In children, one-third of immune thrombocytopenic purpura (ITP) patients follow a chronic course. The present study investigated whether treatment with intravenous immunoglobulin (IVIG) at the time of diagnosis of ITP is of prognostic significance, using data from 1984 children entered in Registry I of the Intercontinental Cooperative ITP Study Group. A matched pairs analysis compared children with thrombocytopenia (platelet count <150 x 10(9)/l) 6 months following diagnosis with children whose platelet count was normal 6 months after diagnosis. It was found that children initially treated with IVIG were more likely to have a normal platelet count 6 months after diagnosis than children not receiving IVIG (odds ratio 1.81; 95% confidence interval: 1.25-2.64). This result was independent of age, gender, country of origin, platelet count at diagnosis or infection preceding the diagnosis of ITP. In a similar analysis, comparing children with a platelet count <50 x 10(9)/l 6 months after diagnosis with children whose platelet count was > or =50 x 10(9)/l at that time point, the former group was less often treated with IVIG than with steroids (P = 0.02). Prospective studies are required to further explore this potential effect of IVIG.