Sensory Motor Retuning: A Behavioral Treatment for Focal
Hand Dystonia of Pianists and Guitarists
Victor Candia, MA, Thomas Scha¨fer, Edward Taub, PhD, Harald Rau, PhD, Eckart Altenmu¨ller, MD,
Brigitte Rockstroh, PhD, Thomas Elbert, PhD
ABSTRACT. Candia V, Scha¨ fer T, Taub E, Rau H,
Altenmu¨ ller E, Rockstroh B, Elbert T. Sensory motor returning: a
behavioral treatment for focal hand dystonia of pianists and
guitarists. Arch Phys Med Rehabil 2002;83:1342-8.
Objective: To evaluate the long-term effectiveness of sen-
sory motor retuning (SMR), a new treatment for focal hand
dystonia in musicians.
Design: Prospective case series with an (adventitious) com-
parison group with 3- to 25-month follow-up in piano and
guitar and 0- to 4-month follow-up in ﬂute and oboe players.
Setting: General community in Germany.
Participants: Eleven professional musicians.
Intervention: Immobilization by splints of 1 or more digits
other than the focal dystonic ﬁnger. This ﬁnger carried out
repetitive exercises in coordination with 1 or more of the other
digits for 1
hours a day for 8 consecutive days under
therapist supervision. The subjects then were instructed to
continue practice for 1 hour daily for 1 year.
Main Outcome Measures: Spectral analysis of the output
of a dexterity-displacement device that continuously recorded
digital displacement during ﬁnger movements and a dystonia
evaluation scale on which patients rated how well they had just
performed dystonic movement sequences and repertoire pas-
Results: The 3 wind players (adventitious placebo controls)
did not improve substantially. However, each pianist and gui-
tarist showed marked and signiﬁcant improvement in sponta-
neous repertoire performance without the splint. The ﬁrst sub-
ject is now 25 months posttreatment.
Conclusions: Results suggest that SMR is of value for the
treatment of focal hand dystonia in pianists and guitarists.
Key Words: Behavioral medicine; Focal dystonia; Hand
injuries; Music; Rehabilitation; Repetitive strain injury; Sen-
sory motor performance.
©2002 by the American Congress of Rehabilitation Medi-
cine and the American Academy of Physical Medicine and
FOCAL HAND DYSTONIA in musicians is a set of con-
ditions generally characterized as occupational hand cramp
or focal task-speciﬁc dystonia.
Widely accepted diagnostic
criteria are lacking, but a differentiation is usually made be-
tween digital incoordination syndromes that are idiopathic in
nature and those that have a known or probable organic basis.
Treatments of this condition have included psychotherapy,
and the ad-
ministration of such agents as anticholinergic drugs,
and botulinum toxin.
of success, no treatment has been beneﬁ-
cial on more than a temporary basis. Comprehensive reviews of
the current status of treatment and understanding of mechanism
have been written by Hallett and Chen.
Work by Byl et al
in new-world monkeys showed that
digital motor incoordination resulting from digital overuse is
associated with an induced disorder in the representation of the
digits in area 3b of the somatosensory cortex. The present
research group, using a noninvasive neuroimaging technique
(magnetic source imaging), found that string instrument play-
ers exhibit a use-dependent alteration in the cortical represen-
tational zones of the digits of the left hand that engage in the
dexterity-demanding task of ﬁngering the strings.
It is of
particular interest that use-dependent central nervous system
plasticity also occurred in musicians with focal dystonia; over-
lap or smearing of the homuncular organization of digits oc-
curred in the somatosensory cortex.
mechanisms apparently underlie both the cortical disorder and
the involuntary incoordination of movement, we hypothesized
that a behavioral intervention focusing on movement could
help reduce or eliminate these conditions.
The treatment we report here features a behavioral therapy
approach that leaves free the digit exhibiting the main dystonic
symptoms. We call the therapy sensory motor retuning (SMR).
