Effect of Muscle Activity Immediately After Botulinum ToxinInjection for Writer’s Cramp

*†Robert Chen, MBBChir, MSc, FRCPC, *Barbara I. Karp, MD, *Susanne R. Goldstein, MD,*William Bara-Jimenez, MD, *Zaneb Yaseen, and *Mark Hallett, MD

*Human Motor Control Section, National Institute of Neurological Disorders and Stroke, Bethesda, Maryland, U.S.A.; and†Division of Neurology, The Toronto Hospital, University of Toronto, Ontario, Canada

Summary: Animal and human studies have shown that nervestimulation enhances some effects of botulinum toxin (btx A)injection. Voluntary muscle activity might work similarly andwould focus the effect of an injection into the active muscles.We studied the effects of exercise immediately after btx Ainjection in eight patients with writer’s cramp with establishedresponse to btx A over two injection cycles with a single-blinded, randomized, crossover design. Immediately after thefirst study injection, they were randomly assigned to write con-tinuously for 30 min or have their hand and forearm immobi-lized for 30 min. Following the second injection, they wereassigned the alternate condition. Patients were assessed justbefore each injection, and at 2 weeks, 6 weeks, and 3 monthspost-injection. Assessment included objective strength testing,self-reported rating of benefit and weakness, and blinded evalu-ation of videotapes and writing samples of the patients writing

a standard passage. Strength testing showed that the maximumweakness occurred at 2 weeks post-injection, but the benefitwas maximum at 6 weeks post-injection. The “write” conditionresulted in greater reduction in strength than the “rest” condi-tion. Btx A treatment led to improvement in self-reported rat-ings, writer’s cramp rating scale scores by blinded raters, andreduction in writing time, but the differences between the“write” and “rest” conditions were not significant. We concludethat voluntary muscle activity immediately after btx A injectionleads to greater reduction in muscle strength. Our findings raisethe possibility that voluntary muscle activation may allow re-duction of btx A doses and favorably alter the balance of ben-efit and side effects of btx A injections.Key Words: Dysto-nia—Botulinum toxin—Treatment—Writer’s cramp—Exercise—Muscle strength.

Local injection of botulinum toxin type A (btx A) is aneffective treatment for hemifacial spasm1,2and focal dys-tonia.3–12 The clinical benefit is likely related to weak-ening of the injected muscles.3 However, to eradicateinvoluntary movement by this mechanism alone, btx Awould have to cause almost complete paralysis of vol-untary movement.13 There is also a poor correlation be-tween the extent of weakness and the beneficial ef-fects.13,14 It has been suggested that btx A may prefer-entially block the neuromuscular junction of the fibersinvolved in the involuntary movements.13,15

Activity of the neuromuscular junction increases someof the effects of btx A. In rat phrenic nerve–diaphragmpreparations, nerve stimulation after exposure to btx A

dramatically decreases the time to paralysis proportion-ate to the rate of nerve stimulation.16–18In people, elec-trical stimulation of the injected muscle after btx A treat-ment leads to greater reduction of the compound muscleaction potential (CMAP)19 and improved efficacy in pa-tients with lower limb spasticity.20 However, the effectsof voluntary muscle activation has not been studied.

In the present study, we investigated the effects ofactivation of dystonic muscles by writing immediatelyafter btx A injection in patients with writer’s cramp.Botulinum toxin has been shown to be an effective treat-ment for writer’s cramp in open-labeled2,9,10,21 anddouble-blinded studies.11,22,23Writer’s cramp is a task-specific dystonia.21,24 We chose patients with writer’scramp for this study because the dystonic muscles are notactive at rest but can be selectively activated by specificactivities such as writing.21,24 We reasoned that thismight enhance the weakness specifically for the dystonicmuscles and might improve efficacy.

Received August 19, 1998; revision received December 4, 1998.Accepted December 15, 1998.

Address correspondence and reprint requests to Mark Hallett, MD,Clinical Director, NINDS, Bldg. 10, Room 5N226, 10 Center Drive,MSC 1428, Bethesda, MD 20892-1428, U.S.A.

