Thrombosis and platelets: an update

Haemostasis and thrombosis are complex, multifactorial processes. There is an evolving understanding of the mechanisms influencing vascular occlusion and the role of inflammation and immunity. Despite major advances in elucidating the mechanistic pathways mediating platelet function and thrombosis, challenges in the treatment of vascular occlusive diseases persist. Pharmacological advances have greatly affected thrombotic outcomes, but this has led to the unwanted side effect of bleeding. Detailed assessment of the impact of non-thrombotic diseases on haemostasis and thrombosis is necessary to better evaluate thrombotic risk and establish optimal treatment. This review will focus on recent advances in understanding the contribution of evolving risk factors to thrombosis.

Platelets , Thrombosis , Coagulation , Cardiovascular , Bleeding

Introduction

Heart disease is the leading cause of death in many developed countries and approximately 50% of all deaths associated with malignant neoplasms are due to thrombotic events. Damage to and extravasation of blood from the vascular circulatory system commonly occurs throughout life. Haemostasis is the process that maintains the regulation of vascular integrity and blood flow. Normal haemostasis may be overwhelmed by pathological factors, leading to uncontrolled clot formation and vessel occlusion in either the arteries or veins. Platelets, together with endothelial cells and circulating coagulation proteins, are crucial mediators of vascular haemostasis and thrombosis.

Arterial thrombosis is the cause of myocardial infarction (MI) and stroke, while venous thrombosis (VT) leads to venous thromboembolism (VTE) and pulmonary embolism (PE). Structurally, arterial and venous thrombi are distinct. Arterial thrombi are rich in platelets and form at the sides of or around ruptured atherosclerotic plaques. Venous thrombi are rich in fibrin and red blood cells and may occur despite an intact endothelial wall. Finally, arterial thrombosis occurs at places of high shear flow while VT occurs in the setting of slow shear flow (Figure 1).

Figure 1

Major differences between arterial and venous thrombosis. (A) Arterial thrombosis occurs under high shear flow when platelet rich thrombi are formed around ruptured atherosclerotic plaques and damaged endothelium. (B) Venous thrombosis occurs under low shear flow and mostly around intact endothelial wall. Venous thrombi are fibrin rich, encapsulating a large amount of red blood cells in addition to activated platelets.

Two major mechanisms mediate vascular homeostasis and thrombosis depending on vascular damage or vessel structure.¹ One is mediated by collagen and the other is tissue factor (TF) dependent. During normal haemostasis, damage to the endothelial cell layer may occur and collagen from the subendothelial space is exposed. Platelets, through their glycoproteins (GP) GPVI and GPIb/V/IX, interact with collagen and von Willebrand factor (vWF). Collagen exposure leads to platelet adhesion and formation of a platelet monolayer. Platelets form a three-dimensional structure by aggregating through their activated GPIIb/IIIa (α_IIbβ₃) integrins. Activated platelets recruit other circulating platelets by secreting aggregatory mediators such as thromboxane A2, ADP,² ultra-large vWF (ulvWF) multimers and serotonin as well as by producing thrombin (Figure 2). Deeper tissue damage leads to release of TF from smooth muscle, adventitial cells, and pericytes. TF mediates the conversion of pro-thrombin to thrombin, fibrin generation, and activation of the clotting cascade¹ (Figure 2). Endothelial cells play a major role in thrombus propagation control and its subsequent resolution.

