Scholarly article on topic 'Daily Repetitive Transcranial Magnetic Stimulation for Poststroke Upper Limb Paresis in the Subacute Period'

Daily Repetitive Transcranial Magnetic Stimulation for Poststroke Upper Limb Paresis in the Subacute Period Academic research paper on "Medical engineering"

CC BY-NC-ND
0
0
Share paper
OECD Field of science
Keywords
{"Repetitive transcranial magnetic stimulation" / "motor cortex stimulation" / rehabilitation / "subacute stroke" / "stroke recovery" / "randomized controlled trial"}

Abstract of research paper on Medical engineering, author of scientific article — Koichi Hosomi, Shayne Morris, Tomosaburo Sakamoto, Junji Taguchi, Tomoyuki Maruo, et al.

Background We conducted a randomized, double-blind, sham-controlled study to assess the efficacy in motor recovery and safety of daily repetitive transcranial magnetic stimulation (rTMS) in subacute stroke patients. Methods Forty-one patients were randomly assigned to a real or sham stimulation group. Each patient underwent regular rehabilitation accompanied by a series of 10 daily 5-Hz rTMS of the ipsilesional primary motor cortex (M1) or sham stimulation. The primary outcome was motor recovery evaluated by the Brunnstrom stages (BS). The secondary outcomes were improvement in the Fugl-Meyer Assessment (FMA), grip power, National Institutes of Health Stroke Scale (NIHSS), Functional Independence Measure (FIM), a quantitative measurement of finger tapping movement, and the incidence of adverse events. Results Thirty-nine patients completed the study and were included in the analyses. The real rTMS group demonstrated additional improvement in the BS hand score at the last follow-up compared to the sham. The grip power, the NIHSS motor score, and the number of finger taps in the affected hand improved in the real stimulation group but not in the sham group. The BS upper limb scores, the FMA distal upper limb score, the NIHSS total score, and the FIM motor score showed improvement from baseline at the earlier time points after the real rTMS. There were no additional improvements in the other scores after the real rTMS compared to the sham. No serious adverse events were observed. Conclusions Our results suggest that dailyhigh-frequency rTMS of the ipsilesional M1 is tolerable and modestly facilitates motor recovery in the paralytic hand of subacute stroke patients.

Academic research paper on topic "Daily Repetitive Transcranial Magnetic Stimulation for Poststroke Upper Limb Paresis in the Subacute Period"

Daily Repetitive Transcranial Magnetic Stimulation for Poststroke Upper Limb Paresis in the Subacute Period

Koichi Hosomi, md, PhD,*t Shayne Morris, md, PhD/t1 Tomosaburo Sakamoto, md, PhD,| Junji Taguchi, md, PhD,§ Tomoyuki Maruo, md, PhD,*t Yu Kageyama, md, PhD,t Yusuke Kinoshita, md, PhD,§ Yuko Goto, md, PhD,*t Toshio Shimokawa, PhD, | Tetsuo Koyama, md, PhD,*! and Youichi Saitoh, md, PhD*t

Background: We conducted a randomized, double-blind, sham-controlled study to assess the efficacy in motor recovery and safety of daily repetitive transcranial magnetic stimulation (rTMS) in subacute stroke patients. Methods: Forty-one patients were randomly assigned to a real or sham stimulation group. Each patient underwent regular rehabilitation accompanied by a series of 10 daily 5-Hz rTMS of the ipsilesional primary motor cortex (M1) or sham stimulation. The primary outcome was motor recovery evaluated by the Brunnstrom stages (BS). The secondary outcomes were improvement in the Fugl-Meyer Assessment (FMA), grip power, National Institutes of Health Stroke Scale (NIHSS), Functional Independence Measure (FIM), a quantitative measurement of finger tapping movement, and the incidence of adverse events. Results: Thirty-nine patients completed the study and were included in the analyses. The real rTMS group demonstrated additional improvement in the BS hand score at the last follow-up compared to the sham. The grip power, the NIHSS motor score, and the number of finger taps in the affected hand improved in the real stimulation group but not in the sham group. The BS upper limb scores, the FMA distal upper limb score, the NIHSS total score, and the FIM motor score showed improvement from baseline at the earlier time points after the real rTMS. There were no additional improvements in the other scores after the real rTMS compared to the sham. No serious adverse events were observed. Conclusions: Our results suggest that daily

From the 'Department of Neuromodulation and Neurosurgery, Osaka University Graduate School of Medicine, Osaka, Japan; tDepartment of Neurosurgery, Osaka University Graduate School of Medicine, Osaka, Japan; JKansai Rehabilitation Hospital, Osaka, Japan; §Takarazuka Rehabilitation Hospital, Hyogo, Japan; ¡Clinical Study Support Center, Wakayama Medical University, Wakayama, Japan; and ^Department of Rehabilitation Medicine, Nishinomiya Kyoritsu Neurosurgical Hospital, Nishinomiya, Hyogo, Japan.

Received August 12, 2015; revision received February 3, 2016; accepted February 16, 2016.

Grant support: This study was partly supported by the Strategic Research Program for Brain Sciences by the Ministry of Education, Culture, Sports, Science and Technology of Japan (15dm0107049h0003), the General Insurance Association of Japan, and Japan Agency for Medical Research and Development (15hk0102029h0001).

