Thromboelastography is predictive of mortality, blood transfusions, and blood loss in patients with traumatic pelvic fractures: a retrospective cohort study.

2020 
PURPOSE In patients with traumatic pelvic fractures, thromboelastography (TEG) is a useful tool to rapidly evaluate and identify coagulation disturbances. The purpose of this study was to examine the coagulation kinetics of patients with traumatic pelvic fractures (pelvic ring and/or acetabulum) by analyzing the TEG results at initial presentation and its relationship with mortality and blood loss. METHODS A retrospective review at our Level-1 trauma center was conducted to identify Full Trauma Team activations (FTTa) with traumatic pelvic and/or acetabular fractures who were evaluated with a TEG on initial presentation between 2012 and 2016. In-hospital mortality, product transfusion, and hemoglobin changes were analyzed. Subgroup analysis was performed based on pelvic fracture type. RESULTS 141 patients with a mean age of 49.0 ± 20.8 years and mean Injury Severity Score (ISS) of 25.18 ± 12.8 met inclusion criteria. PRBC transfusion occurred in 78.0% of patients; a total of 1486 blood products were transfused. A total of 65 patients (46.1%) underwent operative treatment for the pelvic injuries, and 18 patients (12.7%) required embolization. The overall in-hospital mortality rate was 14.9%. The degree of clot lysis at 30 min (LY30) was significantly associated with blood loss (p < 0.0001), units of packed red blood cells (PRBCs) transfused (p < 0.0001), and mortality rate (p = 0.0002). CONCLUSION Increased fibrinolysis evidenced by an elevated LY30 on initial TEG in patients with traumatic pelvic fractures is associated with increased blood loss, blood product transfusions, and mortality. Future studies should evaluate the clinical utility of reversing hyperfibrinolysis on initial TEG. LEVEL OF EVIDENCE Prognostic level III.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    29
    References
    0
    Citations
    NaN
    KQI
    []