Thrombotic Complications in Patients with COVID-19: Pathophysiological Mechanisms, Diagnosis, and Treatment

The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), responsible for the coronavirus disease 2019 (COVID-19), originated in Wuhan, China, in December 2019. SARS-CoV-2 is a new ribonucleic acid (RNA) virus strain from the Coronaviridae family, firstly described by Zhou et al. in February 2020 [1]. In March 2020, COVID-19 was declared a pandemic by the World Health Organization (WHO), involving at the end of May more than 100 countries worldwide, number of cases and deaths rapidly increasing every day. On 25th June 2020, the WHO’s 157th Situation Report of COVID-19 announced that more than 9 million people worldwide had been infected by this virus, which has caused almost 480,000 deaths in barely 6 months. Europe and America have been the most affected regions, accounting for almost 80% of worldwide cases and deaths [2]. Despite SARS-CoV-2 strains having a within-sequence identity of 99.98%, it has been classified in 2 haplogroups (A and B), the first being the predominant worldwide and the second being more prevalent in America and Asia than in Africa and Europe [3]. Nevertheless, it remains unclear whether the severity and outcome of the disease are related to the virus strain.

SARS-CoV-2 fundamentally uses the angiotensin-converting enzyme 2 (ACE2) receptor to enter the cell through the binding with spike-like protein (S protein), although other cell receptors may be alternatively bound. In addition, ACE2 plays a major role in viral proliferation [4]. ACE2 is widely expressed in different organs and tissues, which concurs with the panoply of clinical systemic symptoms and multiorgan dysfunction presented with COVID-19, such as respiratory, renal, hepatic, gastrointestinal, and cardiac damage [5]. Both the down- and upregulation of ACE2 appear to play a major role in heart failure, hypertension, myocardial infarction, and overall cardiovascular disease (CVD) [6]. On those grounds, it is not surprising that patients with underlying CVD have 1.7-fold higher mortality rate caused by COVID-19 than patients with chronic respiratory disease, despite the respiratory system being the principal target of the virus [7].

Emerging evidence points to a strong association between severe clinical presentation of COVID-19 and increased risk of thromboembolism, although the mechanisms behind this are not completely understood [8]. Several risk factors have been described for such an association, such as systemic hyperinflammation driven by the coronavirus infection, hypoxia, and comorbidities associated with hospitalization of critical patients [9]. Whence, understanding the clinical and mechanistic features of COVID-19 severity and mortality is critical to identify and treat patients at the highest thromboembolic risk and mortality. We aim to review the mechanistic relationships between inflammation and thrombosis, discuss the macrovascular and microvascular complications, and summarize the diagnosis, prophylaxis, and treatment of thromboembolism in patients affected by COVID-19.

One-third of patients hospitalized due to severe COVID-19 develop macrovascular thrombotic complications which are associated with an increased risk of in-hospital mortality [43, 44]. These complications include especially venous thromboembolism (VTE), and also stroke and acute myocardial infarction.

Microvascular thrombosis is defined as pathological occlusion of microvessels (arterioles, capillaries, and venules) by platelet- and/or fibrin-rich thrombi [68]. Classically, microvascular thrombosis includes thrombotic microangiopathies (e.g., thrombocytic thrombocytopenic purpura, hemolytic-uremic syndrome) and disseminated intravascular coagulation (DIC) [69]. Microvascular thrombosis is a diagnostic challenge, because (i) microthrombi are difficult to visualize due to their small size (often ≤ 10 μm), (ii) microthrombi often occur only transiently, and (iii) specific biomarkers to detect them are lacking [68]. Clinically, microvessel occlusion leads to ischemia, with the effects ranging from alterations in plasma coagulation markers to severe multiorgan failure [70]. Although in most cases the evidence indicating a causal relationship between microvascular thrombosis and organ failure is difficult to obtain, microvascular thrombosis appears to have a critical importance in the course of COVID-19 [71].

Emerging evidence shows that severe COVID-19 can be complicated with coagulopathy of a prothrombotic character and is associated with poor prognosis [72]. Although many patients with severe COVID-19 present with coagulation abnormalities that mimic thrombotic microangiopathy or DIC, COVID-19-associated coagulopathy has distinct features [73]. The first autopsy series of four patients who died due to severe COVID-19 demonstrated a unique pathological picture, with the presence of diffuse microthrombosis and hemorrhage along with abundant intravascular megakaryocytes in all major organs, including the lungs, heart, kidneys, and liver [74]. Since the gross pulmonary thromboembolism and parenchymal inflammation were absent, this picture has been described as pauci-inflammatory thrombogenic vasculopathy. Another autopsy series of five patients with COVID-19 and acute respiratory distress syndrome (ARDS) demonstrated the lung and skin damage consistent with complement-mediated microvascular injury, whereas the hallmarks of classic ARDS with diffuse alveolar damage and hyaline membranes were not prominent [75]. The distinct respiratory distress syndrome accompanying severe COVID-19 has been termed microvascular COVID-19 lung vessels obstructive thromboinflammatory syndrome (MicroCLOTS) [76] or lung-restricted vascular immunopathology [77].

