Anemia is very common in very low birth weight (VLBW) infants(1,2). It may become severe enough to require red blood cells transfusion. This has been related to blood letting for laboratory analysis and hyporegenerative state of the bone marrow secondary to erythropoietin (Epo) deficiency(1). Recombinant form of erythropoietin or rhEPO became available in mid 1980’s. In VLBW infants, randomized clinical studies with parenteral rhEPO, has been shown to maintain a stable hematocrit and decreased the need for red blood cells transfusion(1-5). While parenteral rhEPO has been shown to enhance erythropoiesis and decrease the need for late blood transfusion in preterm infant, the effects of its enteral administration provided conflicting results(6-10). We report the results of our study evaluating the effects of enteral rhEPO on erythropoiesis in stable preterm infants.

Preterm infants with birth weight less that 1800 g and gestational age less than 34 weeks were enrolled and sequentially openly randomized. Additional criteria for eligibility were stable cardiopulmonary status for at least 7 days and enteral feeding of at least 50 mL/kg/day of human milk. Infants with congenital anomalies, grade III and IV intracranial hemorrhage, seizures, immune-mediated hemolytic anemia and evidence of acquired and congenital infection were excluded. The Study group (n = 7) received three times weekly enteral administration of 400 U rhEPO, added to breast milk, and daily administration of enteral FeSO4 3-6 mg/kg/day until discharge. The Control group (n = 9) received only enteral FeSO₄. Complete blood count, reticulo- cyte counts (corrected with hematocrit), serum erythropoietin (enzyme linked immunosorbant assay, Gamma Counter Clin, LKB Company, Germany) and serum ferritin (quantitated by radioimmuno-assay, Immunotec Company, France) levels were measured at the day of entry; 10 days post initiation of the study and on the day of discharge. All the babies were fed human milk. Adverse events such as feeding intolerance, necrotizing enterocolitis, circulatory instability, temperature instability, thrombocytopenia and leukopenia were recorded. Data were analyzed utilizing SPSS software. Study and Control groups were compared by t-test for repeated measures because the distribution of outcome variables was normal according to Kolmogorov-Sminoroff test. Our institutional review committee for ethical research approved the study. Written informed consent from parents was obtained prior to enrollment of the subjects in the study.

Clinical characteristics of the Study and the Control group are shown in Table I. There were no difference in birth weight, gestational age and the day of enrollment in the study between the Study and the Control group. The Study group received rhEPO from the day of enrolment to discharge. They received rhEPO for 25 ± 10 days (range 15-35 days). Two subjects from the Control group developed recurrent apnea and septicemia. Based on clinical assessment, they were given red blood cells transfusion. They were excluded from further data analysis. No subject from the study group required red blood cells transfusion. No other significant adverse event was noted in the remaining control subjects or in the study group. Table II shows comparative hematologic status, serum erythropoietin and serum ferritin levels between Study and Control group. Reticulocyte counts and serum erythropoietin levels were significantly higher in the Study group after 10 day of rhEPO treatment and at discharge. There were no significant differences in serum ferritin level, hemoglobin and hematocrit between the groups after 10 days of therapy or at discharge.