Recently, the Multicenter Automatic Defibrillator Implantation Trial II (MADIT II) demonstrated that post-infarction patients with advanced left ventricular dysfunction, defined by ejection fraction equal to or < 30%, have a very high 19.6% 20-month mortality which was reduced by 28% with prophylactic implantation of a cardioverter-defibrillators. Over 30% of patients with SCDs receive appropriate therapy for ventricular tachycardia or ventricular fibrillation. Similar reduction in mortality associated with ICD therapy was recently reported from the Sudden Cardiac Death in Heart Failure Trial (SCD-HeFT). Since the population of patients eligible for an ICD following MADIT II indications is very large, there is a great interest in developing methods and algorithms for identifying high- and low-risk individuals among MADIT ll-type patients to prioritize them for ICD therapy. Based on previous experience and recent preliminary analyses from the MADIT II data, we hypothesize that a combination of clinical variables and noninvasive ECG parameters, indicating contribution of different mechanisms that predispose to arrhythmias and sudden death, will allow identification of patients with increased benefit and those with little benefit from ICD therapy. Therefore, the primary aims of this study are: 1) to evaluate the predictive value of a multivariate model consisting of pre-specified clinical and ECG parameters for predicting arrhythmic events in MADIT II type postinfarction patients with severe left ventricular dysfunction; 2) to develop a multivariate risk-stratification model, based on a broader spectrum of pre-specified clinical covariates and ECG parameters, and from it a risk-scoring algorithm identifying high-risk and low-risk patient groups; this algorithm will be validated by a cross-validation study. Such an algorithm will enable an ordering of patients who may benefit most, and benefit least, from ICD therapy. The secondary objectives of this study are: 1) to determine the prognostic significance of clinical and noninvasive ECG variables for predicting non-sudden (non-arrhythmic) cardiac mortality in MADIT II type patients; identifying such individuals will further refine clinical practice and cost-effectiveness of primary prevention of SCO with ICD therapy; 2) to identify ECG predictors of inappropriate therapy delivered for episodes of atrial fibrillation or supraventricular tachyarrhythmias and evaluate the association between inappropriate therapy and the risk of ventricular tachyarrhythmias; 3) to determine whether clinical and ECG predictors contribute to identification of patients with worsening quality of life, and compare changes in quality of life of patients experiencing ICD therapy with those who do not, after adjusting for clinical and ECG parameters.
