Recordings of diaphragmatic electromyograms during cryoballoon ablation for atrial fibrillation accurately predict phrenic nerve injury

Abstract

Background: Cryoballoon ablation has been associated with a significant incidence of phrenic nerve injury (PNI).

Objective: The purpose of this study was to evaluate whether recordings of diaphragmatic compound motor action potentials (CMAP) on a modified lead I during cryoballoon ablation can predict PNI.

Methods: Cryoballoon ablation was performed in 109 patients with atrial fibrillation (AF). During ablation of the right-sided pulmonary veins, the phrenic nerve was paced from the superior vena cava. The right and left arm electrodes from a 12-lead ECG were positioned 5 cm above the xiphoid process and 16 cm along the right costal margin. The amplitude of CMAP was recorded on lead I during ablation.

Results: Cryoballoon was applied 424 times in 211 right-sided veins. PNI occurred in 7 (6.4%) patients. The average CMAP amplitude did not significantly change in patients without PNI from the initial average CMAP amplitude of 0.34 ± 0.18 mV to 0.32 ± 0.17 mV (P = .58). In patients who developed PNI, there was a significant decrease in the initial average CMAP amplitude during the ablation from 0.33 ± 0.14 mV to 0.09 ± 0.05 mV (P <.001). The maximal percent change in the average CMAP amplitude in patients with PNI was higher (70% ± 10%) than in patients without PNI (7.6% ± 7%; P <.001). In any patient without PNI, the CMAP amplitude did not decrease more than 35% from baseline.

Conclusion: Recording of CMAP amplitude on a modified lead I is reliable and could be early and sensitive method for predicting PNI in patients undergoing cryoballoon ablation for AF.

Keywords: Atrial fibrillation; Cryoablation; Phrenic nerve injury.