edicine and Rehabilitation

Archives of Physical M journal homepage: www.archives-pmr.org

Archives of Physical Medicine and Rehabilitation 2013;94:1-8

FEATURED ARTICLE

Effects on Decreasing Upper-Limb Poststroke Muscle ToneUsing Transcranial Direct Current Stimulation: A RandomizedSham-Controlled Study

Dongyu Wu, MD,a Long Qian, PT,a Richard D. Zorowitz, MD,b Lei Zhang, MD,c

Yaping Qu, PT,a Ying Yuan, MSca

From the aDepartment of Rehabilitation, Xuanwu Hospital of Capital Medical University, Beijing, China; bDepartment of Physical Medicineand Rehabilitation, The Johns Hopkins University School of Medicine, Baltimore, MD; and cVisiting scholar at Medical College of Wisconsin,Milwaukee, WI.

Abstract

Objective: To assess the efficacy of transcranial direct current stimulation (tDCS) on decreasing upper-limb (UL) muscle tone after stroke.

Design: A prospective, sham-controlled, randomized controlled trial with 4-weeks follow-up. Randomization into the tDCS group or the control

group.

Setting: Rehabilitation education and research hospital.

Participants: Inpatients (NZ90, 45 per group; age range, 15e70y; 69 men, 21 women; duration of stroke, 2e12mo) with poststroke UL

spasticity. No participant withdrew because of adverse effects.

Intervention: The tDCS group received tDCS to the primary sensorimotor cortex of the affected side with cathodal stimulation, 20 minutes per

day, 5 days per week, for 4 weeks and conventional physical therapy. The control group received sham stimulation (same area as the tDCS group)

and conventional physical therapy.

Main Outcome Measures: Modified Ashworth scale (MAS), Fugl-Meyer Assessment of motor recovery, and Barthel Index. All outcomes were

measured at admission, after treatment, and after follow-up. A clinically important difference (CID) was defined as a reduction of�1 in theMAS score.

Results: Compared with the sham tDCS group, the active tDCS group had significantly more patients with a clinically important difference after

treatment (80% and 78% vs 6% and 9%) and at 4-week follow-up (84% and 82% vs 7% and 4%), and UL motor function and activities of daily

living (ADL) assessment improved more significantly in the active tDCS group (Fugl-Meyer Assessment of motor recovery from 12 [range,

4e26] to 22 [range, 7e50] to 32 [range, 28e41], Barthel Index from 55 [range, 0e85] to 85 [range, 5e100] to 90 [range, 10e100 vs Fugl-Meyer

Assessment of motor recovery from 8 [range, 3e34] to 10 [range, 8e25] to 15 [range, 6e40], Barthel Index from 55 [range, 25e95] to 65 [range,

30e100] to 75 [range, 40e100], respectively, P<.01).

Conclusions: UL muscle tone after stroke can be decreased using cathodal tDCS. Combined with conventional physical therapy, tDCS appears to

improve motor function and ADL. Cathodal tDCS over ipsilesional primary sensorimotor cortex may inhibit primary sensorimotor cortex

hyperactivation, resulting in significant reductions in muscle tone.

Archives of Physical Medicine and Rehabilitation 2013;94:1-8

ª 2013 by the American Congress of Rehabilitation Medicine

Spasticity, a “velocity-dependent increase in muscle tone withexaggerated tendon jerks that result from hyperexcitability of the

Supported by the National Natural Science Foundation of China (grant nos. 30600186 and

81171011).

No commercial party having a direct financial interest in the results of the research supporting

this article has or will confer a benefit on the authors or on any organization with which the authors

are associated.

Clinical Trial Registry. http://www.chictr.org. Registration no.: ChiCTR-TRC-11001367.

0003-9993/13/$36 - see front matter ª 2013 by the American Congress of Re

http://dx.doi.org/10.1016/j.apmr.2012.07.022

stretch reflex,”1(pS1-13) is a condition that results from a number ofneurologic disorders, such as stroke, traumatic brain injury, andmultiple sclerosis. Increased muscle tone may result in muscleimbalance and shortening, leading to abnormal postures. It mayinterfere with nursing care, rehabilitation potential, functionalactivities, and functional recovery.2 It is estimated that nearly halfof stroke survivors develop moderate or severe spasticity of theaffected upper limb (UL).3,4

habilitation Medicine

2 D. Wu et al

Management of spasticity generally is considered essential toprevent deformities, to improve function, and to relieve distressingsymptoms. It may require multiple interventions, including posi-tioning; reducing noxious stimulation; stretching; physicalmodalities; electrical stimulation; oral and transdermal medica-tions; nerve blocks and chemical neurolysis; and surgical inter-ventions, such as intrathecal pumps, neurosurgical procedures, andorthopedic procedures.5 Clinically, such interventions either haveunsatisfactory results (eg, physical modalities, electrical stimula-tion) or may require invasive management (eg, nerve blocks,chemical neurolysis, or surgical interventions). Oral and trans-dermal medications may not always be effective and may bedifficult to tolerate because of adverse effects.

