Early massive transfusion in trauma patients: Canadian single-centre retrospective cohort study.

PURPOSE

To determine associations between red blood cell (RBC) transfusion and early and late clinical outcomes in massively transfused adult trauma patients.

METHODS

A retrospective cohort study (1992-2001) including 260 patients receiving >or=10 RBC units <or=24 hr after admission to a university-affiliated trauma centre. We extracted demographic and clinical data and used multivariable regression to determine independent effects of RBC transfusion on clinical outcomes.

RESULTS

Patients had a high (mean [standard deviation]) injury severity score (ISS) (42.5 [15.1]), a high admission sequential organ failure assessment (SOFA) score (8.4 [3.8]), and a high hospital mortality (58.5%). They received 38 (25-64) (median [interquartile range]) blood components within 48 hr, including 19 (14-28) RBC units. For 143 patients surviving >or=48 hr, the maximum SOFA score was associated with RBC units transfused before 48 hr (linear regression beta coefficient 0.075, P < 0.0001), lower nadir hemoglobin before 48 hr (0.034, P = 0.03), age (0.032, P = 0.015), and admission SOFA (0.59, P < 0.0001). The RBC units transfused by 48 hr were not associated with either hospital mortality (n = 35) among patients surviving >or=48 hr (independent predictors, age [logistic regression odds ratio (OR) 1.06, 95% confidence interval 1.03-1.10], ISS [OR 1.07, 1.02-1.13], and maximum SOFA score [OR 1.56, 1.27-1.93]) or 48-hr mortality (n = 117) (independent predictors, admission SOFA [1.65, 1.45-1.88] and later year of hospital admission [OR 1.15, 1.02-1.29]).

CONCLUSIONS

Hospital mortality is high among massively transfused trauma patients. Among early survivors, 48-hr RBC transfusion volume is associated with increased organ dysfunction, but not hospital mortality. Also, it is not associated with 48-hr mortality. Future research should continue to explore methods to improve hemostasis and minimize the need for RBC transfusion.