Evidence of rTMS for Motor or Cognitive Stroke Recovery: Hype or Hope?

BACKGROUND: Evidence of efficacy of repetitive transcranial magnetic stimulation (rTMS) for stroke recovery is hampered by an unexplained variability of reported effect sizes and an insufficient understanding of mechanisms of action. We aimed to (1) briefly summarize evidence of efficacy, (2) identify critical factors to explain the reported variation in effects, and (3) provide mechanism-based recommendations for future trials. METHODS: We performed a systematic review of the literature according to Cochrane and PRISMA Protocols. We included trials with ≥10 patients per treatment group. We classified outcome measures according to the International Classification of Functioning, Disability, and Health. Meta-analysis was done when at least 3 trials were reported on the same construct. In case of significant summary effect sizes with significant heterogeneity, we used sensitivity analyses to test for correlations and differences between found individual effect sizes and possible effect modifiers such as patient-, repetitive transcranial magnetic stimulation-, and trial characteristics. RESULTS: We included 57 articles (N=2595). Funnel plots showed no publication bias. We found significant effect sizes at the level of body function (upper limb synergies, muscle strength, language functioning, global cognitive functioning, visual/spatial inattention) with repetitive transcranial magnetic stimulation within or beyond 3 months after stroke. We also found significant effect sizes at the level of activities. We found no subgroup differences or significant correlations between individual summary effect sizes and any tested possible effect modifier. CONCLUSIONS: Repetitive transcranial magnetic stimulation holds the potential to benefit a range of motor and cognitive outcomes after stroke, but the evidence of efficacy is challenged by unexplained heterogeneity across many small sampled trials. We propose large trials with the collection of individual patient data on baseline severity and brain network integrity with sufficiently powered subgroup analyses, as well as protocolized time-locked training of the target behavior. Additional neurophysiological and biomechanical data may help in understanding mechanisms and identifying biomarkers of treatment efficacy. REGISTRATION: URL: https://www. CLINICALTRIALS: gov; Unique identifier: CRD42022300330.