Recently, some research in the treatment of brain-injuredpatients has focused on a noninvasive intervention known astranscranial direct current stimulation (tDCS). During tDCS, weakpolarizing direct currents (1e2mA) are delivered to the cortex via2 electrodes placed on the scalp. Cathodal polarization reducescortex excitability, whereas anodal polarization increases it. Thesechanges evolve during tDCS, and remain for up to 1 hour after itceases.6,7 When measured by blood oxygen leveledependentmagnetic resonance imaging, cathodal stimulation over thesensorimotor cortex has been found to significantly decreaseactivation, while anodal stimulation resulted in a nonsignificantincrease in activation.8 Compared with tDCS, which requiresseveral minutes for administration, transcranial magnetic stimu-lation can be applied more focally and with millisecond accuracy,but is more expensive. However, the larger stimulation electrodesize (thus poor spatial resolution) needed for tDCS might be anadvantage over transcranial magnetic stimulation because it canactivate larger areas of the cortex and potentially have a greaterpositive effect on neuroplasticity. In addition, sham tDCS is easierto achieve than sham transcranial magnetic stimulation in double-blinded experimental designs.9

A study of event-related functional magnetic resonanceimaging (fMRI) during slow and fast passive movements of thehand with simultaneous recording of passive movement resistancedemonstrated that stroke survivors activated ipsilesional S1 (area3b) and M1 (area 4a) significantly more than did controls.10

Another study suggested that ipsilesional primary sensorimotorcortex (S1M1) hyperactivation occurred 4 to 12 monthsafter stroke.11

Therefore, the following questions arise: Can cathodal tDCSdecrease the hyperactivation of ipsilesional S1M1 and decreaseUL muscle tone after stroke? Does cathodal tDCS on ipsilesionalS1M1 improve or worsen UL motor performance after stroke?Thus, a prospective, sham-controlled, randomized controlled trialwas designed to answer these questions.

The purpose of this study was to assess the efficacy of cathodaltDCS on ipsilesional S1M1 when coupled with conventionalphysical therapy in patients with UL spasticity after stroke. Our

List of abbreviations:

ADL activities of daily living

CI confidence interval

CID clinically important difference

fMRI functional magnetic resonance imaging

MAS modified Ashworth scale

S1M1 primary sensorimotor cortex

tDCS transcranial direct current stimulation

UL upper limb

hypothesis was that cathodal tDCS combined with conventionalphysical therapy decreases muscle tone significantly more thanconventional physical therapy alone. Therefore, assessment of ULmuscle tone was the main endpoint, and assessments of motorfunction and activities of daily living (ADL) were the secondaryendpoints in this study.

Methods

Participants

The study was performed in the Department of Rehabilitation,Xuanwu Hospital of Capital Medical University, Beijing, China.Consecutive patients admitted to the inpatient rehabilitationfacility for stroke rehabilitation were considered for recruitmentfrom June 2011 to February 2012. Stroke was defined as “anacute event of cerebrovascular origin causing focal or globalneurologic dysfunction lasting more than 24 hours,”12(p548)

diagnosed by a neurologist, and confirmed by computedtomography and magnetic resonance imaging. The HospitalEthics committee approved the study. Informed written consentwas obtained from either the subjects or the guardian or parentsof the subjects.

Inclusion criteria consisted of the following: (1) occurrence ofa stroke at least 2 months before admission, (2) first-ever stroke,(3) muscle tone at the wrist and elbow with a modified Ashworthscale (MAS) score between 1 and 3 (inclusive), (4) no history ofbotulinum toxin injection or other invasive treatment for past 6months, (5) use of antispasticity drugs resulting in an adverseevent or maximized dosing without effect, and (6) no severecognitive or mood disorders that would interfere with the study’spurpose. Subjects were excluded for (1) unstable vital signs or (2)unstable, progressive, or severe neurologic disease, heart condi-tions, or hypertension.

