- Email: [email protected]
A Systematic Review and Meta-Analysis on the Use of Repetitive Transcranial Magnetic Stimulation for Spasticity Poststroke
Accepted Manuscript A Systematic Review and Meta-Analysis on the Use of Repetitive Transcranial Magnetic Stimulation for Spasticity Post Stroke A. McIntyre, M. Mirkowski, S. Thompson, A.M. Burhan, T. Miller, R. Teasell PII:
S1934-1482(17)31359-X
DOI:
10.1016/j.pmrj.2017.10.001
Reference:
PMRJ 1999
To appear in:
PM&R
Received Date: 4 April 2016 Revised Date:
8 August 2017
Accepted Date: 9 October 2017
Please cite this article as: McIntyre A, Mirkowski M, Thompson S, Burhan A, Miller T, Teasell R, A Systematic Review and Meta-Analysis on the Use of Repetitive Transcranial Magnetic Stimulation for Spasticity Post Stroke, PM&R (2017), doi: 10.1016/j.pmrj.2017.10.001. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
A Systematic Review and Meta-Analysis on the Use of Repetitive Transcranial Magnetic Stimulation for Spasticity Post Stroke
McIntyre A, 1Mirkowski M, 1Thompson S, 2,3Burhan AM, 1-,3Miller T, 1-3Teasell R
1
Lawson Health Research Institute, Parkwood Institute, London, ON
2
St. Joseph’s Health Care, Parkwood Institute, London, ON
SC
3
RI PT
1
M AN U
Schulich School of Medicine and Dentistry, University of Western Ontario, London, ON
Corresponding author/reprints: Amanda McIntyre RN MSc BSc BScN Aging, Rehabilitation, and Geriatric Care, Rm. B3-123C Parkwood Institute 550 Wellington Road
TE D
London, ON, N6C 0A7
Fax: 519-685-4036 Tel.: 519-685-4292 x41296
EP
[email protected]
Cover Title: rTMS for Spasticity Post Stroke # Figures: 3
AC C
# Tables: 3 # Words: 3,971
Key Words: Stroke, rTMS, Repetitive Transcranial Magnetic Stimulation, Spasticity, Muscle Tone Conflict of Interest: Dr. Tom Miller and Dr. Robert Teasell have received funding from Allergan Canada Inc. to establish a Centre for Excellence in Spasticity at St. Joseph’s Health Care in London, Ontario.
ACCEPTED MANUSCRIPT
1
A Systematic Review and Meta-Analysis on the Use of Repetitive Transcranial Magnetic
2
Stimulation for Upper Extremity Spasticity Post Stroke
4
Cover Title: rTMS for Spasticity Post Stroke
5
# Tables: 3
6
# Words: 3,392
# Figures: 5
8
SC
7
Key Words: Stroke, rTMS, Repetitive Transcranial Magnetic Stimulation, Spasticity, Muscle Tone
Sources of Funding: None Declared.
AC C
EP
TE D
11
M AN U
9 10
RI PT
3
ACCEPTED MANUSCRIPT 2
ABSTRACT
12
Background: Spasticity is a common and potentially debilitating complication that develops after
14
stroke, arising in approximately 30% of patients.
15
Objective: To evaluate the effectiveness of repetitive transcranial magnetic stimulation (rTMS) in
16
improving spasticity after stroke.
17
Design: Meta-analysis and systematic review.
18
Setting: N/A
19
Patients: A total of 273 post-stroke (hemorrhagic=123, ischemic=150) participants were included
20
with sample sizes ranging from 5 to 80. The majority of participants were male (66.0%) with a
21
mean age ranging 55.0-64.6 years. Mean stroke duration ranged 6 months – 10 years.
22
Methods: A literature search of multiple databases was conducted for articles published in English
23
from January 1980 to April 2015 using select keywords. Studies were included if: 1) the population
24
included was >50% stroke patients; 2) the sample size included ≥4 subjects; 3) the intervention
25
applied was rTMS; and 4) upper extremity spasticity was assessed pre and post intervention.
26
Randomized controlled trials (RCTs) were assessed for methodological quality using the
27
Physiotherapy Evidence Database (PEDro) tool. All research designs were given a level of evidence
28
according to a modified Sackett Scale.
29
Main Outcome Measurements: Modified Ashworth Scale (MAS).
30
Results: Ten studies met the inclusion criteria: two RCTs (PEDro scores 8-9) and eight pre-post
31
studies. Meta-analyses of primarily uncontrolled pre-post studies found significant improvements in
32
MAS for elbow (p<.001), wrist (p<.001), and finger flexors (p<.001). However, a meta-analysis of
33
the two available RCTs failed to find a significant rTMS treatment effect on MAS for the wrist
34
(standardized difference=.34, p=.30).
AC C
EP
TE D
M AN U
SC
RI PT
13
ACCEPTED MANUSCRIPT 3
Conclusions: There is limited available evidence to support the use of rTMS in improving spasticity
36
post stroke. Despite the positive findings reported, better powered and appropriately controlled
37
trials are necessary.
AC C
EP
TE D
M AN U
SC
RI PT
35
ACCEPTED MANUSCRIPT 4
INTRODUCTION
39
Stroke is considered to be the leading cause of adult disability, often causing motor
40
impairment and contributing to long-term neurological disability.1 Spasticity and increased tone2 is
41
a common and potentially debilitating complication that develops after stroke, arising in
42
approximately 30% of patients, with a highly variable timing of onset.3 Spasticity is defined as a
43
motor disorder characterized by a velocity dependent increase in tone, and is part of the upper
44
motor neuron (UMN) syndrome. Spasticity management helps improve the quality of life of many
45
stroke survivors.4 Current modalities to decrease tone in the UMN syndrome have been shown to
46
improve function. Current modalities such as medications or botulinum toxin injections are able to
47
decrease tone generally or locally; however, treating spasticity does not necessarily improve
48
weakness or motor control. As recovery of motor function after stroke is usually incomplete, recent
49
research has focused on novel neurorehabilitation techniques to enhance the beneficial effects of
50
treatment aimed at motor recovery.5,6 Among these are various methods of cortical stimulation,
51
which include interventional paired associative stimulation, transcranial direct current stimulation
52
(tDCS), implanted epidural motor cortex stimulation, and repetitive transcranial magnetic
53
stimulation (rTMS).7
TE D
M AN U
SC
RI PT
38
rTMS is a noninvasive therapeutic intervention involving the generation of a strong
55
magnetic field by a powerful electrical coil that passes unimpeded through the skull to the cerebral
56
cortex producing an electric current that painlessly stimulates the targeted brain area.8 A train of
57
pulses is delivered to a particular cortical region at a given intensity, as well as frequency,
58
modulating brain function by influencing cortical excitability1 and allowing for the improvement of
59
motor performance through facilitation of adaptive brain plasticity.5 The effect of this stimulation
60
on underlying neural tissue depends particularly on the frequency at which it is administered. Lower
61
frequencies, generally those below 1 Hz, suppress motor cortex excitability and thus have an
62
inhibitory effect on the stimulated cortex; conversely, higher frequencies lead to facilitation by
AC C
EP
54
ACCEPTED MANUSCRIPT 5
increasing excitability.9 The results of rTMS may include changes in cortical or subcortical
64
structures that are functionally connected to the site of direct stimulation, such that its effects are
65
not limited to one region.7
66
RI PT
63
The rationale for using rTMS is based upon the plastic effects that it can have on disrupted physiological mechanisms. These disruptions include impaired intracortical inhibition and
68
abnormally increased transcallosal inhibition from the healthy to the lesioned hemisphere.5 In an
69
effort to limit the extent of the loss of function caused by a stroke, the non- or less-affected
70
hemisphere may initially have a beneficial effect on the activity of the affected hemisphere;
71
however, this can subsequently become detrimental and interfere with the recovery process.10
72
Interventional approaches with rTMS are targeted at normalizing the interhemispheric imbalance
73
between the affected and unaffected hemispheres. The therapeutic strategy involves delivering
74
excitatory rTMS on the motor cortex of the lesioned hemisphere to upregulate cortical excitability
75
in this area, thus increasing intracortical facilitation. Additionally, inhibitory stimulation may be
76
applied to the contralesional hemisphere, thereby down-regulating the excitability it has over the
77
ipsilesional motor cortex.5 The modulation of cortical excitability through rTMS may facilitate
78
neuroplastic changes that potentially reestablish disrupted transcallosal inhibitory pathways
79
between the primary motor areas of both hemispheres, thus leading to a restoration of
80
interhemispheric balance and improved motor function.1 Therefore, rTMS serves as a promising
81
complementary treatment used alongside traditional therapies to enhance motor neurorehabilitation
82
in stroke patients. Given that this therapy is relatively novel, few studies have evaluated its
83
effectiveness in improving spasticity. It was our objective to perform a systematic review and meta-
84
analysis to evaluate the effectiveness of rTMS in improving upper extremity spasticity after stroke.
85
METHODS
86
AC C
EP
TE D
M AN U
SC
67
Literature Search
ACCEPTED MANUSCRIPT 6
A literature search of multiple databases (i.e., Pubmed, CINAHL, Scopus, Cochrane, and
87
EMBASE) was conducted for articles published between January 1980 and April 2015 in English.
89
Selected keywords included: repetitive transcranial magnetic stimulation, rTMS, stroke,
90
cerebrovascular accident, spasticity, upper limb, limb disorders, upper motor neuron syndrome,
91
and movement disorders. Variations of keywords were individualized for each scientific database.
92
The references of all retrieved articles were reviewed to ensure all relevant articles were included
93
for data synthesis.
94
Study Selection
SC
RI PT
88
M AN U
Studies were included if they satisfied the following four a priori inclusion criteria:
95 96
1.
the population included was >50% stroke patients;
97
2.
the sample size included four or more subjects;
98
3.
the intervention applied was rTMS; and
99
4.
upper extremity spasticity was assessed pre and post intervention
To determine if the study specifically applied rTMS, a physician (AB) trained in the
TE D
100
administration of this therapy reviewed each article individually to verify that the methods were
102
consistent with typical application. There were no specified criteria in terms of the intensity or
103
duration of therapy. Studies assessing concomitant therapies (e.g., botulinum toxin,
104
pharmacological drugs) were included for review. Only studies providing consistent rTMS
105
protocols were included in the meta-analysis; studies providing concomitant therapy (i.e., botulinum
106
toxin type A) were excluded from meta-analysis but still included for descriptive review. Studies
107
were excluded if information on patient demographics, research design, intervention and/or results
108
could not accurately be extracted from the article. Additionally, studies that assessed the effect of
109
rTMS priming were excluded.
110
Study Appraisal
AC C
EP
101
ACCEPTED MANUSCRIPT 7
Two independent reviewers (AMc; ST) assessed randomized controlled trials (RCTs) for
111
methodological quality using the Physiotherapy Evidence Database (PEDro) scoring system .11 The
113
tool assesses 11 items on study quality that are answered using a ‘yes’ (score=1) or ‘no’ (score=0).
114
The first item is a measure of external validity and is not used in calculating the final score; thus,
115
the maximum score that can be achieved is 10. All research designs were given a level of evidence
116
according to a modified Sackett Scale (Table 1).12 Insert Table 1 about here.
118
SC
117
RI PT
112
Data Synthesis
Data extracted from the studies included author(s), year, country of origin, subject and
M AN U
119
treatment characteristics (e.g., age, gender, stroke onset), study design, intervention/control
121
protocol, outcome measure pre-treatment and post-treatment scores, and adverse effects. The mean
122
and standard deviation (median and interquartile range, as necessary) at baseline and follow-up was
123
extracted for all available spasticity outcome measures. To approximate an effect size, medians
124
were considered to be equal to means when the interquartile range was deemed to represent a
125
normal distribution; to obtain associated standard deviations, interquartile ranges were divided by
126
1.35. Four meta-analyses were completed using Comprehensive Meta-Analysis software (version 2,
127
Biostat Inc., Englewood, NJ, 2005). Pooled analyses were conducted using a fixed or random
128
effects model when there was little or great heterogeneity, respectively.