Movement DisordersVol. 14, No. 2, 1999, pp. 307–312© 1999 Movement Disorder Society

307

METHODS

Patients

Patients with writer’s cramp were recruited from theNINDS botulinum toxin clinic. Inclusion criteria were(1) diagnosis of writer’s cramp24,25 with no dystonia atrest, (2) identical combination of muscles and btx Adoses were used in the last three injection sessions, and(3) improvement following each of the last three injec-tions. Patients were eligible for the study if they rated thebenefit of their last three treatment sessions as mild orhigher on a five-point scale (none, minimal, mild, mod-erate, excellent) andù30% on a visual analog scale (0–100%). Criteria 2 and 3 were used to select patientswhose optimal combination of muscles and btx A doseshad been established. All patients gave their written in-formed consent and the protocol was approved by theInstitutional Review Board.

Study Protocol

Each patient was studied over a period of two injectioncycles with a randomly assigned crossover design. Thepatients were told that we were studying the effects ofbtx A for treatment of writer’s cramp but were unawareof the hypothesis being tested. The identical combinationof muscles and btx A (BOTOX, Allergan, Irvine, CA,U.S.A.; dilution 10 units/0.1 mL) doses were adminis-tered by the same physician in both injection sessions.Electromyographic guidance was used for btx A injec-tion.10,21 After the first injection, the patient was ran-domly assigned to either the “write” or the “rest” condi-tion. The predetermined allocation schedule was basedon block randomization with a block size of four. In the“write” condition, the patient was required to write astandard sentence (“Today is a nice day in Washington”)continuously for 30 minutes immediately after btx Ainjection. In the “rest” condition, the patient’s forearmand hand were immobilized with a splint to prevent wristor hand movements for 30 minutes immediately after btxA injection. The patients returned for the next injectionas the effects of the first injection wore off. They wereinstructed that their participation in the study should notinfluence the timing for the next injection. At the secondinjection session, the patients were assigned the alternatecondition.

Patient Assessment

Patients were assessed immediately before and 2weeks, 6 weeks, and 3 months after each btx A injection.At each assessment, the strength of wrist flexion, wristextension, and hand grip were measured, the writing of astandard passage was videotaped, and the extent of ben-

efit and weakness were determined with the self-administered rating scales. The extent of benefit andweakness were reported on a five-point scale (none,minimal, mild, moderate, excellent/severe) and with avisual analog scale from 0 (no response) to 100% (nor-mal writing). Maximum grip strength was measured witha handheld dynamometer. Maximum wrist flexion andextension strength were measured with a rigid frame dy-namometer (Myometer, Penny & Giles, Christchurch,U.K.). For each strength measurement, the mean of threetrials of maximum effort was used. Writing a standardpassage (“Mary has a little lamb, with fleece as white assnow” five times and 10 interconnected “l”) was video-taped using a standard set of pen, chair and table adjustedto the same height. The patients were informed that thewriting samples would be rated for the writing qualityand speed, and that they should produce their best writ-ing in the shortest possible time.

Dystonia Rating

The videotape segments of writing were rated usingthe writing movement score of the writer’s cramp ratingscale (WCRS)26 by two raters blinded to the interventionassignment and the time before or after injection. Writingtimes were measured with a stopwatch during the play-back of the videotape. Three of 63 videotape segments(those for patient 1 at 6 weeks post-injection; patient 5 at2 weeks post-injection; patient 7 at 3 months post-injection; all for the rest condition) were misplaced andcould not be rated. The quality of the writing was as-sessed by three raters blinded to the intervention assign-ment. Five attributes of writing quality were evaluated:even spacing, smoothness, straight baseline, well-formedloops, and consistency. Each attribute was rated as good(2), moderate (1), or poor (0), giving a total writing qual-ity score ranging from 0 (worst) to 10 (best).

Data Analysis

Wrist flexion, wrist extension, and hand grip strengthwere expressed as a percentage of the pre-injection mea-surement. The movement (wrist flexion, extension, orgrip) most affected by btx A injection, defined as themovement with greatest percentage decline in force inthe post-injection measurements, was identified for eachpatient and was used for the analysis of the effects oftime and condition on strength.

The interrater reliability for each component of theWCRS and the total score of the writing quality scalewas calculated using Cohen’s weighted k coefficient.27

According to Altman,28 k values of 0.21–0.40 indicatefair agreement, 0.41–0.60 indicate moderate agreement,0.61–0.80 indicate good agreement, and 0.8–1.0 indicate

R. CHEN ET AL.308

Movement Disorders, Vol. 14, No. 2, 1999

very good agreement. For comparison between condi-tions and times after injection, the writing movementscores and the writing quality scores from different raterswere averaged and expressed as a percentage of the pre-injection score. Similarly, the writing time for each as-sessment was expressed as a percentage of the pre-injection baseline.

