Abstract Background and aims Post-stroke cognitive impairment (PSCI) affects over half of stroke survivors and drives long-term disability. Although PSCI is diagnosed clinically, the mechanisms underlying specific cognitive deficits remain poorly defined, limiting prognostication and targeted intervention. Conventional lesion analyses focus on infarct volume and location to understand cognitive impact and may overlook disruption of distributed brain networks. We applied lesion network mapping (LNM) to determine whether domain-specific PSCI is explained by network-level circuit disruption beyond focal infarct sites. Methods We analyzed harmonized neuroimaging and cognitive data from 12 international PSCI cohorts in the Meta VCI Map Consortium (n = 2,950). Cognitive assessments within one year post-stroke were harmonized into domain-specific binary impairment scores. To test whether cognitive deficits are linked to network disruption beyond focal lesion location, we related domain-specific impairment to (i) lesion location (VLSM), (ii) structural disconnectivity from normative tractography, and (iii) functional network involvement from normative resting-state connectivity, adjusting for infarct volume (Figure 1). Results While VLSM identified focal, overlapping regions, structural and functional LNM revealed disruption of widely distributed brain circuits. We found that impairment in each cognitive domain (attention/executive function, processing speed, language, verbal memory, visuospatial function and visuospatial memory) mapped to unique, domain-specific structural and functional network architecture extending beyond focal lesion sites (Figure 2). Conclusions Across a large and diverse stroke population, LNM provided insights into neuroanatomical substrates of PSCI not captured by focal lesion models alone. Embedding infarcts within their network context offers complementary mechanistic understanding and may support improved prediction, stratification, and network-targeted interventions. Conflict of interest Marvin Petersen: nothing to disclose: Geert Jan Biessels: nothing to disclose. Figure 1 - belongs to Methods Figure 2 - belongs to Results
Petersen et al. (Fri,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: