ESC Congress 2026 Abstract · Oct 26, 2026
Genetic risk-enhanced CHA2DS2-VASc score predicts ischemic stroke after atrial fibrillation catheter ablation
By Park JW, Hong MH, Shin ES et al.
Backgroud and Aims: Although atrial fibrillation catheter ablation (AFCA) can reduce the risk of ischemic stroke (IS), patients remain at residual risk. The genetic basis of post-AFCA IS is unclear. We investigated single-nucleotide polymorphisms (SNPs) associated with post-AFCA IS and evaluated whether incorporating a genetic risk score (GRS) provides incremental predictive value beyond CHA2DS2-VASc scores in two independent cohorts. Methods: We developed and validated a GRS-enhanced CHA2DS2-VASc model for predicting post-AFCA IS using the UK Biobank (UKB) cohort (n=2,063) and the independent Yonsei Atrial Fibrillation Ablation (YAF) cohort (n=2,897), excluding procedure-related IS (<14 days post-ablation). We created a CHA2DS2-VASc-G score by adding one point to the CHA2DS2-VASc score in patients with high genetic risk (defined as GRS ≥ 1.025, derived from the Youden index). We evaluated the performance and clinical utility of the CHA2DS2-VASc+GRS model. Results: Over a mean follow-up of 5.5 years in the UKB cohort and 5.1 years in the YAF cohort, 46 and 48 patients experienced post-AFCA IS, corresponding to incidence rates of 0.40 and 0.33 per 100 person-years, respectively. Five genomic loci (SMARCC1, FGF10, RUNX1T1, HEMGN, VAC14) were significantly associated with post-AFCA IS and replicated in both cohorts. The GRS was higher in patients with IS (1.9±2.2 vs. 0.3±0.7, p<0.001; hazard ratio 1.80 per 1 SD, [95% CI, 1.63-1.99]). The CHA2DS2-VASc+GRS model demonstrated superior discrimination (C statistic, 0.754 vs. 0.599; p < 0.001) and improved net reclassification (net reclassification index, 21%; 95% CI, 3.3%–37.5%) compared with the CHA2DS2-VASc model alone. During long-term follow-up, the incidence of post-AFCA IS remained high in high-risk patients classified by the CHA2DS2-VASc-G score (≥2; log-rank p < 0.001) compared with low-risk patients, in contrast to patients classified by the CHA2DS2-VASc score alone. Among patients with a CHA2DS2-VASc score <2, the incidence rate of post-AFCA ischemic stroke was higher in those classified as having genetic risk by the CHA2DS2-VASc-G score than in those without genetic risk (0.59 vs. 0.14 per 100 person-years; incidence rate ratio, 4.28; p<0.001). Conclusions: Incorporating GRS into the CHA2DS2-VASc score significantly improved the prediction of IS after AFCA compared with the CHA2DS2-VASc score alone. The CHA2DS2-VASc-G score identified high-risk patients for post-AFCA IS among those with a CHA2DS2-VASc score <2.Performance of modelsCumulative incidence of post-AFCA IS