EP
To examine individuals’ improvement in spasticity of the upper extremity, the modified
AC C
129
TE D
120
130
Ashworth scale (MAS) was assessed. Using the baseline (pre-treatment) and follow-up (post-
131
treatment) means with standard deviations, a pooled mean difference (plus standard error and 95%
132
confidence interval, CI) between baseline and post-intervention was calculated. If a pre- or post-
133
test mean did not have an accompanying standard deviation, a p value or Cohen’s d value was used
134
instead.
135
Outcome Measure
ACCEPTED MANUSCRIPT 8
136
The MAS was designed to assess muscle tone and evaluate an individual’s level of spasticity. The MAS uses a six-point classification scale ranging from 0 to 4 where 0 indicates no
138
increase in muscle tone, 1 indicates a slight increase in muscle tone, manifested by a catch and
139
release or by minimal resistance at the end of the range of motion when the affected part(s) is/are
140
moved in flexion or extension, 1+ indicates a slight increase in muscle tone, manifested by a catch,
141
followed by minimal resistance throughout the remainder (less than half) of the ROM, 2 indicates
142
more marked increase in muscle tone through most of the ROM, but affected part(s) easily moved,
143
3 indicates considerable increase in muscle tone, passive movement difficult, and 4 indicates that
144
the affected part(s) is(are) rigid in flexion or extension.13 The MAS has been validated as an
145
assessment tool for the evaluation of spasticity post stroke.14 Results from other outcome measures
146
were reported in the included studies, such as those that evaluated motor functioning. However,
147
since the objective of this study was to evaluate change in spasticity, these measures were not
148
considered when interpreting findings.
RESULTS
151
TE D
149 150
M AN U
SC
RI PT
137
Study Selection and Quality
For this systematic review and meta-analysis, the literature search yielded 350 studies; the selection process is presented in Figure 1. In total, ten studies published from 2008 to 2015 met
153
inclusion criteria (Figure 1) including one RCT,4 one cross-over RCT,15 and eight pre-post
154
studies.16,17-23 The methodological quality for both RCTs was excellent, with scores ranging 8-9 on
155
the PEDro scale, and providing level 1a evidence. The eight pre-post studies were rated as level 4
156
evidence (Table 2).
158 159 160
AC C
157
EP
152
Insert Figure 1 and Table 2 about here.
Patient Characteristics Table 2 presents patient characteristics for each of the studies involved. A total of 273 poststroke (hemorrhagic=123, ischemic=150) participants were included in all ten studies; however,
ACCEPTED MANUSCRIPT 9
sample sizes ranged considerably from 5 to 80 participants. The majority of participants were male
162
(66.0%) with a mean age ranging 55.0-64.6 years; Yamada et al.16 did not report on participants’
163
gender. While Mally and Dinya17 and Yamada et al.22 did not report which side their participants’
164
hemisphere was affected, among the other studies, hemispheres were affected on the left and right
165
sides in 58 and 60 participants, respectively. The mean stroke duration ranged 6 months - 10.0
166
years.
167
Study Design
SC
168
RI PT
161
A description of the study design and methodology for each article is reported in Table 3. The RCTs by Barros-Galvao et al.4 and Etoh et al.15 compared an intervention group (receiving
170
rTMS) to a sham control group. In both studies, individuals receiving active treatment received 1
171
Hz to the unaffected hemisphere for ten sessions; however, Barros-Galvao et al.4 delivered rTMS at
172
1,500 pulses whereas Etoh et al.15 delivered rTMS at 240 pulses. Both studies provided their active
173
and control participants with physiotherapy and/or occupational therapy (0.5-2 hours).
M AN U
169
The eight pre-post studies followed similar treatment regimens whereby rTMS was
175
delivered to all enrolled participants. Four of these studies were led by the author Kakuda18-20,23 with
176
almost identical treatment protocols. In all four studies, 22 sessions of motor rTMS (1,200 pulses, 1
177
Hz) as well as 1 hour of occupational therapy and 1 hour of self-training were applied to the
178
unaffected hemisphere. Two of their studies provided concomitant therapy to participants. Kakuda
179
et al.19 administered 100 mg daily Levadopa to participants for four weeks pre rTMS until four
180
weeks post rTMS. Kakuda et al.18 injected botulinum toxin type A (maximum 240 U) into muscles
181
of the affected limb four weeks before rTMS. Two pre-post studies were led by Yamada.16,22
182
Yamada et al.22 applied contralesional rTMS (1 Hz, 2,400 pulses), occupational therapy, self-
183
training, and botulinum toxin type A to individuals over 15 days. The study by Yamada et al.16 was
184
the only one to apply rTMS to both the contralesional (1 Hz) and ipsilesional (10 Hz) hemispheres
185
in all patients, in addition to occupational therapy and self-training over 15 days. The remaining
AC C
EP
TE D
174
ACCEPTED MANUSCRIPT 10
pre-post study by Mally and Dinya17 applied motor rTMS (100 pulses; 1 Hz) to the unaffected
187
and/or affected hemisphere for 14 sessions; stimulation sites were determined by testing the effect
188
of one pulse rTMS to the affected/unaffected hemisphere and visualizing for movement in the
189
paretic side prior to treatment.
RI PT
186
The site of rTMS application was the location in which the largest motor-evoked potential
191
could be elicited in the first dorsal interosseous muscle of the unaffected upper limb. In all studies,
192
with the exception of that by Mally and Dinya,17 the intensity of stimulation was set at 90% of
193
motor threshold.
SC
190
Insert Table 3 about here.
195
M AN U
194 Adverse Effects
Eight studies reported that there were no adverse effects as a result of the rTMS
196
intervention.4,15,16,18,20-23 The remaining two studies did not report any information on whether
198
participants experienced adverse effects post intervention.17,19
199
Effectiveness
200
Elbow
TE D
197
The results from each study are reported in Table 3. With the exception of Mally and
201
Dinya17 who assessed upper and lower extremity spasticity, the remaining studies examined only
203
spasticity in the elbow, wrist, and fingers. Four studies15,16,18,20 assessed change in elbow spasticity
204
using the MAS; three of the studies16,18,20 reported significant improvement post intervention and
205
this improvement was maintained at 4-week follow-up.18 However, the cross-over RCT by Etoh et
206
al.15 found no significant improvement in elbow spasticity post intervention or at 4-week follow-up.
207
For the meta-analysis, a fixed effects model was used due to little heterogeneity between the two
208
studies included (Q = 1.75; I2 = 42.9; df = 1). The standard difference in means was significant from
209
baseline to post intervention demonstrating a treatment effect on wrist MAS scores (p=.032; Figure
210
2).
AC C
EP
202
ACCEPTED MANUSCRIPT 11
Insert Figure 2 about here.
211 212
Wrist Nine studies4,15,16,18-23 assessed change in wrist spasticity using the MAS. Etoh et al.15 and
214
Kakuda et al.19 reported no improvement in wrist spasticity post intervention or at 4-week follow-
215
up. However, the remaining seven studies reported significant improvement post intervention.4,16,18-
216
22
217
al.22 reported that while both of their treatment groups improved, there were no significant between-
218
group differences in improvement (rTMS only vs. rTMS + botulinum toxin type A). Six studies4,20-
219
23
220
meta-analysis, a fixed effects model was used due to little heterogeneity between the studies
221
included (Q =11.206; I2 = 55.4; df = 5). The standard difference in means was significant from
222
baseline to post intervention demonstrating a treatment effect on wrist MAS scores (p<.001; Figure
223
3). To examine the contribution of only the two RCTs, a separate meta-analysis was conducted
224
comparing the treatment to sham; a fixed effects model (Q = 1.021; I2 = 2.1; df = 1) demonstrated
225
that the standard difference in means was not significant from baseline to post intervention (p=.300;
226
Figure 4).
RI PT
213
SC
This effect was maintained at four-week follow-up in two studies.18,19 Interestingly, Yamada et
Eight studies15,16,18-23 examined change in finger spasticity using the MAS. Six studies
AC C
229
Fingers
EP
Insert Figure 3 and 4 about here.
227 228
TE D
M AN U
were included in the meta-analysis of rTMS on wrist spasticity as measured by the MAS. For the
230
reported improvement post intervention16,18-22 and three studies at 4-week follow-up.15,18,19 Five
231
studies20-23 were included in the meta-analysis of rTMS on finger spasticity as measured by the
232
MAS. A fixed effects model was used due to little heterogeneity between the studies included for
233
this analysis (Q = 3.392; I2 = 0; df = 4). The standard difference in means was significant from
234
baseline to post intervention demonstrating a treatment effect on finger MAS scores (p<.001;
235
Figure 5).
ACCEPTED MANUSCRIPT 12
236
Mally and Dinya17 measured change in upper extremity spasticity (predominately fingers) using an outcome measure created by the authors (i.e., “Score of Spasticity”) whereby the rating
238
system was similar to that of the MAS but rated on a scale of 0-3 instead of 0-4. The authors
239
reported significant improvement in upper extremity spasticity post intervention, and at 1- and 3-
240
month follow-up for all groups (except for group D at 3 month follow-up; see Table 3). Insert Figure 5 about here.
242
DISCUSSION
SC
241
RI PT
237
This systematic review and meta-analysis aimed to evaluate the effectiveness of rTMS in
244
improving spasticity post stroke. In an assessment of its effect on elbow flexors, three level 4 pre-
245
post studies demonstrated improvement post intervention but this was different than the one level 1a
246
cross-over RCT reporting no improvement; however, the meta-analysis demonstrated a significant
247
treatment effect immediately post intervention (p<.001). Similarly, while one level 4 study reported
248
improvement at 4-week follow-up, the level 1a cross-over RCT found no improvement. Regarding
249
wrist flexors, the results from the first meta-analysis (four level 4 studies and one level 1a RCT)
250
found a significant improvement post intervention; however, the meta-analysis of only the two
251
RCTs did not show significant improvement. Non-meta-analytic data (two level 4 studies)
252
supported significant improvement at 4-week follow-up as well. For finger flexors, the second
253
meta-analysis (four level 4 studies) found a significant improvement in spasticity post intervention;
254
this improvement was shown to be maintained at 4-week follow-up by two level 4 studies and one
255
level 1a cross-over RCT.
TE D
EP
AC C
256
M AN U
243
Spasticity is a complex disorder, encompassing a considerably diverse variety of symptoms,
257
which can have a negative impact on the quality of life of stroke survivors.24 Available therapies to
258
date include physiotherapy modalities as well as pharmacologic agents such as botulinum toxin.
259
While the latter has demonstrated effectiveness in a post-stroke population,25 additional, non-
260
pharmacological novel approaches are necessary since no one treatment appears to be completely
ACCEPTED MANUSCRIPT 13
effective for all patients. The use of rTMS as a therapeutic tool in stroke recovery is relatively new,
262
and has been reported to be of benefit in the rehabilitation of post-stroke aphasia and
263
hemineglect.10,26 Within the confines of the included studies’ limitations, this systematic review and
264
meta-analysis revealed that rTMS may be beneficial in improving spasticity post stroke.