Statistical AnalysisValues are expressed as mean ± standard error. The

strength measurement, subjective ratings of benefit orweakness, dystonia rating scale, and writing times wereanalyzed with analysis of variance with repeated mea-sures. Time (pre-injection, 2 weeks, 6 weeks, 3 months)was the within-subject factor (repeated measure) andcondition (write/rest) was the between-subject factor.Differences were considered significant if pø0.05.

RESULTSTen patients enrolled in the study. Two were excluded

from the analysis. Patient 10 withdrew from the studyafter one injection cycle. Patient 4 reported no effectfrom one of the injections and seven of the 12 nonstudypatients treated on the same day also reported little or noeffects from that session. The reason for the ineffectivetreatment is unclear but may be the result of a defectivebatch of btx A. The demographic information, musclesinjected, btx A doses used, order of intervention assign-ment, time between the first and second injections, andnumber of sentences written in the “write” condition forthe remaining eight patients is summarized in Table 1.

Strength MeasurementFor each patient, the movement (wrist flexion, wrist

extension, or hand grip) most weakened by btx A injec-tion corresponded with the muscles injected, as is shownin Table 1. The strength at 2 weeks, 6 weeks, and 3months after injection, expressed as a percentage of thepre-injection strength, is shown in Figure 1. The weak-

ness induced by btx A was most pronounced at 2 weeksand showed partial recovery at 6 weeks and 3 months. Ateach time interval, there was greater reduction in strengthafter “write” than “rest.” Analysis of variance (ANOVA)showed a significant effect of time (F4 62.3, p 40.001) and condition (F4 6.6, p4 0.037). The time xcondition interaction was not significant.

Self-Reported Benefit and Weakness, Writer’sCramp Rating Scale, Writing Time, and

Writing Quality

The patients’ ratings for benefit and weakness at dif-ferent times after injection, the writing movement scores,the writing time for the standard passage, and the writingquality scores are shown in Table 2. ANOVA showedsignificant effects of time on the benefit and weaknessratings in the five-point scale (benefit: F4 5.0, p 40.023; weakness: F4 4.3, p 4 0.034) and the visualanalog scale (benefit: F4 5.2, p4 0.022; weakness: F4 4.3, p4 0.034) on the writing movement scores (F47.6, p 4 0.004) and on writing time (F4 13.5, p40.004) but not on writing quality. The effect of conditionwas not significant for any of these parameters.

For the writer’s cramp rating scale, the k coefficient

TABLE 1. Summary of patient information

Patient Age, sexMuscles injectedand dose (units)

Order ofintervention

Time betweenfirst and secondinjections (mos.)

No. ofsentenceswritten

Movementmost weakened

by btx A

1 56, F FDS 20 Rest/write 3 157 Grip2 41, F FDS 15, FPL 5 Write/rest 3 104 Grip3 43, F FDS 35, FPL 10, FPB 5 Write/rest 6 102 Grip5 34, M FCU 30, FCR 10, FDS 15, FPL 5 Write/rest 11 76 Wrist flexion6 37, F FDS 20, FDP 5, FPL 5, FPB 2.5 Rest/write 6 112 Grip7 37, F FDS 25, FDP 5, FPL 25 Write/rest 7 110 Grip8 36, F FDS 20, FCU 20, FCR 20 Rest/write 4 91 Grip9 52, M FCU 20, FCR 20, ECU 20 Write/rest 4 79 Wrist flexion

ECU, extensor carpi ulnaris; FCR, flexor carpi radialis; FCU, flexor carpi ulnaris; FDP, flexor digitorium profundus; FDS, flexor digitoriumsuperficialis; FPB, flexor pollicis brevis; FPL, flexor pollicis longus.

FIG 1. Strength at different times after btx A injection. The solid barsrepresent the “write” condition and the hatched bars represent the “rest”condition. Error bars indicate one standard error.

EFFECTS OF EXERCISE AFTER BOTULINUM TOXIN INJECTION 309

Movement Disorders, Vol. 14, No. 2, 1999

for interrater reliability indicates good agreement for thewrist (k 4 0.60) and tremor (k4 0.63) scores; moderateagreement for the finger II (k4 0.56) and latency ofdystonia (k4 0.46) scores, and fair agreement for thefinger I (k 4 0.33) and finger III (k4 0.23) scores.Interrater reliability for the elbow score could not beassessed because none of the patients had dystonic pos-ture at the elbow. The k coefficient for the writing qualityscale scores was 0.46, indicating moderate agreement.

