Patients with sepsis commonly require invasive procedures and frequently have an associated coagulopathy.1 In a recent observational survey conducted in Japan, among 1,895 patients with sepsis treated in intensive care units, 29% were diagnosed with sepsis-induced coagulopathy, a term that is synonymous with disseminated intravascular coagulation (DIC) as defined by laboratory criteria.2 In patients with sepsis, the imbalance in clot generation (coagulation) and clot breakdown (fibrinolysis) is a pivotal response that occurs due to host defense mechanisms but is associated with the development of organ dysfunction.3,4 In a prior Anesthesiology editorial, Gropper suggested that “all of these conditions (in sepsis) likely share a common pathway for the development of multiple system organ failure: diffuse activation of endothelium by proinflammatory cytokines, leukocytes, and other proteins. Activated endothelium becomes prothrombotic in these conditions, leading to the formation of microvascular thrombosis. In addition, fibrinolysis is inhibited, resulting in the buildup of fibrin thrombus, which itself is proinflammatory.”5 This description is also consistent with our understanding of the coagulopathy that occurs in sepsis, which is more commonly described as DIC. Although sepsis is a common cause of DIC, other pathophysiologic states including trauma, cardiogenic shock, or acute ischemic injury can also cause DIC, a pathologic diagnosis and clinical sequelae due to another underlying disease process. Inflammatory responses after surgery, in particular after cardiopulmonary bypass, can produce a systemic inflammatory response syndrome and coagulopathy.
Pathophysiology of Sepsis-induced Coagulopathy and DIC
The major pathways that lead to sepsis-induced coagulopathy and DIC include activation of coagulation, platelets, and other inflammatory cells (e.g., neutrophils, lymphocytes) and vascular endothelial injury.6 Classically, tissue factor, a critical component of the extrinsic coagulation pathway, expressed on macrophages and monocytes, was thought to play a central role coagulation initiation.7 Tissue factor is expressed on extracellular vesicles, and phosphatidylserine residue, an initiator of the contact pathway, expressed on the cell membranes is known to initiate hemostasis and clot formation.8 In sepsis, neutrophils are activated, releasing extracellular traps for limiting infection that consist of DNA, histones (the binding protein for DNA structural integrity), and other neutrophil granule proteins, all of which are highly prothrombotic as shown in figure 1.9 In addition, the release of damage-associated molecular patterns from injured host cells further increase prothrombotic activity. Upon cellular damage and hematopoietic cell activation, cell-free DNA and nuclear proteins are released into the circulation, such as histones, which possess strong procoagulant activity as shown in figure 1.10 To balance this prothrombotic effect, circulating plasma proteins including antithrombin and protein C/S provide important anticoagulant effects.11 The antithrombotic contribution of the vascular endothelium is also important in sepsis but under physiologic conditions to prevent clot formation, maintain vascular integrity, and regulate vascular tone. Vascular endothelial cells release nitric oxide and prostacyclin, thereby contributing to physiologic antithrombotic effects, whereas it promotes prothrombotic effects under septic conditions by expressing tissue factor and releasing von Willebrand factor.
Endothelial dysfunction and derangement of anticoagulation are the hallmarks of sepsis-induced coagulopathy. The vascular endothelial surface is lined by membrane-binding proteoglycans and glycosaminoglycan side chains that provide critical antithrombotic effects by binding to antithrombin.12 Endothelial cells also balance fibrinolysis by producing tissue-type plasminogen activator and plasminogen activator inhibitor-1, and this balance shifts to inhibit fibrinolysis during sepsis.13 Derangement of the coagulation/fibrinolytic system is another hallmark of sepsis-induced coagulopathy.
Definitions and Diagnoses of Sepsis-induced Coagulopathy and DIC
DIC was first described as a unique condition in which thrombosis and bleeding coexist. In patients with septic shock, coagulation is activated, and consumption of clotting factors occurs, which is the hallmark of DIC.14 The International Society on Thrombosis and Haemostasis scientific standardization subcommittee has defined DIC as “an acquired syndrome characterized by intravascular activation of coagulation with loss of localization arising from different causes that can originate from and cause damage to the microvasculature, which if sufficiently severe, can produce organ dysfunction” in 2001.15 The four laboratory tests used to make the diagnosis of DIC as shown in table 1 include elevated levels of fibrin-related markers such as fibrin/fibrinogen degradation products or D-dimers, decreased platelet counts, prolonged prothrombin time, and decreased fibrinogen levels.15
Additional diagnostic criteria were established to categorize an advanced decompensated stage of hemostatic function termed overt DIC to emphasize “systemic intravascular coagulation” and “consumptive coagulopathy” and defined by the scoring system as shown in table 1. If the total DIC score is 5 or more, patients are diagnosed as having overt DIC to represent an advanced, “full-blown coagulopathic state.” Other organizations have used similar scoring systems based on other combinations of coagulation biomarkers; however, because DIC is a conceptual laboratory-based clinical entity, the relevance of these criteria is debatable. The International Society on Thrombosis and Haemostasis overt-DIC criteria have been adopted as the gold standard for the diagnosis of DIC worldwide.16 Overt-DIC scoring should ideally be used to identify patients already at an advanced and possibly irreversible stage of coagulopathy, thereby late from a therapeutic perspective. Thus, identifying patients earlier in the course of their hemostatic dysfunction would be important for potential management, especially in the intensive care unit.17
Coagulopathy in patients with sepsis continues from an initial compensatory phase to the decompensated stage of DIC. Although there is still debate regarding the definition of coagulopathy, it is generally suggested as “a condition in which the ability of the blood to clot is impaired,”18 and conventionally, the term coagulopathy predominantly had represented a hemorrhagic state. However, the term coagulopathy is now used to represent a disordered condition that includes both the hemorrhagic and thrombotic states.19 In regard to managing sepsis-induced coagulopathy, recent studies have suggested that any delay in intervention can be detrimental,20 and the early detection of a coagulation disorder using criteria has currently been recommended by the International Society on Thrombosis and Haemostasis DIC subcommittee.21 Criteria for diagnosing sepsis-induced coagulopathy include thrombocytopenia, prolonged prothrombin time and organ dysfunction as assessed by the Sequential Organ Failure Assessment score (table 1). A comparison of the scoring system for sepsis-induced coagulopathy with that of overt DIC revealed that the sensitivity was approximately twice as much as that of overt DIC, and patients met sepsis-induced coagulopathy criteria earlier than overt International Society on Thrombosis and Haemostasis criteria.22 Another validation study reported that the diagnosis of sepsis-induced coagulopathy may be valuable for identifying appropriate therapeutic approaches for anticoagulant therapy in sepsis, similar to the once-used therapeutic approach of activated protein C.23
The development of thrombocytopenia is often the clue to recognize coagulopathy in sepsis-induced coagulopathy. However, different causes of thrombocytopenia should be considered in critically ill patients because specific management differs depending upon the underlying cause. Important prothrombotic states that may occur in a perioperative presentation include thrombotic microangiopathy and heparin-induced thrombocytopenia.
In patients with thrombotic microangiopathy, extensive thrombus forms within the microcirculation that causes thrombocytopenia and hemolytic anemia and includes thrombotic thrombocytopenic purpura and hemolytic uremic syndrome. There are two types of thrombotic thrombocytopenic purpura: congenital and acquired. Acquired thrombotic thrombocytopenic purpura is an autoimmune disease caused by depletion or inhibition of a cleaving enzyme called ADAMTS13 (a disintegrin and metalloproteinase with a thrombospondin type 1 motif, member 13) that cleaves von Willebrand factor into multimeric units to prevent hypercoagulability. ADAMTS13 depletion is often triggered by infection24 and requires daily therapeutic plasma exchanges to reduce morbidity and mortality.25 Two types of hemolytic uremic syndromes are recognized. The Shiga toxin–producing Escherichia coli–induced type occurs in children, is common in developing countries, and is characterized by severe hemorrhagic colitis, microangiopathic hemolytic anemia, and renal failure with a high mortality.26 The less common form is the atypical hemolytic uremic syndrome, which presents similarly but is caused by complement activation, and therapy is complement inhibition with eculizumab.27 Although the clinical presentations are similar, the differential diagnosis as shown in figure 2 may be helpful.28
