The Influence of Blood Transfusion on Mortality in Multiply Injured Patients

Abstract

Severely injured patients often need large amounts of blood transfusion. Besides the life-saving effect, it is known to have negative side effects as well, affecting multiple organ systems. These side effects may add significantly to the already high mortality rate of these patients. This study was undertaken to determine the independent influence of the speed (= relative) as well as the total (= absolute) amount of blood transfusion given to multiply injured patients on mortality. All consecutive multiply injured patients with an Injury Severity Score (ISS) ≥ 18, based on the Hospital Trauma Index (HTI), treated over a 4-year period in the University Medical Center Utrecht, The Netherlands, were reviewed. Using cross tabulation analysis, the relation between relative and absolute blood transfusion and mortality was described. Potential transfusion-influencing factors were selected on the basis of literature and common knowledge. The independent effect of blood transfusion on mortality was calculated by means of nonparametric tests and binary logistic regression analysis. The mean age of the 668 included patients was 36.8 years (standard deviation [SD] ± 18.5 years) and the median HTI-ISS 30 (range 18–75). 360 patients needed blood transfusion. Their mean relative blood transfusion was 8.5 PRBCrel (SD ± 11) (and their mean absolute blood transfusion 12.7 PRBCabs (SD ± 17). The overall mortality was 18.4%. The patients without blood transfusion (n = 308) had a mortality rate of 14%. The mortality rate of the patients with > 30 PRBCrel relative transfusion (n = 14) was highest with 50%. Both relative as well as absolute transfusion displayed no significant influence on mortality. The HTI-ISS, hemothorax, splenic and hepatic injury, laparotomy, vascular or retroperitoneal injury, and fractures of the pelvis and/or femur were exhibited as having significant influence on blood transfusion. The seemingly significant influence of both relative and absolute blood transfusion on mortality, even in high amounts, disappeared after correction for the significant determinants mentioned above. Blood transfusion itself does not turn the scale in decision-making for treatment of severely injured patients.