In it, 1 or several of the other digits are immobilized and
extensive practice is given in performing individual move-
ments of the focal dystonic ﬁnger in coordination with move-
ments of the other ﬁngers. The exercises are performed on each
individual’s musical instrument. We report here the successful
treatment of 6 pianists and 2 guitarists with this intervention. A
preliminary short report
of the data of the ﬁrst 3 pianists and
the 2 guitarists has appeared elsewhere. The same intervention
was administered to 2 ﬂutists and an oboist with lack of
The 11 subjects were right-handed musicians (6 pianists:
patients 1, 2, 5, 6, 7, 11; 2 guitarists: patients 3, 4; 2 ﬂutists:
patients 8, 9; 1 oboist: patient 10); all were professional mu-
sicians, except patient 8. Patients 1, 5, and 10 were women. All
11 subjects had chronic conditions of 2 to 34 years in duration
(table 1). Each had received multiple courses of therapy for
their condition including acupuncture, massage, eurythmics,
neurolinguistic programming, kinetic therapy, physical ther-
From the Department of Psychology, University of Konstanz, Konstanz (Candia,
Scha¨fer, Rau, Rockstroh, Elbert); Institut fu¨r Musikphysiologie und Musiker-Medi-
zin, Hochschule fu¨r Musik und Theater Hannover, Hannover (Altenmu¨ller), Ger-
many; and Birmingham Veterans Affairs Medical Center (Taub), and Department of
Psychology, University of Alabama (Taub), Birmingham, AL.
Supported by the Deutsche Forschungsgemeinschaft and the Rehabilitation Re-
search Service of the Veterans Administration (grant nos. B95-975R, W98-0410).
Presented in part at the 8th European Congress on Performing Arts Medicine and
Physiology of Music Making, October 13-15, 2000, Mainz, Germany.
No commercial party having a direct ﬁnancial interest in the results of the research
supporting this article has or will confer a beneﬁt upon the author(s) or upon any
organization with which the author(s) is/are associated.
Reprint requests to Victor Candia, MA, Dept of Psychology, University of Konstanz,
Box D25, D-78457 Konstanz, Germany, e-mail: firstname.lastname@example.org.
Arch Phys Med Rehabil Vol 83, October 2002
apy, chiropractic, and administration of anticholinergics, xylo-
caine, corticosteroids, and botulinum toxin.
All of the guitarists and pianists were soloists, except for
patient 2, who was a chamber music pianist. They had to limit
or give up their public playing. Two of the wind players were
professional orchestral musicians; one of the ﬂutists was semi-
professional. Subject characteristics are in table 1. Before
project enrollment, each subject was given a neurologic exam-
ination by a neurologist (EA) experienced in musician’s focal
hand dystonia. Symptoms required to make that diagnosis were
painless loss of ﬁne motor coordination of ﬁnger movements,
which occurred exclusively in the task-speciﬁc context of play-
ing the musical instrument; motor disorder restricted to invol-
untary ﬂexion of single digits and (compensatory) extension of
adjacent digits when executing the movements; absence of
other neurologic signs; no sensory deﬁcit in any sensory test,
including the sense of touch, pain, temperature, joint position,
graphesthesia and vibration; and normal or above average
2-point discrimination at the tips of the ﬁngers, the palm, and
the back of the hand. Nerve compression syndromes were
excluded by median and ulnar nerve neurography. Two other
exclusion criteria were (1) the presence of a neurologic condi-
tion other than the focal hand dystonia and (2) maintenance on
a medication for their dystonia. Informed consent was obtained
from patients after they were fully informed, according to the
declaration of Helsinki, about the nature of the study and their
treatment. The treatment protocol was approved by the univer-
sity’s ethics review board.