Design and procedures

We used a double-blind, sham-controlled, randomized controlleddesign, in which both subjects and assessors were blinded(fig 1).

Sample size was determined on the basis of results of ourpreexperiment: alpha (2-sided, .05), beta (0.9), MAS difference ofthe tDCS group (0.7), and SD (1), resulting in 45 patients per group.

Interventions

Direct current was transferred by a saline-soaked pair of surfacesponge electrodes (4.5cm�5.5cm) and delivered by a speciallydeveloped, battery-driven, constant current stimulator (IS200).a

A constant current with an intensity of 1.2 milliampere wasapplied for 20 minutes once daily, 5 days per week, for 4 weeks.According to Boggio et al’s method,13 the cathodal plate wasplaced over C3 or C4 (international 10/20 electroencephalographysystem). The anodal plate was placed over the unaffected shoulderto avoid any adverse effect. Because some subjects sensed thecurrent as an itching sensation at both electrodes at the beginningof the stimulation, the electrodes were placed in the same positionon subjects receiving the sham stimulation; however, the stimu-lator was turned off after 30 seconds.13,14 For both active andsham tDCS groups, current intensity was gradually increased and

www.archives-pmr.org

Fig 1 Flowchart of the randomized clinical trial of tDCS treatment of poststroke UL muscle tone.

Decreasing muscle tone using noninvasive brain stimulation 3

decreased. This procedure blinded subjects to the respectivestimulation conditions.

Both groups received a conventional physical therapy programfor 30 minutes twice daily, including maintaining good limbposition, chronic stretching via casting or splinting, physicalmodalities and techniques, and movement training.

Blinding

Subjects were randomly assigned using a computer-generatedrandomization list by a single investigator. The assigned randomnumber was inputted into the stimulator device by the sameinvestigator. She did not participate in other parts of the study. Thedevice automatically generated active or sham tDCS according tothe parity of the random number. All other investigators, subjects,and outcome assessors remained blinded to group allocation untilthe completion of the final statistical analyses.

www.archives-pmr.org

Outcome assessment

All outcomes were assessed at baseline, at the conclusion of the4-week treatment protocol, and at 4-week follow-up. TheMASwasused to assess the affected wrist and elbow while subjects wereseated.15 This well-validated instrument characterizes muscle toneon a scale ranging from 0 (no increase in tone) to 4 (limb rigid inflexion or extension).16 A clinically important difference (CID) wasdefined as a reduction of�1 in the MAS score. Motor function wasassessed using the UL motor component of the Fugl-MeyerAssessment (scale range, 0e66).17 ADL were assessed using themodified Barthel Index (scale range, 0e100).18,19

Data analysis

SPSS (version 17)b was used to perform all analyses. Independent ttests and chi-square test were used to compare baseline

Table 1 Baseline characteristics of the 2 groups

Characteristics

Active tDCS

(nZ45)

Sham tDCS

(nZ45) Statistics

Age (y) 45.9�11.2 49.3�12.6 NS, t test

Range (minemax) 15e66 18e70

Sex NS, c2 test

Male 34 35

Female 11 10

Time since stroke (mo) 4.9�3.0 4.9�2.9 NS, t test

Range (minemax) 2e12 2e12

Paretic side NS, c2 test

Left 21 22

Right 24 23

Diagnosis NS, c2 test

Infarction 27 26

Hemorrhage 18 19

NOTE. Values are mean � SD or number of patients.

Abbreviation: NS, not significant.

4 D. Wu et al

characteristics of the 2 groups. Mann-Whitney U test was used tocompare the outcome assessments at admission, after tDCS, and atfollow-up between the 2 groups. Wilcoxon signed-rank test wasused to model the time trend for the 2 groups. For the mainendpoint, chi-square test was used to compare the number ofpatients with CID between the 2 groups. Moreover, the patientswere measured 3 times consecutively so that the outcomes werecorrelated over time. Therefore, repeated-measures analysis ofvariance was applied to model the within-subject correlations aswell as other clinical variables. A P value of .05 was used forstatistical significance.

Results

Ninety inpatients (69 men and 21 women; mean age, 47.6y; range,15e70y) with poststroke UL spasticity were recruited in thisstudy. Sample characteristics are summarized in table 1. The meanduration poststroke was 4.9 months (range, 2e12mo). There wereno significant differences between the 2 groups for age, sex, timesince stroke, paretic side, and diagnosis. No subject withdrewfrom the trial. Most of the subjects had no sensation while per-forming tDCS. Several subjects felt a slight itching or tingling inthe scalp under the tDCS electrode.