Disclosure: The Department of Neuromodulation and Neurosurgery, Osaka University Graduate School of Medicine, is a joint research chair established with sponsorship by Teijin Pharma Limited.

Address correspondence to Youichi Saitoh, MD, PhD, Department of Neuromodulation and Neurosurgery, Osaka University Graduate School of Medicine, 2-2 Yamadaoka, Suita, Osaka 565-0871, Japan. E-mail: neurosaitoh@mbk.nifty.com.

1 These authors contributed equally to this work.

1052-3057/$ - see front matter

© 2016 The Authors. Published by Elsevier Inc. on behalf of National Stroke Association. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

http://dx.doi.org/10.1016Zj.jstrokecerebrovasdis.2016.02.024

Journal of Stroke and Cerebrovascular Diseases, Vol. 25, No. 7 (July), 2016: pp 1655-1664

high-frequency rTMS of the ipsilesional Ml is tolerable and modestly facilitates motor recovery in the paralytic hand of subacute stroke patients. Key Words: Repetitive transcranial magnetic stimulation—motor cortex stimulation— rehabilitation—subacute stroke—stroke recovery—randomized controlled trial. © 2016 The Authors. Published by Elsevier Inc. on behalf of National Stroke Association. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Introduction

Poststroke motor disturbance not only reduces the quality of life and activities of daily living of patients but also has a great social impact through lost productivity. With this in mind, efforts have been made to improve the functional outcomes of poststroke patients undergoing rehabilitation. One such approach is repetitive transcranial magnetic stimulation (rTMS), with the purpose of facilitating poststroke recovery of motor function. Two common approaches have been advocated. One of these is low-frequency rTMS (1 Hz or less) to the contralesional primary motor cortex (M1) area to decrease excessive excitability and thus decrease excessive interhemispheric inhibition to the ipsilesional side. The other approach involves high-frequency rTMS (greater than 1 Hz), or excitatory stimulation, to facilitate the decreased cortical excitability on the stroke-affected side.1-4 In the past, there have been a variety of reports on the use of low-frequency rTMS to the contralesional M1,2,4-8 and high-frequency rTMS to the ipsilesional M11-3,6-12 with the purpose of neuroreha-bilitation. About half of these studies have involved chronic stage infarction patients.2,4,5,7,10-12 Although some studies have shown improvements in acute stroke patients both as a result of low-frequency contralesional stimulation and as a result of high-frequency ipsilesional stimulation,3,6,8,9 a few randomized double-blind controlled trials have investigated the efficacy of high-frequency ipsilesional stimulation in subacute stroke patients.8 As functional recovery after stroke is said to be most pronounced in the period within 3 months after onset,13 we postulated that rather than rTMS at the chronic stage, the add-on effects of rTMS may be greater when it is applied at an earlier stage. To study the add-on effects of rTMS on ischemic and hemorrhagic subacute stroke patients, we undertook a randomized, double-blind, parallel study to test the hypothesis that 10 sessions of daily rTMS, combined with regular rehabilitation, improve the results of recovery of motor function in subacute stroke patients.

Methods

Patients

This was a randomized, double-blind, sham-controlled, parallel study conducted at 3 centers (a university hospital and 2 rehabilitation hospitals) in Japan from September, 2010, to December, 2012. We enrolled patients with the

following conditions: (1) 20 years old and over, (2) motor disturbances in the upper limb caused by ischemic or hemorrhagic stroke (Brunnstrom stages [BS]14 arm < 5 or BS hand < 5), and (3) within 8 weeks of stroke onset. The following conditions excluded patients from participating in the present study: (1) total paralysis of the upper limb (BS arm = 1 and BS hand = 1); (2) contraindications to transcranial magnetic stimulation, such as the implantation of a cardiac pacemaker; (3) previous rTMS; (4) aphasia, dementia, psychological disorders, or suicidal wishes; (5) a history of epilepsy; and (6) pregnancy.

This randomized controlled study was conducted in accordance with the Declaration of Helsinki and Japanese ethical guidelines for clinical studies. The study protocol was thoroughly reviewed and approved by the institutional review boards and the ethics committees of all the participating institutions (approval number, 09278-2). The protocol was finalized on September 1, 2010, and this clinical trial was registered with the Japanese University Hospital Medical Information Network Clinical Trials Registry, number UMIN000007594. All patients provided written informed consent and approval before enrollment.

Randomization

The participants were recruited from 2 hospitals specializing in rehabilitation, where they received daily rehabilitation. Randomization was performed using a computer-generated permuted-block method by a third-party statistician upon confirmation of patient eligibility, prior to the start of the study. Patients were randomly assigned to 1 of 2 treatment groups (real rTMS plus regular rehabilitation therapy versus sham stimulation plus regular rehabilitation therapy) according to age (<65 and > 65 years old), severity of symptoms (BS hand score < 3 [severe] and > 4 [mild]), and institution. The patients were identified by sequential numbers assigned at randomization. An assignment notice was sent only to investigators who conducted the rTMS intervention. The patients and assessors were blinded to group assignment until the study was completed.

