- Email: [email protected]
Bi-directional activation of inflammation and coagulation in septic neonates
Early Human Development 90S1 (2014) S22–S25
Bi-directional activation of inflammation and coagulation in septic neonates Antonio Del Vecchio a, *, Mauro Stronati b , Caterina Franco a , Robert D. Christensen c a
Division of Neonatology, Neonatal Intensive Care Unit, Di Venere Hospital, Bari, Italy Neonatal Intensive Care Unit, Maternal-Infant Department, Fondazione IRCCS Policlinico San Matteo, Pavia, Italy c Women and Newborns Program, Intermountain Healthcare, Salt Lake City, Utah, USA b
A R T I C L E
I N F O
Keywords: Neonate Sepsis Coagulation Inflammatory cytokines Inflammation
A B S T R A C T Neonatal sepsis is frequently accompanied by significant and sometimes life-threatening coagulopathy. More complete understanding is needed of the molecular and cellular mechanisms underlying the interaction of the inflammatory and hemostatic systems. Such information may help focus future studies toward novel ways to improve the outcome of neonates who develop septic coagulopathy. © 2014 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Complex and balanced physiological systems exist for regulating inflammation and hemostasis. Overlap in these two intricate systems occurs because the vascular endothelium and platelet– endothelial and neutrophil–endothelial interactions are central to both systems [1]. Because of this overlap in regulatory mechanisms, it is understandable that activation of inflammatory pathways due to sepsis can lead to significant pathological changes in coagulation. Indeed, neonatal sepsis is often accompanied by significant and sometimes life-threatening coagulopathy [2]. Among ill and preterm neonates, sepsis and coagulopathy are common and often devastating pathological entities, with the two conditions frequently occurring simultaneously. However, the exact mechanisms involved in the inflammation/hemostasis interaction, and means of preventing morbidity from coagulopathy during neonatal sepsis, remain largely undiscovered. This review will focus on developmental aspects of inflammation and coagulation, with the intent of describing a basis for future experimental and clinical studies aimed at improving the outcomes of neonates who develop coagulopathy during sepsis. 2. Inflammation, sepsis, and coagulation Invasion of microbial pathogens into normally sterile tissues results in activation of the innate immune response, triggering a chain of events including the production and secretion of cytokines and chemokines and the activation of macrophages and monocytes. The activated proinflammatory and anti-inflammatory cytokine network is closely linked with other physiological systems including the coagulation–fibrinolytic system, production and regulation of
* Corresponding author: Antonio Del Vecchio, Division of Neonatology, Neonatal Intensive Care Unit, Di Venere Hospital, Via Ospedale Di Venere 1, 70121 Bari, Italy. Tel.: +39 080 5015012; fax: +39 080 5015016. E-mail address: [email protected] (A. Del Vecchio). 0378-3782/$ – see front matter © 2014 Elsevier Ireland Ltd. All rights reserved.
acute-phase and heat-shock proteins, interactions of neutrophils with vascular endothelium and interactions of platelets with vascular endothelium, hypothalamic–pituitary–adrenal axis activation, cell apoptosis, increased nitric oxide (NO) production, and the oxidant-antioxidant pathways [3] (Fig. 1). Gram positive and Gram negative bacteria, viruses, fungi, and protozoa, can immediately affect the coagulation system of neonates. An example of such is the significant link between elevated markers of inflammation at birth (elevated C-reactive protein and elevated leukocyte left shift) and an elevated fibrinogen concentration [4]. Since fibrinogen, central to coagulation, is an acute phase reaction, it is predictable that during perinatal infection the fibrinogen concentration initially rises. However, fibrinogen can be consumed when overwhelming sepsis proceeds to DIC, thus the fibrinogen concentration can increase very early during infection and then fall as the infection evolves into septic coagulopathy. Recent studies show a bidirectional cross-talk between the systems regulating inflammation and coagulation, and some of the mechanisms involved in this cross-talk are becoming clear. For instance, after microbial invasion, cellular pattern recognition receptors (PRRs) such as Toll-like receptor (TLR)-4 and CD-14, become expressed on the surface of monocytes and macrophages. When bacteria interact with these PRRs, proinflammatory cytokines are released. At this stage coagulation also becomes activated, although the mechanisms by which this occurs are just beginning to be understood [3]. Three pathways have been proposed by which bacteria and proinflammatory cytokines activate coagulation: (1) cytokine induction of tissue factor (TF) expression and TF-mediated thrombin generation, (2) inflammation-induced down-regulation of the protein C (PC) system, and (3) inhibition of fibrinolysis. These three changes are expected to result in a net procoagulant state with increased fibrin deposition [1]. As inflammation initiates coagulation, so the coagulation can activate inflammation. An example of the latter occurs when the procoagulant enzyme thrombin activates the acute inflamma-
A. Del Vecchio et al. / Early Human Development 90S1 (2014) S22–S25
S23
Fig. 1. Schematic representation of the interaction between infection, inflammation and coagulation. Invasion of the host by microbial pathogens causes the response of the innate immune system which activates a chain of events that results in the production and secretion of cytokines and chemokines involving other homeostatic pathways, including coagulation and fibrinolysis. The onset of sepsis seems to result from a combination of uncontrolled cascades of coagulation, fibrinolysis and inflammation. PAI, plasminogen activator inhibitor; T, thrombin; TAFI, thrombin activatable fibrinolytic inhibitor; TF, tissue factor; TM, thrombomodulin; t-PA, tissue plasminogen activator.