Masking of the treatment assignment during randomizationwas effective. All subjects in both groups believed that theyreceived active tDCS treatment.

The outcome assessments at admission, after tDCS, and atfollow-up between the 2 groups are listed in table 2. There was

Table 2 Outcome assessments at admission, after tDCS, and at follow

Assessment

At Admission Aft

Active tDCS Sham tDCS Z P Active tDCS Sham

MAS: elbow 2 (1e4) 2 (1e3) �1.60 0.110 1 (0e3) 2 (1

MAS: wrist 2 (1e4) 2 (1e3) �0.47 0.638 1 (0e3) 2 (1

FMA 12 (4e26) 8 (3e34) �1.77 0.077 22 (7e50) 10 (8

BI 55 (0e85) 55 (25e95) �0.83 0.404 85 (5e100) 65 (3

NOTE. Values are median (range). MAS scores are changed from 1, 1þ, 2, 3 t

Abbreviations: BI, Barthel Index; FMA, Fugl-Meyer Assessment.

no significant difference between the 2 groups for all assess-ments at admission. Compared with the sham tDCS group, ULmuscle tone was significantly decreased and UL motor functionand ADL assessment significantly improved in the active tDCSgroup after tDCS and at follow-up. Table 3 lists the result ofWilcoxon signed-rank test to model the time trend for the 2groups. For the active tDCS group, MAS scores of elbow andwrist significantly decreased after tDCS and kept decreasing (notsignificantly) at follow-up. For the sham tDCS group, MASscores almost kept unchanged after tDCS and increased signif-icantly at follow-up. UL motor function and ADL assessmentsignificantly improved in both groups after tDCS and atfollow-up.

Tables 4 and 5 list the number of patients with MAS scoredifferences after treatment and at follow-up. Compared withthe sham tDCS group, the active tDCS group had signifi-cantly more patients with CID after treatment (80% and 78%vs 6% and 9%, elbow and wrist, respectively) and at follow-up (84% and 82% vs 7% and 4%, elbow and wrist,respectively).

The results of repeated-measures analysis of variance tomodel the time trend and comparison between the 2 groups arelisted in table 6. After adjustment for age, duration, sex, loca-tion, and diagnosis, a pretty clear trend could be identified:compared with the sham tDCS group, no significant differenceat admission, presenting significant difference after treatment,and getting more and more significant after follow-up for clin-ical assessments of MAS, Fugl-Meyer Assessment, and BarthelIndex in the active tDCS group (mean difference betweenthe 2 groups, Fugl-Meyer Assessment: 6.71 [95% confidenceinterval (CI), 3.26e10.17], 14.71 [95% CI, 10.98e18.44],P<.01; Barthel Index: 8.16 [95% CI, .24e16.07], 9.86 [95% CI,2.08e17.64], P<.05) (fig 2).

Discussion

This study is the first randomized, sham-controlled trial of whichwe are aware to evaluate the benefits of tDCS for UL muscle tone,motor function, and ADL in stroke survivors with UL dysfunctionand associated activity limitations. The results confirmed thehypothesis that cathodal tDCS over ipsilesional S1M1 coupledwith conventional physical therapy significantly decreased ULmuscle tone and improved motor function and ADL whencompared with the sham control.

In recent years, we have observed strokes in younger patients.In this population, the youngest subjects were 15 and 18 years old.As a result, we did not exclude these young subjects on the basisof age in order that they may be studied.

-up for the 2 groups

er tDCS At Follow-Up

tDCS Z P Active tDCS Sham tDCS Z P

e3) �5.4 <0.001 1 (0e3) 3 (1e3) �7.2 <0.001

e3) �5.0 <0.001 1 (0e3) 3 (2e4) �7.2 <0.001

e25) �4.1 <0.001 32 (28e41) 15 (6e40) �6.2 <0.001

0e100) �2.7 0.006 90 (10e100) 75 (40e100) �3.2 0.001

o 1, 2, 3, 4 for the convenience of statistics.

www.archives-pmr.org

Table 3 Wilcoxon signed-rank test to model the time trend for the 2 groups

Group Statistics

MAS: Elbow MAS: Wrist FMA BI

t2et1 t3et2 t2et1 t3et2 t2et1 t3et2 t2et1 t3et2

Active Z �5.6* �1.4* �5.7* �1.9* �5.8y �5.8y �5.8y �5.4y

tDCS P <0.001 0.157 <0.001 0.059 <0.001 <0.001 <0.001 <0.001

Sham Z �0.6* �5.0y �0.6* �4.6y �5.0y �5.5y �5.5y �5.4y

tDCS P 0.527 <0.001 0.564 <0.001 <0.001 <0.001 <0.001 <0.001

NOTE. t1 (at admission), t2 (after tDCS), t3 (at follow-up).