In SMR therapy, a splint (ﬁg 1A) immobilizes 1 or more
ﬁnger(s) while leaving the remaining digits free (ﬁg 1B). In this
series of patients, we found that the investigators and patients
usually could identify 1 ﬁnger as the focal dystonic digit and 1
or 2 other digits that were involved in performing movements
that compensated for the dysfunction of the focal dystonic digit
(table 1). (In other cases of dystonia, all 4 of the other ﬁngers
can be involved in the compensatory pattern.) The focal dys-
tonic ﬁnger was identiﬁed by immobilizing in turn the other
ﬁngers involved in the abnormal movement pattern and requir-
ing that the subject carries out sequences of alternating move-
ments of the focal dystonic ﬁnger and the remaining digits. In
each case in this series, immobilization of 1 speciﬁcﬁnger
permitted independent movement of the digit the patient had
previously identiﬁed as being dystonic; it was termed the main
compensatory ﬁnger. This was not true after immobilization of
any other digit; immobilization of another compensatory ﬁnger
did enable freer, independent movements of the dystonic ﬁnger
but not nearly to the same extent as immobilization of the main
Treatment involved immobilizing different ﬁnger(s) by
means of the splint. During ﬁxation, the patient held a ﬁnger in
Table 1: Subject Characteristics
Pt Instrument Gender
Dystonia Affected Fingers*
Month of Last
1 Piano Female 40 35 3 Left D2, D3, D4 25
2 Piano Male 47 41 4 Left D3, D4, D5 9
3 Guitar Male 37 25 4 Right D3, D4, D5 19
4 Guitar Male 30 20 3 Right D3, D4 17
5 Piano Female 52 34 8 Right D2, D3, D4 12
6 Piano Male 70 65 34 Right D4, D5 3
7 Piano Male 51 30 3 Right D2, D3 5
8 Flute Male 42 27 5 Left D2, D3, D4 4
9 Flute Male 38 28 2 Left D4 0
10 Oboe Female 38 26 5 Right D2, D3, D4, D5 2
11 Piano Male 39 34 6 Left D3, D4, D5 3
Abbreviation: Pt, patient.
* Focal dystonic ﬁnger is bolded.
Fig 1. Splint device used to immobilize ﬁngers of the dystonic hand.
(A) The device was tailored to the dimensions of each individual’s
hand and the requirements of the instrument he/she played. (B)
Hand in splint with D2 immobilized.
1343BEHAVIORAL TREATMENT FOR FOCAL HAND DYSTONIA, Candia
Arch Phys Med Rehabil Vol 83, October 2002
his/her characteristic rest position for his/her instrument. When
the ﬁnger making the maximum compensatory movements was
immobilized in this position, the focal dystonic ﬁnger could in
each case participate in alternating individual ﬁnger move-
ments with all possible permutations of the other ﬁngers of the
dystonic hand used in the treatment exercises, as was the case
in the earlier diagnostic procedure. In these exercises, the
subject was required to make sequential movements of 2 or 3
digits in extension, including the focal dystonic digit, for a
period of 10 minutes in an ascending and then in a descending
order in continuous repetition; for example, D2, D3, D4, D3,
D2, and so on, with D4 being the focal dystonic ﬁnger and D5
the immobilized main compensatory ﬁnger. After a 2-minute
rest, subjects completed a different sequence of movements of
2or3ﬁngers including the focal dystonic ﬁnger. Five such
blocks of exercises were performed in an hour. In the begin-
ning, performance was paced by a metronome and was begun
at a medium tempo (usually 60bpm); the tempo was then
increased and gradually decreased. (In 7 of the 11 dystonic
musicians, slow controlled movements were more difﬁcult than
fast movements.) The speed sequencing was maintained in an
attempt to force the subject to generate faster and faster and
then slower and slower alternating movements in successive
sequences. The training process was quite fatiguing for each
subject. After completing the ﬁrst 5 blocks of exercises, the
splint was removed and subjects were given a 10-minute rest,
after which they received 4 more 10-minute blocks of exercises
with 2-minute rest intervals between blocks. Various permuta-
tions of possible ﬁnger movements were performed in different
exercise blocks. The subjects were then given a rest of about 40
After the rest, subjects were encouraged to play their instru-
ments without the splint. Initially, they played 15 to 30 seconds
from a self-selected musical piece. If they could not do this,
they were encouraged to try a second time. After 2 successful
repetitions, they were asked to attempt a different segment and
were then asked to play portions of musical pieces of longer
and longer duration until they had played for 15 minutes
(excluding rest intervals). Some subjects were disconcerted by
this process, being surprised by their new facility and also
fearing a return of dystonic movements; however, they were
encouraged to continue. The complexity and duration of the
practice segments within the unsplinted performance period
was based on the therapist’s judgment of therapeutic progress.
