FIBRINOGEN LEVEL AS A PREDICTOR FOR MASSIVE TRANSFUSION AMONG COAGULATION/FIBRINOLYSIS MARKERS IN SEVERE BLUNT TRAUMA PATIENTS : A MULTICENTER RETROSPECTIVE ANALYSIS

Abstract

   [Background] We evaluated whether coagulation/fibrinolysis markers can be a predictor for massive transfusion (MT) in patients with severe blunt trauma. [Methods] This study included blunt trauma patients based on the J–OCTET database. Patients were divided into MT (transfusion of 10 or more units of packed red blood cells in the first 24 hours after admission) and non–MT groups. We evaluated MT predictors (age, sex, vital signs, Glasgow Coma Scale, hemoglobin, platelet count, lactate, PT–INR, D–dimer, and fibrinogen levels on admission) using stepwise multiple logistic analysis. We evaluated MT prediction accuracy by receiver operating characteristic curve, and optimal cut–off value were calculated using Youden index. [Results] Heart rate (HR) (per 10 bpm), body temperature (BT), and fibrinogen (per 10 mg/dL) were independent predictors of MT [Odds ratio (OR), 1.72, p<0.001, OR, 0.70, p=0.013, and OR, 0.89, p<0.001, respectively] . The area under the curve values for MT were 0.81 in HR, 0.60 in BT, and 0.75 in fibrinogen. Optimal cut–off values for HR, BT, and fibrinogen were 96 bpm (sensitivity 73.7%, specificity 76.6%), 36.3°C (66.7%, 48.7%), and 190 mg/dL (61.4%, 80.2%), respectively. [Conclusions] Decreased levels of fibrinogen may be one of the independent predictor of MT, with higher specificity in comparison with HR, BT.