Respective effects of phlebotomy losses and erythropoietin treatment on the need for blood transfusion in very premature infants

Background: The benefit to risk ratio of the treatment with erythropoietin (EPO) as a means of limiting the number of transfusions in very preterm infants during hospitalization, seems to be modest since the adoption of restrictive transfusion criteria and of policy limiting phlebotomy losses. We therefore aim to evaluate the factors associated with the number of late blood transfusion in very preterm infants in a unit where the routine use of EPO has been discontinued. Methods: A comparative "before-after" study was carried out in premature infants born before 32 weeks postmenstrual age (PMA), over a period of one year before (EPO group) and one year after (non-EPO group) the discontinuation of EPO therapy. Results: A total of 48 infants were included in the study (EPO=21; non-EPO=27). The number of infants transfused after the 15 day of life (D15) and the number of transfusions per infant after D15 were not significantly different between the two groups. In a multivariate analysis, the gestational age and the volume of blood drawn off during the first month of life significantly influenced the need for transfusions after the 15th day of life, independently of the treatment with EPO. The hemoglobin levels measured at different times of hospitalization (median postnatal age: 16, 33 and 67 days) were not significantly different between the two groups. Conclusions: Our study shows that the discontinuation of EPO did not change the number of late transfusions. Even when a policy limiting phlebotomy losses is used, blood loss is an important and independent risk factor for late transfusion of very preterm infants.