Neonatal Sepsis and Hemostasis: History
Please note this is an old version of this entry, which may differ significantly from the current revision.
Contributor:

Neonatal sepsis is considered critical for a significant increase in neonatal morbidity and mortality among hospitalized neonates. Neonatal sepsis, in most cases, coexists with coagulopathy, which can prove to be life-threatening.

  • neonatal sepsis
  • inflammation
  • hemostasis
  • viscoelastic tests

1. Introduction

Neonatal sepsis is the third most common cause of newborn mortality [1]. To date, the reported fatalities from neonatal sepsis have reached a rate of 2% in full-term infants and 20% in preterm infants, and it was further complicated by hemodynamic, inflammatory and clinical instability associated with systemic inflammatory response syndrome (SIRS) and meningitis in 30% of cases [2,3,4,5].
Sepsis is frequently related to coagulation disorders and presents with varied severity. In sepsis, both hemostasis and the immune system are activated, sharing common pathways that respond to cellular–endothelium interactions, eventually resulting in microvascular thrombosis. This complex process of thrombosis formation is poorly understood in adults and not at all in neonates [6,7]. Neonatal sepsis can present as mildly prolonged coagulation times and mild thrombocytopenia. However, it is often accompanied by significant and sometimes life-threatening coagulopathy, as well as by disseminated intravascular coagulopathy and thromboembolic events that impair organ function and fulminant bleeding that require prompt and appropriate intervention [8,9,10].

2. Developmental Hemostasis in Neonates

Age-related variations in coagulation components have been well established since the introduction of developmental hemostasis by Andrew et al., and were further validated by Monagle et al. in healthy full-term neonates and children up to 16 years old [11]. Compared with those of older individuals, the coagulation components and the properties of the endothelium in neonates are strongly affected by age [12,13]. Coagulation factors are produced by the fetus starting at 11 weeks’ gestation, and elevated levels are observed as gestation progresses and in postnatal life. Subsequently, preterm neonates exhibit lower coagulation factor levels compared with older individuals and adults. At birth, the vitamin K-dependent coagulation factors are almost at 30–50% of the adult levels for extremely preterm and full-term neonates, respectively, and reach adult values by approximately 6 months of age [14,15]. In contrast, coagulation factors V, VIII (FV, FVIII), and XIII, and von Willebrand factor (vWF) reach almost normal adult levels at birth [16].
The levels of natural anticoagulants, namely antithrombin (antithrombin-III, AT), heparin cofactor II, and proteins C and S, are significantly reduced in both preterm and full-term neonates, reaching almost 50% of adult levels at birth, except for α-macroglobulin, which is markedly increased. These coagulation inhibitors progressively increase near to adult levels by 3 and 6 months, respectively, for full-term and preterm neonates [14,15]. Furthermore, the activity of the plasmin/plasminogen system decreases, along with reduced fibrinolysis and faster thrombin generation capacity to ensure the neonatal hemostasis balance, was detected in preterm neonates in particular [17].
Furthermore, the antithrombotic and fibrinolytic properties of the endothelium are related to age; consequently, the neonatal endothelium is likely to differ from that of adults. In particular, endothelial cells express selectins in an age-dependent pattern similar to developmental hemostasis. Specifically, E-selectin and P-selectin reach adult levels by 32 and 11 weeks’ gestation [18]. Neonatal endothelial cells exhibit a low capacity to reverse oxidative agents, while lower levels of molecules with adhesion properties have been recorded at distinct gestational and postnatal ages [19]. Researchers have shed some light on the glycocalyx in neonates with necrotizing enterocolitis, but the biological structure of the endothelium and glycocalyx components in neonates is still under investigation [20].
Fetal platelets originate in megakaryocytes located in the fetal liver and, in the first trimester of pregnancy, already account for almost 150,000/μL. At 22 to 24 weeks’ gestation, they reach a stable value of around 250,000/μL until delivery in full-term gestation [21]. Additionally, the hyporeactivity of platelets has been reported in neonates during the first 10 days of life, while platelet counts are similar in neonates and adults. A significant impairment in activation and aggregation capability was found in neonatal platelets from cord blood after stimulation with adenosine diphosphate (ADP), epinephrine, collagen, thrombin, and thromboxane analogs in vitro compared with adult platelets [22]. This deficit is dominant among preterm neonates in response to platelet agonists [20]. Neonatal platelets present a reduced number of a2-adrenergic receptors, reduced expression of the thrombin receptors (protease activated receptors, PARs) PAR-1 and PAR-4, and a downregulation of signaling from the thromboxane receptor. Age-related hypersensitivity in the effect of prostaglandin E1 (PGE1) has been observed [23]. In addition, neonatal platelets’ hyporesponsiveness can be attributed to fewer dense granules and functional defects in the alpha granules, which decrease the degranulation and the fibrinogen-binding capacity [24]. Interestingly, counteractive factors, such as the mean platelet volume (MCV), higher hematocrit at birth, and higher concentrations of vWf and its ultra-large multimers, enhance the interaction between platelets and vessels and ultimately keep the neonatal hemostatic status in balance.
However, healthy neonates preserve coagulation homeostasis by maintaining balanced procoagulant protein levels in the plasma, thrombin generation, and fibrinolysis capacity, along with platelet hyporeactivity during the first days of life [25,26]. In contrast to their healthy counterparts, the fine line between bleeding and clotting is apparently disturbed in sick neonates, particularly during episodes of sepsis.

