Clinical factors associated with motor imagery ability in individuals with subacute stroke.

OBJECTIVE: To examine the associations of motor recovery stage, proprioceptive function, and working memory with motor imagery ability in patients with subacute stroke. DESIGN: Multicentre cross-sectional study. SUBJECTS/PATIENTS: Forty patients with subacute stroke who had complete data. METHODS: Motor imagery ability was assessed using mental chronometry during a Purdue Pegboard task. Associations with motor recovery stage, proprioceptive function, and working memory were examined using correlation and multiple regression analyses. An exploratory linear support vector machine analysis was performed to classify participants into groups with higher and lower motor imagery ability using these variables. RESULTS: Motor imagery ability was significantly associated with motor recovery stage and proprioceptive function in the correlation analyses. However, these associations were no longer independently significant after adjustment in the multiple regression analysis. A linear support vector machine achieved an area under the curve of 0.716, with an accuracy of 0.700, sensitivity of 0.800, and specificity of 0.600. Permutation testing confirmed that the classification performance was above chance. CONCLUSION: Motor imagery ability may be related to a combination of motor, sensory, and cognitive functions rather than any single functional domain.