Abbreviations: BI, Barthel Index; FMA, Fugl-Meyer Assessment.

* Based on positive ranks.y Based on negative ranks.

Decreasing muscle tone using noninvasive brain stimulation 5

Possible mechanisms of tDCS treatment

From a pathophysiological view of muscle tone, central nervoussystem lesions affect tracts in both the pyramidal and extrapyra-midal systems. Increased muscle tone arises as a result of neu-roplastic changes (eg, collateral sprouting) and/or release effects(disinhibition) as a result of the lesion.20 In a 1-year longitudinalfMRI study, Tombari et al11 demonstrated an evolution in S1M1activation from early (20d after stroke) contralesional hyper-activation to a later (4mo after stroke) ipsilesional hyperactivationconcomitant to recovery. Another electromyogram-fMRI study of10 chronic stroke survivors with UL dysfunction demonstratedthat fMRI could detect wide bilateral activation in S1M1, thesupplementary motor area, and cerebellum while subjects movedthe paretic hand.21 Therefore, ipsilesional S1M1 hyperactivationplays an important role in spasticity caused by upper motor neuronsyndromes such as stroke.

In this study, cathodal tDCSwasused to inhibit thehyperactivationof ipsilesional S1M1. The results demonstrate that compared withsham tDCS, active tDCS significantly decreased elbow and wristMAS scores. Therefore, it could be inferred that cathodal tDCS overipsilesional S1M1 had the effect of inhibition of S1M1 hyper-activation, which resulted in significant reduction in muscle tone.

Effect of tDCS treatment

Because increased muscle tone is a common manifestation insubacute and chronic stroke survivors, the decision to applycathodal tDCS was motivated by a desire to determine whetherinhibition of S1M1 hyperactivation could decrease UL muscletone and improve UL motor performance.

tDCS effect on muscle tone

Compared with the sham tDCS group, elbow and wrist muscletone significantly decreased and there were significantly more

Table 4 MAS (Elbow) assessment after tDCS and at follow-up

Active tDCS

(nZ45)

Sham tDCS

(nZ45)

P�1 Z0 �1 �1 Z0 �1

After tDCS 36 9 0 3 38 4 <0.001

At follow-up 38 7 0 3 15 27 <0.001

NOTE. Values are number of patients (difference in MAS grade at

admission).

www.archives-pmr.org

patients with CID after treatment and 4-week follow-up in theactive tDCS group. This result confirmed our previous hypothesis;that is, cathodal tDCS combined with conventional physicaltherapy decreases muscle tone significantly more than conven-tional physical therapy alone. The effect of tDCS could last fora minimum of 1 month. Because a majority of the subjects hadMAS scores between 1þ and 2, the results suggest that tDCStreatment may be most beneficial in subjects with mild ormoderate muscle tone.

tDCS effect on motor function and ADL

In this study, UL motor function and ADL assessment signifi-cantly improved in both groups after tDCS and at follow-up,which indicated that conventional physical therapy (with orwithout tDCS) had a positive effect on UL motor function andADL. Conventional positioning, stretching, and other physicalmodalities play an important role in inducing neuroplasticity,thereby maintaining and consolidating the treatment effect.

A number of studies have explored the effects of tDCS in themotor cortex. In stroke survivors, both cathodal stimulation of theunaffected hemisphere and anodal stimulation of the affectedhemisphere (but not sham tDCS) improved motor performancesignificantly.22 Hummel and colleagues14,23 demonstrated thatactivation of the affected hemisphere with anodal tDCS led totransient improvements in motor performance in chronic strokesurvivors. However, in this study, significant decreases in elbowand wrist MAS scores resulted in significantly better UL motorfunction and ADL in the active tDCS group. This result answeredthe question of whether cathodal tDCS on ipsilesional S1M1improved UL motor performance after stroke. Therefore, increasedmuscle tone is a major obstacle that hinders further improvementof UL motor function in subacute and chronic stroke survivors.Ipsilesional S1M1 hyperactivation may produce negative effects onmotor recovery, such as increased muscle tone. tDCS has a bidi-rectional regulatory effect on target cortical regions that both

Table 5 MAS (Wrist) assessment after tDCS and at follow-up

Active tDCS

(nZ45)

Sham tDCS

(nZ45)

P�1 Z0 �1 �1 Z0 �1

After tDCS 35 10 0 4 32 9 <0.001

At follow-up 37 8 0 2 14 29 <0.001

NOTE. Values are number of patients (difference in MAS grade at

admission).