265
RI PT
261
The limitations of included studies are similar to those reported by other systematic reviews and meta-analyses published on rTMS. From a methodological standpoint, the primary limitation
267
among eight studies was the use of a pre-post research design. Since the majority of studies were
268
uncontrolled, the authors’ conclusions are subject to some bias and open to interpretation. In the
269
current review, two studies18,22 employed rTMS concomitantly with either botulinum toxin type A18
270
or Levadopa19 without a control group. This is especially significant as botulinum toxin type A has
271
been shown to significantly and independently improve spasticity as measured using the MAS.25
272
Therefore, the positive findings reported by Kakuda et al.18 and Yamada et al.22 may have resulted
273
from the botulinum toxin type A, the rTMS or both. A complex, rigorous research design such as a
274
multiple-armed RCT can overcome this problem by separating out the individual as well as additive
275
effects of combination therapies. Regardless, it is important to note that significant treatment effects
276
were demonstrated by the meta-analyses which excluded groups receiving concomitant therapies.
M AN U
TE D
From a rehabilitation perspective, increased intensity and frequency of rehabilitation
EP
277
SC
266
sessions has been shown to positively benefit motor outcomes.27 The underlying assumption is that
279
to induce long-term effects, highly intense sessions that are frequent and of long duration are
280
necessary. From a neuromodulation perspective, this viewpoint holds true as well. To induce
281
neuroplastic effects, high pulse counts and a long duration are required. On average, the total
282
number of minutes of rTMS applied to participants in these studies ranged 400-480 minutes over
283
two weeks and the pulse count ranged 100-2,400 per session. Since rehabilitation is a long-term
284
process often lasting months to years, these diverse rTMS protocols may be suboptimal in their
AC C
278
ACCEPTED MANUSCRIPT 14
285
current form. Future trials should consider evaluating the dose-response relationship between rTMS
286
and improvement in spasticity. Considering the mechanism by which rTMS is hypothesized to be effective, all but one
RI PT
287
study included for review utilized rTMS at 1 Hz to inhibit the contralesional hemisphere. An rTMS
289
frequency of 1 Hz on the contralesional side is likely to help restore interhemispheric
290
excitation/inhibition balance thereby facilitating motor recovery.26 The re-balance of inter-
291
hemispheric cortical excitability and its impact on spasticity is unknown. Excitability-increasing
292
high frequency rTMS of ipsilesional M1 or excitability-decreasing low frequency rTMS of
293
contralesional M1 is thought to improve motor abilities in stroke patients.26,28 The positive clinical
294
signs of spasticity include enhanced stretch reflex, increased tone, and exaggerated tendon reflexes,
295
whereas the negative signs include loss of dexterity and slowness of movement and control of
296
muscles and limb segments. The negative symptoms in the UMN syndrome (i.e., weakness) are due
297
to disruption of the pathways between the supraspinal and the motor subsystems. The influence of
298
rTMS may improve the hemispheric balance of excitation and inhibition, thereby reducing tone, as
299
measured by the Ashworth scale.
TE D
M AN U
SC
288
Conceptually, a potentially more efficacious protocol for future studies may involve
301
providing bi-hemispheric stimulation to patients whereby the contralesional hemisphere is inhibited
302
via 1 Hz and the ipsilesional hemisphere is activated via HF TMS 10 Hz. This type of protocol was
303
tested in just one study16 on post-stroke spasticity. Future studies should consider designing RCTs
304
which compare uni- versus bi-hemispheric stimulation. Typical rTMS has been reported to be weak
305
and have only short-term effects lasting 30 minutes. Another evolving neuroplastic paradigm
306
involves theta-burst stimulation (TBS) which is a method for rapid, controllable, consistent, and
307
long-lasting stimulation.29 A pilot study by Huang and Rothwell30 demonstrated 50 Hz TBS to be
308
safe and effective in targeting neurons in the motor cortex. Additional studies should be conducted
309
which examine TBS for spasticity post stroke.
AC C
EP
300
ACCEPTED MANUSCRIPT 15
310
The current study is not without its own limitations. It is confined by its inclusion criteria which excluded non-English trials. Therefore, studies published in alternate languages were not
312
included in the analysis.
RI PT
311
CONCLUSION
314
There is limited available evidence to date to support the use of rTMS in improving
315
spasticity post stroke. Despite the positive findings reported by several studies, they are not without
316
their own significant methodological limitations. The effectiveness of rTMS for use in stroke
317
rehabilitation should be verified further; thus, additional better powered, high-quality RCTs are
318
needed. Furthermore, its use in combination with other tone-reducing modalities, such as botulinum
319
toxin, requires exploration.
AC C
EP
TE D
M AN U
SC
313
ACCEPTED MANUSCRIPT 16
REFERENCES
320 1.
2.
3.
Thibaut A, Chatelle C, Ziegler E, Bruno MA, Laureys S, Gosseries O. Spasticity after stroke: physiology, assessment and treatment. Brain Inj 2013;27(10):1093-105.
326 327
Pandyan AD, Gregoric M, Barnes MP, et al. Spasticity: clinical perceptions, neurological realities and meaningful measurement. Dis Rehab 2005;27(1-2):2-6.
324 325
RI PT
after stroke: a focused review. Amer J Phys Med Rehab 2012;91(3):254-70.
322 323
Corti M, Patten C, Triggs W. Repetitive transcranial magnetic stimulation of motor cortex
4.
SC
321
Barros Galvao SC, Borba Costa dos Santos R, Borba dos Santos P, Cabral ME, MonteSilva K. Efficacy of coupling repetitive transcranial magnetic stimulation and physical
329
therapy to reduce upper-limb spasticity in patients with stroke: a randomized controlled
330
trial. Arch Phys Med Rehab 2014;95(2):222-9. 5.
transcranial magnetic stimulation. Restor Neurol Neurosci 2011;29(6):395-409. 6.
neurorehabilitation after stroke? Lancet Neurol 2006;5(8):708-712.
334 7.
stimulation (rTMS). Neurophysiol Clin 2006;36(3):105-15.
336 337
8.
9.
10.
344
Hao Z, Wang D, Zeng Y, Liu M. Repetitive transcranial magnetic stimulation for
improving function after stroke. Cochrane 2013;5:CD008862.
342 343
Kobayashi M, Pascual-Leone A. Transcranial magnetic stimulation in neurology. Lancet
Neurol 2003;2(3):145-56.
340 341
Talelli P, Rothwell J. Does brain stimulation after stroke have a future? Curr Opin Neurol 2006;19(6):543-50.
338 339
Lefaucheur JP. Stroke recovery can be enhanced by using repetitive transcranial magnetic
EP
335
Hummel FC, Cohen LG. Non-invasive brain stimulation: a new strategy to improve
TE D
332 333
Hoyer EH, Celnik PA. Understanding and enhancing motor recovery after stroke using
AC C
331
M AN U
328
11.
Moseley AM, Herbert RD, Sherrington C, Maher CG. Evidence for physiotherapy practice: a survey of the Physiotherapy Evidence Database (PEDro). Aus J Physio 2002;48(1):43-9.
ACCEPTED MANUSCRIPT 17
12.
teach EBM. Toronto, Ontario: Elsevier Churchill Livingstone; 2005.
346 347
13.
Bohannon RW, Smith MB. Interrater reliability of a modified Ashworth scale of muscle spasticity. Phys Ther 1987;67(2):206-7.
348 349
Straus SE RW, Glasziou P, Haynes RB. Evidencebased medicine: How to practice and
14.
RI PT
345
Blackburn M, van Vliet P, Mockett SP. Reliability of measurements obtained with the modified Ashworth scale in the lower extremities of people with stroke. Phys Ther
351
2002;82(1):25-34.
352
15.
SC
350
Etoh S, Noma T, Ikeda K, et al. Effects of repetitive trascranial magnetic stimulation on repetitive facilitation exercises of the hemiplegic hand in chronic stroke patients. J Rehab
354
Med 2013;45(9):843-7.
355
16.
M AN U
353
Yamada N, Kakuda W, Kondo T, Shimizu M, Mitani S, Abo M. Bihemispheric repetitive
356
transcranial magnetic stimulation combined with intensive occupational therapy for upper
357
limb hemiparesis after stroke: a preliminary study. Inter J Rehab Res 2013;36(4):323-9. 17.
transcranial magnetic stimulation (rTMS). Brain Res Bull 2008;76(4):388-95.
359 360
Mally J, Dinya E. Recovery of motor disability and spasticity in post-stroke after repetitive
TE D
358
18.
Kakuda W, Abo M, Momosaki R, et al. Combined therapeutic application of botulinum toxin type A, low-frequency rTMS, and intensive occupational therapy for post-stroke
362
spastic upper limb hemiparesis. Eur J Phys Rehab Med 2012;48(1):47-55. 19.
occupational therapy with levodopa administration: an intensive neurorehabilitative
364
approach for upper limb hemiparesis after stroke. Int J Neurosci 2011b;121(7):373-8.
365 366
Kakuda W, Abo M, Kobayashi K, et al. Combination treatment of low-frequency rTMS and
AC C
363
EP
361
20.
Kakuda W, Abo M, Kobayashi K, et al. Low-frequency repetitive transcranial magnetic
367
stimulation and intensive occupational therapy for poststroke patients with upper limb
368
hemiparesis: preliminary study of a 15-day protocol. Int J Rehab Res 2010;33(4):339-45.
ACCEPTED MANUSCRIPT 18
369
21.
Kondo T, Kakuda W, Yamada N, Shimizu M, Abo M. Effects of repetitive transcranial magnetic stimulation and intensive occupational therapy on motor neuron excitability in
371
poststroke hemiparetic patients: a neurophysiological investigation using F-wave
372
parameters. Int J Neurosci 2015;125(1):25-31.
373
22.
RI PT
370
Yamada N, Kakuda W, Kondo T, Mitani S, Shimizu M, Abo M. Local muscle injection of botulinum toxin type a synergistically improves the beneficial effects of repetitive
375
transcranial magnetic stimulation and intensive occupational therapy in post-stroke patients
376
with spastic upper limb hemiparesis. Eur Neurol 2014;72(5-6):290-8. 23.
Kakuda W, Abo M, Kobayashi K, et al. Anti-spastic effect of low-frequency rTMS applied
M AN U
377
SC
374
378
with occupational therapy in post-stroke patients with upper limb hemiparesis. Brain Inj
379
2011b;25(5):496-502. 24.
Eur J Neurol 2012;19(1):21-7.
381 382
Ward AB. A literature review of the pathophysiology and onset of post-stroke spasticity.
25.
Dashtipour K, Chen JJ, Walker HW, Lee MY. Systematic literature review of
TE D
380
383
abobotulinumtoxinA in clinical trials for adult upper limb spasticity. Amer J Phys Med
384
Rehab 2015;94(3):229-38. 26.
Lefaucheur JP, André-Obadia N, Antal A, et al. Evidence-based guidelines on the
EP
385
therapeutic use of repetitive transcranial magnetic stimulation (rTMS). Clin
387
Neurphysiol2014;125(11):2150-206.
388
27.
Phys Ther 2015;39(2):95-102.
390
392
Holleran CL, Rodriguez KS, Echauz A, Leech KA, Hornby TG. Potential contributions of
training intensity on locomotor performance in individuals with chronic stroke. J Neurol
389
391
AC C
386
28.
Rossini PM, Burke D, Chen R, et al. Non-invasive electrical and magnetic stimulation of the brain, spinal cord, roots and peripheral nerves: Basic principles and procedures for
ACCEPTED MANUSCRIPT 19
393
routine clinical and research application. J Int Fed Clin Neurophysiol 2015;126(6):1071-
394
107. 29.
the Human Motor Cortex. Neuron 2005;45(2):201-6.
396 397
Huang Y-Z, Edwards MJ, Rounis E, Bhatia KP, Rothwell JC. Theta Burst Stimulation of
30.
RI PT
395
Huang YZ, Rothwell JC. The effect of short-duration bursts of high-frequency, low-
intensity transcranial magnetic stimulation on the human motor cortex. Clin Neurophysiol
399
2004;115(5):1069-75.