tory response by way of intracellular signal transduction initiated through protease activated receptors (PARs) on endothelial cells. It is apparent that activation of hemostasis and inflammation can synergize, in ways still being discovered. The apparent purpose of this synergy is to facilitate the recognition and elimination of invading pathogens and to limit the damage done during this process [3]. Inflammation up-regulates TF synthesis in endothelial cells, monocytes/macrophages and dendritic cells, and induces TF expression on the surface of mononuclear cells, endothelial cells and parenchymal cells, including heart, lung, brain and kidney. The responsible cytokines appear to be predominantly IL-6, TNFα, IL-1, and IL-12. TF binds and activates factor VII which, initiating the extrinsic clotting pathway via factor X, and catalyzes the cleavage of prothrombin (factor II) to thrombin, cleaving fibrinogen to fibrin. IL-1 and TNFα can work in an additional way as procoagulant cytokines. Furthermore, leukocytes, endothelial cells, vascular smooth muscle cells and platelets can shed microparticles of TF in the plasma that are implicated in activation of both coagulation and inflammation in sepsis. These microparticles have the role of transferring TF to cells that do not produce TF and can be considered part of a mechanism of amplification of signals generated during sepsis. Furthermore, the interaction between platelets, leukocytes, and endothelium can produce a pro-coagulant effect enhanced by inflammation. Endotoxin, pro-inflammatory mediators including platelet activating factor, and thrombin itself activate platelets and granulocytes that further stimulate monocyte TF expression. Severe infection impairs the function of the PC system, as a result of decreased synthesis of protein C by the liver, increased consumption of protein C, and impaired activation of protein C by diminished thrombomodulin expression on endothelial cells. Thus the widespread formation of fibrin is further facilitated. PC activated form (APC) inhibit thrombin generation through irreversible inhibition of factors Va and VIIIa, and also acts as inhibitor of fibrinolysis by enhancing the function of two fibrinolysis
inhibitors: thrombin activatable fibrinolytic inhibitor (TAFI) and plasminogen activator inhibitor type I (PAI-1) [5]. Fibrinolysis is considerably involved in the inflammation– coagulation balance. Pro-inflammatory cytokines stimulate PAI secretion by endothelium, and the consequent depression of the fibrinolytic system impairs fibrin removal within circulation. Even though defective late fibrin removal may exacerbate organ damage in sepsis, the inflammation-induced inhibition of fibrinolysis can be considered a rational part of the host defense [5]. Since a crosstalk between inflammation and coagulation exists, the activation of the coagulation system can in turn notably affect the inflammatory response. Proteases released during activation of the clotting cascade have been reported to maintain and reinforce inflammatory reactions, and, during severe sepsis, contribute to multiple organ failure and death. Components of the thrombin/fibrin pathway can also modulate inflammation, as they can affect the inflammatory cell responses. In most cases, endothelial cells, platelets and monocytes/macrophages become activated and secretion of IL-1 and IL-8 increases [6]. The PC pathway, independently of its anticoagulant function, also has anti-inflammatory effects; it has been recently reported that APC acts as a therapeutic drug for severe sepsis [5]. In addition, we can suppose that in the course of severe infection or sepsis patients with thrombophilia may suffer from more severe coagulopathy, and sepsis result in a more serious clinical course and an adverse outcome. 3. Fetal and neonatal development of the coagulation system Hemostasis evolves in utero and physiologic concentrations of coagulation proteins gradually increase with gestational age. In the 1980s Andrew introduced the term “developmental hemostasis” to describe the age-related physiological changes of the coagulation system as it develops progressively from fetal, to neonatal, to pediatric, to adult systems [7].
S24
A. Del Vecchio et al. / Early Human Development 90S1 (2014) S22–S25
Coagulation proteins are synthesized by the fetus and are measurable by 5 to 10 weeks of gestational age and increase gradually with advancing gestational age. Plasma concentrations of the coagulation system proteins in premature and healthy term infants differ markedly from those in adults as is clear from the reference ranges established by Andrew et al. [8,9]. Adults with plasma concentrations of hemostatic protein similar to neonatal plasma concentrations are at risk for pathologic bleeding or thrombosis. At birth, plasma concentrations of the vitamin K-dependent coagulant factors (factors II, VII, IX, and X) and the “contact factors” (factors XI and XII, prekalIekrein, and high-molecularweight kininogen) are less than 70% of adult values. Plasma concentrations of fibrinogen, factor V, factor VIII, von Willebrand’s factor, and factor XIII are greater than 70% of normal adult values. Plasma concentrations of inhibitors of coagulation are also different in neonates compared with adults; protein C and protein S are reduced at birth to 30% of adult plasma concentrations and remain decreased during the first weeks of life. The unique balance of coagulant proteins and inhibitors of coagulation in neonates influences overall generation of thrombin. Furthermore, antithrombotic and fibrinolytic properties of the endothelium are related to age, and consequently the neonatal endothelium is likely to differ from that of adults [10]. Additionally hyporeactivity of platelets has been reported in neonates during the first ten days of life [11]. Potential mechanisms to explain the physiologic differences in the coagulation system in neonates include decreased synthesis, increased clearance, and increased consumption. These mechanisms are incompletely understood [12]. Fetal and neonatal elements of the coagulation system show unique developmentally regulated patterns and times for maturation to normal adult protein quantities and functions. The dynamic factor and protein values reveal the relative immaturity of the neonatal hemostatic system; however, they are physiologic and protect healthy infants from hemorrhagic and thrombotic complications. Despite the immature coagulation system, well term and preterm infants uncommonly have bleeding or clotting problems, and the result of most screening tests of hemostasis vary only slightly from adult normal values [12]. It appears that sick term and preterm infants have a limited capacity to preserve the fine balance between pro- and anticoagulant systems, as it is easily disrupted by iatrogenic conditions and perinatal risk factors, such as infection, asphyxia, dehydration, central venous lines and inherited and acquired thrombophilia. Consequently bleeding and clotting become significant clinical problems in sick neonates, particularly so during neonatal sepsis. 