In Trauma, Conventional ROTEM and TEG Results Are Not Interchangeable But Are Similar in Clinical Applicability

• Published in: Military medicine Vol. 181; no. 5 Suppl; pp. 117 - 126
• Main Authors: Rizoli, Sandro, Min, Arimie, Sanchez, Adic Perez, Shek, Pang, Grodecki, Richard, Veigas, Precilla, Peng, Henry T
• Format: Journal Article
• Language: English
• Published: England Oxford University Press 01.05.2016
• ISSN: 0026-4075; 1930-613X
• DOI: 10.7205/MILMED-D-15-00166

There is growing interest in viscoelastic hemostatic assays rotational thromboelastometry (ROTEM) and thromboelastography (TEG) for trauma. Despite shared features, it is unknown whether their results are interchangeable and whether one is clinically superior in predicting mortality, blood transfusion, and diagnosing early trauma coagulopathy. We conducted a prospective observational study comparing equivalent ROTEM and TEG parameters. Severely injured patients expected to receive massive transfusion were included. Assays were performed simultaneously on admission and repeated over subsequent 12 hours. International normalized ratio ≥1.2 or fibrinogen <1 g/L defined coagulopathy. TEG used kaolin as coagulation initiator and ROTEM used tissue factor (conventional). Spearman nonparametric analysis and Bland-Altman difference mean plot revealed parameter association. Logistic regression and receiver operating characteristic curves measured predictive values. 33 patients (74 ROTEM, 74 TEG) were included; 79% were male, mean Injury Severity Score was 23.5 ± 14, admission international normalized ratio was 1.33 ± 0.4, and 63.4% received blood transfusions. Overall, parameter agreement fell outside acceptable limits, with weak or no association. Clinically, ROTEM maximum clot firmness and TEG maximum amplitude showed reasonable predictive accuracy for mortality, strong accuracy for any or massive blood transfusion, reasonable for plasma transfusion and similar poor predictive accuracy for diagnosing coagulopathy. ROTEM and TEG results are not interchangeable, arguably due to different coagulation triggers. Assays had similar clinical performance.