The influence of coronary artery disease and bypass grafting on survival after valve replacement for aortic stenosis (1975 to 1986, N = 512) was analyzed. Mean follow-up for 30-day survivors was 5.1 years (0.1 to 12.9 years). A total of 205 patients had coronary angiography performed: 122 did not have coronary artery disease, 55 with coronary artery disease underwent bypass grafting, and 28 with coronary artery disease did not. Early mortality rates (less than or equal to 30 days)/5-year cumulative survivals (standard error) were 4.1%/86% (4%), 3.6%/68% (8%), and 17.9%/51% (13%), respectively (p less than 0.05/p less than 0.01). Triple vessel/left main stem disease was more prevalent in patients with coronary disease who underwent bypass grafting (47%) than in those who did not (14%; p less than 0.05). Multivariate analysis revealed that right ventricular failure and omission of bypass grafting in patients with coronary artery disease were independent determinants of early mortality. A Cox regression analysis identified coronary artery disease and aortic valve gradient as determinants of mortality after hospital dismissal, which was not influenced by bypass grafting. On the basis of a coronary artery disease score (positive predictive value for coronary artery disease of 66%) developed on the patients with angiography, 307 patients without angiography were divided into 234 with a low score and 73 with a high score. Early mortality rates/5-year survivals (standard error) were 6.4%/86% (2%) and 16.4%/67% (6%), respectively (p less than 0.01/p less than 0.001). Autopsy revealed stenotic or occlusive coronary artery disease in 92% of 12 early deaths in the group with a high coronary artery disease score and in 33% of 15 in the group with a low score (p less than 0.01). Left ventricular failure and a high coronary artery disease score were independent determinants of early mortality, whereas cardiothoracic index, a high coronary artery disease score, and left ventricular failure were independent predictors of death after hospital dismissal. Despite more severe coronary artery disease, bypass grafting reduced early mortality to a level comparable with that of patients without coronary artery disease, contrasting with a high early mortality rate for unbypassed coronary artery disease. Coronary artery disease increased the late mortality rate, which was not modified by bypass grafting. In the group without angiography, undiagnosed and unbypassed coronary artery disease probably increased both early and late mortality. Coronary angiography should be performed in all adult patients with aortic stenosis, and those with significant coronary artery disease should undergo bypass grafting concomitant with valve replacement.