Severe trauma with tissue damage and shock can rapidly (<30 min) result in abnormal coagulation function which is independent of consumption or dilution effects; this acute traumatic coagulopathy (ATC) has been linked to increased transfusion requirements, morbidity and mortality. It has been suggested that ATC is caused by an increase in activated protein C (aPC; from 40 pM in normal to 175 pM) which has been identified in trauma patients [Cohen MJ, Ann Surg 255:379 (2012)].

aPC inactivates factor Va (FVa) as part of normal hemostasis, but our previous results indicate that aPC at trauma levels is not sufficient to prevent coagulation, particularly when platelets are present, as platelet fVa is resistant to aPC [Campbell JE, PLoS ONE 9:e99181 (2014); Camire RM, Blood 91:2818 (1998)].

Tissue Factor Pathway Inhibitor (TFPI) is an endothelial-bound protein which inhibits coagulation through effects on factor Xa and the factor VIIa-tissue factor complex. It is released in the plasma (normal level 2.5nM, pathophysiologic states including trauma up to 10nM) and is typically cleared by the liver or kidneys. TFPIα accounts for 20% of circulating TFPI and binds FV and FVa [Ndonwi M, J Thromb Haemost 10:1944 (2012)]. Recently, a truncated FV (termed FV-short) produced by an autosomal dominant mutation was found to form complexes with TFPIα, resulting in retention of TFPIα in a 10-fold increase over normal levels in affected individuals [Vincent LM, J Clin Invest 123:3777 (2013)]. FV-short has significantly reduced thrombogenic potential and, by concomitantly raising TFPIα, causes a bleeding disorder.

We hypothesize that the activation of PC in acute trauma may result in the production of FV-degradation products which could stabilize TFPIα similarly to the effect of FV-short. The combination of reduced FVa and increased TFPIα may contribute to ATC.


Whole blood from healthy volunteers was drawn by phlebotomy into ACD-containing tubes. Platelet-rich plasma (PRP) and platelet-poor plasma (PPP) were collected by centrifugation (200g for 10 min and 1000g for 15 min, respectively). Calibrated automated thrombogram (CAT) assays and thromboelastography (TEG) were conducted using PRP and PPP with exogenously delivered aPC (0, 100pM, or 1nM) and TFPI (0, 2.5nM, or 10nM). An immunodepleted FV-deficient (<1% normal) plasma (FVdp) was used as a reference.


CAT assays verified that aPC and TFPI each delay and suppress thrombin generation in PPP in a dose-dependent manner, but the combination of aPC and TFPI had a synergistic effect at the highest doses tested. The endogenous thrombin potential (ETP) was eliminated in PPP (control: 1663 nM-min; 1nM aPC + 10nM TFPI: 0 nM-min; P<0.001); lag time was similarly affected (control: 2.5 min; 1nM aPC + 10nM TFPI: >60 min; P<0.001). FVdp also featured a delayed lag time (11.17 min, P<0.001 versus control), but the delay induced by the addition of 10nM TFPI was not nearly as severe in FVdp (19.67 min) as in the PPP + 10nM aPC sample (>60 min, P<0.001). For PRP, there was no statistical difference between control and the highest doses of aPC and TFPI in ETP (1636 nM-min versus 1388 nM-min) or lag time (8.72 min versus 9.13 min).


In vitro studies suggest that PC activation is not the sole cause of ATC; the concentrations of aPC measured in trauma patient blood have little effect on the coagulation potential of plasma with or without platelets when aPC is delivered exogenously. The results here demonstrate that there is a synergistic effect between aPC and TFPI. When TFPI is added to FV deficient plasma, the thrombin generation is delayed and suppressed; however, in normal plasma digested with aPC (containing fragmented FVa), the addition of TFPI is sufficient to suppress the generation of thrombin beyond the window of measurement (>60 min). The addition of platelets to the milieu eliminated the effects of aPC, TFPI, and the combination on thrombin generation, highlighting the central role of platelets in maintaining hemostatic function.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.