Whether ablation for atrial fibrillation (AF) is, in terms of clinical outcomes, beneficial for Japanese patients has not been clarified. Drawing data from 2 Japanese AF registries (AF Frontier Ablation Registry and SAKURA AF Registry), we compared the incidence of clinically relevant events (CREs), including stroke/transient ischemic attack (TIA), major bleeding, cardiovascular events, and death, between patients who underwent ablation (n = 3451) and those who did not (n = 2930). We also compared propensity-score matched patients (n = 1414 in each group). In propensity-scored patients who underwent ablation and those who did not, mean follow-up times were 27.2 and 35.8 months, respectively. Annualized rates for stroke/TIA (1.04 vs. 1.06%), major bleeding (1.44 vs. 1.20%), cardiovascular events (2.15 vs. 2.49%) were similar (P = 0.96, 0.39, and 0.35, respectively), but annualized death rates were lower in the ablation group than in the non-ablation group (0.75 vs.1.28%, P = 0.028). After multivariate adjustment, the risk of CREs was statistically equivalent between the ablation and non-ablation groups (hazard ratio [HR] 0.89, 95% confidence interval [CI] 0.71-1.11), but it was significantly low among patients who underwent ablation for paroxysmal AF (HR 0.68 [vs. persistent AF], 95% CI 0.49-0.94) and had a CHA2DS2-VASc score < 3 (HR 0.66 [vs. CHA2DS2-VASc score ≥ 3], 95% CI 0.43-0.98]). The 2-year risk reduction achieved by ablation may be small among Japanese patients, but AF ablation may benefit those with paroxysmal AF and a CHA2DS2-VASc score < 3.
Keywords: Ablation; Atrial fibrillation; Bleeding; Clinically relevant events; Stroke.