Antigravity Versus Body-Weight-Supported Treadmill Training in Lower-Limb Arthroplasty Rehabilitation: A Randomized Controlled Pilot Trial

Objective: To evaluate the feasibility of adding antigravity treadmill training (ATT) or harness-based body-weight-supported treadmill training (BWSTT) to standard inpatient rehabilitation after primary hip or knee arthroplasty and to explore preliminary effects on osteoarthritis-related outcomes, balance, and psychological status. Methods: In this single-center, assessor-blinded pilot randomized trial, 60 adults within 3 months after primary hip or knee arthroplasty for osteoarthritis were allocated 1:1:1 to ATT, BWSTT, or standard inpatient rehabilitation over 6 weeks. Feasibility outcomes included recruitment, retention, and adherence. ATT and BWSTT additionally included unloading-based treadmill gait training using lower-body positive pressure or a harness system. Exploratory clinical outcomes included WOMAC total and subscale scores, analyzed using baseline-adjusted ANCOVA estimated marginal means. Secondary exploratory outcomes were BBS, FES-I, PHQ-9, and PSS-10. Results: Post-intervention data were available for 47 participants, with differential attrition across groups. Exploratory ANCOVA suggested between-group differences for WOMAC total (p = 0.004) and WOMAC function (p < 0.001). Compared with standard rehabilitation, ATT showed lower adjusted WOMAC total and function scores (both p < 0.01). ATT versus BWSTT contrasts for WOMAC total and function were statistically significant in the primary exploratory model but attenuated after hypertension adjustment. Exploratory signals were also observed for BBS and FES-I, although FES-I was less robust in sensitivity analysis. No clear between-group differences were observed for WOMAC pain, stiffness, PHQ-9, or PSS-10. No formal multiplicity adjustment was applied across exploratory endpoints. Conclusions: In this single-center pilot randomized trial, ATT suggested preliminary function- and balance-related signals that require confirmation in adequately powered multicenter trials.