AC C
EP
TE D
M AN U
SC
398
ACCEPTED MANUSCRIPT 20
400
Table 1: Modified Sackett Scale
401 Description
1a
More than one randomized controlled trial (PEDro score ≥6)
1b
One randomized controlled trial (PEDro score ≥6)
2
Randomized controlled trial (PEDro score <6), prospective controlled trials, cohort studies
3
Case-control
4
Case series, pre-post or post-test
5
Observational, case report or clinical consensus
Conflicting
In the absence of evidence, agreement by a group of experts on the appropriate treatment
M AN U
SC
RI PT
Level
course
AC C
EP
TE D
402
ACCEPTED MANUSCRIPT
21
Research Country
Year Kondo et al.
Evidence
Age
N
Males Females Design
(PEDro)
Japan
10
Pre-Post
Level 4
8
Brazil
20
RCT
Level 1a (9)
13
2
Barros Galvao et al. 2014 Yamada et al.
Japan
80
Pre-Post
Level 4
50
Yamada et al.
Japan
8
Pre-Post
Level 4
Japan
18
RCT cross-
Level 1a (8)
2013 Kakuda et al.
over Japan
14
Pre-Post
Japan
39
Pre-Post
2012 Kakuda et al. 2011a
Left
mean ± SD (mo)
Hem
57.4 ± 8.1
86.0 ± 52.8
7
3
3
7
E: 57.4 ± 12.0
E: 47.8 ± 43.2
3
17
10
10
C: 64.6 ± 6.8
C: 58.9 ± 27.2
E: 62.9 ± 10.2
E: 62.0 ± 51.7
41
39
N/R
N/R
C: 57.2 ± 15.2
C: 48.0 ± 29.8
Isch
Right
N/R
62.8 ± 4.9
84.3 ± 87.2
4
4
5
3
14
4
59.7 ± 11.0
29.9 ± 18.8
5
13
8
10
Level 4
10
4
54.9 ± 9.2
87.1 ± 48.2
9
5
5
6
Level 4
30
9
56.5 ± 16.0
50.3 ± 37.8
23
16
23
16
AC C
Etoh et al.
Affected Side
N/R
EP
2013
30
TE D
2014
7
Stroke Type
mean ± SD (yr)
M AN U
2015
Stroke Onset
SC
Author,
RI PT
Table 2. Study Subject Characteristics
ACCEPTED MANUSCRIPT
22
Kakuda et al.
Japan
5
Pre-Post
Level 4
3
2
61.0 ± 4.5
Japan
15
Pre-Post
Level 4
10
5
55.0 ± 17.0
Hungary
64
Pre-Post
Level 4
37
27
57.6 ± 10.8
64.0 ± 57.0
4
1
3
2
57.0 ± 55.0
9
6
9
6
120.0 ± 76.8
18
46
N/R
N/R
Kakuda et al.
Mally &
SC
2010
RI PT
2011b
Dinya 2008
M AN U
Note: C=control group; E=experimental group; Hem=hemorrhagic; Isch=ischemic; Mo=month; N=number; N/R=not reported; NS=non-stroke;
AC C
EP
TE D
PEDro=physiotherapy evidence database; S=stroke; SD=standard deviation; Yr=year
ACCEPTED MANUSCRIPT
23
Table 3. Study protocol, outcome measures and results for all included studies Motor rTMS Intervention Protocol Control
Pulses; No. Frequency
Intensity Sessions
Kondo et
Contralesional
1 Hz
al. 2015
Contralesional
1 Hz
Galvao et
Rehab/Session
90%
2,400; 12, 40 min
120 min OT +
RMT
sessions over 15
120 min self
days Barros
Outcome
90%
1,500; 10 sessions
RMT
at 3/wk
None
30 min PT
Sham
1 Hz
90%
2,400; 12 , 40 min
RMT
MAS
Elbow
Post rTMS
MAS
Wrist Fingers +
Elbow
+
Post sham
--
Post BGD
+
4-week rTMS
+
4-week sham
--
4-week BGD
--
rTMS
sessions over 15
120 min self +
without
days
botulinum toxin
botulinum
type A
toxin type
Post rTMS only
A
Post BGD
MAS Post rTMS/ botulinum toxin type A
Elbow
+
Wrist Fingers
Post rTMS
120 min OT +
AC C
al. 2014
Contralesional
EP
TE D
al. 2014
Yamada et
Results
Protocol
SC
Target
M AN U
Year
Additional
RI PT
Author,
Wrist Fingers ++
++
++
++
--
+
ACCEPTED MANUSCRIPT
24
Bihemispheric
al. 2013
1 Hz contra
90%
2,000 each; 10, 40
120 min OT +
10 Hz ips
RMT
min sessions over
120 min self
2013
Kakuda et
Contralesional
1 Hz
al. 2012
90%
240; 10, 40 min
60-120 min
RMT
sessions over 14
voluntary
days
PT/OT
90%
1,200; 22, 20 min
RMT
sessions over 15 days
Sham
SC
1 Hz
M AN U
Contralesional
60 min OT +
None
60 min self +
botulinum toxin
Contralesional
1 Hz
al. 2011a
TE D
type A
Kakuda et
90%
1,200; 22, 20 min
60 min OT +
RMT
sessions over 15
60 min self +
None
1 Hz
90%
1,200; 22, 20 min
AC C
al. 2011b
Contralesional
EP
days
Kakuda et
RMT
60 min OT +
MAS Post rTMS
15 days Etoh et al.
None
RI PT
Yamada et
None
Elbow
Wrist Fingers
+
+
+
Elbow Wrist
Fingers
Post rTMS
--
--
--
Post sham
--
--
--
4-week BGD
--
--
+
Elbow
Wrist
Fingers
++
++
++
+
+
+
Elbow
Wrist Fingers
MAS
MAS Post rTMS/ botulinum toxin type A 4-week FU MAS Post rTMS
++
++
4-week FU
+
+
MAS
Elbow
Wrist Fingers
sessions over 15
60 min self +
Post rTMS/Levadopa
--
--
days
Levadopa
4-week FU
--
--
ACCEPTED MANUSCRIPT
25
Contralesional
1 Hz
al. 2010
90%
1,200; 22, 20 min
60 min OT +
RMT
sessions over 15
60 min self
days Mally &
Bihemispheric
30%
100; 2 sessions per
2.3T
day for 1 wk
None
MAS Post rTMS
None
M AN U
SC
Dinya 2008
1 Hz
None
RI PT
Kakuda et
a
Spasticity
Elbow
Wrist
Fingers
++
+
+
b
A
B
C
D
Post rTMS
++ ++ ++ +
1 month
++ ++ ++ +
3 month
++ ++ ++ - -
Note. + = statistically significant improvement at p<.05; ++ = statistically significant improvement at p<.01; - - = No significant difference at p>.05 a
Score of Spasticity – outcome created by the authors whereby 0=none, 1=slight, 2=fingers in flexion and passive extension to elbow difficult, 3=expressive flexion
b
A, B, C, D, refer to groups A) movement produced when stimulation was applied to both sides simultaneously, B) stimulation of either side produced no movements, C)
TE D
stimulation of contralateral hemisphere induced movement, and D) stimulation of ipsilateral hemisphere induced movement.
AC C
EP
BGD = Between-Group Difference; Contra = FU = Follow-up; OT = occupational therapy; PT = physical therapy; RMT = resting motor threshold
ACCEPTED MANUSCRIPT
26
Figure Captions
RI PT
Figure 1. Study Selection Process Figure 2. Meta-analysis of two uncontrolled studies assessing within-group improvements in MAS of the elbow flexors from baseline to post intervention
SC
Figure 3. Meta-analysis of five uncontrolled studies assessing within-group improvements in MAS of the wrist flexors from baseline to post
M AN U
intervention
Figure 4. Meta-analysis of two RCTs assessing between-group improvements in MAS of the wrist flexors from baseline to post intervention Figure 5. Meta-analysis of four uncontrolled studies assessing within-group improvements in MAS of the finger flexors from baseline to
AC C
EP
TE D
post intervention
ACCEPTED MANUSCRIPT
RI PT
Records identified through database searching (n = 350)
Additional records identified through other sources (n = 0)
SC
Identification
PRISMA 2009 Flow Diagram
M AN U
Records screened (n = 341)
Records excluded (n = 328)
Full-text articles assessed for eligibility (n = 13)
Full-text articles excluded, based on not satisfying inclusion criteria (n = 3)
TE D EP AC C
Included
Eligibility
Screening
Records after duplicates removed (n = 341)
Studies included in qualitative synthesis (n = 10)
Studies included in quantitative synthesis (meta-analysis) (n = 10)
From: Moher D, Liberati A, Tetzlaff J, Altman DG, The PRISMA Group (2009). Preferred Reporting Items for Systematic Reviews and MetaAnalyses: The PRISMA Statement. PLoS Med 6(6): e1000097. doi:10.1371/journal.pmed1000097
For more information, visit www.prisma-statement.org.
Study name
Statistics for each study Std diff in means
Standard error
Variance
Lower limit
Upper limit
RI PT
ACCEPTED MANUSCRIPT
Std diff in means and 95% CI
Z-Value
p-Value
0.000
0.471
0.222
-0.924
0.924
0.000
1.000
0.733
0.291
0.085
0.163
1.303
2.520
0.012
0.531
0.247
0.061
0.046
1.016
2.145
SC
Etoh et. al 2013 Kakuda et al. 2010
0.032
AC C
EP
TE D
M AN U
-1.00
-0.50
0.00
0.50
1.00
Statistics for each study Std diff Standard in means error
Variance
Lower limit
Upper limit
Std diff in means and 95% CI
Z-Value p-Value
SC
Study name
0.693
0.352
0.124
0.003
1.384
1.969
0.049
Barros Galvao et al. 2014
1.441
0.451
0.204
0.556
2.326
3.192
0.001
M AN U
Kondo et al. 2015
Yamada et al. 2014
0.849
0.189
0.036
0.478
1.220
4.487
0.000
Kakuda et al. 2011a
0.329
0.164
0.027
0.007
0.651
2.002
0.045
Kakuda et al. 2010
0.667
0.285
0.081
0.107
1.226
2.336
0.020
0.333
0.111
-0.653
0.653
0.000
1.000
0.165
0.027
0.286
0.932
3.701
0.000
EP
TE D
0.000 0.609
AC C
Etoh et al. 2013
RI PT
ACCEPTED MANUSCRIPT
-3.00
-1.50
0.00
1.50
3.00
Statistics for each study Std diff Standard in means error
Variance
Lower limit
Upper limit
Std diff in means and 95% CI
Z-Value p-Value
0.665
0.459
0.211
-0.235
1.566
1.448
Etoh et al. 2013
0.000
0.471
0.222
-0.924
0.924
0.000
0.341
0.329
0.108
-0.304
0.986
1.037
TE D EP
0.148
1.000
M AN U
Barros Galvao et al. 2014
AC C
SC
Study name
RI PT
ACCEPTED MANUSCRIPT
0.300
-1.00
-0.50
0.00
0.50
1.00
Statistics for each study Std diff Standard in means error
Variance
Lower limit
Upper limit
Std diff in means and 95% CI
Z-Value p-Value
SC
Study name
Kondo et al. 2015
0.768
0.360
0.130
0.063
1.474
2.135
Yamada et al. 2014
0.749
0.184
0.034
0.389
1.109
4.080
Kakuda et al. 2011a
0.336
0.165
0.027
0.013
0.658
2.041
0.041
Kakuda et al. 2010
0.465
0.272
0.074 -0.068
0.997
1.710
0.087
Etoh et al. 2013
0.675
0.485
0.235 -0.275
1.625
1.393
0.164
0.540
0.104
0.011
0.744
5.181
0.000
AC C
EP
0.033
0.000
M AN U
TE D
0.336
RI PT
ACCEPTED MANUSCRIPT
-1.00
-0.50
0.00
0.50
1.00
S1934-1482(17)31359-X
DOI:
10.1016/j.pmrj.2017.10.001
Reference:
PMRJ 1999
To appear in:
PM&R
Received Date: 4 April 2016 Revised Date:
8 August 2017
Accepted Date: 9 October 2017
Please cite this article as: McIntyre A, Mirkowski M, Thompson S, Burhan A, Miller T, Teasell R, A Systematic Review and Meta-Analysis on the Use of Repetitive Transcranial Magnetic Stimulation for Spasticity Post Stroke, PM&R (2017), doi: 10.1016/j.pmrj.2017.10.001. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
A Systematic Review and Meta-Analysis on the Use of Repetitive Transcranial Magnetic Stimulation for Spasticity Post Stroke
McIntyre A, 1Mirkowski M, 1Thompson S, 2,3Burhan AM, 1-,3Miller T, 1-3Teasell R
1
Lawson Health Research Institute, Parkwood Institute, London, ON
2
St. Joseph’s Health Care, Parkwood Institute, London, ON
SC
3
RI PT
1
M AN U
Schulich School of Medicine and Dentistry, University of Western Ontario, London, ON
Corresponding author/reprints: Amanda McIntyre RN MSc BSc BScN Aging, Rehabilitation, and Geriatric Care, Rm. B3-123C Parkwood Institute 550 Wellington Road
TE D
London, ON, N6C 0A7
Fax: 519-685-4036 Tel.: 519-685-4292 x41296
EP
[email protected]
Cover Title: rTMS for Spasticity Post Stroke # Figures: 3
AC C
# Tables: 3 # Words: 3,971
Key Words: Stroke, rTMS, Repetitive Transcranial Magnetic Stimulation, Spasticity, Muscle Tone Conflict of Interest: Dr. Tom Miller and Dr. Robert Teasell have received funding from Allergan Canada Inc. to establish a Centre for Excellence in Spasticity at St. Joseph’s Health Care in London, Ontario.
