Background The circumferential pulmonary vein ablation (CPVA) has been proved effective for atrial fibrillation (AF) treatment and is becoming more widely accepted and practiced. This study aims to evaluate the characteristics of the CARTO and the Ensite/NavX system and draw a comparison between them on the aspects of procedural parameters and clinical effectiveness.Methods Seventy-five cases with paroxysmal or chronic symptomatic AF were randomly assigned to CPVA procedure guided by the Ensite/NavX system (group Ⅰ, n=40) and by the CARTO system (group Ⅱ, n=35). After successful transseptal procedure, the geometry of left atrium was created under the guidance of the two systems. Radiofrequency energy was applied to circumferentially ablate tissues out of pulmonary veins (PVs) ostia. In cases with chronic AF, linear ablation was applied to modify the substrate of left atrium (LA). The endpoint of the procedure was complete PVs isolation. Results Seventy-five cases underwent the procedure successfully. The total procedure and fluoroscopic durations in group Ⅱ were significantly shorter than in group Ⅰ [(150±23) min and (18±17) min versus (170±34) min and (25±16) min, P=0.03 and 0.04, respectively]. There was no significant difference in the fluoroscopic and procedure durations for geometry creation between group Ⅰ and group Ⅱ [(8±4) min and (16±11) min versus (5±4) min and (14±8) min, respectively]. The fluoroscopic durations for CPVA were (15±5) min in group Ⅰ versus (10±6) min in group Ⅱ (P=0.05), and the CPVA procedural durations were significantly shorter in group Ⅱ than in group Ⅰ [(18±11) min versus (25±10) min, P=0.04]. AF was terminated by radio frequency delivery in 14 cases (35%) in group Ⅰ versus 5 cases (14%) in group Ⅱ (P=0.035). After CPVA complete PV isolation was attained in 26 cases (65%) in group Ⅰ versus 11 cases (31%) in group Ⅱ (P=0.004). During a mean follow-up of 7 months, 32 (80%) cases in group Ⅰ and 24 (69%) cases in group Ⅱ were arrhythmia-free (P=0.06). One case developed pericardium effusion and another one case was found to have intestinal artery thrombosis in group Ⅱ. One case had moderate hemothorax in group Ⅰ. All the complications were cured by proper treatment. No PV stenosis was observed. Conclusions The CPVA procedure for atrial fibrillation is effective and safe. Although there is difference between the CARTO and the Ensite/NavX system, the CPVA procedure guided by either of them yields similar clinical results.