DISCUSSION

This is the first demonstration in people that voluntarymuscle activity enhances the effect of btx A. Two pre-vious studies have shown that electrical stimulation overthe target muscle increases the effect of btx A. Hesse etal.20 reported that in spastic patients, electrical stimula-tion of lower limb muscles for 30 minutes six times perday for 3 days following btx A injection resulted ingreater improvement in gait velocity and reduction inmuscle tone. Eleopra et al.19 showed that periodic elec-trical stimulation over the target muscle for 24 hoursafter btx A injection led to greater reduction in CMAPs.Compared with electrical stimulation, voluntary muscleactivation is easier to use and more acceptable to patientsthan electrical stimulation. It also selectively activatesthe dystonic muscles, including deep muscles which maybe inaccessible to direct muscle stimulation. Althoughprevious studies used stimulation for 3 days20 or 24hours,19 we found that increasing the activity of the neu-romuscular junction for only 30 minutes after btx A in-jection is sufficient to cause increased weakness. This isconsistent with the findings in the rat phrenic nerve–

diaphragm preparation in which the effects of nervestimulation were most pronounced immediatelyafter contact with btx A, and the effects diminishedwith a longer time between toxin contact and nervestimulation.16

The interaction between btx A and the neuromuscularjunction initially involves binding to the motor nerveterminal membrane followed by internalization of themolecule by receptor-mediated endocytosis and an intra-cellular poisoning step which involves selective cleavingof the synaptic protein SNAP-25, thereby preventing re-lease of acetylcholine from the motor nerve termi-nal.17,18,29,30Nerve stimulation accelerates the internal-ization of btx A into the nerve terminal.18 It is likely theneuromuscular junction activity induced by writing ledto increased btx A uptake, particularly into those motorunits involved in the dystonic movements.

The objective strength measurements showed greaterreduction in strength at 2 weeks than 6 weeks post-injection consistent with previous observations that themaximum reduction of CMAP occurs 1–3 weeks post-injection followed by slow recovery to the pre-injectionlevel.19,31However, the extent of improvement as shownby self-reported ratings, WCRS, and writing time mea-surements were more marked at 6 weeks than at 2 weekspost-injection. It thus appears that the optimal balancebetween benefit and weakness is achieved in the pro-longed recovery phase after maximum weakness ratherthan the time of maximum weakness. It has been sug-gested that the late improvement may also represent re-organization in the spinal cord or at the cortical levelfollowing btx A injection.14,32,33

TABLE 2. Effects of condition and time after injection on different measurements

Measurement

Write Rest

2 wks 6 wks 3 mos 2 wks 6 wks 3 mos

Self-reported benefit(5-point scale)*

2.1 ± 0.2 3 ± 0.0 2.1 ± 0.5 2 ± 0.4 2.6 ± 0.2 2.2 ± 0.4

Self-reported benefit(visual analog scale)†

40 ± 6.6% 48 ± 6.5% 43 ± 10% 39 ± 8.4% 55 ± 3.8% 41 ± 8.7%

Self-reported weakness(5-point scale)*

2.3 ± 0.3 1.9 ± 0.2 1.1 ± 0.4 1.9 ± 0.4 2.2 ± 0.3 1.6 ± 0.3

Self-reported weakness(visual analog scale)†

39 ± 7.2% 33 ± 7.2% 21 ± 6.6% 36 ± 8.8% 38 ± 6.5% 24 ± 6.5%

Writing movement score‡ 4.9 ± 1.1 5.7 ± 0.7 7.8 ± 0.7 4.9 ± 1.1 6.8 ± 1.6 8.1 ± 1.3(63 ± 17%) (69 ± 13%) (92 ± 9%) (76 ± 25%) (88 ± 22%) (110 ± 19%)

Writing time (sec)‡ 147 ± 18 138 ± 13 143 ± 12 140 ± 10 134 ± 10.6 149 ± 15(98 ± 1.7%) (94 ± 0.4%) (99 ± 5.5%) (100 ± 5.7%) (92 ± 3.7%) (102 ± 2.6%)

Writing quality scale‡ 6.0 ± 0.7 6.2 ± 0.6 6.0 ± 0.5 6.3 ± 0.3 6.3 ± 0.4 5.9 ± 0.7(111 ± 5.5%) (117 ± 8.7%) (112 ± 7.4%) (110 ± 10%) (129 ± 17%) (124 ± 25%)

Values are mean ± standard error.* 1 4 none, 24 minimal, 34 mild, 4 4 moderate, 54 excellent/severe.† From 0 (no improvement or no weakness) to 100% (normal use or inability to move).‡ The data in the parentheses are the percentages of the preinjection baseline values, calculated separately for each patient and then averaged.

