Validity of movement smoothness metrics for upper limb reaching movements in people with moderate to severe subacute stroke

Abstract

Abstract Background: Movement smoothness is a potential kinematic biomarker of upper extremity (UE) movement quality and recovery after stroke; however, the validity of available smoothness metrics has not been determined in this group. We aimed to measure the responsiveness to change, reliability, convergent and criterion (concurrent and predictive) validity of several smoothness metrics. Methods: This ancillary study of the REM-AVC trial included 31 participants with hemiparesis in the subacute phase of stroke (median time since stroke: 38 days). Assessments performed at inclusion (Day 0, D0) and at the end of a rehabilitation program (Day 30, D30) included the UE Fugl Meyer Assessment (UE-FMA), the Action Research Arm Test (ARAT), and 3D motion analysis of the UE during three reach-to-point movements at a self-selected speed to a target located in front at shoulder height and at 90% of arm length. Four smoothness metrics were computed: a frequency domain smoothness metric, spectral arc length metric (SPARC); and three temporal domain smoothness metrics (TDSM): log dimensionless jerk (LDLJ); number of submovements (nSUB); and normalized average rectified jerk (NARJ). Results: At D30, movements were significantly shorter in duration and trajectory, straighter, faster and smoother (highest effect size for smoothness change: SPARC, 0.76). Intra-subject coefficients of variation were <10% for SPARC and LDLJ, and >30% for nSUB and NARJ. SPARC was strongly correlated with all TDSM, and the TDSM were very strongly correlated with each other. Concurrent validity at D0 was higher for SPARC than TDSM in terms of correlation with proximal UE-FMA, ARAT and index of curvature (r: 0.56, 0.68 and 0.87 respectively, all p<0.01). At D30, concurrent validity was similar between all smoothness metrics. TDSM were very strongly correlated with movement duration at D0 and D30. Finally, SPARC had the highest predictive validity among the four smoothness metrics. Conclusions: Of the four smoothness metrics, the SPARC had the highest sensitivity to change, reliability, construct and criterion validity for the evaluation of the UE in people with moderate to severe stroke in the subacute phase. Among the TDSM, LDLJ was the most reliable. Trial Registration: NCT01383512, https://clinicaltrials.gov/, June 27, 2011