Design and methods: We hypothesized that treatment with recombinant human erythropoietin (r-HuEPO) would stimulate erythropoiesis and would thereby reduce the need for erythrocyte transfusions in preterm infants. We treated 157 preterm infants born at 26.9 +/- 1.6 weeks of gestation who weighed 924 +/- 183 g at birth with either subcutaneous r-HuEPO (100 U/kg/d, 5 days per week) or placebo for 6 weeks in a randomized, double-blind, controlled clinical trial. All patients received oral iron and were managed according to uniform conservative transfusion guidelines.
Results: Treatment with r-HuEPO was associated with fewer erythrocyte transfusions (1.1 +/- 1.5 per infant in the r-HuEPO group versus 1.6 +/- 1.7 per infant in the placebo group; P = .046) and with a reduction in the volume of packed erythrocytes transfused (16.5 +/- 23.0 mL versus 23.9 +/- 25.7 mL per infant; P = .023). Overall, 43% of the infants in the r-HuEPO group and 31% of placebo-treated infants were transfusion-free during the study (P = .18). The volume of blood removed for laboratory tests and the need for respiratory support at the start of treatment had major effects on transfusion requirements independent of r-HuEPO. Reticulocyte counts were higher during treatment in the r-HuEPO group (P = .0001), and r-HuEPO-treated infants had higher hematocrit values at the end of the study (32% versus 27.3% in the placebo group; P = .0001). We found no differences in the incidence of major complications of prematurity between the treatment groups.
Conclusion: We conclude that treatment with r-HuEPO at a weekly dose of 500 U/kg stimulates erythropoiesis, moderates the course of anemia, is associated with a reduction in erythrocyte transfusions, and appears safe in very low birth weight preterm infants who are receiving iron supplements. Conservative transfusion criteria, minimization of phlebotomy losses, and treatment with r-HuEPO are complementary strategies to reduce erythrocyte transfusions in these infants.