R. CHEN ET AL.310

Movement Disorders, Vol. 14, No. 2, 1999

The significant effect of time after injection on WCRSscores and writing time confirms previous findings thatbtx A injection is a useful treatment for writer’scramp2,9–11,21–23and that the WCRS can detect changesin patients following treatment.26 The interrater reliabil-ity for most subcomponents of the WCRS was in thegood to moderate range, similar to the report of Wissel etal.26 The writing quality scale we used also showed mod-erate interrater agreement. Although there was a trendtoward improvement in writing quality following btx Atreatment, the changes were not statistically significant.This may be the result of our small sample size and mayalso reflect the relatively low sensitivity of the ratingscale or compensation by the patient by changing writingspeed or adopting different postures. Rather than return-ing writing legibility to normal, btx A injection may easethe writing process, which is better assessed by videotapeor direct observation.

The conditions of “write” or “rest” had no significanteffect on the self-reported ratings of the benefit or weak-ness, the WCRS scores, or writing times. One possiblereason is that the subjective scores are likely to be morevariable than the more precise and objective strengthmeasurements. The three missing videotape segmentsalso reduced the sensitivity for detecting changes withthe WCRS scores or writing times. In addition, the ben-efit of btx A injection depends on a balance betweenreduction of dystonic movements and muscle weakness.Because we used the previously established optimal btxA doses, any further increase in weakness may have ledto decreased rather than increased benefit. Clearly, thereis not a one-to-one relationship between weakness andbenefit. However, our findings raise the possibility thatvoluntary muscle activation after btx A injection mightallow a reduction of the btx A dose (and lessen the like-lihood of antibody production).34 While we were notable to demonstrate that voluntary muscle activation canfavorably alter the balance of benefit and side effects ofbtx A injection, our results encourage further study.

Acknowledgments:The authors thank Dr. Irene Litvan forassistance with statistical analysis and Devera Schoenberg, MS,for skillful editing. Supported by the NINDS intramuralprogram.

REFERENCES

1. Brin MF, Fahn S, Moskowitz C, et al. Localized injections ofbotulinum toxin for the treatment of focal dystonia and hemifacialspasm.Mov Disord1987;2:237–254.

2. Jankovic J, Schwartz K, Donovan DT. Botulinum toxin treatmentof cranial–cervical dystonia, spasmodic dysphonia, other focal dys-tonias and hemifacial spasm.J Neurol Neurosurg Psychiatry1990;53:633–639.

3. Jankovic J, Brin MF. Therapeutic uses of botulinum toxin.N EnglJ Med1991;324:1186–1194.

4. Tsui JKC, Eisen A, Stoessl AJ, Calne S, Calne DB. Double-blindstudy of botulinum toxin in spasmodic torticollis.Lancet1986;2:245–247.

5. Gleb DJ, Lowenstein DH, Aminoff MJ. Controlled trial of botuli-num toxin injections in the treatment of spasmodic torticollis.Neu-rology 1989;39:80–84.

6. Greene P, Kang U, Fahn S, Brin M, Moskowitz C, Flaster E.Double-blind, placebo-controlled trial of botulinum toxin injec-tions for the treatment of spasmodic torticollis.Neurology1990;40:1213–1218.

7. Blackie JD, Lees AJ. Botulinum toxin treatment in spasmodictorticollis. J Neurol Neurosurg Psychiatry1990;53:640–643.

8. Jankovic J, Orman J. Botulinum toxin A for cranial–cervical dys-tonia: a double-blind, placebo controlled study.Neurology1987;37:616–623.

9. Rivest J, Lees AJ, Marsden CD. Writer’s cramp: treatment withbotulinum toxin injections.Mov Disord1991;6:55–59.

10. Karp BI, Cole RA, Cohen LG, Grill S, Lou JS, Hallett M. Long-term botulinum toxin treatment of focal hand dystonia.Neurology1994;44:70–76.

11. Cole R, Hallett M, Cohen LG. Double-blind trial of botulinumtoxin for treatment of focal hand dystonia.Mov Disord1995;10:466–471.

12. Pullman SL, Greene P, Fahn S, Pedersen SF. Approach to thetreatment of limb disorders with botulinum toxin A.Arch Neurol1996;53:617–624.

