Treatment with recombinant human erythropoietin (rHuEPO) stimulates erythropoiesis and reduces the need for transfusions in hospitalized preterm infants. The aim of our study was to follow very low birth weight infants after the initial 6 weeks of rHuEPO treatment.

We randomly assigned 97 very low birth weight infants with a gestational age of 31 weeks or less and birth weight of 1500 gm or less to receive rHuEPO, 300 units/kg per week (erythropoietin [EPO] 300, n = 33), rHuEPO, 750 units/kg per week (EPO 750; n = 28), or no treatment (control, n = 36). The rHuEPO was administered from the first week of life for 6 weeks. After EPO therapy was discontinued, 75 neonates were followed weekly until discharge and at 3, 6, and 12 months of age.

Mean numbers (±SD) of packed erythrocyte transfusions per patient from the time rHuEPO therapy was discontinued until discharge were 0.38 ± 0.64 (EPO 300), 0.23 ± 0.52 (EPO 750), 0.9 ± 1.1 (control) (p <0.05 in both EPO groups vs control). Mean reticulocyte counts at the sixth week were 6% ± 2.2% (EPO 300), 6.9% ± 2.2% (EPO 750), and 3.1% ± 2.6% (control) in the three groups (p <0.01 in both EPO groups vs control), and at the eighth week were 4.7% ± 2.8% (EPO 300), 5.4% ± 2.7% (EPO 750), and 2.6% ± 2.2% (control) (p <0.01 in both EPO groups vs control). Serum ferritin levels were significantly higher at the sixth week, and the percentage of hemoglobin F was significantly lower at 6, 8, and 10 weeks in the control group versus EPO groups. At 3, 6, and 12 months of age, there were no differences in reticulocytes, ferritin, HbF, and growth among groups.

Preterm infants who received rHuEPO had a normal pattern of erythropoiesis after the drug was discontinued. These data provide strong evidence that the anemia of prematurity is the result of a transient developmental abnormality in EPO production. (J PEDIATR 1995;127:291-7)