3. Neonatal Sepsis and Hemostasis

Inflammation and coagulation are mutually regulated [27]. Cytokines and chemokines, especially interleukin (IL) 6 and tumor necrosis factor alpha (TNF-α), which are expressed early at the onset and throughout the duration of sepsis, trigger the coagulation process through a cascade that leads to the activation of coagulation factors and anticoagulant proteins. In particular, the proinflammatory cytokines activate coagulation through upregulation of tissue factor (TF) expression and TF-mediated thrombin generation, and also through downregulation of the protein C system and enhanced fibrinolysis inhibition [6]. Coagulation factors interact with PARs, thus contributing to the inflammatory process. Elements of hemostasis, mainly thrombin, the TF-VIIa complex, and factor Xa, bind to PARs and activate the inflammatory response via intracellular endothelial cell signaling [28]. This is an excellent model for supporting the claim that inflammation promotes coagulation, after which coagulation intensifies the inflammatory process. Although this has been sufficiently elucidated in adults, it is still being investigated in neonates.
Animal studies have shown that T cells during fetal and neonatal life present a rather “toleragenic” (a TH2-skewed response) and anti-inflammatory phenotype, which is consistent with the developmental maturation of hemopoietic stem cells, and that neonates are highly prone to infection, in accordance with gestational age and postnatal health status [29]. Neonatal animal models presented with a weakened inflammatory response, including downregulation of the TNF-a related genes and a defective production of pro-inflammatory cytokines IL-23, IL-6, and IL-10, along with lower absolute concentrations of plasma cytokines and chemokines namely IL-1a/b, IL-12, granulocyte-macrophage colony-stimulating factor and sargramostim (GM-CSF), macrophage inflammatory protein (MIP)-1b, and IFN g [26].
During toxinemia, TF is expressed in the macrophages and monocytes, and initiates thrombin generation, thus promoting clot formation via extracellular vesicles. The TF contained in microparticles circulating in the plasma is transferred to cells that do not produce it or produce it in part, such as platelets, thereby enhancing the coagulation cascade during sepsis. This activation process arises from procoagulant cytokines, mainly IL-1 and TNF-a [30]. Neonates with an early predominance of inflammatory cytokines during sepsis have an increased risk of developing DIC [31]. This finding is consistent with the elevated levels of IL-6 in the serum and the high frequency of DIC seen with disseminated viral infection [2].
In order to counteract the spread of clotting, antithrombin, alongside its anticoagulation properties, provides downregulation of the cytokine receptors by linking to inflammatory cells [32]. Antithrombin interacts with endothelial cells, monocytes, neutrophils, and lymphocytes, and enhances prostacyclin release. The latter inhibits the interaction between endothelial cells and inflammatory cells, and also reduces the production of various cytokines and chemokines by endothelial cells [33]. In addition, activation of tissue factor induces an increase in the levels of the thrombin/antithrombin III (ATIII) complex (TAT), the plasminogen activator inhibitor (PAI), and the plasmin-α2–antiplasmin complex (PAP). Elevated levels of TAT and PAP in septic VLBW infants at 26–32 weeks’ gestation have been recorded, but no clear trend towards either thrombosis or hemorrhage has been shown [34]. In a prospective cohort study of full-term neonates with sepsis, severe infection was associated with activation of the contact system and consumption of anticoagulant proteins, in parallel with increased levels of the proteins of the complement system. Moreover, protein S was inactivated and anticoagulant proteins, including the TAT complex, increased, while fibrinolysis was inhibited, establishing a hypercoagulable state which resolved after antibiotic therapy among survivors [35]. Similarly, in a prospective case–control study, a significant simultaneous reduction in the TAT complex and protein C with elevated levels of inactive protein S was observed in full-term neonates with confirmed sepsis, while protein C levels were most markedly reduced in those who fatally developed DIC [36]. It has also been proven that neonates in the early stages of sepsis are prone to a prothrombotic state due to the consumption of coagulation inhibitors and activation of the coagulation cascade through cytokine release. This hypercoagulable state could mostly be resolved after administration of appropriate therapy. As there is ample evidence to show that activated protein C induces a reduction in TNF-α, IL-1β, IL-6, and IL-8 by blocking monocytes/macrophages, protein C has been suggested as potential therapeutic agent in neonatal sepsis [37]. In children with sepsis, fibrinolysis is profoundly inhibited, which is mostly attributed to an increase in plasma activity of the fibrinolysis inhibitor plasminogen activator inhibitor-1 (PAI-1) during sepsis, leading to severe sepsis and septic shock. Plasminogen activator inhibitor-1 has also been proposed as a promising treatment in pediatric sepsis, but relevant studies in neonates are still lacking [38].
Sepsis-induced endothelial dysfunction expressed as the disruption of antithrombotic properties results in the accumulation of fibrinogen [39]. During sepsis, disturbance in the endothelial glycocalyx structure modulates endothelium–neutrophil–platelet interactions, leading to thrombus formation and also to exacerbated fibrin formation and circulatory disorders. Glycocalyx impairment, along with inflammation during sepsis, leads to capillary leakage and vascular damage, which enhances inflammation and hypercoagulation. These aberrations result in increased vascular permeability, altered blood flow, impaired oxygen delivery, and ultimately to organ dysfunction [40]. Disturbances in the endothelial glycocalyx function induces disorders in tissue factor activation; thus disturbing the production of the tissue-type plasminogen activator and the plasminogen activator inhibitor-1. Moreover, the expression of glycosaminoglycans (such as heparan sulfates) of the injured glycocalyx is also diminished. Recently, reduced levels of endothelial glycocalyx components have been highlighted for reducing the hemostatic response of the endothelium. Furthermore, in adults, syndecan-1 levels are associated with the severity of sepsis and the development of DIC [41]. The attenuation of the anticoagulant properties of the glycosaminoglycans will directly impair the anticoagulant effect of the endothelium. Consequently, the fine line between thrombosis and bleeding becomes apparent [42]. Compared with adults, a decrease in TNF-a production from endothelial cells after inflammatory stimulation was shown in neonatal mouse models with Pneumocystis carinii infection of the respiratory tract. The diminished TNF-a production failed to enhance the expression of adhesion molecules in the surface of endothelial cells and finally attenuated the T cells’ migration capacity and the host’s defense response to infection [43]. This point underlines the importance of molecules deriving from endothelial cells and the glycocalyx during neonatal sepsis. In this context, TNFα has been proposed as a potential immunomodulator in neonatal sepsis [43]. Certainly, strong evidence regarding the contribution of the glycocalyx and endothelial cells to managing neonatal sepsis is still lacking, but this is a fairly promising research field.

