Repetitive transcranial magnetic stimulation (rTMS) has been extensively used in poststroke rehabilitation over the last two decades. This priming technique is believed to facilitate cortical excitability, thereby promoting neuroplasticity and functional recovery after stroke.1–7 Since the mid-1990s, researchers have published more than 174 articles describing the effects of rTMS on poststroke motor impairment and functional activity, predominantly in the upper extremities. Most studies found substantial improvements of arm/hand movements and muscle tone on completion of the rTMS course. The best outcomes were observed in patients with acute and subacute strokes8,9 or with lesions located in subcortical versus cortical brain structures.10,11
Despite the growing number of reports citing the positive effects of rTMS after stoke, questions remain about the utility of rTMS in poststroke rehabilitation. A meta-analysis by Hao et al.1 (2013), which included 19 clinical studies encompassing 588 patients, showed that neither low- nor high-frequency stimulation had significant effects on arm motor or functional recovery after stroke. Several factors could explain the discrepancy between results of rTMS in different studies. Although rTMS can be applied with different frequencies (0.1–50 Hz) to different parts of the motor cortex responsible for arm/hand movements, which delivery mode of rTMS is most effective remains unknown. A group of experts reviewing this method assigned a moderate level of evidence (B) to the use of low-frequency stimulation of the primary motor cortex (M1 zone) of the nonaffected hemisphere in patients in the late-recovery phase after stroke (after 6 months). A lower level of evidence (C) was assigned to high-frequency stimulation of the M1 zone of the affected hemisphere in patients in the acute and subacute phases.12 Nevertheless, the mechanisms of rTMS action were not well articulated. Inconsistency of reported results also may be related to the spatial accuracy of locating a cortical area for rTMS application.
As rTMS technologies advance, some of the above concerns can be partially addressed by applying navigated rTMS for noninvasive brain stimulation of motor areas. Using MRI scans, this technology allows the stimulation site to be located with greater accuracy and precision that is important, considering that motor representation areas can be remapped after stroke. Also, the same area (with accuracy up to 2 mm) can be stimulated in each consecutive session over the entire therapy course that reduces variability of motor responses.13,14 These features make the navigated systems much more reliable and user-friendly, compared with regular hand-held coils equipment. Despite the potential advantages of navigated rTMS, its efficacy has not been tested extensively yet, and mainly positive results have been reported in a few studies. Bashir et al.15 (2016) reported increased cortical excitability of the ipsilesional hemisphere and motor response of the hemiparetic hand in patients with stroke after navigated rTMS of the contralesional hemisphere at low frequency. Ayache et al.16 (2016) induced analgesia by applying stimulation with and without navigation in patients with focal neuropathic pain of the upper or lower extremities. Stimulation was applied with high frequency to the primary motor cortex, contralateral to pain side. Both procedures were effective, but the navigated approach produced a more prolonged effect.
To test the superiority of navigated over nonnavigated rTMS, the Nexstim company designed a multicenter phase III clinical trial, Navigated Inhibitory rTMS in Contralesional Hemisphere Evaluation, which was started in 12 different sites in the United States. However, after the second interim analysis of data from 138 patients showed futility of further trial continuation, the Data Safety and Monitoring Board recommended termination of the trial. However, active treatment in the Navigated Inhibitory rTMS in Contralesional Hemisphere Evaluation trial gained clinically meaningful results, with over two-thirds of patients responding positively. Thus, navigated rTMS shows potential to be an effective rehabilitation technology, but practical evidence of the system is lacking. This study was designed as an exploratory clinical trial to test the effects of different navigated rTMS stimulation modes (1, 10 Hz, and combination of 1–10 Hz) on poststroke motor impairment and functional limitations. The inhibitory low-frequency stimulation was applied to the nonaffected hemisphere, whereas the excitatory high-frequency stimulation was applied to the affected hemisphere; both were supposed to increase the excitability of motor cortical representation of the affected arm. These stimulation modes are comparable with other studies that used navigated and nonnavigated rTMS.8,9,15–17 Results of this study will be used for designing a large-scale clinical trial.
