Some conflicting results of the efficacy of the inducibility test used in the catheter ablation of atrial fibrillation (AF) have been reported. The aim of this study was to investigate the inducibility and efficacy of circumferential ablation with pulmonary vein isolation (PVI) in patients with paroxysmal AF and its relationship to the atrial substrate. This study consisted of 88 patients with paroxysmal AF who underwent catheter ablation. Electroanatomic mapping using a NavX system was performed and the biatrial voltage was obtained during sinus rhythm. After successful circumferential ablation with PVI, an inducibility test was performed to determine the requirement for creating left atrial (LA) ablation line. After procedure, patients with inducible AF had a higher recurrence rate than did those with noninducibility of AF (55% vs 18%, P = 0.02). The patients with inducible AF after the PVI had a lower biatrial voltage than did those with negative inducibility. The patients with inducible AF after the final procedure who had a recurrence had a lower LA voltage (1.3 +/- 0.4 vs 1.8 +/- 0.4 mV, P = 0.02) and longer LA total activation time (99 +/- 18 vs 88 +/- 13 msec, P = 0.02) than did those with noninducible AF and no recurrence. None of the patients had occurrence of LA flutter during the follow-up. After a single procedure of circumferential ablation with PVI and noninducibility, 82% patients did not have recurrence of AF. The inducibility of AF was related to the recurrence of AF. The atrial substrate affected the outcome of the inducibility.