Effects of Platelet Transfusion on Plasma Cytokine Levels and Neutrophil Extracellular Traps (NETs) in Neonates

Thrombocytopenia affects 22-35% of all neonates admitted to the neonatal intensive care unit (NICU), and prophylactic platelet transfusions are administered to thrombocytopenic premature neonates at higher platelet count thresholds than used in children/adults in an attempt to decrease their bleeding risk. However, the largest neonatal platelet transfusion threshold trial found an increased incidence of major bleeding and/or death in neonates randomized to a higher vs. a lower platelet count threshold. The mechanisms mediating these outcomes are unknown, but we hypothesized that they are related to the functional differences between adult (transfused) and neonatal platelets, particularly the higher pro-inflammatory potential of adult platelets (Thom, Davenport et al., JTH 2024). To test this, we measured a panel of 48 cytokines/growth factors by multiplex assay and neutrophil extracellular trap (NET) formation by MPO/DNA ELISA in plasma samples collected immediately before and either 2- or 4-hours after a platelet transfusion in 20 neonates admitted to the NICU at Boston Children's Hospital or Beth Israel Deaconess Medical Center. For 15 transfusions, the platelet unit supernatant was also stored for cytokines, extracellular vesicles (EVs), and free mitochondrial DNA (mtDNA) quantification. All studies were conducted with IRB approval and written informed consent was obtained from parents. The 20 infants studied (45% female) had an average gestational age of 36.9 weeks (range 23.1-40.7) and a birth weight of 2,589g (range 430-4,770). All transfused platelet units were apheresis derived, the median donor age was 65 years (range 22-75) and 55% of donors were male. Compared to pre-transfusion plasma levels, platelet transfusions were associated with significant increases in PDGF-a and b two hours post-transfusion (p=0.018 and p=0.046, respectively). Four hours post-transfusion, PDGF-a and b remained significantly elevated (p=0.004 and p=0.007, respectively) but RANTES (CCL5) exhibited the greatest increase (6.5-fold, p=0.034). These factors are all stored in platelets and released during storage, and indeed their concentrations were significantly higher in the supernatants of the transfused units than in the pre-transfusion neonatal plasma samples. Thus, to determine whether the observed post-transfusion increases could be explained by the transfused amounts of those factors, we compared the measured to the expectedpost-transfusion values. For most infants, the measured post-transfusion levels were similar to or below the expected values for PDGF-a and b, but were higher than expected for RANTES. These findings suggested that the post-transfusion increases in PDGF-a and b could be accounted for by the quantity of these factors transfused in the platelet unit, but that an additional source contributed to the RANTES elevation. Platelets also interact with neutrophils and promote the formation of NETs, extracellular web-like filamentous structures consistent of DNA and neutrophil proteins, which trap and kill pathogens but also increase platelet aggregation and microthrombosis, causing tissue damage. In the same pre-and post-transfusion samples, platelet transfusions were associated with a significant increase in NET formation (p=0.007), although the magnitude of this effect was variable. We next examined the correlation between various donor, recipient, and platelet unit characteristics and the changes in RANTES and NETs concentrations, and found that only platelet unit storage time significantly correlated with changes in RANTES (p=0.03). Storage time also weakly correlated with mtDNA levels (p=0.07) but not total EV or platelet-derived EV levels. In summary, our study identified novel inflammatory effects that may mediate the negative outcomes associated with platelet transfusions in neonates, and potential factors influencing these inflammatory responses. While platelet transfusions remain lifesaving treatments for specific populations of neonates, a better understanding of their pro-inflammatory effects is essential to improve outcomes.

Yost:University of Utah: Patents & Royalties: patent no. 10,232,023 B2. Sola-Visner:Johnson and Johnson: Consultancy; Sysmex America: Other: research equipment.