F1000Res. 2026 May 19;15:762. doi: 10.12688/f1000research.180596.1. eCollection 2026.
ABSTRACT
INTRODUCTION: Stroke is a primary contributor to global adult morbidity, frequently resulting in debilitating gait disturbances that impede independence. While standard neurorehabilitation remains essential, the advent of robot-assisted gait training (RAGT) provides a platform for high-intensity, reproducible interventions designed to stimulate neuroplasticity. This systematic review evaluated the literature from 2015 to 2025 to determine the influence of RAGT on neural reorganization and functional outcomes in stroke survivors.
METHODOLOGY: A systematic search of MEDLINE, PubMed, Scopus, and Cochrane Library was performed for studies published between 2015 and 2025. The inclusion criteria were randomized controlled trials, systematic reviews, and meta-analyses investigating RAGT in acute, subacute, and chronic stroke patients. The evaluation metrics included clinical functional scales (e.g., the Berg Balance Scale and the 10-Meter Walk Test) and neurophysiological markers (e.g., fMRI, fNIRS, QEEG, and BDNF).
RESULTS: Analysis of 23 randomized trials and multiple meta-analyses indicated that RAGT combined with conventional therapy yielded significant improvements in gait speed (standardized mean difference [SMD] = 0.47) and balance (mean difference [MD] = 4.58). Neuroimaging revealed increased activation of the primary motor cortex and supplementary motor areas. Electrophysiological data confirmed reductions in the Power Ratio Index, suggesting normalized cortical activity. End-effector systems have demonstrated superior efficacy in the subacute phase compared to exoskeletons.
CONCLUSION: RAGT serves as a potent driver of cortical and spinal neuroplasticity by facilitating high-repetition task-specific stimuli. Early intervention during the subacute phase maximizes functional recovery. Future research should prioritize standardizing the dosing and longitudinal monitoring of neural connectivity. Trial Registration Status: Not Applicable.
PMID:42817934 | PMC:PMC13620352 | DOI:10.12688/f1000research.180596.1