Robot-assisted gait training in people with subacute and chronic stroke: a randomized controlled trial

Introduction The effectiveness of robot-assisted gait training (RAGT) after stroke remains uncertain, particularly regarding its role across different recovery stages. This single-blind, randomized, parallel-group trial examined whether a 4-week inpatient program of end-effector-based RAGT plus conventional physiotherapy (RAGT+CPT) improved walking ability and basic activities of daily living (ADL) more than conventional physiotherapy alone (CPT), and whether effects differed by stroke phase. Methods One hundred people with subacute (>7 days to ≤6 months) or chronic (>6 months) stroke were stratified by stroke phase and randomly allocated (1:1) to RAGT+CPT or CPT. Interventions were delivered four times weekly for 4 weeks (45–50 min RAGT/CPT + 25 min CPT). Primary outcomes were walking ability (Functional Ambulation Categories) and basic ADL (Barthel Index). Secondary outcomes included walking speed and distance, functional mobility, concerns about falling, fatigue, health-related quality of life (HRQoL), intervention acceptability, and RAGT gait parameters. Outcomes were assessed at baseline, post-intervention, and follow-up (Week 12). Analyses followed the intention-to-treat principle. Results Median (range) baseline FAC was 4 (RAGT+CPT: 0–4; CPT: 1–4) in the subacute groups and 3 (RAGT+CPT: 0–4; CPT: 1–4) in the chronic groups. Cumulative link mixed models demonstrated significant improvements over time in walking ability and basic ADL, with no evidence of differential treatment effects between RAGT+CPT and CPT. Improvements in walking endurance were greater in subacute- than chronic-phase participants. Additional analyses adjusting for baseline ambulatory function did not materially alter the intervention-related conclusions. Self-perceived HRQoL assessed with the EQ-5D-5L Visual Analogue Scale improved significantly more following RAGT+CPT than CPT, whereas most other secondary outcomes were comparable between interventions. No study-related serious adverse events occurred. Discussion This trial found no clear superiority of RAGT+CPT over CPT for walking ability or basic ADL. Both interventions produced meaningful improvements over time across both stroke phases, supporting intensive inpatient rehabilitation beyond the early recovery phase. Although objective functional outcomes were comparable between interventions, RAGT+CPT was associated with greater improvement in self-perceived HRQoL. Overall, RAGT may be considered a complementary rehabilitation approach rather than a universally superior alternative to conventional physiotherapy. Clinical trial registration https://drks.de/search/en/trial/DRKS00029910/details , identifier DRKS00029910.