In August 1985, the Percutaneous Transluminal Coronary Angioplasty Registry of the National Heart, Lung, and Blood Institute reopened at its previous sites to document changes in angioplasty strategy and outcome. The new registry entered 1802 consecutive patients who had not had a myocardial infarction in the 10 days before angioplasty. Patient selection, technical outcome, and short-term major complications were compared with those of the 1977 to 1981 registry cohort. The new-registry patients were older and had a significantly higher proportion of multivessel disease (53 vs. 25 percent, P less than 0.001), poor left ventricular function (19 vs. 8 percent, P less than 0.001), previous myocardial infarction (37 vs. 21 percent, P less than 0.001), and previous coronary bypass surgery (13 vs. 9 percent, P less than 0.01). The new-registry cohort also had more complex coronary lesions, and angioplasty attempts in these patients involved more multivessel procedures. Despite these differences, the in-hospital outcome in the new cohort was better. Angiographic success rates according to lesion increased from 67 to 88 percent (P less than 0.001), and overall success rates (measured as a reduction of at least 20 percent in all lesions attempted, without death, myocardial infarction, or coronary bypass surgery) increased from 61 to 78 percent (P less than 0.001). In-hospital mortality for the new cohort was 1 percent, and the nonfatal myocardial infarction rate was 4.3 percent. Both rates are similar to those for the old registry. The long-term efficacy of current angioplasty remains to be determined.