A new martensitic nitinol stent with improved flexibility and radiopacity was tested to evaluate whether these differences improve initial or long-term outcome. Patients who underwent percutaneous revascularization of a discrete native coronary lesion were randomly assigned to the new stent (PARAGON, n = 349) or to the first-generation Palmaz-Schatz (PS) stent (n = 339). The primary end point was target vessel failure at 6 months (a composite of cardiac or noncardiac death, any infarction in the distribution of the treated vessel, or clinically indicated target vessel revascularization). Secondary end points were, among others, device and procedural success and angiographic restenosis. Mean age was 62 years; diabetes was present in 21% of patients, prior bypass surgery in 6%, and recent infarction in 22% (p = NS for comparison between the 2 randomized arms). The PARAGON stent group had smaller reference vessels (2.97 vs 3.05 mm, p = 0.05), more prior restenosis (8.0% vs 4.5%, p = 0.07), and a longer average stent length (21.3 vs 19.4 mm, p < 0.05). Device success was significantly higher in the PARAGON arm (99.1% vs 94.3%, p < 0.05). Death and infarction at 6-month follow-up were infrequent in both groups. There was no significant difference in death (2.0% vs 1.2%, p = 0.546), but a higher rate of infarction for the PARAGON cohort (9.2% vs 4.7%, p = 0.025). Although target vessel failure (20.3% vs 12.4%, p = 0.005) and target lesion revascularization (12.0% vs 5.9%, p = 0.005) were higher in the PARAGON group, there was no significant difference in 6-month follow-up in in-stent minimal lumen diameter or in the rate of binary angiographic restenosis. Both PARAGON and PS stents are safe and associated with infrequent adverse events. The PARAGON stent can be delivered more frequently than the first-generation PS stent. Although there was no significant difference in in-stent minimal lumen diameter or the frequency of angiographic restenosis, clinical restenosis was more frequent in the PARAGON group.