The study was designed as an exploratory interventional four-arm randomized, sham-controlled trial. The trial was registered in the ClinicalTrials.gov database (NCT01652677) and conducted between December 2011 and October 2014 with the CONSORT flow diagram (Fig. 1). All eligible patients who were willing to participate in the study were randomized into one of four groups and received no stimulation (sham group), stimulation of the nonaffected hemisphere at a frequency of 1 Hz (low-frequency group), stimulation of the affected hemisphere at a frequency of 10 Hz (high-frequency group), or stimulation with a combination of 1–10 Hz high- and low-frequency stimulations (high–low frequency group). Randomization was done in blocks using sealed envelopes.
Ninety-four participants who were undergoing treatment at the Research Center of Neurology (Moscow, Russia) were recruited and screened for the study. Eligibility criteria included a history of stroke confirmed by MRI in the area of the middle cerebral artery, stroke onset between 1 and 12 months before study enrollment, patient age between 18 and 70 years, and stroke severity of four points or less as determined by the Modified Rankin Scale.18 Participants were excluded if they presented severe somatic pathology (e.g., acute myocardial infarction, lower extremity deep vein thrombosis, or episodes of pulmonary embolism), had an implanted cardiostimulator, intracardiac catheter, or electronic pump, or had metallic staples, vascular sutures, metal plates closing skull defects, or metallic foreign bodies in the cranial cavity. All individuals signed an informed consent form, approved by the local ethics committee of the Research Center of Neurology, before enrollment.
Of the 92 recruited and screened individuals, 64 patients met the eligibility criteria and were enrolled in the study (Fig. 1). The rTMS intervention was completed by 42 patients (26 men), with an average (mean ± standard deviation) age of 58.5 ± 10.7 years. Participants were distributed among groups as follows: sham (n = 10), low-frequency (n = 11), high-frequency (n = 13), and high–low frequency (n = 8) groups. Nine patients withdrew from the study at different stages because they developed epileptic seizures, reported subjective “bad” or “tired” feelings after rTMS session(s) or for no cause. Thirteen participants were excluded from analysis, mainly because of incomplete data sets.
All 42 patients who completed the study had experienced a single unilateral stroke, with the average time since a stroke onset of 6.45 ± 5.91 months, ranging from 6 weeks to 12 full months. As there is no agreement on the time windows of poststroke recovery phases, a stroke onset greater than 6 weeks was considered as chronic, as suggested by several researchers previously.19,20 Average stroke severity at the time of admission was 2.80 ± 0.85 points, according to the Modified Rankin Scale (range: 1–4 points; score of six points indicates death, Table 1). All participants presented with arm and hand motor impairments and activity limitations. Scores on clinical scales ranged as follows: (1) 6 to 63 points on the arm and hand section of the Fugl–Meyer Stroke Assessment Scale (FM21), with 20 points indicating severe and 44 points indicating moderate arm paresis; (2) 0 to 3 points on the modified Ashworth Scale of Muscle Spasticity (MAS22); with four points indicating severe muscle tone in flexion or extension; and (3) 25 to 100 points on the Barthel Index (BI23) with 100 points indicating independence in all activities of daily living (ADLs). Participants in all groups were matched by age, time since stroke onset, and clinical manifestations (Table 1).
Intervention consisted of 10 sessions of navigated rTMS, scheduled 5 times a week over two consecutive weeks. Repetitive transcranial magnetic stimulation was delivered with the Magstrim Rapid two system and calibrated with a navigation system NBS eXimia Nexstim system (Nexstim Ltd, Helsinki Finland). Stimulation was applied to a cortical representation of the abductor pollicis brevis (APB) muscle in the primary motor area of the affected or nonaffected hemisphere, depending on group assignment.