We introduced the requirement for periods of unsplinted per-
formance for 2 reasons: ﬁrst, because it represents the real-
world context, and second, because it was very rewarding for
the musicians to experience an unexpected ability to play
frequently practiced musical pieces with a reduced number of
abnormal movements or postures of the ﬁngers. Consequently,
it encouraged continuation of and compliance with the details
of the therapy. After a 5-minute rest, and if the subject was not
too fatigued, the splint was replaced and a second series of
alternating digital maneuvers, each for a duration of 5 minutes,
was performed. This regimen continued for 8 consecutive days
(14d for patient 6, 6d for patient 11 because of an injury while
The therapeutic exercises for the ﬂute players were much the
same as for the pianists and guitarists. The wind players blew
into their instruments and produced tones only at infrequent
intervals during the exercises. We decided that it was unnec-
essary for these individuals to repeatedly sustain the very
taxing, prolonged, breath exhalations that would enable the
same type of digital exercises peformed by pianists and gui-
tarists for their dystonic hands; improved hand function, not
breathing ability, was the intended objective.
On the last treatment day, the subjects were given the splint
constructed for them and asked to practice the exercises per-
formed during therapy while wearing the splint for 1 hour daily
for 1 year after therapy. They were also instructed to engage in
unsplinted repertoire practice for 10% of the time that was their
usual custom before the onset of dystonia. It was recommended
that this period be increased a further 10% in each succeeding
month if they had no deterioration of the motor control that had
been attained. Follow-up was performed for 25 months in the
patient treated ﬁrst and 3 months in the most recently treated
patient (table 1). The intention is to gradually fade out the
therapeutic exercises in the second posttreatment year accord-
ing to a schedule worked out individually with each patient.
Patients’compliance with the home exercises was assessed by
regular phone interviews.
Patient status was quantiﬁed with 2 measurement instru-
ments: a dexterity and displacement device and a dystonia
evaluation scale (DES). The dexterity and displacement device
continuously recorded digital displacement during metronome-
paced movements of 2 ﬁngers performed for 50 seconds;
of the record provided information concern-
ing the smoothness of the movements before, during, and at the
end of the 8 days of treatment. Measurements were done at the
beginning of the treatment session on days 1, 4, and 8. To
quantify the subject’s performance, we divided the spectral
power in the frequency of the metronome (0.9 –1.2Hz) by the
power in the side bands (0.1–0.9Hz plus 1.2–1.9Hz). The side
bands deﬁned in this manner contained the record of move-
ments that were irregular and did not conform to the instruction
to follow the metronome beat (values are missing for patients
1 and 11 because of procedural problems).
A DES was used to rate how well patients performed (with-
out the splint) movement sequences and passages from their
repertoire that had tended to generate dystonic movements in
the past; this, in effect, provided information on the clinical
status of the patients. The average of the ratings for the selected
movement sequences and repertoire passages represented the
DES score for a single day. The ratings on the scale are as
follows: 0, dystonia as bad as at its worst; 1, slightly improved;
2, moderately improved; 3, almost normal; and 4, normal.
The number of movement sequences and repertoire passages
that were rated varied between patients (median, 20). The ﬁrst
rating occurred before treatment and was then repeated after
the ﬁrst 1-hour block of exercises on the ﬁrst treatment day.
This protocol enabled the patient to recognize how much
improvement in the relief of dystonic movements had been
achieved during that brief period. In subsequent sessions, the
DES was administered at the beginning of each treatment day
after a warm-up of approximately 3 minutes and by telephone
in follow-up. Follow-ups were obtained after 1 month, 3 to 4
months, about 6 months, and the time of this writing (month of
last follow-up, see table 1).
The DES scores are given in ﬁgure 2 for the following time
points: pretreatment, posttreatment, and last follow-up. Each of
the 6 pianists (closed inverted triangles) and 2 guitarists (closed
upright triangles) improved very substantially from pretreat-
ment to posttreatment while the 2 ﬂutists (open circles) did not.
