial magnetic stimulation. Invest. Radio!., 33: 336-340. Buchel, C and Friston, K (2001) Interactions among neuronal systems assessed with functional neuroimaging. Rev. Neural. (Paris), 157: 807-815. Byl, NN and McKenzie, A (2000) Treatment effectiveness for patients with a history of repetitive hand use and focal hand dystonia: a planned, prospective follow-up study. J. Hand Ther., 13: 289-301. Byl, NN, Merzenich, MM and Jenkins, WM (1996) A primate genesis model of focal dystonia and repetitive strain injury: I. Learning-induced dedifferentiation of the representation of the hand in the primary somatosensory cortex in adult monkeys. Neurology, 47: 508-520. Dannon, PN, Dolberg, OT, Schreiber, Sand Grunhaus, L (2002) Three and six-month outcome following courses of either ECT or rTMS in a population of severely depressed individuals - preliminary report. Bio!. Psychiatry, 51: 687-690. Dietz, V, Colombo, G and Jensen, L (1994) Locomotor activity in spinal man. Lancet, 344: 1260-1263. Eidelberg, D, Moeller, JR, Antonini, A et al. (1998) Functional brain networks in DYTl dystonia. Ann. Neurol., 44: 303-312. Fraser, C, Power, M, Hamdy, S et al. (2002) Driving plasticity in human adult motor cortex is associated with improved motor function after brain injury. Neuron, 34: 831-840. George, JS, Aine, CJ, Mosher, JC et al. (1995) Mapping function in the human brain with magnetoencephalography, anatomical magnetic resonance imaging, and functional magnetic resonance imaging. J. Clin. Neurophysiol., 12: 406-431. George, MS, Sallee, FR, Nahas, Z, Oliver, NC, Wassermann, EM (2001) Transcranial magnetic stimulation (TMS) as a research tool in Tourette syndrome and related disorders. Adv. Neuro!., 85: 225-235. Hamdy, S, Rothwell, JC, Aziz, Q, Singh, KD and Thompson, DG (1998) Long-term reorganisation of human motor cortex driven by short-term sensory stimulation. Nat. Neurosci., 1(1): 64-68. Jahanshahi, M and Rothwell, J (2000) Transcranial magnetic stimulation studies of cognition: an emerging field. Exp. Brain Res., 131: 1-9. Johansen-Berg, H, Rushworth, MF, Bogdanovic, MD, Kischka, D, Wimalaratna, S and Matthews, PM (2002) The role of ipsilateral premotor cortex in hand movement after stroke. Proc. Natl. Acad. Sci. USA, 99: 14518-14523. Krack, P and Vercueil, L (2001) Review of the functional surgical treatment of dystonia. Eur. J. Neurol., 8: 389-399.

RESEARCH STUDIES IN NORMAL SUBJECTS AND PATIENTS: CURRENT AND FUTURE

Krakow, K, Allen, PJ, Lemieux, L, Symms, MR and Fish, DR (2000) Methodology: EEG-correlated fMRI. Adv, Neurol. 83: 187-20L Loo, C, Sachdev, P, Elsayed, H et al. (2001) Effects of a 2to 4-week course of repetitive transcranial magnetic stimulation (rTMS) on neuropsychologic functioning, electroencephalogram, and auditory threshold in depressed patients, Biol. Psychiatry, 49: 615-623, Mally, J and Stone, TW (1999) Therapeutic and "dosedependent" effect of repetitive microelectroshock induced by transcranial magnetic stimulation in Parkinson's disease. J. Neurosci. Res., 57: 935-940. Marsden, CD and Obeso, JA (1994) The functions of the basal ganglia and the paradox of stereotaxic surgery in Parkinson's disease. Brain, 117(4): 877-897. McKay, DR, Ridding, MC, Thompson, PD and Miles, TS (2002) Induction of persistent changes in the organisation of the human motor cortex. Exp. Brain Res., 143: 342-349. Meunier, S, Garnero, L, Ducorps, A et al. (2001) Human brain mapping in dystonia reveals both endophenotypic traits and adaptive reorganization. Ann. Neurol., 50: 521-527. Mink, JW (1996) The basal ganglia: focused selection and inhibition of competing motor programs. Prog. Neurobiol., 50: 381--425. Muellbacher, W, Richards, C, Ziemann, U et al. (2002) Improving hand function in chronic stroke. Arch. Neurol.,59: 1278-1282.

Nielsen, JB, Tijssen, MA, Hansen, NL et al. Corticospinal transmission to leg motoneurons in human subjects with deficient glycinergic inhibition. J. Physiol., 544: 631-640. Nitsche, MA and Paulus, W (2000) Excitability changes induced in the human motor cortex by weak tran-

723

scranial direct current stimulation. J. Physiol., 527(3): 633-639. Nitsche, MA and Paulus, W (2001) Sustained excitability elevations induced by transcranial DC motor cortex stimulation in humans. Neurology, 57: 1899-1901. Paus, T, Jech, R, Thompson, CJ, Comeau, R, Peters, T and Evans, AC (1997) Transcranial magnetic stimulation during positron emission tomography: a new method for studying connectivity of the human cerebral cortex. J. Neurosci., 17: 3178-3184. Siebner, HR and Rothwell, J (2003) Transcranial magnetic stimulation: new insights into representational cortical plasticity. Exp. Brain Res., 148: 1-16. Siebner, HR, Tormos, JM, Ceballos, BA et al, (1999) Lowfrequency repetitive transcranial magnetic stimulation of the motor cortex in writer's cramp. Neurology, 52: 529-537. Staudt, M, Grodd, W, Gerloff, C, Erb, M, Sitz, J, Krageloh-Mann, I (2002) Two types of ipsilateral reorganization in congenital hemiparesis: a TMS and fMRI study. Brain, 125: 2222-2237. Stefan, K, Kunesch, E, Cohen, LG, Benecke, R and Classen, J (2000) Induction of plasticity in the human motor cortex by paired associative stimulation. Brain, 123(3): 572-584. Vitek, JL (2002) Pathophysiology of dystonia: a neuronal model, Mov. Disord., 17 (Suppl, 3): S49-S62. Walsh, V and Cowey, A (2000) Transcranial magnetic stimulation and cognitive neuroscience. Nat. Rev. Neurosci., I: 73-79. Wichmann, T and DeLong, MR (1993) Pathophysiology of parkinsonian motor abnormalities. Adv. Neurol., 60: 53-61. Ziemann, D, Hallett, M and Cohen, LG (1998) Mechanisms of deafferentation-induced plasticity in human motor cortex. J. Neurosci., 18: 7000--7007.

