Is a prophylactic treatment by erythropoietin relevant to reduce red blood cell transfusion in the pediatric intensive care unit?

OBJECTIVE: An adult trial reported the efficacy of recombinant human erythropoietin in critically ill patients with a 19% decrease in red blood cell transfusion. Our aim was to evaluate the relevance of this prophylactic treatment in children hospitalized in a pediatric intensive care unit (PICU). DESIGN: Cohort study from January 1995 to December 2004. SETTING: University hospital PICU. PATIENTS: Children between 1 month and 18 yrs of age. INTERVENTIONS: We searched through a prospective databank for all children hospitalized in the PICU for > or =4 days (potential recipients of erythropoietin, as proposed in the adult trial) and transfused with red blood cells after day 7 following PICU entry (in whom erythropoietin might prevent anemia, according to results of the adult trial). MEASUREMENTS AND MAIN RESULTS: We found that 799 of 2,578 children (31%) were hospitalized for > or =4 days. The study group comprised 787 patients who were hospitalized for > or =4 days in the PICU and for whom full records were available. One hundred eighty-three children in this study group were transfused during their stay in the PICU (median age, 7 months; weight, 6.60 kg). Hemoglobin levels before transfusion (mean +/- sd) were 7.7 +/- 1.5 g/dL. These transfused children represented 23% of the study group and 7% of the total PICU admissions. Forty-seven children (6% of the study group, 2% of the total PICU admissions) were transfused with red blood cells after 7 days of hospitalization and could have benefited from a prophylactic treatment with erythropoietin. Relative risk to benefit of a prophylactic treatment by erythropoietin was higher in cases of mechanical ventilation (relative risk, 1.18) and inotropic treatment (relative risk, 1.72) and if the main diagnosis involved dermatological (relative risk, 3.03) or oncologic disease (relative risk, 3.94). CONCLUSIONS: If we applied the results of the adult trial to our PICU, we would have to treat 31% of the children with prophylactic erythropoietin and thereby expect a reduction of one red blood cell transfusion for every 17 treated patients.