4. Effects of neonatal sepsis on platelet counts and platelet function An obvious effect of sepsis on platelet count involves platelet consumption during DIC. In fact a falling platelet count is so typical of DIC as to be essential for the diagnosis. Semeraro et al. recently found that the presence of DIC during sepsis is an independent predictor of septic mortality [13]. Because of this association, perhaps finding means of recognizing DIC at its earliest stages could be used to somehow intensify treatments and reduce morbidity and mortality. Platelet counts can also fall during infection when DIC is not present [14]. The fall in count probably occurs too rapidly to be ascribed to reduced platelet production, and therefore is likely due to platelet consumption. Animal studies and clinical experiments have shown platelet dysfunction during infection. A peculiar reduction in adhesion of platelets to sub-endothelial surfaces has been described during infection, but an increase has been observed in platelet–platelet aggregation, induced by collagen, ADP, ristocetin or thrombin
[15]. Perhaps platelet aggregation during infection, with resulting platelet clumps filtered out by the spleen, explains some cases of septic thrombocytopenia in the absence of DIC. 5. Relationship between sepsis and coagulation system in neonates The link between inflammation and coagulation has been clearly elucidated in adults, but developmental differences that affect the inflammatory, coagulation, and immune responses make it difficult to extrapolate data from adult studies to neonates. It is challenging to determine how inflammation and hemostasis interact in neonate, because the two developmental pathways have unique features in fetal and neonatal life. Thus, it remains unclear how the increased susceptibility to sepsis and the immature coagulation system of the newborn influence one another. Developmental deficiencies of the host defense system, including a delayed maturation of the specific humoral and cellular immune response of neonatal B and T cells, and a decreased competence of neonatal cells to secrete cytokines involved in inflammatory response have been reported. Moreover, a defective activation of the complement cascade and deficiencies of neonatal myelopoiesis as responsible for compromised functions of the innate immune system have been described. Most studies conducted in neonates to evaluate the systemic inflammatory status aimed to recognize one or more cytokines as markers to identify infected neonates. Plasma levels of TNF-α, IL-1β, IL-4, IL-6, IL-8, and IL-10 are elevated in septic neonates, but it is difficult to know whether this is of pathogenic importance or is an epiphenomenon of the disease process [16]. Schultz et al. reported an increased production of proinflammatorycytokines in term and preterm infants, spontaneously and after endotoxin challenge, indicates a well-developed inflammatory response. They ascribed the greater susceptibility of the neonate for sepsis to an imbalance between pro- and anti-inflammatory cytokines in favor of proinflammatory cascade [17]. It has been demonstrated that these proinflammatory cytokines, triggered by infection or hypoxia, are involved in a final common pathway within a molecular cascade that may lead to perinatal brain damage, and may be also associated in the pathophysiology of perinatal stroke. Fetal or neonatal coagulation can be activated by maternal inflammation. Maternal diabetes, pre-eclampsia and chorioamnionitis may play a pivotal role in perinatal stroke and neonatal cerebral sinovenous thrombosis, and may be related to intracranial hemorrhage in preterm and term neonates. Additionally, a higher rate of placental thrombosis has been associated with the presence of antiphospholipid antibodies (APA) in pregnant women. APA can interfere with the activation of protein C and fibrinolysis, or determine the increase of proinflammatory cytokines and of complement activation. These inflammatory risk factors as well as transplacentally-derived maternal APA can be appreciated before birth, and permit early screening and intervention for those neonates recognized at risk [18]. The effect of sepsis and its severity on the physiologic inhibition system of coagulation, including protein C, protein S and antithrombin III, has been investigated in neonates. These deficiencies are associated with disseminated intravascular coagulation (DIC), thrombosis and poor outcome. Protein C is indicated as a very useful biomarker in severe sepsis, and as a possible tool for monitoring treatment with activated protein C. In a different study, low protein C plasma levels in low birth weight septic neonates have been related to higher rate of mortality [19,20]. Increased thrombin/ATIII complex (TAT) and/or plasmin/α2antiplasmin complex (PAP) values were demonstrated in very-lowbirth-weight infants with gestational age 26–32 weeks, even though
A. Del Vecchio et al. / Early Human Development 90S1 (2014) S22–S25
