Background Phrenic nerve injury (PNI) is recognized as an important complication during atrial fibrillation ablation. This study aimed to investigate the incidence and outcome of PNI during superior vena cava isolation (SVCI) and circumferential pulmonary vein isolation (CPVI) using radiofrequency (RF) energy and the factors associated with its occurrence. Methods and results Five hundred sixty-seven consecutive patients who underwent SVCI after CPVI without substrate modification who completed a 12-month follow-up were retrospectively analyzed. Point-by-point RF applications were applied with maximum energy settings of 35 W and 30 seconds for the SVCI. In the former 210 patients, sites where pacing captured the PN were avoided whenever possible; however, the maximum power was 35 W. In the latter 357 patients, RF energy was delivered regardless of PN capture; however, the power at PN capture sites was limited to 10 W during continuous diaphragmatic movement monitoring on fluoroscopy. Circumferential pulmonary vein isolation and SVCI were successfully achieved in all. Twelve patients (2.1%) had PNI during SVCI but not during CPVI. Phrenic nerve injury completely recovered in all patients a median of 8.0 months after the procedure. The prevalence was higher in the former period (3.8% vs 1.1%; P = .03). A multivariate logistic regression analysis revealed that the study period (odds ratio 3.546; 95% CI 1.051-11.965; P = .041) was the sole independent predictor for identifying patients with PNI during SVCI. Conclusions Phrenic nerve injury occurred in 2.1% of the patients. All occurred during SVCI but not during contemporary CPVI. Energy titration and continuous diaphragmatic movement monitoring significantly decreased the incidence during SVCI.