The objective of this study was to evaluate the relation between runoff vessels and the prognosis in patients who have had an occlusion of a previously placed peripheral arterial bypass graft. Over a 4-year period 77 occluded synthetic grafts were treated with urokinase and reconstruction after angiographic study. Follow-up ranged from 1 to 1627 days. The cases were divided into three groups according to the number of patent tibial vessels. Group I consisted of patients who had no vessel runoff. Group II consisted of patients who had single vessel runoff. Group III consisted of patients who had two or three runoff vessels. In Group I, the 1, 2, and 3-year patency rates were 35.8%, 8.9%, and 8.9%; and the limb salvage rates were 50.2%, 40.2%, and 40.2% for 1, 2, and 3 years, respectively. The Group II patency rates were 31.2%, 26.0%, and 13.0% and the limb salvage rates were 72.3%, 62.9%, and 62.9%. The Group III patency rates were 50.0%, 26.1%, and 20.9%; and the limb salvage rates were 93.1%, 79.1%, and 79.1%. There is no statistically significant difference in patency rates among any of the groups. The limb salvage rate was significantly increased in Group III, compared to I (P < 0.01) and Group II (P < 0.05), and in Group II, compared to Group I (P < 0.05). These results indicate that the higher rate of limb salvage in this study statistically correlated with the number of runoff vessels.