Bleeding after open heart surgery is still a great concern for the surgeon, especially when the surgical field has been revised accurately and hemostatic stitches and electrical cauterization have been used extensively. Among non-surgical adjuncts, aprotinin has been reported as very effective in reducing complications. At the time we started using this drug, we intended to test two different dosages lower than those reported in the literature. We evaluated three groups of 18 patients: the first (A) received about 350 mg of aprotinin from the start of anesthesia up to the end of operation (140 mg in the priming of cardio-pulmonary bypass and 70 mg/h i.v. during the procedure; the second (A/2) received half that dose (i.e. 70 mg and 35 mg, respectively), and the third (C) did not receive aprotinin. We compared in these groups: postoperative bleeding, blood transfusions, red blood cells, hemoglobin, hematocrit, platelets. The results were good only in the A group: bleeding was reduced and few transfusions were required. The patients in the A/2 and C groups did not show significant differences. From our observations we conclude that aprotinin is a useful adjunct, but has to be given in the proper dose.