B-PO03-145 POST LVAD ATRIAL ARRHYTHMIAS IN PATIENTS WITH NO PRIOR HISTORY OF ATRIAL ARRHYTHMIAS

Data on incidence of post LVAD atrial arrhythmias (AA) in patients with no prior history of AA remain sparse. We sought to assess the incidence and predictors of de novo AA in continuous flow LVAD patients. A multi-center LVAD arrhythmia registry was queried for patients with no history of AA before LVAD implant. AA was defined as atrial tachycardia, flutter, or fibrillation lasting >6 hours or >1% burden. Univariate demographic and pre-operative risk factors were evaluated for association with post-LVAD AA. A multivariate regression model was used to identify independent factors associated with incidence of post-LVAD AA. A total of 298 patients (56±14 years, 73% male) with no history of AA underwent LVAD implantation. Patients with post LVAD AA (31%, n=92) were older (61 vs 58 years). There was no statistically significant difference between post LVAD AA and No AA groups in terms of gender (female 44% v 39%, p=.5), COPD (42 v 41%, p=.8), sleep apnea (43 v 40%, p=.6) or diabetes (42 v 51%, p=.2). Patients with higher INTERMACS profiles had a higher incidence of post LVAD AA (for class 1, 2, 3, 4+ respectively 33%, 28%, 35%, 56%, p=.04) whereas those with pulmonary hypertension (PHTN) were less likely to have post LVAD AA (33% v 50%, p=.01). The regression model identified higher INTERMACS class as an independent predictor of the post LVAD AA (OR=1.36, p=.02), Table 1. Patients with no history of AA were more likely to have de novo AA post LVAD with increasing INTERMACS profile and age whereas presence of PHTN was associated with lesser incidence of post LVAD AA. Gender, BMI, presence of COPD or sleep apnea were not significant predictors of post LVAD AA.