Impact of the robotic-assistance level on upper extremity function in stroke patients receiving adjunct robotic rehabilitation: sub-analysis of a randomized clinical trial

Background Robotic therapy has been demonstrated to be effective in treating upper extremity (UE) paresis in stroke survivors. However, it remains unclear whether the level of assistance provided by robotics in UE training could affect the improvement in UE function in stroke survivors. We aimed to exploratorily investigate the impact of robotic assistance level and modes of adjustment on functional improvement in a stroke-affected UE. Methods We analyzed the data of 30 subacute stroke survivors with mild-to-severe UE hemiplegia who were randomly assigned to the robotic therapy (using ReoGo System) group in our previous randomized clinical trial. A cluster analysis based on the training results (the percentage of each stroke patient’s five assistance modes of robotics used during the training) was performed. The patients were divided into two groups: high and low robotic assistance groups. Additionally, the two groups were sub-categorized into the following classes based on the severity of UE functional impairment: moderate-to-mild [Fugl-Meyer Assessment (FMA) score ≥ 30] and severe-to-moderate class (FMA < 30). The outcomes were assessed using FMA, FMA-proximal, performance-time in the Wolf motor function test (WMFT), and functional assessment scale (FAS) in WMFT. The outcomes of each class in the two groups were analyzed. A two-way analysis of variance (ANOVA) was conducted with robot assistance level and severity of UE function as explanatory factors and the change in each outcome pre- and post-intervention as the objective factor. Results Overall, significant differences of the group × severity interaction were found in most of the outcomes, including FMA-proximal (p = 0.038, η 2 = 0.13), WMFT-PT (p = 0.021, η 2 = 0.17), and WMFT-FAS (p = 0.045, η 2 = 0.14). However, only the FMA score appeared not to be significantly different in each group (p = 0.103, η 2 = 0.09). Conclusion An optimal amount of robotic assistance is a key to maximize improvement in post-stroke UE paralysis. Furthermore, severity of UE paralysis is an important consideration when deciding the amount of assistance in robotic therapy. Trial registration Trial enrollment was done at UMIN (UMIN 000001619, registration date was January 1, 2009)