Inflammation and Coagulation: Implications for the Septic Patient

REVIEW ARTICLE

Inﬂammation and Coagulation: Implications for the Septic Patient

R. Phillip Dellinger Downloaded from https://academic.oup.com/cid/article/36/10/1259/307497 by guest on 30 September 2021 Critical Care Section, Cooper Health System, Camden, New Jersey

Sepsis with acute organ dysfunction (severe sepsis) is common, frequently fatal, and associated with a significant national health/economic burden. In addition to standard care, investigators have focused on interrupting the inflammatory and anti-inflammatory cascade associated with this disease. Unfortunately, despite promising preclinical results, interventions directed at the inflammatory elements have not reduced the morbidity and mortality associated with this disease. Inflammation and coagulation are tightly linked. In fact, sepsis-associated coagulopathy is almost universal in patients with severe sepsis. Preclinical observations indicate that antithrombotic-targeted therapy has the potential to reduce morbidity and mortality in patients with this disease. Treatment with recombinant human activated protein C (drotrecogin a [activated]) was the first antithrombotic-targeted therapy to significantly reduce 28-day all-cause mortality in patients with severe sepsis. The pathophysiological and clinical significance of this evidence and the relationship of coagulation to inflammation are discussed, as are positive and negative results of clinical trials of antithrombotic therapy.

Once the proinflammatory cascade was identified as an proinflammatory cytokine, IL-6. In addition, investi- important cause of morbidity and mortality in patients gators appreciated that targeting inflammatory com- with sepsis, investigators targeted the cytokine cascade ponents might actually be harmful [2]. for innovative antisepsis interventions. Focus on the More recently, investigators recognized the intimate inflammatory component of the host response has con- relationship between procoagulant and proinflamma- tinued despite identification of a significant compen- tory activity and hypothesized that antithrombotic satory anti-inflammatory response syndrome (CARS) strategies might also offer an approach to reduce sepsis- that can also increase mortality [1]. Attempts were associated mortality [3–6]. Because this relationship is made to integrate this new information and to char- integral to the pathophysiology of sepsis, it has led to acterize patients who would be more likely to benefit preclinical and clinical studies that target both procoagulant and proinflammatory portions of the host re- from anti-inflammatory therapy. Therefore, investiga- sponse to infection [7]. Here, I describe the inflammatory tors developed strategies to identify patients with severe and counterinflammatory elements of the host’s response sepsis and septic shock who were still in a proinflam- to sepsis and discusses unsuccessful attempts to apply matory state. These included enrolling patients soon anti-inflammatory interventions to patients with severe after the inflammatory process was identified and tar- sepsis. I also discuss the tight linkage between inflam- geting patients with increased circulating levels of the mation and the host’s hemostatic response. Finally, I describe the results of clinical trials of antithrombotic interventions. Received 18 July 2002; accepted 28 January 2003; electronically published 8 May 2003. Reprints or correspondence: Dr. R. Phillip Dellinger, Critical Care Section, Div. of Cardiovascular Diseases and Critical Care Medicine, Cooper Health System, 1 INFLAMMATION AND SEPSIS Cooper Plaza, Camden, NJ 08103 ([email protected]). Clinical Infectious Diseases 2003;36:1259–65 2003 by the Infectious Diseases Society of America. All rights reserved. As has been demonstrated in animal models of sepsis, 1058-4838/2003/3610-0008$15.00 inflammation, tissue injury, and organ dysfunction are

