BackgroundTo re-evaluate comparative cardiovascular (CV) safety of febuxostat versus allopurinol among patients with gout following recent accumulated use of febuxostat. MethodsUsing 2011-2019 Korea National Health Insurance database, we conducted a cohort study comparing gout patients initiating febuxostat versus allopurinol, 1:1 matched on a propensity-score (PS) for >60 covariates. The primary outcome was a composite endpoint of myocardial infarction, coronary revascularization, and stroke. Secondary outcomes were individual components of the primary outcome, hospitalized heart failure, and all-cause mortality. Subgroup analyses were done for those at high CV risk, long-term users (follow-up >3 years), and those without chronic kidney disease. We used Cox proportional hazard models to estimate hazard ratios (HRs) and 95% confidence intervals (CIs). ResultsWe included 160,930 PS-matched pairs of febuxostat and allopurinol users (mean age 59.3 years, 79.6% male). Incidence rates of the primary outcome were 2.06 and 2.27 per 100 person-years for febuxostat and allopurinol users, respectively, with a HR [95% CI] of 1.03 [0.95-1.12] comparing febuxostat versus allopurinol initiators. We also observed similar risks for secondary outcomes, except for reduced all-cause mortality among febuxostat users (HR [95% CI] of 0.84 [0.78-0.91]). Subgroup analyses also showed non-inferior CV safety of febuxostat. ConclusionIn this population-based cohort study including the largest number of febuxostat users to date, we found non-inferior CV safety of febuxostat versus allopurinol. There was a 16% reduction in all-cause mortality among febuxostat users.