not always associated with DIC and clinical evidence of thrombosis. In addition, it has been demonstrated that neonatal sepsis induces a hypercoagulable state that decreases after a correct treatment [2]. The link between hemostatic imbalance and inflammatory response has also been elucidated by Grant in his study with thromboelastography (TEG); authors concluded that TEG changes (R- and K-time prolongation; MA reduction) may be used as early indicators of sepsis, enabling timely treatment before the onset of DIC. In addition, TEG seems reliable in detecting any underlying coagulopathy in neonates with established sepsis, and in effectively guiding specific therapy [21]. Thornburg and colleagues reported the association in neonates between catheter-related thrombosis and infections, supporting the hypothesis of a bidirectional relationship between thrombosis and inflammation induced by infection; the tip of the catheter may be covered by fibrin which provide a sort of nidus for bacterial adherence and growth, and the inflammation may activate coagulation inducing thrombin formation on indwelling catheters [22]. Moreover, some polymorphisms in genes coding for coagulation proteins can modify the host response to inflammation. The factor XIII val34leu polymorphism, which results in thinner fibrin fibrils with defective fibrinolysis, was associated with a higher rate of neonatal sepsis, and the factor VII-323del/ins (323 A1/A2) promoter polymorphism, which is associated with a 20 decrease in plasma factor VII activity, was associated with a reduced risk of the inflammatory disorder bronchopulmonary dysplasia [23]. 6. Implications for the use of anticoagulants during neonatal sepsis Neonatal purpura fulminans can be due to homozygous deficiency of protein C or protein S, or it can be acquired as a rare complication of neonatal sepsis. In either instance, marked hemorrhage and diffuse micro-thrombosis co-exist. The appearance of a neonate with this condition is distinctive, because of the diffuse areas of cutaneous necrosis and purpuric markings, including petechiae, and concomitant bleeding from venipuncture sites and mucous membranes. Anticoagulation has been recommended for neonates with purpura fulminans as a means of limiting the thrombotic and thromboembolic consequences. However, the mortality rate is very high in this condition, even with successful anticoagulation. The risks and benefits of anticoagulation have not been adequately evaluated in septic neonates who do not have purpura fulminans. Whether anticoagulants might improve outcomes of selected septic neonates seems a reasonable question, but will require a more complete understanding of the underlying responsible mechanisms, and clinical trials should be preceded by promising studies in experimental neonatal animals. 7. Conclusions Advances are needed toward understanding the interactions between the regulators of inflammation and coagulation. The common occurrence of coagulopathy during infection, and the poor prognostic significance of DIC during neonatal sepsis, underscores the importance of making basic and clinical advances in this area of study. Considering the crosstalk between the inflammatory and coagulation systems, it might be possible to devise new ways to recognize a procoagulant state early during infection, to identify
S25
DIC at its earliest stages during infection, and to intervene at early stages to prevent coagulopathy from becoming part of the clinical problem. Conflict of interest The authors report no conflict of interest. The authors alone are responsible for the content and writing of the paper. References [1] van der Poll T, Levi M. Crosstalk between inflammation and coagulation: the lessons of sepsis. Curr Vasc Pharmacol 2012(5);10:632–8. [2] Roman J, Velasco F, Fernandez F, Fernandez M, Villalba R, Rubio V, Vicente A, Torres A. Coagulation, fibrinolytic and kallikrein systems in neonates with uncomplicated sepsis and septic shock. Haemostasis 1993;23(3):142–8. [3] Levi M, van der Poll T. Inflammation and coagulation. Crit Care Med 2010;38(2 Suppl):S26–34. [4] Christensen RD, Baer VL, Lambert DK, Henry E, Ilstrup SJ, Bennett ST. Reference intervals for common coagulation tests of preterm infants. Transfusion. 2013 Jul 9. doi: 10.1111/trf.12322. [Epub ahead of print]. [5] Fourrier F. Severe sepsis, coagulation, and fibrinolysis: Dead end or one way? Crit Care Med 2012;40:2704–8. [6] van der Poll T, de Jonge E, Levi M. Regulatory role of cytokines in disseminated intravascular coagulation. Semin Thromb Hemost 2001;27:639–51. [7] Andrew M. The relevance of developmental hemostasis to hemorrhagic disorders of newborns. Semin Perinatol 1997;21:70–85. [8] Andrew M, Paes B, Milner R, Johnston M, Mitchell L, Tollefsen DM, Powers P. Development of the human coagulation system in the full-term infant. Blood 1987;70(1):165–72. [9] Andrew M, Paes B, Milner R, Johnston M, Mitchell L, Tollefsen DM, Castle V, Powers P. Development of the human coagulation system in the healthy premature infant. Blood 1988;72:1651–7. [10] Edstrom CS, Christensen RD, Andrew M. Developmental aspects of blood hemostasis and disorders of coagulation and fibrinolysis in the neonatal period. In: Christensen RD (ed.), Hematologic Problems of the Neonate. Philadelphia, PA: WB Saunders Co; 2000, pp. 247–9. [11] Sola-Visner M. Platelets in the neonatal period: developmental differences in platelet production, function, and hemostasis and the potential impact of therapies. Hematology Am Soc Hematol Educ Program 2012;2012:506–11. [12] Manco-Johnson MJ. Hemostasis in the Neonate. Neoreviews 2008;9;e119–23. [13] Semeraro N, Ammollo CT, Semeraro F, Colucci M. Sepsis-associated disseminated intravascular coagulation and thromboembolic disease. Mediterr J Hematol Infect Dis 2010;13:e2010024. [14] Eissa DS, El-Farrash RA. New insights into thrombopoiesis in neonatal sepsis. Platelets 2013;24:122–8. [15] Woth G, Varga A, Ghosh S, Krupp M, Kiss T, Bogár L, Mühl D. Platelet aggregation in severe sepsis. J Thromb Thrombolysis 2011;31:6–12. [16] Del Vecchio A, Laforgia M, Capasso M, Iolascon A, Latini G. The role of molecular genetics in the pathogenesis and diagnosis of neonatal sepsis. Clin Perinatol 2004;31:53–67. [17] Schultz C, Rott C, Temming P, Schlenke P, Moller JC, Bucsky P. Enhanced interleukin-6 and interleukin-8 synthesis in term and preterm infants. Pediatr Res 2002;51:317–22. [18] Armstrong-Wells JL, Manco-Johnson MJ. Coagulation Disorders: Inflammation and Thrombosis. In G. Buonocore et al. (eds.), Neonatology. A Practical Approach to Neonatal Diseases. Springer-Verlag Italia; 2012, pp. 770–4. [19] El Beshlawy A, Alaraby I, Abou Hussein H, Abou-Elew HH, Mohamed Abdel Kader MS. Study of protein C, protein S, and antithrombin III in newborns with sepsis. Pediatr Crit Care Med 2010;11(1):52–9. [20] Petäjä J, Manco-Johnson MJ. Protein C pathway in infants and children. Semin Thromb Hemost 2003;29:349–62. [21] Radicioni M, Mezzetti D, Del Vecchio A, Motta M. Thromboelastography: might work in neonatology too? J Matern Fetal Neonatal Med 2012;25(S4):18–21. [22] Thornburg CD, Smith PB, Smithwick ML, et al. Association between thrombosis and bloodstream infection in neonates with peripherally inserted catheters. Thromb Res 2008;122:782–5. [23] Härtel C, König I, Köster S, et al. Genetic polymorphisms of hemostasis genes and primary outcome of very low birth weight infants. Pediatrics 2006;118:683–9.
