Multicenter propensity score-matched comparison of Farapulse, cryoballoon, and radiofrequency ablation for paroxysmal atrial fibrillation with 2-year follow-up
G Vetta, A Almorad, M Magnocavallo, C Audiat, V Schillaci, C Monaco, J Osca, L Marcon, J Sieira, P Rossi, P Jais, G B Chierchia, C De Asmundis, D G Della RoccaAbstract
Introduction
Pulmonary vein isolation (PVI) represents the cornerstone of rhythm control in paroxysmal atrial fibrillation (PAF). Thermal ablation modalities—radiofrequency (RF) and cryoballoon (CRYO)—are effective but limited by collateral tissue injury and variable durability. Pulsed field ablation (PFA), a non-thermal electroporation-based technology, has demonstrated favourable safety and efficacy in pivotal randomized trials. However, long-term comparative data from real-world multicenter settings remain limited. This study evaluates procedural performance, safety, and 2-year efficacy of PFA compared with CRYO and RF ablation in a large propensity-matched cohort of PAF patients.
Methods
Consecutive patients undergoing first-time PVI with PFA, CRYO, or RF across eight European centers (July 2022–October 2024) were included. Propensity score matching was applied in a 1:2:2 ratio (PFA:CRYO:RF) to balance demographic and clinical covariates. Primary endpoints was freedom from any atrial tachyarrhythmia at 2-year follow-up. Safety endpoints included major and minor procedural complications.
Results
Among 2897 patients (mean age 58.6 ± 10.4 years; 39.3% female), 1320 were propensity-matched (PFA n=264; CRYO n=528; RF n=528). Acute PVI was achieved in all 5240 pulmonary veins. First-pass isolation was significantly higher with PFA (98%) compared with CRYO (90.6%) and RF (89.8%) (p<0.001). Procedural (57 vs 64.5 vs 75 min) and left atrial dwelling times (36.5 vs 45.8 vs 55 min) were significantly shorter with PFA (all p<0.001), whereas RF showed the lowest fluoroscopy time (p<0.001).
Major complications were low and comparable (PFA 1.5%, CRYO 1.1%, RF 0.9%; p=0.77). Overall and minor complications were numerically lower with PFA (4.2% and 2.7%) than with CRYO (8.0% and 6.8%) and RF (5.5% and 4.5%), driven by the absence of thermal-related phrenic nerve palsy (exclusive to CRYO) and reduced pericarditis (more common with RF).
At 2-year follow-up, freedom from any atrial tachyarrhythmia was significantly higher with PFA (75.8%) compared with CRYO (68.0%) and RF (66.7%) (log-rank p=0.022)(Figure 1). Freedom from atrial fibrillation was also superior with PFA (83.3% vs 72.7% vs 72.5%; p=0.001) (Figure 2). Among patients undergoing redo procedures, pulmonary vein reconnection was significantly lower following PFA (20.9%) compared with CRYO (27.5%) and RF (34.9%) (p=0.005), with markedly reduced reconnection at the right inferior pulmonary vein.
Conclusions
In this multicenter propensity-matched analysis, PFA demonstrated superior procedural efficiency, fewer minor adverse events, and enhanced long-term rhythm outcomes compared with CRYO and RF ablation.Figure 1Figure 2