Idiopathic thrombocytopenic purpura occurring in pregnancy has a risk for the fetus of severe thrombocytopenia. The handling of these cases obstetrically is controversial. Our experience and that derived from reading the literature suggests that counting the number of maternal platelets does not reflect on the risk for the fetus and the neonate. On the other hand it does seem that estimating the levels of circulating platelet antibodies in the maternal plasma does give an indication of platelet destruction in the fetus. Corticosteroid treatment improves greatly the number of maternal platelets and is probably good treatment for the fetus in utero. Giving gamma-globulins intravenously to the pregnant woman or to the newborn is an interesting therapeutic measure seeing how it is free of danger. How useful this treatment is in pregnancy has yet to be determined.