Robotic-Assisted Shoulder Rehabilitation Therapy Effectively Improved Poststroke Hemiplegic Shoulder Pain: A Randomized Controlled Trial

Objective The purpose of this study was to investigate the therapeutic effects of a newly developed shoulder robot on poststroke hemiplegic shoulder pain. Design Prospective, single-blind randomized controlled trial. Setting Inpatient department of a tertiary university hospital. Participants Hemiplegic shoulder pain patients (N=38) were consecutively recruited and randomly assigned to an intervention or control group. Interventions A newly developed robot was designed to perform joint mobilization and stretching exercises with patients lying in the supine position. Conventional physical therapy directed at both improving upper extremity mechanics and reducing neurologic injury was performed twice per day in both groups. In the intervention group, additional robotic-assisted shoulder rehabilitation therapy was administered for 30 minutes per day, 5 times per week for 4 weeks. Main Outcome Measures The visual analog scale was the primary outcome, and the pain-free passive range of motion of the shoulder joint, the Korean version of the Shoulder Disability Questionnaire, and ultrasonographic grades were the secondary outcomes. The outcomes were evaluated at baseline (T0), postintervention (T1), and a 4-week follow-up (T2). Results Significant time and group interaction effects were found on the visual analog scale, in the abduction passive range of motion, and on the Shoulder Disability Questionnaire (F2,33=16.384, P=.002; F2,33=10.609, P=.012; F2,33=32.650, P=.008, respectively). Significantly higher improvements in these outcome measures were observed in the intervention group than in the control group at T1 after post hoc analysis (P<0.05, all). These improvements were sustained at T2 when the intervention group was compared with the control group (P<.05, all). Conclusions A prototype shoulder rehabilitation robot as an adjuvant therapy improves hemiplegic shoulder pain and self-reported shoulder-related disability.