OGTT was performed in 28 liver transplants maintained with tacrolimus to investigate carbohydrate metabolism and assess risk factors for development of PTDM. None had PTDM that was detected by OGTT. Early PTDM in four cases (14.3%) resolved in follow-up. Five new cases (17.9%) demonstrated DCM (DCM = IGT +/- hyperinsulinemia). Fasting measurements were normal in two hyperinsulinemic cases. With one (20%, p > 0.05) exception none of the children with DCM were overweight or had a family history of diabetes. All five (100%) children with DCM had been given high cumulative dosage of steroids 18 (78.3%) - without DCM (p > 0.05). The median age of children with DCM was greater [4.3 (12.7-18.0) vs. 7.0 (2.3-18.0) yr, p < 0.01] and duration of follow-up longer [5.3 (2.3-7.0) vs. 2.5 (0.7-7.3) yr, p < 0.05]. Four children (80%) with DCM were pubertal (p < 0.05). However, neither age nor duration of follow-up or pubertal stage had significant effect on DCM development. Early PTDM is a transient phenomenon and is not predictive for future development of diabetes. DCM is frequently observed in liver transplanted children. Albeit the children with DCM were given high cumulative dose of steroids, were older, mostly were pubertal, and had longer duration of follow-up, we cannot draw firm conclusions on effects of the risk factors on carbohydrate metabolism because of the small sample size and relatively short duration of follow-up. Unlike fasting measurements, OGTT can detect all children with DCM.