Abstract 11084: Prediction of Cause Specific Mortality in Ischemic Cardiomyopathy: A Competing Risk Model to Differentiate Sudden From Non-Sudden Cardiac Death
Background: Mortality in patients with ischemic cardiomyopathy is usually due to cardiac causes; however, the mechanism of death is almost equally split between sudden, primarily arrhythmic events and non-sudden cardiac death due to pump failure. The present study was designed to determine what specific risk factors can differentially predict cause specific cardiac mortality (sudden versus non-sudden).
Methods: A competing risk analysis was performed using the results of the PAREPET trial (Prediction of ARrhythmic Events with Positron Emission Tomography). This multivariate analysis with backward selection simultaneously evaluated risk factors for both sudden cardiac arrest (SCA) and non-sudden cardiac death. PAREPET is an NIH-sponsored, prospective observational trial of subjects with ischemic cardiomyopathy (n = 204) who were eligible for an implantable cardiac defibrillator (ICD) for the primary prevention of SCA. The primary end-point was adjudicated SCA (including ICD discharge for VF or VT >240 bpm).
Results: Mean age was 67 ± 11 years (±SD) with an EF of 27 ± 9%, and 90% were male. Over a median of 4.1 years, there were 33 SCA and 36 non-sudden cardiac deaths. The Table summarizes the results of the competing risk analysis. SCA was differentially predicted by the volume of denervated myocardium (defect of the PET sympathetic nerve tracer 11C-hydroxyephedrine), lack of angiotensin inhibition therapy, higher BNP and greater LVEDVI. In contrast, non-sudden cardiac death was preferentially predicted by higher resting HR, older age, greater LAVI and lower EF.
Conclusions: In patients with ischemic cardiomyopathy, cause specific mortality can be differentially predicted by various risk factors. Specifically, greater denervated myocardium and BNP preferentially predict SCA, but low EF and older age favors non-sudden cardiac death. Thus, a multi-variable prediction tool may be helpful for targeting ICD therapy to those at greatest risk of SCA.
- © 2013 by American Heart Association, Inc.