Activated protein C plays no major roles in the inhibition of coagulation or increased fibrinolysis in acute coagulopathy of trauma-shock: a systematic review

Coagulation and fibrinolysis are innate immune responses; they develop nonspecifically start following infectious and noninfectious insults, such as sepsis and trauma [1]. These responses occur at the site of the insult to limit host damage and they have roles in the compartmentalization of danger- and pathogen-associated molecular patterns, which suppress dissemination of these patterns into the circulation [1‐3]. These processes are referred to as hemostasis and wound healing in trauma. During normal hemostasis, anticoagulant pathways, tissue factor pathway inhibitor (TFPI), antithrombin, protein C, and thrombomodulin restrict clot dissemination at intact sites. Protein C and thrombomodulin have important roles in controlling coagulation, fibrinolysis and inflammation via activated protein C, thrombin-activatable fibrinolysis inhibitor (TAFI) and endothelial protein C receptor (EPCR) [4, 5]. These physiological systems cannot work in severely injured trauma patients, which is associated with endothelial injury; thus, sever trauma leads to systemic thrombin generation, a condition that is called disseminated intravascular coagulation (DIC) [6].

In 2007, a condition called “acute coagulopathy of trauma-shock” (ACOTS) was newly proposed [7‐9]. Another paper [10] and a review approved this hypothesis [11], and thereafter ACOTS was established as major pathophysiological concept in trauma-induced coagulopathy [12]. The pathogenesis of ACOTS is summarized as follows, based on several review articles [7, 11, 13‐17]. ACOTS, in which activated protein C plays central roles, only occurs at the very early period after injury in only patients with shock and severe acidosis. Traumatic shock slows the clearance of thrombin, increasing its binding to newly expressed thrombomodulin on the normal endothelium and soluble thrombomodulin with full domains and a 100% activity in the circulation. Thrombin and thrombomodulin complexes induce the systemic production of activated protein C, which inactivates activated Factors V (FVa) and VIII (FVIIIa) and neutralizes plasminogen activator inhibitor-1 (PAI-1). This is followed by suppression of thrombin generation and an increase in the production of tissue-type plasminogen activator (t-PA). In addition, the fibrinogen levels always remain normal, indicating that less thrombin is available to cleave fibrinogen.

However, a number of researchers have expressed doubts in this theory. One group noted that the immediate massive generation of thrombin and it activation in the systemic circulation, insufficient anticoagulant pathways associated with endothelial injury and suggested that this is the main pathophysiological mechanisms of trauma-induced coagulopathy, which coincides with the definition of DIC reported by the International Society on Thrombosis and Haemostasis (ISTH) [18‐23]. In addition to secondary fibrinolysis due to DIC, the accelerated release of t-PA from the injured endothelium due to traumatic shock-induced hypoperfusion, the consumption of α2-antiplasmins, and increased levels of neutrophil elastase synergistically mediate systemic fibrin (nogen) olysis. The time delay between the immediate release of t-PA from the endothelium and the expression of PAI-1 mRNA enhances fibrin (ogen) olysis immediately after injury. In addition, the massive generation of tissue factors cause fibrin (ogen) olysis. These conditions, the coexistence of DIC and pathological systemic fibrin (ogen) olysis are referred to as DIC with the fibrinolytic phenotype [24].

Another group also expressed doubts regarding the theory that ACOTS is driven by activated protein C [25‐28]. They acknowledged that the currently available evidence suggests that ACOTS occurs due to the massive stimulation of thrombin generation, platelet and fibrinogen consumption, and fibrinolysis by damaged tissues, and they suggested that these data indicate a consumptive coagulopathy. Furthermore, the levels of activated protein C observed in ACOTS are far from the concentration that can cleave both platelet and plasma FVa. In plasma, PAI-1 is completely bound to vitronectin. It is therefore unlikely that activated protein C inactivates the PAI-1/vitronectin complex and leads to PAI-1 depletion. Instead, an enormous increase in the release of t-PA by endothelial cells due to shock-induced hypoperfusion is the likely cause of increased fibrinolysis. These mechanisms are very similar to DIC with the fibrinolytic phenotype; however, the lack of evidence of intravascular clot formation rules out this hypothesis according to the theory proposed by this group.

This systematic review was performed in accordance with the protocol of the Preferred Reporting Items for Systematic Review and Meta-Analysis Protocol (PRISMA-P) 2015 [30, 31]. The flow of information through the different phases of the systematic review was constructed based on the PRISMA Statement 2009 [32]. No systematic reviews have been conducted on this theme; thus, this is not an amendment of a previously completed protocol. At the point of completing the data extraction, this study did not meet the inclusion criteria of the International Prospective Register of Systematic Review (PROSPERO); thus, this systematic review was not registered.