Intervention began with mapping of the APB muscle area in the primary motor cortex in both hemispheres. Each participant underwent high-resolution T1-weighted structural MRI scanning. Images were imported into the NBS eXimia Nexstim system software for automatic three-dimensional brain reconstruction and for finding a potential cortical representation of the APB muscle. Once found, this cortical representation was stimulated with a 70-mm figure-eight-shaped BiPulse Nexstim coil with a maximal possible magnetic field strength of 199 V/m and magnetic pulse duration of 280 μs. During stimulation, a magnetic field of 80 to 100 V/m was generated to detect motor evoked potentials (MEPs) from the contralateral APB muscle. Motor-evoked potentials were recorded by the Nextim EMG system (Nexstim, EMD, Helsinki, Finland) with 0.8-cm2 surface electrodes, placed on the skin overlying the APB muscle, and a reference electrode attached to the distal interphalangeal joint of the same hand. The resting motor threshold (RMT) of the muscle was defined based on the motor-evoked potentials as the lowest stimulation intensity able to evoke motor responses of at least 50 μV peak-to-peak amplitude in 5 of 10 trials with the muscle at rest. The RMT was determined with the APB muscle being completely relaxed that was monitored using continuous EMG signal. Resting motor threshold was measured as the percentage of the maximum intensity of the magnetic stimulus (1.5 T). Stimulation with a magnetic field of 110% of the MT was applied to confirm the cortical area where stimulation-evoked motor responses of at least 50 μV and the point (hot-spot) where stimulation elicited the maximum amplitude of motor response in the contralateral hand. Average mean values of RMT in the affected and nonaffected hemispheres are presented in Table 1 for each group.
After APB muscle cortical mapping, rTMS was applied to the cortical hot-spot of the APB muscle with a Magstim Rapid2 stimulator, according to a protocol developed for each group. The low-frequency group received a single train of 1,200 stimuli of 1-Hz rTMS at 100% resting RMT to the nonaffected hemisphere for 20 minutes per session. The high-frequency group received 200 stimuli of 10-Hz rTMS at 80% of resting RMT to the affected hemisphere for 10 minutes. The high–low frequency group had a consecutive combination of the above sets of stimuli (10 minutes of 10-Hz rTMS at 80%, followed by 20 minutes 1-Hz rTMS at 100% of resting RMT), beginning with the high-frequency stimulation. The sham group received pseudo-stimulation, in which the Magstim coil was placed over the confirmed cortical representation of the APB muscle but was not connected to the stimulator. The sham procedure imitated the high-frequency stimulation protocol, and the patient could hear all the real sounds but without actual exposure to the electromagnetic field.
During the procedure, patients sat in a comfortable recliner chair with their hands in a supine position at the hand rests. They were instructed to remain relaxed and silent during the whole intervention to avoid speech-induced modulation of cortical excitability. They were monitored for drowsiness and asked to keep their eyes open; this condition was checked regularly by the experimenter. Concurrently with navigated rTMS, all patients received a course of 10 physical therapy sessions, each lasting for 45 to 55 minutes, and scheduled 5 times a week on two consecutive weeks. The physical therapy focused on correcting poststroke motor and functional deficits, according to a standard protocol established in the research center, with consideration of each patient's individual limitations.
Clinical Outcome Measures
Regardless of group assignment, each participant was evaluated twice: before the intervention began (pretest) and immediately after its completion (posttest). Outcome measures included scores of three valid and reliable clinical scales (FM, MAS, and BI), described in the Subjects subsection. Evaluations were performed by experienced clinicians who were masked to each patient's group assignment.
Data normality was verified with the Kolmogorov–Smirnov test (P > 0.05). Pretest clinical outcome measures and RMTs were compared between groups (sham, low-frequency, high-frequency, and high–low frequency) by 1-way ANOVA. Dependent t-test was used for within-group comparisons of pre- and post-test clinical scores. A minimum significance level of P < 0.01 was set for all comparisons.