Table 6 RMANOVA to model the time trend and comparison between treatment and control groups

Outcomes Covariates

At Admission After tDCS At Follow-Up

D 95% CI P D 95% CI P D 95% CI P

MAS-elbow Treatment 0.3 0.04 to 0.56 0.026 �0.6 �0.81 to �0.39 <0.001 �1.19 �1.42 to �0.95 <0.001

Infarction 0 �0.3 to 0.30 0.993 �0.07 �0.31 to 0.17 0.544 �0.15 �0.42 to 0.12 0.272

Male �0.06 �0.38 to 0.20 0.698 0.13 �0.13 to 0.39 0.325 0.22 �0.07 to 0.51 0.132

Left 0.07 �0.2 to 0.35 0.592 0.09 �0.13 to 0.31 0.425 0.09 �0.15 to 0.34 0.462

Age 0 �0.01 to 0.01 0.993 0 �0.01 to 0.01 0.846 0.01 0 to 0.02 0.27

Duration 0.04 �0.01 to 0.08 0.099 0.05 0.02 to 0.09 0.005 0.04 0 to 0.08 0.072

MAS-wrist Treatment 0.18 �0.13 to 0.48 0.252 �0.65 �0.9 to �0.4 <0.001 �1.38 �1.66 to �1.11 <0.001

Infarction �0.1 �0.44 to 0.25 0.569 0.12 �0.17 to 0.4 0.427 �0.2 �0.51 to 0.12 0.212

Male �0.17 �0.54 to 0.20 0.375 0.13 �0.18 to 0.44 0.4 0.36 0.03 to 0.7 0.035

Left 0 �0.31 to 0.32 0.986 0.14 �0.13 to 0.4 0.306 0.13 �0.16 to 0.41 0.377

Age 0 �0.01 to 0.02 0.654 �0.01 �0.02 to 0.01 0.372 0 �0.01 to 0.01 0.753

Duration 0.06 0.01 to 0.11 0.021 0.07 0.03 to 0.12 0.001 0.06 0.01 to 0.1 0.022

BI Treatment �5.69 �12.99 to 1.61 0.125 8.16 0.24 to 16.07 0.043 9.86 2.08 to 17.64 0.014

Infarction 15.28 6.93 to 23.63 <0.001 17.16 8.11 to 26.21 <0.001 13.02 4.12 to 21.92 0.005

Male 8.9 �0.08 to 17.88 0.052 4.64 �5.1 to 14.37 0.346 5.92 �3.66 to 15.49 0.222

Left 1.28 �6.33 to 8.89 0.739 3.04 �5.21 to 11.28 0.466 5.52 �2.59 to 13.63 0.18

Age �0.57 �0.88 to �0.26 <0.001 �0.5 �0.84 to �0.16 0.004 �0.39 �0.73 to �0.06 0.021

Duration 1.05 �0.19 to 2.30 0.097 0.18 �1.17 to 1.53 0.793 �0.11 �1.44 to 1.22 0.869

FMA Treatment 1.12 �2.14 to 4.39 0.495 6.71 3.26 to 10.17 <0.001 14.71 10.98 to 18.44 <0.001

Infarction 5.81 2.07 to 9.54 0.003 6.52 2.58 to 10.47 0.001 7.49 3.22 to 11.75 0.001

Male 0.28 �3.74 to 4.30 0.89 0.09 �4.15 to 4.34 0.965 2.25 �2.34 to 6.83 0.333

Left 1.86 �1.54 to 5.27 0.279 1.95 �1.65 to 5.54 0.285 2.49 �1.4 to 6.38 0.206

Age �0.17 �0.31 to �0.03 0.019 �0.21 �0.36 to �0.06 0.006 �0.27 �0.43 to �0.11 0.001

Duration 0.51 �0.05 to 1.07 0.072 0.3 �0.29 to 0.89 0.317 0.36 �0.27 to 1.0 0.258

NOTE. D represents the mean difference between the 2 groups.

Abbreviations: CI, confidence interval; FMA, Fugl-Meyer Assessment; RMANOVA, repeated-measures analysis of variance.

6 D. Wu et al

anodal and cathodal tDCS can influence, depending on theappropriate tDCS treatment at the appropriate time.

