We sought to determine if advances in percutaneous coronary intervention (PCI) are associated with better outcomes among patients with diabetes mellitus (DM). Patients with DM enrolled in the National Heart, Lung, and Blood Institute (NHLBI) early PTCA Registry (1985-1986) were compared to those in the subsequent contemporary Dynamic Registry (1999-2002) for in-hospital and one-year cardiovascular outcomes. The study population included 945 adults with DM, 325 from the PTCA Registry and 620 from the Dynamic Registry. Multivariable Cox regression models were built to estimate the risk of clinical events. Dynamic Registry patients were older, had more noncardiac comorbidities, and a lower mean ejection fraction (50.5% vs 57.8%, P < or = 0.001) compared to the PTCA Registry patients. The incidence of in-hospital mortality (1.9% vs 4.3%, P < or = 0.05), myocardial infarction (MI) (1.0% vs 7.4%, P <or = 0.001), and coronary artery bypass grafting (CABG) (0.8% vs 6.2%, P < or = 0.001) were all significantly lower and independent of the use of stents. One-year adverse events including MI (4.9% vs 11.0%, P < or = 0.001), CABG (6.4% vs 15.0%, P < or = 0.001), and need for repeat revascularization (18.7% vs 33.3%, P < or = 0.001) were all lower in the Dynamic Registry. The relative risk of death at 1 year was significantly less for patients in the Dynamic Registry (RR 0.56, 0.34; 0.92, P = 0.02). Although Dynamic Registry patients with diabetes had more advanced coronary disease, in-hospital and late adverse events were lower. A combination of the use of stents and an increase in adjunctive medical therapy are likely responsible for the observed improvements in outcomes in contemporary PCI.