COVID-19 and Thrombosis: Pathophysiological Mechanisms and Therapeutic Update

Abstract

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is a pro-inflammatory and prothrombogenic virus with a high mutagenic profile, which produces active infection of variable duration in various organs and systems, and it has been observed that patients who have already suffered from the disease, especially in its more severe forms such as bilateral pneumonia or respiratory distress, present symptoms and signs of chronic multi-organ involvement. However, little is known about the molecular mechanisms that generate endothelial damage (chronic reactive endotheliitis) and subsequent thrombosis in SARS-CoV-2 infection are still not sufficiently elucidated, and in this chapter, we explore these mechanisms and therapeutic options to reduce prothrombosis and multiple vascular involvement that cause morbidity and mortality in this disease. In particular, we will evaluate heparin doses according to the stage of infection and its correlation with improved survival.

1. Introduction

SARS-CoV-2 causing coronavirus disease (COVID-19) is a proinflammatory and prothrombogenic virus with a high mutation rate, producing active infection of variable duration in various organs and systems [1, 2]. The disease is highly heterogeneous in its manifestations and, although in most cases resolve asymptomatically or mildly, it can lead to severe symptoms and even death [3, 4]. In hospitalised patients with COVID-19, due to the procoagulant state and the increased risk of thromboembolic events, the use of anticoagulation for prophylactic purposes is recommended [5], and heparin is of great benefit in the treatment and prevention of venous and arterial micro- and macro-thrombosis [6]. Given reports of excess thrombotic risk, dose-escalating anticoagulation strategies have been incorporated into some COVID-19 clinical guidelines [7]. However, the effectiveness and safety of therapeutic/intermediate/prophylactic anticoagulation doses in COVID-19 are uncertain and remain under study [8].

2. Pathophysiology of the disease

2.1 Angiotensin-converting enzyme 2 receptor interaction

Angiotensin-converting enzyme 2 (ACE2) has been identified as a functional receptor for coronaviruses [6]. The ACE2 receptor is present throughout multiple cellular organs such as the heart, kidney, lungs, as well as in the central and peripheral nervous system [9]. As with other respiratory viruses, respiratory tract symptoms are the most common. The gradient of receptor expression has been directly correlated with the ability of SARS-CoV-2 to infect cells throughout the respiratory tract. The highest concentration of ACE2 receptors is found in the hair cells of the nasal mucosa and is 80% lower in trachea, bronchi and lung tissue [10]. SARS-CoV-2 binds to the transmembrane ACE2 protein to enter type II pneumocytes; due to this tropism, SARS-CoV-2 can interact with a large area of the pulmonary microvasculature. In addition, it can infect pericytes and perivascular cells on the surface of microvessels. ACE2 receptor expression on endothelial cells increases their vulnerability to SARS-CoV-2 binding, causing infection and subsequent vascular injury, dysfunction and endotheliitis [6]. Likewise, interaction with ACE2 receptors in the peripheral nervous system may contribute to the development of myopathies and neuropathies [9].

2.2 COVID-19-associated coagulopathy (CAC)

Most critically ill patients with COVID-19 present with isolated respiratory failure, usually acute respiratory distress syndrome (ARDS). In COVID-19 deaths, the predominant lung damage is diffuse alveolar damage, which includes hyaline membrane formation, capillary congestion, inflammation and pneumocyte necrosis [4, 11]. In addition, fibrin-platelet thrombi in small arterial vessels are also identified in many of these cases [4]. However, about 20–30% of these patients have multi-organ involvement. Among the extra respiratory complications present in COVID-19 are vascular alterations, among which coagulopathies are of great relevance [2]. In general, COVID-19-associated coagulopathy (CAC) is characterised by moderate thrombocytopenia, mildly increased prothrombin time (PT), elevated D-dimer and fibrinogen levels [4].

