New Dynamic Prediction Model Could Help Guide Platelet Transfusions in Preterm Infants

September 30, 2025

Researchers at Leiden University Medical Center (LUMC) in the Netherlands have developed and validated a dynamic prediction model that may help clinicians individualize prophylactic platelet transfusion decisions for preterm infants with severe thrombocytopenia. The findings were published Sept. 15 in JAMA.

Researchers developed the model using data from 1,042 preterm infants across 14 neonatal intensive care units in Germany, the Netherlands and Sweden. The model estimates the risk of major bleeding or mortality under two hypothetical transfusion strategies: receiving a platelet transfusion within six hours or withholding transfusion for three days. It incorporates multiple clinical factors, including gestational and postnatal age, small-for-gestational-age status, necrotizing enterocolitis, sepsis, mechanical ventilation, vasoactive medication and prior platelet transfusions. The model updates continuously as the infant’s condition changes.

Validation in a separate cohort of more than 600 infants from Dutch NICUs showed that predicted risks closely matched observed outcomes. Data also indicated that the potential benefits and harms of prophylactic transfusions varied substantially depending on each infant’s clinical profile.

While the model has been successfully tested, Hilde van der Staaij, MD, the study’s lead author, cautioned that the model is not yet ready for clinical use.

“It’s a first step. Follow-up research is needed to determine at what risk level a transfusion should or should not be given,” she said. “It also needs to be proven that the model truly leads to better care and fewer bleedings and deaths.”

Van der Staaij and her co-authors also noted that while the model was developed for preterm infants, the underlying approach could be applied to other patient populations and medical decisions.