The effect of short-term prostacyclin (PGI2) administration on the incidence of restenosis after coronary angioplasty was studied in a prospective single-blind randomized trial of 286 patients. Of the 270 patients in whom dilation was successful, 134 received prostacyclin and 136 received placebo. Intracoronary prostacyclin was administered before and after dilation and then intravenously for 48 h. The control group received intracoronary placebo infusions before and after dilation. All patients received aspirin and dipyridamole before and after angioplasty, at least until follow-up angiography. Follow-up angiograms were obtained in 93% of patients in whom angioplasty was successful. Restenosis of one or more lesions was present in 34 patients (27%) who were given prostacyclin compared with 40 patients (32%) in the control group (p = NS). Acute vessel closure and ventricular tachyarrhythmias were more common in the control group than in the patients who received prostacyclin (acute vessel closure occurred in 14 [10.3%] of 136 versus 4 [3.0%] of 134, respectively, p less than 0.01; ventricular tachyarrhythmias occurred in 5 [3.4%] of 147 versus 0 of 139 respectively, p less than 0.05). Short-term administration of prostacyclin did not significantly lower the risk of restenosis after coronary angioplasty.