Tranexamic acid and trauma

To the Editor: Gruen and colleagues1 have not accurately represented the current safety data on the risk of vascular occlusive events (VOE) with tranexamic acid (TxA), including deep vein thrombosis (DVT) and pulmonary embolism (PE).

A systematic review of data for 10 488 surgical patients showed lower mortality with TxA and no difference in DVT (risk ratio [RR], 0.86; 95% CI, 0.53–1.39; P = 0.54) or PE (RR, 0.61; 95% CI, 0.25–1.47; P = 0.27).2 The CRASH-2 (Clinical Randomisation of an Antifibrinolytic in Significant Haemorrhage 2)3 trial involving 20 211 patients showed there was a trend to fewer VOE in the tranexamic acid group compared with placebo (168/10 060 [1.7%] v 201/10 067 [2.0%]; P = 0.084). Gruen et al implied that the low rate of PE (0.7%) and DVT (0.4%) in this study renders these findings suspect. Low detection of DVT and PE applied equally to both groups and only affects study power — which is not an issue in this large study.

Gruen and colleagues stated that “in contrast” the observational MATTERs (Military Application of Tranexamic Acid in Trauma Emergency Resuscitation)4 trial showed PE and DVT rates that were, respectively, nine and 12 times the rates for the control group. They neglected to mention that the rates in the control group were 2/603 (0.3%) and 1/603 (0.2%) — lower than the rates in the less severely injured CRASH 2 patients. The MATTERs study was unblinded, so the difference could easily have been due to more investigations in the TxA group.

Finally, the MATTERs study reported that the inhospital mortality in the TxA group was 17.4% compared with 23.9% in the group not given TxA (P = 0.03). Unlike non-fatal DVT and PE, death is not a condition that requires investigations to diagnose.