Movement Disorders Handbook of Clinical Neurophysiology, Vol. 1 M. Hallett (Ed.) © 2003 Elsevier n.v All rights reserved

727 CHAPTER 44

Future clinical applications of clinical neurophysiology in movement disorders G. Deuschl":" and M. Hallettb b

"Neurologische Klinik, Christian-Albrechts-Universitdt, Kiel, Germany Human Motor Control Section, National Institute of Neurological Disorders and Stroke, National Institutes ofHealth, Bethesda, MD, USA

Without a crystal ball, it is not easy to predict future developments in the field of clinical applications of Movement Disorders. We will try to formulate some expectations for future developments in this challenging application of clinical neurophysiology, knowing well that the development of neurology and clinical neurosciences is erratic, and new and promising ideas have often changed the expected developments in a favorable way. Our approach was to put together the open questions in the field of movement disorders and the possible answers that clinical neurophysiology could provide. In general, clinical neurophysiology has been strongest in the study of the pathophysiology of movement disorders. This has been helpful for understanding of the different conditions, for suggesting rational therapies and for helping with diagnosis. For the clinic at the present time, clinical neurophysiology offers some help in the diagnosis of movement disorders and monitoring of disease progression and therapy. Clinical neurophysiology is clearly already extremely valuable for the diagnosis of a number of conditions. An excellent example is in determining that a quick movement is myoclonus and making the differential diagnosis as to what type of myoclonus it is. Neurophysiological techniques allow measurements in the millisecond range, needed for the differential diagnosis, and not possible with even expert clinical evaluation. Neurophysiological methods are also useful for the study of tremors,

* Correspondence to: Prof. Dr. G. Deuschl, Neurologische Klinik, Christian-Albrechts-Universitat Kie1, Niemannsweg 147,24105 Kie1, Germany. E-mail address:[email protected] Tel.: 0431-597-2610; fax: 0431-597-2712.

allowing, for example, the separation of essential tremor and exaggerated physiological tremor most of the time. In many circumstances, even when there have been good physiological studies of certain conditions, the results have been shown only for group differences. For clinical diagnostic purposes, since patients come into the laboratory one at a time, it will be necessary to know the sensitivity and specificity of these tests. And, if the sensitivity and specificity is not adequate, the test might well be redesigned. In our view, this is one of the most important tasks for the future. There is still room for improvement for the diagnosis and differential diagnosis of a number of different Movement Disorders. This applies especially for: - Parkinson's disease and non-idiopathic parkinsonian syndromes. What is the most sensitive method for identifying early parkinsonism? Can it be identified prior to clinical symptoms? Can idiopathic disease be identified compared with the Parkinson-plus conditions? - Tremors and their differential diagnosis. For example, mild essential tremor cannot always be differentiated from exaggerated physiological tremor. - Tics vs. other hyperkinetic movement disorders (e.g. stereotypies, mannerisms). - Other non-rhythmic movement disorders. Will it be possible to identify Huntington chorea from Sydenham chorea? Will it be possible to separate tardive dyskinesia from other dyskinesias? - Organic vs. psychogenic movement disorders. Psychogenic movement disorders are very common, and while there are some very useful

728

physiological tools, more are needed. For example, psychogenic dystonia is particularly difficult to diagnose. Further tasks are appearing with the development of genetic methods for the study of movement disorders. New genetic tests will help provide early diagnosis for hereditary movement disorders, but questions may arise whether a particular patient is already affected by the diagnosed hereditary condition. Therefore, there is much room for the early diagnosis of the actual manifestation of these conditions - a suitable task for clinical neurophysiology. Another variant of the same problem comes with the need for more precise phenotyping of hereditary conditions. The study of movement disorders with genetic methods relies on the exact diagnostic classification. In trying to find a gene for a condition, for example, false positives are devastating. Clinical neurophysiology can play an important role in this field. For example, studies so far have failed to find genes for essential tremor or Tourette syndrome. One problem could be false positives. Monitoring of movement disorders is an important task for the future care of patients. Up to now this field has just been opened with methods of measurement that are often crude. Such monitoring might be useful with abnormal movement conditions that cause complaints such as tremor, rigidity, hyperkinesias, disturbances of coordination and spasticity. It will be necessary to study the course over time of such abnormal movements including the effects of therapy. Such techniques should monitor the target symptoms, and more elaborate computer programs to quickly analyze the results are needed. Several conditions where long-term monitoring over extended time periods would be desirable include Parkinson's disease, essential tremor, myoclonus and dystonia. Clinical neurophysiology is just beginning to establish a role in the treatment of neurologic disorders. This is already established for the presurgical diagnostic procedures in epileptology, but is just starting for the field of movement disorders. Two major areas have been opened. One is the use of botulinum toxin with electromyography. Many muscles can only be injected under electromyographic guidance. It is the task of clinical neurophysiology to explore the possible advances of such a procedure. There are in fact only a few studies available

G. DEUSCHL AND M. HALLETT

demonstrating superior therapeutic results when botulinum toxin is applied under EMG guidance. Another important application is intraoperative recording with microelectrodes for the correct placement of deep brain electrodes for the treatment of movement disorders or pain. New indications are being developed actively for deep brain stimulation. In order to qualify for this task, the clinical neurophysiologist must develop clinical skills for these different tasks. In the future these developments may change the profile of the clinical neurophysiologist who will now need to know about the intraoperative clinical assessment of patients with Parkinson's disease, tremors or dystonia or the indication and management of dystonic movement disorders or other conditions to be treated with botulinum toxin. This development will broaden the profile of clinical neurophysiology. Many fields for future research will be opened for clinical neurophysiology with therapeutic applications. For example, the therapeutic potential for procedures such as transcranial magnetic stimulation can be profitably explored. Clinical neurophysiology will be able to enter therapeutic fields only by proving its innovative potential in appropriate carefully controlled prospective studies. Every time period has its hot topics of scientific research. When new methods become available, all of a sudden new insights are possible. This applies for the physiology of movement disorders especially with transcranial magnetic stimulation, techniques of coherence analysis for the EEG or EMG, and with the new techniques of functional imaging. It may be possible by combining these techniques to improve simultaneous temporal and spatial resolution of brain events. Advanced mathematical techniques will probably play an important role in further development in this area. We hope also to understand better the basal ganglia and the brainstem/cerebellum. Any kind of noninvasive method to study the function of these areas would be very productive. Other available and fully developed methods like polysornnography and autonomic testing have not even reached the field of movement disorders, where many applications are waiting for them. Another challenge is the integration of cognitive and motor research. Training programs for clinical neurophysiology do not yet include very much teaching in the movement disorders related fields. Already at the

