Genetic risk score of cerebral infarction in atrial fibrillation genome-wide association study.

Abstract

Stroke is a leading cause of death and the primary cause of adult-acquired disability. Patients with cardiogenic embolic stroke also have higher mortality and recurrence rates than patients with other stroke subtypes. Atrial fibrillation (AF) is a major risk factor for cerebral infarction (CI). The large-scale study identified 32 loci in the MEGASTROKE study. However, few studies have attempted to identify novel stroke risk variants in patients with a history of AF. Our overall aim was to identify novel CI risk variants in AF cases and explore whether their associations with the CI risk were affected by the CHADS2 and CHA2DS2-VASc scores. We performed association study with CI using 8181 AF cases in previous genome-wide association study (GWAS) and imputation data without controls. We classified AF cases into those with or without past history of CI, and the genetic associations with the CI risk were examined. GWAS identified eight associated loci. The generated genetic risk score (GRS) for the eight loci was significantly associated with CI in patients with AF (1.46 × 10-8 ). We estimated bivariate logistic regression model which contained GRS and CHADS2 score (GRS: p-Value = 7.41 × 10-9 , CHADS2 score: p-Value <2.0 × 10-16 ) or CHA2DS2-VASc scores (GRS: p-Value = 2.52 × 10-10 , CHA2DS2-VASc score: p-Value <2.0 × 10-16 ). We identified eight genetic variants that were potentially associated with the risk of CI of AF cases and the significant GRS, whose associations were independent of the CHADS2 or CHA2DS2-VASc score.