Graft thrombosis is an important cause of early (<4 weeks) renal graft loss. Reports show that heparin reduces the incidence of early renal allograft thrombosis. Routine peri-operative administration of unfractionated heparin was introduced in our unit in 1994. We conducted a retrospective study of 254 transplants, undertaken in children, between 1987 and 2000. There were 126 children who did not receive heparin (group 1) and 128 who did (group 2). Recipient characteristics and immunosuppression were similar in both groups. The incidence of graft loss secondary to thrombosis was compared between the groups. Variables previously identified with increased risk of graft loss, including donor age, recipient age, cold ischaemia time (CIT), multiple donor vessels, surgical complications, and side of graft donation, were examined using logistic regression. Thrombosis occurred in 14 grafts in group 1 and 11 grafts in group 2 (odds ratio 0.7, 95% confidence interval 0.3-1.6, P=not significant). The mean time to graft loss was similar in groups 1 and 2 (6.6, SD 3.9, range 2-12 days and 7.9, SD 4.4, range 1-14 days, respectively) ( P=0.445). Young recipient age ( P=0.006), young donor age ( P=0.009), increasing CIT ( P=0.007), and surgical complications ( P=0.002) increased the risk of graft thrombosis. A reduction in the incidence of early renal allograft thrombosis upon introduction of heparin was not demonstrated.