Procedures

Stimulation sessions were undertaken daily for 10 consecutive days except for weekends, after which follow-up evaluations were undertaken over the next 2 weeks

Figure 1. The intervention and evaluation schedule. Abbreviations: BS, Brunnstrom stage; FIM, Functional Independence Measure; FMA, Fugl-Meyer Assessment; NIHSS, National Institutes of Health Stroke Scale; rTMS, repetitive transcranial magnetic stimulation.

at each rehabilitation hospital. In principle, the stimulation began on a Monday (day 1) and finished on the Friday of the following week (day 12), with follow-ups until the Monday of the fourth week (day 29). Standard rehabilitation was undertaken on a daily basis, including weekends and public holidays, during and after the study period. Physical therapy and occupational therapy were provided, with speech therapy also undertaken by patients who required it. Therapies started in most cases within 1 hour of the completion of rTMS sessions. Daily rehabilitation consisted of 8 therapy sessions each lasting for 20 minutes. Of these sessions, occupational therapy made up 3 sessions per day, including gross motor training in the proximal upper extremity, motor training of hand dexterity, training of coordinated movement with both hands, and practices for activities of daily living.

The evaluations were performed by assessors who were blinded to the group assignment. Figure 1 shows the time schedule of the evaluations. All evaluations except for the finger tapping measurement were undertaken prior to rehabilitation sessions at baseline, and on days 5 (Assessment 1), 12 (Assessment 2), and 29 (Assessment 3). They included BS arm, BS hand, BS lower limb,14 Fugl-Meyer Assessment (FMA) total score, FMA proximal upper limb score (shoulder, elbow, and forearm motor functions), FMA distal upper limb score (wrist and hand motor functions),15 handgrip of both hands, National Institutes of Health Stroke Scale (NIHSS) total score, NIHSS motor arm score,16 Functional Independence Measure (FIM) motor score, and FIM cognitive score.17 Three days prior to the start of stimulations, objective estimations of finger tapping movements were obtained using a system with

magnetic sensors (UB-1; Finger Tapping Movement Analyzer; Hitachi Corporation, Tokyo, Japan) that continuously monitored the distance between 2 coils via a calibration method.18 Using this system, we quantitatively measured the total distance traveled, the mean maximum amplitude, the mean maximum opening velocity, the mean maximum closing velocity, an estimate of total consumed energy (sum of squares of velocity), and the number of finger taps during 30 seconds with respect to the movement of the index finger and the thumb of both hands.19 The same tests were undertaken after the full completion of the individual's stimulation series (days 15-17).

Prior to the stimulation period, the location of the M1 hand knob in the affected hemisphere was located using a transcranial magnetic stimulation navigation system (Brainsight; Rogue Research Inc., Montreal, Quebec, Canada). This location was then marked on the scalp and measured so that the same location could be rapidly determined without a navigation system. The rTMS was applied using a figure-8 coil (MC B-70; Medtronic Functional Diagnostics A/S, Skovlunde, Denmark; or no. 9925-00; Magstim Co Ltd, Whitland, United Kingdom) connected to a magnetic stimulator (MagPro, Medtronic Functional Diagnostics A/S; or Magstim Rapid, Magstim), which provided repetitive biphasic pulses. The patients were positioned in the supine position, and their heads were fixed. The predetermined target was located from the previously made marking, and fine adjustments in the coil location were made to confirm the optimal spot according to visual detection of muscle twitches if muscle twitches were observed. The resting motor threshold was defined as the minimal intensity necessary to induce at least 1 visible muscle twitch in the affected hand on a session by session basis, which corresponds to that determined using an electromyogram,20 and the intensity of real rTMS was set to 90% of the resting motor threshold for that day. In patients without muscle twitches, the intensity of the real rTMS was set to 100 A/ |is for the MagPro or maximum output for the Magstim Rapid, which is approximately equivalent to the stimulus intensity. Five hundred pulses per session were delivered to the M1 hand in the affected hemisphere (10 trains of 5 Hz for 10 seconds with a 50-second intertrain interval). Sham stimulations were applied with the same parameters as real stimulations, but the coil was placed at a 90° angle to the scalp.21 Standard guidelines on the safe use of rTMS were followed in the present study.22

The finger tapping measurement and the localization of the M1 hand knob were undertaken at Osaka University Hospital. Other evaluations, daily rTMS, and rehabilitation were undertaken at the 2 rehabilitation hospitals.

Statistical Analysis

The primary end point was the BS. The secondary end points were the FMA score, handgrip strength score, NIHSS

score, FIM score, the finger tapping measurement, and the incidence of adverse events. A target number of 20 real and 20 sham subjects were decided upon based on a previous rTMS study,6 from which we expected BS improvements of 1.3 in a real stimulation group and of .35 in a sham stimulation group with the same standard deviation of 1.0. We calculated this sample size with a power of 80% at an a level of .05 to detect the effect of rTMS, allowing for a drop-out rate of 5%. The projected study period was 2 years, and no interim analysis was planned. Differences in baseline patient characteristics and scores between the 2 assigned groups were assessed with a 2-sample i-test for continuous data, Mann-Whitney's U-test for ordinal data (BS, FMA, NIHSS, and FIM scores), and Fisher's exact test for nominal data. Regarding analyses of rTMS efficacy, first, the improvement over time with respect to baseline scores was evaluated in each group using a paired i-test for the finger tapping measurement, a repeated measures analysis of variance (within-subject factor, day [baseline, days 5, 12, and 29]) for handgrip, and Friedman test for the other evaluations with ordinal scales (BS, FMA, NIHSS, and FIM). Second, in the improved scores except for the finger tapping measurement, we used the Dunnett multiple comparisons or Wilcoxon signed rank tests with the Bonferroni correction as post hoc analyses to evaluate improvement from the baseline score at each time point after stimulation. Third, differences in the improvement at the last evaluation time point between the 2 groups were evaluated by a 2-sample i-test for the finger tapping measurement and handgrip and Mann-Whitney's U-tests for the other evaluations. For all comparisons, findings with P values less than .05 were considered statistically significant. Data were analyzed with the JMP Pro 11.2.1 software (SAS Institute Inc., Cary, NC), and the Statistics Toolbox implemented in MATLAB 8.3.0 (MathWorks Inc., Natick, MA).