4. Neonatal Platelets in Sepsis

Platelets are major players in sepsis-induced coagulopathy. During systemic inflammation, P-selectin is expressed on the platelet surface, facilitating the platelets’ adhesion to leukocytes and platelet aggregation, in parallel with tissue factor expression on monocytes [44]. Recently, the enhanced expression of GPIIb/IIIa receptors on activated platelet surfaces has been recognized in association with infection by Staphylococcus aureus and Escherichia coli, thereby demonstrating direct platelet activation in response to bacterial invasion and simultaneously introducing thromboinflammation and immunothrombosis [45,46,47]. The etiopathology and management of immunothrombosis in infancy and early childhood still lack sufficient evidence, as little research has been conducted in the pediatric population [48].
Thrombocytopenia noted in septic patients is mainly attributed to platelet consumption during clot propagation and thrombus formation through the activated endothelium [49]. In neonates, thrombocytopenia driven by LPS in Gram-negative sepsis is thought to be related to diminished expression of platelet Toll-like receptor 4 (TLR4) and is linked to elevated mortality rates [50]. Platelet activation was associated with high expression of platelet CD40L following endothelium inflammation, and higher platelet aggregation was observed after LPS stimulation (mostly in Gram-negative sepsis) [51]. Higher CD40L expression levels in platelets from cord blood samples were observed in premature neonates with histologically proven chorioamnionitis [52].
Moreover, research on neonatal thrombopoiesis during sepsis strongly suggested that neonates respond to sepsis by upregulating thrombopoietin (Tpo) production; although the degree of upregulation is modest, neonates present with a hypercoagulant profile at the onset of infection. In septic neonates, elevated levels of circulating megakaryocyte progenitors (CMPs) have also been observed. Simultaneous measurements of serum TPO levels and reticulated platelets (RP%) are helpful for discriminating hyperdestructive from hypoplastic thrombocytopenia in septic neonates [53,54].
As thrombocytopenia is linked to increased morbidity and mortality in ICU admissions, the delineation of platelet functionality may potentially alter the threshold levels of platelet transfusions [55]. Septic preterm neonates, when compared with healthy individuals, present a lower platelet adhesion capacity, which is mostly attributed to deficiencies in the intrinsic platelet properties rather than to an impairment in the concentrations or function of vWf [50]. These studies demonstrated that platelet activation and degranulation may follow thrombocytopenia, and that this phenomenon should be further investigated by means of accurate qualitative modalities for mapping distinct platelet phenotypes in patients with sepsis.
The missing part of this complex interplay between inflammation and coagulation is quantification of this model of cell-based coagulation triggered by an agent, such as sepsis, by means of a validated and practical tool for use in everyday clinical settings.

This entry is adapted from the peer-reviewed paper 10.3390/diagnostics12020261

This entry is offline, you can click here to edit this entry!
Video Production Service