Inflammation and Coagulation • CID 2003:36 (15 May) • 1259 driven by a cascade of proinflammatory cytokines, such as TNF- CLINICAL TRIALS OF ANTI-INFLAMMATORY a and IL-1 [8–10]. In patients with severe sepsis, host response THERAPY IN PATIENTS WITH SEVERE SEPSIS can be characterized as an excessive proinflammatory reaction. Strategies have been tested to counteract the runaway pro- Resulting high levels of circulating cytokines can potentiate inflammatory state seen in sepsis. These strategies targeted en- organ damage by endothelial injury and a multitude of other dotoxin and mediators, such as TNF-a, IL-1, PAF, and bra- routes. Endothelial damage is associated with activation of neu- dykinin. Other investigational strategies have focused on trophils and expression of neutrophil and endothelial adhesion corticosteroid administration, prostaglandin inhibition, nitric molecules (e.g., E- and P-selectins and intercellular adhesion oxide inhibition, and immunoglobulin therapy. Hemofiltration molecule–1). These molecules help localize WBCs to the area studies have also evaluated the efficacy of cytokine removal as of injury; however, before leaving the vessel, inflammatory mol- a therapeutic intervention in severe sepsis. ecules released by activated neutrophils can produce additional endothelial injury. Release of endothelial cell molecules, such Downloaded from https://academic.oup.com/cid/article/36/10/1259/307497 by guest on 30 September 2021 as IL-8 and platelet-activating factor (PAF), attract and activate ENDOTOXIN INHIBITION other cells, amplifying the inflammatory response. This cycle Endotoxin interacts with cells through specific membrane or results in focal and ongoing microvascular injury. Other in- soluble receptors. Binding of endotoxin to receptors, such as flammatory factors (e.g., eicosanoids) activate circulating neu- membrane CD14 or Toll-like receptors, on monocytes and mac- trophils; complement receptors on these activated inflamma- rophages results in the production of numerous proinflam- tory cells increase neutrophil adhesion to the endothelium. matory cytokines (e.g., TNF-a, IL-1, and IL-6) and anti- These observations provide a strong theoretical basis for the inflammatory cytokines (e.g., IL-10 and IL-12). Furthermore, role of anti-inflammatory strategies in patients with severe sep- the interaction upregulates expression of cell adhesion mole- sis and septic shock. cules, inducible nitric oxide synthase, phospholipase A2, and inducible cyclooxygenase. Therefore, there is a strong theoretical basis for administration of endotoxin antagonists to patients with CARS IN SEPSIS severe sepsis of gram-negative bacterial origin. Antiendotoxin strategies have included murine monoclonal antibody (E-5) and Interventions for sepsis that focus on the inflammatory com- mouse-human hybrid monoclonal antibody (HA-1A) against en- ponent of the host response ignore the body’s CARS [1]. The dotoxin. Angus et al. [12] found that, despite adequate sample CARS includes mediator-neutralizing molecules, such as sol- size and patient enrollment, treatment with E-5 did not improve uble cytokine receptor and cytokine receptor antagonist. These short-term survival. Similarly, McCloskey et al. [13] found that molecules counteract the excessive release of proinflammatory treatment with HA-1A was not effective at reducing 14-day mor- cytokines by binding to cytokines and interfering with inter- tality in patients with bacteremia and septic shock due to gram- actions between molecules and membrane-bound receptors. negative organisms. Neutralizing molecules produced by recombinant technology might be used to downregulate the effects of CARS, potentially reducing severe sepsis-associated morbidity and mortality. De- TNF-a INHIBITION creased expression of human leukocyte antigens (HLA-DR) by TNF-a is a central proinflammatory cytokine. Anti–TNF-a monocytes obtained from patients with severe sepsis and CARS strategies include monoclonal antibodies against TNF-a and indicates a state of immunosuppression [1]. Therefore, patients use of neutralizing soluble TNF receptors [2, 14–16]. Mono- whose monocytes show decreased HLA-DR expression might clonal antibody against TNF-a failed to show statistically sig- benefit from an immunostimulant. This concept was evaluated nificant benefit, as did the P-55 construct soluble TNF-a re- in severely septic [11] patients treated with filgrastim (G-CSF). ceptor (type 1). The P-75 (type 2) soluble TNF receptor Although HLA-DR studies were not performed, administration produced deleterious effects, as did nonspecific blocking of of G-CSF as an immunostimulant was not associated with dif- nitric oxide synthase [2, 17]. ferences in morbidity or mortality. Although there are also potential benefits to the suppression of the proinflammatory IL-1 INHIBITION response, some investigators believe that, if anti-inflammatory molecules are present in excessive amounts, as in CARS, the IL-1 is another proinflammatory molecule that is central to the proinflammatory state is already neutralized. Therefore, therapy host’s response to infection. The anti–IL-1 approach was tested targeting the proinflammatory state might produce deleterious with IL-1ra. A 72-h, continuous intravenous infusion failed to immunosuppression [1]. result in a statistically significant reduction in mortality [18].

1260 • CID 2003:36 (15 May) • Dellinger CORTICOSTEROIDS infectious insults is overzealous, activation of the coagulation cascade can occur. Corticosteroids downregulate inflammation through a variety In sepsis, toxins cause direct activation of coagulation via of processes. These include inhibition of inducible nitric oxide the effect of chemical mediators on the endothelium and mono- synthase and downregulation of nuclear factor–kB (NF-kB) cytes as well as indirect activation through the proinflammatory with reduced transcription of proinflammatory cytokines and cascade. Activation of coagulation by toxins occurs directly increased transcription of IL-1ra [19]. Although early trials of through upregulation of tissue factor (TF) [35]. TF then ac- corticosteroids failed to show benefit [20, 21], a recent 300- tivates factor VII of the extrinsic system, leading to thrombin patient study by Annane et al. [22] found that patients with formation and generation of fibrin clots (figure 1). Thrombin septic shock and relative adrenal insufficiency (defined as an production is amplified through the intrinsic system when fac- increase in the serum cortisol level of !10 mg/dL after a high- tor VIIa of the extrinsic system activates factor IX. Thrombin dose adrenocorticotropic hormone stimulation test) had a sig-

