Metabolic Dysfunction‐Associated Steatotic Liver Disease, With and Without Increased Alcohol Intake, and Incident Atrial Fibrillation in Older Korean Adults: A Competing‐Risk Cohort Study

AIM: Older adults carry the largest burden of atrial fibrillation (AF) and a high competing risk of death, yet metabolic dysfunction-associated steatotic liver disease (MASLD) and MASLD with increased alcohol intake (MetALD) have rarely been examined separately in this population. We examined the associations of fatty liver index (FLI)-defined MASLD and MetALD with incident AF in adults aged ≥ 60 years, accounting for competing death. METHODS: We analyzed 216 292 participants of the Korean National Health Insurance Service-Senior cohort screened in 2009-2010 without AF recorded before follow-up. Using FLI ≥ 30 as the phenotype threshold, participants were classified into FLI < 30 reference, MASLD, and MetALD phenotypes. Incident AF was ascertained from claims from 2011 to 2019. Cause-specific hazard ratios (HRs) were estimated with Cox regression and subdistribution HRs (SHRs) with Fine-Gray regression; MetALD was compared with MASLD using the ratio of adjusted estimates, and E-values were computed. RESULTS: Over 1 833 772 person-years, 4292 AF events and 25 713 deaths without AF occurred. Compared with the reference phenotype, Model 2 HRs were 1.04 (95% confidence interval [CI], 0.96-1.12) for MASLD and 1.27 (95% CI, 1.09-1.48) for MetALD; SHRs were 1.02 (95% CI, 0.95-1.10) and 1.26 (95% CI, 1.08-1.47). The MetALD estimate was 1.22 times (95% CI, 1.05-1.42) the MASLD estimate. E-values were 1.23 for MASLD and 1.85 for MetALD. CONCLUSIONS: In older Korean adults, the MASLD-AF association was uncertain after adjustment, whereas the MetALD association persisted under both cause-specific and subdistribution hazard frameworks. The MASLD CI included the null, whereas the MetALD estimate was relatively more robust to unmeasured confounding.