The pathophysiological mechanisms underlying COVID-19-associated coagulopathy are summarized in Fig. 1. These mechanisms seem to follow Virchow’s triad, including (i) diffuse endothelial cell injury, (ii) abnormal blood flow dynamics, and (iii) uncontrolled platelet activation [71]. SARS-CoV-2 enters the target cells through the ACE2 receptors, which are widely expressed on the surface of lung epithelial cells, and arterial and venous endothelial cells in multiple organs [78]. Virus infection causes direct endothelial injury, dysfunction, and prothrombotic gene expression [79]. In addition to the direct effect, the virus triggers innate immune responses, including the activation of monocytes and complement cascade, responsible for a massive local release of proinflammatory cytokines and aggravation of endothelial injury and microvascular thrombosis, termed immunothrombosis [76]. The major mediators of immunothrombosis include tissue factor (TF) and neutrophils. TF, exposed particularly by activated monocytes and monocyte-derived extracellular vesicles, is a major initiator of thrombosis in vivo [80]. Neutrophils, in turn, stabilize microthrombi via the release of neutrophil extracellular traps (NETs) and neutrophil elastase, which immobilize inflammatory cells and promote intravascular fibrin formation via degradation of TF antagonist, a tissue factor pathway inhibitor [81].

Abnormal flow dynamics results from high blood concentrations of vWF, cell-free DNA, histones, and viral RNA in course of COVID-19, which all cause factor XI activation, thrombin generation, and intravascular fibrin formation, leading to change in blood flow rheology [82]. Platelet activation is induced both indirectly by NETs and directly by virus infection. NETs lead to platelet activation via toll-like receptors on platelets and other cells, therefore integrating pulmonary infection, inflammation, and thrombosis [83, 84]. There is data showing that platelets and megakaryocytes may have receptors for viruses [85‐87] and could be infected by earlier SARS-CoV [88]. Although currently there is no evidence of direct infection of platelets or megakaryocytes by SARS-CoV-2, the abundance of megakaryocytes in the lungs at autopsy seems to co-localize with platelet-rich thrombi [7]. Altogether, the immunothrombosis pathway in the lungs might be a starting point for the systemic propagation of COVID-19 infection, but more research is needed to elucidate this process.

Altogether, the mechanisms underlying microvascular thrombosis in course of COVID-19 may be related to diffuse vascular endothelial cell injury, hyperimmune reaction of the host, and maladaptive platelet aggregation. Clinician awareness of microvascular thrombosis during COVID-19 infection is crucial, since prompt recognition may be life-saving [89].

Thrombosis and thromboembolic complications in patients hospitalized with COVID-19 should be assessed by integrating physical examination findings, patient history, results of laboratory tests reflecting hemostasis, and/or results of focused imaging investigations. Patients who exhibit signs and symptoms of DVT and have risk factors for thrombosis (previous thrombotic events, active malignancy, or current use of hormonal therapy), unexplained and sudden deterioration in respiratory status, and unexplained hypotension and/or tachycardia should be further clinically evaluated to rule out thrombotic complications. In patients with high suspicion of thrombosis, most readily available imaging studies such as bilateral compression ultrasonography of the legs and/or bedside point-of-care cardiac ultrasonography should be used to detect proximal DVT, myocardial thrombus, clot-in-transit to the pulmonary trunk or right ventricular dilatation [92‐95]. However, negative findings of these tests are insufficient to fully exclude thrombosis [92‐95]. If the underlying etiology of persisting hemodynamic and respiratory compromise is not detected by less complex imaging modalities, a CTPA should be undertaken. On the other hand, a normal D-dimer test performed with high-sensitivity assays in this population reasonably rules out VTE, while positive D-dimer test does not confirm thrombosis, and therefore, a further imaging workup is warranted. Table 1

