Pulmonary vein isolation: The impact of pulmonary venous anatomy on long-term outcome of catheter ablation for paroxysmal atrial fibrillation

Background Circumferential pulmonary vein (PV) isolation is the cornerstone of catheter ablation for atrial fibrillation (AF); however, PV reconnection remains problematic. Objective To assess the impact of PV anatomy on outcome after AF ablation. Methods One hundred two patients with paroxysmal AF underwent cardiac magnetic resonance (60%) or computed tomography (40%) before AF ablation. PV anatomy was classified according to the presence of common PVs, accessory PVs, PV branching pattern, and the dimensions of the PV ostia, intervenous ridges (IVRs), and the left PV-left atrial appendage ridge. Results Four discrete PVs were present in 48(47%) of the patients: a left common PV in 38(37%), a right common PV in 2(2%), an accessory right PV in 20(20%), and left PV in 4(4%). At a mean follow-up of 12 ± 4 months, 75 of 102 (74%) patients were free of recurrent AF. A LCPV was associated with an increase in freedom from AF (87% vs 66% for 4 PV anatomy; P = .03). Greater left IVR length (16.9 ± 3.5 mm vs 14.0 ± 3.0 mm; P ≤ .001) and width (1.4 ± 0.6 mm vs 1.1 ± 0.6 mm; P = .02) were associated with increased AF recurrence. After multivariate analysis, abnormal anatomy (LCPV or accessory PV) and left IVR length were found to be the only independent predictors of freedom from AF. Conclusions Four discrete PVs are present in the minority of patients with paroxysmal AF undergoing PV isolation. The presence of a LCPV is associated with an increased freedom from AF after catheter ablation. PV anatomy may in part explain the variable outcome to electrical isolation in patients with paroxysmal AF. Circumferential pulmonary vein (PV) isolation is the cornerstone of catheter ablation for atrial fibrillation (AF); however, PV reconnection remains problematic. To assess the impact of PV anatomy on outcome after AF ablation. One hundred two patients with paroxysmal AF underwent cardiac magnetic resonance (60%) or computed tomography (40%) before AF ablation. PV anatomy was classified according to the presence of common PVs, accessory PVs, PV branching pattern, and the dimensions of the PV ostia, intervenous ridges (IVRs), and the left PV-left atrial appendage ridge. Four discrete PVs were present in 48(47%) of the patients: a left common PV in 38(37%), a right common PV in 2(2%), an accessory right PV in 20(20%), and left PV in 4(4%). At a mean follow-up of 12 ± 4 months, 75 of 102 (74%) patients were free of recurrent AF. A LCPV was associated with an increase in freedom from AF (87% vs 66% for 4 PV anatomy; P = .03). Greater left IVR length (16.9 ± 3.5 mm vs 14.0 ± 3.0 mm; P ≤ .001) and width (1.4 ± 0.6 mm vs 1.1 ± 0.6 mm; P = .02) were associated with increased AF recurrence. After multivariate analysis, abnormal anatomy (LCPV or accessory PV) and left IVR length were found to be the only independent predictors of freedom from AF. Four discrete PVs are present in the minority of patients with paroxysmal AF undergoing PV isolation. The presence of a LCPV is associated with an increased freedom from AF after catheter ablation. PV anatomy may in part explain the variable outcome to electrical isolation in patients with paroxysmal AF.