Objective: We examined the risk factors of cortical cerebral microinfarcts (CMIs) on 3T MRI and their association with cognitive impairment. Methods: Participants (aged 60 years and older) from the multiethnic Epidemiology of Dementia In Singapore Study underwent detailed neuropsychological testing and 3T brain MRI. Cortical CMIs were graded using a previously validated protocol. Cognitive impairment was categorized into cognitive impairment, no dementia (CIND)-mild, CIND-moderate, and dementia. Cognitive function was summarized as composite and domain-specific z scores. Results: Among 861 participants, 54 (6.3%) had ≥1 cortical CMI. In multivariate-adjusted models, the risk factors of cortical CMIs were increasing age, Malay ethnicity, hypertension, diabetes, history of stroke, and markers of both large (cortical infarcts and intracranial stenosis) and small (lacunar infarcts, white matter hyperintensities, cerebral microbleeds) vessel disease. Presence of cortical CMIs was associated with CIND-moderate (odds ratio: 3.12; 95% confidence interval [CI]: 1.18-8.58), dementia (odds ratio: 16.92; 95% CI: 3.37-85.05), and poorer cognitive function (mean difference in composite z score: -0.42; 95% CI: -0.62 to -0.21). Additional adjustments for vascular risk factors and other MRI markers did not alter these associations. Conclusions: Cortical CMIs are a novel MRI marker of cerebrovascular disease and are independently associated with cognitive impairment and dementia. These findings provide new insights into the burden of cerebrovascular disease in cognitive impairment. Future research is needed to establish the additional etiologic and prognostic significance of cortical CMIs.

Additional Metadata
Persistent URL dx.doi.org/10.1212/WNL.0000000000003110, hdl.handle.net/1765/93862
Journal Neurology
Hilal, S, Sikking, E. (Emiel), Shaik, M.A. (Muhammad Amin), Chan, Q.L. (Qun Lin), Van Veluw, S.J. (Susanne J.), Vrooman, H.A, … Ikram, M.K. (2016). Cortical cerebral microinfarcts on 3T MRI: A novel marker of cerebrovascular disease. Neurology, 87(15), 1583–1590. doi:10.1212/WNL.0000000000003110