A reflex technique to measure presynaptic inhibition in cerebral stroke

The purpose of this study was to evaluate the feasibility of reflex excitability measurement techniques in the partial measurement of spasticity related to cerebral stroke. Techniques involved the testing of the soleus H-reflex at specific ankle positions during passive dorsiflexing movements with and without background plantarflexing contractions; conditions attempted to simulate the terminal stance phase of gait. Testing of 12 stroke subjects, having cerebrovascular lesions related only to occlusion of the middle cerebral artery, demonstrated significantly (p < 0.01) less inhibition of the H-reflex during passive ankle dorsiflexion compared to 12 matched, healthy controls. However, evocation of the H-reflex during a low-level, voluntary plantarflexing contraction concomitant with passive dorsiflexion, did not reflect a statistical difference between the two groups. The two conditions were thought to each represent measures of faulty presynaptic inhibition as indicators of cerebral spasticity. A Chi-square calculation of sensitivity for the passive ankle movement without background plantarflexing contraction condition, was shown to significantly differentiate (p < 0.05) between the stroke and normal groups. A positive, but weak, correlation was found for stroke subjects between this reflex measure and the Ashworth clinical measure of spasticity (r = 0.49). Although stroke subjects exhibited increased joint stiffness when the full range of passive ankle dorsiflexion movement was considered, in comparison to the matched healthy control subjects, no significant increase in passive stiffness was found at the joint position of the reflex evocation. Size of the cerebral lesion, as determined from CT or MRI scan, was not related to the spasticity measures. Therefore, in a homogeneous stroke sample, a component of cerebral spasticity i.e., faculty Ia presynaptic inhibition, has been measured during a simulated functional movement in the lower extremity and was shown to differentiate this group from a matched, healthy, control sample. Joint stiffness did not contaminate the measures.