Introduction:
Despite recent tremendous developments in atrial fibrillation (AF) catheter ablation by the advent of pulsed field ablation (PFA), radiofrequency ablation (RFA) remains one of the standard ablation technologies at least for the near future. Relevant improvements of RFA strategies and catheter technology helped to overcome important safety issues. However, in contrast to PFA, radiofrequency ablation at left atrial posterior wall (LAPW) is associated with a certain risk of esophageal injury.
The aim of the study was to identify risk factors predicting ablation-induced endoscopically detected esophageal lesions (EDEL) in patients undergoing high-power and very high-power short-duration atrial fibrillation RFA.
Methods and Results:
Consecutive patients undergoing AF RFA and post-ablation esophageal endoscopy (EE) were included at the Department of Interventional Electrophysiology, University Hospital Oldenburg, from March 2021 to November 2025. Patients undergoing LA catheter ablation without ablation at LAPW were excluded.
Ablation procedures were performed according to our standard approach using mostly propofol sedation. Two different types of ablation catheters developed for the use of different maximum power levels were used (Smart Touch SF and QDOT Micro, Biosense Webster, CA, USA). Catheter ablation (CA) was performed using either high-power short-duration (HPSD) ablation (Smart Touch SF: 50 W, ablation index guided with an target ablation index of 320, 350 or 380 at all LAPW sites) or very-high-power short-duration (vHPSD) ablation (QDOT Micro: 90 W, 4 sec at all LAPW sites) following the CLOSE protocol.
A total of 1003 consecutive patients (age 66.6±9.8 years, 58% male, 42% paroxysmal AF, 27% redo procedures) undergoing AF ablation and planned for an EE after CA were included in the analysis. A total of 49 patients (4.9%) demonstrated an endoscopically detected esophageal lesion (EDEL) in postprocedural endoscopy (EDEL category 1: 42 (4.2%) patients; category 2: 7 (0.7%)). Use of very-high-power 90 W ablation at LAPW sites was associated with a significantly lower rate of EDEL compared to AI-targeted 50 W ablation (OR 0.310; CI 0.144 - 0.669; p=0.003).
Conclusion:
The incidence of EDEL was 5% in the total study cohort and significantly lower in patients undergoing PVI using 90 W vHPSD ablation (2.1% EDEL) compared to 50 W HPSD ablation (6.6% EDEL).
