• Thomas Deneke, Karin Nentwich, Artur Berkovitz, Kai Sonne, Elena Ene, Borche Pavlov, Franziska Fochler, Anne-Katrin Bötsch, Rainer Kuhn, and Philipp Halbfaß.
• Heart Center Bad Neustadt, Clinic for Interventional Electrophysiology, Germany (T.D., K.N., A.B., K.S., E.E., B.P., F.F., A.-K.B., P.H.).
• Circ Arrhythm Electrophysiol. 2018 Nov 1; 11 (11): e006681.

BackgroundEndoscopically detected thermal esophageal lesions (EDEL) after ablation of atrial fibrillation may be precursors of atrioesophageal fistula and esophageal luminal temperature monitoring has previously failed to decrease thermal damage.MethodsSixty-three patients undergoing their first pulmonary vein isolation using radiofrequency point-by-point catheter ablation were prospectively included in the HEAT-AF study (High-Resolution Esophageal Assessment of Esophageal Temperature During Atrial Fibrillation Ablation) and esophageal temperatures were continuously monitored using a novel infrared thermography system (IRTS). Peak esophageal temperature (Tpeak) was correlated to postablation endoscopy results characterizing patients as EDEL positive or negative.ResultsTwelve patients had EDEL (19%). Comparing EDEL positive to negative patients, Tpeak was significantly higher (56.3±4.6°C versus 45.7±5.5°C, P<0.0001). Logistic regression analysis demonstrated Tpeak was a statistically significant predictor ( P=0.0008) of EDEL and yielded an odds ratio of 1.52; 95% CI, (1.24-2.05). Receiver operator curve analysis demonstrated Tpeak as a highly accurate binary classifier with an area under the curve of 93%.ConclusionsFor the first time esophageal temperature monitoring using a high resolution, high-fidelity IRTS allowed accurate prediction of postablation EDEL suggesting that Tpeak alone is an excellent binary classifier of patients at risk of EDEL. The logistic regression model and associated receiver operator curve will aid in the selection of optimal temperature thresholds in future prospective studies.

29 keywords...

hide…