The other wind player (oboist, open square) showed a modest
improvement from pre- to posttreatment but 1 month into
follow-up had regressed and remained close to pretreatment
levels at the 2-month follow-up point. A 1-way analysis of
variance (ANOVA) that included all 11 musicians revealed
1344 BEHAVIORAL TREATMENT FOR FOCAL HAND DYSTONIA, Candia
Arch Phys Med Rehabil Vol 83, October 2002
that, even when the 3 wind players are included, the scores at
pretreatment, posttreatment, and last follow-up differed signif-
icantly from each another (F
⫽16.5, P⬍.0001). Post hoc
Scheffe´tests indicated that pretreatment scores differed signif-
icantly from posttreatment (P⬍.0001) and last follow-up
(P⬍.0001) scores. At the time of this writing, follow-up had
proceeded for different lengths of time for different subjects,
depending on when their treatment had taken place (table 1,
column 8). One guitarist (patient 3) and a pianist (patient 1)
showed marked improvement at the end of treatment and then
continued to improve into the normal range at 25 and 19
months, respectively, after treatment. Patient 11 (piano) scored
very close to normal at the end of treatment. These 3 patients
and patients 5 and 7 (piano) performed the posttreatment home-
practice exercises at frequent intervals. Patient 2 (piano) per-
formed home practice approximately half the time for the ﬁrst
5 months posttreatment. At 4 months posttreatment, this indi-
vidual, although only partially compliant, had retained the
substantial therapeutic gains he had obtained during treatment.
After 5 months, he stopped home practice entirely, and by 9
months he had relapsed to pretreatment levels. Patient 6 (piano)
was compliant with the posttreatment home practice require-
ment for 3 months but then the demands of professional activ-
ities reduced his compliance substantially for the next 3
months. As ﬁgure 2 indicates, there has been as yet no regres-
sion in this patient’s performance. Patient 4 (guitar) and patient
5 have stabilized in follow-up with a large improvement com-
pared with very low pretreatment levels. Patient 7 exhibited a
good response to treatment, and has improved somewhat in the
5 months during which he has been followed to date. In the
opinion of these patients, treatment produced signiﬁcant and
long-term improvement of digital coordination, and these re-
sults were clearly superior to the effects attained by previous
therapy with anticholinergic drugs or with local injections of
botulinum toxin type A.
The data for the wind players contrasted to those of the
pianists and guitarists. Pretreatment, their DES scores did not
differ signiﬁcantly from the other types of instrumentalists.
However, a 2-way ANOVA with group (wind players vs pia-
nists and guitarists) and time as factors indicated that a signif-
icant difference occurred between the 2 groups over time (pre-
to posttreatment to last follow-up; interaction group ⫻time,
The results on the DES and the clinical evaluation were
conﬁrmed by the laboratory ﬁndings obtained from the dexter-
ity and displacement device (ﬁg 3). During paced movements,
the smoothness of the movements of the nondystonic hand did
not signiﬁcantly change from pre- to posttreatment, as revealed
by spectral power analysis. These analyses served as control
measurements for the dystonic hand. The ﬁnger movements of
the pianists’and guitarists’dystonic hand were clearly
smoother after treatment (t⫽2.9, P⬍.05). In contrast, no im-
provement occurred from pre- to posttreatment for the wind
players; the difference in treatment effect between the 2 groups
was signiﬁcant for the dystonic hand (F
⫽10.0, P⬍.05 for the
interaction of group ⫻time). Figure 3 is a scatterplot showing
the relationship between change in DES score from pre- to
posttreatment and change in the relative fast Fourier transform
power on the dexterity and displacement device scores from
pre- to posttreatment for 9 subjects. The correlation is requal
to .70 (P⬍.05) showing that the measures exhibit considerable
concordance in showing an improvement or lack of improve-
ment in the dexterity of musical performance after treatment.
The 2-ﬁngered dexterity and displacement device exercise
did not always elicit the maximum amount of dystonic move-
ment. For patient 6 (piano), a 3-ﬁngered pattern yielded a
particularly dramatic demonstration of treatment effect. This
individual had a pronounced dystonic condition of 34 years in
duration. The required alternating movements in his 3-ﬁngered
test exercises always involved the focal dystonic ﬁnger (D4),
the maximal compensatory ﬁnger (D5), and the least-involved
digit of the dystonic hand (D2) in 3 different orders and the
homologous ﬁngers of the nondystonic hand in the same 3
orders. Each sequence continued for 75 seconds with a
3-minute rest between sequences. Figure 4 shows the dexterity
and displacement device record for the last 10 seconds of the
ﬁrst sequence (D2, D4, D5) for the nondystonic and dystonic
hands on the fourth day of treatment (panels 3, 4) and the day
after the end of treatment (panels 1, 2). The tracings represent
Fig 2. Mean DES ratings of all 11 patients for the pre- and posttreat-
ment and last follow-up periods. Rating scale: 0, dystonia as bad as
at its worst; 1, slightly improved; 2, moderately improved; 3, almost
normal; and 4, normal.