One important thrombotic microangiopathy that perioperative clinicians need to be aware of is heparin-induced thrombocytopenia, an autoimmune disease caused by platelet-activating antibodies that recognize the multimolecular complexes of platelet factor 4 bound to heparin. This interesting disease state is when a commonly used anticoagulant (heparin) produces a prothrombotic effect caused by platelet activation. The clinical probability of heparin-induced thrombocytopenia can be evaluated using the clinical “4 Ts” scoring system (thrombocytopenia, timing of onset, thrombosis, and other causes of thrombocytopenia).29 Diagnosis of heparin-induced thrombocytopenia is supported by a positive platelet factor 4–dependent immunoassay.30,31 In clinical practice, heparin-induced thrombocytopenia can be easily confused with sepsis-associated coagulopathy, especially when it occurs in association with organ dysfunction such as shock liver, adrenal hemorrhage, and skin/limb necrosis.32 Therapy includes stopping heparin and initiating a non-heparin anticoagulant therapy when appropriate, usually a direct thrombin inhibitor: either argatroban or bivalirudin. Other conditions also mimic DIC, and practical guidelines for the diagnosis and management of these conditions have been recently reported.33
Biomarkers and Other Tests
DIC is diagnosed based on laboratory testing that can be performed in most local hospitals as routine tests, and so-called global hemostatic tests have been adopted as diagnostic criteria, although additional biomarkers are needed.34 Other molecular markers, such as the thrombin–antithrombin complex, soluble fibrin, and prothrombin fragment 1.2, have been adopted in a proposed Japanese Society on Thrombosis and Hemostasis criteria.35 One of the unique features of sepsis-associated coagulopathy is the suppression of fibrinolysis caused by increased plasminogen activator inhibitor-1 production, and its circulating levels have been reported to be useful markers of the severity in sepsis.36 Suppression of fibrinolysis could also be evaluated by measuring the thrombin–antithrombin/plasmin–plasmin inhibitor complex ratio, and this ratio has been reported to be significantly higher in patients with severe sepsis.37
One of the hallmarks of sepsis-induced coagulopathy is consumption of physiologic anticoagulants such as antithrombin and protein C, circulating anticoagulants that are thought to be important in the pathogenesis, and plasma levels have been reported to correlate with the disease severity.38 The mechanisms underlying the decreases in the levels of these anticoagulants are multifactorial and include increased consumption, decreased production, and excess extravasation caused by increased vascular permeability.
Monitoring of the dynamic changes of coagulation and fibrinolysis provides information to determine the stage of the coagulation disorder, and viscoelastic testing is frequently used as a point-of-care device.39 Thromboelastography and rotational thromboelastometry are used in managing patients with trauma-associated coagulopathy to detect fibrinolysis. In sepsis, early inhibition of plasmin activity leads to acute shutdown of fibrinolysis, and its association with increased morbidity and mortality in sepsis patients has been reported.40 However, the usefulness of these tests remains to be established.41 Rotational thromboelastometry correlates well with DIC scoring, and the diagnostic ability increases when combined with other biomarkers that conclude protein C levels or antithrombin.42
In patients with sepsis, components of the glycocalyx are shed from the vascular endothelium into the circulation. Measuring their levels in plasma has been proposed as a method for measuring endothelial damage.43 However, the reliability of this test has not been established yet, and further work is needed.
Extracellular vesicles are other potential novel biomarkers, and leukocyte-derived extracellular vesicle count has been reported as an early and relevant biomarker of sepsis-associated coagulopathy.44 However, the clinical relevance of such tests is hampered by the lack of standardized assay procedures.
Other important coagulation biomarkers in sepsis, including circulating damage–associated molecular patterns such as cell-free DNA and histones, have been demonstrated to influence hemostasis by promoting coagulation activation, platelet aggregation, and inhibition of fibrinolysis.10 However, their usefulness remains to be examined.
Critical to the management of sepsis-induced coagulopathy is the rapid and timely treatment of the underlying infection. Other strategies that have been extensively studied are directed at suppressing the prothrombotic effects. Heparin and heparinoids are the most popularly used anticoagulants for a variety of thromboembolic diseases; however, their effectiveness for treating sepsis-induced coagulopathy or DIC still are debated and generally limited to the prevention of deep vein thrombosis.