Figure 2

Mechanisms of platelet-mediated thrombosis. (A) Collagen-mediated platelet thrombus formation occurs when subendothelial collagen becomes exposed to the circulation. Platelets, through their glycoproteins, interact with collagen and collagen-deposited vWF, change their shape and adhere to the site of injury. The attachment leads to secretion of ADP, serotonin and thromboxane (TxA2) leading to the recruitment and activation of more platelets. Activated platelets release thrombin (IIa) leading to platelet aggregation and three-dimensional clot formation. In certain instances, damage to the vessel may extend beyond the endothelium into the adventitial layer. In those instances, thrombosis is mediated through (B) Tissue factor (TF). Active TF is expressed by smooth muscle and adventitial cells and is able to generate thrombin (IIa). TF-generated thrombin, in turn, activates platelets, fibrin generation and the coagulation cascade to form a thrombus. Of note, circulating active TF may also be secreted by monocytes and is present in tumor-secreted microparticles. (C) Platelet thrombosis mediated by ultra large vWF multimers (ULvWF). Endothelial cells activated by high shear stress and/or epinephrine secrete large multimers of vWF forming strings that catch and crosslink platelets to the surface of the morphologically intact endothelium. The contact between the ULvWF and GPIb leads to platelet activation. Similarly as in (A). adherent platelets release ADP and TxA2 which lead to activation of more platelets and ultimately three-dimensional clot formation and retraction. (D) Neutrophil extracellular trap (NET)-mediated platelet thrombosis. In the presence of pathogens, platelets and neutrophils collaborate to form NETs that are highly thrombotic and resistant to tissue plasminogen activator-mediated fibrinolysis. Generally, the size of a thrombus formation is regulated by limiting the level of clot propagation. The endothelium plays a major role by expressing certain thrombo-regulators (NO, prostacyclins, and CD39-ectonucleotidase) that prevent the clot from spreading. Clots are resolved by initiation of fibrinolysis through generation of plasmin from plasminogen and breakdown of fibrin. Depending on the vessel in which the clot forms, the thrombus can be platelet-rich (white clots, form in arteries) or fibrin- and red blood cell-rich (red clots, form in veins). Each step in thrombus formation can lead to uncontrolled clot formation that can result in arterial or venous thrombosis, ultimately presenting a risk for MI, stroke or VTE.

Arterial thrombosis

Mechanism of activation

Arterial thrombosis results from clot formation in the setting of atherosclerotic plaque rupture, leading to platelet aggregation, thrombus formation, vessel occlusion and possible MI or ischemic stroke. Thus, arterial thrombi are treated with therapies that target platelet activation and aggregation (Figure 1). In recent years, there has been an increase in the prevalence of angina and no obstructive coronary artery disease (ANOCA) or MI and non-obstructed coronary arteries (MINOCA). The underlining mechanism or optimal antithrombotic therapies are unclear, however it has been proposed that in these cases arterial thrombosis occurs in the absence of plaque rapture.

Pathophysiologically, thrombus formation has been linked to the secretion of protein disulfide isomerases (PDI, ERp5, ERp57) by platelets and activated endothelial cells.¹ PDIs can react with nitric oxide (NO) and reactive oxygen species (ROS) contributing to the initiation of thrombus formation.^1,3 PDIs are known to contribute to thrombus formation by activating TF and increasing generation of fibrin.³ In addition, PDI function is essential for platelet aggregation⁴ and regulates the rapid increase in thrombin production on the surface of activated platelets.⁵

Healthy arterial endothelial cells limit thrombosis by releasing NO and prostacyclins leading to control of clot size. Endothelial cells also express CD39, an enzyme that hydrolyzes ADP to AMP. This process of ADP elimination prevents further pro-thrombotic platelet activation. In addition to CD39, endothelial cells express CD73, an enzyme that converts AMP to adenosine. Adenosine, in turn, limits thrombosis by blocking platelet activation and acting as an anti-inflammatory mediator through its receptors.⁶ Endothelial cells located around atherosclerotic lesions lose their ability to regulate thrombotic propagation due to compromised release of NO and prostacyclin.^7,8