Fig 3. Scatterplot of pre- to posttreatment change for the DES and
for the fast Fourier transform (FFT) power values for spectral anal-
ysis of dexterity and displacement device performance (which gives
the smoothness of movement) for the focal dystonic ﬁngers.
1345BEHAVIORAL TREATMENT FOR FOCAL HAND DYSTONIA, Candia
Arch Phys Med Rehabil Vol 83, October 2002
the excursion and duration of the key depression for the focal
dystonic ﬁnger (D4, dark lines) and the maximal compensatory
ﬁnger (D5, light lines) and for the same ﬁngers on the nondys-
tonic hand. On day 4 of treatment, when this subject had
already improved considerably, the movements of the digits of
the dystonic hand were still markedly abnormal. The very
substantial improvement that had taken place after 8 days of
treatment may be seen by comparing panel 1 with panel 3. The
ﬁgure by itself, however, does not show the most noteworthy
change that occurred. On day 4 of treatment, by the time 45 to
50 seconds had elapsed during each of the 3 exercises with the
dystonic hand, hyperﬂexion of D4 and D5 had developed to
such an extent that this subject had to complete the exercises by
playing with his knuckles. It is remarkable and a tribute to his
skill that he could maintain the metronome-speciﬁed tempo as
well as he did. Posttreatment, his movements were not normal
(as indicated by comparison to the data from the nondystonic
hand—panels 2, 4), but they were greatly improved. Playing on
the knuckles did not occur at all. From time to time, a tendency
emerged for D4 or D5 or both to hyperﬂex, but in each case the
subject was able to bring the 2 digits into a more appropriate
extended position and complete each exercise without further
It should be noted that patients 3 and 6 reported that per-
forming home practice with the splint for 1 hour every day over
an extended duration was fatiguing. They found that practicing
the therapeutic exercise every other day appeared to give them
a better result for repertoire performance. Four pianists and
guitarists have passed the 1-year follow-up (patients 1, 5, 3, 4).
Each of these patients stopped carrying out the therapeutic
exercise at 21, 12, and (contrary to instructions) 4, and 2
months, respectively. None of them has experienced a decre-
ment from the maximum treatment gains that they exhibited.
Figure 5A shows a characteristic dystonic pattern before
treatment with right D4 and D5 in ﬂexion spasm as the pianist
attempts to depress the keys with the ﬁngers of the dystonic
hand. In ﬁgure 5B, the same pianist playing the same piece
after treatment is able to depress the piano keys with D4 and
D5 in a more correct extended posture.
The basic principles of SMR therapy are as follows.
1. The focal dystonic ﬁnger was not splinted.
2. The main compensatory ﬁnger was splinted ﬁrst in the
sequence of daily exercise blocks, at an angle similar to
the patient’s normal resting angle of that ﬁnger on the
instrument. (In other patients, there may be more than 1
main compensatory ﬁnger.)
3. Subsequently, exercises were performed with the other
ﬁngers on the dystonic hand splinted.
4. The speed at which the dystonic ﬁnger was required to
move in concert with the other ﬁngers of the hand was
increased and then decreased with progressively more
exacting requirements (eg, to extend the ﬁngers as much
as possible when playing).
5. It was valuable to have daily repertoire practice. This
practice helped to sustain patient motivation and to ac-
complish transfer of the exercise-induced improvement in
motor control to the target behavior.
6. Repertoire practice of the dystonic hand was intensive but
not to the point when excessive fatigue was induced.
7. Home practice of the treatment exercises after conclusion
of therapist-administered treatment is important to main-
tain therapeutic gains and continued improvement.
In 5 cases, the performers’musical careers were not inter-
rupted by the dystonia because they resorted to well-known
Fig 4. The extent and duration of the movements of the focal
dystonic (D4, dark line) and maximal compensatory ﬁnger (D5, light
line) of the dystonic hand and the corresponding ﬁngers of the
nondystonic hand in patient 6 recorded during and after treatment.
Each record is for the last 10 seconds of the ﬁrst exercise, which was
the same at the 2 time points and for both hands. Note the improve-
ment in the dystonic hand (panels 1, 3).
