Phrenic nerve palsy during ablation of atrial fibrillation using a 28-mm cryoballoon catheter: predictors and prevention

Abstract

The purposes of this study were to determine whether predictors of phrenic nerve palsy (PNP) exist and to test whether a standardized ablation protocol may prevent PNP during cryoballoon (CB) ablation using the 28 mm CB. Three-dimensional (3D) geometry of the pulmonary veins (PV) and their relationship to the superior vena cava (SVC) was analyzed. Phrenic nerve (PN) stimulation was performed during ablation of the right-sided PVs with a 28-mm CB. The freezing cycle was immediately terminated in case of loss of PN capture. Sixty-five patients (age, 58 ± 11 years; ejection fraction, 0.59 ± 0.06; left atrial size, 40 ± 5 mm) with paroxysmal atrial fibrillation were included. No persistent PNP was observed. Transient PNP occurred in 4 of 65 patients (6 %). PN function normalized within 24 h in all four patients. A short distance between the right superior PV and the SVC was significantly associated with PNP, but left atrial and 3D PV anatomy were not. Low temperature early during the freezing cycle (<-41 °C at 30 s) predicted PNP with a sensitivity and a specificity of 100 and 98 %, respectively. The anatomical relationship between the right superior PV and the SVC is a preprocedural predictor for the development of transient PNP, and low temperature early during ablation at the right superior PV is a sensitive warning sign of impending PNP. Despite the use of the 28 mm CB, transient PNP occurred in 6 % of patients undergoing CB ablation.