Lower-limb joint-coordination and coordination variability during gait in children with cerebral palsy

Background: Children with cerebral palsy present with poor motor control, altering their ability to perform tasks such as walking. Continuous relative phase analysis is a popular method to quantify motor control impairments via inter-joint coordination and coordination variability; however, it has not been explored in children with cerebral palsy. Methods: 45 children with cerebral palsy and 45 typically developing children walked while fit with retroreflective markers. Continuous relative phase analysis for knee-hip and ankle-knee joint pairs quantified inter-joint coordination and coordination variability. The Gait Profile Score estimated gait pathology. Group differences were assessed with unpaired t-tests for coordination amplitude and variability (knee-hip, ankle-knee) across gait events. For the cerebral palsy group, correlations assessed the relation between the gait profile score and coordination metrics. Findings: The cerebral palsy group showed more in-phase patterns for knee-hip coupling compared to the typically developing group (initial contact, loading response, mid-stance, terminal swing) (p <= 0.03). The cerebral palsy group showed lower knee-hip coordination variability (mid-stance, mid-swing) (p <= 0.037) and lower ankle-knee coordination variability (initial contact, loading response, terminal swing) (p < 0.001). The gait profile score correlated weakly to moderately (r = [0.323-0.472]), and negatively with the knee-hip inter-joint coordination (initial contact, loading response, mid-stance, terminal swing) (p <= 0.042). Interpretation: Children with cerebral palsy showed a more in-phase gait strategy during challenging transitional gait cycle phases (beginning and end) and less flexible and adaptable motor behaviors. Moreover, the correlation between in-phase joint patterns and increased gait deviations (gait profile score) reinforces the relevance of coordination analysis to assess motor control impairment.