• Pierre Esnault, Quentin Mathais, Erwan D'Aranda, Ambroise Montcriol, Mickaël Cardinale, Pierre-Julien Cungi, Philippe Goutorbe, Christophe Joubert, Arnaud Dagain, and Eric Meaudre.
• Intensive Care Unit, Sainte Anne Military Hospital, Toulon, France. pierre.esnault@gmail.com.
• Neurocrit Care. 2020 Aug 1; 33 (1): 182-195.

BackgroundProgressive hemorrhagic injury (PHI) is common in patients with severe traumatic brain injury (TBI) and is associated with poor outcomes. TBI-associated coagulopathy is frequent and has been described as risk factor for PHI. This coagulopathy is a dynamic process involving hypercoagulable and hypocoagulable states either one after the other either concomitant. Fibrin monomers (FMs) are a direct marker of thrombin action and thus reflect coagulation activation. This study sought to determine the ability of FM to predict PHI after severe TBI.MethodsWe conducted a prospective, observational study including all severe TBI patients admitted in the trauma center. Between September 2011 and September 2016, we enrolled patients with severe TBI into the derivation cohort. Between October 2016 and December 2018, we recruited the validation cohort on the same basis. Study protocol included FM measurements and standard coagulation test at admission and two computed tomography (CT) scans (upon arrival and at least 6 h thereafter). A PHI was defined by an increment in size of initial lesion (25% or more) or the development of a new hemorrhage in the follow-up CT scan. Multivariate logistic regression analysis was applied to identify predictors of PHI.ResultsOverall, 106 patients were included in the derivation cohort. Fifty-four (50.9%) experienced PHI. FM values were higher in these patients (151 [136.8-151] vs. 120.5 [53.3-151], p < 0.0001). The ROC curve demonstrated that FM had a fair accuracy to predict the occurrence of PHI with an area under curve of 0.7 (95% CI [0.6-0.79]). The best threshold was determined at 131.7 μg/ml. In the validation cohort of 54 patients, this threshold had a negative predictive value of 94% (95% CI [71-100]) and a positive predictive value of 49% (95% CI [32-66]). The multivariate logistic regression analysis identified 2 parameters associated with PHI: FM ≥ 131.7 (OR 6.8; 95% CI [2.8-18.1]) and Marshall category (OR 1.7; 95% CI [1.3-2.2]). Coagulopathy was not associated with PHI (OR 1.3; 95% CI [0.5-3.0]). The proportion of patients with an unfavorable functional neurologic outcome at 6-months follow-up was higher in patients with positive FM: 59 (62.1%) versus 16 (29.1%), p < 0.0001.ConclusionsFM levels at admission had a fair accuracy to predict PHI in patients with severe TBI. FM values ≥ 131.7 μg/ml are independently associated with the occurrence of PHI.

40 keywords...

hide…