Functional connectivity between homotopic perisylvian regions correlates with functional status after stroke

Goal setting in stroke rehabilitation requires assessing patients' functional status, which involves both recovery and compensatory processes following stroke, underscoring the importance of identifying valid biomarkers. Functional connectivity measures derived from functional magnetic resonance imaging (fMRI) signals, which reflect underlying biological states, may offer value for characterizing functional status.

We tested whether functional connectivity correlates with functional status and discriminates post-stroke functional independence.

Observational study.

Inpatient rehabilitation ward at Fujita Health University Hospital, Japan.

Fifty-eight patients with stroke.

Resting-state fMRI data were used to compute group-sparse inverse covariance matrices, which were negated to estimate partial correlations representing direct functional connectivity. Correlations between connectivity measures and functional independence measure (FIM) scores were examined. Principal component analysis (PCA) was applied to extract dominant connectivity features from connections associated with functional status. Connectivity-based discriminative performance was evaluated using receiver operating characteristic analysis.

Connectome mapping revealed significant positive correlations between total FIM scores and interhemispheric connectivity in homotopic perisylvian regions, including the insulae, superior temporal sulci and temporoparietal junctions. PCA extracted a major connectivity component whose strength demonstrated within-sample discrimination of functional independence. Lesions were predominantly located in subcortical regions.

The results support a significant association between homotopic functional connectivity within perisylvian, multifunctional regions located at the intersection of lobes and post-stroke daily functioning. Intrinsic interhemispheric functional connectivity, assessed using fMRI, may offer valuable insights into post-stroke functional status.

These findings highlight a promising direction for developing connectivity-based rehabilitation strategies for stroke-related disability.