BACKGROUND: Although the recent surgical treatment of ischemic heart failure substudy reported that revascularization of viable myocardium did not improve survival, these results were limited by the viability imaging technique used and the lack of inducible ischemia information. We examined the relative impact of stress-rest rubidium-82/F-18 fluorodeoxyglucose positron emission tomography identified ischemia, scar, and hibernating myocardium on the survival benefit associated with revascularization in patients with systolic dysfunction. METHODS AND RESULTS: The extent of perfusion defects and metabolism-perfusion mismatch was measured with an automated quantitative method in 648 consecutive patients (age, 65±12 years; 23% women; mean left ventricular ejection fraction, 31±12%) undergoing positron emission tomography. Follow-up time began at 92 days (to avoid waiting-time bias); deaths before 92 days were excluded from the analysis. During a mean follow-up of 2.8±1.2 years, 165 deaths (27.5%) occurred. Cox proportional hazards modeling was used to adjust for potential confounders, including a propensity score to adjust for nonrandomized treatment allocation. Early revascularization was performed within 92 days of positron emission tomography in 199 patients (33%). Hibernating myocardium, ischemic myocardium, and scarred myocardium were associated with all-cause death (P=0.0015, 0.0038, and 0.0010, respectively). An interaction between treatment and hibernating myocardium was present such that early revascularization in the setting of significant hibernating myocardium was associated with improved survival compared with medical therapy, especially when the extent of viability exceeded 10% of the myocardium. CONCLUSIONS: Among patients with ischemic cardiomyopathy, hibernating, but not ischemic, myocardium identifies which patients may accrue a survival benefit with revascularization versus medical therapy.