Background: Prior studies show that ischemic cardiomyopathy (ICM) patients with substantial viable myocardium have better survival with coronary revascularization (CR) than medical therapy (MT). When myocardial perfusion imaging (MPI) is used, the analysis is often based on visual scoring. We sought to determine the value of automated quantitative viability analysis in guiding management and predicting outcome. Methods: We identified 246 consecutive ICM patients who had rest-redistribution gated SPECT thallium-201 MPI. Size and severity of perfusion defects were assessed by automated method. Regions with <50% activity vs normal were considered nonviable. Mortality was verified against the social security death index database. Results: Of the 246 patients, 37% underwent CR within 3 months of MPI. The initial images showed a total perfusion defect size of 32 ± 17%, redistribution of 3.5 ± 4.6% and nonviable myocardium of 13 ± 14%LV. Using multivariate logistic regression analysis, independent predictors of CR included chest pains (OR 2.74) and rest-delayed transient ischemic dilatation (OR 4.49), while a prior history of CR or ventricular arrhythmias favored MT. The cohort was followed-up for 41 ± 30 m during which 111 patients (45%) died. Survival was better with CR than MT (P < .0001). For CR, survival was better for those with a smaller area of nonviable myocardium (risk of death increased by 5%/1% increase in size of nonviable myocardium, P = .009) but this was not seen in MT. CR had a mortality advantage over MT when the area of nonviable myocardium was ≥20%LV but not larger. Conclusions: Automated quantitative analysis of MPI is useful in predicting survival in ICM, but the decision for or against CR is a complex one as it depends on multiple other factors and "viability testing" is just one variable that needs to be incorporated in the decision-making process. © 2010 American Society of Nuclear Cardiology.