To Örebro University

BACKGROUND: Dronedarone and sotalol are antiarrhythmic drugs (AADs) recommended in similar populations per atrial fibrillation (AF) guidelines; however, comparative safety data are limited.

OBJECTIVES: The goal of this study was to assess the safety of dronedarone vs sotalol for treatment of AF in AAD-naive patients.

METHODS: This was a prespecified noninterventional meta-analysis of 4 retrospective observational cohort studies from 4 databases (Optum Clinformatics Data Mart, Merative MarketScan, Veterans Health Administration Electronic Health Record, and the Swedish National Patient Register) conducted by using one master protocol. Each analysis emulated the target trial using an active comparator (dronedarone vs sotalol), new user design with an as-treated approach. Primary outcomes were tested hierarchically for dronedarone vs sotalol: first for statistical significance of cardiovascular (CV) hospitalization, and then for statistical significance of ventricular arrhythmias. Propensity score matching (PSM) was used for confounding control, and negative control outcomes were used to assess residual confounding. Outcomes were evaluated by using Cox proportional hazards regression; meta-analysis was performed by using fixed effects models.

RESULTS: The dronedarone and sotalol cohorts were well balanced within databases before and after PSM (after PSM mean age range: 62.5-70.9 years; mean CHA2DS2-VASc score range: 1.81-3.15). Negative control outcomes exhibited little-to-no evidence of residual confounding. Meta-analysis found significantly lower rates of CV hospitalization (pooled HR: 0.91; 95% CI: 0.85-0.97) and ventricular arrhythmias (pooled HR: 0.77; 95% CI: 0.69-0.85) with dronedarone vs sotalol.

CONCLUSIONS: In this retrospective meta-analysis, dronedarone exhibited significantly lower rates of CV hospitalization and ventricular arrhythmias compared with sotalol. These findings provide real-world evidence to support selection of the most appropriate first-line AAD for rhythm control in patients with AF.