During the intervention period, two patients (3%) developed secondary generalized epileptic seizures. One female patient (3 months after a left middle cerebral artery stroke) presented with seizures during single-pulse stimulation for diagnostic mapping. The second patient (4 months after a right middle cerebral artery stroke) developed seizures during the first session of high-frequency rTMS. Both seizures could have been prevented by excluding these patients, as the epileptiform signs were seen on EEG records during initial screening. Twenty-five patients (39%) reported short-lasting moderate headache, requiring no additional therapy. Twenty-nine patients (45%), including 73% of patients receiving high-frequency stimulation alone, showed increased paroxysmal or newly emerged epileptiform EEG activity after 10 sessions. One patient experienced an aggravated somatic status involving lower extremity deep vein thrombosis and thrombus flotation. In total, nine patients withdrew from the study at different stages due to one or more of the above side effects or for no reason. Thirteen participants were excluded from analysis, mainly because of incomplete data sets. Despite the high dropout rate of 22 participants (34%), the final sample size had at least 80% power at the P = 0.5 level.
Participants in all groups had no significant between-group differences in pretest scores of arm and hand functions on the FM (F 3,38 = 0.09, P = 0.96), arm spasticity on the MAS (F 3,38 = 0.82, P = 0.51), and performance of ADLs on the BI (F 3,38 = 0.39, P = 0.75), as confirmed by a 1-way ANOVA. There was no difference in RMTs of the affected and nonaffected hemispheres between groups (Table 1). On completion of the experiment, most of the 42 participants reported subjective improvements related to the navigated rTMS intervention. Perceived improvements were reported even by patients in the sham group. The low-frequency group improved their FM scores by 11.3 points or 35% from the pretest to posttest (t-test, P = 0.001), the high-frequency group by 7.6 points or 23% (P = 0.000), and the high–low frequency group by 9.7 points or 33% (P = 0.018), with no difference in the sham group (P = 0.164, Fig. 2). All improvements exceeded the minimal clinically important difference range of 4.25 to 7.25 points established for the arm section of the FM.24 Arm spasticity decrease on the MAS of 0.55 (43%) points was observed in the low-frequency group (t-test p = 0.006) and of 0.69 points (37%) in the high-frequency group (P = 0.001), with improvements in both groups being below the minimal detectable changes.25 No posttest changes in arm spasticity were found in the sham and high–low frequency group (P = 0.591 and P = 0.033, respectively). Finally, the BI, which characterizes performance of ADLs, improved by 12.7 points (17%) in the low-frequency group (P = 0.003) and by 4.61 points (5.51%) in the high-frequency group (P = 0.008), with no changes in two other groups. Barthel Index score increases of 4.02 and 1.85 points are the minimal detectable changes for patients with chronic stroke and acute stroke, respectively.26
Adding navigated rTMS intervention to a course of conventional rehabilitation facilitated recovery of patients with stroke. On completion of the intervention, participants in three groups, receiving rTMS in different modes, improved their arm and hand functions on the FM. Arm spasticity and the BI scores were significantly improved in the groups receiving low- or high-frequency stimulation alone. Thus, the results of the study suggest that navigated rTMS can be an effective supplement to conventional poststroke rehabilitation programs and positively affect motor and functional recovery. This is consistent with other studies that used low- and high-frequency stimulation after stroke, although nonnavigated.1–4
All groups, receiving actual rTMS improved their scores after stimulation to some extent. The low- and high-frequency groups showed changes in all three outcome measures, whereas the high–low frequency group improved only arm and hand functions on the FM. Difference between groups has not been analyzed because it seems to be negligible. This was probably due to several limitations of this study, with some of these limitations being practically unavoidable but others caused by a lack of understanding the mechanisms of rTMS action after stroke. Noninvasive brain stimulation is believed to change the cortical excitability, thereby facilitating mechanisms of neural repair. However, affected and nonaffected hemispheres may respond differently to stimulation. Two models have been proposed to explain a role of ipsilateral versus contralateral cortical projections in motor recovery after stroke. The model of interhemispheric competition suggests that the nonaffected hemisphere frequently becomes hyperactive after stroke, leading to inhibition of the affected hemisphere via transcallosal fibers. This situation creates interhemispheric competition,27,28 which suppresses activity in the affected hemisphere and negatively correlates with recovery of motor function.29 According to this model, stimulation of the nonaffected hemisphere, for