Multicenter Validation of a Simplified Score to Predict Massive Transfusion in Trauma

• DOI: 10.1097/ta.0b013e3181e42411
• Publication Date: 2010-07-12T10:34:55Z
• Authors: Bryan A. Cotton; Lesly A. Dossett; Elliott R. Haut; Shahid Shafi; Timothy C. Nunez; Brigham K. Au; Victor Zaydfudim; Marla Johnston; Patrick Arbogast; Pampee P. Young
• Keywords: Trauma Center; Major trauma; Severe trauma

Background: Several studies have described predictive models to identify trauma patients who require massive transfusion (MT). Early identification of lethal exsanguination may improve survival in this patient population. The purpose the current study was validate a simplified score predict MT at multiple Level I centers. Methods: All adult treated three centers from July 2006 June 2007 (1) were transported directly scene, (2) activations, and (3) received any blood transfusions during admission included. Assessment Blood Consumption (ABC) developed using same inclusion criteria for admitted single center (Vanderbilt University Medical Center [VUMC]-1) between 2005 2006. ABC calculated by assigning value (0 or 1) each four parameters: penetrating mechanism, positive focused assessment with sonography fluid, arrival pressure <90 mm Hg, pulse >120 bpm. A 2 used as “positive” MT. Area under receiver-operating characteristic curve compare ability institution. Results: There 586 developmental (VUMC-1), 513 1 (VUMC-2), 372 (PMH), 133 3 (Johns Hopkins Hospital). rate similar centers: 14% 15%. Sensitivity specificity predicting ranged 75% 90% 67% 88%, respectively. Correctly classified area curve, however, 84% 87% 0.83 0.90, Conclusions: is valid instrument early patient's care across various demographically diverse Future research should focus on score's prospectively will receive