Bi-directional activation of inflammation and coagulation in septic neonates Antonio Del Vecchio a, *, Mauro Stronati b , Caterina Franco a , Robert D. Christensen c a
Division of Neonatology, Neonatal Intensive Care Unit, Di Venere Hospital, Bari, Italy Neonatal Intensive Care Unit, Maternal-Infant Department, Fondazione IRCCS Policlinico San Matteo, Pavia, Italy c Women and Newborns Program, Intermountain Healthcare, Salt Lake City, Utah, USA b
A R T I C L E
I N F O
Keywords: Neonate Sepsis Coagulation Inflammatory cytokines Inflammation
A B S T R A C T Neonatal sepsis is frequently accompanied by significant and sometimes life-threatening coagulopathy. More complete understanding is needed of the molecular and cellular mechanisms underlying the interaction of the inflammatory and hemostatic systems. Such information may help focus future studies toward novel ways to improve the outcome of neonates who develop septic coagulopathy. © 2014 Elsevier Ireland Ltd. All rights reserved.
1. Introduction Complex and balanced physiological systems exist for regulating inflammation and hemostasis. Overlap in these two intricate systems occurs because the vascular endothelium and platelet– endothelial and neutrophil–endothelial interactions are central to both systems [1]. Because of this overlap in regulatory mechanisms, it is understandable that activation of inflammatory pathways due to sepsis can lead to significant pathological changes in coagulation. Indeed, neonatal sepsis is often accompanied by significant and sometimes life-threatening coagulopathy [2]. Among ill and preterm neonates, sepsis and coagulopathy are common and often devastating pathological entities, with the two conditions frequently occurring simultaneously. However, the exact mechanisms involved in the inflammation/hemostasis interaction, and means of preventing morbidity from coagulopathy during neonatal sepsis, remain largely undiscovered. This review will focus on developmental aspects of inflammation and coagulation, with the intent of describing a basis for future experimental and clinical studies aimed at improving the outcomes of neonates who develop coagulopathy during sepsis. 2. Inflammation, sepsis, and coagulation Invasion of microbial pathogens into normally sterile tissues results in activation of the innate immune response, triggering a chain of events including the production and secretion of cytokines and chemokines and the activation of macrophages and monocytes. The activated proinflammatory and anti-inflammatory cytokine network is closely linked with other physiological systems including the coagulation–fibrinolytic system, production and regulation of
* Corresponding author: Antonio Del Vecchio, Division of Neonatology, Neonatal Intensive Care Unit, Di Venere Hospital, Via Ospedale Di Venere 1, 70121 Bari, Italy. Tel.: +39 080 5015012; fax: +39 080 5015016. E-mail address: [email protected] (A. Del Vecchio). 0378-3782/$ – see front matter © 2014 Elsevier Ireland Ltd. All rights reserved.
acute-phase and heat-shock proteins, interactions of neutrophils with vascular endothelium and interactions of platelets with vascular endothelium, hypothalamic–pituitary–adrenal axis activation, cell apoptosis, increased nitric oxide (NO) production, and the oxidant-antioxidant pathways [3] (Fig. 1). Gram positive and Gram negative bacteria, viruses, fungi, and protozoa, can immediately affect the coagulation system of neonates. An example of such is the significant link between elevated markers of inflammation at birth (elevated C-reactive protein and elevated leukocyte left shift) and an elevated fibrinogen concentration [4]. Since fibrinogen, central to coagulation, is an acute phase reaction, it is predictable that during perinatal infection the fibrinogen concentration initially rises. However, fibrinogen can be consumed when overwhelming sepsis proceeds to DIC, thus the fibrinogen concentration can increase very early during infection and then fall as the infection evolves into septic coagulopathy. Recent studies show a bidirectional cross-talk between the systems regulating inflammation and coagulation, and some of the mechanisms involved in this cross-talk are becoming clear. For instance, after microbial invasion, cellular pattern recognition receptors (PRRs) such as Toll-like receptor (TLR)-4 and CD-14, become expressed on the surface of monocytes and macrophages. When bacteria interact with these PRRs, proinflammatory cytokines are released. At this stage coagulation also becomes activated, although the mechanisms by which this occurs are just beginning to be understood [3]. Three pathways have been proposed by which bacteria and proinflammatory cytokines activate coagulation: (1) cytokine induction of tissue factor (TF) expression and TF-mediated thrombin generation, (2) inflammation-induced down-regulation of the protein C (PC) system, and (3) inhibition of fibrinolysis. These three changes are expected to result in a net procoagulant state with increased fibrin deposition [1]. As inflammation initiates coagulation, so the coagulation can activate inflammation. An example of the latter occurs when the procoagulant enzyme thrombin activates the acute inflamma-
A. Del Vecchio et al. / Early Human Development 90S1 (2014) S22–S25
S23
Fig. 1. Schematic representation of the interaction between infection, inflammation and coagulation. Invasion of the host by microbial pathogens causes the response of the innate immune system which activates a chain of events that results in the production and secretion of cytokines and chemokines involving other homeostatic pathways, including coagulation and fibrinolysis. The onset of sepsis seems to result from a combination of uncontrolled cascades of coagulation, fibrinolysis and inflammation. PAI, plasminogen activator inhibitor; T, thrombin; TAFI, thrombin activatable fibrinolytic inhibitor; TF, tissue factor; TM, thrombomodulin; t-PA, tissue plasminogen activator.
