RBC transfusions in children requiring intensive care admission after traumatic injury*.

Objective: To describe packed RBC utilization patterns in trauma patients admitted to a PICU and study associated outcomes while controlling for severity.

Design: Retrospective cohort study.

Setting: The PICU of a tertiary care children's hospital.

Patients: All pediatric trauma patients admitted to Helen DeVos Children's Hospital PICU between June 2007 and July 2010, either directly from the emergency department or transferred from another institution.

Interventions: None.

Measurements And Main Results: Of 389 trauma patients, 107 patients (27.5%) transferred to the PICU were transfused with blood products. Of these transfusions, 81 were packed RBC transfusions and 26 were other blood products. Only 73 of the packed RBC transfusions had a documented time of transfusion: 17 (23.3%) were transfused prior to PICU admission, seven (9.5%) both before and after PICU, and 49 (67.1%) only after PICU admission. After adjusting for injury severity score, transfused patients had higher odds of needing mechanical ventilation (odds ratios, 9.2; 95% CI, 3.6-23.3) and higher risk of mortality (odds ratios, 8.6; 95% CI, 2.6-28.6), when compared with nontransfused patients. Mean age of packed RBC was 19.6 ± 9.3 days (mean ± SD). The impact of age of packed RBCs on mortality was examined as a categorical variable at 14, 21, and 28 days. Packed RBCs more than 28 days old (14/61 patients) were associated with longer lengths of stay (13 ± 12 vs 7 ± 6; p < 0.03), lower discharge Glasgow Coma Scale score (9 ± 6 vs 13 ± 4; p< 0.03), and more mortality (43% vs 13%; p < 0.02) when compared with blood less than 28 days old.

Conclusions: In pediatric trauma patients, transfusion of packed RBC and use of older RBC units are associated with higher risk of adverse outcomes independent of injury severity.