We performed a retrospective study of 26 pregnancies with chronic immune thrombocytopenic purpura (ITP) or incidental ITP. Thirteen pregnancies were followed without the use of percutaneous umbilical blood sampling (PUBS) and 13 were followed sampling PUBS at 36–40 weeks of gestation. The overall prevalence of neonatal thrombocytopenia was 27%: 50% in chronic ITP, 12.5% in incidental ITP. Symptomatic thrombocytopenia occurred only in infants born to mothers with chronic ITP. Without the use of PUBS, 2 symptomatic thrombocytopenic fetuses were vaginally delivered and there was a high rate of cesarean sections in normal fetuses. Owing to PUBS, a decrease of the cesarean section rate in normal fetuses was observed, but one fetal bradycardia (due to the technique) was encountered. PUBS resulted helpful to indicate the best route of delivery and reducing unnecessary cesarean sections in women with chronic ITP, but in presence of incidental ITP the use of PUBS remains an open questio