Smooth muscle cell proliferation induced by intimal denudation similar to that produced by coronary angioplasty has been decreased by early and brief heparin administration in animal models. Heparin might also decrease the incidence of thrombus-induced postdenudation arterial obstruction, albiet at risk of bleeding. To evaluate the risks and benefits of heparin after coronary angioplasty, 416 patients (469 vessel sites) with successful coronary angioplasty without large dissection were randomized to heparin (titrated to partial thromboplastin time 1.5 to 2.5 times normal) or dextrose administered for 18 to 24 hours. Patients also had received heparin during angioplasty (usually 10,000 to 15,000 units), and they received aspirin for a period of 6 months. Heparin and dextrose groups were not different in the percentage of patients with male sex (74% in the heparin group and 75% in the dextrose group, p = NS), postangioplasty tear or dissection (30% in the heparin group and 30% in the dextrose group, p = NS), or postangioplasty percent stenosis greater than 35% (31% in the heparin group and 30% in the dextrose group, p = NS). Partial thromboplastin time during treatment in the heparin group was 56 +/- 22 seconds and in the dextrose group 27 +/- 9 seconds (p less than 0.001). Late angiographic follow-up (180 +/- 81 days) was achieved in 58.4% in the heparin group and 64.5% in the control group. Of patients with late angiographic follow-up, 41.2% and 36.7% randomized to heparin and dextrose, respectively, had documented restenosis (greater than or equal to 50% diameter stenosis at one or more the sites dilated) (p = NS).(ABSTRACT TRUNCATED AT 250 WORDS)