Can TMS help stroke recovery?
Yes, with realistic expectations. This is a systematic review and meta-analysis of 57 trials involving 2,595 stroke patients, and it found that repetitive transcranial magnetic stimulation produced significant improvements in both motor and cognitive outcomes after stroke. The authors framed the question as “hype or hope?” and landed closer to hope, with an important caveat: the evidence is challenged by unexplained heterogeneity across many small trials.
TMS is thought to work for stroke recovery by stimulating healthy brain areas to take over functions from damaged regions, a process called neuroplasticity. The magnetic stimulation can either enhance activity in the affected hemisphere or reduce overactivity in the unaffected side that may be interfering with recovery. The authors of this review note that the mechanisms of action are still insufficiently understood.
What the data show:
- Motor improvement: Significant effect sizes for upper limb synergies and muscle strength at the level of body function
- Cognitive benefits: Significant effect sizes for language functioning, global cognitive functioning, and visual and spatial inattention
- Everyday function: Significant effect sizes at the level of activities, not just at the level of body function
- Treatment timing: Significant effects were found both within and beyond 3 months after stroke
- The main limitation: Unexplained heterogeneity across many small trials, which is why the authors call for large trials with individual patient data
This review, published in Stroke in October 2023, followed Cochrane and PRISMA protocols, included only trials with at least 10 patients per treatment group, and classified outcomes using the International Classification of Functioning, Disability, and Health. Funnel plots showed no publication bias.
Dr. Kumar’s Take
This review gives a refreshingly honest assessment of TMS for stroke recovery. The authors are right to ask “hype or hope?” because there has been a lot of overenthusiastic promotion of TMS in this space. My read is that the signal is real: significant effects appeared across motor and cognitive domains, and importantly at the level of activities, not just laboratory measures of body function. But the heterogeneity across trials is unexplained, and the trials are small. That is the difference between a finding I would act on cautiously in a rehabilitation program and one I would promise a patient. I tell patients that TMS is a reasonable adjunct to rehabilitation, not a substitute for it, and that the size of the benefit for any individual person is still uncertain.
What the Research Shows
The review was a systematic search of the literature performed according to Cochrane and PRISMA protocols, restricted to trials with at least 10 patients per treatment group. Fifty-seven articles met criteria, covering 2,595 patients. Meta-analysis was performed when at least three trials reported on the same construct.
Outcomes were classified using the International Classification of Functioning, Disability, and Health, which separates effects at the level of body function from effects at the level of activities. That distinction matters: a change in muscle strength on a testing table is not the same thing as a change in what a person can do.
Results in Real Numbers
At the level of body function, the review found significant effect sizes for upper limb synergies, muscle strength, language functioning, global cognitive functioning, and visual and spatial inattention. Significant effect sizes were also found at the level of activities.
These significant effects were present with rTMS delivered both within 3 months after stroke and beyond 3 months after stroke.
Where the picture gets harder: several summary effect sizes came with significant heterogeneity. The authors ran sensitivity analyses testing whether individual effect sizes correlated with, or differed by, possible effect modifiers including patient characteristics, rTMS characteristics, and trial characteristics. They found no subgroup differences and no significant correlations with any modifier they tested. So the variation between trials remains unexplained.
Funnel plots showed no publication bias, which argues against the variation being an artifact of selective reporting.
Who Benefits Most
This is the honest answer from this review: it could not identify who benefits most. The sensitivity analyses looked for exactly that, testing patient, stimulation, and trial characteristics as possible effect modifiers, and none of them explained the differences in effect between trials.
That is why the authors recommend large trials that collect individual patient data on baseline severity and brain network integrity, with sufficiently powered subgroup analyses. Until that work is done, patient selection for TMS after stroke rests on clinical judgement rather than on demonstrated predictors.
Safety, Limits, and Caveats
The main limitation here is evidentiary rather than clinical. The efficacy signal is challenged by unexplained heterogeneity across many small sampled trials, and the mechanisms of action remain insufficiently understood. The review reported no conflicts of interest.
The authors also recommend protocolized, time-locked training of the target behavior in future trials, meaning the stimulation should be paired with practicing the specific function you are trying to restore. They suggest that additional neurophysiological and biomechanical data may help identify biomarkers of treatment efficacy.
Practical Takeaways
- Treat TMS as an adjunct within a rehabilitation program, paired with training of the specific function being targeted
- Do not rule out TMS on the basis of time since stroke: significant effects were found both within and beyond 3 months
- Do not expect a validated way to predict responders yet, since no patient, stimulation, or trial characteristic explained the variation in effect
- Look for benefits at the level of activities, not only at the level of body function, since the review found significant effects at both
- Consider both motor targets, such as upper limb synergies and muscle strength, and cognitive targets, such as language, global cognition, and visual and spatial inattention
- Interpret any single small trial cautiously given the unexplained heterogeneity across this literature
Related Studies and Research
Episode 31: Depression Explained, The Biology Behind the Darkness
Episode 32: Depression Recovery Roadmap: A Step-by-Step, Evidence-Based Plan
Default Mode Network Mechanisms of Transcranial Magnetic Stimulation in Depression
FAQs
What kind of improvement does this review support for stroke patients?
Significant effect sizes at the level of body function, covering upper limb synergies, muscle strength, language functioning, global cognitive functioning, and visual and spatial inattention, plus significant effect sizes at the level of activities.
When is the best time to start TMS after a stroke?
This review found significant effects with rTMS both within 3 months after stroke and beyond 3 months, so timing did not separate responders from non-responders in the trials analyzed.
Can TMS replace traditional stroke rehabilitation?
The authors recommend protocolized, time-locked training of the target behavior alongside stimulation in future trials, which points to TMS as an addition to rehabilitation rather than a replacement for it.
Why do the authors call the evidence uncertain?
Because the effect sizes varied across trials in a way none of the tested factors explained, and most of the trials were small. That is the core reason they call for large trials with individual patient data on baseline severity and brain network integrity.
Bottom Line
This review found that rTMS produced significant improvements across a range of motor and cognitive outcomes after stroke, both within and beyond 3 months from the event, and at the level of activities as well as body function. It leans toward hope. The honest limit is that the size of the benefit varies between trials for reasons nobody has yet explained, and the trials are small. I would offer TMS as an adjunct to rehabilitation while being straightforward with patients that the strength of the evidence is not yet where it needs to be.
