Are routine cryoballoon procedural characteristics predictive of atrial arrhythmia recurrence in the long term?

BACKGROUND: Cryoballoon ablation is an effective procedure to treat atrial fibrillation (AF). However, AF recurrence rate at 1-year follow-up is approximately 20% despite improvements in ablation technique. AIM: To find factors predictive of AF recurrence following a first pulmonary vein isolation procedure using a second-generation cryoballoon (PVI-2CB). METHODS: This was an observational, retrospective, single-centre study. From June 2012 to April 2017, all patients who had a PVI-2CB procedure and a scheduled follow-up at Rouen University Hospital were included. The primary endpoint was atrial arrhythmia (AA) recurrence (e.g. AF, flutter or tachycardia), considering a blanking period of 2 months following the procedure. Secondary endpoints were procedural variables for each pulmonary vein (successful isolation, time to disconnection, total cryoballoon application time, number of cryoballoon applications, level of occlusion during cryoballoon application leading to successful disconnection and lowest temperature reached during successful cryoballoon application), occurrence of redo procedures, use of antiarrhythmic drugs and adverse events. RESULTS: The initial population consisted of 239 patients; six were excluded for lack of procedural variable data, giving an analysed population of 233 patients. The AA recurrence rate was 36.9% (mean follow-up 25±14 months). Mean time to AA recurrence was 10±12 months. No procedural variable was found to be predictive of AA recurrence. Only major left atrial enlargement (defined as diameter>50mm or left atrial area>30cm2 or left atrial volume>50mL/m2) was predictive (odds ratio 2.70, 95% confidence interval 1.54-4.72; P=0.001). Forty-one patients had redo procedures (17.6% of analysed population); in this subgroup, 75.6% had at least one pulmonary vein reconnected, mainly the right inferior pulmonary vein. CONCLUSIONS: At long-term follow-up, up to one-third of patients had AA recurrence after PVI-2CB. Important atrial remodelling was the only factor predictive of AA recurrence, whereas no procedural variable was found to be predictive.