There is no globally approved therapeutic agent at present for the specific treatment of sepsis-induced coagulopathy. The effectiveness of natural anticoagulants, such as antithrombin, activated protein C, and tissue factor pathway inhibitors, have been studied in large-scale controlled trials in the early 21st century. Unfortunately, the primary goal could not be achieved in any of these trials, although it must be remembered that all of these trials targeted sepsis and not sepsis-associated coagulopathy and DIC. In contrast, post hoc analyses performed to examine the effects in the DIC patient subgroup showed favorable effects of antithrombin and recombinant activated protein C.45,46 Although recombinant activated protein C has been withdrawn from the world market, the efficacy of antithrombin for septic DIC has been extensively studied in Japan, and nationwide-database studies have repeatedly demonstrated a potential survival benefit of supplemental doses of antithrombin supplementation.47 In addition, a meta-analysis demonstrated a trend toward better survival associated with antithrombin use in septic patients with DIC.48 However, the use of antithrombin has not been recommended in the international sepsis guidelines.49 The important lesson to learn from these studies is that anticoagulant therapy may be expected to be beneficial for sepsis patients who developed coagulopathy but not in those without a coagulation disorder.
Although the role of fresh frozen plasma as a therapeutic approach in patients with sepsis-induced DIC may have potential theoretical benefits, there are no data supporting its application unless there is a specific therapy to treat including bleeding or factor depletion beyond antithrombin. Based on this consideration, the effect of recombinant thrombomodulin was recently examined in a multinational, randomized controlled phase III trial in sepsis patients with coagulopathy (with a platelet count less than 150 × 109/l and a prothrombin time ratio of more than 1.4). The 28-day mortality improved by 2.6% in a total of 800 sepsis patients, although this result was not found to reach statistical significance. It was also reported that more than 20% of the patients recovered even before the initiation of the treatment, and a subgroup analysis performed in the patients who fulfilled the entry criteria at baseline (approximately 600 cases) revealed a reduction of the 28-day all-cause mortality by 5.4%.50 In summary, the effectiveness of natural anticoagulants has not yet been proven, and continued research is warranted.
Finally, we admit that there are some unanswered questions in this review. First, coagulopathy in sepsis is a dynamic phenomenon, with heterogeneous presentations, and we still do not have an ideal tool to assess this dynamic change in sepsis. Second, we do not have an established therapy worldwide for sepsis-induced coagulopathy at present. The above important issues should be addressed in the future.
Coagulopathy is an important and common complication in patients with sepsis and contributes to the development of organ dysfunction. Sepsis is a common cause of vascular injury and thrombocytopenia and can progress to DIC, which is synonymous with sepsis-induced coagulopathy. DIC is a laboratory-based diagnosis representing the decompensated status of coagulopathy that can also arise from other causes, including traumatic injury, cardiogenic shock, and multiorgan injury. Diagnosing coagulopathy in sepsis is often not established early or correctly, and there are different classifications for DIC. Recently, a two-step approach for the rapid diagnosis of DIC has been reported (table 1).21 Other conditions in critically ill patients such as thrombotic thrombocytopenic purpura and hemolytic uremic syndrome can also occur in patients with sepsis, conditions that should be differentiated for therapy. Although the diagnosis of sepsis-induced coagulopathy is made based on the combination of hemostatic biomarkers, new candidates that would allow simpler diagnosis could emerge in the future. Continued efforts to develop new therapeutics to treat these complicated disorders and a more target-specific approach are needed. Until new therapies are developed, a multimodal approach is needed that includes timely antibiotic therapy of the underlying infection, cardiopulmonary resuscitation, and appropriate management of the underlying coagulopathy if bleeding or thrombotic sequela have occurred.
Toshiaki Iba, M.D., is the Chairperson of the International Society on Thrombosis and Haemostasis (Carrboro, North Carolina) disseminated intravascular coagulation subcommittee.
Supported by a Grant-in-Aid from the Ministry of Health, Labour and Welfare of Japan (Tokyo, Japan) and from the Ministry of Education, Culture, Sports, Science and Technology of Japan (Tokyo, Japan).
Dr. Levy serves on research steering committee data safety monitoring boards or advisory boards for CSL Behring (King of Prussia, Pennsylvania), Instrumentation Laboratories (Bedford, Massachusetts), Octapharma (Lachen, Switzerland), Leading Biosciences (Carlsbad, California), and Merck (Kenilworth, New Jersey). The other author declares no competing interests.