FUTURE CLINICAL APPLICATIONS OF CLINICAL NEUROPHYSIOLOGY IN MOVEMENT DISORDERS

present state of development such a training program should be developed. Much can be done in the highly interesting and fast-developing area of movement disorders and the clinical neurophysiologist can already play a central role in the analysis of

729

patients. This role will likely expand with new techniques, and clinical neurophysiologists can certainly facilitate the continuing development of the field itself.

Movement Disorders Handbook of Clinical Neurophysiology, Vol. 1 M. Hallett (Ed.) © 2003 Elsevier B. V. All rights reserved

731

Subject Index Italic page numbers indicate in-depth treatment Accelerometry, 181-189 accelerometers, 181 EMG,186 motion in 3-D space, 183-185 signal conditioning, 182-183 sleep disorders, 186 spectral analysis, 186-189 technical specifications, 181-182 tremor, 185-186,367,381,404,409 Adverse drug effects focal injection therapy, 66~63 Alcohol focal injection therapy, 655-676 Alien hand sign parkinsonian syndrome, 445 Alzheimer's disease brain imaging, 165 myoclonus, 536 reaction/movement time, 221 transcranial magnetic stimulation (TMS), 102, 106 Amyotrophic lateral sclerosis (ALS) TMS, 102, 104-106 Apraxia event-related (de-)synchronization, 25 kinesiology, 194-196 Astereognosis kinesiology, 198 Asterixis myoclonus, 537 Athetosis, 574 Attention deficit hyperactivity disorder tics, 550-553 Balance disorders posturography, 328 Ballism, 571, 577 Basal ganglia deep brain stimulation, 722 dystonia, 457-458 hemiparesis, 608 tics, 555 tremor, 383-385 Bell's palsy cranial hyperkinetic syndromes, 562

Bereitschaftspotential cerebellar ataxia, 501 electrocorticography, 34-38 myoclonus, 545 Parkinson's disease, 18,420 tics, 552, 555. Blepharospasm cranial hyperkinetic syndromes, 563-566 focal injection therapy, 670-671 Blink reflex cranial hyperkinetic syndromes, 561, 565 Parkinson's disease, 576 Botulinum toxin focal injection therapy, 655-676 Bradykinesia Huntington's disease, 573 Parkinson's disease, 417-428 parkinsonian syndrome, 438-439 Brain imaging, 163-172, see also PET, MR!, SPECT, etc. Alzheimer's disease, 165 functional imaging brain mapping, 168-171 fMRI bold imaging, 168 movement disorders, 169-171 normal motor control, 169 PET cerebral blood flow, 168 hyperkinetic movement disorders, 166-168 hypokinetic movement disorders, 165-166 proton magnetic resonance spectroscopy, 167-168 radionuclide imaging, 164-167 imaging therapy ablative surgery, 171 deep brain stimulation, 171-172 fetal transplantation, 172 structural imaging, 163-164 tics, 554 Brain injury/lesion focal injection therapy, 668 Holmes' tremor, 400 long-latency reflexes, 290 reaction/movement time, 220 Brain plasticity dysphagia, 720 neurophysiology, 719-722

732

Brainsurgery imaging therapy, 171 stereotactic surgery microelectrode recording, 127-136

Brainstem disorders Huntington's disease, 572 hyperekplexia, 483-486 Parkinsonian syndrome, 439

Casting focal injection therapy, 665

Cerebellar ataxia, 491-516 afferents, 492 autonomic centers, 497-498 Bereitschaftspotential, 501 clinical practice, 498-500 clinical signs, 498 inherited ataxias, 499 spinocerebellar ataxia, 499 sporadic forms, 498 cortex, 493 definition, 498 efferents dentate nucleus, 496 fastigial nucleus, 496 interpositus nucleus, 496 thalamic nuclei, 496 electric stimulation, 515 eye movements optokinetic responses, 501 saccades, 501 smooth pursuit, 500 vestibula-ocular responses, 501 Friedreich's disease, 499 gait disorders, 645 inferior olive, 494 limb movements Bereitschaftspotential, 501 multi-joint movements (arm) angular kinematics, 505 dynamics, 506 spatial tuning of EMG, 507 reaction and movement time, 502 single-joint movements aberrant recovery, 504 accuracy, 502 hypermetria, 504 hyperventilation, 503 inertia, 503 visuomotor coordination, 507 mossy fibers acoustic/visual/trigeminal, 496 pontine nuclei, 496 reticular nuclei, 495

SUBJECT INDEX

spinocerebellar, 494 vestibular projections, 496 motor learning, 507-511 adaptation eye blink conditioning, 508-509 prisms, 509 visuomotor gain, 510 skills serial reaction time test, 510 myoclonus, 513-515 neuroanatomical pathways, 491-498 neurophysiological aspects, 500-516 nucleus, 494 posture and gait, 515 anterior lobe damage, 515 posterior vermal split syndrome, 515 TMS, 104-105,513,515 tremors kinetic, 511 palatal, 511 postural,511-513 TMS,513

Cerebellum cerebellar tremor, 403-405, 511-515 long-latency reflexes, 291

Cerebral palsy focal injection therapy, 671, 674

Cerebrovascular disease palatal tremor, 402

Choreas, 571-574, see also Huntington's disease EMG,1O

Coherence, see Cortico-cortical coherence, and Cortlco-muscular coherence Computertomography deep brain stimulation, 700 Parkinson's disease, 700