potentiates the prothrombic state when it activates thrombin- Downloaded from https://academic.oup.com/cid/article/36/10/1259/307497 by guest on 30 September 2021 nificantly reduced risk of death after treatment with low doses activatable fibrinolysis inhibitor (TAFI) [41, 42]. Therefore, of hydrocortisone and fludrocortisone. These results are en- thrombin not only produces clots, but also inhibits their re- couraging and, one hopes, will be confirmed in larger trials moval. In addition, activation of coagulation in sepsis can occur [23, 24]. indirectly through the activation of the contact system and release of the coagulation inducers TNF-a and interleukins; OTHER APPROACHES bacterial products, such as lipid A (gram-negative product) and

NF-kB is a key intracellular mediator that is responsible for generation of proinflammatory cytokines [25, 26]. After stimulation of various cell receptors with endotoxin, exotoxin, TNF- a, IL-1 oxygen radicals, et cetera, a cytoplasmic kinase is activated, which cleaves the NF-kB inhibitor (IkB) from NF-kB through proteolysis. NF-kB is then translocated to the nucleus and enables the translation, transcription, and production of mRNA targeted for production of a variety of proinflammatory cytokines. Last, anti-inflammatory interventions, such as hemofiltration, immunoglobulin therapy, PAF receptor antagonist, or bradykinin antagonists, have not produced significant benefits or have only been evaluated in small trials [27–30].

COAGULATION AND SEPSIS

There is no debate that inflammation is a major component of sepsis. In 1992, Bone et al. [31] characterized this inflammatory element as the systemic inflammatory response syndrome (SIRS). The authors noted 2 important points: SIRS can be seen in patients with a noninfectious cause, and future work on the inflammatory response to infection would shed light on additional cellular and immunologic mechanisms of sepsis. Since that time, additional work on inflammation has lead to the discovery of the importance of coagulation in sepsis [7, 32–35]. Events that lead to SIRS and eventually the activation of the coagulation cascade can be caused by a myriad of noninfectious insults, including thermal injury, pancreatitis, and trauma Figure 1. The coagulation cascade. The coagulation cascade is broken down into 2 parts: the intrinsic and extrinsic systems. The primary initiating [36–38]. Coagulation activation in patients without infection pathway is the extrinsic system, which is amplified by the intrinsic system. can be indirectly linked to inflammation through thrombin The primary initiator of the extrinsic cascade is tissue factor, which is only production, TNF-a release, and activation of the complement exposed after tissue injury. The outcome of the cascade is the polymerization system [39, 40]. When the proinflammatory response to non- of fibrin monomers into a clot. Adapted from [45, 62].

Inﬂammation and Coagulation • CID 2003:36 (15 May) • 1261 lipoteichoic acids (gram-positive product), can each cause their study, the mortality rate was 32% in patients without dissem- release [35, 43, 44]. inated intravascular coagulation (n p 16 ), and it was 77% in Sepsis-induced procoagulant activity is generally more severe patients with disseminated intravascular coagulation (n p 44 ). than that which can be produced by trauma. The prothrombic In addition, subclinical consumptive coagulopathy can be dem- diathesis is systemic—that is, thrombin is generated not only onstrated by increased D-dimers in almost all patients with by the endothelium, but also by circulating activated mono- severe sepsis, with a decrease in the protein C level in 180% cytes. This creates an essentially unlimited supply of TF. Gen- [55]. Therefore, a coagulopathy of sepsis can be seen in essen- eralized activation of coagulation depletes the body’s natural tially all patients with severe sepsis. The earliest signs of con- antithrombotic factors: protein C, antithrombin (AT), and TF sumptive coagulopathy in sepsis are a decrease in protein C pathway inhibitor (TFPI) [45, 46]. Activated protein C is a and an increase in D-dimer. In patients with more-severe con- serine protease that inhibits clotting factors Va and VIIIa. AT sumptive coagulopathy, prothrombin time and partial throm-

