Abstract 17147: Prognostic Significance of Atrial Fibrillation in Acute Coronary Syndromes. A Propensity Score Matching Analysis

Abstract

Introduction: Conflicting data exits with respect to the prognostic implication and management of atrial fibrillation (AF) in acute coronary syndromes (ACS). Some studies suggest AF as a mortality confounding factor in ACS that reflects its association with heart failure. We sought to evaluate the real involvement of AF in mortality.

Methods: We designed a prospective study of patients from 40 Andalucian centers, previously included in ARIAM registry (Análisis del Retraso en el Infarto Agudo de Miocardio) from 2001 to 2011. Non adjusted and Adjusted Cox multivariate models were conducted. Propensity score matching (PSM) analysis was performed to adjust baseline characteristic of AF group (AFg)

Results: From 39237 patients analyzed, 2852 patients (7,3 %) presented AF. AFg were older (71±9,8 vs 63±12; p<0,00001), more frequently women (34,6 vs 25,6%; p< 0,00001), more likely had cardiovascular risk factors at baseline, and had an adverse outcome during hospitalization: malignant arrhythmias (8,9 vs 3,7%; p<0,0001), cardiogenic shock (16,7 vs 2,8 %; p<0,00001) and in-hospital mortality (12,4 vs 5,1 %; p<0,00001). In non-adjusted Cox model, AF (HR 2,16; p<0,0001), age (HR 1,82; p<0,0001), diabetes (HR 1,65; p0,0001), cardiogenic shock (HR 22,38; p<0,0001) and heart rate (HR) at admission (HR 1,18; p=0,005) were predictors of in-hospital mortality. Ejection fraction (EF) (HR 0,67; p=0,006, every units of 5%) and beta-blockers (BB) use (HR 0,2; p<0,00001) were protective variables. In adjusted model, age, HR, malignant arrhythmias, BB use, and cardiogenic shock persisted as an independent predictors of mortality. After removing the two variables of greatest prognostic impact (shock and arrythmias), only new-onset AF continued as an independent predictors of mortality (HR 1,81; p<0,0001). In adjusted Cox model of 3816 PSM patients, only AF (HR 1,32; p=0,006), cardiogenic shock (HR 15,6; p<0,00001),and malignant arrhythmias (HR 1,35; p=0,009) remained as independent predictors of in-hospital mortality.

Conclusions: AF during ACS is associated with a higher rate of in-hospital complications and is an independent predictor of mortality. The occurrence of AF in the course of ACS should be considered as a factor with prognostic implications.