Comparison of Upper Limb Motor Therapy Outcomes in Post-Stroke Patients: A Randomized Controlled Trial of Therapist-Led Versus Robot-Assisted Rehabilitation

Abstract High-dose, task-specific upper limb training is essential for post-stroke motor recovery. EMG-triggered robot-assisted therapy enables intention-driven, high-repetition practice; however, evidence from randomized trials directly comparing this approach with dose-matched conventional rehabilitation remains limited. This single-blind randomized controlled trial compared short-term outcomes of EMG-triggered robot-assisted upper limb training using the Luna EMG system with standardized therapist-led therapy in patients after stroke. Fifty-eight patients in the early post-stroke phase were randomly assigned to robot-assisted therapy (n = 29) or therapist-assisted therapy (n = 29). Both groups received identical upper limb training with respect to duration, intensity, frequency, and number of repetitions over six weeks within the same inpatient rehabilitation program. Outcomes were assessed at baseline, weeks three and six, and at three-week follow-up using the Fugl–Meyer Assessment for Upper Extremity, Box and Block Test, Modified Ashworth Scale, Numerical Rating Scale, and EQ-5D-5L. Both groups showed significant within-group improvements in motor function, pain, and quality of life over time. When therapy dose is strictly controlled, no statistically significant between-group differences were observed. A single between-group difference was observed in the EQ-5D-5L domain of usual activities at follow-up, favoring the robot-assisted group. Trial registration: ClinicalTrials.gov, NCT07002463. Registered May 30, 2025 (retrospectively registered)