Dear Sirs,

Severe bleeding is a catastrophic complication of haemostasis. In contrast to acute thrombosis, where low-molecular-weight-heparin (LMWH) (eventually combined with urokinase or t-PA) is the physiologic drug of first choice in the emergency situation, there has been no safe drug to treat severe bleeding in routine intensive care medicine. Tranexamic acid might now be considered such a drug for patients at risk of severe bleeding [1].

The beneficial action of tranexamic acid on severe bleeding is usually attributed to its inhibitory action on blood fibrinolysis: about 0.3 mM tranexamic acid inhibits clot lysis by 50% [2]. However, fibrinolysis normally acts in hours or days. The rapid stopping of bleeding has to be attributed to another mechanism. Since tranexamic acid is a potent trigger of intrinsic coagulation, enhancing intrinsic thrombin generation twofold at 0.02-0.2 mM [3], bleeding could be stopped by contact phase activated thrombin generation. Tranexamic acid as a xenobiotic folds factor 12 into factor 12a or prekallikrein into kallikrein (see figure attached)[4].

Current practice is to infuse 1 g tranexamic instantaneously, followed by 1 g over 8 hours [1]. Unfortunately, there is great inter-individual difference in triggering the contact phase of coagulation [5], which is why it is not easy to predict which dosage of contact trigger is exactly necessary for the individual patient. A fixed initial dosage could be given in the emergency room and the following appropriate dosage might be determined by performing differentiated ultra-specific, ultra-sensitive recalcified thrombin generation assays (RECA) in the plasma of the individual patient [6]. Instead of tranexamic acid another contact trigger such as hydroxyethyl starch (HES) [7] could be given to stop severe bleeding. An alternative to infusion of a thrombin generating drug would be the application of ultrasound exactly at the position of the bleeding if known [8].

Acknowledgements
There was no specific funding and there are no conflicts of interest. The rights of the article are transferred upon acceptance to BMJ.

Institute PD Dr. med. Thomas Stief of the Laboratory Medicine and Pathobiochemistry, University Hospital of Marburg, Baldingerstr. D-35043 Marburg, Germany

References
1. Roberts I, Perel P, Prieto-Merino D, Shakur H, Coats T, Hunt BJ, Lecky F, Brohi K, Willett K; on behalf of the CRASH-2 collaborators. Effect of tranexamic acid on mortality in patients with traumatic bleeding: prespecified analysis of data from randomised controlled trial. BMJ. 2012; 345: e5839. doi: 10.1136/bmj.e5839.
2. Stief TW. Modulation of granulocyte-mediated thrombolysis. Hemostasis Laboratory 2008; 1: 77-102.
3. Stief TW. Tranexamic acid triggers thrombin generation. Hemostasis Laboratory 2009; 2: 73-82.
4. Stief TW. Drug - induced thrombin generation: the breakthrough. Hemostasis Laboratory 2010; 3: 3-6.
5. Stief TW. Thrombin generation by therapeutic fibrinogen. Hemostasis Laboratory 2011; 4: 467-482.
6. Stief TW. Thrombin – applied clinical biochemistry of the main factor of coagulation. In: Thrombin: function and pathophysiology. Stief T, ed.; Nova science publishers; New York; 2012; pp. vii-xx.
7. Stief TW. Hydroxy-ethyl-starch (HES) triggers plasmatic thrombin generation. Hemostasis Laboratory 2008; 1: 179-184.
8. Stief TW, Klingmüller V. The ultrasound frequency determines the degree of intrinsic coagulation activation. Blood Coagulation and Fibrinolysis 2012; 23: 440-4.

Competing interests: None declared