Subacute default mode network dysfunction in the prediction of post-stroke depression severity

To identify patterns of rest functional connectivity (FC) in the whole brain with the default mode network (DMN) soon after stroke and to explore the predictive accuracy of the strength of rest FC in specific areas on poststroke severity of depression and anxiety symptoms.

The protocol was accepted by the local ethics board, and all patients provided informed consent to participate. Resting-state functional magnetic resonance (MR) images were acquired 10 days after a first stroke in 24 patients without a history of psychiatric illness. Independent component analysis was used to isolate the DMN in each subject. Hamilton Depression Rating Scale (HDRS) 17 and Hamilton Anxiety Rating Scale (HARS) were recorded 10 days and 3 months after the stroke. Associations between severity of anxiety or depression symptoms and DMN functional connectivity were investigated with whole-brain analyses by using statistical parametric mapping software and were adjusted for age, sex, manual laterality, and National Institutes of Health Stroke Severity scores. Correlations were considered significant if P<.001, with a cluster size of more than 50 voxels.

Ten days after stroke, anxiety severity was correlated with functional connectivity in the middle temporal cortex and the anterior midcingulate cortex, while at 3 months after stroke, a correlation was observed with the middle temporal cortex and the posterior cingulate cortex. Poststroke depressive symptom severity did not correlate with functional connectivity changes at 10-day follow-up, while the HDRS 17 score was correlated with functional connectivity in the left middle temporal cortex and precuneus at 3-month follow-up.

These results suggest that a dysfunction of DMN functional connectivity involved in emotional control is associated with the severity of poststroke depression. Further studies are necessary to determine the mechanisms of this functional impairment.