Results

Figure 2 shows the trial profile. Forty-one patients were enrolled in the present study. Twenty patients were assigned to the real stimulation group and 21 to the sham stimulation group. Two subjects in the real stimulation group failed to complete the study. One patient did not like the stimulation sensation on the scalp after one 10-second train of stimulations on the first day of stimulations, and declined following stimulations. The other patient also temporarily refused other forms of treatment such as regular rehabilitation during the first week of the stimulation. Thirty nine patients were finally included in the analysis after the removal of these 2 patients because evaluations could not be performed at all after the beginning of interventions. Table 1 shows the baseline patient characteristics of the 39 patients analyzed. The mean age of the participants was 62.9 ± 13.8

Enrolled and randomized (n=41)

Allocated real stimulation group (n=20; Hospital A: 12, B: 8)

Discontinued intervention (n=2; Hospital A: 2)

Completed allocated intervention and evaluation (n-18)

Allocated sham (n=21; Hospit timulation group al A: 13, B: 8)

Completed allocated intervention and evaluation (n-21)

Figure 2. Trial profile. Flowchart shows the organizational structure of the study, with the number of patients initially enrolled (n = 41), and the number of patient dropouts during the study. The final number of patients in each assigned group is indicated.

years old, and the mean postonset duration at the start of the intervention was 45.5 ± 9.0 days (range, 25-56 days). One patient with partial damage of the M1 hand area was included in each group. There were no differences in the baseline characteristics, scores, and lesion size between the real and sham stimulation groups, except for the FIM cognitive score.

Improvemenis in the BS (Primary Ouicome)

The BS of all regions (arm, hand, and lower limb) improved significantly over time with respect to baseline scores for both real and sham stimulation groups (Table 2). Figure 3 indicates changes over baseline scores in the BS. The multiple comparisons showed that the real stimulation group achieved earlier improvement from baseline in the BS arm and hand scores when compared to the sham stimulation group. The BS arm score was significantly improved at day 29 in the real stimulation group but not in the sham stimulation group. The BS hand score was significantly improved at days 12 and 29 in the real stimulation group but only at day 29 in the sham stimulation group. The BS lower limb score did not show a significant improvement at any of the time points after stimulation (Table 3). Improvement over baseline scores in the BS hand score at day 29 was significantly greater in the real stimulation group than in the sham stimulation group (P = .037). Although improvement in the BS arm score tended to be greater in the real stimulation group, the difference between the real and sham groups was not significant for the arm and lower limb scores (P = .294 and P = .747, respectively).

Improvemenis in the FMA, Handgrip Strength,

NIHSS, and FIM Scores

The FMA total score, FMA proximal upper limb score, FMA distal upper limb score, NIHSS total score, and FIM subscores improved significantly over time with respect to baseline scores for both real and sham stimulation

Table 1. Patient characteristics

Characteristics Real (n = 18) Sham (n = 21) P value

Sex (female), n (%) 8 (44) 8 (38) .94

Age (years) 62.4(15.5) 63.2(12.5) .87

Type of stroke, n (%)

Hemorrhage 6 (33) 9 (43) .78

Infarction 12(67) 12 (57)

Distribution of stroke lesion, n (%)

Subcortex 17 (94) 16(71) .26

Involvement of cortex 1(6) 5(29)

Affected hemisphere

Right 11 13 .78

Left 7 8

Postonset duration (days) 46.1 (8.7) 45.1 (9.5) .79

Brunnstrom stages

Arm 3.8 (1.1) 4.0 (1.3) .52

Hand 3.6 (1.5) 3.7 (1.5) .87

Lower limb 4.6 (1.2) 4.4(1.4) .74

Fugl-Meyer Assessment score

Total 176.3 (29.6) 169.1 (37.9) .92

Motor 62.3 (23.9) 61.9 (28.4) .96

Handgrip strength score (kg)

Affected hand 4.9 (5.7) 7.1 (8.6) .38

Nonaffected hand 26.1 (9.1) 26.3 (9.6) .94

NIHSS score 4.4 (2.8) 4.6 (3.2) .99

FIM score

Motor 63.9 (13.4) 62.7 (17.6) .92

Cognitive 31.8 (4.8) 28.8 (4.6) .02*

Resting motor threshold (%)t 67.7 (16.1) 72.5 (16.0) .62

Study hospital

A B 10 8 13 8 .94

Abbreviations: FIM, Functional Independence Measure; NIHSS, National Institutes of Health Stroke Scale; SD, standard deviation.