Established and evolving risk factors for arterial thrombosis

Various factors increase the risk of developing arterial thrombosis. Classically, the risk factors implicated in thrombosis have been hypertension, high levels of low-density lipoprotein (LDL)-cholesterol, and smoking. However, diabetes, pregnancy, age, chemotherapeutics, infectious burden, human immunodeficiency virus (HIV) and high vWF in plasma also pose a risk. Recently, studies report additional risk factors that may contribute to thrombosis. Low activity of ADAMTS13, an enzyme that cleaves vWF multimers, was associated with an increased risk of ischemic stroke and improved the accuracy of risk predictions for ischemic stroke beyond traditional risk factors.⁹ Stillbirth and loss of multiple pregnancies increase the risk of ischemic stroke and MI.¹⁰ A large retrospective cohort study showed an association between VT /anticoagulation therapy and the risk for arterial thrombosis (in ∼1.5% of all patients).¹¹

Systemic lupus erythematosus (SLE) is now a well-recognized risk for thrombosis. The incidence of thrombosis in patients with SLE is 25–50-fold higher than in the general population. In patients with recent onset of SLE, the incidence rate for thrombosis is 31/1000 patients per year. Risk for VT in SLE patients is higher than for arterial thrombosis and it is mostly independent from lupus anticoagulant therapy.¹² Arterial (2.4%) and VT(3.6%) are also increased in paediatric patients with SLE.¹³ Factors contributing to thrombosis in this paediatric population are vasculitis, avascular necrosis, or antiphospholipid antibody.¹³

Hormone therapy (HT) is also known to be associated with increased risk of both arterial (MI and stroke) and VT, particularly, in the first years of therapy.^14,15 The origin for thrombosis in this case is multifactorial. A recent cross-sectional study of 2787 post-menopausal women on HT suggests that elevated levels of estradiol and sex hormone binding globulin were associated with elevated levels of C-reactive protein (CRP) and lower levels of TF pathway inhibitor, both of which contribute to the prothrombotic events in HT users.¹⁶

Hereditary thrombophilia is another risk factor associated with a slight increase in risk of arterial thromboembolism. Presence of cardiovascular risk factors such as diabetes mellitus, however, may lead to stronger risk association between thrombophilia and arterial thrombosis. It would be beneficial to establish the thrombophilia risk in populations with high prevalence of traditional cardiovascular risk factors.¹⁷

Venous thrombosis

Mechanism of activation

The mechanisms mediating VT are distinct and venous thrombi contain an abundance of red blood cells and fibrin in addition to platelets (Figure 1) and are typically treated with drugs targeting proteins mediating coagulation. The balance between blood flow and composition together with venous endothelial health must be altered for a venous clot to form. Venous thrombi are not tightly adherent to the endothelium and can easily dislodge, leading to distant vessel occlusive disease such as PE.

Mechanistically, it has been proposed that the venous endothelium first becomes activated and, as a result of inflammation, surface selectin expression increases¹⁸ and autophagy regulates vWF secretion.¹⁹ Endothelial activation causes attachment of platelets and leukocytes. Attached leukocytes become activated and initiate expression of TF that, in turn, activates the coagulation cascade. The protective anticoagulant effect of the endothelial surface is negated by low blood flow.¹⁸ Low blood flow may lead to hypoxic conditions that are described to increase the expression of endothelial adhesion molecules and consequent attachment of leukocytes. More recent studies have shown that red blood cells play a role clot formation.²⁰ Erythrocytes in the clot change shape from double concave to polyhedral allowing for tight packaging and resistance to fibrinolysis.²⁰

Established and evolving risk factors for venous thrombosis

An array of different factors contributes to the risk of VT. It is notable that women and men of all ages, races, and ethnicities are at risk for VTE. Age and obesity are important risk factors for VTE and, after the age of 40, the risk for VTE doubles with each decade of life. Prior episodes of VTE and atherothrombosis also contribute to the increased risk. A recent study demonstrated that MI is associated with an increased risk of transient VTE and PE independent of traditional atherosclerotic risk factors.²¹

Fibrin degradation products, as measured by plasma D-dimer levels are associated with acute VTE. The concentration of D-dimer remains increased in VTE patients even after treatment. Higher basal level of plasma D-dimer is a strong, long-term risk marker for first VTE.²²

