Maternal thrombocytopenia is not predictive of neonatal thrombocytopenia: a single-center Irish study.

BACKGROUND

Maternal thrombocytopenia during pregnancy is common. However, the relationship between maternal and neonatal thrombocytopenia is poorly understood.

OBJECTIVES

We aimed to determine whether an association exists between platelet counts of neonates born to mothers with moderate-to-severe thrombocytopenia (<100 × 10/L) and neonatal platelet counts.

METHODS

We identified records from 557 patients with moderate-to-severe thrombocytopenia (maternal platelet count <100 × 10/L) and the 338 associated newborn charts from 2018 to 2022 in a single large maternity center. Pregnant people with a platelet count of <100 × 10/L prior to delivery during present gestation were included. Any thrombocytopenia that occurred outside of pregnancy or in the postpartum period was excluded. A logistic regression was then generated to examine the association between maternal thrombocytopenia and neonatal thrombocytopenia. A receiver operating characteristic (ROC) curve was generated to assess accuracy of (i) lowest maternal platelet count and (ii) trimester of thrombocytopenia onset in predicting neonatal thrombocytopenia.

RESULTS

A total of 550 full blood count assessments were taken in neonates of pregnant people with thrombocytopenia. Sixteen neonates with clinically significant thrombocytopenia (platelet count <100 × 10/L) were identified. A binomial logistic regression was fitted that demonstrated limited association between lowest maternal platelet count and trimester of onset of maternal thrombocytopenia and the development of neonatal thrombocytopenia. An ROC curve was generated to determine the accuracy of maternal platelet count at identifying neonatal thrombocytopenia. The coordinates of the best platelet count threshold for this dataset were then derived from the ROC curve and determined that a threshold of 77.5 × 10/L maternal platelets offered the best accuracy.

CONCLUSION

Neonatal full blood count assessment based on maternal platelet counts of <100 × 10/L has a poor diagnostic yield with no statistically significant association in this cohort on logistic regression analysis. A lower threshold of 77.5 × 10/L may be of higher clinical utility and improve laboratory and clinical workflow.