ACCEPTED MANUSCRIPT
1
A Systematic Review and Meta-Analysis on the Use of Repetitive Transcranial Magnetic
2
Stimulation for Upper Extremity Spasticity Post Stroke
4
Cover Title: rTMS for Spasticity Post Stroke
5
# Tables: 3
6
# Words: 3,392
# Figures: 5
8
SC
7
Key Words: Stroke, rTMS, Repetitive Transcranial Magnetic Stimulation, Spasticity, Muscle Tone
Sources of Funding: None Declared.
AC C
EP
TE D
11
M AN U
9 10
RI PT
3
ACCEPTED MANUSCRIPT 2
ABSTRACT
12
Background: Spasticity is a common and potentially debilitating complication that develops after
14
stroke, arising in approximately 30% of patients.
15
Objective: To evaluate the effectiveness of repetitive transcranial magnetic stimulation (rTMS) in
16
improving spasticity after stroke.
17
Design: Meta-analysis and systematic review.
18
Setting: N/A
19
Patients: A total of 273 post-stroke (hemorrhagic=123, ischemic=150) participants were included
20
with sample sizes ranging from 5 to 80. The majority of participants were male (66.0%) with a
21
mean age ranging 55.0-64.6 years. Mean stroke duration ranged 6 months – 10 years.
22
Methods: A literature search of multiple databases was conducted for articles published in English
23
from January 1980 to April 2015 using select keywords. Studies were included if: 1) the population
24
included was >50% stroke patients; 2) the sample size included ≥4 subjects; 3) the intervention
25
applied was rTMS; and 4) upper extremity spasticity was assessed pre and post intervention.
26
Randomized controlled trials (RCTs) were assessed for methodological quality using the
27
Physiotherapy Evidence Database (PEDro) tool. All research designs were given a level of evidence
28
according to a modified Sackett Scale.
29
Main Outcome Measurements: Modified Ashworth Scale (MAS).
30
Results: Ten studies met the inclusion criteria: two RCTs (PEDro scores 8-9) and eight pre-post
31
studies. Meta-analyses of primarily uncontrolled pre-post studies found significant improvements in
32
MAS for elbow (p<.001), wrist (p<.001), and finger flexors (p<.001). However, a meta-analysis of
33
the two available RCTs failed to find a significant rTMS treatment effect on MAS for the wrist
34
(standardized difference=.34, p=.30).
AC C
EP
TE D
M AN U
SC
RI PT
13
ACCEPTED MANUSCRIPT 3
Conclusions: There is limited available evidence to support the use of rTMS in improving spasticity
36
post stroke. Despite the positive findings reported, better powered and appropriately controlled
37
trials are necessary.
AC C
EP
TE D
M AN U
SC
RI PT
35
ACCEPTED MANUSCRIPT 4
INTRODUCTION
39
Stroke is considered to be the leading cause of adult disability, often causing motor
40
impairment and contributing to long-term neurological disability.1 Spasticity and increased tone2 is
41
a common and potentially debilitating complication that develops after stroke, arising in
42
approximately 30% of patients, with a highly variable timing of onset.3 Spasticity is defined as a
43
motor disorder characterized by a velocity dependent increase in tone, and is part of the upper
44
motor neuron (UMN) syndrome. Spasticity management helps improve the quality of life of many
45
stroke survivors.4 Current modalities to decrease tone in the UMN syndrome have been shown to
46
improve function. Current modalities such as medications or botulinum toxin injections are able to
47
decrease tone generally or locally; however, treating spasticity does not necessarily improve
48
weakness or motor control. As recovery of motor function after stroke is usually incomplete, recent
49
research has focused on novel neurorehabilitation techniques to enhance the beneficial effects of
50
treatment aimed at motor recovery.5,6 Among these are various methods of cortical stimulation,
51
which include interventional paired associative stimulation, transcranial direct current stimulation
52
(tDCS), implanted epidural motor cortex stimulation, and repetitive transcranial magnetic
53
stimulation (rTMS).7
TE D
M AN U
SC
RI PT
38
rTMS is a noninvasive therapeutic intervention involving the generation of a strong
55
magnetic field by a powerful electrical coil that passes unimpeded through the skull to the cerebral
56
cortex producing an electric current that painlessly stimulates the targeted brain area.8 A train of
57
pulses is delivered to a particular cortical region at a given intensity, as well as frequency,
58
modulating brain function by influencing cortical excitability1 and allowing for the improvement of
59
motor performance through facilitation of adaptive brain plasticity.5 The effect of this stimulation
60
on underlying neural tissue depends particularly on the frequency at which it is administered. Lower
61
frequencies, generally those below 1 Hz, suppress motor cortex excitability and thus have an
62
inhibitory effect on the stimulated cortex; conversely, higher frequencies lead to facilitation by
AC C
EP
54
ACCEPTED MANUSCRIPT 5
increasing excitability.9 The results of rTMS may include changes in cortical or subcortical
64
structures that are functionally connected to the site of direct stimulation, such that its effects are
65
not limited to one region.7
66
RI PT
63
The rationale for using rTMS is based upon the plastic effects that it can have on disrupted physiological mechanisms. These disruptions include impaired intracortical inhibition and
68
abnormally increased transcallosal inhibition from the healthy to the lesioned hemisphere.5 In an
69
effort to limit the extent of the loss of function caused by a stroke, the non- or less-affected
70
hemisphere may initially have a beneficial effect on the activity of the affected hemisphere;
71
however, this can subsequently become detrimental and interfere with the recovery process.10
72
Interventional approaches with rTMS are targeted at normalizing the interhemispheric imbalance
73
between the affected and unaffected hemispheres. The therapeutic strategy involves delivering
74
excitatory rTMS on the motor cortex of the lesioned hemisphere to upregulate cortical excitability
75
in this area, thus increasing intracortical facilitation. Additionally, inhibitory stimulation may be
76
applied to the contralesional hemisphere, thereby down-regulating the excitability it has over the
77
ipsilesional motor cortex.5 The modulation of cortical excitability through rTMS may facilitate
78
neuroplastic changes that potentially reestablish disrupted transcallosal inhibitory pathways
79
between the primary motor areas of both hemispheres, thus leading to a restoration of
80
interhemispheric balance and improved motor function.1 Therefore, rTMS serves as a promising
81
complementary treatment used alongside traditional therapies to enhance motor neurorehabilitation
82
in stroke patients. Given that this therapy is relatively novel, few studies have evaluated its
83
effectiveness in improving spasticity. It was our objective to perform a systematic review and meta-
84
analysis to evaluate the effectiveness of rTMS in improving upper extremity spasticity after stroke.
85
METHODS
86
AC C
EP
TE D
M AN U
SC
67
Literature Search
ACCEPTED MANUSCRIPT 6
A literature search of multiple databases (i.e., Pubmed, CINAHL, Scopus, Cochrane, and
87
EMBASE) was conducted for articles published between January 1980 and April 2015 in English.
89
Selected keywords included: repetitive transcranial magnetic stimulation, rTMS, stroke,
90
cerebrovascular accident, spasticity, upper limb, limb disorders, upper motor neuron syndrome,
91
and movement disorders. Variations of keywords were individualized for each scientific database.
92
The references of all retrieved articles were reviewed to ensure all relevant articles were included
93
for data synthesis.
94
Study Selection
SC
RI PT
88
M AN U
Studies were included if they satisfied the following four a priori inclusion criteria:
95 96
1.
the population included was >50% stroke patients;
97
2.
the sample size included four or more subjects;
98
3.
the intervention applied was rTMS; and
99
4.
upper extremity spasticity was assessed pre and post intervention
To determine if the study specifically applied rTMS, a physician (AB) trained in the
TE D
100
administration of this therapy reviewed each article individually to verify that the methods were
102
consistent with typical application. There were no specified criteria in terms of the intensity or
103
duration of therapy. Studies assessing concomitant therapies (e.g., botulinum toxin,
104
pharmacological drugs) were included for review. Only studies providing consistent rTMS
105
protocols were included in the meta-analysis; studies providing concomitant therapy (i.e., botulinum
106
toxin type A) were excluded from meta-analysis but still included for descriptive review. Studies
107
were excluded if information on patient demographics, research design, intervention and/or results
108
could not accurately be extracted from the article. Additionally, studies that assessed the effect of
109
rTMS priming were excluded.
110
Study Appraisal
AC C
EP
101
ACCEPTED MANUSCRIPT 7
Two independent reviewers (AMc; ST) assessed randomized controlled trials (RCTs) for
111
methodological quality using the Physiotherapy Evidence Database (PEDro) scoring system .11 The
113
tool assesses 11 items on study quality that are answered using a ‘yes’ (score=1) or ‘no’ (score=0).
114
The first item is a measure of external validity and is not used in calculating the final score; thus,
115
the maximum score that can be achieved is 10. All research designs were given a level of evidence
116
according to a modified Sackett Scale (Table 1).12 Insert Table 1 about here.
118
SC
117
RI PT
112
Data Synthesis
Data extracted from the studies included author(s), year, country of origin, subject and
M AN U
119
treatment characteristics (e.g., age, gender, stroke onset), study design, intervention/control
121
protocol, outcome measure pre-treatment and post-treatment scores, and adverse effects. The mean
122
and standard deviation (median and interquartile range, as necessary) at baseline and follow-up was
123
extracted for all available spasticity outcome measures. To approximate an effect size, medians
124
were considered to be equal to means when the interquartile range was deemed to represent a
125
normal distribution; to obtain associated standard deviations, interquartile ranges were divided by
126
1.35. Four meta-analyses were completed using Comprehensive Meta-Analysis software (version 2,
127
Biostat Inc., Englewood, NJ, 2005). Pooled analyses were conducted using a fixed or random
128
effects model when there was little or great heterogeneity, respectively.