An increase in oestrogen levels due to pregnancy, obesity, or, oral contraceptive use is also a risk factor for VTE. Elevated levels of oestrogen lead to a rise in coagulation factors which are crucial to prevent blood loss during child birth but concomitantly increase risk for deep vein thrombosis (DVT).¹⁸ The risk is further increased in overweight or obese women who use oral contraceptives.¹⁸ Similarly, the risk for cerebral VT (CVT) was associated with an increased body mass index (BMI).²³ The dose-dependent association between BMI and CVT was not found in women who did not take oral contraceptives.²³ Multiple pregnancies and older maternal age are risk factors for VTE. Immobility due to bed rest, long distance travel or surgery is associated with a higher risk of VTE.

Anaemia may be an important risk factor for CVT. A stronger association between anaemia and CVT was found in men as compared to women and haemoglobin was inversely associated with CVT.²⁴ It has been suggested that endothelial hypoxia may be responsible for the increased VT.¹⁸

Advanced stages of cancer as well as chemotherapy treatment are also associated with increased risk for VTE. Cancer patients have an approximately four-fold increased risk of VTE as compared with the general population, and cancer patients with VTE have reduced survival. This VTE risk is notable for pancreatic and cerebral cancer followed by stomach and bladder cancer.²⁵ Certain haematological malignancies such as acute leukaemia are also associated with a high incidence of VTE.²⁶ Evidence shows that TF-containing microvesicles secreted by malignant tissues are able to activate platelets via thrombin and lead to increased coagulability and VTE.²⁷ The process could be further complicated by increases in platelet and leukocyte number, soluble P-selectin, and D-dimer all contributing to platelet pro-thrombotic properties and VTE.²⁸

In addition to environmental and acquired risk factors, there are particular genetic mutations that also increase the risk for VTE. A classic example is Factor V Leiden mutation that leads to hypercoagulability. This particular variation leads to the inability of activated protein C to degrade and inactivate factor V. Other genetic risk factors that increase risk for VTE are mutations in prothrombin (G20210A) and fibrinogen (C10034T)¹⁸ and mutations in proteins mediating anticoagulation. The last are mutations in antithrombin, protein C and protein S. In addition, non-O blood types have also been associated with increased risk for VTE.²⁹ The mechanism by which these blood group increases VTE risk is not well understood; however, it has been proposed that people with blood groups different than O have higher levels of factor VIII, protein C, and vWF.²⁹

Using meta-analyses and genome-wide association studies (GWAS), new mutations have been suggested to increase thrombotic risk. An alanine to cytosine mutation (A1298C) in the methylenetetrahydrofolate reductase (MTHFR) gene is associated with DVT.³⁰ The 4G/5G polymorphism in the plasminogen activator inhibitor (PAI-1, responsible for fibrinolysis) is associated with increased risk of DVT³⁰ and coronary artery disease risk.³¹ Recently, trauma exposure and post-traumatic stress disorder have also been linked to MI and stroke in women and lead to a two-fold increased risk for VTE.³²

Mean platelet volume and thrombosis

As previously discussed, platelets vary in size. Large platelets are more reactive, have greater prothrombotic potential, and are more resistant to inhibition with aspirin and clopidogrel (P2Y12). Large platelets are presumed to be immature and their number increases when there is a rise in factors affecting platelet turnover. Recent studies have shown that elevated mean platelet volume (MPV) is associated with increased risk for DVT and MI. A meta-analysis of cohort-studies suggests that MPV might be a useful prognostic marker in patients with cardiovascular disease (CVD). Elevated MPV in these analyses was significantly higher in patients with acute MI and or coronary angioplasty.³³ Elevated MPV is also identified as a predictor for VTE, particularly VTE of unprovoked origin.³⁴

