To test the hypothesis that coronary artery bypass grafting (CABG) is not routinely required in patients undergoing aortic valve replacement (AVR) who have coexistent coronary artery disease (CAD), we compared the results of operation in 55 consecutive symptomatic patients who had CAD and underwent AVR without CABG with results in another 142 patients without CAD who underwent AVR during the same period, and with published results from other centers in which CABG was used in patients with CAD who underwent AVR. Operative mortality was 4% in patients with CAD and 5% in patients without CAD. Late survival was not significantly different between the two groups when analyzed for the entire population (80% survival at 3 years in CAD patients, 82% for non-CAD patients), or for the subgroup of patients with aortic stenosis, aortic regurgitation or aortic stenosis plus regurgitation. Eight patients with CAD (15%) developed recurrent angina after AVR (mean follow-up 43 months); only three patients (6%) required CABG because of medically refractory angina (12-43 months). Operative mortality, operative infarction (9%), recurrent angina and long-term survival in patients with CAD after AVR were similar to those at other centers after AVR plus CABG. These data suggest that preoperative detection of CAD does not necessitate CABG in all patients at the time of AVR.