Congenital mirrormovement TMS, 104, 107-108

Contingent negativevariation cued movements, 19-20 electrocorticography, 38-39 event-related (de-)synchronization, 25 Parkinson's disease, 421-424 tics, 551

Continuous motor unit activity stiffness, 463-475 Corticobasal (ganglionic) degeneration brain imaging, 166 long-latency reflexes, 290 myoclonus, 537 parkinsonian syndrome, 437, 445 reaction/movement time, 219-220

Cortico-cortical coherence, 77-84 applications, 83-84

733

SUBJECT INDEX

methodology, 78-83 advanced computations, 81-83 coherence estimation, 78 data acquisition, 83 multiple frequencies, 79-81 task- and event-related coherence, 78-79, 82

Cortico-muscular coherence factors with effect, 90-91 generator mechanism, 89-90 methodology, 87-89 coherence estimation, 87-88 phase analysis, 88 recording technique, 88-89 pathological conditions, 91

Cranial hyperkinetic syndromes Bell's palsy, 562 blepharospasm, 563-566 electrophysiology, 559-567 EMG,559 facial action myoclonus, 567 facial myokymia, 567 hemifacial spasm, 559-562 hemimasticatory spasm, 566 post-facial palsy synkinesia, 562-563 Cranial nerve reflexes, 247-262 jaw reflexes jaw jerk, 252-253 central pathway, 253 physiology and normative data, 253 recording technique, 253 masseter inhibitory reflex central pathway, 255 physiology and normative data, 254 recording technique, 254 levator palpebrae inhibitory reflex recording technique, 251 physiology and normative data, 251-252 motor function assessment conduction examination facial nerve, 256 hypoglossal nerve, 256 spinal accessory nerve, 256 recording of movements, 256-257 orbicularis oculi reflexes blink reflex acoustic, 251 central pathways, 248-249 photic, 251 physiology and normative data, 248 recording technique, 247-248 somatosensory, 251 corneal reflex central pathways, 250 physiology and normative data, 250

recording technique, 249 recovery curves to paired stimuli, 258-262 levator palbebrae inhibitory reflex, 261-262 masseter inhibitory reflex, 262 orbicularis oculireflexes, 260-261

Creutzfeldt-Jakob disease myoclonus, 536 TMS, 108

Deafferentation kinesiology, 197

Deep brain nuclei microelectrode recording, 127-136

Deep brain stimulation basal ganglia, 722 brain plasticity, 721-722 cerebellar tremor, 403--405 dystonia, 722 Parkinson's disease, 697-709, 722 beneficial effects, 703-705 complications, 705 cost, 707 depression and suicide, 709 electrode implantation, 699-701 electrode location, 701-703 indications, 698-699 lesion, 707 mood and behavior changes, 709 neuroprotection hypothesis, 708 patient selection, 698-699 precision, 706 quality, 707 side effects, 705 speed, 706 stimulation or lesion, 707 subthalamic nuclear stimulation, 707-708 tremor, 722

Depression TMS, 110

Diurnal movement disorders polysornnography, 147

Dyskinesia deep brain stimulation, 703-704 differential diagnosis, 727 levodopainduced,574-576,703-704 polysornnography, 144

Dysphagia brain plasticity, 720

Dystonia, 451-458 abnormal brain activity, 455 basal ganglia disorder, 457--458 brain imaging, 167 brain plasticity, 720 cranial hyperkinetic syndromes, 563

SUBJECT INDEX

734 cranial nerve reflexes, 251 cued movements, 19 deep brain stimulation, 722 diagnosis, 451 dystonic tremor, 405 EMG, 10,451 event-related (de-)synchronization, 25 focal injection therapy, 672-673 gait disorders, 645 historical perspective, 452 long-latency reflexes, 291 reaction/movement time, 219-220 reflex abnonnality?, 454 sensory/subcortical abnormalities, 452 somatosensory evoked potentials, 64-65 spinal reflexes, 242 stiffness, 470 llJS,103-106,455-466 treatment, 456-457

Electric/magnetic fields cued movements, 18-20 event-related (de-)synchronization, 23-26 clinical application, 25-26 method of recording, 24 normal findings, 24-25 involuntary movements, 20-23 cortical reflex myoclonus, 22-23 EEG (MEG)/EMG polygraphic recording, 20-21 jerk-locked back averaging, 21-22 negative myoclonus, 23 self-initiated movements, 16-18 clinical application, 18 method of recording, 16 muscle relaxation normal findings, 16-18 principle, 16 steady-state movement, 18

Electrocardiography electrocorticography, 49 polysornnography, 141 tremor, 358

Electrocorticography, 31-40 Bereitschaftspotential, 34-38 motor mapping in neurosurgery, 37-38 multiple cortical generators, 36-37 precentral gyrus, 36 scalp recorded, 36 supplementary motor area, 37 contingent negative variation, 38-39 cortico-muscular coherence, 92 motor control, 31-41 movement disorders, 31-41 paroxysmal dyskinesias, 39-40

recording technique, 32-36 amplifier and recording conditions, 33-34 data analysis and interpretation, 35 electrode placement, 34 electrodes, 32-33 intraoperative, 36 patient condition, 34-35

Electroencephalography accelerometry, 186 brain plasticity, 720 combination with fMRIIPET, 718-719 correlation with EMG, 15-26 cortico-cortical coherence, 79, 81, 84 cortico-muscular coherence, 89-92 electrocorticography, 31-41 gait disorders, 644 hemiparesis, 601 hyperekplexia, 482 motor cortex output, 109 myoclonus, 528 Parkinson's disease, 420-426 periodic limb movements, 585 polysornnography, 140 psychogenic movement disorder, 633 restless leg syndrome, 585 startle reflex, 269 tics, 551 tremor, 368

Electromyography, 7-13 accelerometry, 186 ballism, 577 cranial hyperkinetic syndromes, 559 cerebellar ataxia, 502 chorea, 10 correlation with EEGIMEG, 15-26 cortico-cortical coherence, 81 cortico-muscular coherence, 89-92 cranial nerve reflexes, 257 dystonia, 10 electrocorticography, 49 gait disorder, 347-349, 643 Huntington's disease, 572-574 hyperekp1exia, 480, 483-486 kinesiology, 197 myoclonus, 10, 524 Parkinson's disease, 422, 576 periodic limb movements, 585 restless leg syndrome, 585 polysornnography, 141 posturography, 303, 307-315 reaction time, 217 spasticity, 619-621 spinal reflexes, 240 stiffness, 472

