Efficacy of Robot-assisted Training on Upper Limb Motor Function After Stroke: A Systematic Review and Network Meta-analysis.

Objective: To systematically evaluate the efficacy of robot-assisted training (RAT) in the treatment of upper limb motor function in patients with stroke, and to compare the efficacy of different types of RAT. Data Sources: We searched 5 English database (PubMed, Web of Science, Embase, Cochrane Library, and Scopus) and 4 Chinese databases (China National Knowledge Infrastructure [CNKI], SinoMed [CBM], VIP, and Wanfang Data) from inception to May 2024. Study Selection: Randomized controlled trials (RCTs) of RAT combined with conventional rehabilitation therapy (CRT) on patients with upper extremity motor dysfunction after stroke. Two independent researches screened 3461 records, resulting in 31 eligible RCTs. Data Extraction: Two independent researches extracted data into a preformed table, including first author, year of publication, baseline characteristics of the literature (including sample size, age, disease duration, interventions, outcome indicators, and adverse reactions) and quality evaluation information. Data Synthesis: Traditional meta-analysis and network meta-analysis were performed using RevMan 5.4 software and Stata 14.0 software, respectively. A total of 31 RCTs were included, involving 1537 patients. The results of traditional meta-analysis showed that in terms of upper limb function, RAT combined with RCT could effectively reduce the elbow flexion Modified Asworth Score (mean difference [MD] -0.63, 95% confidence interval (95% CI) -0.75 to -0.51), improve the Fugl-Meyer Assessment-upper extremity score (MD 5.92, 95% CI 3.52-8.32) and elbow extension angle (MD 16.14, 95% CI 14.56-17.73) in patients with stroke. In terms of activities of daily living, RAT combined with RCT could effectively improve the Modified Barthel Index score (MD 8.03, 95% CI 5.54-10.52) and Functional Independence Measure score (standardized MD 0.95, 95% CI 0.32-1.59) in patients with stroke. The results of network meta-analysis showed that the best probability of improving upper limb motor function was end-effector robot-assisted therapy (EE-RAT)>exoskeleton robot-assisted therapy (Exo-RAT)>CRT, and the best probability of improving daily living ability was Exo-RAT>EE-RAT>CRT. Conclusions: The current evidence showed that RAT combined with routine rehabilitation therapy can effectively improve the upper limb motor function and activities of daily life of patients with stroke. Among them, the EE-RAT had the best effect in improving the upper limb motor function of patients with stroke, and the Exo-RAT had the best effect in improving the ability of daily life of patients with stroke.