Data are means (SD), numbers (%), or numbers. There were no differences in the baseline characteristics and scores between real and sham stimulation groups.

*Significant difference between 2 groups.

tResting motor threshold is shown as percentage of maximum output. Data from 7 patients in each group because of an absence of motor evoked potentials and motor twitches.

groups. Handgrip strength scores for the nonaffected hand did not significantly improve over time for the real or for the sham stimulation group. Meanwhile, in the affected hand, only the real stimulation group showed an improvement in handgrip strength score. The NIHSS motor arm score significantly improved only in the real stimulation group but not in the sham stimulation group (Table 2). The multiple comparisons showed that the real stimulation group achieved earlier improvement from baseline in the FMA distal upper limb score, NIHSS total score, and FIM motor score when compared to the sham stimulation group. The FMA distal score significantly improved at days 12 and 29 in the real stimulation group but only at day 29 in the sham stimulation group. The NIHSS total score was significantly improved at all time points in the real stimulation group but not in the sham stimul ation

group (Table 3). There were no significant differences between the real and sham stimulation groups in improvements in any of the secondary end points over baseline scores at day 29.

Finger Tapping Motion

There was a significant increase in the number of taps made in the real stimulation group (P = .006), whereas no significance was observed in the sham stimulation group (P = .092). The change from the baseline was not significantly different between the 2 groups (P = .068). No significant improvement was seen in the total distance traveled, mean maximum amplitude, mean maximum opening velocity, mean maximum closing velocity, or the estimate of total consumed energy (Table 4).

Table 2. Results of clinical scores

Clinical scores Baseline Day 5 Day 12 Day 29 P value

Brunnstrom stages Arm Real 4 [3-5] 4 [3.3-5] 4 [4-5] 5 [4-5] <.001*

3.8 (1.1) 3.9(1.0) 4.2(1.1) 4.3 (1.2)

Sham 5 [3-5] 5 [3-5] 5 [3-5] 5 [3-5] .004*

4.0(1.3) 4.0(1.3) 4.2(1.2) 4.3 (1.2)

Hand Real 4 [2-5] 4 [3-5] 5 [3.3-5] 5 [4-5] <.001*

3.6 (1.5) 3.9(1.1) 4.2(1.1) 4.4(1.2)

Sham 4 [2-5] 4 [2-5] 4 [3-5] 4 [3-5] .001*

3.7 (1.5) 3.9(1.5) 4.0(1.4) 4.1 (1.4)

Lower limb Real 5 [4-5.8] 5 [4-6] 5 [4.3-6] 5 [5-6] .013*

4.6 (1.2) 4.7 (1.2) 4.8 (1.2) 4.9(1.0)

Sham 5 [3-6] 4 [4-6] 5 [4-6] 5 [4-6] .004*

4.4(1.4) 4.5 (1.2) 4.7 (1.2) 4.8 (1.2)

Fugl-Meyer Assessment score Total Real 183 [158-198] 185 [170-201] 193.5 [176-209] 194 [172-211] <.001*

Sham 174 [133-205] 191 [138-208] 187 [158-209] 191 [163-211] <.001*

Proximal upper limb Real 26.5 [19-32] 30 [20-34] 31.5 [22-34] 31.5 [21-35] <.001*

Sham 25 [11-35] 29.5 [12-35] 29 [14-35] 33 [16-36] <.001*

Distal upper limb Real 13 [2.3-22] 13.5 [3.8-22] 15 [4.8-22] 17 [3-22] <.001*

Sham 12 [2-20] 18 [2-23] 18 [3-22] 17 [2.8-23] <.001*

Handgrip strength score (kg) Affected hand Real 4.9 (5.7) 5.4 (5.5) 6.1 (5.9) 6.5 (6.8) .041*

Sham 7.1 (8.6) 7.8 (8.4) 8.2 (8.9) 8.3 (8.6) .077

Nonaffected hand Real 26.1 (9.1) 27.2 (8.0) 27.3 (8.8) 27.7 (8.4) .159

Sham 26.3 (9.6) 26.4 (9.7) 25.7 (9.7) 26.8 (9.8) .388

NIHSS score Total Real 3.5 [2.3-6] 3 [2-5] 3 [1-4.8] 2 [1-4.8] <.001*

Sham 3 [2-7] 3 [2-6] 3 [1-6] 3 [1-6] .004*

Motor arm Real 2 [0-2] 1 [0-2] 0 [0-2] 0 [0-2] <.001*

Sham 0 [0-3] 0 [0-2] 0 [0-2] 0 [0-2] .097

FIM score Motor Real 65 [53-75] 68 [56-75] 69 [61-79] 71 [64-80] <.001*

Sham 64 [48-78] 71 [52-81] 71 [52-80] 73 [55-82] <.001*

Cognitive Real 34 [31-35] 34 [31-35] 34 [31-35] 35 [31-35] .003*

Sham 28 [25-34] 29 [25-34] 31 [26-34] 31 [27-35] <.001*

Abbreviations: FIM, Functional Independence Measure; IQR, interquartile range; NIHSS, National Institutes of Health Stroke Scale; SD, standard deviation. ^

