In Reply

We thank Prof. Tanaka and colleagues for their interest in our article¹  and for their thoughts on the interpretation of our findings. We believe that overall they are in agreement with the main thrust of our article, which is that activation of protein C (aPC) is central to the pathogenesis of acute traumatic coagulopathy (ATC). Most of their points address matters of interpretation regarding the relative extent that aPC affects different parts of the coagulation system.

First, they highlight our use of D-dimers rather than a specific fibrin degradation marker and state that the high level of D-dimer must simply be a reflection of high thrombin generation. Our extensive reading of the literature and our understanding of fibrinogen and fibrin cleavage is that these processes are not linear and that no specific test differentiates between the different ways in which they are broken down. Indeed, they themselves say that D-dimers represent ongoing fibrin degradation. In either case, we have shown that products of fibrin/fibrinogen breakdown are dramatically higher in patients with ATC and that blocking aPC (with thrombomodulin knock-in mice) essentially abolishes this response (despite equivalent markers of thrombin generation).

Second and central to their concerns is our finding that, although factor V levels are affected by aPC, this does not appear to have a large effect on overall functional measures of anticoagulant activity. Conversely, they state that there is a general reduction in all procoagulant factors and therefore the factor V effect is of minimal relevance. In our patient and mouse data there are clear differential reductions in factor V levels in patients with ATC to a greater extent than the procoagulant proteases. It is true that our endogenous thrombin potential assay may have missed subtle effects of factor V inhibition, and the correspondents are right to highlight that lower levels of tissue factor may reveal these effects. We absolutely did not say that there was no effect of aPC on the anticoagulant pathways. However, from patient functional assays and from our mechanistic data, these aspects of aPC activation seem to have less of an impact on overall hemostatic function than fibrinolysis and possibly fibrinogenolysis.

Finally, the authors discuss our findings in the context of our previous work, which included elevations of thrombin activatable fibrinolysis inhibitor (TAFI) activity levels. Much of this part of their argument is conjecture, and we believe that they are missing the differential effects of TAFI within a formed clot and that found in the systemic circulation. All of this requires additional study, because our previous data were very preliminary, and we have not explored TAFI activity in this mechanistic model. Throughout the correspondence we believe that their interpretation of our findings is based primarily on in vitro hemostasis biology being applied to the circulating plasma samples obtained from our trauma patients who will have multiple derangements in blood and endothelial function. Although we fully agree that high thrombin generation is required to generate aPC, we believe that our mechanistic data fully support a central role for this dysregulated aPC generation in the overactivation of systemic fibrinolysis and fibrinogenolysis rather than this being a purely consumptive phenomenon. We fully agree that this mechanism may drive other clinical coagulopathies, and additional studies in these areas are warranted.