Catheter ablation as first versus second-line therapy for atrial fibrillation: a frequentist and bayesian meta-analysis with reconstructed time-to-event and trial sequential analysis
A Anghel, B Maille, J C DeharoAbstract
Background
Catheter ablation (CA) is a cornerstone therapy for atrial fibrillation (AF), but the optimal timing of intervention - whether as first-line or after antiarrhythmic drug (AAD) failure - remains uncertain.
Purpose
This meta-analysis aimed to compare one-year atrial arrhythmia recurrence rates after catheter ablation as first versus second-line therapy in patients with paroxysmal and persistent atrial fibrillation.
Methods
We performed a systematic search of PubMed, Embase, and Cochrane databases to identify studies comparing CA as first-line versus second-line therapy in AF. Second-line therapy was defined as ablation following failure of ≥1 AAD. The primary endpoint was atrial arrhythmia recurrence at 1 year after the first ablation procedure, excluding the 3-month blanking period. When event rates were unavailable, 1-year data were reconstructed from Kaplan–Meier curves or provided by study authors. Analyses included frequentist random-effects, reconstructed time-to-event, Bayesian, and trial sequential approaches.
Results
Seven observational studies including 10,739 patients were analyzed. In the frequentist model, first-line CA was associated with a lower risk of atrial arrhythmia recurrence compared with second-line therapy (OR 0.81, 95% CI 0.66 to 0.99; p = 0.04) (Figure 1A). Trial sequential analysis indicated sufficient cumulative evidence (required information size = 2958) (Figure 2). Pooled reconstructed time-to-event data from four studies yielded consistent results (HR 0.78, 95% CI 0.69 to 0.87; p < 0.001) (Figure 1B). Bayesian analysis (half-Cauchy prior) supported these findings with a 96.6% posterior probability of benefit :OR < 1(Figure 1C), 95% CrI 0.64 to 1.02 (Figure 1D). Subgroup analyses suggested that first-line ablation was associated with lower recurrence risk both in paroxysmal (HR 0.85, 95% CI 0.73 to 0.97; p = 0.022) and persistent AF (HR 0.77, 95% CI 0.64 to 0.93; p = 0.005), consistent across frequentist and Bayesian frameworks. Egger’s test did not indicate any potential risk of bias by funnel plot asymmetry(intercept=0.1,p=0.57).
Conclusion
Catheter ablation as an initial rhythm-control strategy is associated with a lower rate of atrial arrhythmia recurrence compared with ablation performed after AAD failure. However, whether the difference in outcomes is explained by earlier ablation or by more advanced disease remains to be established. Further randomized studies are warranted to confirm our results.