• Herb A Phelan, Richard P Gonzalez, Hetal D Patel, Jamie B Caudill, Rachel K Traylor, Lydia R Yancey, Jason L Sperry, Randall S Friese, and Paul A Nakonezny.
• Division of Burns/Trauma/Critical Care, Department of Surgery, UT Southwestern Medical Center, Dallas, Texas 75390-9158, USA. phelan@utsouthwestern.edu
• J Trauma. 2010 Aug 1;69(2):330-7.

BackgroundPrevious studies have demonstrated that the transfusion of older blood is independently associated with higher rates of infectious complications, multiple organ failure, and mortality. Putative mechanisms implicate leukocytes in stored blood that generate immunomodulatory mediators as the stored blood ages. The purpose of this retrospective cohort study was to describe the effect of prestorage leukoreduction (PS-LR) on the detrimental clinical effects of increasing age on blood products used in trauma patients.MethodsAll patients receiving >or=6 units of packed red cells and surviving >or=48 hours since May 1999 when institutional universal PS-LR was begun were identified. Transfusion requirements, demographic data, and causes of death were collected. Blood bank records were reviewed to determine the age of each unit of blood transfused. Multivariate logistic regression was used to determine the relationship between the age of PS-LR transfused blood and mortality after adjusting for total transfusion requirement, patient age, Injury Severity Score, head Abbreviated Injury Score, mechanism of injury, and gender. A subgroup analysis was performed excluding those patients in whom care was withdrawn at 48 hours to 72 hours postinjury for brain death or neurologic devastation.ResultsA total of 399 patients, receiving 6,603 units of blood, met inclusion criteria. Mortality analysis showed that increasing Injury Severity Score, patient age, head Abbreviated Injury Score, and number of units of packed red cells transfused were all independently associated with an increased risk of death. When mean age of blood was analyzed as a continuous variable, a significant reduction in the risk of death with increasing mean age of transfused PS-LR blood was noted (odds ratio [OR], 0.959; 95% confidence interval [CI], 0.924-0.996). Both of these findings persisted when the mean age of blood was dichotomized at 14 days (OR, 0.426; 95% CI, 0.182-0.998) and 21 days (OR, 0.439; 95% CI, 0.225-0.857). The area under the curve for the receiver operating characteristics of our mortality model was 0.90. After excluding 13 patients in whom care was withdrawn 48 hours to 72 hours postinjury for brain death or neurologic devastation, the mortality analysis still showed that increasing injury severity, number of units of packed red cells transfused, and age were all independently associated with an increased risk of death. The protective effect of receiving older blood seen in the all-cause mortality analysis disappeared because no association was found between odds of dying and increasing age of packed red blood cells units transfused. This was true whether the mean age of transfused blood was dichotomized at 14 days (OR, 0.93; CI, 0.30-2.83) or at 21 days (OR, 0.54; CI, 0.25-1.16).ConclusionOur data suggest that the deleterious effects of aging on banked blood are ameliorated by PS-LR. We are currently conducting a prospective observational study in an effort to duplicate the findings of this retrospective investigation.

39 keywords...

hide…