185 The Utility of Thromboelastography for Predicting the Risk of Progression of Intracranial Hemorrhage in Traumatic Brain Injury Patients

Abstract INTRODUCTION: Progression of traumatic intracranial hemorrhage (TICH) is associated with increased morbidity and mortality. Half of patients with TICH progression have normal traditional coagulation laboratory studies. We hypothesized that thromboelastography (TEG), which directly measures clot formation and strength, would predict progression of TICH, the need for neurosurgical procedure, and mortality in patients with TICH. METHODS: A prospective observational study in patients with TICH at a level 1 trauma center was undertaken. Progression of TICH was defined as >30% increase in TICH volume or by radiologist interpretation. Baseline demographic and clinical data were analyzed. TEG was performed at admission and 6 hours postadmission. Regression models were developed to determine whether admission TEG, the change between admission and 6-hour TEG, and TEG dichotomized as normal vs abnormal predict each outcome, controlling for important covariates. RESULTS: In 169 subjects, 56% developed progression of TICH between admission and 48 hours; 25% underwent a neurosurgical procedure. Overall mortality was 7%. Progressors had higher injury severity scores than nonprogressors (median = 26, interquartile range [IQR] 17-30 vs median = 17, IQR 10-26; P < .01) and were more likely to have admission Glasgow Coma Scale [GCS] < 8 (P < .01). There were no differences in international normalized ratio, partial thromboplastin time, fibrinogen, platelets, or D-dimer between progressors and nonprogressors. There were no differences in median TEG values based on progression, mortality, or receipt of neurosurgical procedure status. For every 1% increase in LY30 during the first 6 hours postadmission, the odds of receiving a neurosurgical procedure increased by 17% (P = .04). In addition, subjects with an abnormal K time had 4.7 times the odds of mortality as those with a normal K time (95% confidence interval 1.09-20.44, P = .04). CONCLUSION: In patients with TBI, TEG values obtained shortly after admission do not predict progression of TICH, need for neurosurgical procedure, or mortality. We caution against relying on TEG to guide clinical decisions related to TICH progression.