PO-701-06 PREDICTORS OF SUDDEN CARDIAC DEATH AND ARRHYTHMOGENICITY IN PATIENTS WITH MARFAN`S SYNDROME

Abstract

Marfan´s syndrome caused by FBN1 mutation is associated with sudden cardiac death and ventricular arrhythmias. Data evaluating predictors for these events is scarse. We evaluated predictors of rhythm events and sudden cardiac death in patients with Marfan`s syndrome. In this prospective study 94 adult marfan patients (60 % male, 42 ± 15 years) with FBN1 mutation and a follow up of at least 1 months (median: 5 years, IQR: 3.1 - 10) were included. Baseline serum N-terminal pro-brain natriuretic peptide (NT-proBNP), transthoracic echocardiogram, 12-lead resting ECG and epidemiological parameters were assessed for further analysis and predictor selection. Primary endpoint was composed of ventricular tachycardia, arrhythmogenic syncope or sudden cardiac death. In a mean follow-up time of 6.1 ± 4 years 15 (16 %) marfan patients suffered from rhythm event or sudden cardiac death.Left atrial volume, age, mitral-valve prolaps, impaired left ventricular ejection fraction, NT-proBNP, gender, QRS duration, T-peak-T-end intervals with derivatives and aortic bulbus diameter index were included in further model selection process. We then performed statistical learning methods with best subset selection for model evaluation to identify predictors of composite endpoint. Subsequent multivariate Cox-regression analysis revealed hazard ratios. LA volume (ml) (HR 1.03 per ml, p<0.001), impaired left ventricular ejection fraction (HR: 41.7, p<0.001), male-gender (HR: 4.5, p=0.06), QRS duration (HR: 1.03 per ms, p=0.02) and TpTe maximum (HR: 1.08 per ms, p=0.02) were identified as independent predictors for the combined primary endpoint. NT-proBNP was highly correlated to LA volume and therefore was not significant in the multivariate model. In a statistical learning model combination of echocardiographic and electrocardiographic parameters performed best in prediction power of rhythm events and sudden cardiac death in patients with Marfan´s syndrome.