BACKGROUND Limited erythropoietin (Epo) production diminishes neonates' ability to regenerate blood removed by phlebotomy. Neonates requiring surgery are at risk to receive multiple transfusions. We sought to determine if recombinant Epo administration to neonates requiring surgery would stimulate erythropoiesis. METHODS Infants were randomized in double-masked fashion to receive Epo (200 units kg(-1) d(-1)) or placebo for 14 days. Complete blood
count, absolute reticulocyte count (ARC), phlebotomy losses, and transfusions were measured during the study period. Infants were transfused using a strict transfusion protocol. RESULTS In the Epo group (n = 10, 2034 +/- 308 g, 8 +/- 2 days old; mean +/- SEM), ARC increased significantly, whereas in the placebo group (n = 10, 2400 +/- 184 g, 7 +/- 2 days old), ARC remained low. Hematocrits in the Epo group trended upward from 34. 4 1. 7% to 37. 3 1. 9% (although not statistically significant) despite phlebotomy losses of 53 +/- 12 mL/kg. Hematocrits in the placebo group were 35. 9 1. 8% and 33. 2 1. 6% on days 1 and 15, respectively, with phlebotomy losses of 27 +/- 5 mL/kg. There were no differences in absolute neutrophil counts or platelet counts between groups at the end of the study. No adverse effects were noted. CONCLUSIONS Infants randomized to Epo increased reticulocyte counts and hematocrits without adverse effects. Erythropoietin administration may provide an adjunct to present care in decreasing or eliminating erythrocyte transfusions in surgical neonates.