735

SUBJECT INDEX

tics, 10, 550 tremor, 8-10, 359-362, 367, 371, 381, 388, 398 Electro-oculography periodic limb movements, 585 restless leg syndrome, 585 Electrophysiology cranial hyperkinetic syndromes, 559-567 Encephalomyelitis stiffness, 464-469 Epilepsy brain imaging, 165 electrocorticography, 31-32, 38-39 hyperekplexia, 481-482 muscle relaxation, 18 myoclonus, 522-523, 538-539 reaction/movement time, 221 startle reflex, 269 TMS, 102, 104-106, 108, 110 Event-related (de-)synchronization apraxia, 25 electric/magnetic fields, 23-26 Parkinson's disease, 424, 425, 427

Facial palsy cranial hyperkinetic syndromes, 562 TMS, 103 Focal injection therapy, 655-676 alcohol adverse effects/contraindications, 661-663 histology, 655-656 historical aspects, 651, 652 pharmacology, 653 physiology, 657-658 use/effects/indications, 665-667 anesthetics adverse effects/contraindications, 660-661 histology, 655 historical aspects, 651 pharmacology, 653 physiology, 65tH;57 use/effects/indications, 663 botulinum toxin adverse effects/contraindications, 663 histology, 656 historical aspects, 652 pharmacology, 654 physiology, 656 use/effects/indications, 667-673 electrical stimulation technique, 673-675 anatomical location, 675 intramuscular targeting, 674 phenol adverse effects/contraindications, 661-663 histology, 655-656

historical aspects, 652 pharmacology, 654 physiology, 657-658 use/effects/indications, 665-667 rationale, 651 Friedreich's disease cerebellar ataxia, 499 long-latency reflexes, 291

Gait disorders, 337-352 cautious gait, cerebellar gait, 645 classification, 642-643 cycle, 338-341 dystonic gait, 645 equipment and methods electromagnetic tracking system, 345 EMG, 347-349 force measurement, 344-346 inverse dynamic problem, 347 joint dynamics, 346 joint motion, 344 movement measurement, 342-344 observation, 341 freezing gait, 644 frontal gait, 645 historical perspective, 337-338 interpretation of data, 349 locomotion anatomy/physiology, 641-642 myopathy and weakness, 646 neurophysiologic tests, 647-648 Parkinson's disease, 430 psychogenic, 646 sensory ataxia, 646 stiff-legged gait, 643 treatment planning/assessment, 350 Gilles de la Tourette syndrome, see Tourette syndrome Globus pallidus microelectrode recording, 130-132 Glycine receptor hyperekplexia, 481 Hemifacial spasm cranial hyperkinetic syndromes, 559-562 focal injection therapy, 671

Hemiparesis, 601-610 neural reorganization, 605 tremor, 370 Hemiplegia focal injection therapy, 665 Hemispherectomy kinesiology, 199 Hemispheric lesions somatosensory evoked potentials, 61-62

736

Huntington's disease brain imaging, 166 brainstem function, 572 cortical function cortical motor stimulation, 573 movement-related potentials, 572 voluntary movements, 572 differential diagnosis, 727 involuntary movements, 571 long-latency reflexes, 290-291, 572 myoclonus, 3 prepulse effects, 277 reaction/movement time, 219-220 somatosensory evoked potentials, 63, 572 spinal cord function, 571 Sydenham's chorea, 574 TMS, 104--105 Hydrocephalus TMS, 107 5-Hydroxytryptophan, 3 Hyperekplexia, 479-486 differential diagnosis, 481--485 hereditary hyperekplexia clinical aspects, 479--480 genetics, 480, 719 minor form, 481 neurophysiolgy, 719 physiology, 483--486 sporadic hyperekplexia, 481 TMS,109 Tourette syndrome, 482 treatment, 483 with stiffness stiff person syndrome, 483 stiffness in the newborn, 483 tetanus and strychnine, 483 without stiffness conversion disorder, 482 culture-bound syndromes, 482 reflex myoclonus, 482 startle-induced epilepsy, 481--482 Tourette syndrome, 482 Hyperkinetic movement disorders brain imaging, 166-167 Hypersensitivity focal injection therapy, 661 Hypokinetic movement disorders brain imaging, 165-166 Hypophonia deep brain stimulation, 704 Information processing reaction time, 214 somatosensory evoked potentials, 66

SUBJECT INDEX

Injection therapy, see Focal injection therapy Involuntary movement disorder polysornnography, 144 Isaac's syndrome, see Neuromyotonia Kinesiology, 191-200 apraxia, 194--196 astereognosis, 198 bilateral synergies, 198-200 coordinate systems, 192-193 dynamics, 196 levels of observation, 193 limb and body configuration, 196 manipulative hand movements, 197-198 patient populations, 193 spatial kinematics, 195 technical principles, 191-192 temporal kinematics, 193-194 temporal movement organization, 194--195 Language processing cortico-cortical coherence, 83 Learning cortico-cortical coherence, 83 event-related (de-)synchronization, 26 Lennox-Gastaut syndrome myoclonus, 538 Levodopa dyskinesias, see Parkinson's disease Limb amputation TMS, 102-103, 106 Locomotion gait analysis, 337-352 Long-loop reflexes Parkinson's disease, 429 Long-latency reflexes, 285-292 clinical applications intra-axial lesions, 289 Friedreich's disease, 291 pattern and physiology, 285-289 methods, 288 thenar reflexes, 288-289 Magnetic resonance imaging, see also Brain imaging combination with EEG, MEG, TMS, 718-719 deep brain stimulation, 700 gait disorders, 647 motor cortex output, 109 Parkinson's disease, 700 periodic limb movements, 592 reaction time, 212-214 restless leg syndrome, 592 tics, 554 Magnetoencephalography brain plasticity, 720 combination with fMRIIPET, 718-719