tory response by way of intracellular signal transduction initiated through protease activated receptors (PARs) on endothelial cells. It is apparent that activation of hemostasis and inflammation can synergize, in ways still being discovered. The apparent purpose of this synergy is to facilitate the recognition and elimination of invading pathogens and to limit the damage done during this process [3]. Inflammation up-regulates TF synthesis in endothelial cells, monocytes/macrophages and dendritic cells, and induces TF expression on the surface of mononuclear cells, endothelial cells and parenchymal cells, including heart, lung, brain and kidney. The responsible cytokines appear to be predominantly IL-6, TNFα, IL-1, and IL-12. TF binds and activates factor VII which, initiating the extrinsic clotting pathway via factor X, and catalyzes the cleavage of prothrombin (factor II) to thrombin, cleaving fibrinogen to fibrin. IL-1 and TNFα can work in an additional way as procoagulant cytokines. Furthermore, leukocytes, endothelial cells, vascular smooth muscle cells and platelets can shed microparticles of TF in the plasma that are implicated in activation of both coagulation and inflammation in sepsis. These microparticles have the role of transferring TF to cells that do not produce TF and can be considered part of a mechanism of amplification of signals generated during sepsis. Furthermore, the interaction between platelets, leukocytes, and endothelium can produce a pro-coagulant effect enhanced by inflammation. Endotoxin, pro-inflammatory mediators including platelet activating factor, and thrombin itself activate platelets and granulocytes that further stimulate monocyte TF expression. Severe infection impairs the function of the PC system, as a result of decreased synthesis of protein C by the liver, increased consumption of protein C, and impaired activation of protein C by diminished thrombomodulin expression on endothelial cells. Thus the widespread formation of fibrin is further facilitated. PC activated form (APC) inhibit thrombin generation through irreversible inhibition of factors Va and VIIIa, and also acts as inhibitor of fibrinolysis by enhancing the function of two fibrinolysis
inhibitors: thrombin activatable fibrinolytic inhibitor (TAFI) and plasminogen activator inhibitor type I (PAI-1) [5]. Fibrinolysis is considerably involved in the inflammation– coagulation balance. Pro-inflammatory cytokines stimulate PAI secretion by endothelium, and the consequent depression of the fibrinolytic system impairs fibrin removal within circulation. Even though defective late fibrin removal may exacerbate organ damage in sepsis, the inflammation-induced inhibition of fibrinolysis can be considered a rational part of the host defense [5]. Since a crosstalk between inflammation and coagulation exists, the activation of the coagulation system can in turn notably affect the inflammatory response. Proteases released during activation of the clotting cascade have been reported to maintain and reinforce inflammatory reactions, and, during severe sepsis, contribute to multiple organ failure and death. Components of the thrombin/fibrin pathway can also modulate inflammation, as they can affect the inflammatory cell responses. In most cases, endothelial cells, platelets and monocytes/macrophages become activated and secretion of IL-1 and IL-8 increases [6]. The PC pathway, independently of its anticoagulant function, also has anti-inflammatory effects; it has been recently reported that APC acts as a therapeutic drug for severe sepsis [5]. In addition, we can suppose that in the course of severe infection or sepsis patients with thrombophilia may suffer from more severe coagulopathy, and sepsis result in a more serious clinical course and an adverse outcome. 3. Fetal and neonatal development of the coagulation system Hemostasis evolves in utero and physiologic concentrations of coagulation proteins gradually increase with gestational age. In the 1980s Andrew introduced the term “developmental hemostasis” to describe the age-related physiological changes of the coagulation system as it develops progressively from fetal, to neonatal, to pediatric, to adult systems [7].
S24
A. Del Vecchio et al. / Early Human Development 90S1 (2014) S22–S25
Coagulation proteins are synthesized by the fetus and are measurable by 5 to 10 weeks of gestational age and increase gradually with advancing gestational age. Plasma concentrations of the coagulation system proteins in premature and healthy term infants differ markedly from those in adults as is clear from the reference ranges established by Andrew et al. [8,9]. Adults with plasma concentrations of hemostatic protein similar to neonatal plasma concentrations are at risk for pathologic bleeding or thrombosis. At birth, plasma concentrations of the vitamin K-dependent coagulant factors (factors II, VII, IX, and X) and the “contact factors” (factors XI and XII, prekalIekrein, and high-molecularweight kininogen) are less than 70% of adult values. Plasma concentrations of fibrinogen, factor V, factor VIII, von Willebrand’s factor, and factor XIII are greater than 70% of normal adult values. Plasma concentrations of inhibitors of coagulation are also different in neonates compared with adults; protein C and protein S are reduced at birth to 30% of adult plasma concentrations and remain decreased during the first weeks of life. The unique balance of coagulant proteins and inhibitors of coagulation in neonates influences overall generation of thrombin. Furthermore, antithrombotic and fibrinolytic properties of the endothelium are related to age, and consequently the neonatal endothelium is likely to differ from that of adults [10]. Additionally hyporeactivity of platelets has been reported in neonates during the first ten days of life [11]. Potential mechanisms to explain the physiologic differences in the coagulation system in neonates include decreased synthesis, increased clearance, and increased consumption. These mechanisms are incompletely understood [12]. Fetal and neonatal elements of the coagulation system show unique developmentally regulated patterns and times for maturation to normal adult protein quantities and functions. The dynamic factor and protein values reveal the relative immaturity of the neonatal hemostatic system; however, they are physiologic and protect healthy infants from hemorrhagic and thrombotic complications. Despite the immature coagulation system, well term and preterm infants uncommonly have bleeding or clotting problems, and the result of most screening tests of hemostasis vary only slightly from adult normal values [12]. It appears that sick term and preterm infants have a limited capacity to preserve the fine balance between pro- and anticoagulant systems, as it is easily disrupted by iatrogenic conditions and perinatal risk factors, such as infection, asphyxia, dehydration, central venous lines and inherited and acquired thrombophilia. Consequently bleeding and clotting become significant clinical problems in sick neonates, particularly so during neonatal sepsis. 