Can inter-stride variability capture signs of mixed tone in individuals with cerebral palsy? An exploratory study

Introduction: The identification of dystonia in addition to spasticity (mixed-tone) for individuals with cerebral-palsy (CP) is important, as it can alter clinical management. This study aims to examine if the inter-stride variability of conventionally used gait features can be used for recognizing mixed-tone during gait in individuals with CP. Methods: Retrospective treadmill-based 3D gait-analysis data for 20 individuals (mean ± SD age 10.4 ± 3.3 years) with mixed-tone CP were extracted (mixed-tone-group). A control group of individuals diagnosed with spastic-CP and no dystonia during gait were individually matched (spastic-group). Gait-kinematics were evaluated using Spatiotemporal characteristics and the Gait-Profile-Score (GPS). Selective-motor-control was assessed by the dynamic-motor-control-index (walk-DMC). Inter-stride variability was calculated per-individual using the coefficient-of-variation (CV; (SD/mean)∗100). Results: The mixed-tone-group presented with significantly smaller step-length and higher CV only in spatiotemporal parameters (p < 0.050). After controlling for walking-speed, only the CV for cadence remained significant (p < 0.001); a cut-off of 11.5 % CV in cadence could identify individuals with mixed-tone CP with 65 % sensitivity and 85 % specificity. Interpretation: Larger inter-stride variability was identified for spatiotemporal characteristics in individuals with mixed-tone CP, compared to individuals with spastic CP. Capturing the highly variable movements may be a biomarker of dystonia during gait. Further prospective studies with larger sample size are needed.