737

SUBJECT INDEX

correlation with EMG, 15-26 hemiparesis, 601 tremor, 368 McArdle's disease stiffness, 474 Microelectrode recording amplification and filtering, 129 commercial systems, 129-130 deep brain nuclei, 127-136 extracellular spikes, 128-129 globus pallidus, 130-132 microelectrode assembly, 127-128 subthalamic nucleus, 132-135 surgical technique, 130 thalamus, 135 Microneurography, 153-159 clinical studies dystonia, 159 Parkinson's disease, 158 spasticity, 157-158 clinical value, 159 fusimotor involvement, 153-157 gamma drive to resting muscle, 153 reflex reinforcement, 154-155 spindle discharge, 153-154 voluntary contractions, 155-156 skin afferents and motor control, 157 technique, 153 Migraine TMS, 104, 107 Mitochondrial disorders TMS, 98-101 Motor control cortico-muscular coherence, 87-92 electrocorticography, 31-41 gait analysis, 337-352 posturography, 295-330 reaction time, 212-214 Motor evoked potential hemiparesis, 602 Motor parasomnias polysornnography, 144 Motor programming reaction time, 203-221 Motor seizure event-related (de-)synchronization, 26 Movement disorders clinical neurophysiology, 717-722, 727-729 electric/magnetic fields, 16-23 electrocardiography, 31-41 imaging therapy, 171-172 microelectrode recording, 127-136 microneurography, 153-159 polysornnography,139-151

reaction time, 215-220 somatosensory evoked potentials, 45-66 surgery, 127-136 TMS,102 Movement-related cortical potential electric/magnetic fields, 16-18 electrocorticography, 33-35, 38-39,41 Multiple sclerosis cerebellar tremor, 404 gait analysis, 350 long-latency reflexes, 291 TMS, 98-101, 106-107 Multiple system atrophy cerebellar ataxia, 499 parkinsonian syndrome, 437 reaction/movement time, 219-220 Myoclonus, 521-545 Alzheimer's disease, 536 cerebellar ataxia, 513-515 classification (etiologic), 521-524 epileptic myoclonus, 522-523 essential myoclonus, 522-523 physiologic myoclonus, 521-523 symptomatic myoclonus, 524 classification (physiologic), 524, 530-544 absence seizures, 538 Alzheimer's disease, 536 asterixis, 537 cortical with reflex, 530-532 cortical without reflex, 532 corticobasal degeneration, 537 Creutzfeldt-Jakob disease, 536 epilepsy, 538 essential myoclonus, 539 focal motor seizures, 534 opsoclonus-myoclonus syndrome, 540 peripheral myoclonus, 543 primary generalized epileptic myoclonus, 539 primary generalized seizures, 538 propriospinal myoclonus, 540 reticular reflex myoclonus, 540 segmental myoclonus, 542 subacute sclerosing panencephalitis, 536 subcortical reflex myoclonus, 542 cortical reflex myoclonus, 22-23 definition, 521 EEG,528 EMG, 10, 12, 524 facial action myoclonus, 567 hyperekplexia, 482 jerk-locked back averaging, 21-22 long-latency reflexes, 291 methodology multichannel surface EMG, 525-528

738 basic properties, 525 long latency, 528 multichannel surface, 525 spatial aspects, 527 timing, 527 EEG, 528-529 EEG-EMG polygraphy, 528 somatosensory evoked potential, 529 negative myoclonus, 23 neurophysiology, 521-545 Parkinson's disease, 534 parkinsonian syndrome, 445 psychogenic myoclonus, 633, 543-544 research methods, 545 somatosensory evoked potentials, 62-63, 529 TMS, 107, 109

Myopathy gait disorders, 646

Myotonia, see Stiffness Myotonic dystrophy stiffness, 473 TMS, 98-101 Narcolepsy multiple sleep latency test, 149

Nerve stimulation long-latency reflexes, 285-292 Neurodegenerative disease parkinsonian syndrome, 437-447 TMS, 98-101 Neuromuscular dysfunction gait analysis, 352 Neuromyotonia stiffness, 470 Neurophysiology (clinical) brain plasticity, 719-722 future applications, 727-729 genetics, 719 MRIlPET with EEG, MEG, TMS, 718-719 myoclonus, 521-546 parkinsonian tremor, 377-390 research studies, 717-722 Nystagmus focal injection therapy, 673

Obsessive compulsive disorder tics, 550 Optic ataxia kinesiology, 194 Pain disorders focal injection therapy, 673

SUBJECT INDEX

Parkinson's disease, 417-428, see also Parkinsonian syndrome ablative surgery, 171 Bereitschaftspotential, 18, 422 bradykinesia, 417-428 ballistic movements, 419 complex movements, 419-420 extra deficits, 421 EEG studies Bereitschaftspotential, 422 contingent negative variation, 423, 424 event-related (de-)synchronization, 424 lateralized readiness potential, 423 somatosensory evoked potentials, 425 movement execution muscle weakness, 418 rigidity, 418 variability, 418 reaction time, 417 sensorimotor processing, 421 transcranial magnetic stimulation before movement, 425 interruption of movement, 427 thresholds, inhibition and silent period, 425 brain imaging, 165, 169-171 cortico-cortical coherence, 83 cortico-muscular coherence, 91 cranial nerve reflexes, 251 cued movements, 19 deep brain stimulation, 697-709, 722 differential diagnosis, 727 event-related (de-)synchronization, 25 fetal transplantation, 172 focal injection therapy, 667 gait disorders, 430, 644 idiopathic, 437 kinesiology, 196 levodopa dyskinesias, 574-576 blink reflex, 576 EMG,576 neurophysiology, 576 long-latency reflexes, 290-291 myoclonus, 534 posture gait, 430 quiet stance, 428 response to perturbations, 430 posturography, 329 prepulse effects, 277 reaction/movement time, 215-219 rigidity, 427-428 somatosensory evoked potentials, 63-64 spinal reflexes, 242 stiffness, 470

SUBJECT INDEX

TMS, 101, 104-105, 110 tremor, 377-390, 408-411 Parkinsonian syndrome, 437-447, see also Parkinson's

disease bradykinesia and rigidity, 438-439 brainstem disorders, 439 corticobasal ganglionic degeneration alien hand sign, 446 myoclonus, 445 differential diagnosis, 727 focal injection therapy, 671 multisystem atrophy autonomic dysfunction, 443-444 minipolymyoclonus, 445 sphincter EMG, 444 progressive supranuclear palsy, 440-442 eye(-lid) movement disorders, 440 startle reaction, 441 Parkinson-plus conditions, see Parkinsonian

syndrome Paroxysmal dyskinesias electrocorticography, 39-40

Pathophysiology parkinsonian tremor, 377-390 Peripheral neuropathy long-latency reflexes, 290

