Association between acute infarct topography and cerebral microbleed location in acute ischemic stroke: a retrospective cross-sectional magnetic resonance imaging study.

BACKGROUND: Cerebral microbleeds (CMBs) are important imaging markers of cerebral small vessel disease and are frequently detected in patients with stroke. However, imaging evidence regarding spatial associations between infarct topography and CMBs detected on the index magnetic resonance imaging (MRI) examination in patients with acute ischemic stroke (AIS) remains limited. This study aims to investigate the association between acute infarct location and CMB location on brain MRI. METHODS: This retrospective cross-sectional study included cranial MRI data from patients with AIS. Acute infarcts were identified on diffusion-weighted imaging/apparent diffusion coefficient (DWI/ADC) and classified into six predefined neuroanatomical regions. CMBs detected on the index MRI examination were identified on susceptibility-weighted imaging (SWI) and recorded using the same regional scheme. Two experienced readers independently interpreted the images under blinded conditions, and disagreements were resolved by consensus adjudication. Multivariable logistic regression was performed to evaluate factors associated with the presence of CMBs after limited adjustment for sex, age, and infarct location. Exploratory regression analyses were further performed to assess associations between infarct location and region-specific CMBs. RESULTS: A total of 138 patients with AIS were included, among whom 79 (57.2%) had CMBs detected on SWI. In multivariable logistic regression analysis, after limited adjustment for sex, age, and infarct location, corona radiata infarction was associated with the presence of CMBs [odds ratio (OR) =3.61, 95% confidence interval (CI): 1.52-8.61, P=0.004]. Exploratory univariable analyses showed that cortical infarction was associated with cortical CMBs (OR =2.90, 95% CI: 1.37-6.13, P=0.005), while corona radiata infarction was associated with corona radiata CMBs (OR =3.82, 95% CI: 1.83-8.00, P<0.001). In addition, increasing age (OR =1.11, 95% CI: 1.02-1.20, P=0.013) and thalamic infarction (OR =5.36, 95% CI: 1.18-24.46, P=0.030) showed exploratory associations with cerebellar CMBs. CONCLUSIONS: In this retrospective single-center MRI study, corona radiata infarction was associated with CMBs detected on the index MRI examination after limited adjustment for demographic variables and infarct location. Region-specific associations between acute infarct location and CMB location were exploratory and should be interpreted cautiously, given the cross-sectional design, limited covariate adjustment, small sample size, and potential scanner heterogeneity. Further prospective multicenter studies using standardized MRI protocols and more comprehensive clinical and imaging covariates are needed to validate these findings.