Importantly, elevations of laboratory parameters indicating hypercoagulability at the admission of patients with COVID-19 are associated with an increased incidence of in-hospital thromboembolic events and mortality [44, 52, 72, 96‐98]. For example, in a recent report including 2377 consecutive adults hospitalized with COVID-19, 76% of patients had elevated D-dimer at presentation which had prognostic value for the severe course of COVID-19 infection, including 2-fold higher odds for critical illness, thrombotic events, acute kidney injury, and death. Importantly, the rates of adverse events increased with the magnitude of D-dimer elevation [99]. The initial prothrombotic cascade in patients with COVID-19 is similar as in critical patients with DIC and consumptive coagulopathy; however, there are clear differences between COVID-19-associated coagulopathy and patients that have classical DIC. Notably, coagulopathy in COVID-19 patients is primarily characterized by the marked elevations in D-dimer and fibrin/fibrinogen degradation product levels while thrombocytopenia and prolongations of activated partial thromboplastin time (aPTT) and/or prothrombin time (PT) are less common or usually mild [100]. Accordingly, a hyperfibrinolytic consumptive DIC with a bleeding diathesis is a rare and unlikely event in these patients [100]. A worsening trend of coagulation parameters during an ICU stay generally indicates impending clinical deterioration, development of cytokine storm, and progression to multiorgan failure [101]. The procoagulant pattern in severe COVID-19 patients, assessed by standard laboratory and viscoelastic tests, revealed increased clot strength (CS), increased platelet and fibrinogen contribution to CS, elevated D-dimer levels, and hyperfibrinogenemia [102]. Similarly, a complete lack of fibrinolysis of a clot at 30 min and exuberant increase in D-dimer levels were strongly associated with thromboembolic events and renal failure [97]. Furthermore, levels of C-reactive protein (CRP), IL-6, factor VIII, vWF, and plasma viscosity are increased in these patients while antithrombin is modestly decreased thus clearly showing a cross-link between infection-induced inflammatory response and thrombosis, known as infection-induced thromboinflammation [100, 102‐104]. A summary of the most important laboratory parameters reflecting hemostasis derangement in patients with COVID-19 is shown in Table 2.

In terms of laboratory workup, measurements of D-dimer levels, PT/aPTT, thrombocyte count, and fibrinogen levels (if assays are available) are advised to identify hospitalized COVID-19 patients at a high risk of poor in-hospital outcomes, or to triage patients that would require hospital admission and close monitoring [105, 106]. Repeated assessments (every 2–3 days) of D-dimer, PT, and the platelet count at the minimum should be integrated into the diagnostic approach, as recommended by the documents of relevant societies [73, 106]. PT expressed as an international normalized ratio (INR) should not be used to monitor hemostasis in COVID-19 patients, since it mostly lies in the normal range. Early recognition and vigilant monitoring of coagulation abnormalities among hospitalized COVID-19 patients are critical measures to (i) identify patients at a high risk of poor prognosis, (ii) guide antithrombotic prophylaxis or treatment to prevent thromboembolic complications, and (iii) improve patients’ clinical outcomes.

Most of the guidelines and consensus documents published on behalf of professional societies focused on thrombosis and hemostasis advocate the use of prophylactic doses of anticoagulants in all patients hospitalized with COVID-19 in the absence of contraindications [105‐108]. Likewise, extended-duration thromboprophylaxis with low molecular weight heparin (LMWH) or a non-vitamin K antagonist oral anticoagulants (NOAC) for 2 to 6 weeks post-hospital discharge might be applied to hospitalized COVID-19 patients, especially those with no bleeding risk factors and with high VTE risk post-discharge (most commonly defined as advanced age, long stay in the ICU, malignancy, prior VTE history, thrombophilia, severe immobility, D-dimer levels > 2 times of upper reference range, an IMPROVE VTE score ≥ 4, and similar). Such recommendations are made due to observed hypercoagulability and a robust number of studies that consistently showed a high incidence of thromboembolic complications, especially venous thromboembolism in hospitalized and particularly ICU-treated COVID-19 patients [45, 46, 48, 49, 52]. Likewise, most of the available documents support the possibility to initiate intermediate-dose LMWH in selected severe COVID-19 patients by individually balancing thrombotic and bleeding risk. Finally, in hospitalized COVID-19 patients with verified VTE, the aforementioned consensus statements stimulate the use of full therapeutic doses of antithrombotic drugs (LWMH as most preferred, followed by unfractionated heparin) in all patients without contraindications, with carefully monitoring renal function, thrombocyte count, and concomitant medications. Antithrombotic prophylactic and therapeutic regimens in various clinical scenarios as recommended by relevant societies are shown in Table 3.