4. Effects of neonatal sepsis on platelet counts and platelet function An obvious effect of sepsis on platelet count involves platelet consumption during DIC. In fact a falling platelet count is so typical of DIC as to be essential for the diagnosis. Semeraro et al. recently found that the presence of DIC during sepsis is an independent predictor of septic mortality [13]. Because of this association, perhaps finding means of recognizing DIC at its earliest stages could be used to somehow intensify treatments and reduce morbidity and mortality. Platelet counts can also fall during infection when DIC is not present [14]. The fall in count probably occurs too rapidly to be ascribed to reduced platelet production, and therefore is likely due to platelet consumption. Animal studies and clinical experiments have shown platelet dysfunction during infection. A peculiar reduction in adhesion of platelets to sub-endothelial surfaces has been described during infection, but an increase has been observed in platelet–platelet aggregation, induced by collagen, ADP, ristocetin or thrombin
[15]. Perhaps platelet aggregation during infection, with resulting platelet clumps filtered out by the spleen, explains some cases of septic thrombocytopenia in the absence of DIC. 5. Relationship between sepsis and coagulation system in neonates The link between inflammation and coagulation has been clearly elucidated in adults, but developmental differences that affect the inflammatory, coagulation, and immune responses make it difficult to extrapolate data from adult studies to neonates. It is challenging to determine how inflammation and hemostasis interact in neonate, because the two developmental pathways have unique features in fetal and neonatal life. Thus, it remains unclear how the increased susceptibility to sepsis and the immature coagulation system of the newborn influence one another. Developmental deficiencies of the host defense system, including a delayed maturation of the specific humoral and cellular immune response of neonatal B and T cells, and a decreased competence of neonatal cells to secrete cytokines involved in inflammatory response have been reported. Moreover, a defective activation of the complement cascade and deficiencies of neonatal myelopoiesis as responsible for compromised functions of the innate immune system have been described. Most studies conducted in neonates to evaluate the systemic inflammatory status aimed to recognize one or more cytokines as markers to identify infected neonates. Plasma levels of TNF-α, IL-1β, IL-4, IL-6, IL-8, and IL-10 are elevated in septic neonates, but it is difficult to know whether this is of pathogenic importance or is an epiphenomenon of the disease process [16]. Schultz et al. reported an increased production of proinflammatorycytokines in term and preterm infants, spontaneously and after endotoxin challenge, indicates a well-developed inflammatory response. They ascribed the greater susceptibility of the neonate for sepsis to an imbalance between pro- and anti-inflammatory cytokines in favor of proinflammatory cascade [17]. It has been demonstrated that these proinflammatory cytokines, triggered by infection or hypoxia, are involved in a final common pathway within a molecular cascade that may lead to perinatal brain damage, and may be also associated in the pathophysiology of perinatal stroke. Fetal or neonatal coagulation can be activated by maternal inflammation. Maternal diabetes, pre-eclampsia and chorioamnionitis may play a pivotal role in perinatal stroke and neonatal cerebral sinovenous thrombosis, and may be related to intracranial hemorrhage in preterm and term neonates. Additionally, a higher rate of placental thrombosis has been associated with the presence of antiphospholipid antibodies (APA) in pregnant women. APA can interfere with the activation of protein C and fibrinolysis, or determine the increase of proinflammatory cytokines and of complement activation. These inflammatory risk factors as well as transplacentally-derived maternal APA can be appreciated before birth, and permit early screening and intervention for those neonates recognized at risk [18]. The effect of sepsis and its severity on the physiologic inhibition system of coagulation, including protein C, protein S and antithrombin III, has been investigated in neonates. These deficiencies are associated with disseminated intravascular coagulation (DIC), thrombosis and poor outcome. Protein C is indicated as a very useful biomarker in severe sepsis, and as a possible tool for monitoring treatment with activated protein C. In a different study, low protein C plasma levels in low birth weight septic neonates have been related to higher rate of mortality [19,20]. Increased thrombin/ATIII complex (TAT) and/or plasmin/α2antiplasmin complex (PAP) values were demonstrated in very-lowbirth-weight infants with gestational age 26–32 weeks, even though
A. Del Vecchio et al. / Early Human Development 90S1 (2014) S22–S25
