BCIwiki (bciwiki.com) — Researchers affiliated with Southwest Medical University and Zigong Vocational and Technical College, China, conducted a meta-analysis of BCI rehabilitation after stroke, published in the Journal of NeuroEngineering and Rehabilitation on August 24, 2026. The authors reported 16 randomized trials involving 1,761 patients, comparing BCI rehabilitation with conventional rehabilitation.
The researchers searched seven databases and used FMA-UE as the primary outcome. BCI rehabilitation had an SMD of 0.49 versus conventional rehabilitation, with a 95% confidence interval of 0.26–0.73. SMD expresses a standardized between-group difference, not a gain of 0.49 raw points or a 49% improvement.
For ARAT, the SMD was 0.53 with a 95% confidence interval of 0.14–0.91. The authors also reported gains in neurophysiological markers and daily living performance assessed by the Modified Barthel Index. Outcomes pooled subsets of the trials; 1,761 is not the sample size for every measure.
Subgroup analyses favored motor recovery in patients in the subacute phase. The authors suggested that the timing of post-stroke neural remodeling might explain this finding, but this was not direct verification of the mechanism. They also reported improvements with both 10–15 and 20–30 sessions. Although the latter schedule had a relatively larger effect size, the difference between schedules was not statistically significant. No significant differences were detected between intervention approaches either; this does not establish equivalent efficacy.
Fourteen of the 16 trials were rated at high risk of bias for participant and personnel blinding, and most trials had fewer than 30 participants per group. Small subgroups limit interpretation; subgroup associations do not determine an individual patient’s optimal number of sessions.