Interventions combined with task-specific training to improve upper limb motor recovery following stroke: a systematic review with meta-analyses

Background: Upper limb (UL) hemiparesis is a common, disabling and persistent problem, and a major contributor to poor well-being and quality of life in persons after stroke. Conventional UL rehabilitation has had limited success. Novel combined interventions are being investigated in an effort to stimulate greater recovery. Objective: To identify and assess the efficacy of interventions combined with task-specific training aimed at UL motor recovery after stroke. Methods: A systematic search was undertaken in databases including MEDLINE, MEDLINE InProcess, EMBASE, AMED, CINAHL, OTseeker, and PEDro. Key inclusion criteria were: peerreviewed articles published in English, adults after stroke, and an intervention combined with task-specific training targeted to improve motor function and/or impairment of the UL following stroke. Findings from included studies were synthesized qualitatively and metaanalyzed where there was sufficient homogeneity. Results: From 3494 citations identified, 120 papers (72 randomized controlled trials and 4 pseudo-randomized controlled trials) were included. Adjunctive interventions (21 categories) identified included electrical stimulation, transcranial magnetic stimulation, robotic devices, mental practice, action observation, trunk restraint, virtual reality, and resistance training. Of the interventions meta-analyzed, only peripheral nerve stimulation demonstrated small additional benefits over those of task-specific training alone for UL impairment, as measured by the Fugl-Meyer scale (MD 2.69, 95% 0 1.12, 4.26). Several individual studies found benefits for other interventions combined with task-specific training, but further investigations are needed to provide more comprehensive evidence of their efficacy. Conclusion: To date, there is little evidence that adding another intervention to TST confers additional benefits and therefore there is no evidence to guide rehabilitation professionals. Further research is required as heterogeneity of studies limited ability to conduct meta-analyses.