With the aim to evaluate acute thrombogenicity, segments of human umbilical vein grafts (Dardik Biograft, Meadox Medicals, Inc., Oakland, N.I.), nonheparinized, heparin-alcohol-treated, or covalently heparinized, were implanted into the carotid arteries of sheep. Autologous carotid arteries were used as control grafts. Flow was restricted to 25 ml/ min. Accumulation of autologous a2P-labeled platelets was registered at both anastomotic and midgraft regions for 240 minutes. At the end of the perfusion period, grafts were removed, opened longitudinally, and the thrombus-free surface (TFS) and thrombus weight determined. The ex vivo determinations showed good correlation with results obtained with in vivo registration. In vivo accumulation of platelets was significantly larger in nonheparinized human umbilical vein grafts compared with heparin-alcohol treated grafts. Covalently heparinized grafts showed a tendency toward lower radioactive values, but visual examination after perfusion revealed areas of "intimal" damage with thrombotic deposits, probably caused by heparinization procedure. Autologous arteries showed the largest TFS and the smallest thrombi, Platelet accumulation in autologous arteries decreased almost to reference values after an initial rapid increase. In conclusion, heparin-alcohol treatment of human umbilical vein grafts reduces acute thrombogenicity. Although covalent heparin bonding seems to act in the same way, the fragility of the graft appears to preclude use of this method. Autologous arteries exhibit excellent antithrombotic characteristics. (J VASe SURG 1985; 2:434-42.)The autologous vein is considered the best available graft for use in small-diameter artery bypass surgery) When not available, human umbilical vein grafts and expanded polytetrafluoroethylene (PTFE) grafts appear to be the best alternatives. 2-4 In a recent randomized clinical trial, human umbilical vein grafts performed significantly better than PTFE grafts in fcmoropopliteal bypass surgery. SThe procedure for irrigating human umbilical vein grafts immediately prior to implantation includes flushing with heparin, and it has been demonstrated in an experimental canine model that hcparin treatment of the graft surface decreased the acute thrombogenicity of the graft, particularly if the luminal surface was incubated with ethyl alcohol after exposure to heparin. 6 It thus seems advisable to stabilize the hcparin layer. In a similar experimental