EP
To examine individuals’ improvement in spasticity of the upper extremity, the modified
AC C
129
TE D
120
130
Ashworth scale (MAS) was assessed. Using the baseline (pre-treatment) and follow-up (post-
131
treatment) means with standard deviations, a pooled mean difference (plus standard error and 95%
132
confidence interval, CI) between baseline and post-intervention was calculated. If a pre- or post-
133
test mean did not have an accompanying standard deviation, a p value or Cohen’s d value was used
134
instead.
135
Outcome Measure
ACCEPTED MANUSCRIPT 8
136
The MAS was designed to assess muscle tone and evaluate an individual’s level of spasticity. The MAS uses a six-point classification scale ranging from 0 to 4 where 0 indicates no
138
increase in muscle tone, 1 indicates a slight increase in muscle tone, manifested by a catch and
139
release or by minimal resistance at the end of the range of motion when the affected part(s) is/are
140
moved in flexion or extension, 1+ indicates a slight increase in muscle tone, manifested by a catch,
141
followed by minimal resistance throughout the remainder (less than half) of the ROM, 2 indicates
142
more marked increase in muscle tone through most of the ROM, but affected part(s) easily moved,
143
3 indicates considerable increase in muscle tone, passive movement difficult, and 4 indicates that
144
the affected part(s) is(are) rigid in flexion or extension.13 The MAS has been validated as an
145
assessment tool for the evaluation of spasticity post stroke.14 Results from other outcome measures
146
were reported in the included studies, such as those that evaluated motor functioning. However,
147
since the objective of this study was to evaluate change in spasticity, these measures were not
148
considered when interpreting findings.
RESULTS
151
TE D
149 150
M AN U
SC
RI PT
137
Study Selection and Quality
For this systematic review and meta-analysis, the literature search yielded 350 studies; the selection process is presented in Figure 1. In total, ten studies published from 2008 to 2015 met
153
inclusion criteria (Figure 1) including one RCT,4 one cross-over RCT,15 and eight pre-post
154
studies.16,17-23 The methodological quality for both RCTs was excellent, with scores ranging 8-9 on
155
the PEDro scale, and providing level 1a evidence. The eight pre-post studies were rated as level 4
156
evidence (Table 2).
158 159 160
AC C
157
EP
152
Insert Figure 1 and Table 2 about here.
Patient Characteristics Table 2 presents patient characteristics for each of the studies involved. A total of 273 poststroke (hemorrhagic=123, ischemic=150) participants were included in all ten studies; however,
ACCEPTED MANUSCRIPT 9
sample sizes ranged considerably from 5 to 80 participants. The majority of participants were male
162
(66.0%) with a mean age ranging 55.0-64.6 years; Yamada et al.16 did not report on participants’
163
gender. While Mally and Dinya17 and Yamada et al.22 did not report which side their participants’
164
hemisphere was affected, among the other studies, hemispheres were affected on the left and right
165
sides in 58 and 60 participants, respectively. The mean stroke duration ranged 6 months - 10.0
166
years.
167
Study Design
SC
168
RI PT
161
A description of the study design and methodology for each article is reported in Table 3. The RCTs by Barros-Galvao et al.4 and Etoh et al.15 compared an intervention group (receiving
170
rTMS) to a sham control group. In both studies, individuals receiving active treatment received 1
171
Hz to the unaffected hemisphere for ten sessions; however, Barros-Galvao et al.4 delivered rTMS at
172
1,500 pulses whereas Etoh et al.15 delivered rTMS at 240 pulses. Both studies provided their active
173
and control participants with physiotherapy and/or occupational therapy (0.5-2 hours).
M AN U
169
The eight pre-post studies followed similar treatment regimens whereby rTMS was
175
delivered to all enrolled participants. Four of these studies were led by the author Kakuda18-20,23 with
176
almost identical treatment protocols. In all four studies, 22 sessions of motor rTMS (1,200 pulses, 1
177
Hz) as well as 1 hour of occupational therapy and 1 hour of self-training were applied to the
178
unaffected hemisphere. Two of their studies provided concomitant therapy to participants. Kakuda
179
et al.19 administered 100 mg daily Levadopa to participants for four weeks pre rTMS until four
180
weeks post rTMS. Kakuda et al.18 injected botulinum toxin type A (maximum 240 U) into muscles
181
of the affected limb four weeks before rTMS. Two pre-post studies were led by Yamada.16,22
182
Yamada et al.22 applied contralesional rTMS (1 Hz, 2,400 pulses), occupational therapy, self-
183
training, and botulinum toxin type A to individuals over 15 days. The study by Yamada et al.16 was
184
the only one to apply rTMS to both the contralesional (1 Hz) and ipsilesional (10 Hz) hemispheres
185
in all patients, in addition to occupational therapy and self-training over 15 days. The remaining
AC C
EP
TE D
174
ACCEPTED MANUSCRIPT 10
pre-post study by Mally and Dinya17 applied motor rTMS (100 pulses; 1 Hz) to the unaffected
187
and/or affected hemisphere for 14 sessions; stimulation sites were determined by testing the effect
188
of one pulse rTMS to the affected/unaffected hemisphere and visualizing for movement in the
189
paretic side prior to treatment.
RI PT
186
The site of rTMS application was the location in which the largest motor-evoked potential
191
could be elicited in the first dorsal interosseous muscle of the unaffected upper limb. In all studies,
192
with the exception of that by Mally and Dinya,17 the intensity of stimulation was set at 90% of
193
motor threshold.
SC
190
Insert Table 3 about here.
195
M AN U
194 Adverse Effects
Eight studies reported that there were no adverse effects as a result of the rTMS
196
intervention.4,15,16,18,20-23 The remaining two studies did not report any information on whether
198
participants experienced adverse effects post intervention.17,19
199
Effectiveness
200
Elbow
TE D
197
The results from each study are reported in Table 3. With the exception of Mally and
201
Dinya17 who assessed upper and lower extremity spasticity, the remaining studies examined only
203
spasticity in the elbow, wrist, and fingers. Four studies15,16,18,20 assessed change in elbow spasticity
204
using the MAS; three of the studies16,18,20 reported significant improvement post intervention and
205
this improvement was maintained at 4-week follow-up.18 However, the cross-over RCT by Etoh et
206
al.15 found no significant improvement in elbow spasticity post intervention or at 4-week follow-up.
207
For the meta-analysis, a fixed effects model was used due to little heterogeneity between the two
208
studies included (Q = 1.75; I2 = 42.9; df = 1). The standard difference in means was significant from
209
baseline to post intervention demonstrating a treatment effect on wrist MAS scores (p=.032; Figure
210
2).
AC C
EP
202
ACCEPTED MANUSCRIPT 11
Insert Figure 2 about here.
211 212
Wrist Nine studies4,15,16,18-23 assessed change in wrist spasticity using the MAS. Etoh et al.15 and
214
Kakuda et al.19 reported no improvement in wrist spasticity post intervention or at 4-week follow-
215
up. However, the remaining seven studies reported significant improvement post intervention.4,16,18-
216
22
217
al.22 reported that while both of their treatment groups improved, there were no significant between-
218
group differences in improvement (rTMS only vs. rTMS + botulinum toxin type A). Six studies4,20-
219
23
220
meta-analysis, a fixed effects model was used due to little heterogeneity between the studies
221
included (Q =11.206; I2 = 55.4; df = 5). The standard difference in means was significant from
222
baseline to post intervention demonstrating a treatment effect on wrist MAS scores (p<.001; Figure
223
3). To examine the contribution of only the two RCTs, a separate meta-analysis was conducted
224
comparing the treatment to sham; a fixed effects model (Q = 1.021; I2 = 2.1; df = 1) demonstrated
225
that the standard difference in means was not significant from baseline to post intervention (p=.300;
226
Figure 4).
RI PT
213
SC
This effect was maintained at four-week follow-up in two studies.18,19 Interestingly, Yamada et
Eight studies15,16,18-23 examined change in finger spasticity using the MAS. Six studies
AC C
229
Fingers
EP
Insert Figure 3 and 4 about here.
227 228
TE D
M AN U
were included in the meta-analysis of rTMS on wrist spasticity as measured by the MAS. For the
230
reported improvement post intervention16,18-22 and three studies at 4-week follow-up.15,18,19 Five
231
studies20-23 were included in the meta-analysis of rTMS on finger spasticity as measured by the
232
MAS. A fixed effects model was used due to little heterogeneity between the studies included for
233
this analysis (Q = 3.392; I2 = 0; df = 4). The standard difference in means was significant from
234
baseline to post intervention demonstrating a treatment effect on finger MAS scores (p<.001;
235
Figure 5).
ACCEPTED MANUSCRIPT 12
236
Mally and Dinya17 measured change in upper extremity spasticity (predominately fingers) using an outcome measure created by the authors (i.e., “Score of Spasticity”) whereby the rating
238
system was similar to that of the MAS but rated on a scale of 0-3 instead of 0-4. The authors
239
reported significant improvement in upper extremity spasticity post intervention, and at 1- and 3-
240
month follow-up for all groups (except for group D at 3 month follow-up; see Table 3). Insert Figure 5 about here.
242
DISCUSSION
SC
241
RI PT
237
This systematic review and meta-analysis aimed to evaluate the effectiveness of rTMS in
244
improving spasticity post stroke. In an assessment of its effect on elbow flexors, three level 4 pre-
245
post studies demonstrated improvement post intervention but this was different than the one level 1a
246
cross-over RCT reporting no improvement; however, the meta-analysis demonstrated a significant
247
treatment effect immediately post intervention (p<.001). Similarly, while one level 4 study reported
248
improvement at 4-week follow-up, the level 1a cross-over RCT found no improvement. Regarding
249
wrist flexors, the results from the first meta-analysis (four level 4 studies and one level 1a RCT)
250
found a significant improvement post intervention; however, the meta-analysis of only the two
251
RCTs did not show significant improvement. Non-meta-analytic data (two level 4 studies)
252
supported significant improvement at 4-week follow-up as well. For finger flexors, the second
253
meta-analysis (four level 4 studies) found a significant improvement in spasticity post intervention;
254
this improvement was shown to be maintained at 4-week follow-up by two level 4 studies and one
255
level 1a cross-over RCT.
TE D
EP
AC C
256
M AN U
243
Spasticity is a complex disorder, encompassing a considerably diverse variety of symptoms,
257
which can have a negative impact on the quality of life of stroke survivors.24 Available therapies to
258
date include physiotherapy modalities as well as pharmacologic agents such as botulinum toxin.
259
While the latter has demonstrated effectiveness in a post-stroke population,25 additional, non-
260
pharmacological novel approaches are necessary since no one treatment appears to be completely
ACCEPTED MANUSCRIPT 13
effective for all patients. The use of rTMS as a therapeutic tool in stroke recovery is relatively new,
262
and has been reported to be of benefit in the rehabilitation of post-stroke aphasia and
263
hemineglect.10,26 Within the confines of the included studies’ limitations, this systematic review and
264
meta-analysis revealed that rTMS may be beneficial in improving spasticity post stroke.
265
RI PT
261
The limitations of included studies are similar to those reported by other systematic reviews and meta-analyses published on rTMS. From a methodological standpoint, the primary limitation
267
among eight studies was the use of a pre-post research design. Since the majority of studies were
268
uncontrolled, the authors’ conclusions are subject to some bias and open to interpretation. In the
269
current review, two studies18,22 employed rTMS concomitantly with either botulinum toxin type A18
270
or Levadopa19 without a control group. This is especially significant as botulinum toxin type A has
271
been shown to significantly and independently improve spasticity as measured using the MAS.25
272
Therefore, the positive findings reported by Kakuda et al.18 and Yamada et al.22 may have resulted
273
from the botulinum toxin type A, the rTMS or both. A complex, rigorous research design such as a
274
multiple-armed RCT can overcome this problem by separating out the individual as well as additive
275
effects of combination therapies. Regardless, it is important to note that significant treatment effects
276
were demonstrated by the meta-analyses which excluded groups receiving concomitant therapies.
