Non-invasive brain stimulation for stroke-related motor impairment and disability: an umbrella review of systematic review and meta-analysis

Introduction: Non-invasive brain stimulation (NIBS) techniques, particularly repetitive transcranial magnetic stimulation (rTMS) and transcranial direct current stimulation (tDCS), have shown potential in stroke rehabilitation. However, systematic reviews often reach conflicting conclusions, highlighting the need for an umbrella review. Objective: To synthesize, based on the principal domains of the International Classification of Functioning, Disability and Health (ICF) framework, the best available evidence on the effectiveness and safety of NIBS for improving motor impairment and disability after stroke. Methods: We conducted an umbrella review (PROSPERO: CRD42021239577) that included meta-analyses of controlled trials investigating NIBS effects in stroke survivors, retrieved from PubMed/MEDLINE from February 2020 to July 2025. Methodological quality was appraised using AMSTAR-2 and certainty of evidence using GRADE. Outcomes were mapped to ICF body structure/function and activity domains. Results: Fifty-six studies were included (2–48 primary trials each; 54–1,654 participants per meta-analysis). All included studies evaluated only rTMS and tDCS; no meta-analyses of other NIBS modalities met inclusion criteria. Methodological quality was high or moderate in 85.7% of the meta-analyses. Certainty of evidence was low or very low for 14/50 studies; only one rTMS review provided moderate-certainty evidence for activities of daily living. rTMS showed improvement in activities of daily living (ADL; SMD = −0.82, 95% CI −1.05 to −0.59), upper-limb motor impairment (SMD = −0.32, 95% CI −0.55 to −0.09) and variable effects on mobility from small (SMD = −0.35, 95% CI −0.45 to −0.24) to large (SMD = −0.97, 95% CI −1.28 to −0.66). tDCS was supported by very-low-certainty evidence: small effects were found for motor impairment (SMD = −0.22, 95 % CI −0.32 to −0.12) and upper-limb activity (SMD = −0.31, 95% CI −0.55 to −0.01), while a much smaller subset of trials suggested a large effect (SMD = −1.54, 95% CI −2.78 to −0.29). Effects on ADL and mobility with tDCS were inconsistent and generally non-significant. Conclusion: rTMS was more frequently associated with moderate to large effect sizes for body structure/function outcomes, particularly general neurological function. In contrast, tDCS demonstrated small effects on motor recovery, though evidence certainty was very low due to heterogeneity, imprecision, and protocol variability. Within the activity domain, NIBS showed modest effects, with rTMS showing more consistent benefits for ADL. tDCS effects were generally limited and supported by low to very low certainty of evidence.