example by applying inhibitory low-frequency rTMS could benefit recovery. The other, vicariation model predicts that the poststroke recovery mostly occurs at the cost of residual neural cells in the nonaffected hemisphere, in which inhibition could be counterproductive. Later, these two concepts have been elaborated by Di Pino et al. (2014), who proposed a bimodal balanced model for neuronal recovery. The model introduces a term “structural reserve,” describing the remaining after stroke neuronal substrate for functional movements.30 This substrate dictates a contribution of the affected versus nonaffected hemisphere to functional recovery. When the structural reserve is high, the interhemispheric competition model dominates over the vicariation model, and vice versa when reserve is low. In our study, by applying low-frequency stimulation to the nonaffected hemisphere, we attempted to suppress its cortical excitability and to reduce an inhibitory effect. However, such indirect signs of structural reserve as the size of lesion, severity of motor impairments, and amplitude of motor responses have not been taken into consideration when assigning participants to different experimental groups. An attempt was made to make the groups as homogeneous as possible. On average, all experimental groups had similar demographic and clinical characteristics, but within-group variability was not controlled. Therefore, participants with different severities of poststroke impairments in the same group could respond differently to a similar stimulation mode, counterbalancing each other's scores on average and reducing between-group differences. This was a major limitation of this study.
Muscle spasticity was reduced in groups receiving low- or high-frequency stimulation alone. Although statistically significant, changes in muscle tone on the MAS were less than a 1-point increment, reflecting a clinically significant improvement for this scale for stroke patients.25 This result may be explained by the relatively mild spasticity of study participants. A positive effect of rTMS stimulation on muscle tone was reported by other studies, confirming our results.31–33 Muscle tone increases gradually after stroke, mainly because of reduced corticoreticular regulation of the modified descending pathways originating in the brainstem. These pathways, reticulospinal, vestibulospinal, and rubrospinal (almost ineffective in humans), provide balanced excitatory and inhibitory regulation of inter- and lower motor neurons, setting a threshold for muscle stretch reflexes and tone, respectively.34 Cortical stimulation may disinhibit spared or promote development of new neurons and their cortireticular projections, normalizing activity in modified descending pathways. Thus, the mechanisms of muscle tone reduction and motor functional recovery after brain stimulation could use different descending pathways, triggered by similar mechanism of neuroplasticity.35 Further research is needed to elucidate the mechanisms of rTMS action on cortical spasticity.
In addition to the improved motor function and tone, the low- and high-frequency groups showed increased values on the BI, which measures independence in performance of ADLs. Improvements of motor functions and muscle tone likely created a compound effect, leading to transition to a new level of independence, as evidenced by the increased BI scores. Application of high–low stimulation protocol did not change the level of spasticity and ADLs, and overall resulted in fewer effects compared with two other rTMS stimulation modes. According to an initial idea, stimulating consequently the affected and nonaffected hemispheres is supposed to increase an intervention dosage and to use at least two different recovery mechanisms, as discussed above. However, application of excitatory and inhibitory protocols to the same subject could be counterproductive by “washing out” each other's effect, as suggested by the bimodal balanced model.29
Other major limitations of the study included a high attrition rate, small sample size, and unequal distribution of participants between groups, all of which were related to unanticipated adverse effects. The greater number of adverse effects (e.g., seizures and increased epileptiform activity in EEG) were observed after high-frequency stimulation. These complications can be prevented in future studies by performing EEG screening in all patients before enrolling them to receive rTMS. Results could have also been affected by a time window of stroke onsets in study participants that ranged from one to 12 full months. This period allows maximal neuroplasticity-based motor recovery, while excluding the period of active spontaneous recovery (3–4 weeks from stroke onset). Even within this time window, patients may recover at different paces. Increased synaptic neuroplasticity characterize the first three months of poststroke recovery that then slows down in later phases.36 Having patients with different times of stroke onsets in the same group could also increase variability of an average group response to selected stimulation mode. All these issues will be addressed in future research.
