Unappreciated Risks Surround Red Blood Cell Transfusions

The authors from the Children’s National Medical Center, Washington, DC, conducted a retrospective cohort study of children treated in 5 pediatric intensive care units (PICU) from 1996 through 1999. Children were included if they had a hemoglobin level ≤9 gm/dL during their PICU stay. Patients were excluded if physiologic data 48 hours after admission was not available or if they were treated with a hyper-transfusion protocol for their routine care. The lowest hemoglobin level for each patient was considered for study entry. Diagnoses were divided into 14 mutually exclusive categories and severity of illness was assessed using the Pediatric Risk of Mortality III (PRISM III)¹ at 12 hours and the PRISM III Acute Physiology Score (APS) 24 hours after admission. The PRISM III APS includes the physiological data in PRISM III as well as respiratory rate, diastolic pressure, hemoglobin, calcium and sodium levels.² Outcome variables included mortality as well as hospital resource use, including the …