Virtual reality therapy (VRT) is an interactive intervention that induces cortical activation. However, when this tool may aid in motor function recovery after stroke is not yet established. The aim of this study was to evaluate the effects of VRT on upper limb function after stroke. Neuroimaging features were also assessed to investigate a predictor of better response to this therapy.

Patients with stroke were consecutively selected and underwent VRT. Clinical neurological characteristics, function of the upper limb, and quality of life were evaluated before and after VRT. Statistical analysis was performed using a general linear model (GLM) comparing pre- and post-intervention outcomes. Age, sex, and thrombolytic therapy were used as confounding factors. High-resolution computed tomography (CT) scans were obtained in the post-stroke phase. CT images were registered into standard space using a lesion mask protocol and flipped based on the side of the stroke. Voxel-based lesion symptom mapping (VLSM) was performed using GLM and searching for correlations between the clinical data and CT lesion.

Eighteen patients were studied (8 women; mean age, 55.5±13.9 years). Quality of life, functional independence, functional characteristics, and dexterity of the upper limb showed improvement after the intervention (P<0.001). The mean volume of lesions was 24,172±17,302mm³. VLSM analysis showed negative correlations mainly between the posterior portion of the internal capsule and functional recovery (Fig. 1).

VRT showed benefits for patients with stroke. Subcortical strokes with involvement of the internal capsule presented the worst response to therapy.