2.2.1 Chronic reactive endotheliitis and von Willebrand factor

There are different prothrombotic mechanisms in SARS-CoV-2 infection. It has been established that the virus induces oxidative stress at the endothelial level, causing the release of von Willebrand factor (vWF) multimers causing hypercoagulability, despite the thrombopenia caused by the virus, leading to a state of prothrombosis through increased thrombin and D-dimer levels [4, 12, 13, 14]. Thus, vWF may be implicated in CAC, due to its direct relationship with homeostasis, inflammation and endothelial cell activation and damage [4]. vWF is a large multimeric glycoprotein whose gene is located on the short arm of chromosome 12 at position 13.3 and has a length of 180 kb and 52 exons [15]. The VWF gene belongs to the endogenous ligand gene family and genetic differences between individuals are associated with vWF levels. This includes polymorphisms in the 5′ homeostatic factor regulatory region, which contributes to the level of vWF present in plasma and, consequently, to the risk of thrombotic events [15]. This protein is present in blood plasma, platelet-α-granules, subendothelial connective tissue and endothelium [4, 15]. vWF is synthesised and stored primarily in endothelial cells, megakaryocytes and platelet precursors in the bone marrow [15]. Upon synthesis in endothelial cells, the sequence of the vWF propeptide acts by aligning the 2 units of the molecule to ensure optimal multimerisation. Post-translational modifications involve the removal of the propeptide, glycosylation and the addition of blood group determinants, and then, a multitude of ultra-long vWF (UL-vWF) molecules are synthesised. When endothelial cells are activated, UL-vWF molecules are released and can remain free in plasma or bound to the endothelial surface. UL-vWF molecules exhibit greater prothrombotic activity than smaller vWF multiples. Simultaneously to the release of UL-vWF molecules, ADAMTS-13 (thrombospondin type 1 metalloprotease, member 13) cleaves these molecules into smaller multimers, to stop unwanted thrombus formation [4, 15]. This protein has a functional duality, as it is involved in both homeostasis and thrombosis [4, 15].

vWF plays a major role in primary haemostasis. When damage to the vascular wall occurs, the subendothelium to which vWF is bound is exposed. This protein interacts with platelets and promotes the recruitment of circulating platelets to the site of damage. Platelet-vWF binding is an adhesive interaction capable of binding platelets to the endothelial surface. Although the binding between platelets and vWF is unstable, it promotes a stronger and prolonged adhesion to the endothelium, which is mediated primarily by vWF [4, 15]. As for the process of secondary haemostasis, vWF also plays an important role and that process involves coagulation factors and the coagulation cascade to produce fibrin networks in areas of vascular damage. vWF promotes the process of secondary haemostasis by two mechanisms: Firstly, vWF acts as a transporter of coagulation factor VIII, stabilising it and extending its half-life in plasma. Secondly, it releases and concentrates factor VIII at the site of endothelial damage. Factor VIII is a coagulation factor that, when activated, complements other factors to generate fibrin [4].

During the inflammatory process, different mediators are released as inflammatory molecules activate endothelial cells to release their contents, such as vWF. UL-vWF molecules that remain attached to the cell surface will bind to platelets and serve as a surface to interact with leukocytes. Inflammation also promotes association between vWF molecules, leading to an increase in platelet adhesiveness and a decrease in ADAMTS-13 cleavage. In addition, high-density lipoprotein (HDL) levels decrease during the inflammatory process. HDL plays a key role in preventing the association of vWF molecules with inflammation, as well as decreasing the risk of thrombus under normal circumstances. Increased release of vWF can induce a prothrombotic state [4], which is a pathological state of the haemostatic process. Thrombi are composed of numerous elements including endothelial cells, plasma, proteins and alterations in haemodynamic stress [15]. There is evidence of an association between increased levels of vWF and increased risk of thrombosis [4, 16]; therefore during the inflammatory process, there is an increased risk of thrombosis due to the imbalance in which vWF level and activity are elevated due to over-activation of endothelial cells.

The increase in vWF levels in COVID-19 could be due to the release of this molecule from pulmonary endothelial cells as a result of the pathophysiological process of COVID-19 itself. Infection of endothelial cells by SARS-CoV-2 or their activation in response to inflammatory mediators results in the release of prothrombotic factors, such as vWF. vWF either binds to endothelial cells or circulates in plasma to promote platelet aggregation and thrombus formation (Figure 1) [4].

Figure 1.

Adapted from Mei et al., J. Appl. Lab. Med. 2021: Mechanism and characteristics of CAC in mild and severe cases. (A) CAC in mild cases. Localised infection and minimal systemic inflammation increase endothelial cell activation. Infection and inflammation remain well regulated. HDL and ADAMTS-13 mechanisms remain largely unchanged, with only a slight increase in thrombotic events. (B) CAC in severe cases. Infection and inflammation are deregulated, leading to an extremely high level of activated endothelial cells. In addition, HDL and ADAMTS-13 levels are decreased, leading to a much greater increase in thrombotic events. (C) Thrombotic thrombocytopenic Purpura (TTP). In TTP, ADAMTS-13 activity levels are significantly lower than in CAC. TTP leads to the elevated levels of UL-vWF. Consequently, platelet-binding levels increase and, consequently, the risk of thrombosis increases. Addendum: In the NIH COVID-19 treatment guidelines panel’s statement on anticoagulation in hospitalized patients with COVID-19, last updated the 5th of January 2022, the panel recommends using therapeutic-dose heparin for patients who have a D-dimer above the upper limit of normal, require low-flow oxygen, and have no increased bleeding risk. They recommend continue therapeutic-dose heparin treatment for 14 days or until hospital discharge, whichever comes first [17].