M AN U
TE D
From a rehabilitation perspective, increased intensity and frequency of rehabilitation
EP
277
SC
266
sessions has been shown to positively benefit motor outcomes.27 The underlying assumption is that
279
to induce long-term effects, highly intense sessions that are frequent and of long duration are
280
necessary. From a neuromodulation perspective, this viewpoint holds true as well. To induce
281
neuroplastic effects, high pulse counts and a long duration are required. On average, the total
282
number of minutes of rTMS applied to participants in these studies ranged 400-480 minutes over
283
two weeks and the pulse count ranged 100-2,400 per session. Since rehabilitation is a long-term
284
process often lasting months to years, these diverse rTMS protocols may be suboptimal in their
AC C
278
ACCEPTED MANUSCRIPT 14
285
current form. Future trials should consider evaluating the dose-response relationship between rTMS
286
and improvement in spasticity. Considering the mechanism by which rTMS is hypothesized to be effective, all but one
RI PT
287
study included for review utilized rTMS at 1 Hz to inhibit the contralesional hemisphere. An rTMS
289
frequency of 1 Hz on the contralesional side is likely to help restore interhemispheric
290
excitation/inhibition balance thereby facilitating motor recovery.26 The re-balance of inter-
291
hemispheric cortical excitability and its impact on spasticity is unknown. Excitability-increasing
292
high frequency rTMS of ipsilesional M1 or excitability-decreasing low frequency rTMS of
293
contralesional M1 is thought to improve motor abilities in stroke patients.26,28 The positive clinical
294
signs of spasticity include enhanced stretch reflex, increased tone, and exaggerated tendon reflexes,
295
whereas the negative signs include loss of dexterity and slowness of movement and control of
296
muscles and limb segments. The negative symptoms in the UMN syndrome (i.e., weakness) are due
297
to disruption of the pathways between the supraspinal and the motor subsystems. The influence of
298
rTMS may improve the hemispheric balance of excitation and inhibition, thereby reducing tone, as
299
measured by the Ashworth scale.
TE D
M AN U
SC
288
Conceptually, a potentially more efficacious protocol for future studies may involve
301
providing bi-hemispheric stimulation to patients whereby the contralesional hemisphere is inhibited
302
via 1 Hz and the ipsilesional hemisphere is activated via HF TMS 10 Hz. This type of protocol was
303
tested in just one study16 on post-stroke spasticity. Future studies should consider designing RCTs
304
which compare uni- versus bi-hemispheric stimulation. Typical rTMS has been reported to be weak
305
and have only short-term effects lasting 30 minutes. Another evolving neuroplastic paradigm
306
involves theta-burst stimulation (TBS) which is a method for rapid, controllable, consistent, and
307
long-lasting stimulation.29 A pilot study by Huang and Rothwell30 demonstrated 50 Hz TBS to be
308
safe and effective in targeting neurons in the motor cortex. Additional studies should be conducted
309
which examine TBS for spasticity post stroke.
AC C
EP
300
ACCEPTED MANUSCRIPT 15
310
The current study is not without its own limitations. It is confined by its inclusion criteria which excluded non-English trials. Therefore, studies published in alternate languages were not
312
included in the analysis.
RI PT
311
CONCLUSION
314
There is limited available evidence to date to support the use of rTMS in improving
315
spasticity post stroke. Despite the positive findings reported by several studies, they are not without
316
their own significant methodological limitations. The effectiveness of rTMS for use in stroke
317
rehabilitation should be verified further; thus, additional better powered, high-quality RCTs are
318
needed. Furthermore, its use in combination with other tone-reducing modalities, such as botulinum
319
toxin, requires exploration.
AC C
EP
TE D
M AN U
SC
313
ACCEPTED MANUSCRIPT 16
REFERENCES
320 1.
2.
3.
Thibaut A, Chatelle C, Ziegler E, Bruno MA, Laureys S, Gosseries O. Spasticity after stroke: physiology, assessment and treatment. Brain Inj 2013;27(10):1093-105.
326 327
Pandyan AD, Gregoric M, Barnes MP, et al. Spasticity: clinical perceptions, neurological realities and meaningful measurement. Dis Rehab 2005;27(1-2):2-6.
324 325
RI PT
after stroke: a focused review. Amer J Phys Med Rehab 2012;91(3):254-70.
322 323
Corti M, Patten C, Triggs W. Repetitive transcranial magnetic stimulation of motor cortex
4.
SC
321
Barros Galvao SC, Borba Costa dos Santos R, Borba dos Santos P, Cabral ME, MonteSilva K. Efficacy of coupling repetitive transcranial magnetic stimulation and physical
329
therapy to reduce upper-limb spasticity in patients with stroke: a randomized controlled
330
trial. Arch Phys Med Rehab 2014;95(2):222-9. 5.
transcranial magnetic stimulation. Restor Neurol Neurosci 2011;29(6):395-409. 6.
neurorehabilitation after stroke? Lancet Neurol 2006;5(8):708-712.
334 7.
stimulation (rTMS). Neurophysiol Clin 2006;36(3):105-15.
336 337
8.
9.
10.
344
Hao Z, Wang D, Zeng Y, Liu M. Repetitive transcranial magnetic stimulation for
improving function after stroke. Cochrane 2013;5:CD008862.
342 343
Kobayashi M, Pascual-Leone A. Transcranial magnetic stimulation in neurology. Lancet
Neurol 2003;2(3):145-56.
340 341
Talelli P, Rothwell J. Does brain stimulation after stroke have a future? Curr Opin Neurol 2006;19(6):543-50.
338 339
Lefaucheur JP. Stroke recovery can be enhanced by using repetitive transcranial magnetic
EP
335
Hummel FC, Cohen LG. Non-invasive brain stimulation: a new strategy to improve
TE D
332 333
Hoyer EH, Celnik PA. Understanding and enhancing motor recovery after stroke using
AC C
331
M AN U
328
11.
Moseley AM, Herbert RD, Sherrington C, Maher CG. Evidence for physiotherapy practice: a survey of the Physiotherapy Evidence Database (PEDro). Aus J Physio 2002;48(1):43-9.
ACCEPTED MANUSCRIPT 17
12.
teach EBM. Toronto, Ontario: Elsevier Churchill Livingstone; 2005.
346 347
13.
Bohannon RW, Smith MB. Interrater reliability of a modified Ashworth scale of muscle spasticity. Phys Ther 1987;67(2):206-7.
348 349
Straus SE RW, Glasziou P, Haynes RB. Evidencebased medicine: How to practice and
14.
RI PT
345
Blackburn M, van Vliet P, Mockett SP. Reliability of measurements obtained with the modified Ashworth scale in the lower extremities of people with stroke. Phys Ther
351
2002;82(1):25-34.
352
15.
SC
350
Etoh S, Noma T, Ikeda K, et al. Effects of repetitive trascranial magnetic stimulation on repetitive facilitation exercises of the hemiplegic hand in chronic stroke patients. J Rehab
354
Med 2013;45(9):843-7.
355
16.
M AN U
353
Yamada N, Kakuda W, Kondo T, Shimizu M, Mitani S, Abo M. Bihemispheric repetitive
356
transcranial magnetic stimulation combined with intensive occupational therapy for upper
357
limb hemiparesis after stroke: a preliminary study. Inter J Rehab Res 2013;36(4):323-9. 17.
transcranial magnetic stimulation (rTMS). Brain Res Bull 2008;76(4):388-95.
359 360
Mally J, Dinya E. Recovery of motor disability and spasticity in post-stroke after repetitive
TE D
358
18.
Kakuda W, Abo M, Momosaki R, et al. Combined therapeutic application of botulinum toxin type A, low-frequency rTMS, and intensive occupational therapy for post-stroke
362
spastic upper limb hemiparesis. Eur J Phys Rehab Med 2012;48(1):47-55. 19.
occupational therapy with levodopa administration: an intensive neurorehabilitative
364
approach for upper limb hemiparesis after stroke. Int J Neurosci 2011b;121(7):373-8.
365 366
Kakuda W, Abo M, Kobayashi K, et al. Combination treatment of low-frequency rTMS and
AC C
363
EP
361
20.
Kakuda W, Abo M, Kobayashi K, et al. Low-frequency repetitive transcranial magnetic
367
stimulation and intensive occupational therapy for poststroke patients with upper limb
368
hemiparesis: preliminary study of a 15-day protocol. Int J Rehab Res 2010;33(4):339-45.
ACCEPTED MANUSCRIPT 18
369
21.
Kondo T, Kakuda W, Yamada N, Shimizu M, Abo M. Effects of repetitive transcranial magnetic stimulation and intensive occupational therapy on motor neuron excitability in
371
poststroke hemiparetic patients: a neurophysiological investigation using F-wave
372
parameters. Int J Neurosci 2015;125(1):25-31.
373
22.
RI PT
370
Yamada N, Kakuda W, Kondo T, Mitani S, Shimizu M, Abo M. Local muscle injection of botulinum toxin type a synergistically improves the beneficial effects of repetitive
375
transcranial magnetic stimulation and intensive occupational therapy in post-stroke patients
376
with spastic upper limb hemiparesis. Eur Neurol 2014;72(5-6):290-8. 23.
Kakuda W, Abo M, Kobayashi K, et al. Anti-spastic effect of low-frequency rTMS applied
M AN U
377
SC
374
378
with occupational therapy in post-stroke patients with upper limb hemiparesis. Brain Inj
379
2011b;25(5):496-502. 24.
Eur J Neurol 2012;19(1):21-7.
381 382
Ward AB. A literature review of the pathophysiology and onset of post-stroke spasticity.
25.
Dashtipour K, Chen JJ, Walker HW, Lee MY. Systematic literature review of
TE D
380
383
abobotulinumtoxinA in clinical trials for adult upper limb spasticity. Amer J Phys Med
384
Rehab 2015;94(3):229-38. 26.
Lefaucheur JP, André-Obadia N, Antal A, et al. Evidence-based guidelines on the
EP
385
therapeutic use of repetitive transcranial magnetic stimulation (rTMS). Clin
387
Neurphysiol2014;125(11):2150-206.
388
27.
Phys Ther 2015;39(2):95-102.
390
392
Holleran CL, Rodriguez KS, Echauz A, Leech KA, Hornby TG. Potential contributions of
training intensity on locomotor performance in individuals with chronic stroke. J Neurol
389
391
AC C
386
28.
Rossini PM, Burke D, Chen R, et al. Non-invasive electrical and magnetic stimulation of the brain, spinal cord, roots and peripheral nerves: Basic principles and procedures for
ACCEPTED MANUSCRIPT 19
393
routine clinical and research application. J Int Fed Clin Neurophysiol 2015;126(6):1071-
394
107. 29.
the Human Motor Cortex. Neuron 2005;45(2):201-6.
396 397
Huang Y-Z, Edwards MJ, Rounis E, Bhatia KP, Rothwell JC. Theta Burst Stimulation of
30.
RI PT
395
Huang YZ, Rothwell JC. The effect of short-duration bursts of high-frequency, low-
intensity transcranial magnetic stimulation on the human motor cortex. Clin Neurophysiol
399
2004;115(5):1069-75.
