The efficacy of iloprost, a stable prostacyclin analog, was investigated in a placebo-controlled trial in 109 diabetics with ischemic lesions. 56 patients were randomly allocated to iloprost and 53 patients to placebo. Iloprost was intravenously applied for 6 hours daily on 28 consecutive days at an individually tolerated dose up to 2 ng/kg/min. The control group received identical solvent volumes. In addition all patients had an intensive basic, mainly local, therapy. At the end of the treatment in the iloprost group 31 of 50 patients (62%) showed partial (greater than 30%) or total healing of the lesion(s). In the placebo group this was the case in 12 of 51 patients (22.5%). The difference of 38.5% was statistically significant (p less than 0.05, chi 2-test, alpha = 0.05, beta = 0.1). The percentage of patients who were free of pain increased from 23% to 42% (+19%) in the iloprost group and from 38% to 48% (+10%) in the placebo group. After dose-titration iloprost was well tolerated. Flush, headache and abdominal complaints were the most frequent side effects. Heart rate and blood pressure were not influenced and the control of diabetes was not altered.