OBJECTIVE The aim of this study was to evaluate retrospectively our strategies in monitoring and treating pregnant women with idiopathic thrombocytopenic purpura (ITP). METHODS Medical records were reviewed for diagnosis, clinical course, treatment, and neonatal outcome in 35 Finnish women with ITP giving birth to 55 neonates during 53 pregnancies. The outcome of the first (i.e. index) pregnancy was used in the statistical analyses. The platelet immunofluorescence test (PIFT) was used for detection of platelet autoantibodies. The correlation between neonatal platelet counts and results of PIFT was calculated with the Pearson's correlation coefficient and the Fisher's exact test. RESULTS There were no serious bleeding complications although five of 35 women had platelet counts of less than 50 x 10(9)/l in the third trimester of the index pregnancy. Prophylactic platelet transfusions were given to six of 15 women delivered by cesarean section. Five of 35 (14.3%; 95% confidence interval, 2.6 to 25.8%) neonates had platelet counts of less than 50 x 10(9)/l median 3 days after delivery versus only one of 28 (3.6%; 95% confidence interval, 0.1 to 10.5%) at birth. No infant showed any clinical signs of intracranial hemorrhage. No significant correlation was encountered between neonatal thrombocytopenia and maternal platelet autoantibodies. The history of a previous infant with thrombocytopenia was the only important information in estimating the risk of fetal thrombocytopenia. CONCLUSIONS To avoid unnecessary and possibly harmful monitoring and treatment, we need further tests for predicting the perinatal risks in pregnant women with ITP.