not always associated with DIC and clinical evidence of thrombosis. In addition, it has been demonstrated that neonatal sepsis induces a hypercoagulable state that decreases after a correct treatment [2]. The link between hemostatic imbalance and inflammatory response has also been elucidated by Grant in his study with thromboelastography (TEG); authors concluded that TEG changes (R- and K-time prolongation; MA reduction) may be used as early indicators of sepsis, enabling timely treatment before the onset of DIC. In addition, TEG seems reliable in detecting any underlying coagulopathy in neonates with established sepsis, and in effectively guiding specific therapy [21]. Thornburg and colleagues reported the association in neonates between catheter-related thrombosis and infections, supporting the hypothesis of a bidirectional relationship between thrombosis and inflammation induced by infection; the tip of the catheter may be covered by fibrin which provide a sort of nidus for bacterial adherence and growth, and the inflammation may activate coagulation inducing thrombin formation on indwelling catheters [22]. Moreover, some polymorphisms in genes coding for coagulation proteins can modify the host response to inflammation. The factor XIII val34leu polymorphism, which results in thinner fibrin fibrils with defective fibrinolysis, was associated with a higher rate of neonatal sepsis, and the factor VII-323del/ins (323 A1/A2) promoter polymorphism, which is associated with a 20 decrease in plasma factor VII activity, was associated with a reduced risk of the inflammatory disorder bronchopulmonary dysplasia [23]. 6. Implications for the use of anticoagulants during neonatal sepsis Neonatal purpura fulminans can be due to homozygous deficiency of protein C or protein S, or it can be acquired as a rare complication of neonatal sepsis. In either instance, marked hemorrhage and diffuse micro-thrombosis co-exist. The appearance of a neonate with this condition is distinctive, because of the diffuse areas of cutaneous necrosis and purpuric markings, including petechiae, and concomitant bleeding from venipuncture sites and mucous membranes. Anticoagulation has been recommended for neonates with purpura fulminans as a means of limiting the thrombotic and thromboembolic consequences. However, the mortality rate is very high in this condition, even with successful anticoagulation. The risks and benefits of anticoagulation have not been adequately evaluated in septic neonates who do not have purpura fulminans. Whether anticoagulants might improve outcomes of selected septic neonates seems a reasonable question, but will require a more complete understanding of the underlying responsible mechanisms, and clinical trials should be preceded by promising studies in experimental neonatal animals. 7. Conclusions Advances are needed toward understanding the interactions between the regulators of inflammation and coagulation. The common occurrence of coagulopathy during infection, and the poor prognostic significance of DIC during neonatal sepsis, underscores the importance of making basic and clinical advances in this area of study. Considering the crosstalk between the inflammatory and coagulation systems, it might be possible to devise new ways to recognize a procoagulant state early during infection, to identify
S25
DIC at its earliest stages during infection, and to intervene at early stages to prevent coagulopathy from becoming part of the clinical problem. Conflict of interest The authors report no conflict of interest. The authors alone are responsible for the content and writing of the paper. References [1] van der Poll T, Levi M. Crosstalk between inflammation and coagulation: the lessons of sepsis. Curr Vasc Pharmacol 2012(5);10:632–8. [2] Roman J, Velasco F, Fernandez F, Fernandez M, Villalba R, Rubio V, Vicente A, Torres A. Coagulation, fibrinolytic and kallikrein systems in neonates with uncomplicated sepsis and septic shock. Haemostasis 1993;23(3):142–8. [3] Levi M, van der Poll T. Inflammation and coagulation. Crit Care Med 2010;38(2 Suppl):S26–34. [4] Christensen RD, Baer VL, Lambert DK, Henry E, Ilstrup SJ, Bennett ST. Reference intervals for common coagulation tests of preterm infants. Transfusion. 2013 Jul 9. doi: 10.1111/trf.12322. [Epub ahead of print]. [5] Fourrier F. Severe sepsis, coagulation, and fibrinolysis: Dead end or one way? Crit Care Med 2012;40:2704–8. [6] van der Poll T, de Jonge E, Levi M. Regulatory role of cytokines in disseminated intravascular coagulation. Semin Thromb Hemost 2001;27:639–51. [7] Andrew M. The relevance of developmental hemostasis to hemorrhagic disorders of newborns. Semin Perinatol 1997;21:70–85. [8] Andrew M, Paes B, Milner R, Johnston M, Mitchell L, Tollefsen DM, Powers P. Development of the human coagulation system in the full-term infant. Blood 1987;70(1):165–72. [9] Andrew M, Paes B, Milner R, Johnston M, Mitchell L, Tollefsen DM, Castle V, Powers P. Development of the human coagulation system in the healthy premature infant. Blood 1988;72:1651–7. [10] Edstrom CS, Christensen RD, Andrew M. Developmental aspects of blood hemostasis and disorders of coagulation and fibrinolysis in the neonatal period. In: Christensen RD (ed.), Hematologic Problems of the Neonate. Philadelphia, PA: WB Saunders Co; 2000, pp. 247–9. [11] Sola-Visner M. Platelets in the neonatal period: developmental differences in platelet production, function, and hemostasis and the potential impact of therapies. Hematology Am Soc Hematol Educ Program 2012;2012:506–11. [12] Manco-Johnson MJ. Hemostasis in the Neonate. Neoreviews 2008;9;e119–23. [13] Semeraro N, Ammollo CT, Semeraro F, Colucci M. Sepsis-associated disseminated intravascular coagulation and thromboembolic disease. Mediterr J Hematol Infect Dis 2010;13:e2010024. [14] Eissa DS, El-Farrash RA. New insights into thrombopoiesis in neonatal sepsis. Platelets 2013;24:122–8. [15] Woth G, Varga A, Ghosh S, Krupp M, Kiss T, Bogár L, Mühl D. Platelet aggregation in severe sepsis. J Thromb Thrombolysis 2011;31:6–12. [16] Del Vecchio A, Laforgia M, Capasso M, Iolascon A, Latini G. The role of molecular genetics in the pathogenesis and diagnosis of neonatal sepsis. Clin Perinatol 2004;31:53–67. [17] Schultz C, Rott C, Temming P, Schlenke P, Moller JC, Bucsky P. Enhanced interleukin-6 and interleukin-8 synthesis in term and preterm infants. Pediatr Res 2002;51:317–22. [18] Armstrong-Wells JL, Manco-Johnson MJ. Coagulation Disorders: Inflammation and Thrombosis. In G. Buonocore et al. (eds.), Neonatology. A Practical Approach to Neonatal Diseases. Springer-Verlag Italia; 2012, pp. 770–4. [19] El Beshlawy A, Alaraby I, Abou Hussein H, Abou-Elew HH, Mohamed Abdel Kader MS. Study of protein C, protein S, and antithrombin III in newborns with sepsis. Pediatr Crit Care Med 2010;11(1):52–9. [20] Petäjä J, Manco-Johnson MJ. Protein C pathway in infants and children. Semin Thromb Hemost 2003;29:349–62. [21] Radicioni M, Mezzetti D, Del Vecchio A, Motta M. Thromboelastography: might work in neonatology too? J Matern Fetal Neonatal Med 2012;25(S4):18–21. [22] Thornburg CD, Smith PB, Smithwick ML, et al. Association between thrombosis and bloodstream infection in neonates with peripherally inserted catheters. Thromb Res 2008;122:782–5. [23] Härtel C, König I, Köster S, et al. Genetic polymorphisms of hemostasis genes and primary outcome of very low birth weight infants. Pediatrics 2006;118:683–9.