Association of a kinematic measure of upper limb reaching and StartReact in chronic hemiparetic stroke

Abstract

Background: After stroke, recovery of upper limb reaching movements may partly depend on the level of activation of the reticulospinal tract (RST), but few clinical studies have explored this. Here we examined the association between the strength of reticulospinal connections and extent of reaching in post-stroke patients. Methods: Fifteen patients (all male) with right hemiparesis who had suffered a stroke at least six months prior to the assessment were selected based on predefined selection criteria. Video recordings of the reaching task from the Action Research Arm Test (ARAT) were processed for two-dimensional kinematic analysis using the markerless motion tracking software DeepLabCut. We defined a novel Index of Elbow Extension (IoEE), and validated it by comparison between simultaneously obtained two- and three-dimensional datasets in healthy people. Strength of reticulospinal outputs was estimated using the ‘StartReact’ paradigm, which measures the speeding up of reaction time by a loud sound cue. Results: We observed a significant negative correlation between the IoEE and StartReact (rho = 0.9, p < 0.05). There was no correlation in this cohort between ARAT and StartReact. Conclusion: We speculate that the negative correlation between reaching performance and StartReact is a consequence of the variable compensatory activation of the reticulospinal tract (RST) in response to different levels of initial damage. This study reinforces the application of freely available computer vision technology for assessment of two- dimensional kinematics in a clinical scenario.