Ciprostene, a chemically stable prostacyclin analog was studied for its effects on restenosis in patients with coronary artery disease undergoing therapeutic percutaneous transluminal coronary angioplasty (PTCA). In a double-blind, randomized trial 32 patients were randomized to receive either ciprostene or the respective placebo. The infusion started intracoronarily at a rate of 40 ng/kg/min 20 min before introduction of the balloon catheter into the coronary artery. Thereafter infusion was continued intravenously for 36 hours at a rate of 120 ng/kg/min and a tapering off period until 48 hours. The quantitative analyses of the degree of coronary artery stenoses on the angiographic films before PTCA, after PTCA and after 6 month of follow-up was performed in 24 patients available. In patients receiving placebo (n = 12) coronary artery stenoses was 81 +/- 3% before PTCA and was reduced to 34 +/- 3% by angioplasty. At the 6 month follow up angiography stenoses diameter was measured as 63 +/- 8%, being not significantly different from the % stenoses before PTCA. In contrast, coronary artery stenoses in patients receiving ciprostene (n = 12) measured 83 +/- 3% before PTCA, 31 +/- 4% after PTCA and 55 +/- 9% at 6 month, being still significantly different from pre-PTCA value (P less than 0.05). When patients were characterized according to their clinical status, these differences were accounted for by patients with unstable angina receiving ciprostene. Ciprostene seems to reduce restenosis 6 month after coronary angioplasty in patients with unstable angina. The infusion rate of 40 ng/kg/min i.c. followed by 120 ng/kg/min i.v. was tolerated well, although the incidence of catheter associated bleeding was increased.