ABSTRACT
INTRODUCTION: A hybrid convergent approach (endocardial and epicardial ablation) demonstrated superior effectiveness in a recent randomized study for long-standing persistent atrial fibrillation (LSPAF). Yet, there is a lack of real-world, long-term evidence as to which patients are best candidates for a hybrid convergent approach compared to standard endocardial cryoballoon pulmonary vein isolation (CB PVI). METHODS AND RESULTS: This single-center, retrospective analysis spanning from 2010 to 2015 compared two distinctly different atrial fibrillation (AF) cohorts; one treated with stand-alone cryoablation and one treated with a hybrid convergent approach. Baseline characteristics described candidates for each approach. The following criteria were utilized to determine CB PVI candidacy: (1) paroxysmal AF (PAF) (stage 3A) with failed class I/III antiarrhythmic drug (AAD) or (2) persistent/LSPAF (stage 3B/3C/3D) with failed class I/III AAD unwilling to undergo hybrid procedure. Selection criteria for the hybrid procedure included: (1) PAF refractory to both class I/III AAD and prior CB PVI (stage 3D) or (2) persistent/LSPAF (stage 3B/3C/3D) with failed class I/III AAD agreeable to hybrid procedure. Prior sternotomy was excluded. Serial electrocardiograms and continuous monitoring evaluated primary efficacy outcome of time-to-first recurrence of atrial arrhythmia after a 90-day blanking period. Secondary outcomes were procedure-related complications and AAD use (at discharge, 12, and 36 months). Kaplan-Meier methods evaluated arrhythmia recurrence. Of 276 patients, 197 (64.2 ± 10.6 years old; 66.5% male; 74.1% 3A-PAF; 18.3% 3B/3D-persistent AF; 1.0% 3C-LSPAF; 6.6% undetermined) underwent CB PVI and 79 (61.4 ± 8.1 years old; 83.5% male; 41.8% 3D-PAF; 45.5% 3B/3D-persistent AF; 12.7% 3C/3D-LSPAF) underwent hybrid procedure. Arrhythmia freedom through 36 months was 55.2% for CB PVI and 50.4% for hybrid (p = .32). Class I AAD utilization at discharge occurred in 38 (19.3%) patients in the CB PVI group and 5 (6.3%) patients in the hybrid group (p = .01). CB PVI class I AAD utilization at 12 months occurred in 14 (9.0) patients versus 0 patients for hybrid convergent (p = .004). Patients with one or more adverse event were as follows: two (1.0%) in the CB PVI group (both transient phrenic nerve palsy) and three (3.7%) in the hybrid group (two with significant bleeding and one with wound infection) (p = .14). CONCLUSION: This study demonstrated that patients with more complex forms of AF (3D-PAF or 3B/3C/3D-persistent/LSPAF) could be well managed with a convergent approach. In a real-world evaluation, outcomes match safety and efficacy thresholds achieved for patients with earlier, less complex AF etiologies treated by CB PVI alone.