Atrial Fibrillation and Risk of Incident Cognitive Impairment: The REasons for Geographic and Racial Differences in Stroke Study

Atrial fibrillation (AF) and cognitive impairment will each double in prevalence over the next 20 years. Most studies on AF and cognitive disorders have focused on dementia, with less research on cognitive impairment generally. We assessed the association of AF with incident cognitive impairment (ICI) and whether inflammation biomarkers or anticoagulant use attenuated this. The REasons for Geographic and Racial Differences in Stroke (REGARDS) study enrolled 30,239 adults ≥45 years old in 2003-07. Among those without baseline cognitive impairment, ICI was identified by standardized telephone assessments. Hazard ratios (HRs) of ICI were calculated using Cox proportional hazards models. Differences in associations by prevalent stroke, race, and oral anticoagulant use were tested using interaction terms. Among 23,638 participants (mean age 64 years, 56% women, 38% Black), 7% developed ICI over 13 years. AF was associated with ICI among those with prevalent stroke (adjusted HR: 1.69, 95% CI: 1.11-2.56) but not without (HR: 1.05, 95% CI: 0.88-1.27; p interaction = 0.07). The association was not attenuated by anticoagulant use and did not differ by race. Among those with prevalent stroke, there was a small-to-modest attenuation after adjusting for inflammation markers, with the largest attenuation by albumin (15%). In conclusion, in this large cohort, AF was associated with ICI in those with - but not in those without - prevalent stroke. Inflammation biomarkers had modest attenuating effects, and anticoagulation use did not. Results underscore the importance of considering cognitive impairment after stroke in those with AF and identifying underlying causes and preventive treatments.