Parameters of tDCS used in this study

To minimize the discomfort of tDCS, a constant current of 0.7milliampere intensity was applied for 20 minutes resulting ina current density of 28mA/cm2 (half of the current intensity forother studies using tDCS) at the skin.

How many cathodal tDCS sessions are appropriate is yet to bedetermined. On the one hand, cortical neuroplastic changes arehard to maintain with fewer cathodal tDCS sessions. On the otherhand, ipsilesional S1M1 might be overinhibited with more cath-odal tDCS sessions. Perhaps 4 weeks of treatment (20 cathodaltDCS sessions) is suitable. However, future investigation needs toaddress appropriate dosing and duration of treatment.

Safety of tDCS treatment

The safety of tDCS treatment is closely related to current intensity,electrode size, and stimulation duration.24,25 In this study,a constant current intensity of 0.7 milliampere with an electrodesize of 25cm2 applied for 20 minutes was quite safe.

In previous studies, the most common adverse events duringand after stimulation were a mild tingling sensation and moderatefatigue. Other adverse events using tDCS included mild itching,burning, and pain associated with switching on and off thestimulator. Headache, nausea, nervousness, and difficulties in

concentrating were rare.26,27 In this study, most of the subjectshad no sensation while performing tDCS, while several subjectsfelt a slight itching or tingling in the scalp beneath thetDCS electrode.

Study limitations

(1) The follow-up of this study was relatively short. (2) Functionalimaging methods (eg, fMRI) could be used to evaluate the changeof activation in S1M1 before and after active tDCS treatment. (3)The relatively small number of patients with severe muscle tone(MAS scoreZ3) was another limitation of this study.

Conclusions

UL muscle tone after stroke can be significantly decreased usingcathodal tDCS. Combined with conventional positioning,stretching, and other physical modalities, tDCS may improvemotor function and ADL. The possible mechanism may be relatedto the effect of inhibition of S1M1 hyperactivation, resulting insignificant reductions in muscle tone. tDCS appears to be a novelnoninvasive means to treat increased muscle tone and seemsworthy of further investigation.

Suppliers

a. Zhineng Electronics, Inc, No. 7, Yu Lin Bei Lu, Chengdu610041, China.

b. SPSS, Inc, 233 S Wacker Dr, 11th Fl, Chicago, IL 60606.

www.archives-pmr.org

Fig 2 Graphic display of active tDCS and sham tDCS group after adjustment for age, duration, sex, location, and diagnosis. A pretty identifiable

trend: no significant difference at admission, presenting significant difference after treatment, and getting more and more significant after

follow-up for clinical assessments of MAS, FMA, and BI. Error bar: The confidence interval is 95%. Abbreviations: BI, Barthel Index; FMA, Fugl-

Meyer Assessment.

Decreasing muscle tone using noninvasive brain stimulation 7

Keywords

Electric stimulation therapy; Muscle spasticity; Rehabilitation;Stroke; Upper-limb

Corresponding author

Dongyu Wu, MD, Dept of Rehabilitation, Xuanwu Hospital ofCapital Medical University, No. 45, Changchun St, XuanwuDistrict, PO Box 100053, Beijing, China. E-mail address:wudongyu73@yahoo.com.cn.

www.archives-pmr.org

Acknowledgments

We thank Huijun Zhou, PhD, Saw Swee Hock School of PublicHealth, Singapore, for his help in statistical analysis.

References

1. Mayer NH. Clinicophysiologic concepts of spasticity and motor

dysfunction in adults with an upper motoneuron lesion. Muscle Nerve

Suppl 1997;6:S1-13.

2. Katz RT, Rymer WZ. Spastic hypertonia: mechanisms and measure-

ment. Arch Phys Med Rehabil 1989;70:144-55.

8 D. Wu et al

3. Urban PP, Wolf T, Uebele M, et al. Occurence and clinical predictors

of spasticity after ischemic stroke. Stroke 2010;41:2016-20.

4. WatkinsCL,LeathleyMJ,Gregson JM,MooreAP,SmithTL,SharmaAK.

Prevalence of spasticity post stroke. Clin Rehabil 2002;16:515-22.

5. Elovic E, Bogey R. Spasticity and movement disorder. In: DeLisa JA,

Gans BM, Walsh NE, et al, editors. Physical medicine and rehabili-

tation: principles and practice. 4th ed. Philadelphia: Lippincott Wil-

liams & Wilkins; 2004. p 1428-41.

