This study evaluated the effect of heart failure (HF) and ejection fraction (EF) at baseline on long-term cardiac mortality in patients undergoing coronary revascularization and investigated the effect of diabetes mellitus (DM) on mortality. We evaluated long-term outcomes of patients without HF, HF and a preserved EF, and HF and a decreased EF who underwent revascularization with percutaneous transluminal coronary angioplasty or coronary artery bypass graft surgery after enrollment in the Bypass Angioplasty Revascularization Investigation (BARI) trial. Ten years after initial revascularization, cumulative rates of freedom from cardiac death were 90% in patients without HF, 75% in patients with HF and a preserved EF, and 59% in patients with HF and a decreased EF (p <0.001, 3-way comparison). In diabetic patients with HF and a preserved EF, there was a significant increase in cardiac mortality compared with patients without HF (p <0.001); however, this relation was not seen in patients without DM. In conclusion, patients with HF and a preserved EF have increased mortality over 10 years compared with those without HF. Only in patients with DM did HF with preserved EF confer additional risk.