INTRODUCTION Isolation of all pulmonary veins (PV) is advocated for treatment of paroxysmal atrial fibrillation (PAF). However, the superior PVs are responsible for most AF triggers, whereas the inferior PVs carry the higher risk for ablation-induced ostial stenosis. The aim of this study was to compare a superior PV isolation approach with isolation of all PVs for treatment of PAF. METHODS AND RESULTS Fifty-two patients with PAF were randomized to either left superior pulmonary vein (LSPV) isolation followed by additional isolation of the right superior pulmonary vein (RSPV) in case of AF recurrence (group A, n = 27) or isolation of all four PVs followed by a repeat procedure in case of recurrence (group B, n = 25). At 1-year follow-up, 11 patients (41%) in group A and 8 patients (32%) in group B had AF relapse (P = 0.55). No significant differences in AF relapse were detected between groups at 3 and 12 months (log rank = 0.36, P = 0.54) and by Cox proportional hazards model analysis (P = 0.62). Nonsignificant PV stenosis was detected in two patients from group B. Total radiofrequency energy delivery and fluoroscopy and procedure times were lower in group A: 8.9 +/- 1.4 minutes vs 25.6 +/- 3.7 minutes (P < 0.001), 22.2 +/- 6.8 minutes vs 62 +/- 10.3 minutes (P < 0.001), and 131.8 +/- 26.5 minutes vs 222.2 +/- 32.3 minutes (P < 0.001), respectively. CONCLUSION A staged superior PVs isolation approach confers equal success rates but with reduced radiofrequency energy delivery and fluoroscopy and procedure times compared to isolation of all PVs at the initial ablation attempt.