Rhythm Control of Persistent Atrial Fibrillation in Systolic Heart Failure: A Bayesian Network Meta-Analysis of Randomized Controlled Trials

Int J Heart Fail. 2021 Jun 11;3(3):179-193. doi: 10.36628/ijhf.2021.0008. eCollection 2021 Jul.

Abstract

Background and objectives: Persistent atrial fibrillation (PeAF) with heart failure (HF) arguably constitutes the sickest subset of atrial fibrillation (AF) patients.

Methods: A systematic search was made in PubMed, Embase, and Scopus databases. Network meta-analysis (NMA) of PeAF patients with systolic HF comparing all-cause mortality, change in HF-related quality of life (QoL) and hospitalization due to heart failure (HHF) were performed among catheter ablation (CA) of AF, rate-controlling drugs (RCDs), anti-arrhythmic drugs (AADs), and atrio-ventricular nodal ablation (AVNA) using Bayesian random effect model.

Results: Ablation strategies resulted significantly lower mortality than medical therapies (odds ratio [OR], 0.51; 95% confidence interval [CI], 0.35 to 0.76). CA of AF was associated with lower trend of mortality (OR, 0.78; 95% credible interval [CrI], 0.08 to 7.63) in comparison to AVNA in the Bayesian NMA. Rhythm control strategies resulted significantly higher improvement of QoL than rate control strategies (mean difference [MD], -12.78; 95% CI, -21.26 to -4.31). Bayesian NMA showed that CA of AF was better than AAD (MD, -7.98; 95% CrI, -27.68 to 8.27), however ranked AVNA to be lowest. Ablation strategies provided significantly lower HHF than medical therapies (OR, 0.42; 95% CI, 0.30 to 0.58). Bayesian NMA showed that CA of AF performed not only better than AAD (OR, 0.33; 95% CrI, 0.09 to 1.3) to reduce HHF, but also than AVNA (OR, 0.20; 95% CrI, 0.00 to 4.76). Of note, RCD ranked lowest with regard to mortality and HHF.

Conclusions: CA of AF remains the best strategy even for the sickest group of PeAF patients with systolic HF in regards to all-cause mortality, HF-related QoL and HHF.

Keywords: Catheter ablation; Heart failure; Persistent atrial fibrillation.