PO-636-04 ABLATION OF ATRIAL FIBRILLATION BEYOND PULMONARY VEIN ISOLATION: DO ADDITIONAL ABLATION LESIONS IMPACT LEFT ATRIAL FUNCTION?

Abstract

Electrical isolation of pulmonary veins (PVI) is a cornerstone for Atrial Fibrillation (AF) ablation therapy. Elimination of non-pulmonary vein (NPV) triggers in addition to PVI was shown to decrease long-term AF recurrence. The overall effect of AF ablation on left atrial (LA) function is poorly understood and it is unknown if additional ablation lesions can affect it. Our aim was to determine if LA function is different in patients after extensive LA ablation compared to PVI only. We hypothesized that addition of NPV ablation lesions in LA do not further deteriorate LA function compared to PVI alone. Out of consecutive 994 patients who underwent AF ablation at our center in years 2018-2019, we included 68 patients in our retrospective analysis who had echocardiograms (TTE) performed within 12 months prior to AF ablation and 1-12 months after. Redo ablations, history of mitral valve interventions were excluded. Patients were stratified into 2 groups: PVI only and PVI with additional LA ablation lesions (PVI+). Primary outcome was LA reservoir strain (LASr). We applied non-inferiority analysis with 90% CI for an overall alpha level of 0.05. Mean LASr in patients after AF ablation is reported 19 ±9%, we used a conservative 6% to define non-inferiority as a change in LASr. Patients in our study cohort had higher rates of history of HTN, HFrEF, DM, HLD, ESRD compared to all patients who underwent AF ablation. A higher rate of paroxysmal AF in the PVI only group was noted (70% vs 30%). The PVI+ group was observed to have a slightly higher increase in LASr compared to PVI alone (5.0%vs 4.3%), with 90% CI (-4.2 to 2.9). The upper bound for the true difference of 2.9% did not cross the pre-set margin of 6% (p<0.01 for test of non-inferiority). These findings were consistent when the LASr was adjusted for age, sex, hx of CAD, HLD, paroxysmal vs persistent AF, rhythm at pre-procedure TTE in a multivariable linear regression model, 90% CI (-5.46; 2.04), p<0.01. LA functional improvement evaluated by LASr is statistically non-inferior after PVI with additional LA ablation lesions compared to PVI alone. These findings were confirmed when adjusted for confounding clinical variables.