LAUSR

Supplemental Digital Content is available in the text. Objective: To test the hypothesis that early RBC transfusion is associated with duration of organ dysfunction in critically ill septic children. Design: Secondary analysis of a single-center prospective observational study. Multivariable negative binomial regression was used to determine relationships between RBC transfusion within 48 hours of sepsis onset and number of days in 14 with organ dysfunction, or with multiple organ dysfunction syndrome. Setting: A PICU at a quaternary care children’s hospital. Patients: Children less than 18 years old with severe sepsis/septic shock by consensus criteria were included. Patients with RBC transfusion prior to sepsis onset and those on extracorporeal membrane oxygenation support within 48 hours of sepsis onset were excluded. Interventions: None. Measurements and Main Results: Ninety-four patients were included. Median age was 6 years (0–13 yr); 61% were male. Seventy-eight percentage had septic shock, and 41 (44%) were transfused RBC within 48 hours of sepsis onset (early RBC transfusion). On multivariable analyses, early RBC transfusion was independently associated with 44% greater organ dysfunction days (adjusted relative risk, 1.44 [1.04–2.]; p = 0.03), although risk differed by severity of illness (interaction p = 0.004) and by shock severity (interaction p = 0.04 for Vasoactive Inotrope Score and 0.03 for shock index). Relative risks for multiple organ dysfunction syndrome days varied by shock severity (interaction p = 0.008 for Vasoactive Inotrope Score and 0.01 for shock index). Risks associated with early RBC transfusion were highest for the children with the lowest shock severities. Conclusions: In agreement with previous studies, early RBC transfusion was independently associated with longer duration of organ dysfunction. Ours is among the first studies to document different transfusion-associated risks based on clinically available measures of shock severity, demonstrating greater transfusion-associated risks in children with less severe shock. Larger multicenter studies to verify these interaction effects are essential to plan much-needed RBC transfusion trials for critically ill septic children.