In summary, the results of this study are consistent with other works and provide data necessary for designing a large-scale clinical trial. The proposed trial will compare the effects of different modes of navigated rTMS in high- versus low-functioning individuals with a unilateral stroke occurred from 3 to 6 months previously. The trial will consider all the shortages of this study, which may help to detect a clearer difference or no difference between navigated rTMS stimulation modes.
1. Hao Z, Wang D, Zeng Y, Liu M. Repetitive transcranial magnetic stimulation for improving function after stroke. Cochrane Database Syst Rev 2013;5:CD008862.
2. Hsu WY, Cheng CH, Liao KK, Lee IH, Lin YY. Effects of repetitive transcranial magnetic stimulation on motor functions in patients with stroke: a meta-analysis. Stroke 2012;43:1849–1857.
3. Le Q, Qu Y, Tao Y, Zhu S. Effects of repetitive transcranial magnetic stimulation on hand function recovery and excitability of the motor cortex after stroke: a meta-analysis. Am J Phys Med Rehabil 2014;93:1–9.
4. Dafotakis M, Grefkes C, Eickhoff SB, Karbe H, Fink GR, Nowak DA. Effects of rTMS on grip force control following subcortical stroke. Exp Neurol 2008;211:407–412.
5. Grefkes C, Nowak DA, Wang LE, Dafotakis M, Eickhoff SB, Fink GR. Modulating cortical connectivity in stroke patients by rTMS assessed with fMRI and dynamic causal modeling. Neuroimage 2010;50:233–242.
6. Mansur CG, Fregni F, Boggio PS, et al. A sham stimulation-controlled trial of rTMS of the unaffected hemisphere in stroke patients. Neurology 2005;64:1802–1804.
7. Nowak DA, Grefkes C, Dafotakis M, et al. Effects of low-frequency repetitive transcranial magnetic stimulation of the contralesional primary motor cortex on movement Kinematics and neural activity in subcortical stroke. Arch Neurol 2008;65:741–747.
8. Khedr EM, Ahmed MA, Fathy N, Rothwell JC. Therapeutic trial of repetitive transcranial magnetic stimulation after acute ischemic stroke. Neurology 2005;65:466–468.
9. Khedr EM, Abdel-Fadeil MR, Farghali A, Qaid M. 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.
10. Khedr EM, Etraby AE, Hemeda M, Nasef AM, Razek AAE. Long-term effect of repetitive transcranial magnetic stimulation on motor function recovery after acute ischemic stroke. Acta Neurol Scand 2010;121:30–37.
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. Lefaucheur JPP, 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.
13. Treister R, Lang M, Klein MM, Oaklander AL. Non-invasive transcranial magnetic stimulation (TMS) of the motor cortex for neuropathic pain-at the tipping point? Rambam Maimonides Med J 2013;4:e0023.
14. Gugino LD, Rafael Romero J, Aglio L, et al. Transcranial magnetic stimulation coregistered with MRI: a comparison of a guided versus blind stimulation technique and its effect on evoked compound muscle action potentials. Clin Neurophysiol 2001;112:1781–1792.
15. Bashir S, Vernet M, Najib U, et al. Enhanced motor function and its neurophysiological correlates after navigated low-frequency repetitive transcranial magnetic stimulation over the contralesional motor cortex in stroke. Restor Neurol Neurosci 2016;34:677–689.
16. Ayache SS, Ahdab R, Chalah MA, et al. Analgesic effects of navigated motor cortex rTMS in patients with chronic neuropathic pain. Eur J Pain 2016;20:1413–1422.
17. Sasaki N, Mizutani S, Kakuda W, Abo M. Comparison of the effects of high- and low-frequency repetitive transcranial magnetic stimulation on upper limb hemiparesis in the early phase of stroke. J Stroke Cerebrovasc Dis 2013;22:413–418.
