Effectiveness of Catheter and Standalone Surgical Ablation Procedures for Atrial Fibrillation: A Bayesian-Network...

Ablation procedures for atrial fibrillation (AF), including catheter (CA) and surgical ablation, are effective rhythm control therapies. This study is a Bayesian network meta-analysis evaluating randomised evidence on the invasive treatment of AF, focusing on freedom from atrial tachyarrhythmias (ATAs) while evaluating the potential trade-off in morbidity and mortality.

This study was registered in PROSPERO (CRD42025632171). Randomised controlled trials (RCTs) were included comparing any of the 4 treatments; CA, isolated thoracoscopic, hybrid thoracoscopic ablation, and the Convergent procedure. Primary outcome was freedom from ATA at 12 months. Secondary outcomes were mortality, stroke, and bleeding. A Bayesian-network meta-analysis was performed. The combined effects of the primary and secondary outcomes were studied in a bivariate analysis. Treatments were ranked and their effects were summarised using surface under the cumulative ranking curves (SUCRAs).

Ten RCTs were included in the analysis (n = 877 patients, predominantly persistent AF). Using CA as a reference, the pooled network odds ratios for freedom from ATA for hybrid thoracoscopic, isolated thoracoscopic, and Convergent were 4.95 (95% credible interval [CrI] 2.16-13.46), 2.23 (95% CrI 1.23-4.48), and 2.23 (95% CrI 0.90-6.69), with SUCRAs for hybrid thoracoscopic, isolated thoracoscopic, Convergent, and CA of 95.5%, 50.8%, 52.1%, and 1.5%, respectively. No increase in periprocedural morbidity or mortality was observed. Results were robust across various sensitivity analyses.

In this Bayesian-network meta-analysis consisting exclusively of randomised data, surgical ablation in general and hybrid ablation in particular provide superior outcome in terms of 1-year freedom from ATA. Both CA and surgical procedures are characterised by a favourable safety profile.