Author:

Abstract:

Introduction Three-dimensional motion analysis (3DMA) thus far mainly focused on feature analysis of waveforms by extracting scalars or performing point-by-point comparison. However, these analyses do not take into account the dependency between the time-points of the waveforms. Statistical Parametric Mapping (SPM) has been shown to offset those limitations by using random field theory to identify field regions taking into account the co-variance among the waveform time-points. Here, we compared upper limb waveforms between typically developing children (TDC) and children with unilateral cerebral palsy (uCP) during a reach-to-grasp task. Patients and methods Twenty TDC (mean age 10y11m (SD 2y11m), 2 left-handed) and 21 children with uCP (mean age 11y1m (SD 2y1m), 10 right-hemiplegia) were included. Non-dominant and affected upper limb were tested. Children sat in a custom-made chair and were instructed to grasp a vertical cylinder, positioned at arm-length. SPM was used to compare the waveforms of both groups in 13 angles. Results SPM{t} tests depicted differences between groups for wrist flexion (uCP>TDC during 89% of the movement cycle, p<0.001), elbow flexion (uCP>TDC during 56%, p<0.001), shoulder elevation (TDC>uCP during 37%, p<0.001), elbow supination (TDC>uCP during 43%, p<0.001), scapula protraction (uCP>TD during 100%, p<0.001) and scapula medial rotation (uCP>TDC during 9%, p=0.04). Conclusion Children with uCP have abnormal upper limb movement patterns over a large percentage of the 3DMA waveforms for the wrist, elbow, shoulder and scapula compared to TDC during a reach-to-grasp task. These results show the additional value of SPM-analysis to further understand upper limb movement pathology in uCP.