is a glycoprotein that inactivates most serine proteases and boplastin time increase, with a decrease in the fibrinogen level Downloaded from https://academic.oup.com/cid/article/36/10/1259/307497 by guest on 30 September 2021 thrombin. In addition, TFPI inhibits the activity of the VIIa/ and platelet count. Fibrinolysis is also impaired in severe sepsis. TF complex of the extrinsic system. By inactivating these fac- Thus, the net result is a proinflammatory and prothrombic tors, natural antithrombotics dampen coagulation. Generation disorder with impaired fibrinolysis. This pathophysiologic triad of thrombin also initiates fibrinolysis, primarily through the is an attractive target for antisepsis interventions. release of tissue plasminogen activator. Once plasmin is acti- Two approaches have potential benefits in patients with sep- vated, it degrades fibrin, producing fibrin degradation/split sis-associated coagulopathy: (1) administration of TFPI or a products, including D-dimers [47]. VIIa/TF inhibitor to block TF-mediated activation of the ex- The acute respiratory distress syndrome (ARDS) is a com- trinsic system, and (2) restoration of endogenous natural an- mon accompaniment of severe sepsis and illustrates the role tithrombotic mechanisms through administration of AT con- of prothrombic changes in organ injury/dysfunction [48]. In centrate or activated protein C. All of these compounds would ARDS, proinflammatory cytokines activate endothelial cells, be expected to have both antithrombotic and anti-inflamma- which leads to activation and accumulation of neutrophils in tory effects. the pulmonary vasculature. Endothelial cells damaged by inflammatory cytokines and activated neutrophils lose their abil- ity to regulate vascular permeability and become prothrombic. RECOMBINANT HUMAN TFPI This results in an increase in interstitial water, pulmonary After stimulation by bacterial products, monocytes and perhaps edema, and deposition of intravascular and intraalveolar fibrin, a subset of endothelial cells upregulate expression of TF [56, which enhances the inflammatory reaction. Bronchoalveolar 57]. Extrinsic pathway activation is tightly regulated by TFPI, lavage fluid from acute lung injury and ARDS reveal increased a potent inhibitor of TF. Other inhibitors of the VIIa/TF com- levels of soluble TF and increased expression of TF by mac- plex include antifactor VII monoclonal antibody and inacti- rophages [49, 50]. Levels of plasminogen activator inhibitor–1 vated factor VIIa (DEGR-FVIIa), a form of factor VIIa with its are also increased in bronchoalveolar lavage fluid during active center blocked, preventing interaction of factor VII with ARDS [51]. TF. In addition, low-molecular-weight inhibitors, such as the Not only is severe sepsis associated with activation of both synthetic protease inhibitor nafamostat, have also been tested the coagulation and inflammatory cascades, but these 2 systems in sepsis models [58]. also function in a positive feedback loop [52, 53]. Therefore, After successful preclinical trials, a phase 2 study of TFPI a vicious cycle of inflammation and coagulation can result, revealed promising data as to the potential for clinical outcome which has the potential to lead to progressive organ dysfunction benefit [59]. Unfortunately, results of a large phase 3 trial of and death. Therefore, the hemostatic system is an appropriate TFPI in patients with severe sepsis completed in late 2001 failed target for intervention to reduce morbidity and mortality in to demonstrate a mortality benefit for recombinant TFPI [60]. patients with severe sepsis.

RECOMBINANT HUMAN AT CLINICAL TRIALS OF ANTITHROMBOTIC THERAPY IN PATIENTS WITH SEVERE SEPSIS AT is a serine protease inhibitor of hepatic origin [61]. It inhibits multiple components of the intrinsic, extrinsic and final A new understanding of sepsis pathophysiology incorporates common coagulation pathways (IIa, IXa, Xa, XIa, XIIa, VIIa/ coagulation and fibrinolysis. Fourrier et al. [54] identified con- TF, and kallikrein) [45, 62]. In addition to its antithrombotic sumptive coagulopathy as a strong predictor of death and mul- properties, AT bound to endothelial cell surface heparans stim- tiple-organ dysfunction in patients with septic shock. In that ulates the endothelium to release prostacyclin, a molecule with

1262 • CID 2003:36 (15 May) • Dellinger anti-inflammatory properties. In animal models, AT success- was associated with receipt of drotrecogin a (activated) treat- fully blocks endotoxin-induced pulmonary leukocyte accu- ment (P p .005 ) [69]. The treatment effect was primarily in mulation and vascular permeability changes [63]. Despite these the third and fourth APACHE II quartiles, which demonstrated encouraging preclinical data, a large (n p 2314 ) phase 3 clinical an absolute risk reduction of 13.0% and a relative risk reduction trial of AT in patients with severe sepsis failed to show overall of 29.0% [70]. benefit [64]. A subgroup analysis suggested that patients who There was a non–statistically significant increase in cases of were receiving concomitant heparin had significantly increased severe bleeding in the drotrecogin a (activated) group (P p bleeding, whereas a trend in improved 28-day survival, which .06); therefore, increased risk of severe bleeding with use of this became significant at 90 days, was demonstrated in a subgroup drug should be assumed. Bleeding occurred primarily during analysis of patients not receiving heparin. This suggests, per- the drug infusion and was procedure related. Although there haps, that binding of the natural and therapeutic AT prevented is an increased risk of bleeding, the population as a whole,

