BACKGROUND Associations between coronary artery disease (CAD) and outcomes in systolic heart failure (HF) and that between coronary artery bypass graft (CABG) surgery and outcomes in patients with HF and CAD have not been examined using propensity-matched designs. METHODS Of the 2707 patients with advanced chronic systolic HF in the Beta-Blocker Evaluation of Survival Trial (BEST), 1593 had a history of CAD, of whom 782 had prior CABG. Using propensity scores for CAD we assembled a cohort of 458 pairs of CAD and no-CAD patients. Propensity scores for prior CABG in those with CAD were used to assemble 500 pairs of patients with and without CABG. Matched patients were balanced on 68 baseline characteristics. RESULTS All-cause mortality occurred in 33% and 24% of matched patients with and without CAD respectively, during 26 months of median follow-up (hazard ratio {HR} when CAD was compared with no-CAD, 1.41; 95% confidence interval {CI}, 1.11-1.81; P=0.006). HR's (95% CIs) for CAD-associated cardiovascular mortality, HF mortality, and sudden cardiac death (SCD) were 1.53 (1.17-2.00; P=0.002), 1.44 (0.92-2.25; P=0.114) and 1.76 (1.21-2.57; P=0.003) respectively. CAD had no association with hospitalization. Among matched patients with HF and CAD, all-cause mortality occurred in 32% and 39% of those with and without prior CABG respectively (HR for CABG, 0.77; 95% CI, 0.62-0.95; P=0.015). CONCLUSIONS In patients with advanced chronic systolic HF, CAD is associated with increased mortality, and in those with CAD, prior CABG seems to be associated with reduced all-cause mortality but not SCD.