After sporadic reports of renal graft artery thromboses, prophylaxis against thrombosis (PAT) was given to all of our r-HuEPO-treated patients (n = 35) during a period of 2 years. No thromboembolic events (TEE) occurred in the r-HuEPO-treated group receiving PAT. However, the PAT-protocol (500 ml dextran on days 0, 1, 3 and 5, followed by low doses of aspirin, 160-250 mg daily) resulted in a 54.3% incidence of bleeding complications, of which 22.9% were major (i.e., necessitating multiple transfusions or invasive procedures). A group of renal graft recipients (n = 83), who were not treated with r-HuEPO and were not given PAT, showed a 10.8% incidence of bleeding complications of which 2.4% major. Two cases of TEE were noted in the untreated group. The difference in bleeding complications between the two groups was statistically significant (0.025 > P > 0.01 ). The