Data are expressed as medians [IQR] or means (SD). Both medians and means are indicated for each score of the Brunnstrom stages. Improvement over time with respect to a baseline score was § statistically evaluated and each P value is shown. P

*Significant improvement over time with respect to baseline. &

Figure 3. Improvements in the BS score. Improvement over baseline scores in the BS of the arm (A), hand (B), and lower limb (C). The BS arm score significantly improved at day 29 in the real stimulation group but not in the sham stimulation group. The BS hand score significantly improved at days 12 and 29 in the real stimulation group, but only at day 29 in the sham stimulation group. The BS lower limb score did not show a significant improvement at any of the time points after stimulation. Improvement over baseline scores in the BS hand score at day 29 was significantly greater in the real stimulation group than in the sham group (P = .037). However, the difference between the 2 groups at day 29 was not significant for the other scores. Abbreviation: BS, Brunnstrom stage; *, P <.05; **, P <.001; —*— real; - » - sham.

Adverse Events

There were no serious adverse effects observed during or after the stimulations during the 2-week follow up period after completion of stimulations. As was previously mentioned, 1 subject withdrew from the protocol due to uncomfortableness. Incidentally, he later apparently demanded continuation of stimulations, after which he reportedly found the same stimulation pattern relaxing and enjoyable.

Discussion

This double-blind randomized controlled study of daily rTMS targeting the M1 hand area demonstrated a facilitation effect on motor recovery in the paralytic hand of subacute ischemic and hemorrhagic stroke patients without any serious adverse events. The real stimulation provided better improvement in the BS hand score at the last follow-up compared to the sham. The handgrip strength score, the NIHSS motor arm score, and the number of taps significantly improved only after the real stimulation. The real stimulation showed earlier improvement in the BS arm score, BS hand score, FMA distal upper limb score, NIHSS total score, and FIM motor score. However, it did not show an effect on motor recovery in the paralytic leg, or scores for activities of daily living.

The effect of rTMS on poststroke motor recovery has been studied in around a dozen articles, in which high-frequency rTMS on the ipsilesional M11-3,6-12 or low-frequency rTMS on the contralesional M12,4-8 was applied according to the interhemispheric balance hypothesis. A recent meta-analysis reported that rTMS had a positive effect on motor recovery in patients with stroke, especially for those with subcortical stroke.23 Among these studies using high-frequency rTMS, 3 studies from the same group involved poststroke patients at the acute stage,3,6,9 one at the subacute stage,1 three at the chronic stage,7,10,12 as well as two at various stages.2,11 Two studies for chronic stage patients failed to demonstrate any positive effects of high-frequency rTMS on motor function,7,12 whereas the others reported positive effects. However, three of these were non- or pseudorandomized studies,1,10,11 and the others were regarded as having an unclear risk of selection bias in random sequence generation and/or allocation concealment procedures by the Cochrane review and meta-analysis.24 A recent evidence-based guideline issued by a group of European experts stated there may be a possible effect of low-frequency rTMS on the contralesional M1 in acute motor stroke (recommendation level C) and a probable effect in chronic motor stroke (level B), while there may be a possible effect from high-frequency rTMS on the ipsilesional M1 in acute and chronic motor stroke (level C).25 In our study, additional improvements in motor function were demonstrated in a double-blind, randomized manner. The results from our study indicated that the benefits of rTMS were more localized to the particular area being stimulated, in our cases, the affected hand (stimulation of the M1 hand knob). This finding is consistent with the results of a previous report, which tested the effects of high-frequency rTMS on the M1 corresponding to the paretic hand in poststroke patients in the subacute period.1 Chang et al. reported that real rTMS, in conjunction with motor practice, had produced a greater improvement in the arm score of the Motricity Index, but not in the lower limb score, up to 3 months after onset of stroke. Moreover, grip strength

Table 3. Results of multiple comparisons

P value

Clinical scores Day 5 Day 12 Day 29

Brunnstrom stages Arm Real 1.000 .094 .023*

Sham 1.000 .375 .094

Hand Real .211 .003* <.001*

Sham .375 .094 .047*

Lower limb Real 1.000 .188 .094

Sham 1.000 .094 .229

Fugl-Meyer Assessment score Total Real .003* <.001* <.001*

Sham .002* <.001* <.001*

Proximal upper limb Real .045* .007* .003*

Sham .035* .018* .003*

Distal upper limb Real .709 .004* .006*

Sham .182 .051 .002*

Handgrip strength score (kg) Affected hand Real .795 .136 .024*

NIHSS score Total Real .041* .006* .006*

Sham 1.000 .076 .053

Motor arm Real .094 .012* .047*

FIM score Motor Real .001* <.001* .001*

Sham .070 <.001* <.001*

Cognitive Real .375 .188 .047*

Sham 1.000 .188 .012*

Abbreviations: FIM, Functional Independence Measure; NIHSS, National Institutes of Health Stroke Scale. Improvement from a baseline score was statistically evaluated by multiple comparisons and each P value is shown. *Significant improvement from a baseline score.

in the affected hand improved only in the real stimulation group over baseline in that study. Sasaki et al8 also reported improvement in grip strength and finger tapping frequency in acute or subacute stroke patients after a 5-session high-frequency rTMS, which is consistent with our results. Our results could reinforce the evidence of the positive effects on motor recovery after multisession

high-frequency rTMS during the subacute period while the patient was undergoing rehabilitation.

