Left Atrial Remodeling After Pacemaker Implantation Predicts Incident Atrial Fibrillation: Comparison of Leadless and Transvenous Devices

BACKGROUND: Atrial fibrillation (AF) after pacemaker (PM) implantation is associated with adverse outcomes; however, the relationship between left atrial remodeling, device type, and ventricular pacing burden remains unclear. OBJECTIVE: We aimed to investigate (1) whether post-implantation changes in left atrial diameter (ΔLAD) are associated with incident AF; (2) differences in AF and ΔLAD between TVP and LP; and (3) whether mean ventricular pacing percentage (Vp%) modifies device-related effects. METHODS: This retrospective study included patients without a prior history of AF who underwent PM implantation. The primary outcome was defined as incident clinical AF, which was analyzed using Cox hazard models. Mechanistic models incorporating Vp% and paced QRS duration were also evaluated. Interaction analyses were performed using mean-centered Vp % (device × Vp) and Vp%-stratified analyses. To mitigate reverse causality, a landmark analysis excluding AF events within 1 year after implantation was conducted. RESULTS: In the pooled Cox analyses, ΔLAD was independently associated with incident AF (p < 0.0001). No significant difference in ΔLAD was observed between the TVP and LP groups (p = 0.417). In models adjusted for ΔLAD, LP was associated with a significantly lower risk of AF (device, p = 0.049; ΔLAD, p < 0.0001). Vp% and paced QRS duration were not significant main effects. However, a significant device × Vp interaction was observed in the mean-centered interaction models (p = 0.022). In landmark analyses, ΔLAD remained associated with AF beyond 1 year (p = 0.0013). CONCLUSION: Left atrial remodeling independently predicted incident AF after pacemaker implantation. LP was associated with a lower AF risk, with effects modified by the pacing burden.