Fig 5. The hands of a pianist playing the same passage (A) before
and (B) after treatment. The lines outline the ﬂexor spasm of right
D3, D4, and D5 before treatment and the correct positioning after
1346 BEHAVIORAL TREATMENT FOR FOCAL HAND DYSTONIA, Candia
Arch Phys Med Rehabil Vol 83, October 2002
tricks (eg, see Lederman
) that allowed them to perform some
(but not other) concert pieces publicly, though accomplished
with great difﬁculty. The tricks are substitute maneuvers such
as using different and atypical ﬁngerings to avoid use of the
focal dystonic digit, and adopting special postures to allay
incoordination-inducing maneuvers. Although their concert
performance had never been completely halted, SMR initiated
a dramatic improvement in instrument-related manual dexterity
for these 5 musicians. All judged their performance after treat-
ment to be superior and much easier than before therapy. The
tricks were either no longer necessary or the need for them was
greatly reduced. For 4 of the 8 pianists and guitarists (patients
1, 3, 7, 11) who continued performance, there was a return
close to the level of performance carried out by the dystonic
hand before the onset of pathology. Because follow-up pro-
ceeded from 3 to 25 months and is 12 months or more in 4
cases, the treatment effect appears to be long-term.
The 3 wind players can be viewed as treatment failures,
given the original study objectives. However, because they
were treatment failures, they can be viewed as a comparison
group (although entirely unplanned and unintentional on our
part) for the following reasons: (1) the wind players were
musicians with focal hand dystonia, just as were the pianists
and guitarists; (2) they were as strongly motivated to improve
as the pianists and guitarists, (3) we ﬁrmly that believed they
would improve; and (4) they received the same basic treatment
procedure as the pianists and guitarists, who did improve.
Thus, all the nonspeciﬁc factors associated with treatment of
the wind players were the same as for the pianists and guitar-
ists. Consequently, if the good treatment effects with pianists
and guitarists were a placebo effect, then the wind players
should have improved as well. That they did not indicates that
the positive outcome obtained with the experimental subjects
would appear to be because of an interaction between the
treatment and the type of instrument played rather than to a
The question arises as to why SMR treatment was successful
for pianists and guitarists and not for wind players. The ﬁrst
possibility is that the need for ﬁnger-mouth coordination in
playing wind instruments creates an effect on brain mecha-
nisms that is not adequately addressed by the present therapy,
which focuses only on movements of the digits and not the
mouth. For example, it is possible that, in focal dystonic wind
players, not only do the cortical representational zones of the
ﬁngers overlap and fuse, as occurs in pianists and guitarists,
but also the cortical representational zones of the digits and
mouth move closer together and meld. This possibility could be
experimentally evaluated with neuroimaging techniques. In
addition, it would be of interest to require wind players to make
movements of the ﬁngers while simultaneously blowing into
their instruments; the latter element was not part of the thera-
peutic regimen reported here. The exercises would be designed
to facilitate an increased differentiation between digit and
mouth cortical representations.
A second explanation for the lack of results with the wind
players might be that the therapy is not effective for musical
performance that involves exerting a fairly constant and ﬁrm
force during performance with D1 to hold the instrument and
maintain its correct orientation. The combination of the re-
quired static load involved in holding the instrument and
skilled movement in the same hand might hinder successful
learning of digital performance in patients with focal dystonia.
(Although guitarists hold the neck of their instruments with the
thumb of their left hand, both guitarists treated in this experi-
ment exhibited focal dystonic patterns in their right hand,
which is used to pluck the instrument’s strings.) If this expla-
nation is correct, a therapy designed to reduce the pressure
exerted by D1 during performance might be effective. A third
possibility is that the posture of the digits and the nature of the
ﬁngering involved in playing wind instruments is not amenable
to a reduction of focal dystonic patterns in the same way as is
the posture of the digits and the nature of the ﬁngering during
performance on the piano or guitar.
The present results from 11 subjects must be considered
preliminary. However, they should be interpreted in the light of
2 considerations. First, each patient had a chronic condition
that had persisted for years, in 1 case 34 years. Second, the
condition had become intractable for each of the patients and
was not responsive to a variety of commonly used medical
treatments; they therefore, in effect, constitute placebo treat-
ment in these patients. Thus, each individual case assumes
increased therapeutic interest.