the recombinant AT from exerting an anti-inflammatory effect. regardless of the relationship placed on risk of serious bleeding Downloaded from https://academic.oup.com/cid/article/36/10/1259/307497 by guest on 30 September 2021 versus benefit of reduced mortality, demonstrated a good benefit-to-risk profile associated with administration of drotrecogin RECOMBINANT HUMAN ACTIVATED a (activated) [69, 71]. PROTEIN C Phase 3 trial data indicated that 1 additional life would be saved for every 16 patients treated with drotrecogin a (acti- Protein C is converted to activated protein C by thrombin vated) and 1 life would be saved per 8 patients treated in the bound to thrombomodulin—a reaction that is facilitated by third and fourth APACHE II quartiles (Eli Lilly, data on file) the endothelial protein C receptor. In patients with severe sepsis, [70]. In fact, it is not unreasonable to compare drotrecogin a conversion of protein C to its activated form is impaired, re- (activated) with tissue plasminogen activator therapy for acute sulting from endothelial cell dysfunction/damage. Exogenous myocardial infarction [72]. Both therapies save lives. Both ther- administration of protein C is ineffective in raising activated apies are associated with the possibility of serious bleeding, protein C levels in the presence of significant endothelial dam- including intracranial hemorrhage. Both drugs are associated age [65]. Therefore, activated protein C must be available to with significant expense. With both drugs, the risk of bleeding counter the sepsis-induced prothrombic state. Activated protein and the expense are well worth the potential benefit, if proper C has also been demonstrated in vitro to have significant anti- care and deliberation are provided before deciding in favor of inflammatory properties associated with a decrease in pro- administration. inflammatory cytokines and a reduction of endothelial neutrophil rolling. Activated protein C also inactivates factors Va and VIIIa, downregulating thrombin formation and thereby indi- POSITIVE AND NEGATIVE TRIALS rectly preventing the activation of TAFI [66]. Activated protein WITH ANTITHROMBOTIC THERAPY C also inactivates plasminogen activator inhibitor–1 by forming a complex with the latter [67]. Experimentally, recombinant The explanations for the success of drotrecogin a (activated) human activated protein C also inhibits endothelial cell ap- and the failures of both TFPI and AT are not entirely clear. optosis [68]. Therefore, this molecule demonstrates many de- The difference may be related to trial design issues, to dosing sirable characteristics of potential clinical utility in patients with differences, or, more likely, to inherent differences in the an- severe sepsis and septic shock. ticoagulant and anti-inflammatory properties of the molecules. On the basis of encouraging results from a phase 2 trial of recombinant human activated protein C (drotrecogin a [ac- CONCLUSION tivated]), an international phase 3 trial of activated protein C was conducted [69]. This randomized, double-blind, placebo- During the past decade, considerable advances have been made controlled, multicenter trial enrolled patients with severe sepsis. in the knowledge of the pathophysiology of severe sepsis. In Most patients were undergoing mechanical ventilation and re- addition, randomized clinical trials of critical care interven- quired vasopressors. A dosage of 24 mg/kg/h of drotrecogin a tions, such as low tidal volume ventilation and glucose control, (activated) was administered for a total duration of 96 h. The have reduced mortality more effectively than conventional ther- primary efficacy end point was 28-day all-cause mortality. apy [73, 74]. Advances in supportive care have not been par- The trial was stopped at the time of the second interim alleled by success with antisepsis interventions. Many trials of analysis by an independent data and safety monitoring board agents solely targeting the inflammatory cascade have failed to because of overwhelming efficacy that met predefined stopping reduce 28-day all-cause mortality. With the knowledge that the rules. At that time, 1728 patients were enrolled in the trial. A innate immune system is closely linked to the coagulation cas- 6.1% absolute and a 19.4% relative risk reduction in mortality cade, the antithrombotic approach now offers great promise

