Between November 1982 and March 2006, 67 children with body weight < or =10 kg had a primary liver transplantation from deceased donors in our unit. The aim of this study was to analyze the outcome in terms of patient and graft survival and to search for factors affecting this outcome. Overall, one-, three-, five-, and 10-yr primary patient and graft survival rates were 73%, 71%, 66%, 63% and 59%, 56%, 53%, 48%, respectively. Twenty-four of 67 (36%) children died and in the remaining 22 (33%), the first grafts failed and they were retransplanted. Cox regression analysis revealed that a need for retransplantation and urgent transplantation were important predictors for patient survival (p = 0.04 and p = 0.001, respectively). To assess whether the need for retransplantation can be influenced, all study variables were compared between surviving grafts and failed grafts. Cox regression analysis showed that only donor/recipient (D/R) weight ratio proved to be independent predictor for graft survival (p = 0.004). After comparison of graft survival with the long rank test according to different D/R weight ratios (3.0-7.0), the cut-off point for significantly different graft survival approached 4.0. The one-, three-, five-, and 10-yr graft survival for technical variant grafts with a D/R weight ratio <4.0 was 85%, 68%, 68%, and 68% compared with a D/R weight ratio >4.0 was 44%, 38%, 38%, and 30%, respectively (p = 0.02). In summary, patient survival in children with body weight < or =10 kg is determined by urgent transplantation and the need for retransplantation. Graft loss and retransplantation in small children can be prevented by adequate size matching of donor and recipient whereby a D/R weight ratio <4.0 seems to offer the favorable outcome.