Symptom Network and Lesion Mapping of Post-Stroke Depression, Cognitive Impairment, and Dysfunction: A Multicenter...

This study aimed to examine the interplay among depressive symptoms, cognitive impairment, and dysfunction after ischemic stroke using a network analysis approach, and to investigate whether selected network-derived targets show convergent neuroanatomical correlates using voxel-based lesion-symptom mapping (VLSM) in an imaging subset.

We prospectively enrolled 589 acute ischemic stroke patients and constructed a 3-month post-stroke symptom network. Depression, cognition, and dysfunction were assessed via 17-item Hamilton Depression Rating Scale (HDRS), Mini-Mental State Examination (MMSE), and modified Rankin Scale (mRS), respectively. Flow analysis quantified dysfunction's direct association with neuropsychiatric symptoms. In an imaging subset (n = 429), VLSM localized neuroanatomical substrates of key network targets.

Depressed mood and Psychiatric anxiety were central symptoms, while Suicidality, Loss of interest, Psychiatric anxiety, and mRS score acted as bridge symptoms. mRS score was positively associated with Suicidality and Loss of interest, and negatively associated with Language. VLSM identified a significant cluster in the left superior corona radiata for Language, whereas Suicidality and Loss of interest showed no focal anatomical correlates.

Identifying core and bridge symptoms (e.g., Depressed mood, Psychiatric anxiety, Suicidality, Loss of interest, mRS score) may serve as candidate indicators for rehabilitation planning. Suicidality, Loss of interest and Language may serve as promising candidates for monitoring and rehabilitation. Notably, while Language maps to the left superior corona radiata, the absence of focal lesions for Suicidality and Loss of interest suggests reliance on distributed network dysfunction.