Extrapolation of improvements in percutaneous coronary intervention (PCI) and outcomes to patients with diabetes has not been systematically examined in clinical practice. Two waves of consecutive patients (n = 4,629) who underwent PCI from July 1997 to June 1999 enrolled in the National Heart, Lung, and Blood Institute Dynamic Registry comprise the study population. There were 1,058 patients with treated diabetes and 3,571 patients without clinically evident diabetes. As a group, patients with diabetes tended to have more clinical, angiographic, and procedural risk factors. Although crude in-hospital mortality rates were higher in patients with diabetes (diabetics 2.3%, nondiabetics 1.3%; p = 0.02), the adjusted risk of in-hospital death (odds ratio 1.46, 95% confidence interval [CI] 0.80 to 2.66) was not significantly different. At 1 year, patients with diabetes had a significantly higher adjusted risk of mortality (risk ratio [RR] 1.80, 95% CI 1.35 to 2.41) and need for repeat revascularization (RR 1.40, 95% CI 1.13 to 1.74). There was a significant interaction between stent use and diabetic status with the need for repeat revascularization (adjusted RR in nondiabetics 0.73, 95% CI 0.61 to 0.88; adjusted RR in patients with diabetes 1.20, 95% CI 0.88 to 1.65). Beta blockers at the time of hospital discharge were significantly associated with reduced mortality rates at 1 year in both groups.