Inflammation and Coagulation • CID 2003:36 (15 May) • 1263 for bedside application of antisepsis therapy that will improve 21. Effect of high-dose glucocorticoid therapy on mortality in patients with clinical signs of systemic sepsis. The Veterans Administration Systemic survival. Sepsis Cooperative Study Group. N Engl J Med 1987; 317:659–65. 22. Annane D, Sebille V, Charpentier C, et al. Effect of treatment with low doses of hydrocortisone and fludrocortisone on mortality in patients References with septic shock. JAMA 2002; 288:862–71. 23. Opal SM. Corticosteroids for patients with septic shock. JAMA 1. Bone RC. Sir Isaac Newton, sepsis, SIRS, and CARS. Crit Care Med 2003; 289:41–2. 1996; 24:1125–8. 24. Zijlstra JG, van der Horst IC, Tulleken JE, et al. Corticosteroids for 2. Fisher CJ Jr, Agosti JM, Opal SM, et al. Treatment of septic shock with patients with septic shock. JAMA 2003; 289:42–4. the tumor necrosis factor receptor: Fc fusion protein. The Soluble TNF 25. Baldwin AS Jr. The NF-kappa B and I kappa B proteins: new discoveries Receptor Sepsis Study Group. N Engl J Med 1996; 334:1697–702. and insights. Annu Rev Immunol 1996; 14:649–83. 3. Hack CE, Zeerleder S. The endothelium in sepsis: source of and a 26. Bohrer H, Qiu F, Zimmermann T, et al. Role of NFkappaB in the target for inflammation. Crit Care Med 2001; 29:S21–7. mortality of sepsis. J Clin Invest 1997; 100:972–85. 4. Aird WC. Vascular bed-specific hemostasis: role of endothelium in 27. Dhainaut JF, Tenaillon A, Hemmer M, et al. Confirmatory platelet- sepsis pathogenesis. Crit Care Med 2001; 29:S28–34. activating factor receptor antagonist trial in patients with severe gram- Downloaded from https://academic.oup.com/cid/article/36/10/1259/307497 by guest on 30 September 2021 5. Vallet B, Wiel E. Endothelial cell dysfunction and coagulation. Crit negative bacterial sepsis: a phase III, randomized, double-blind, pla- Care Med 2001; 29:S36–41. cebo-controlled, multicenter trial. BN 52021 Sepsis Investigator Group. 6. Meijers JC, Bouma BN. New concepts in activation of the clotting Crit Care Med 1998; 26:1963–71. cascade in sepsis. Sepsis 1999; 3:87–91. 28. Kox WJ, Volk T, Kox SN, et al. Immunomodulatory therapies in sepsis. 7. Kidokoro A, Iba T, Fukunaga M, et al. Alterations in coagulation and Intensive Care Med 2000; 26(Suppl 1):S124–8. fibrinolysis during sepsis. Shock 1996; 5:223–8. 29. Dominioni L, Bianchi V, Imperatori A, et al. High-dose intravenous 8. Burdon D, Tiedje T, Pfeffer K, et al. The role of tumor necrosis factor IgG for treatment of severe surgical infections. Dig Surg 1996; 13:430–4. in the development of multiple organ failure in a murine model. Crit 30. Rogiers P. Hemofiltration treatment for sepsis: is it time for controlled Care Med 2000; 28:1962–7. trials? Kidney Int Suppl 1999; 72:S99–103. 9. Murphey ED, Traber DL. Pretreatment with tumor necrosis fac- 31. Bone RC, Balk RA, Cerra FB, et al. Definitions for sepsis and organ tor–alpha attenuates arterial hypotension and mortality induced by failure and guidelines for the use of innovative therapies in sepsis. The endotoxin in pigs. Crit Care Med 2000; 28:2015–21. ACCP/SCCM Consensus Conference Committee. American College of 10. Mancilla J, Garcia P, Dinarello CA. The interleukin-1 receptor antag- Chest Physicians/Society of Critical Care Medicine. Chest 1992; 101: onist can either reduce or enhance the lethality of Klebsiella pneumoniae 1644–55. sepsis in newborn rats. Infect Immun 1993; 61:926–32. 32. Vervloet MG, Thijs LG, Hack CE. Derangements of coagulation and 11. Wunderink R, Leeper K Jr, Schein R, et al. Filgrastim in patients with fibrinolysis in critically ill patients with sepsis and septic shock. Semin pneumonia and severe sepsis or septic shock. Chest 2001; 119:523–9. Thromb Hemost 1998; 24:33–44. 