Robot-Assisted Gait Training for Functional Mobility After Spinal Cord Injury: A Systematic Review and Meta-Analysis

Background: Persistent limitations in walking, functional mobility, and balance restrict independence and participation after spinal cord injury (SCI). Robot-assisted gait training (RAGT) can deliver repetitive and intensive gait practice, but its advantages over non-robotic rehabilitation remain uncertain. Objective: This systematic review and meta-analysis evaluated whether RAGT improves functional mobility and balance compared with study-specific control interventions in adolescents and adults with SCI. Methods: PubMed, PEDro, Cochrane Library, Scopus, Web of Science, EBSCOhost, ScienceDirect, and Wiley Online Library were searched from inception to March 2026. RAGT was defined as gait training delivered by a powered or actuated electromechanical system that actively guided or assisted lower-limb joint movement during treadmill-based or overground walking. Randomized controlled trials involving individuals aged ≥15 years with SCI were eligible. Quantitative synthesis was restricted to clinically compatible RAGT-versus-non-robotic-control contrasts; balance was synthesized narratively. Methodological quality was assessed using the PEDro scale, and certainty was evaluated using GRADE. Results: Twelve RAGT trials involving 362 participants met the operational definition. Three trials (n = 67) contributed to the primary functional mobility meta-analysis. The pooled effect was small and non-significant (SMD = 0.27, 95% CI −0.30 to 0.85; I2 = 27%). All retained trials were classified as high quality using the prespecified PEDro threshold (≥6/10); GRADE certainty remained low for functional mobility and very low for balance. Balance evidence was sparse and was synthesized narratively. Conclusions: Current evidence does not demonstrate that RAGT is superior to non-robotic control interventions for improving functional mobility after SCI. RAGT may be a feasible adjunct for standardized gait practice, but its comparative effectiveness and effect on direct postural balance remain uncertain.