P672 Predictors of ventricular arrythmias and mortality after implantation of primary prevention antitachycardia devices

Abstract Background Patients (pts) with reduced left ventricular (LV) systolic function have high risk of sudden cardiac death and benefit from implantable cardioverter-defibrillators (ICDs/CRT-Ds). However, the risk for arrhythmic events and device therapies is extremely heterogeneous in this population, so more accurate tools for risk stratification are required. Purpose To assess predictors of mortality and arrhythmic events in pts receiving primary prevention ICDs/ CRTs. Methods Retrospective analysis of 150 pts submitted to primary prevention ICD/ CRT-D implantation with remote monitoring between 2014-2018. Demographic, clinical and echocardiographic data from implantation and follow-up period were retrieved. Arrhythmic events and device therapies were retrieved from remote monitoring and clinic visits. Univariate analysis was performed followed by a multivariate Cox analysis to evaluate predictors of events. p < 0.05 were considered significant. Results 150 pts, 80.7% male, with a mean age of 64.30 ± 12.9 years (Y) and a mean follow-up (FU) time of 38 ± 15 months. 66% of pts implanted an ICD. 52.0% of pts presented with an ischemic cardiomyopathy and 41.3% had atrial fibrillation. 35.3% had chronic kidney disease (GFR < 60mL/min) and 24.0% were diabetic. Mean BNP value of 449.6 ± 631.3pg/mL and mean peak VO2 of 15.3mL/kg/min. Mean LV ejection fraction (LVEF) during FU of 35.9 ± 12.1% and a mean average global longitudinal strain (GLS) of -8.7 ± 5.5%. 63pts (42.0%) suffered a ventricular arrhythmia, mostly non-sustained ventricular tachycardia, of which 47.6% received appropriate therapies. Mortality rate of 13.3% during follow-up (20 pts). Baseline diabetes (p = 0.040) and post-procedural pulmonary artery systolic pressure (PASP) (p = 0.002) were independent predictors of overall mortality in the follow-up. Male gender (p = 0.041), baseline diabetes (p = 0.011) and atrial fibrillation (p = 0.038) were associated with ventricular events. In patients with CRT-D, a percentage of biventricular pacing superior to 95% was found to be protective against ventricular arrhythmias. Interestingly despite being associated with a higher overall mortality (p = 0.028), a reduced LVEF wasn’t related to the arrhythmic burden of our population, neither the GLS nor the LV mechanical dispersion were predictors of ventricular arrhythmias. Conclusion Baseline diabetes and PASP were independent predictors of mortality in our population of ICD/CRT-D pts implanted in primary prevention setting. An increased percentage of biventricular pacing was associated to improved clinical outcomes in patients receiving cardiac resynchronization therapy. Identification of predictors of events in this population can help individualize its management.