Acquisition of Stable Coordination for Reciprocal Upper-Extremity Movement: A Study of Kinematics and Electromyography in Hemiparesis
This thesis examines the nature of impaired motor control in persons with chronic post-stroke hemiparesis. Stable neuromuscular and performance related benchmarks for reciprocal coordination were established in healthy elders (n=11) and these data used as reference for assessment of motor-recovery following therapeutic intervention in persons post-stroke (n=18). All subjects performed a kinematically constrained elbow trajectory-tracking task at three target frequencies. To dissociate contributions of strength and coordination, the task involved a gravity neutral, minimal force configuration. Surface EMG was recorded from the biceps brachii, brachioradialis, and triceps muscles in the exploration of reciprocal coordination. The root-mean-square difference of the vertical displacement between the criterion and performed trajectories was calculated as the primary performance measure (RMSe). Familiarization and test-retest stability were demonstrated through high correlation of within day test trials (p < 0.05) and inter-session mean RMSe (p < 0.05). Signal cross-correlation coefficients (SCC) were calculated as evidence of ability to recreate the required shape (mean SCC=0.985 ± 0.013). To indicate appropriate timing, the time offset to optimal cross-correlation (LAG) was computed and found to not differ from zero (p < 0.01). A suite of EMG metrics for each muscle was computed from the aggregate data across subjects. A rich information timing metric (RITM) was created that correlated strongly with kinematic outcome in healthy elders (|?|=0.829, p=0.042). Further metrics were developed to characterize the timing and extent of activation of individual muscles and agonist antagonist co-contraction in appropriate coordination. These methods were used to probe motor recovery in persons with post-stroke hemiparesis. Hemiparetic subjects were stratified into high and low functioning groups and randomized to receive 12 sessions of Standard Functional Rehabilitation (SFR) followed by a 4 week washout period and 12 additional sessions of dynamic high-intensity resistance training (STR) (CONtrol group). An experimental group (EXP) received STR prior to the washout followed by SFR. RMSe correlated with a subset of the Wolf Motor Function Test focusing on the elbow (p < 0.0001). Both groups demonstrated significant improvement in RMSe at six month follow-up (p < 0.05). SCC was highly correlated to RMSe (|?|=0.814, p < 0.001) indicating that performance improvements were largely due to improvements in producing the requisite trajectory shape. LAG remained uncorrelated to RMSe but localized changes revealed a shift toward more appropriate timing strategies. Significant correlation between RMSe and select EMG metrics (p < 0.05) demonstrate the importance of biceps modulation in reciprocal control. Low RITM scores, relative to normals, demonstrate its effectiveness in detecting pathology. Biceps and triceps co-contraction was found to be uncorrelated to RMSe (|?|=0.139, p < 0.179) indicating that co-contraction is not the primary impairment of coordinated motor performance following stroke. The EMG metrics were expanded to identify a possible strategy contributing to improved motor function. Irrespective of treatment order, the effects of STR produce improved motor control and do not exacerbate co-contraction. Last updated 05/23/2006 |