Association between non-ablated left atrial surface area and rhythm outcome in patients treated with cryoballoon and radiofrequency ablation

Abstract

BACKGROUND Cryoballoon (CB) pulmonary vein isolation (PVI) offers outcomes comparable to radiofrequency PVI (RF-PVI) in patients with atrial fibrillation (AF) but has limitations for wide circumferential PVI and extra-pulmonary vein (PV) trigger (ExPVT) ablations. OBJECTIVE This study aimed to compare long-term outcomes of CB-PVI vs RF-PVI in patients without ExPVT and explore underlying electroanatomical mechanisms. METHODS We identified 1902 patients undergoing de novo AF ablation without ExPVT. After propensity matching for age, sex, AF type, and left atrium anteroposterior (LAAP) diameter in patients, we compared AF recurrence in 403 CB-PVI and 403 RF-PVI cases, considering AFtype and LAAP diameter. Using a Cox model, we identified the optimal LAAP diameter cutoff for differentiating outcomes and examined the relationship between PVI modality and reduction in electrically active LA area via computational modeling. RESULTS During a median follow-up of 24 months, CB-PVI had poorer rhythm outcomes than RF-PVI in propensity-matched patients (log-rank P = .009). Outcomes were comparable in those with an LAAP diameter <40 mm or paroxysmal AF. However, CB-PVI was associated with higher AF recurrence in patients with a LAAP diameter >= 40 mm (hazard ratio [HR] 1.54 [1.01-2.36]; log-rank P = .047) or persistent AF (HR 2.17 [1.36-3.45]; log-rank P = .001). In computational modeling, a larger non-ablated LA area post-PVI was independently related to a higher AF recurrence risk. RF-PVI reduced LA surface area more than CB-PVI, especially in patients with a large LA or persistent AF. CONCLUSION CB-PVI showed inferior rhythm outcomes compared with RF-PVI in patients with a LAAP diameter >= 40 mm or persistent AF, possibly because of a smaller reduction in LA critical mass.