Advertisement

CA-528-04 ANTERIOR WALL TEMPERATURE OF ESOPHAGUS DURING CATHETER ABLATION OF THE LA POSTERIOR WALL IS MARKEDLY HIGHER COMPARED TO LUMINAL TEMPERATURE

      Background

      Esophageal injuries (ulceration, denuding of tissue or fistula development) are well-known complications from catheter ablation. Few studies have evaluated energy transfer between the posterior wall of the left atrium (LA), interstitium and esophagus.

      Objective

      To study energy transfer and lag time between tissues, we developed a porcine ex vivo heart-esophageal model to evaluate temperatures at critical regions during catheter ablation of the posterior LA wall.

      Methods

      We built a heart-esophageal model to perform ex vivo catheter ablation on the posterior wall of the LA, with juxtaposed interstitial tissue and esophagus. Circulating saline (3.5-5 L/min) was used to mimic blood flow along the LA and alteration of ionic content to vary impedance. Thermistors along the region of interest were used to analyze temperature gradients. Varying time and power, multiple RF ablations were applied with an externally irrigated ablation catheter. Ablation strategies were divided into standard approaches (SA, 25-35W, 30s) or high-power short duration (HSPD, 40-50W, 10s).

      Results

      At contact forces ranging from 10-15g, for both SA and HSPD, maximum temperature rise from baseline was markedly higher at the anterior wall (AW) of the esophagus compared to the esophageal lumen (SA: 3.47°C vs. 0.98°C; HPSD: 2.51°C vs. 0.31°C). Compared to HPSD (Figure), SA approaches exhibited significantly higher temperature rise (relative to baseline) at both the AW (3.47°C vs. 2.51°C, p<0.01) and within the esophageal lumen (0.98°C vs. 0.31°C, p<0.02). For SA, time from ablation onset to a 1°C rise from baseline was 19.2 sec longer in the lumen relative to AW (45.25 sec vs. 26.04 sec, p<0.05). For HPSD, time from ablation onset to a 0.5°C rise from baseline was 34.8 sec longer in the esophageal lumen relative to AW (39.10 sec vs. 4.31 sec, p<0.005).