BCIwiki (bciwiki.com) — Brain-computer interface (BCI)-coupled robotic rehabilitation ranked highest of 4 interventions for restoring arm function after stroke in a systematic review and network meta-analysis of 22 independent study cohorts with 727 participants, researchers at Beijing Tiantan Hospital, Capital Medical University, and Hainan Provincial People's Hospital reported in Neurological Sciences on September 25, 2026 (PMID 42789169; DOI 10.1007/s10072-026-09375-9).
The team searched 11 databases from inception to March 31, 2026, and included 23 reports of randomized controlled trials in adults with post-stroke upper-limb motor impairment. The primary outcome was the Fugl-Meyer Assessment of Upper Extremity (FMA-UE); secondary outcomes were the Action Research Arm Test (ARAT), Wolf Motor Function Test (WMFT) and Modified Barthel Index (MBI). The network showed no significant inconsistency, and a consistency model was used.
The BCI-robot approach had a P-score of 0.992, followed by robot-assisted therapy (0.641), motor imagery (0.240) and conventional rehabilitation (0.127). In direct comparisons, it improved FMA-UE by a mean 6.80 points over conventional rehabilitation (95% CI 4.29 to 9.30). The point estimate exceeded the 5.25-point minimal clinically important difference (MCID), but the lower confidence limit did not. Against robot-assisted therapy, the gain was 2.04 points (95% CI 0.29 to 3.78), with the point estimate and entire interval below the MCID.
On secondary outcomes, the BCI-robot approach improved WMFT (MD 8.94, 95% CI 5.22 to 12.65) and MBI (MD 3.37, 95% CI 0.10 to 6.63) versus conventional rehabilitation. An MBI benefit versus robot therapy (MD 8.61, 95% CI 1.76 to 15.46) did not hold up in sensitivity analysis, and no significant advantage emerged for ARAT. The approach stayed favorable over conventional rehabilitation in both the under-3-month and 3-month-or-longer follow-up subgroups, but showed no follow-up advantage over robot therapy. Exploratory analyses suggested its added benefit over robot therapy was clearer in studies delivering more than 4 sessions per week.
The authors concluded the approach may offer a clinically important benefit over conventional rehabilitation, though the size remains uncertain, while its added benefit over robot-assisted therapy was small and unlikely to be clinically important. Evidence on longer-term effects and on activity-level and daily-function outcomes is limited, they said, and the findings should be interpreted cautiously.