Effects of Lokomat-based robotic-assisted gait training and physical therapy on gait quality in children with cerebral palsy: A severity-stratified analysis.

Abstract

Background Robotic-assisted gait training (RAGT) is increasingly used in paediatric rehabilitation to promote intensive walking practice in children with cerebral palsy (CP). However, evidence regarding its effectiveness on gait quality remains limited. Research question This retrospective observational study aims to: i) assess the effects of Lokomat-based RAGT and physical therapy (PT) on spatiotemporal gait parameters and overall gait quality; (ii) investigate whether treatment response differs by motor impairment severity, classified by the Gross Motor Function Classification System (GMFCS levels I, II, III). Methods The study included 52 children with CP (GMFCS levels I-III) who underwent a 4-week, 20-day rehabilitation program: each day, participants attended one Lokomat-based RAGT session and one conventional PT session, for 20 RAGT and 20 PT sessions overall. Instrumented gait analysis was performed before (T0) and after treatment (T1). Spatiotemporal parameters, Gait Profile Score (GPS), and Gait Variable Scores (GVS) were computed and compared across time and GMFCS subgroups. Gross motor function was assessed using the Gross Motor Function Measure-88 (GMFM-88) total score as complementary measure. Results GMFM-88 total scores and most spatiotemporal parameters did not change significantly. Step length significantly increased in the overall CP group. GPS decreased post-treatment, with greater relative improvements in children with greater baseline gait deviation. Significant kinematic changes varied by GMFCS level, shifting from predominantly proximal joints in GMFCS-I to mainly distal joints in GMFCS-III. Significance Lokomat-based RAGT combined with PT induces severity-dependent and joint-specific kinematic improvements in children with CP. These changes were detected by instrumented gait analysis and summary gait indices, whereas the GMFM-88 total score, a global gross motor function measure, may be less sensitive to gait-specific changes over 4 weeks. Therefore, these findings should be interpreted as evidence of gait-specific kinematic changes rather than proof that the intervention improves gait more than overall gross motor function.