Background and Purpose- Atrial fibrillation (AF) is associated with dementia independent of clinical stroke. The mechanisms underlying this association remain unclear. In a community-based cohort, the ARIC study (Atherosclerosis Risk in Communities), we evaluated (1) the longitudinal association of incident AF and (2) the cross-sectional association of prevalent AF with brain magnetic resonance imaging (MRI) abnormalities. Methods- The longitudinal analysis included 963 participants (mean age, 73±4.4 years; 62% women; 51% black) without prevalent stroke or AF who underwent a brain MRI in 1993 to 1995 and a second MRI in 2004 to 2006 (mean, 10.6±0.8 years). Outcomes included subclinical cerebral infarctions, sulcal size, ventricular size, and, for the cross-sectional analysis, white matter hyperintensity volume and total brain volume. Results- In the longitudinal analysis, 29 (3.0%) participants developed AF after the first brain MRI. Those who developed AF had higher odds of increase in subclinical cerebral infarctions (odds ratio [OR], 3.08; 95% CI, 1.39-6.83), worsening sulcal grade (OR, 3.56; 95% CI, 1.04-12.2), and worsening ventricular grade (OR, 9.34; 95% CI, 1.24-70.2). In cross-sectional analysis, of 969 participants, 35 (3.6%) had prevalent AF at the time of the 2004 to 2006 MRI scan. Those with AF had greater odds of higher sulcal (OR, 3.9; 95% CI, 1.7-9.1) and ventricular grade (OR, 2.4; 95% CI, 1.0-5.7) after multivariable adjustment and no difference in white matter hyperintensity or total brain volume. Conclusions- AF is independently associated with increase in subclinical cerebral infarction and worsening sulcal and ventricular grade-morphological changes associated with aging and dementia. More research is needed to define the mechanisms underlying AF-related neurodegeneration.
Keywords: atrial fibrillation; cognitive dysfunction; dementia; magnetic resonance imaging; research.