18. Farrell B, Godwin J, Richards S, Warlow C. The United Kingdom transient ischaemic attack (UK-TIA) aspirin trial: final results. J Neurol Neurosurg Psychiatry 1991;54:1044–1054.
19. Bahn MM, Oser AB, Cross DT. CT and MRI of stroke. J Magn Reson Imaging 1996;6:833–845.
20. Adeyemo BO, Simis M, Macea DD, Fregni F. Systematic review of parameters of stimulation, clinical trial design characteristics, and motor outcomes in non-invasive brain stimulation in stroke. Front Psychiatry 2012;3:88.
21. Fugl-Meyer AR, Jääskö L, Leyman I, Olsson S, Steglind S. The post-stroke hemiplegic patient. 1. a method for evaluation of physical performance. Scand J Rehabil Med 1975;7:13–31.
22. Bohannon RW, Smith MB. Inter rater reliability of a modified Ashworth Scale of muscle spasticity. Phys Ther 1987;67:206–207.
23. Collin C, Wade DT, Davies S, Horne V. The Barthel ADL Index: a reliability study. Int Disabil Stud 1988;10:61–63.
24. Page SJ, Fulk GD, Boyne P. Clinically important differences for the upper-extremity Fugl-Meyer scale in people with minimal to moderate impairment due to chronic stroke. Phys Ther 2012;92:791–798.
25. Shaw L, Rodgers H, Price C, et al. BoTULS: a multicentre randomised controlled trial to evaluate the clinical effectiveness and cost-effectiveness of treating upper limb spasticity due to stroke with botulinum toxin type A. Health Technol Assess 2010;14:1–113.
26. Hsieh YW, Wang CH, Wu SC, Chen PC, Sheu CF, Hsieh CL. Establishing the minimal clinically important difference of the Barthel Index in stroke patients. Neurorehabil Neural Repair 2007;21:233–238.
27. Ward NS, Cohen LG. Mechanisms underlying recovery of motor function after stroke. Arch Neurol 2004;61:1844–1848.
28. Nowak DA, Grefkes C, Ameli M, Fink GR. Interhemispheric competition after stroke: brain stimulation to enhance recovery of function of the affected hand. Neurorehabil Neural Repair 2009;23:641–656.
29. Grefkes C, Ward NS. Cortical reorganization after stroke: how much and how functional? Neuroscientist 2014;20:56–70.
30. Di Pino G, Pellegrino G, Assenza G, et al. Modulation of brain plasticity in stroke: a novel model for neurorehabilitation. Nat Rev Neurol 2014;10:597–608.
31. Barros Galvão SC, Borba Costa Dos Santos R, Borba Dos Santos P, et al. Efficacy of coupling repetitive transcranial magnetic stimulation and physical therapy to reduce upper-limb spasticity in patients with stroke: a randomized controlled trial. Arch Phys Med Rehabil 2014;95:222–229.
32. Kakuda W, Abo M, Kobayashi K, et al. Anti-spastic effect of low-frequency rTMS applied with occupational therapy in post-stroke patients with upper limb hemiparesis. Brain Inj 2011;25:496–502.
33. Málly J, Dinya E. Recovery of motor disability and spasticity in post-stroke after repetitive transcranial magnetic stimulation (rTMS). Brain Res Bull 2008;76:388–395.
34. Korzhova J, Sinitsyn D, Chervyakov A, et al. Transcranial and spinal cord magnetic stimulation in treatment of spasticity. A literature review and meta–analysis. Eur J Phys Rehabil Med 2016 [Epub ahead of print].
35. Li S. Spasticity, motor recovery, and neural plasticity after stroke. Front Neurol 2017;3:120.
36. Maulden SA, Gassaway J, Horn SD, Smout RJ, DeJong G. Timing of initiation of rehabilitation after stroke. Arch Phys Med Rehabil 2005;86(12 Suppl 2):S34–S40.