Érythropoïétine humaine recombinante : analyse d’une politique de prescription dans une population hospitalière de nouveau-nés de faible poids de naissance

Abstract

Aim of the study. – To evaluate a policy of treatment with human recombinant erythropoietin (rhEPO) and to describe factors related to red blood cell transfusions (RBCTs) in treated neonates. Study. – Prospective, observative study. Patients and methods. – One-hundred and sixty-five neonates with gestational age (GA) < 30 weeks and/or birthweight < 1000g admitted between may 1998 and october 1999. Ninety were excluded (congenital malformations n = 6, deaths n = 16, referral to a general hospital before discharge n = 67, ECMO n = 1). Data about the characteristics of the population, the severity of the neonatal period, hemoglobin at birth, blood loses, treatment with rhEPO, number of red blood cells transfusions (RBCTs) and donors were recorded in all infants. Results. – Thirty-eight in seventy-five (51%) neonates received 112 blood transfusions. Eighty-eight were prescribed after day 15. In most of the cases ( n = 68), RBCTs were done according to the protocol. In 20 cases (23%) infants were transfused during a late-onset infection. No difference was observed between the non-transfused (group I) and the transfused neonates (group II) with regards to the drug administration: first dose on day 3 ± 2, number of injections (17 ± 4 vs 18 ± 1, ns). The start of oral supplementation with iron was late (12j ± 8 vs 19j ± 10, ns). Infants in group II had a lower birthweight (850 ± 240 vs 1050 ± 160g, p < 0,01) for a similar GA (28 ± 1SA vs 28 ± 2SA, ns) in association with an increased number of small for date babies ( p = 0.03). Antenatal steroïds administration (89 vs 74%, ns), administration of surfactant (59 vs 81%, ns) were similar in the two groups. The Clinical Risk Index for Babies was higher in group II: 5 ± 3 vs 2 ± 1 ( p < 0,001) as was the duration of oxygen delivery (53 ± 44 vs 14 ± 20 days, p < 0,01) and postnatal administration of corticosteroïds ( 38% vs 3%, p < 0.01). Conclusion. – The quality of iron administration, RBCTs and the limitation of donors could be improved in our population. Transfusions among neonates born before 30 weeks and/or with a birthweight of less than 1000g and treated with rhEPO are associated with intrauterine malnutrition and a worse clinical condition on admission. Early identification of at risk neonates could improve prevention of RBCTs and the efficacy of rhEPO administration to preterm infants.