500 Section - Robot-assisted therapy and upper-limb spasticity after stroke: systematic review and meta-analysis of randomized controlled trials

Objectives : Robot-assisted therapy (RAT) is widely used in post-stroke upper-limb rehabilitation, but its impact on spasticity remains inconsistent across trials. The aim is evaluate whether Robotic Assisted therapy (RAT) reduces upper-limb spasticity compared with non-robotic comparators and to synthesize how effects vary across different joints and intervention characteristics. Secondary objectives included safety and functional outcomes.

Design: We performed a systematic review and meta-analysis of randomized controlled trials in adults after stroke comparing RAT with conventional therapy. Spasticity was synthesized primarily using Modified Ashworth Scale (MAS). Secondary outcomes included upper-limb motor impairment (FMA-UE), activities (Barthel Index), quality of life, and adverse events. Risk of bias was assessed using the Cochrane tool; certainty of evidence was rated with GRADE.

Results: We included 29 trials. Overall, RAT produced reductions in spasticity, with effects more consistently observed for elbow (MD −0.25) and wrist (MD −0.25) than for shoulder (MD −0.13). When spasticity was reported as composite or non–joint-specific MAS (MD −0.13). Multi-joint devices demonstrated a modest benefit (SMD −0.26), whereas hand-only devices did not (SMD −0.22). Dose-matched trials retained a small effect (SMD −0.17). Motor impairment improved significantly (MD 1.97) whereas no significant improvement was observed for Barthel index (MD = 1.35). Quality-of-life effects were limited by sparse reporting. Adverse events (OR 1.18) were infrequent and did not indicate increased risk with RAT. GRADE certainty for was generally low to very low, mainly due to risk of bias and heterogeneity.

Conclusion: RAT appears safe and may yield small reductions in upper-limb spasticity, most evident when spasticity is measured and reported at the joint level, with clearest effects observed in elbow flexors. Future trials should standardize joint-specific muscle tone reporting and provide detailed protocol descriptions to enable stronger spasticity-focused comparisons.