Ultramassive Transfusion for Trauma in the Age of Hemostatic Resuscitation: A Retrospective Single-Center Cohort From a Large US Level-1 Trauma Center, 2011–2021

Abstract

BACKGROUND: Uncontrolled bleeding is a leading cause of death in trauma. In the last 40 years, ultramassive transfusion (UMT; ≥20 units of red blood cells [RBCs]/24 hours) for trauma has been associated with 50% to 80% mortality; the question remains as to whether the increasing number of units transfused in urgent resuscitation is a marker of futility. We asked whether the frequency and outcomes of UMT have changed in the era of hemostatic resuscitation. METHODS: We performed a retrospective cohort study of all UMTs in the first 24 hours of care over an 11-year period at a major US level-1 adult and pediatric trauma center. UMT patients were identified, and a dataset was built by linking blood bank and trauma registry data, then reviewing individual electronic health records. Success in achieving hemostatic proportions of blood products was estimated as (units of plasma + apheresis-platelets-in-plasma + cryoprecipitate-pools + whole blood]/[all units given] ≥0.5. Demographics, injury type (blunt or penetrating), severity (Injury Severity Score [ISS]), severity pattern (Abbreviated Injury Scale score for head [AIS-Head] ≥4), admitting laboratory, transfusion, selected emergency department interventions, and discharge status were assessed using χ2 tests of categorical association, the Student t-test of means, and multivariable logistic regression. P <.05 was considered significant. RESULTS: Among 66,734 trauma admissions from April 6, 2011 to December 31, 2021, we identified 6288 (9.4%) who received any blood products in the first 24 hours, 159 of whom received UMT (0.23%; 154 aged 18–90 + 5 aged 9–17), 81% in hemostatic proportions. Overall mortality was 65% (n = 103); mean ISS = 40; median time to death, 6.1 hours. In univariate analyses, death was not associated with age, sex, or more RBC units transfused beyond 20 but was associated with blunt injury, increasing injury severity, severe head injury, and failure to receive hemostatic blood product ratios. Mortality was also associated with decreased pH and evidence of coagulopathy at admission, especially hypofibrinogenemia. Multivariable logistic regression showed severe head injury, admission hypofibrinogenemia and not receiving a hemostatic resuscitation proportion of blood products as independently associated with death. CONCLUSIONS: One in 420 acute trauma patients at our center received UMT, a historically low rate. A third of these patients lived, and UMT was not itself a marker of futility. Early identification of coagulopathy was possible, and failure to give blood components in hemostatic ratios was associated with excess mortality.