2.2.2 Other agents: immunocomplexes, lupus anticoagulants, β-2 glycoprotein 1 (B2GPI) and cytokines

The proinflammatory and prothrombotic state at the endothelial level in the microvasculature may remain in some patients due to immunocomplex formation, causing chronic reactive endotheliitis with multiple vascular involvement, especially in the lungs and central nervous system (Figure 2) [13].

Figure 2.

Neurological lesions mediated by chronic reactive endotheliitis and multiple vascular involvement.

On the one hand, the presence of direct neurological damage by SARS-CoV-2 through the advancement into the CNS from the periphery via retrograde and transsynaptic neuronal transport, especially via vagus nerve afferent pathways, has been demonstrated. Supported by the increasing findings that SARS CoV-2 infects cells in the gastrointestinal tract, the neuroinvasive potential could even encompass the enteric nervous system [18].

On the other hand, the pivotal role of lupus anticoagulants (LA) in the thrombogenesis of SARS-CoV-2 has also been observed [19, 20] and an increased prothrombin time has been reported in COVID-19 patients. This may be indicative of a coagulation factor deficiency or the presence of an inhibitor, either specific such as the factor VIII antibody or non-specific like LA [20]. The β-2 glycoprotein 1 (B2GPI), involved in thrombogenesis, promotes LA activity, thereby stimulating platelet adhesion, tissue factor release and subsequent activation of the coagulation cascade, leading to an often irreversible prothrombotic state in advanced stages [19]. Ultimately, cytokine storm and immune abnormalities also contribute to the inflammatory process. These immune abnormalities have been associated with the severity of COVID-19 and are considered a cause of mortality in this disease [3, 11, 19]. A correlation between severe COVID-19 and abnormalities in circulating immune cells has been described [3]. In severe cases, COVID-19 can trigger an excessive immune response known as a cytokine storm, which is potentially fatal. It is characterised by over-activation of immune cells and excessive production of pro-inflammatory cytokines and chemical mediators [11]. Cytokine storm also amplifies platelet production, leading to an increased formation of disseminated microthrombi in various vascular territories, and is directly involved in the pathogenesis of thrombosis in this disease [19].

2.2.3 Immune response and T lymphocytes

In viral infections, “innocent bystander” activation of CD8+ T cells occurs, which consists of activation of CD8+ memory cells independent of T-cell receptor (TCR) stimulation. Active lymphocytes can migrate to the site of infection and kill infected cells. This form of early response, which begins before symptoms develop, may be associated with disease resolution and avoid progression to severe disease. However, persistent bystander activation of CD8+ T cells has been associated with inflammatory pathology in both chronic infections and autoimmune processes. Therefore, it is possible that persistent T cell activation in severe COVID-19 may influence the development of lung pathology and autoimmune manifestations observed in this disease [3]. On the other hand, patients with more severe clinical manifestations have higher-circulating TNF-α and IL-6, and higher gene expression of proinflammatory pathways [3, 11].

3. Therapeutic proposals: antithrombotic treatment

The recognition of thrombosis as a key contributor to clinical deterioration and death has led to a worldwide interest in the study of optimal antithrombotic treatment doses for patients. Clinical trials have shown that the efficacy and safety of these treatments vary according to the time course of the disease [21]. For this reason, treatment should be started early in the Emergency Department in all hospitalised patients and assessed according to thrombotic and haemorrhagic risk factors [22]. Haematological and coagulation parameters (e.g. D-dimer, prothrombin time, platelet count, fibrinogen) are commonly measured. Evidence shows that elevated D-dimer values correlate positively with disease severity and prognosis [23]. However, at present, there is insufficient evidence to recommend for or against the use of these data to guide management decisions [7, 24]. Dosing recommendations are dynamic and have changed throughout the pandemic. The results of recent scientific publications have generated controversy about the best strategy for antithrombotic prophylaxis and treatment, which is reflected in the variability of consensus recommendations published by different scientific organisations and societies [25]. In terms of treatment, we differentiate between non-hospitalised and hospitalised patients.

3.1 Non-hospitalised patients

For non-hospitalised patients with COVID-19, antithrombotic therapy or antiplatelet therapy for the prevention of venous thromboembolism (VTE) or arterial thrombosis should not be initiated unless the patient has other indications for therapy (prior PTE or DVT, recent surgery, trauma or immobilisation) or is participating in a clinical trial. Patients with COVID-19 receiving warfarin/acenocoumarol who are isolated and therefore unable to proceed with standardised monitoring of anticoagulation levels may be candidates for switching to direct oral anticoagulation therapy [26].