The classifications shown in Table 1 were used in order to clarify the complicated terminology of trauma-induced coagulopathy in this systematic review [22]. Although other terms such as “acute coagulopathy of trauma” or “acute traumatic coagulopathy” etc. are now used to describe ACOTS, the original term, ACOTS, is mostly used in this systematic review. The other terms are used, as appropriate, when discussing the studies in which they are used.

Activated protein C has both anticoagulant and profibrinolytic effects through the proteolytic inactivation of FVa and FVIIIa and through the activation of TAFI and the neutralization of PAI-1 [4, 5, 49]. Although these activities are important in normal physiological hemostasis at the injured site, the suggestion that the same properties are systemically active in pathological conditions such ACOTS is in doubt. Campbell et al. [50] suggested that the levels of activated protein C in ACOTS were insufficient to inactivate platelet and plasma FVa. Another in vitro study confirmed that 300 to 2000 ng/mL of activated protein C was needed to suppress activities of FV and FVIII, and to prolong PT and APTT [51]. These levels were extremely high in comparison to those observed in clinical studies, which indirectly shows that pathomechanisms of ACOTS are not robust [36, 37, 39].

Only 2 studies simultaneously compared the activated protein C and thrombin generation levels between ACOTS (acute traumatic coagulopathy) and non-ACOTS (non-acute traumatic coagulopathy) [35, 41]. Thrombin generation (PF1 + 2 or TAT) was preserved without changes in the activated protein C levels [35] or was significantly increased with high activated protein C levels being observed [41]. One study confirmed normal prothrombinase activity with markedly elevated thrombin generation [42]. Importantly, in studies that measured activated protein C levels, the levels were reported to be inconsistently normal, decreased, or elevated [35‐42]. Although the activated protein C levels were not measured, the clinical studies consistently showed significant systemic thrombin generation in ACOTS and severely injured trauma patients [43‐46]. These results suggest that - irrespective of activated protein C levels - systemic thrombin generation is increased immediately after trauma, especially in ACOTS and in patients with severe trauma.

Two studies simultaneously compared the activated protein C and PAI-1 levels between ACOTS (acute traumatic coagulopathy) and non-ACOTS (non-acute traumatic coagulopathy) [35, 41]. There were no differences in the levels of PAI-1 between patients with and without ACOTS or acute traumatic coagulopathy, irrespective of the activated protein C levels. Another study confirmed the same results in ACOTS with severe head and neck, and extracranial injuries [38]. Although the activated protein C levels was not measured, severe fibrinolysis, evidenced by increased levels of plasmin and α2-antiplasmin complex and t-PA, was not associated with decreased PAI-1 levels [46]. One study clearly demonstrated that the decrease in the PAI-1 levels in patient with trauma-induced coagulopathy was driven by an increase in t-PA, rather than by the degradation of PAI-1 by activated protein C [48]. These results suggest that activated protein C is highly unlikely to neutralize PAI-1.

A PT ratio of > 1.2 was announced as a clinically relevant definition of acute traumatic coagulopathy [52]. However, only two studies used this as diagnostic criteria for ACOTS [40, 42]. As shown in Tables 2, 3, 4, 5, 6 and 7, various criteria were used to diagnose coagulopathy in the studies that were selected for this systematic review; furthermore, various names were used to refer to the conditions. This makes the collection of uniform patients group very difficult. As a consequence, the evidence to support a uniform concept of ACOTS is very fragile.

The main pathological definition of ACOTS is the activated protein C-mediated inactivation of FVa and FVIIIa, which stops the generation of thrombin. Thus, the APTT that represents coagulation pathway including FV and FVIII, but not PT, is considered to be more suitable for the diagnosis of ACOTS [20]. The lack of a clear pathological definition and diagnostic criteria based on activated protein C dynamics leads to these authors to be skeptical of the existence of the condition referred to as ACOTS.

This systematic review failed to establish hypotheses or complete the aims stated in the Objective section.

This systematic review demonstrated that there have been no studies that showed the direct cause and effect relationships between activated protein C and the suppression of systemic thrombin generation or between activated protein C and increased fibrinolysis via the neutralization of PAI-1. Systemic thrombin generation was always increased and increased fibrinolysis was observed independently from the PAI-1 levels in patients with ACOTS. The lack of a clear definition and diagnostic criteria, and various terminologies that are used to refer to ACOTS may be one of the reasons for these results. This systematic review denies pathophysiological definition of ACOTS that was proposed in 2007 and concludes that the condition that is referred to as “ACOTS” based on activated C dynamics unlikely to exist.