This work exempliﬁes how changing the way body parts are
used can elucidate and possibly remediate pathologic condi-
tions resulting from overuse. This is especially the case in light
of recent neuroimaging results
indicating that focal hand
dystonia in musicians involves an alteration in brain organiza-
tion; this may also be the case in some other overuse syn-
Acknowledgment: We thank Dr. Andreas Keil for comments and
help in the data analysis.
1. Sheehy MP, Marsden CD. Writers’cramp—a focal dystonia.
Brain 1982;105(Pt 3):461-80.
2. Rosenbaum F, Jankovic J. Focal task-speciﬁc tremor and dystonia:
categorization of occupational movement disorders. Neurology
3. Bejjani F, Kaye G, Benham M. Musculoskeletal and neuromus-
cular conditions of instrumental musicians. Arch Phys Med Re-
4. Jankovic J, Shale H. Dystonia in musicians. Neurology 1989;9:
5. Fry H. The treatment of overuse injury syndrome. Md Med J
6. Newmark J, Hochberg FH. Isolated painless manual incoordina-
tion in 57 musicians. J Neurol Neurosurg Psychiatry 1987;50:
7. Lederman R. Occupational cramp in instrumental musicians. Med
Probl Perform Art 1988;3:45-51.
8. LaVine W. Behavioral and biofeedback therapy for a functionally
impaired musician: a case report. Biofeedback Self Regul 1983;
9. Fahn S, Marsden C. The treatment of dystonia. 2nd ed. London:
10. Bejjani F. Performing artists’occupational disorders. In: DeLisa J,
editor. Rehabilitation medicine: principles and practice. 2nd ed.
Philadelphia: Lippincott Williams & Wilkins; 1993. p 1165-90.
11. Cohen LG, Hallett M, Geller BD, Hochberg F. Treatment of focal
dystonias of the hand with botulinum toxin injections. J Neurol
Neurosurg Psychiatry 1989;52:355-63.
12. Jankovic J, Brin M. Therapeutic uses of botulinum toxin. N Engl
J Med 1991;324:1186-94.
13. Cole R, Cohen L, Hallett M. Treatment of musician’s cramp with
botulinum toxin. Med Probl Perform Art 1991;6:137-43.
14. Altenmu¨ller E. Fokale Dystonien bei Musikern: eine Herausfor-
derung fu¨r die Musiker-Medizin. Musikphysiol Musikermed
15. Hallett M. The neurophysiology of dystonia. Arch Neurol 1998;
16. Chen R, Hallett M. Focal dystonia and repetitive motion disorders.
Clin Orthop 1998;Jun(351):102-6.
1347BEHAVIORAL TREATMENT FOR FOCAL HAND DYSTONIA, Candia
Arch Phys Med Rehabil Vol 83, October 2002
17. Byl NN, Merzenich MM, Jenkins WM. 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 1996;47:508-
18. Byl NN, Merzenich MM, Cheung S, Bedenbaugh P, Nagarajan
SS, Jenkins WM. A primate model for studying focal dystonia and
repetitive strain injury: effects on the primary somatosensory
cortex. Phys Ther 1997;77:269-84.
19. Elbert T, Pantev C, Wienbruch C, Rockstroh B, Taub E. Increased
use of the left hand in string players associated with increased
cortical representation of the ﬁngers. Science 1995;270:305-7.
20. Elbert T, Candia V, Altenmu¨ller E, et al. Alteration of digital
representations in somatosensory cortex in focal hand dystonia.
21. Bara-Jimenez W, Catalan MJ, Hallett M, Gerloff C. Abnormal
somatosensory homunculus in dystonia of the hand. Ann Neurol
22. Candia V, Elbert T, Altenmuller E, Rau H, Schafer T, Taub E.
Constraint-induced movement therapy for focal hand dystonia in
musicians [letter]. Lancet 1999;353(9146):42.
23. Hesselmann N. The fundamentals of discrete Fourier analysis. In:
Huston P, Weitkunat R, editors. Digital biosignal processing. New
York: Elsevier; 1991. p 59-80.
1348 BEHAVIORAL TREATMENT FOR FOCAL HAND DYSTONIA, Candia
Arch Phys Med Rehabil Vol 83, October 2002