Age of transfused blood: an independent predictor of mortality despite universal leukoreduction.

BACKGROUND The transfusion of relatively older stored blood has been associated with an increased risk of multiple organ failure, infection, and death. It remains unknown whether this phenomenon is mitigated by transfusion of leukoreduced red cell units. The purpose of this study was to evaluate the influence of stored blood age on mortality in injured patients who universally received leukoreduced blood. METHODS Trauma patients who received > or = 1 unit of blood during the first 24 hours after hospital arrival were selected for inclusion. Patients were stratified both according to total units and "old" units (> or = 14 days) versus "young" units (< 14 days) received in the initial 24 hours. Odds ratios and 95% confidence intervals (CIs) were calculated for the association between mortality and the age and amount of blood transfused, adjusted for age, sex, injury severity, injury mechanism, number of units transfused, and length of stay. RESULTS Over 7.5 years, 1,813 patients met study criteria. Among patients who received a total of 1 to 2 or 3 to 5 units in the first 24 hours, there was no association between the amount and age of transfused blood and mortality. For patients who received a total of > or = 6 units, the presence of > or = 3 units of young blood was associated with a 3.8-fold increased odds of death (CI: 1.1-12.7), compared with a 7.8-fold (CI: 2.3-26.3) increased odds of death associated with the presence of > or = 3 units of old blood (p = 0.0024). CONCLUSION Although larger volumes of blood, irrespective of age, are associated with increased odds of mortality, the transfusion of blood stored beyond 2 weeks appears to potentiate this association despite a practice of universal leukoreduction. For patients who receive relatively smaller transfusion volumes, blood age appears to have no effect on mortality.