To identify pretransplant factors that are influencing survival after orthotopic liver transplantation a Cox proportional hazards regression model was applied to 118 children with chronic terminal liver failure transplanted at Medical School Hannover during the period of 1978 to 1994. The response variable was survival, as covariates a total of 19 pretransplant variables were entered--i.e. age, diagnosis (biliary cirrhosis, metabolic cirrhosis, postnecrotic cirrhosis, cryptogenetic cirrhosis) sex, laparotomy prior to OLT, height, weight, standard deviation scores for height and weight, date of first OLT, serum alanine aminotransferase, asparagine aminotransferase, albumin, total bilirubin, cholinesterase activity, glomerular filtration rate, and prothrombin time. Significant independent predictors of survival after OLT were bilirubin (P=0.0024), SDS for weight (P=0.034), and albumin (P=0.039). In a subsequent discriminant analysis cut off points for these variables could be identified--i.e., bilirubin >340 micromol/L, SDS for weight <-2.2 and albumin < 33 g/L. Patients with one or more of these risk factors were grouped as urgent indication group (n=76) and those with no risk factor as elective indication group (n=42). Comparing the posttransplantation survival in these groups there is a statistically significant difference at 1 year (57% vs. 90.5%) and 4 years (49% vs. 90.5%) after OLT (P=0.0001, log rank test). It is concluded that the risk of OLT is much higher if liver function is very poor. Optimal nutritional support prior to transplantation is mandatory to optimise the clinical status of the children and to improve the results of OLT.