Effect of High-Dose Erythropoietin on Blood Transfusions in Extremely Low Gestational Age Neonates

Importance: Extremely preterm infants are among the populations receiving the highest levels of transfusions. Erythropoietin has not been recommended for premature infants because most studies have not demonstrated a decrease in donor exposure. Objectives: To determine whether high-dose erythropoietin given within 24 hours of birth through postmenstrual age of 32 completed weeks will decrease the need for blood transfusions. Design, Setting, and Participants: The Preterm Erythropoietin Neuroprotection Trial (PENUT) is a randomized, double-masked clinical trial with participants enrolled at 19 sites consisting of 30 neonatal intensive care units across the United States. Participants were born at a gestational age of 24 weeks (0-6 days) to 27 weeks (6-7 days). Exclusion criteria included conditions known to affect neurodevelopmental outcomes. Of 3266 patients screened, 2325 were excluded, and 941 were enrolled and randomized to erythropoietin (n = 477) or placebo (n = 464). Data were collected from December 12, 2013, to February 25, 2019, and analyzed from March 1 to June 15, 2019. Interventions: In this post hoc analysis, erythropoietin, 1000 U/kg, or placebo was given every 48 hours for 6 doses, followed by 400 U/kg or sham injections 3 times a week through postmenstrual age of 32 weeks. Main Outcomes and Measures: Need for transfusion, transfusion numbers and volume, number of donor exposures, and lowest daily hematocrit level are presented herein. Results: A total of 936 patients (488 male [52.1%]) were included in the analysis, with a mean (SD) gestational age of 25.6 (1.2) weeks and mean (SD) birth weight of 799 (189) g. Erythropoietin treatment (vs placebo) decreased the number of transfusions (unadjusted mean [SD], 3.5 [4.0] vs 5.2 [4.4]), with a relative rate (RR) of 0.66 (95% CI, 0.59-0.75); the cumulative transfused volume (mean [SD], 47.6 [60.4] vs 76.3 [68.2] mL), with a mean difference of -25.7 (95% CI, 18.1-33.3) mL; and donor exposure (mean [SD], 1.6 [1.7] vs 2.4 [2.0]), with an RR of 0.67 (95% CI, 0.58-0.77). Despite fewer transfusions, erythropoietin-treated infants tended to have higher hematocrit levels than placebo-treated infants, most noticeable at gestational week 33 in infants with a gestational age of 27 weeks (mean [SD] hematocrit level in erythropoietin-treated vs placebo-treated cohorts, 36.9% [5.5%] vs 30.4% [4.6%] (P < .001). Of 936 infants, 160 (17.1%) remained transfusion free at the end of 12 postnatal weeks, including 43 in the placebo group and 117 in the erythropoietin group (P < .001). Conclusions and Relevance: These findings suggest that high-dose erythropoietin as used in the PENUT protocol was effective in reducing transfusion needs in this population of extremely preterm infants. Trial Registration: ClinicalTrials.gov Identifier: NCT01378273.