Retrospective Review of Erythropoietin Therapy Versus Supportive Care in Newborns with Hereditary Spherocytosis (HS): Difficulties Implementing Trial Results Into Clinical Practice

Most newborns with HS have a family history of disease since the most common inheritance pattern is autosomal dominant. Therefore, hematologists often see these infants before their physiologic hemoglobin nadir, which is exaggerated in comparison to healthy infants and can cause a symptomatic anemia requiring red cell transfusion(s) during the first few months of life. In 2000, a small multicenter prospective study demonstrated that recombinant erythropoietin (EPO) therapy can aid infants through this nadir and prevent transfusions (Tchernia et al., Hematol J. 2000; 1(3):146–152). The goal of the present study was to evaluate the frequency of implementation and cost of EPO versus transfusion therapy in infants with HS at a pediatric academic center.

After IRB approval, 59 infants with HS were identified from hematology department records and from an electronic clinic note search tool at Children's Hospital Boston. Subjects with a date of birth from July 2000 through June 2011 were included. Demographic, clinical, laboratory, and transfusion data were collected through a retrospective chart review. Statistical analysis was performed using R 2.13.1.

Only 9 of 59 (15%) infants with HS were treated with EPO therapy in the decade since the study by Tchernia et al. was first published. Table I shows the demographic and clinical data for EPO vs. non-EPO treated newborns. The mean age at the start of treatment was 5.3 ± SD 3.6 weeks. In our academic center of >15 clinical hematologists, two commenced EPO therapy for over half of the treated patients. Of those treated with EPO, 6 (67%) had an older sibling with HS of whom 3 (33%) were treated with EPO. The nadir hemoglobin was not different in the EPO vs. non-EPO treated groups (7.6 vs 8.1 g/dl, p=0.2). The difference in the number of transfusions needed in the first year of life in the two groups was not statistically significantly (0.9 vs 1.1, p=0.52), although 6 of the 9 EPO treated patients received a red cell transfusion prior to starting therapy. EPO was discontinued at a mean age of 23.5 weeks (12–37 weeks). The estimated cost of the average course of EPO therapy was approximately $5437 per infant (AWP plus repeated injections) versus the cost of a single blood transfusion of $1783 per infant (estimated gross charges at our center).

Since a 2000 report of the efficacy of EPO to limit transfusions in newborns with HS, only 15% of infants at our center have been treated with EPO, despite common belief among hematologists that this is a frequent treatment modality. The decision to start EPO appeared arbitrary, based on hematologist preference and previous experiences of families. The estimated cost of EPO therapy is greater than a single outpatient red cell transfusion, and the majority of EPO treated infants also received a blood transfusion. It remains unclear whether EPO decreases the number of transfusions required in infants with HS. The lack of implementation of EPO therapy at our institution is likely due to a combination of financial and logistical challenges, in addition to individual practice customs, which often impede translation of prospective trials into clinical practice. A larger prospective study accounting for these factors could establish “best practice” for infants with HS.

Off Label Use: Recombinant erythropoietin for anemia in infancy associated with hereditary spherocytosis.