Polygenic risk scores for atrial fibrillation and heart failure and the risk of stroke and dementia

Atrial fibrillation and heart failure have both been suggested to increase stroke and dementia risk. However, in observational studies, reversed causation and unmeasured confounding may occur. To mitigate these issues, this study aims to investigate if higher genetic risk for atrial fibrillation and heart failure increases dementia and stroke risk. Data were obtained from the population-based Gothenburg H70 Birth Cohort Studies in Sweden. Participants (<i>N</i> = 984) were born in 1930 with baseline examinations at age 70, 75, 79 or 85 and follow-ups until age 88-89. Polygenic risk scores at the 5 × 10^-8, 1 × 10^-5, 1 × 10^-3 and 1 × 10^-1 thresholds were generated for atrial fibrillation and heart failure. Stroke was diagnosed based on self-reports, close-informant interviews, and the National Patient Register. Dementia was diagnosed based on neuropsychiatric examinations, close-informant interviews, and the National Patient Register. Cox regression analyses were performed, adjusted for sex, age at baseline and the first five principal components to correct for population stratification. Those within the highest atrial fibrillation-polygenic risk score tertile had a 1.5 (95% CI 1.09-2.03) increased risk of dementia (at the 1 × 10^-5 threshold) and a 1.5 (95% CI 1.07-2.03) increased risk of stroke (at the 1 × 10^-3 threshold) compared to the lowest tertile. Those within the highest heart failure-polygenic risk score tertile had a 1.6 (95% CI 1.19-2.27) increased risk of dementia (at the 5 × 10^-8 threshold), but no increased risk of stroke (HR 1.2; 95% CI 0.83-1.60 at the 1 × 10^-5 threshold), compared to the lowest tertile. When analysing the polygenic risk scores as a continuous variable, the associations were in the same direction, although weaker. This study, investigating genetic risk of atrial fibrillation and heart failure in relation to stroke and dementia, supports the increasing body of evidence suggesting that atrial fibrillation is associated with both stroke and dementia risk. Whether heart failure increases dementia risk is less established, but the present study found that genetic risk of heart failure increased dementia risk. The finding that genetic risk for heart failure did not increase stroke risk needs to be interpreted with caution, as it may be due to a lack of statistical power. There are guidelines on how to best treat atrial fibrillation to prevent stroke, but more knowledge is needed on how to treat atrial fibrillation and heart failure to prevent dementia.