Early recurrence after pulsed field ablation as a potential predictor of late arrhythmia: a systematic review and meta-analysis

Background Current atrial fibrillation (AF) guidelines and expert consensus documents from major societies discuss early arrhythmia recurrence (ER) within the first 3 months after catheter ablation as predictive of late recurrence (LR). These recommendations are based almost exclusively on thermal ablation. With pulsed field ablation (PFA) emerging as a distinct modality with different biophysics and lesion characteristics, the prognostic relevance of ER after PFA remains uncertain. Purpose To explore whether ER after PFA (<3 months) is a predictor for LR at 12 months follow-up. Methods We performed a comprehensive search of PubMed, Embase, and Cochrane databases. ER was defined as AF / atrial flutter / atrial tachycardia >30 s within 90 days; LR as recurrence beyond 3 months. A prognostic meta-analysis was conducted using random-effects models, reporting risk ratios (RR) with 95% confidence intervals (Cl). Prespecified sensitivity analyses included sequential leave-one-out evaluation, and subgroup analyses for paroxysmal, persistent AF and Farapulse-only studies. Results The pooled cohort included 1637 patients (67.7% with paroxysmal AF), with a mean age of 62.5 ± 15.7 years. All studies have a follow-up duration of 12 months. Across all five studies (one randomized trial and four observational), ER showed an association with increased LR (RR 3.61, 95% CI 2.62-4.99; I² = 57%, P = 0.0004). Sequential exclusion of studies reduced heterogeneity to 27% (excluding Mohanty) and 0% (excluding both Mohanty and Di Biase), while maintaining statistical significance, showing great stability (Figure 1). Subgroup analyses showed similar associations in paroxysmal AF (RR 3.27, 95% CI 2.29-4.69) and persistent AF (RR 2.06, 95% CI 1.01-4.18), with no measurable heterogeneity, and within Farapulse-only cohorts (RR 3.83, 95% CI 1.95-7.56, I² = 59%) (Figure 2). Heterogeneity appeared related to differences in monitoring intensity and PFA platforms, while age and AF type showed less variability. Conclusion In this exploratory pooled analysis, ER showed a consistent association with LR. Given that the evidence is predominantly observational and heterogeneous in monitoring intensity, these results should be interpreted cautiously and considered hypothesis-generating. Further prospective randomized trials are needed to clarify whether PFA may require a tailored definition of the blanking period.