13. Hallett M, Glocker FX, Deuschl G. Mechanisms of action of botu-linum toxin. Ann Neurol1994;36:449–450.

14. Giladi N. The mechanism of action of botulinum toxin type A infocal dystonia is most probably through its dual effect on efferent(motor) and afferent pathways at the injected site.J Neurol Sci1997;152:132–135.

15. Glocker FX, Guschlbauer B, Lu¨cking CH, Deuschl G. Effects oflocal injection of botulinum toxin on electrophysiological param-eters in patients with hemifacial spasm: role of synaptic activityand size of motor units.Neurosci Lett1995;187:161–164.

16. Hughes R, Whaler BC. Influence of nerve-ending activity and ofdrugs on the rate of paralysis of rat diaphragm preparations by Cl.botulinum type A toxin.J Physiol (Lond)1962;160:221–233.

17. Simpson LL. Kinetic studies on the interaction between botulinumtoxin type A and the cholinergic neuromuscular junction.J Phar-macol Exp Ther1980;212:16–21.

18. Black JD, Dolly JO. Interaction of 125I-labeled botulinum neuro-toxins with nerve terminals. II. Autoradiographic evidence for itsuptake into motor nerves by acceptor-mediated endocytosis.J CellBiol 1986;103:535–544.

19. Eleopra R, Tugnoli V, De Grandis D. The variability in the clinicaleffect induced by botulinum toxin type A: the role of muscleactivity in humans.Mov Disord1997;12:89–94.

20. Hesse S, Jahnke MT, Luecke D, Mauritz KH. Short-term electricalstimulation enhances the effectiveness of botulinum toxin in thetreatment of lower limb spasticity in hemiparetic patients.NeurosciLett 1995;201:37–40.

21. Cohen LG, Hallett M, Geller BD, Hochberg F. Treatment of focaldystonia of the hand with botulinum toxin injection.J NeurolNeurosurg Psychiatry1989;52:355–363.

22. Yoshimura DM, Aminoff MJ, Olney RK. Botulinum toxin for limbdystonia.Neurology1992;42:627–630.

23. Tsui JKC, Bhatt M, Calne S, Calne DB. Botulinum toxin in thetreatment of writer’s cramp: a double-blind study.Neurology1993;43:183–185.

24. Sheehy MP, Marsden CD. Writers’ cramp—a focal dystonia.Brain1982;105:461–480.

25. Cohen LG, Hallett M. Hand cramps: clinical features and electro-myographic patterns in a focal dystonia.Neurology1988;38:1005–1012.

26. Wissel J, Kabus C, Wenzel R, et al. Botulinum toxin in writer’s

EFFECTS OF EXERCISE AFTER BOTULINUM TOXIN INJECTION 311

Movement Disorders, Vol. 14, No. 2, 1999

cramp: objective response evaluation in 31 patients.J Neurol Neu-rosurg Psychiatry1996;61:172–175.

27. Cohen J. Weighted kappa: nominal scale agreement with provisionfor scale disagreement or partial credit.Psychol Bull1968;70:213–220.

28. Altman DG.Practical Statistics for Medical Research.London:Chapman and Hill, 1992.

29. Coffield JA, Considine RV, Simpson LL. The site and mechanismsof action of botulinum neurotoxin. In: Jankovic J, Hallett M, eds.Therapy with Botulinum Toxin.New York, NY: Marcel Dekker,1994:3–13.

30. Blasi J, Chapman ER, Link E, et al. Botulinum neurotoxin Aselectively cleaves the synaptic protein SNAP-25.Nature 1993;365:160–163.

31. Hamjian JA, Walker FO. Serial neurophysiological studies of in-tramuscular botulinum A toxin in human.Muscle Nerve1994;17:1385–1392.

32. Priori A, Berardelli A, Mercuri B, Manfredi M. Physiological ef-fects produced by botulinum toxin treatment of upper limb dysto-nia. Brain 1995;118:801–807.

33. Byrnes ML, Thickbroom GW, Wilson SA, et al. The corticomotorrepresentation of upper limb muscles in writer’s cramp andchanges following botulinum toxin injection.Brain 1998;121:977–988.

34. Chen R, Karp BI, Hallett M. Botulinum toxin type F for treatmentof dystonia. Long-term experience.Neurology 1998;51:1494–1496.

R. CHEN ET AL.312

Movement Disorders, Vol. 14, No. 2, 1999