HIgh Power short duration radiofrequency ablation or cryoballoon ablation for paroxysmal Atrial Fibrillation (HIPAF trial)

AIMS: Pulmonary vein isolation (PVI) is a first-line treatment option for paroxysmal atrial fibrillation (PAF). Radiofrequency ablation (RFA) or cryoballoon ablation (CBA) are commonly used modalities. Recent studies demonstrated the superiority and potential benefits of very high-power short-duration (vHPSD) RFA using 70 W compared to conventional RFA (<50 W). Prospective randomized data comparing vHPSD RFA with 70 W with the frequently used CBA in the setting of PAF are lacking. METHODS AND RESULTS: We conducted a randomized non-inferiority trial involving 170 patients undergoing de novo PVI for PAF. Patients were randomly assigned in a 1:1 ratio to undergo vHPSD RFA or to receive CBA. The composite primary endpoint consisted of (i) any atrial arrhythmia, (ii) new antiarrhythmic drug (AAD) onset, and (iii) re-ablation during 1 year after index procedure. The non-inferiority margin was predefined as a 10% lower 1-year event-free survival rate in vHPSD compared to CBA (delta = -0.1). A total of 170 patients with symptomatic PAF were enrolled and assigned to undergo de novo PVI, with 84 receiving vHPSD and 86 undergoing CBA. The overall study population had a mean age of 65 ± 11 years and included 50.6% women. For vHPSD PVI a 70 W/7 s anterior and 70 W/5 s posterior protocol including 3D mapping was used. Cryoballoon ablation was performed as usual. Successful PVI was achieved in all patients. Overall procedure time for vHPSD was significantly longer (81.1 ± 20.0 vs. 67.7 ± 17.2 min; P < 0.001). However, the mere ablation time was comparable (39.3 ± 15.5 vs. 36.7 ± 14.5 min; P = 0.285). Fluoroscopy time and amount of contrast medium were significantly lower for vHPSD PVI (9.2 ± 3.6 vs. 10.5 ± 4.3 min; P = 0.031; 15.5 ± 5.8 vs. 43.1 ± 30.0 mL; P < 0.001). Complication rates were comparable between groups. One pulmonary vein stenosis occurred after vHPSD. Three pericardial effusions and two transient ischaemic attack were reported after CBA. After a median follow-up of 367 days, 73.8% [n = 62, 95% confidence interval (CI): 63.1-82.8%] of patients in the vHPSD PVI group and 81.4% (n = 70, 95% CI: 71.6-89.0%) in the CBA group remained free of any event. Non-inferiority of vHPSD PVI compared to CBA PVI could not be demonstrated, with a difference of -0.076 [95% CI: (-0.201 to 0.049)] in event-free survival rates off AADs, as the 95% CI includes the delta of -0.1. CONCLUSION: In this randomized non-inferiority trial comparing vHPSD RFA to CBA for PVI in patients with PAF, non-inferiority of vHPSD RFA could not be shown. Both methods showed comparable safety outcome with a shorter procedure time for CBA.