AC C
EP
TE D
M AN U
SC
398
ACCEPTED MANUSCRIPT 20
400
Table 1: Modified Sackett Scale
401 Description
1a
More than one randomized controlled trial (PEDro score ≥6)
1b
One randomized controlled trial (PEDro score ≥6)
2
Randomized controlled trial (PEDro score <6), prospective controlled trials, cohort studies
3
Case-control
4
Case series, pre-post or post-test
5
Observational, case report or clinical consensus
Conflicting
In the absence of evidence, agreement by a group of experts on the appropriate treatment
M AN U
SC
RI PT
Level
course
AC C
EP
TE D
402
ACCEPTED MANUSCRIPT
21
Research Country
Year Kondo et al.
Evidence
Age
N
Males Females Design
(PEDro)
Japan
10
Pre-Post
Level 4
8
Brazil
20
RCT
Level 1a (9)
13
2
Barros Galvao et al. 2014 Yamada et al.
Japan
80
Pre-Post
Level 4
50
Yamada et al.
Japan
8
Pre-Post
Level 4
Japan
18
RCT cross-
Level 1a (8)
2013 Kakuda et al.
over Japan
14
Pre-Post
Japan
39
Pre-Post
2012 Kakuda et al. 2011a
Left
mean ± SD (mo)
Hem
57.4 ± 8.1
86.0 ± 52.8
7
3
3
7
E: 57.4 ± 12.0
E: 47.8 ± 43.2
3
17
10
10
C: 64.6 ± 6.8
C: 58.9 ± 27.2
E: 62.9 ± 10.2
E: 62.0 ± 51.7
41
39
N/R
N/R
C: 57.2 ± 15.2
C: 48.0 ± 29.8
Isch
Right
N/R
62.8 ± 4.9
84.3 ± 87.2
4
4
5
3
14
4
59.7 ± 11.0
29.9 ± 18.8
5
13
8
10
Level 4
10
4
54.9 ± 9.2
87.1 ± 48.2
9
5
5
6
Level 4
30
9
56.5 ± 16.0
50.3 ± 37.8
23
16
23
16
AC C
Etoh et al.
Affected Side
N/R
EP
2013
30
TE D
2014
7
Stroke Type
mean ± SD (yr)
M AN U
2015
Stroke Onset
SC
Author,
RI PT
Table 2. Study Subject Characteristics
ACCEPTED MANUSCRIPT
22
Kakuda et al.
Japan
5
Pre-Post
Level 4
3
2
61.0 ± 4.5
Japan
15
Pre-Post
Level 4
10
5
55.0 ± 17.0
Hungary
64
Pre-Post
Level 4
37
27
57.6 ± 10.8
64.0 ± 57.0
4
1
3
2
57.0 ± 55.0
9
6
9
6
120.0 ± 76.8
18
46
N/R
N/R
Kakuda et al.
Mally &
SC
2010
RI PT
2011b
Dinya 2008
M AN U
Note: C=control group; E=experimental group; Hem=hemorrhagic; Isch=ischemic; Mo=month; N=number; N/R=not reported; NS=non-stroke;
AC C
EP
TE D
PEDro=physiotherapy evidence database; S=stroke; SD=standard deviation; Yr=year
ACCEPTED MANUSCRIPT
23
Table 3. Study protocol, outcome measures and results for all included studies Motor rTMS Intervention Protocol Control
Pulses; No. Frequency
Intensity Sessions
Kondo et
Contralesional
1 Hz
al. 2015
Contralesional
1 Hz
Galvao et
Rehab/Session
90%
2,400; 12, 40 min
120 min OT +
RMT
sessions over 15
120 min self
days Barros
Outcome
90%
1,500; 10 sessions
RMT
at 3/wk
None
30 min PT
Sham
1 Hz
90%
2,400; 12 , 40 min
RMT
MAS
Elbow
Post rTMS
MAS
Wrist Fingers +
Elbow
+
Post sham
--
Post BGD
+
4-week rTMS
+
4-week sham
--
4-week BGD
--
rTMS
sessions over 15
120 min self +
without
days
botulinum toxin
botulinum
type A
toxin type
Post rTMS only
A
Post BGD
MAS Post rTMS/ botulinum toxin type A
Elbow
+
Wrist Fingers
Post rTMS
120 min OT +
AC C
al. 2014
Contralesional
EP
TE D
al. 2014
Yamada et
Results
Protocol
SC
Target
M AN U
Year
Additional
RI PT
Author,
Wrist Fingers ++
++
++
++
--
+
ACCEPTED MANUSCRIPT
24
Bihemispheric
al. 2013
1 Hz contra
90%
2,000 each; 10, 40
120 min OT +
10 Hz ips
RMT
min sessions over
120 min self
2013
Kakuda et
Contralesional
1 Hz
al. 2012
90%
240; 10, 40 min
60-120 min
RMT
sessions over 14
voluntary
days
PT/OT
90%
1,200; 22, 20 min
RMT
sessions over 15 days
Sham
SC
1 Hz
M AN U
Contralesional
60 min OT +
None
60 min self +
botulinum toxin
Contralesional
1 Hz
al. 2011a
TE D
type A
Kakuda et
90%
1,200; 22, 20 min
60 min OT +
RMT
sessions over 15
60 min self +
None
1 Hz
90%
1,200; 22, 20 min
AC C
al. 2011b
Contralesional
EP
days
Kakuda et
RMT
60 min OT +
MAS Post rTMS
15 days Etoh et al.
None
RI PT
Yamada et
None
Elbow
Wrist Fingers
+
+
+
Elbow Wrist
Fingers
Post rTMS
--
--
--
Post sham
--
--
--
4-week BGD
--
--
+
Elbow
Wrist
Fingers
++
++
++
+
+
+
Elbow
Wrist Fingers
MAS
MAS Post rTMS/ botulinum toxin type A 4-week FU MAS Post rTMS
++
++
4-week FU
+
+
MAS
Elbow
Wrist Fingers
sessions over 15
60 min self +
Post rTMS/Levadopa
--
--
days
Levadopa
4-week FU
--
--
ACCEPTED MANUSCRIPT
25
Contralesional
1 Hz
al. 2010
90%
1,200; 22, 20 min
60 min OT +
RMT
sessions over 15
60 min self
days Mally &
Bihemispheric
30%
100; 2 sessions per
2.3T
day for 1 wk
None
MAS Post rTMS
None
M AN U
SC
Dinya 2008
1 Hz
None
RI PT
Kakuda et
a
Spasticity
Elbow
Wrist
Fingers
++
+
+
b
A
B
C
D
Post rTMS
++ ++ ++ +
1 month
++ ++ ++ +
3 month
++ ++ ++ - -
Note. + = statistically significant improvement at p<.05; ++ = statistically significant improvement at p<.01; - - = No significant difference at p>.05 a
Score of Spasticity – outcome created by the authors whereby 0=none, 1=slight, 2=fingers in flexion and passive extension to elbow difficult, 3=expressive flexion
b
A, B, C, D, refer to groups A) movement produced when stimulation was applied to both sides simultaneously, B) stimulation of either side produced no movements, C)
TE D
stimulation of contralateral hemisphere induced movement, and D) stimulation of ipsilateral hemisphere induced movement.
AC C
EP
BGD = Between-Group Difference; Contra = FU = Follow-up; OT = occupational therapy; PT = physical therapy; RMT = resting motor threshold
ACCEPTED MANUSCRIPT
26
Figure Captions
RI PT
Figure 1. Study Selection Process Figure 2. Meta-analysis of two uncontrolled studies assessing within-group improvements in MAS of the elbow flexors from baseline to post intervention
SC
Figure 3. Meta-analysis of five uncontrolled studies assessing within-group improvements in MAS of the wrist flexors from baseline to post
M AN U
intervention
Figure 4. Meta-analysis of two RCTs assessing between-group improvements in MAS of the wrist flexors from baseline to post intervention Figure 5. Meta-analysis of four uncontrolled studies assessing within-group improvements in MAS of the finger flexors from baseline to
AC C
EP
TE D
post intervention
ACCEPTED MANUSCRIPT
RI PT
Records identified through database searching (n = 350)
Additional records identified through other sources (n = 0)
SC
Identification
PRISMA 2009 Flow Diagram
M AN U
Records screened (n = 341)
Records excluded (n = 328)
Full-text articles assessed for eligibility (n = 13)
Full-text articles excluded, based on not satisfying inclusion criteria (n = 3)
TE D EP AC C
Included
Eligibility
Screening
Records after duplicates removed (n = 341)
Studies included in qualitative synthesis (n = 10)
Studies included in quantitative synthesis (meta-analysis) (n = 10)
From: Moher D, Liberati A, Tetzlaff J, Altman DG, The PRISMA Group (2009). Preferred Reporting Items for Systematic Reviews and MetaAnalyses: The PRISMA Statement. PLoS Med 6(6): e1000097. doi:10.1371/journal.pmed1000097
For more information, visit www.prisma-statement.org.
Study name
Statistics for each study Std diff in means
Standard error
Variance
Lower limit
Upper limit
RI PT
ACCEPTED MANUSCRIPT
Std diff in means and 95% CI
Z-Value
p-Value
0.000
0.471
0.222
-0.924
0.924
0.000
1.000
0.733
0.291
0.085
0.163
1.303
2.520
0.012
0.531
0.247
0.061
0.046
1.016
2.145
SC
Etoh et. al 2013 Kakuda et al. 2010
0.032
AC C
EP
TE D
M AN U
-1.00
-0.50
0.00
0.50
1.00
Statistics for each study Std diff Standard in means error
Variance
Lower limit
Upper limit
Std diff in means and 95% CI
Z-Value p-Value
SC
Study name
0.693
0.352
0.124
0.003
1.384
1.969
0.049
Barros Galvao et al. 2014
1.441
0.451
0.204
0.556
2.326
3.192
0.001
M AN U
Kondo et al. 2015
Yamada et al. 2014
0.849
0.189
0.036
0.478
1.220
4.487
0.000
Kakuda et al. 2011a
0.329
0.164
0.027
0.007
0.651
2.002
0.045
Kakuda et al. 2010
0.667
0.285
0.081
0.107
1.226
2.336
0.020
0.333
0.111
-0.653
0.653
0.000
1.000
0.165
0.027
0.286
0.932
3.701
0.000
EP
TE D
0.000 0.609
AC C
Etoh et al. 2013
RI PT
ACCEPTED MANUSCRIPT
-3.00
-1.50
0.00
1.50
3.00
Statistics for each study Std diff Standard in means error
Variance
Lower limit
Upper limit
Std diff in means and 95% CI
Z-Value p-Value
0.665
0.459
0.211
-0.235
1.566
1.448
Etoh et al. 2013
0.000
0.471
0.222
-0.924
0.924
0.000
0.341
0.329
0.108
-0.304
0.986
1.037
TE D EP
0.148
1.000
M AN U
Barros Galvao et al. 2014
AC C
SC
Study name
RI PT
ACCEPTED MANUSCRIPT
0.300
-1.00
-0.50
0.00
0.50
1.00
Statistics for each study Std diff Standard in means error
Variance
Lower limit
Upper limit
Std diff in means and 95% CI
Z-Value p-Value
SC
Study name
Kondo et al. 2015
0.768
0.360
0.130
0.063
1.474
2.135
Yamada et al. 2014
0.749
0.184
0.034
0.389
1.109
4.080
Kakuda et al. 2011a
0.336
0.165
0.027
0.013
0.658
2.041
0.041
Kakuda et al. 2010
0.465
0.272
0.074 -0.068
0.997
1.710
0.087
Etoh et al. 2013
0.675
0.485
0.235 -0.275
1.625
1.393
0.164
0.540
0.104
0.011
0.744
5.181
0.000
AC C
EP
0.033
0.000
M AN U
TE D
0.336
RI PT
ACCEPTED MANUSCRIPT
-1.00
-0.50
0.00
0.50
1.00