Phenol focal injection therapy, 655-676 Periodic limb movements, 583-594, see also Restless leg syndrome autonomic function, 594 cyclic alternating pattern, 587 definition and history, 583-585 diagnosis, 585 neurophysiology, 585 pathophysiology, 585 diagnosis, 585-587 locus, 590-592 pharmalogical probes, 593 polysomnography, 144-146,585 severity assessment, 588 sleep/wake stage, 587 therapeutic impact, 590 tics, 553 Polysomnography, 139-149 artifacts and pitfalls, 148-149 maintenance of wakefulness test actigraphy, 150 multiple sleep latency test indications, 149-150 reliability and validity, 150 technique, 149 noctural movement disorders, 145-148

739 computerization, 147-148 dissociative disorders, 146 diurnal movement disorders, 147 parasomnias, 146 periodic limb movement, 145 restless leg syndrome, 145 recording techniques, 139-142 beginning and ending, 142 body position, 142 EEG, 140-141 electrocardiography, 141 electrooculography, 141 EMG,141 respiratory monitoring, 141-142 snoring, 142 sleep disorders, 139-151 sleep stage scoring, 142-145 indications, 143-145 periodic limb movements, 142-143 Positron emission tomography see also Brain imaging combination with EEG, MEG, TMS, 718-719 motor cortex output, 109 periodic limb movements, 593 psychogenic movement disorder, 636 reaction time, 212-214 restless leg syndrome, 593 tics, 554 tremor, 372, 382 Posturography, 295-330 dynamic posturography moving platforms, 297-300 postural perturbation types, 300 limitations, 327-328 manipulations afferent feedback signals, 315-323 combined, 320 proprioceptive, 318-320 vestibular, 320 visual, 315-318 weightlessness, 320 postural and cognitive set, 323-325 physiological and clinical utility, 328-330 balance disorders, 328-330 diagnosis, 329 normal postural control, 328 patient management selective lesions, 328 recording equipment EMG, 307-315 general principles, 301-302 kinematics, 302 kinetics, 303-307 multimodal posturography, 315

740 safety, 325-327 static posturography, 297 static vs. dynamic, 296 Prepulse effects, see also Startle reflex blink and other facial reflexes, 274-276 blink reflex excitability, 277 physiology, 276-277 postpulse effects, 278-279 startle reaction, 274

Progressive supranuclear palsy brain imaging, 166 event-related (de-)synchronization, 25 focal injection therapy, 672 parkinsonian syndrome, 437 reaction/movement time, 219-220 Psychogenic movement disorder, 629-637 clinical aspects medical history, 630 physical examination, 630-632 functional imaging, 636 jerks, 633-634 neurophysiology, 632 spasms, 634-636 stiffness, 470 treatment, 637 tremor, 632-633

SUBJECT INDEX

Response initiation reaction time, 203-221 Rigidity parkinsonian syndrome, 438 stiffness, 467, 470 Restless leg syndrome, 583-594, see also Periodic limb

movements circadian rhythm, 588 definition and history, 583-585 diagnosis, 585 neurophysiology, 585 pathophysiology, 585 diagnosis, 585-587 locus, 592 pharrnalogical probes, 593 polysornnography, 144-146,585 severity assessment, 588 therapeutic impact, 590 TMS,106

Schizophrenia TMS, 107

Schwartz-Jampel syndrome stiffness, 473

Seizure po1ysomnography, 144

Sensory ataxia Radiculopathy long-latency reflexes, 290

Reaction time, 203-221 brain injury, 22Q.-.221 cerebellar ataxia, 502, 510 common tasks go/no go, 209 sequence length effect, 208 simple, uncued and precued choice, 206-208 functional anatomy, 212-214 historical background, 203-204 influencing factors, 209-211 procedural factors, 210 subject-/operator-related,210 information processing, 214 measurement, 204-205 motor programs, 205-206 movement disorders effectoflevodopa,218-219 other movement disorders, 219-220 Parkinson's disease, 215-221 movement time, 204, 215-220, 502 startle reflex, 272-274

Reciprocal inhibition curve tremor, 372

REM sleep, see Sleep disorders

gait disorders, 646

Sensory organization test posturography, 323

Single photon emission tomography, see also Brain imaging Huntington's disease, 164-165 periodic limb movements, 592 psychogenic movement disorder, 636 restless leg syndrome, 592

Sleep disorders acce1erometry, 186 periodic limb movements, 583-594 polysornnography, 139-151 restless leg syndrome, 583-594 tics, 553 Somatosensory evoked potentials, 45-66 brain plasticity, 720 dystonia, 64-65, 453 hemispheric lesions, 61-62 Huntington's disease, 63, 572 lower limb stimulation, 57-61 N33 scalp potential, 59 N50 and P60 potentials, 61 N7 potential, 58 P39 potential, 59-61 scalp far-field P30 potential, 59 spinal potentials, 58-59

SUBJECT INDEX

motor cortex output, 109 movement disorders, 61-66 myoclonus, 52, 529 Parkinson's disease, 63-64, 427 recording procedures, 45-49 analysis time and sampling, 48 electrical stimuli, 45-47 electrode placement, 48-49 natural stimuli, 47 stimulus rate, 47-48 stimulus types and peripheral fibers, 45 startle reflex, 269 upper limb stimulation, 49-57 conducted cervical NIl potential, 51 early cortical potentials, 54 far-field positive scalp potentials, 51-53 frontal N30 potential, 55-57 Nl8 scalp potential, 53-54 N20-P20 and P22 potentials, 54-55 parietal P24 and P27 potentials, 55 peripheral N9 potential, 50 spinal segmental N 13 potential, 50-51 upper cervical N 13 potential, 51 Spastic paraplegia focal injection therapy, 666 TMS,106 Spasticity, 615-627 cranial hyperkinetic syndromes, 559-562 diseases with, 622-623 focal injection therapy, 665-666, 670 long-latency reflexes, 290 measurement, 620 microneurography, 158 motor disability, 619 pathophysiology, 617-619 psychogenic movement disorder, 634-636 spastic syndrome, 615 terminology, 615 treatment, 622-627 botulinum A toxin, 626 injection of neurolytic drugs, 625 intrathecal baclofen, 626 oral therapy, 624 upper motor neuron syndrome, 615 Sphincter disorder focal injection therapy, 672 Spinal cord injury long-latency reflexes, 290 stiffness, 464 TMS,103 Spinal reflexes, 231-243 cutaneomuscular reflexes, 238-239 silent period, 238 flexor reflexes, 238-239