We focused on the subacute period after stroke in the present study, because the period within several months after stroke onset is thought of as a golden period for initiating exogenous restorative therapies, as endogenous repair-related events reach peak levels,13 and

Table 4. Finger tapping measurement

Measurement items Baseline Post stimulation

Total distance (mm) Real 2608 (3001) 3568 (3832)

Sham 3759 (3904) 3866 (3986)

Mean maximum amplitude (mm) Real 44.4 (38.1) 33.3 (25.7)

Sham 50.1 (35.3) 41.3 (35.0)

Mean maximum opening velocity (m/s) Real .35 (.45) .33 (.34)

Sham .33 (.35) .36 (.38)

Mean maximum closing velocity (m/s) Real .35 (.41) .40 (.40)

Sham .48 (.50) .43 (.41)

Total energyt Real 140 (228) 199 (328)

Sham 220 (303) 226 (309)

Number of taps Real 20.7 (20.1) 39.2* (33.3)

Sham 26.6 (23.6) 32.4 (30.9)

Data are expressed as means (SD).

*The number of taps significantly increased in the real stimulation group (P = .006). fSum of squares of velocity.

functional reorganization and plastic changes are seen in the brain.26 In other studies, earlier neurorehabilitation has been proven to result in better outcomes,27 and most improvement generally occurs within the first 3 months after stroke onset.13 Moreover, high-frequency rTMS on the ipsilesional M1 demonstrated favorable results for the acute stage in poststroke patients,3,6,8,9 whereas no significant benefit for the chronic stage was seen in 2 studies.7,12 This would indicate that high-frequency rTMS should probably be initiated at least by the subacute period.

A recent study has suggested a spinal mechanism for the effect of rTMS. It is reported that high-frequency rTMS of the ipsilesional M1 had suppressed F-waves, which was presumed to result from an enhanced inhibitory effect on spinal excitability.28 In our study, the quantitative measurement of finger tapping movement showed that significant increases were seen in the number of taps in the real stimulation group, which has also been observed in a previous study.8,11 This may be partially due to the attenuation of spasticity after high-frequency rTMS.

In the present study, the direct comparison between the real and sham stimulation groups revealed that the significant, but modest, positive effect was limited to improvement in the BS hand score. Although components of the upper limb in other scores tended to show greater improvement after the real stimulation, there were no significant differences in FMA, NIHSS, and FIM scores between the 2 groups. Similar findings were seen in the above-mentioned study reported by Chang et al.1 Additional motor recovery seen in the present study was still modest; therefore, the efficacy of rTMS needs to be improved. To improve the efficacy of rTMS, there are some possible methods that can be utilized. As suprathreshold stimulation has been said to provide more favorable results,3,6,9 the efficacy of rTMS may be improved by a higher intensity of stimulation within the guidelines on the safe use of rTMS. Moreover, some researchers have recently examined the efficacy of coupling inhibitory and facilitatory rTMS suggesting more favorable outcomes when compared to single-session rTMS alone.7

One of major clinical advantages of rTMS is its non-invasive nature. The present study and previous studies have not reported any serious adverse events after rTMS for the treatment of poststroke patients.23 The use of daily high-frequency rTMS during the subacute period seems to be safe in poststroke patients.

Our study has several potential limitations. First, our study was limited to evaluations over 4 weeks. Outcomes over a longer follow-up period should be considered for evaluation in future studies. Second, the small positive result in the BS hand score should be interpreted with caution. The dissociation between the results of the BS hand score and FMA distal upper limb score may be caused by the difference in characteristics of each score; the BS score is a 6-point scale, whereas the FMA distal upper limb score consists of 3-point scales in 8 motor

tasks (total score of 24). Alternatively there is a possibility of overestimation due to rough evaluation of BS. Third, baseline cognitive condition in activities of daily living was unbalanced in the allocated groups. The stroke type and location of our subjects was heterogeneous, and the number of subjects was small for subgroup analysis. Three stratification factors were not optimal for the small number of subjects in the present study. Further studies in larger populations with optimal stratification factors such as a stroke location should be conducted to clarify the various roles of rTMS in poststroke patients.

Conclusions

Our results suggest that daily high-frequency rTMS of the ipsilesional M1 is tolerable and modestly facilitates motor recovery in the paralytic hand of subacute isch-emic and hemorrhagic stroke patients. Further studies investigating more effective conditions are also required to establish rTMS therapy as a practical clinical utility.

Acknowledgments: The authors would like to express appreciation to Ms. Yuko Fukumoto for her administrative assistance, to the administrative and rehabilitation staff of the participating institutions for their efforts, and also to all of the study participants for their enthusiastic participation in the study.

References

1. Chang WH, Kim YH, Bang OY, et al. Long-term effects of rTMS on motor recovery in patients after subacute stroke. J Rehabil Med 2010;42:758-764.

2. Emara TH, Moustafa RR, Elnahas NM, et al. Repetitive transcranial magnetic stimulation at 1 Hz and 5 Hz produces sustained improvement in motor function and disability after ischaemic stroke. Eur J Neurol 2010;17:1203-1209.

3. Khedr EM, Etraby AE, Hemeda M, et al. Long-term effect of repetitive transcranial magnetic stimulation on motor function recovery after acute ischemic stroke. Acta Neurol Scand 2010;121:30-37.

