The effect of esophageal cooling on esophageal injury during radiofrequency catheter ablation of atrial fibrillation.

Document Type

Article

Publication Date

6-2020

Publication Title

Journal of interventional cardiac electrophysiology : an international journal of arrhythmias and pacing

Abstract

INTRODUCTION: Catheter ablation of atrial fibrillation (AF) may lead to collateral damage to the esophagus. We tested the hypothesis that luminal esophageal temperature (LET)-guided esophageal cooling might reduce the incidence of esophageal thermal lesions (ETL).

METHODS: Seventy-six patients from August 2015 to March 2017 with paroxysmal or persistent AF underwent a first-time catheter ablation procedure with or without LET-guided active esophageal cooling through an orogastric tube placed in the esophagus. Esophageal cooling occurred if and only if LET exceeded 0.5 °C from baseline while ablating the LA posterior wall. All patients underwent esophagogastric endoscopy the next day.

RESULTS: Of the 76 patients studied, 38 (50%) patients underwent esophageal cooling. Baseline characteristics of the non-cooled and cooled groups were comparable. Of these, 59% of patients had ETL. There was a non-significant trend for more severe lesions (grades 3, 4) in the non-cooled group (29% vs. 13.5%, p = 0.10). Average power delivered on the left atrial posterior wall (27 ± 1.8 W vs. 27 ± 3.8 W, p = 0.34) and average force of contact (10.1 g vs. 9.8 g, p = 0.38) were similar in both groups while more time was spent ablating on the posterior wall in the non-cooled group (24.6 ± 7.3 min vs. 20.4 ± 5.9 min, p = 0.014). In a multivariate analysis, esophageal cooling had no significant effect on the esophageal lesion grade post-ablation.

CONCLUSION: The incidence of ETL in patients undergoing left atrial posterior wall isolation is substantial. Our method of esophageal cooling did not decrease the incidence of ETL. There was a non-significant trend toward fewer severe lesions with cooling, but one cannot conclude the value of cooling from this pilot study.

Volume

58

Issue

1

First Page

43

Last Page

50

DOI

10.1007/s10840-019-00566-3

ISSN

1572-8595

PubMed ID

31154536

Link to Full Text

Share

COinS