3.2 Patients hospitalised for COVID-19

All patients hospitalised with COVID-19 should receive prophylactic dose antithrombotic therapy with low-molecular-weight heparin unless contraindicated (e.g. active bleeding or severe thrombocytopenia) [7, 26]. Low-molecular-weight (LMW) heparin is preferred, but unfractionated heparin may be used if LMW heparin is not available or if renal function is severely impaired (creatinine clearance <30 ml/min). Individuals with a history of heparin-induced thrombocytopenia (HIT) or active HIT should not receive low-molecular-weight heparin or unfractionated heparin; instead, Fondaparinux is advised. Aspirin is not indicated outside the usual standard indications [7, 26]. Patients chronically receiving anticoagulant or antiplatelet therapies for underlying conditions should continue these medications unless they have a bleeding or contraindication to them. In patients on chronic anticoagulant therapy, heparin is preferable to oral anticoagulants because of its shorter half-life, the possibility of intravenous or subcutaneous administration, and the reduced presence of drug interactions [26]. In addition, it has antithrombotic, anti-inflammatory and possibly antiviral properties [27].

In the case of suspected venous thromboembolic disease in patients with COVID-19, the recommendations proposed by the scientific societies for any patient with suspected and confirmed thrombotic disease should be followed [7, 26]. For most hospitalised patients with COVID-19, prophylactic dosing is supported. However, following the publication of several large and well-conducted randomised trials, the relative benefits of prophylactic, intermediate or therapeutic dosing continue to generate debate [28]. This is partly due to the increased risk of bleeding attributed to intermediate and therapeutic regimens [29].

Within inpatients, we differentiate those with moderate/non-critical illness from those with critical illness.

3.2.1 Moderately ill or non-critically ill patients

Patients with moderate/non-critical illness are those with clinical features that would normally result in admission to an inpatient ward without the need for advanced clinical support in intensive care unit (ICU). Examples include patients with mild to moderate dyspnoea or hypoxia [5, 21]. In recent months, several randomised clinical trials (ACTIV-4, REMAP-CAP and ATTACC, RAPID and HEP COVID) have shown that in non-critically ill hospitalised patients with COVID-19, heparin at therapeutic doses may be beneficial, with a high probability of reducing the need for organ support and progression to intubation and death [26, 30, 31]. Other trials, however (ACTION, BEMICOP), found no benefit in therapeutic versus prophylactic dosing [32, 33]. The multi-platform randomised clinical trial (ACTIV-4, REMAP-CAP and ATTACC) showed that in non-critically ill hospitalised patients with COVID-19, an initial strategy of therapeutic anticoagulation with heparin increased the likelihood of survival to hospital discharge with a reduced need for ICU-level organ support at 21 days compared with usual care thromboprophylaxis. Therapeutic dose anticoagulation was beneficial regardless of the patient’s baseline D-dimer level. Therapeutic anticoagulation was administered according to local protocols for the treatment of acute venous thromboembolism for up to 14 days or until recovery. The need for organ support was defined as the need for high-flow nasal oxygen, invasive or non-invasive mechanical ventilation, vasopressor therapy or extracorporeal membrane oxygenation (ECMO) [30]. A subsequent randomised clinical trial (RAPID) demonstrated that the use of therapeutic heparin in moderately ill patients with COVID-19 and increased D-dimer levels (above the upper limit set by the local laboratory) were not associated with a significant reduction in the primary composite outcome of death, non-invasive or invasive mechanical ventilation, or ICU admission. However, it decreased the probability of death at 28 days and showed a low risk of haemorrhage [31]. Finally, in the randomised HEP-COVID clinical trial, therapeutic-dose LMWH in hospitalised COVID-19 patients with D-dimer levels at least four times the upper limit of normal reduced the outcome of thromboembolism and death at 30 days compared with standard heparin thromboprophylaxis, without increasing the risk of major bleeding [34]. The effectiveness of anticoagulation appears to depend on the type of anticoagulant; for example, the Coronavirus Anticoagulation Trial (ACTION) used 15–20 mg rivaroxaban in 94% of patients assigned to therapeutic anticoagulation and found no benefit [32]. Finally, the randomised controlled clinical trial BEMICOP randomised adult patients with non-critical COVID and elevated D-dimer to receive heparin at therapeutic versus prophylactic doses for 10 days; however, the use of a short course of heparin at therapeutic doses did not improve the primary outcome (death, death in the intensive care unit), intensive care unit admission, need for mechanical ventilation support, the development of moderate/severe acute respiratory distress and venous or arterial thrombosis) over the subsequent 10 days compared with the use of heparin at a prophylactic dose [33].