741 H-reflexes: methodology motor neuron excitability, 235-236 recruitment curves, 233-234 techniques, 234-235 H-reflexes: vibration inhibition, 237-238 mixed nerve silent period, 240 reciprocal inhibition, 241-243 single motor units peristimulus time histograms, 236 unitary H-reflex, 237 stretch: T-waves and H-reflexes, 231-232 Startle disease, see Hyperekplexia Startle reflex, 267-279, see also Prepulse effects audiospinal reflex, 274 circuits, 267-269 long-latency aspects, 269 PI potential, 270-271 reaction time, 272-274 tics, 553 variability and habituation, 271 Stereotactic surgery tremor, 366 Stiff person syndrome, 464--469, see also Stiffness with continuous motor unit activity gait disorders, 644 hyperekplexia, 483 psychogenic movement disorder, 636 TMS, 104, 106 Stiffness with continuous motor unit activity, 463--475, see also Rigidity central stiffness differential diagnosis, 469 spinal cord pathology, 464 stiff person plus syndrome, 466-469 stiff person syndrome, 464-466 tetanus, 463 treatment, 469 peripheral stiffness differential diagnosis, 472-474 McArdle's disease, 474 neuromyotonia,470-472 other causes, 473-474 Schwartz-Jampe! syndrome, 473 treatment, 472 Strabismus focal injection therapy, 673 Stretch reflexes long-latency reflexes, 285-292 Stroke brain plasticity, 720 cortico-muscular coherence, 91 event-related (de-)synchronization, 26 hemiparesis, 601-610

SUBJECT INDEX

742 TMS, 98-103,105,107-108,110 tremor, 370

Subacute sclerosing panencephalitis myoclonus, 537

Subthalamic nucleus deep brain stimulation, 697-709 microelectrode recording, 132-135 Parkinson's disease, 697-709

Sydenham chorea differential diagnosis, 727

Tardive dyskinesia differential diagnosis, 727

Tetanus hyperekplexia, 483 stiffness, 463 TMS,104 Thalamus, see also Subthalamic nucleus microelectrode recording, 136 tremor, 366, 384, 404 Tics, 549-555 ADHD, 550-553 basal ganglia, 555 blink reflex, 553 clinical features, 549 cortical inhibition, 554 differential diagnosis, 727 EEG and cortical potentials, 551-552 EMG, 10,550 neuroimaging, 554 physiology of movement, 550 sleep, 553 startle reflex, 552-553 synthesis and speculation, 555 TMS,106 Tourette syndrome, 549-555 treatment, 555

Torticollis focal injection therapy, 667, 672

Tourette syndrome hyperekplexia, 482 tics, 549-555

Transcranial electric stimulation cerebellar ataxia, 515 TMS,96

Transcranial magnetic stimulation, 95-110 Alzheimer's disease, 102, 106 amyotrophic lateral sclerosis, 102, 104-106 brain plasticity, 720-721 cerebellar ataxia, 513, 515 combination with fMRIlPET, 718-719 corticospinal tract, 95-101 central motor conduction time, 96-98 single motor unit recordings, 99-101

technical principles, 95-96 triple stimulation technique, 99 dystonia, 455-456 hemiparesis, 601 motor cortex connectivity, 106-109 cerebellum, 107 cutaneous and muscle afferents, 108 different motor representations, 106 distant target structures, 109 inter-hemispheric, 107 ipsilateral spinal motoneurons, 107-108 photic, auditory, nociceptive, 108-109 premotor cortex and SMA, 106-107 motor cortex excitability, 101-106 cortical silent period, 103-104 i-wave facilitation, 106 motor evoked potential intensity curve, 102 motor evoked potential mapping, 102-103 motor threshold, 101-1 02 paired-pulse inhibition, 104-106 paired-pulse excitability, 104 repetitive TMS, 109-110 restless leg syndrome, 592

Tremor accelerometry, 185-186,367,381,404,409 cerebellar tremor, 403-405,511-515 cortico-muscular coherence, 91 deep brain stimulation, 703, 722 differential diagnosis, 727 drug and toxic tremors, 407-408 dystonic tremor, 405 EMG, 8-11, 359-362 essential tremor, 365-369 alcohol and therapy, 365-366 clinical features, 365 definition, 365 functional imaging, 368-369 genetics, 366 pathophysiology, 366 focal injection therapy, 671 hemiparesis, 370 Holmes' tremor, 399-401 long-latency reflexes, 290-291 palatal tremor, 401-403, 511 parkinsonian tremor, 377-390, 408-411, 703 animal models, 386-387 basal ganglia cells, 383-385 clinical neurophysiology, 378-383 central oscillators, 380-381 coherence analysis, 381-382 functional imaging, 382-383 recording reflex pathways, 379-380 electrophysiological diagnosis, 387-389

743

SUBJECT INDEX

pathophysiology, 389-390 stereotactic surgery, 385-386 peripheral neuropathy, 405-407 physiologic (enhanced) tremor, 357-362, 381 clinical guidelines, 361 neurogenic oscillation, 358-360 passive mechanical oscillation, 357 posturography, 330 primary orthostatic tremor, 397-399 primary writing tremor, 369-373 alcohol and therapy, 370-371 clinical features, 369-370 definition, 369 etiology, 370 functional imaging, 372-373 pathophysiology, 371

psychogenic tremor, 408--411, 631-632 task and position tremors, 408 uncommon tremors, 397--411

Ultrasound kinesiology, 192

Walking, see Gait disorders Wolff-Parkinson-White syndrome focal injection therapy, 667 Writer's cramp, see also Tremor, and Dystonia dystonia, 452 movement-related cortical potential, 18 TMS,11O