Effectiveness of esophagus detection by three-dimensional electroanatomical mapping to avoid esophageal injury during ablation of atrial fibrillation

Aims: Esophageal-left atrial (LA) fistula during atrial fibrillation (AF) ablation is a fatal event. We explored the relation of the esophagus-to-ablated point distance and esophageal temperature rise.

Methods: Consecutive patients (n=106) underwent complex fractionated atrial electrogram-guided AF ablation using CartoMerge; the pulmonary veins were isolated in 23 patients. Maximum radiofrequency (RF) power near the esophagus was 15 W. Ablated points with esophageal temperature rise (monitored with a probe) to ≥38.0°C were tagged; if ≥39.0°C, RF was discontinued.

Results: Of 1647 ablated points near the esophagus, 274 were associated with a temperature rise to 38.0-38.9°C and 241 points to ≥39.0°C. Distances (mm) from points to esophagus were 5.1 ± 0.6 (no rise), 4.2±3.1 (38.0-38.9°C), 2.9 ± 2.5 (≥39.0°C). Altogether, 15.5% of points in the upper LA posterior wall, 41.5% in the middle, and 30.2% in the lower caused rises to ≥38.0°C; 8.7%, 24.6%, and 11.0% caused rises to ≥39.0°C. The middle wall was most affected (p<0.01), as shown by multiple logistic regression analysis (both temperatures). Points causing a rise increased significantly as distance decreased (p<0.001). The odds ratio for rise to ≥38.0°C compared with <4.0 to >5.0 mm distance was 2.28 (p=0.004). The longest distance for ≥38.0°C rise was 18.5 mm.

Conclusion: Distance is an important predictor of esophageal temperature rise. The middle LA posterior wall is most vulnerable. A dose of 15 W is too high for ablation, especially <4.0 mm from the esophagus. Points >20.0 mm away are relatively safe.