12. Angus DC, Birmingham MC, Balk RA, et al. E5 murine monoclonal 33. Lorente JA, Garcia-Frade LJ, Landin L, et al. Time course of hemostatic antiendotoxin antibody in gram-negative sepsis: a randomized con- abnormalities in sepsis and its relation to outcome. Chest 1993; 103: trolled trial. E5 Study Investigators. JAMA 2000; 283:1723–30. 1536–42. 13. McCloskey RV, Straube RC, Sanders C, et al. Treatment of septic shock 34. Carvalho AC, Freeman NJ. How coagulation defects alter outcome in with human monoclonal antibody HA-1A: a randomized, double- sepsis: survival may depend on reversing procoagulant conditions. J blind, placebo-controlled trial. CHESS Trial Study Group. Ann Intern Crit Illn 1994; 9:51–75. Med 1994; 121:1–5. 35. Tapper H, Herwald H. Modulation of hemostatic mechanisms in bac- 14. Abraham E, Wunderink R, Silverman H, et al. Efficacy and safety of terial infectious diseases. Blood 2000; 96:2329–37. monoclonal antibody to human tumor necrosis factor alpha in patients 36. Vincent JL. Dear SIRS, I’m sorry to say that I don’t like you… Crit with sepsis syndrome: a randomized, controlled, double-blind, mul- Care Med 1997; 25:372–4. ticenter clinical trial. TNF-alpha MAb Sepsis Study Group. JAMA 37. Dhainaut JF, Vallet B. Combined procoagulant and innate immune 1995; 273:934–41. responses to infection: toward more potent drugs in septic patients. 15. Abraham E, Anzueto A, Gutierrez G, et al. Double-blind randomised Crit Care Med 2001; 29:205–7. controlled trial of monoclonal antibody to human tumour necrosis 38. Bone RC, Grodzin CJ, Balk RA. Sepsis: a new hypothesis for patho- factor in treatment of septic shock. NORASEPT II Study Group. Lancet genesis of the disease process. Chest 1997; 112:235–43. 1998; 351:929–33. 39. Esmon CT. The endothelial cell protein C receptor. Thromb Haemost 16. Abraham E, Glauser MP, Butler T, et al. p55 Tumor necrosis factor 2000; 83:639–43. receptor fusion protein in the treatment of patients with severe sepsis 40. Esmon CT. Does inflammation contribute to thrombotic events? Hae- and septic shock: a randomized controlled multicenter trial. Ro 45- mostasis 2000; 30(Suppl 2):34–40. 2081 Study Group. JAMA 1997; 277:1531–8. 41. Levi M, Ten Cate H, van der Poll T, et al. Pathogenesis of disseminated 17. Grover R, Lopez A, Lorente A, et al. Multicenter, randomized, placebo- intravascular coagulation in sepsis. JAMA 1993; 270:975–9. controlled, double-blind study of the nitric oxide synthase inhibitor 42. Rosenberg RD, Aird WC. Vascular bed-specific hemostasis and hy- 546C88: effect on survival in patients with septic shock [abstract 1]. percoagulable states. N Engl J Med 1999; 340:1555–64. Crit Care Med 1999; 27:A33. 43. Hoffman M, Cooper ST. Thrombin enhances monocyte secretion of 18. Opal SM, Fisher CJ Jr, Dhainaut JF, et al. Confirmatory interleukin-1 tumor necrosis factor and interleukin-1 beta by two distinct mecha- receptor antagonist trial in severe sepsis: a phase III, randomized, dou- nisms. Blood Cells Mol Dis 1995; 21:156–67. ble-blind, placebo-controlled, multicenter trial. The Interleukin-1 Re- 44. Bone RC. The sepsis syndrome: definition and general approach to ceptor Antagonist Sepsis Investigator Group. Crit Care Med 1997; 25: management. Clin Chest Med 1996; 17:175–81. 1115–24. 45. Goyette RE. Hematology: a comprehensive guide to the diagnosis and 19. Carlet J. From mega to more reasonable doses of corticosteroids: a treatment of blood disorders. Los Angeles: PMIC, 1997. decade to recreate hope. Crit Care Med 1999; 27:672–4. 46. Iversen N, Strekerud FG, Abildgaard U. Tissue factor pathway inhibitor 20. Sprung CL, Caralis PV, Marcial EH, et al. The effects of high-dose (TFPI) in disseminated intravascular coagulation: low levels of the corticosteroids in patients with septic shock: a prospective, controlled activated factor X-TFPI complex. Blood Coagul Fibrinolysis 2000; 11: study. N Engl J Med 1984; 311:1137–43. 591–8.

