AB-452654-2 EASY AF: ESOPHAGUS DEVIATION DURING RADIOFREQUENCY ABLATION OF ATRIAL FIBRILLATION

Abstract

Thermal injury to the esophagus (Eso) occurs in a high percent of pts undergoing atrial fibrillation (AF) ablation. This international, multicenter (n=12), prospective, randomized, double blinded, controlled trial compared the incidence of ablation-related Eso lesions in pts undergoing an index radiofrequency AF ablation. Pts were randomly assigned to Control group [luminal esophagus temperature (LET) monitoring alone, n=64] or to Deviation group [Eso deviation device + LET n= 56]. The 11mm diameter Deviation device has a distal deflecting arm and holes overlying the arm. Vacuum suction is applied through these holes making the Eso walls adhere to the device; then, the deflecting arm is used to deviate a segment of Eso, including the trailing edge. Deviation was measured on fluoroscopy and used at the discretion of the operator. In the Control group, the LET was moved to measure Eso temperature. Within 72 hours of ablation, endoscopy was performed to assess for any Eso pathology, including Eso injury related to ablation and / or to Deviation device. Endoscopy images were adjudicated by 3 blinded gastroenterologists. The Deviating device was in Eso a mean of 100.3min and suction force applied 24.2min. Total deviations was 6.7/pt. Compared to Control, Deviation group had significantly greater distance between RF and Eso (28.5 vs 9.4mm, p<0.001), and had greater percent of pts with >20mm distance (85% vs 19%, p<0.001). 54% of Control pts had ≤5mm between RF and Eso. Peak change in LET in Control was significantly higher than Deviation: 1.8±1.30 vs 0.7±0.70, p<0.001. The Data Safety and Monitoring Board recommended stopping the study early after randomizing 120 pts due to Deviating device efficacy. The primary study endpoint, ablation injury to Eso mucosa, was significantly less in the Deviation group (5.7%) in comparison to the Control group (35.4%, p < 0.0001). Control pts had significantly higher severity and number of ablation lesions per pt. LET monitoring did not predict Eso ablation lesions. The only adverse event that was blindly adjudicated to be due to the Device was an Eso injury noted in a pt randomized to Control, in whom the device was not inserted. Use of an Eso deviating device, with vacuum suction force and mechanical deviating arm, was able to successfully move a segment of the Eso and its trailing edge, resulting in a significant reduction in ablation-related Eso lesions and without any adverse events.