Neurophysiological and kinematic correlates of improved motor-function in complex therapy movements in chronic stroke

The most common outcome post-stroke is motor impairment which is typically assessed using clinical tools mainly based on categorical and subjective scoring. This thesis investigated longitudinal changes using quantitative measures of muscle activation and movement kinematics to understand therapy-induced improvements of motor-function in chronic stroke. Electromyography (EMG) and tri-axial accelerometry were recorded from 6 sensors on the more-affected upper body of 24 patients during early- and late-therapy of an intensive 14-day program of Wii-based Movement Therapy, and for a subset of 13 patients at 6-month follow-up. Patients were classified according to residual voluntary motor capacity with low, moderate or high motor-function. Clinical motor-function was assessed pre- and post-therapy and at follow-up. Neurophysiological parameters including EMG area under the curve and muscle synergies, and movement kinematics were measured to investigate the improved motor-function. Finally, the correlates and potential predictors of therapy-induced and longitudinal changes were investigated using a multivariate model. Clinical assessments showed improved motor-function and independence in everyday life over time. The pattern of EMG change by late-therapy was variable within and between classifications of motor-function. Muscle synergy analysis revealed fewer synergies for patients with low compared to those with high motor-function at early-therapy, with a pattern of increased synergies by late-therapy. Movement acceleration magnitude increased over time but was significant only at proximal sensors; there was an effect of motor-function on acceleration at distal sensors. The multivariate model demonstrated that depression scores could predict 31% of improved motor-function, while baseline muscle synergy counts predicted 33% of the increase in acceleration magnitude. Different patterns of EMG and kinematic changes were observed according to motor-function level despite heterogeneity within each level that was not evident with clinical assessments. The combination of EMG and kinematic changes suggest improved motor control which was reflected in significantly greater independence in everyday life. This thesis demonstrates that patients in chronic stroke have the capacity to improve, and that there is no one pattern of improvement. The different patterns of improvement can be revealed by quantitative neurophysiological and kinematic measures but not clinical tools. Pooling data may obscure the changes in chronic stroke