Ab-452663-2 left atrial box isolation guided by force-power-time formula (ablation index): high prevalence of epicardial conduction via septopulmonary bundle in atrial fibrillation patients

Circumferential, continuous, transmural atrial lesions are desired to facilitate pulmonary vein (PV) isolation. However, atrial wall thickness around the PV is variable (5-6 mm in the superior/anterior segments and 2-4 mm in the inferior/posterior segments). Previous experimental studies have shown that radiofrequency (RF) lesion depth and diameter can be predicted by a logarithmic formula using contact force (CF), RF power and application time (Force-Power-Time Formula, Ablation Index, AI). The purpose of this study was to test the ability of AI guided ablation to achieve left atrial (LA) box isolation by a single RF encirclement and examine the location of residual conduction (gap) within the box in patients with atrial fibrillation (AF). In 42 patients with AF, LA box isolation was performed using a 7.5Fr CF catheter (ThermoCool SmartTouch SF, Biosense Webster) and CARTO system. Using RF power (35-40Watts) and CF (5-45g), a target AI value of 380-400 was used for posterior/inferior segments with an inter-lesion distance (ILD) of 4mm and a target AI value of 550-600 for anterior/superior/septal segments with an ILD of 6mm along the box circumference (Figure). RF was delivered until AI reached target values or terminated prematurely when esophageal temperature reached <38.5°C. After the first encirclement along the box line, activation mapping was performed to confirm box isolation and identify gap locations. Total RF time for box isolation was 27.2±4.6 min and box circumference length was 21.2±2.5 cm. Complete box isolation was obtained by the first encirclement in 17/42 (40%) pts. In the remaining 25/42 (60%) pts, activation mapping within the box lesion set identified earliest atrial activation along the ablation line in only 4/42 (10%) pts (indicating endocardial gaps), whereas activation mapping within the box identified earliest atrial activation at a site distant from the ablation line in 21/42 (50%) pts, suggesting epicardial conduction through a septopulmonary bundle (Figure). Focal ablation at endocardial gaps along the ablation line or at the epicardial focal breakouts within the box resulted in compete box isolation in all 25 pts. LA box isolation guided by AI effectively prevented endocardial gaps along the ablation line. However, epicardial conduction via a septopulmonary bundle was often observed (50%).