Abnormal Muscular Coordination in the Lower Extremity of Cerebral Palsy Subjects Darryl G. Thelen, Scott A. Riewald*, Scott L. Delp Biomechanical Engineering Division; Stanford University, Stanford, CA *Sensory Motor Performance Program, Rehabilitation Institute of Chicago, Chicago, IL Email: dthelen@stanford.edu , http://www.stanford.edu/group/nmbl/ INTRODUCTION Subjects with cerebral palsy frequently walk with excessive adduction and internal rotation of the hip during stance (Sutherland et al. 1978), when extension of the limb is required. This pattern resembles the “extension synergy” commonly observed in patients with hemiplegia following stroke (Sawner et al. 1992). The extension synergy is thought to arise from primitive neural control that couples the motions of knee extension to hip extension, adduction, and internal rotation. While these combined motions are often observed in subjects with cerebral palsy, it is not known if neural coupling between these degrees of freedom contributes to the observed motions. The purpose of this study was to determine if maximum isometric exertions of the hip or knee extensors produced the other components of the extension synergy in subjects with cerebral palsy. METHODS Nine subjects with spastic cerebral palsy and six neurologically intact control subjects participated in this study. Subjects performed maximum isometric lower- extremity exertions while seated in a chair that provided support for the pelvis and restrained the upper body (Figure 1). A cast was placed around the ankle and was fixed to a six degree of freedom load cell. The load cell measurements were used along with measured limb geometry to determine the three-dimensional moments about the hip and flexion-extension moments about the knee. Bipolar surface electrodes were used to monitor the activities of eight muscles: gluteus maximus, gluteus medius, adductor magnus, rectus femoris, vastus lateralis, vastus medialis, semitendinosus and biceps femoris long head. Subjects were asked to generate either a maximum hip extension moment or a maximum knee extension moment. They were free to use whatever strategy they desired to maximize this primary moment, which was displayed to them during a 4 second trial. The maximum primary moment, averaged over a moving 300 ms window, was determined in post-hoc analysis. The associated moments about the other degrees of freedom and muscle activities over this same 300 ms window were used to characterize the coordination strategy adopted by the subject. To facilitate comparisons between subjects, the moments for each subject were normalized to the maximum moment they could generate for each degree of freedom. The EMG activity of each muscle was normalized to the maximum EMG recorded over all trials for each subject. A repeated measures ANOVA was used to test for significant differences between the control and CP subjects in the normalized moments and EMG. EMG DATA Load Cell Figure 1. Experimental setup.