Trends in anti-thrombotic and -coagulant therapies

Although beyond the scope of this review, there has been great progress in the development of antithrombotic therapies for the treatment of acute and chronic CVD, atrial fibrillation, and VT. For decades aspirin, the irreversible inhibitor of platelet cyclooxygenase activity, has provided effective secondary prevention of arterial thromboembolic events for a wide range of arterial occlusive diseases with recent data also suggesting a role in venous disease.³⁵ P2Y12 inhibitors are valuable treatment options for many patients. Adjunctive antiplatelet or anticoagulant therapies, such as the PAR-1 inhibitor, vorapaxar, or the FXa inhibitor, rivaroxaban, respectively, have also been recently introduced. The mechanism behind rivaroxaban benefit in acute coronary syndrome (ACS) is not clear, as rivaroxaban does not interfere with platelet activation by classical agonists (Brodde and Kehrel, unpublished).

Adherence to the European Society of Cardiology guidelines on antithrombotic management has led to significantly better outcomes of cardiovascular events, including those related to mortality and bleeding.³⁶ Additionally, the use of clinical tools that consider both patient genetics and evidence-based medicine have led to better clinical outcomes, particularly when using anticoagulants. Unfortunately, there are no long-term risk assessment tools that can aid the prevention of reoccurring events in patients with previous MI.³⁷

Emerging mechanisms: pathogens, immunothrombosis, and platelets

There is rising and convincing evidence that various pathogenic infections pose significant risk for thrombosis that includes arterial thrombosis, VTE, and atherosclerosis.^38–40 In response to blood borne pathogens and tissue damage there is a coordinated intravascular coagulation recently termed ‘immunothrombosis’.⁴¹ This allows platelets and immune cells to form a physical barrier of confinement to prevent dissemination of pathogens and to activate the immune system.⁴¹ Platelets, in turn, carry the transcripts for all pathogen responsive toll-like receptors.⁴² During certain bacterial infections, platelets are able to induce prothrombotic events, secrete cytokines, chemokines, and antimicrobial peptides, leading to sequestration and destruction of bacteria.² Platelets are also known to engulf certain viruses such as HIV, hepatitis C virus (HCV) and encephalomyocarditis (EMCV)⁴³ and to interact with bacteria such as Staphylococcusaureus⁴⁴ and Staphylococcuspneumoniae.⁴⁵

Viruses such as HIV, HCV, and Dengue are also known to cause elevated levels of thrombosis. It is unclear if the thrombosis during viral infections is reactive or if it serves a similar function as in bacterial infections. Immune cells predominantly contribute to prothrombotic risk during immunothrombosis. Monocytes are known to carry the pro-coagulant TF and microvesicles with activated TF are observed during flu infection.⁴⁶ Neutrophils, in turn, are able to release their DNA forming highly prothrombotic neutrophil extracellular traps (NETs) in a process called netosis. The released DNA enables neutrophils to trap and neutralize bacteria and to mediate the outcome of viral infection with poxvirus.⁴⁷ Although the process of netosis is beneficial for a successful immune response, it is also a highly prothrombotic process. NETs are known to increase thrombin levels, activate platelets and coagulation.⁴⁸ Further, netosis significantly increases the risk for DVT.⁴⁸ Elevated levels of DNA and chromatin in the circulation are independently associated with severe atherosclerosis and thrombosis.⁴⁹ Clots removed from the coronary arteries of patients with ST-elevation ACS show that neutrophils undergo netosis at the culprit lesion site. Coronary NET burden and DNase activity are shown to be predictors of myocardial infarct size.⁵⁰ Thus, immunothrombosis may be an efficient way of enabling the immune system to fight diverse infections; however, it significantly contributes to thrombosis and the overall cardiovascular burden.

