A. Duong1, X. Zhang1, J. Karri2, B. Cotton1, C. Wade1, J. Cardenas1 1McGovern Medical School,Surgery,Houston, TX, USA 2Baylor College Of Medicine,Department Of Physical Medicine And Rehabilitation,Houston, TX, USA
Introduction:
Progressive hemorrhagic injury (PHI), the early expansion of intracranial hemorrhage (ICH) following a traumatic brain injury (TBI), is associated with a five-fold increase in the risk of neurological decline, complications, and mortality. Our lab, and others, had previously demonstrated fibrinolysis to be an important mechanism underlying PHI and that admission levels of D-dimer, a byproduct of fibrinolysis, may be used to screen for PHI. In this study, we aimed to independently validate the previously proposed D-dimer of 3.6 µg/mL threshold for PHI identification and hypothesized it would accurately differentiate patients with PHI from those with stable hemorrhage (SH).
Methods:
This was a single institution, retrospective analysis of prospectively collected data between September 2013 and March 2018. All highest-level trauma activation patients between ages of 18-55 years, with an initial head CT scan at admission showing severe TBI, defined as presence of ICH, and a repeat scan within 6 hours following, were included. This limit on age was set to minimize inclusion of patients on anticoagulants. Prisoners, pregnant women, and patients with non-survivable head injury were excluded. Differentiability of D-dimer was evaluated by receiver operating curve analysis. Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated for the previously identified cut point of 3.6 µg/mL.
Results:
From September 2013 to March 2018, 244 patients between 18-55 years old were admitted to Memorial Hermann Hospital with severe TBI. An additional 75 patients were excluded due to non-survivable injuries, lack of CT scan, did not consent, or plasma was not obtained. This left 169 patients for analysis. Of those, 54% had SH and 46% had PHI. Patients who developed PHI received more red blood cell transfusions (p =0.02) and had significantly fewer ventilator-, intensive care unit-, and hospital-free days compared to SH patients (all p <0.01). Furthermore, PHI patients had significantly higher D-dimer levels than those with SH (8.1 (4.6, 17.1) μg/mL vs 5.7 (1.8, 11.7) μg/mL, respectively, p <0.01). The D-dimer threshold of 3.6 μg/mL yields a sensitivity of 0.83 (95% CI 0.74-0.91), specificity of 0.40 (95% CI 0.30-0.50), positive predictive value of 0.53 (95%CI 0.44-0.62), and negative predictive value of 0.74 (95% CI 0.61-0.86).
Conclusion:
Clinical use of D-dimer on admission in TBI patients could provide a sensitive screening tool to identify patients at high risk for PHI. Such a tool could be used for early administration of life-saving interventions for high risk patients and also eliminate costly, invasive screening procedures for patients at low risk of PHI.