6. Nitsche MA, Paulus W. Excitability changes induced in the human

motor cortex by weak transcranial direct current stimulation. J Physiol

2000;527:633-9.

7. Nitsche MA, Seeber A, Frommann K, et al. Modulating parameters of

excitability during and after transcranial direct current stimulation of

the human motor cortex. J Physiol 2005;568:291-303.

8. Baudewig J, Nitsche MA, Paulus W, Frahm J. Regional modulation

of BOLD MRI responses to human sensorimotor activation by

transcranial direct current stimulation. Magn Reson Med 2001;45:

196-201.

9. Gandiga PC, Hummel FC, Cohen LG. Transcranial DC stimulation

(tDCS): a tool for double-blind sham-controlled clinical studies in

brain stimulation. Clin Neurophysiol 2006;117:845-50.

10. Lindberg PG, Gaverth J, Fagergren A, Fransson P, Forssberg H,

Borg J. Cortical activity in relation to velocity dependent movement

resistance in the flexor muscles of the hand after stroke. Neurorehabil

Neural Repair 2009;23:800-10.

11. Tombari D, Loubinoux I, Pariente J, et al. A longitudinal fMRI study:

in recovering and then in clinically stable sub-cortical stroke patients.

Neuroimage 2004;23:827-39.

12. FoulkesMA,WolfPA,PriceTR,Mohr JP,HierDB.TheStrokeDataBank:

design, methods, and baseline characteristics. Stroke 1988;19:547-54.

13. Boggio PS, Castro LO, Savagim EA, et al. Enhancement of non-

dominant hand motor function by anodal transcranial direct current

stimulation. Neurosci Lett 2006;404:232-6.

14. Hummel F, Celnik P, Giraux P, et al. Effects of non-invasive cortical

stimulation on skilled motor function in chronic stroke. Brain 2005;

128:490-9.

15. Bohannon RW, Smith MB. Interrater reliability of a modified Ash-

worth scale of muscle spasticity. Phys Ther 1987;67:206-7.

16. Gregson JM, Leathley M, Moore AP, Sharma AK, Smith TL,

Watkins CL. Reliability of the Tone Assessment Scale and the

modified Ashworth scale as clinical tools for assessing poststroke

spasticity. Arch Phys Med Rehabil 1999;80:1013-6.

17. Fugl-Meyer AR, Jaasko L, Leyman I, Olsson S, Steglind S. The post-

stroke hemiplegic patient, 1: a method for evaluation of physical

performance. Scand J Rehabil Med 1975;7:13-31.

18. Mahoney FI, Barthel DW. Functional evaluation: the Barthel Index.

Md State Med J 1965;14:61-5.

19. Shah S, Vanclay F, Cooper B. Improving the sensitivity of the Barthel

Index for stroke rehabilitation. J Clin Epidemiol 1989;42:703-9.

20. Pandyan AD, Hermens H, Johnson GR. Spasticity. In: Squire LR,

editor. Encyclopedia of neuroscience. Oxford: Academic Pr; 2009.

21. Manganotti P, Acler M, Formaggio E, et al. Changes in cerebral

activity after decreased upper-limb hypertonus: an EMG-fMRI study.

Magn Reson Imaging 2010;28:646-52.

22. Fregni F, Boggio PS, Mansur CG, et al. Transcranial direct current

stimulation of the unaffected hemisphere in stroke patients. Neuro-

report 2005;16:1551-5.

23. Hummel F, Cohen LG. Improvement of motor function with nonin-

vasive cortical stimulation in a patient with chronic stroke. Neuro-

rehabil Neural Repair 2005;19:14-9.

24. Iyer MB, Mattu U, Grafman J, Lomarev M, Sato S, Wassermann EM.

Safety and cognitive effect of frontal DC brain polarization in healthy

individuals. Neurology 2005;64:872-5.

25. Nitsche MA, Nitsche MS, Klein CC, Tergau F, Rothwell JC, Paulus W.

Level of action of cathodal DC polarisation induced inhibition of the

human motor cortex. Clin Neurophysiol 2003;114:600-4.

26. Nitsche MA, Boggio PS, Fregni F, Pascual-Leone A. Treatment of

depression with transcranial direct current stimulation (tDCS):

a review. Exp Neurol 2009;219:14-9.

27. Poreisz C, Boros K, Antal A, Paulus W. Safety aspects of transcranial

direct current stimulation concerning healthy subjects and patients.

Brain Res Bull 2007;72:208-14.

www.archives-pmr.org