A Predictive Model for Massive Transfusion in Combat Casualty Patients

(MT) is associated with increased morbidity and mortality in severely injured patients. Early and aggressive use of blood products in these patients may correct coagulopathy, control bleeding, and improve outcomes. However, rapid identification of patients at risk for MT has been difficult. We postu-lated that evaluation of clinical variables routinely assessed upon admission would al-low identification of these patients for ear-lier, more effective intervention. Methods: A retrospective cohort study was conducted at a single combat support hospital to identify risk factors for MT in patients with traumatic inju-ries. Demographic, diagnostic, and labo-ratory variables obtained upon admission were evaluated. Univariate and multivar-iate analyses were performed. An algo-rithm was formulated, validated with an independent dataset and a simple scoring system was devised. Results: Three thousand four hun-dred forty-two patient records were re-viewed. At least one unit of blood was transfused to 680 patients at the combat support hospital. Exclusion criteria in-cluded age less than 18 years, transfer from another medical facility, designation as a security internee, or incomplete data fields. The final number of patients was 302, of whom 26.5 % (80 of 302) received a MT. Patients with MT had higher mortal-ity (29 vs. 7 % [p < 0.001]), and an in-creased Injury Severity Score (25 11.1 vs. 18 16.2 [p < 0.001]). Four indepen-dent risk factors for MT were identified: heart rate>105 bpm, systolic blood pres-sure <110 mm Hg, pH <7.25, and hemat-ocrit <32.0%. An algorithm was created to analyze the risk of MT (area under the curve [AUC] 0.839). In an independent data set of 396 patients the ability to accu-rately identify those requiring MT was 66% (AUC 0.747). Conclusions: Independent predic-tors for MT were identified in a cohort of severely injured patients requiring trans-fusions. Patients requiring a MT can be identified with variables commonly ob-tained upon hospital admission