4. Takeuchi N, Chuma T, Matsuo Y, et al. Repetitive transcranial magnetic stimulation of contralesional primary motor cortex improves hand function after stroke. Stroke 2005;36:2681-2686.

5. Fregni F, Boggio PS, Valle AC, et al. A sham-controlled trial of a 5-day course of repetitive transcranial magnetic stimulation of the unaffected hemisphere in stroke patients. Stroke 2006;37:2115-2122.

6. Khedr EM, Abdel-Fadeil MR, Farghali A, et al. Role of 1 and 3 Hz repetitive transcranial magnetic stimulation on motor function recovery after acute ischaemic stroke. Eur J Neurol 2009;16:1323-1330.

7. Takeuchi N, Tada T, Toshima M, et al. Repetitive transcranial magnetic stimulation over bilateral hemispheres enhances motor function and training effect of paretic hand in patients after stroke. J Rehabil Med 2009;41:1049-1054.

8. Sasaki N, Mizutani S, Kakuda W, et al. Comparison of the effects of high- and low-frequency repetitive transcranial magnetic stimulation on upper

limbhemiparesis in the early phase of stroke. J Stroke Cerebrovasc Dis 2013;22:413-418.

9. Khedr EM, Ahmed MA, Fathy N, et al. Therapeutic trial of repetitive transcranial magnetic stimulation after acute ischemic stroke. Neurology 2005;65:466-468.

10. Kim YH, You SH, Ko MH, et al. Repetitive transcranial magnetic stimulation-induced corticomotor excitability and associated motor skill acquisition in chronic stroke. Stroke 2006;37:1471-1476.

11. Ameli M, Grefkes C, Kemper F, et al. Differential effects of high-frequency repetitive transcranial magnetic stimulation over ipsilesional primary motor cortex in cortical and subcortical middle cerebral artery stroke. Ann Neurol 2009;66:298-309.

12. Malcolm MP, Triggs WJ, Light KE, et al. Repetitive transcranial magnetic stimulation as an adjunct to constraint-induced therapy: an exploratory randomized controlled trial. Am J Phys Med Rehabil 2007;86:707-715.

13. Cramer SC. Repairing the human brain after stroke: I. Mechanisms of spontaneous recovery. Ann Neurol 2008;63:272-287.

14. Brunnstrom S. Motor testing procedures in hemiplegia: based on sequential recovery stages. Phys Ther 1966; 46:357-375.

15. Fugl-Meyer AR, Jaasko L, Leyman I, et al. The post-stroke hemiplegic patient. 1. a method for evaluation of physical performance. Scand J Rehabil Med 1975;7:13-31.

16. Brott T, Adams HP Jr, Olinger CP, et al. Measurements of acute cerebral infarction: a clinical examination scale. Stroke 1989;20:864-870.

17. Ottenbacher KJ, Hsu Y, Granger CV, et al. The reliability of the functional independence measure: a quantitative review. Arch Phys Med Rehabil 1996;77:1226-1232.

18. Kandori A, Sano Y, Miyashita T, et al. Estimation method of finger tapping dynamics using simple magnetic detection system. Rev Sci Instrum 2010;81:054303.

19. Maruo T, Hosomi K, Shimokawa T, et al. High-frequency repetitive transcranial magnetic stimulation over the primary foot motor area in Parkinson's disease. Brain Stimul 2013;6:884-891.

20. Hanajima R, Wang R, Nakatani-Enomoto S, et al. Comparison of different methods for estimating motor threshold with transcranial magnetic stimulation. Clin Neurophysiol 2007;118:2120-2122.

21. Hirayama A, Saitoh Y, Kishima H, et al. Reduction of intractable deafferentation pain by navigation-guided repetitive transcranial magnetic stimulation of the primary motor cortex. Pain 2006;122:22-27.

22. Rossi S, Hallett M, Rossini PM, et al. Safety, ethical considerations, and application guidelines for the use of transcranial magnetic stimulation in clinical practice and research. Clin Neurophysiol 2009;120:2008-2039.

23. Hsu WY, Cheng CH, Liao KK, et al. Effects of repetitive transcranial magnetic stimulation on motor functions in patients with stroke: a meta-analysis. Stroke 2012;43:1849-1857.

24. Hao Z, Wang D, Zeng Y, et al. Repetitive transcranial magnetic stimulation for improving function after stroke. Cochrane Database Syst Rev 2013;(5):CD008862.

25. Lefaucheur JP, Andre-Obadia N, Antal A, et al. Evidence-based guidelines on the therapeutic use of repetitive transcranial magnetic stimulation (rTMS). Clin Neurophysiol 2014;125:2150-2206.

26. Johansson BB. Brain plasticity and stroke rehabilitation. The Willis lecture. Stroke 2000;31:223-230.

27. Indredavik B, Bakke F, Slordahl SA, et al. Treatment in a combined acute and rehabilitation stroke unit: which aspects are most important? Stroke 1999;30:917-923.

28. Wupuer S, Yamamoto T, Katayama Y, et al. F-wave suppression induced by suprathreshold high-frequency repetitive trascranial magnetic stimulation in poststroke patients with increased spasticity. Neuromodulation 2013;16:206-211, discussion 211.