1264 • CID 2003:36 (15 May) • Dellinger 47. Esmon CT. The protein C pathway. Crit Care Med 2000; 28:S44–8. thrombosis: basic principles and clinical practice. Philadelphia, PA: 48. Estenssoro E, Dubin A, Laffaire E, et al. Incidence, clinical course, and Lippincott Williams & Wilkins, 2001. outcome in 217 patients with acute respiratory distress syndrome. Crit 63. Uchiba M, Okajima K, Murakami K, et al. Attenuation of endotoxin- Care Med 2002; 30:2450–6. induced pulmonary vascular injury by antithrombin III. Am J Physiol 49. Compeau CG, Ma J, DeCampos KN, et al. In situ ischemia and hypoxia 1996; 270:L921–30. enhance alveolar macrophage tissue factor expression. Am J Respir Cell 64. Warren BL, Eid A, Singer P, et al. Caring for the critically ill patient. Mol Biol 1994; 11:446–55. High-dose antithrombin III in severe sepsis: a randomized controlled 50. Welty-Wolf KE, Carraway MS, Idell S, et al. Tissue factor in experi- trial. JAMA 2001; 286:1869–78. mental acute lung injury. Semin Hematol 2001; 38:35–8. 65. Faust SN, Levin M, Harrison OB, et al. Dysfunction of endothelial 51. Abraham E. Coagulation abnormalities in acute lung injury and sepsis. protein C activation in severe meningococcal sepsis. N Engl J Med Am J Respir Cell Mol Biol 2000; 22:401–4. 2001; 345:408–16. 52. Opal SM. Phylogenetic and functional relationships between coagu- 66. Bajzar L, Nesheim ME, Tracy PB. The profibrinolytic effect of activated lation and the innate immune response. Crit Care Med 2000; 28: protein C in clots formed from plasma is TAFI-dependent. Blood S77–80. 1996; 88:2093–100. 53. Opal SM, Thijs LG, Cavaillon JM, et al. Roundtable I: relationships 67. de Fouw NJ, van Hinsbergh VW, de Jong YF, et al. The interaction of between coagulation and inflammatory responses. Crit Care Med activated protein C and thrombin with the plasminogen activator in- Downloaded from https://academic.oup.com/cid/article/36/10/1259/307497 by guest on 30 September 2021 2000; 28:S81–2. hibitor released from human endothelial cells. Thromb Haemost 54. Fourrier F, Chopin C, Goudemand J, et al. Septic shock, multiple organ 1987; 57:176–82. failure, and disseminated intravascular coagulation: compared patterns 68. Joyce DE, Gelbert L, Ciaccia A, et al. Gene expression profile of an- of antithrombin III, protein C, and protein S deficiencies. Chest tithrombotic protein c defines new mechanisms modulating inflam- 1992; 101:816–23. mation and apoptosis. J Biol Chem 2001; 276:11199–203. 55. Yan SB, Helterbrand JD, Hartman DL, et al. Low levels of protein C 69. Bernard GR, Vincent JL, Laterre PF, et al. Efficacy and safety of re- are associated with poor outcome in severe sepsis. Chest 2001; 120: combinant human activated protein C for severe sepsis. N Engl J Med 915–22. 2001; 344:699–709. 56. Hack CE. Tissue factor pathway of coagulation in sepsis. Crit Care 70. Ely EW, Laterre P-F, Angus DC, et al. Drotrecogin alfa (activated) Med 2000; 28:S25–30. administration across clinically important subgroups of patients with 57. Abraham E. Tissue factor inhibition and clinical trial results of tissue severe sepsis. Crit Care Med 2003; 31:12–9. factor pathway inhibitor in sepsis. Crit Care Med 2000; 28:S31–3. 71. Drotrecogin alfa (activated) (recombinant human activated protein C 58. Uchiba M, Okajima K, Murakami K, et al. Effect of nafamostat mesilate rhAPC), Xigris, BLA 125029/0. FDA briefing information. US Food on pulmonary vascular injury induced by lipopolysaccharide in rats. and Drug Administration. Available at: http://www.fda.gov/ohrms/ Am J Respir Crit Care Med 1997; 155:711–8. dockets/ac/01/briefing/3797b1_01_Sponsor.htm. Accessed 21 January 59. Abraham E, Reinhart K, Svoboda P, et al. Assessment of the safety of 2003. recombinant tissue factor pathway inhibitor in patients with severe 72. The GUSTO Angiographic Investigators. The effects of tissue plas- sepsis: a multicenter, randomized, placebo-controlled, single-blind, minogen activator, streptokinase, or both on coronary-artery patency, dose escalation study. Crit Care Med 2001; 29:2081–9. ventricular function, and survival after acute myocardial infarction. N 60. Chiron Corporation. Press release: Chiron announces results of phase Engl J Med 1993; 329:1615–22. III study of tifacogin in severe sepsis. Available at: http://www.chiron. 73. The Acute Respiratory Distress Syndrome Network. Ventilation with com/investors/news/index.htm. Accessed 21 January 2003. lower tidal volumes as compared with traditional tidal volumes for 61. Opal SM, Thijs LG. New potential therapeutic modalities: antithrombin acute lung injury and the acute respiratory distress syndrome. N Engl III. Sepsis 1999; 3:153–9. J Med 2000; 342:1301–8. 62. Bock SC. Antithrombin III and heparin cofactor II. In: Colman RW, 74. van den Berghe G, Wouters P, Weekers F, et al. Intensive insulin therapy Hirsh J, Marder VJ, Clowes AW, George JN, eds. Hemostasis and in the critically ill patients. N Engl J Med 2001; 345:1359–67.

Inﬂammation and Coagulation • CID 2003:36 (15 May) • 1265