In view of the discordance in different clinical trials, recent meta-analyses have attempted to clarify whether higher dosing (intermediate or therapeutic) could be of benefit over prophylactic dosing. Both groups of studies, the ones that included cohort studies and clinical trials [35], as well as those that included only randomised clinical trials [24, 28], have shown that higher anticoagulation dose (intermediate or therapeutic) is not associated with lower in-hospital mortality or incidence of thrombotic events, but increases the risk of bleeding events. There is currently insufficient evidence of survival benefit of therapeutic or intermediate dose anticoagulation compared with prophylactic dose anticoagulation in hospitalised patients with COVID-19.

The meta-analysis by Jorda et al. [24] included 10 randomised controlled open-label trials with a total of 5753 patients. The risk of death was similar between therapeutic dose versus prophylactic dose anticoagulation (RR 0.92, 95% CI 0.69–1.21, P = 0.54) and between intermediate dose versus prophylactic dose anticoagulation (RR 1.01, 95% CI 0.63–1.61, P = 0.98). In patients with markedly increased D-dimer levels, higher-dose anticoagulation was also not associated with a lower risk of death compared with prophylactic dose anticoagulation (RR 0.86, 95% CI 0.64–1.16, P = 0.34). In the meta-analysis by Ortega Paz L et al. [28] including seven randomised controlled trials with 5154 patients, prophylactic dose-escalated anticoagulation was not associated with a reduction in all-cause death [17.8% vs. 18.6%, hazard ratio (HR) 0.96, 95% confidence interval (CI) 0.78–1.18], but was associated with an increase in major haemorrhage.

In conclusion, the literature that has so far been published provides evidence for the use of prophylactic anticoagulation at standard doses over a dose-escalating regimen in routine care for patients hospitalised for COVID-19, irrespective of disease severity. There may be a subgroup of patients with moderate disease where heparin at therapeutic doses may be beneficial; however, further studies are needed to define this subgroup of patients.

3.2.2 Severely ill/critically ill patients

Patients with severe/critical COVID-19-related illness are those requiring respiratory or cardiovascular support (oxygen via high-flow nasal cannula, non-invasive or invasive mechanical ventilation, extracorporeal life support, vasopressors or inotropes) usually in an intensive care unit [5]. For thromboprophylaxis in critically ill hospitalised COVID-19 patients, prophylactic dosing is recommended rather than more intensive (intermediate or therapeutic) dosing. Randomised trials published to date have not consistently demonstrated better outcomes with more intensive dosing in critically ill patients, but have found an association with a higher likelihood of side effects (bleeding) and most institutions have adopted prophylactic dosing standards [7]. In the multi-platform ACTIV-4, REMAP-CAP and ATTACC study, a parallel analysis of the same trial mentioned above, in patients with COVID-19 critical illness, empirical anticoagulation at therapeutic doses was not beneficial in this group of patients [8]. Nor did heparin at anticoagulant doses show benefit in the group of critically ill patients in the HEP-COVID study [34]. The randomised clinical trial (INSPIRATION) in critically ill patients with COVID-19 showed no benefit of intermediate-dose heparin either [36]. Patients with COVID-19 who require ECMO or continuous renal replacement therapy or who have catheter or extracorporeal filter thrombosis should be treated according to standard institutional protocols for non-COVID-19 patients [26].

4. Conclusions

COVID-19 is associated with a hypercoagulable state with acute inflammatory changes. Fibrinogen and D-dimer are increased, with a typical mild prolongation of prothrombin time (PT) and modest thrombocytopenia. The pathogenesis of these abnormalities is not fully understood, and there may be many other related factors contributing to the acute inflammatory response to the disease. Routine thromboprophylactic medication is not recommended in outpatients. For thromboprophylaxis in hospitalised patients with COVID-19, prophylactic antithrombotic dosing with low-molecular-weight heparin is recommended in both moderately and critically ill patients. There may be a subgroup of patients with moderate disease where heparin at therapeutic doses could be beneficial in reducing the need for organ support and death; however, further studies are needed to precisely define this subgroup of patients due to heterogeneity of results.

In critically ill patients, randomised trials with intensive dosing have shown no benefit, but an increased risk of bleeding.

Acknowledgments

The authors wish to express their great gratitude to Mrs. Charo Ibáñez González, secretary of the Internal Medicine Department for all these years shared and for her great dedication to the continuous improvement of the Department.

Conflict of interest

The authors declare that they have no conflicts of interest.