Thrombosis and platelets: future implications

Despite major advances in understanding the mechanistic pathways of platelet function and the interaction of coagulation and thrombosis, challenges in the treatment of vascular occlusive diseases continue to persist. This is due to the complexity of these diverse diseases and the impact of immunological and inflammatory processes on haemostasis and thrombosis. However, the growing understanding of these processes is also opening up new avenues for the use of antithrombotics. Despite the successful development of new classes of drugs, there is a great-unmet clinical need for developing more effective and safe antithrombotic agents. These drugs may incorporate important mechanistic platelet and coagulation-based discoveries, while others will utilize targets established by traditional herbal medicines (Table2). Additionally, pharmacological advances have greatly impacted thrombotic outcomes but have led to the unwanted side effect of bleeding. Compounding this issue is the lack of clear biomarkers to balance risk and benefit of treatments, particularly when used in combination. Future studies are necessary to evaluate the balance of thrombotic risk, impact of non-vascular diseases, progressive treatments and side effects.

Table 1

Summary of updated risk factors for arterial or venous thrombosis

Arterial thrombosis-risk factors		Venous thrombosis-risk factors
Atderosclerosis51	Hypertension51	Age51	Obesity51
LDL51	Smoking51	Atherothrombotic events51	Prior VTE episodes51
Chemotherapeutics51	Diabetes51	Non-O blood type29	Oestrogen18
Infections39,40	Pregnancy/multiple pregnancies10	Pregnancy/multiple pregnancies10	Oral contraceptives18,51
Thrombophilia17	SLE12,13	Older maternal age51	Immobility/long distance travel51
Stillbirth10	Hormone therapy14,15,18	Anaemia24	Cancer/chemotherapy24–28
Low levels of ADAMTS139	MPV33	PTSD32 MPV34	Mutations in Factor V (Leiden), prothrombin and fibrinogen18,51

PTSD, post-traumatic stress disorder; MPV, mean platelet volume; VTE, venous thromboembolism; LDL, low-density lipoprotein; SLE, Systemic lupus erythematosus; ADAMTS13, a disintegrin and metalloproteinase with a thrombospondin type 1 motif, member 13.

Table 2

Existing and novel antithrombotic agents in use and under development

Antithrombotic agents	Specific Class	Subclasses/Specific Therapies
Platelet Inhibitors	P2Y12 Receptor Blockade	Clopidogrel Prasugrel Ticagrelor Cangrelor
	Glycoprotein IIb/IIIa inhibition	Abciximab Eptifibatide Tirofiban
	Cyclooxygenase inhibition	Aspirin
	Thrombin Receptor Antagonists (PAR-1)	Vorapaxar Atopaxar
	Phosphodiesterase II inhibitor	Cilostazol
	Adenosine uptake blockade	Dipyridamole
Anticoagulants	Vitamin K Antagonists	Coumarin Acenocoumarol
	Antithrombin III activation	Heparin Low molecular weight heparin
	Synthetic pentasaccharide inhibitors of factor Xa	Fondaparinux Idraparinux
	Thrombin inhibitor	Dabigatran
	Factor Xa inhibitor	Rivaroxaban Apixaban Edoxaban
Novel Antithrombotic and Antiplatelet Targets	Synergistic inhibition of P2Y1 and P2Y12 ADP Receptors
	Immunotherapy	Complement Innate immunity
	Platelet Inhibitors: Targets of tumor-metastasis	Aspirin
	Platelet Inhibitors: Targets of tumor-metastasis	Platelet-derived exosomes
	Nucleotide based inhibitors
	Second generation selective thrombin inhibitors
	Alternative 2nd generation GPIIb/IIIa oral inhibitors
	Medicinal herbals	Allium sativum (garlic) Zingiber officiante (Ginger) Curcuma longa (Turmeric)

The authors do hereby declare that all illustrations and figures in the manuscript are entirely original and do not require reprint permission.

Acknowledgements

We apologize to all whose work was not cited due to limited space and limited number of citations. This work was supported by National Institute of Health grant U01HL126495 (to J.E.F.) and by American Heart Association grant 16SDG30450001 (to M.K.).

Conflict of interest: none declared.