Trauma-induced coagulopathy : an investigation of fibrinolysis and the effect of tranexamic acid

Haemorrhage is a leading cause of trauma morbidity and mortality, with many deaths potentially preventable. Hyperfibrinolysis is a central characteristic of trauma-induced coagulopathy (TIC) which develops rapidly and is associated with poor outcomes. Tranexamic acid (TXA) improves survival in trauma haemorrhage but its uptake worldwide remains variable, in part because its effects on the coagulation system during trauma haemorrhage have not been described. Further uncertainty regarding patient selection for TXA therapy has emerged following the description of an early viscoelastic haemostatic assay (VHA) diagnosed hypofibrinolytic phenotype in whom TXA may potentiate thrombotic complications. The patient characteristics and mechanisms leading to this apparent hypofibrinolytic phenotype are poorly understood. Over 900 trauma patients prospectively recruited to a multicentre observational cohort study had blood drawn within 2-hours of injury for VHA and fibrinolysis plasma protein analysis. Patients were categorised according to VHA maximum lysis (ML) and D-dimer (DD) levels. Patients with MLLOW exhibited heterogeneity in clinical and injury characteristics and outcomes. Those who died were severely injured, with a high incidence of traumatic brain injury and a 7-fold higher D-dimer. Patients with MLLOW+DDHIGH had a hyperfibrinolytic biomarker profile, with the fibrinolytic mediator S100A10 identified as a potential driver of fibrinolysis, which can ex-vivo artificially reduce ML. Empiric TXA could benefit this occult hyperfibrinolytic phenotype. Over two subsequent observational studies, the effects of TXA on the coagulation system during trauma haemorrhage and the effect of TXA infusion and timing of treatment on thrombotic events were investigated. Early empiric TXA avoided VHA-hyperfibrinolysis and provided a degree of protection from TIC. Whilst univariate analysis suggested increased thromboses with later TXA treatment in patients receiving TXA bolus+infusion, neither the TXA infusion nor time to bolus were associated with thrombotic events after multivariate analysis. A single TXA bolus may provide a lower effective therapeutic dose with reduced complications.