Impact of variant pulmonary vein anatomy and image integration on long-term outcome after catheter ablation for atrial fibrillation

To investigate the impact of variant pulmonary vein (PV) anatomy and the use of three-dimensional image integration (3D-II) on long-term efficacy of catheter ablation for atrial fibrillation (AF). Consecutive procedures from 2002 to 2007 were analysed from a prospective database. All patients underwent wide area circumferential ablation, with linear lesions added and complex fractionated electrograms targeted for persistent AF. Imaging was segmented on Carto to assess PV anatomy. Three hundred and fifty patients underwent 1.9 +/- 0.9 procedures. The mean age was 57 +/- 11 years, 73% males, and 55% paroxysmal AF. Freedom from AF/atrial tachycardia was 42% for paroxysmal AF and 20% for persistent AF at 3.1 years after the first procedure, or 86 and 66%, respectively, at 2.5 years after the last procedure. The Kaplan-Meier analysis showed a trend towards improved single-procedure efficacy with 3D-II (8.9% difference, P = 0.087) and a reduction in the number of procedures per patient from 2.1 +/- 1.1 to 1.8 +/- 0.9 (P < 0.0001). The use of 3D-II improved single-procedure efficacy with Carto (13.3% difference, P = 0.018), but not with Ensite NavX. Variant PV anatomy was identified in 28% and was associated with a lower single-procedure efficacy (10.0% difference, P = 0.024) but with no effect on final outcome. Multivariate analysis confirmed the impact of 3D-II [hazard ratio (HR) for recurrence of AF 0.67, P = 0.020] and variant PV anatomy (HR 1.37, P = 0.044). The use of 3D-II improves single-procedure efficacy of PV isolation for AF. Variant PV anatomy was associated with a lower single-procedure success rate.