A treatment rationale for recommendations elaborated above is also partially based on the reported survival benefits, if the anticoagulation regimen is initiated in these patients. Of note, one study showed that treatment with prophylactic doses of LMWH was associated with lower 28-day mortality among patients with severe COVID-19 that had sepsis-induced coagulopathy score ≥ 4 or D-dimers > 6-fold of an upper limit of normal, compared with LMWH non-users [109]. However, the crude difference between the unselected group of patients that received LMWH vs. non-receivers in 28-day mortality endpoint was not significantly different. Similarly, another study found the significantly increased probability of survival in 2773 hospitalized COVID-19 patients if systemic anticoagulation was initiated, compared with if it was not, especially among patients that required mechanical ventilation [110]. One of the putative explanations for the beneficial effect of heparin and its derivatives in COVID-19 is based on the fact that heparins exhibit pleiotropic actions beyond anticoagulation, including direct anti-inflammatory action, as demonstrated by the decreased levels of inflammatory biomarkers in pre-clinical and clinical settings [111, 112]. Given that the severe inflammatory response in COVID-19 is the principal driver of coagulopathy, the concept of using heparin and its derivatives seems clinically validated as they likely blunt both thrombotic and inflammation pathways, thus improving outcomes in these patients.

COVID-19 caused by SARS-CoV-2, at the severest stages might be considered as a thromboinflammatory disease rather than an infection. In fact, the presence of microthrombotic pulmonary disease led to the suspicion that pulmonary macro- or microembolism could play a role in the pathophysiology and clinical deterioration of COVID-19 patients [1, 113]. The diagnosis of PE or MicroCLOTS may be challenging, because there are some overlapping features between pulmonary embolism and severe COVID-19 disease, such as dyspnoea, high levels of D-dimers, right ventricle with systolic dysfunction or enlargement, and ARDS. Most of the inpatients with COVID-19 should receive at least prophylactic anticoagulation in the absence of contraindication, particularly those at ICU, due to high reported incidence of VTE in this context [8, 89]. There are some observational studies suggesting that heparin improves the disease prognosis when the indication bias would suggest a worse prognosis (i.e., patients that received anticoagulation probably had an indication that could worsen the prognosis but the data suggests otherwise) [89, 109]. As there is no guidance for deciding the intensity of anticoagulation (prophylactic, intermediate, or therapeutic), the decision-making process should be made in individual-case basis, while studies are ongoing (NCT04377997, NCT04362085, NCT04367831). The suspicion of PE should be based in clinical grounds (unexplained chest pain, unexplained RV dysfunction, unilateral lower limb swelling) and not only in biomarkers such as D-dimers. It is known that D-dimers are frequently high in COVID-19 inpatients [62], they are a prognosis biomarker [114], but it is not clear if they reflect the existence of macrovascular thrombosis and/or the need to screen systematically VTE in these patients [108].

As for myocardial injury/infarction, there are also challenges because elevated troponin levels are often seen in COVID-19 patients, and the cause of troponin release may be multifactorial. First, both plaque rupture can occur during severe COVID-19 (type 1 myocardial infarction), and other causes of troponin release can be present, including imbalance between blood/oxygen supply and myocardial requirements, SARS-CoV-2 myocarditis, Tako-tsubo cardiomyopathy, or right heart failure (type 2 myocardial infarction). Patients with ST-elevation myocardial infarction (STEMI) or high-risk non-ST-elevation acute coronary syndrome (NSTE-ACS) should go to the catheterization laboratory, preferably the one dedicated to COVID-19, and followed by an ICU isolation bed [115]. For patients with COVID-19 and STEMI in a non-primary PCI center, fibrinolysis can be considered depending on the current timings for patient transfer for a primary PCI center [115]. In the remaining patients with COVID-19 and NSTE-ACS, physicians can opt out for optimal medical treatment alone for clinical stabilization while further testing, or coronary angiography should be performed after the infection is resolved [115].

Stroke in COVID-19 patients was not as common as VTE [8]. The approach to stroke is the same as non-COVID-19 patients, pursuing the vessel recanalization when indicated and when resources are available [116]. Ideally, these patients should be at COVID-19 stroke units, but logistic difficulties have been seen during the COVID-19 season [117]. In patients with severe COVID-19 disease, the development of stroke signs and symptoms should also raise the suspicion of paradoxical embolism through atrial septum defect.

In summary, COVID-19 is an infectious disease that leads to a proinflammatory and prothrombotic state, resulting in both micro- and macrovascular thrombosis, and both arterial and venous thrombotic events. Therefore, the early recognition and vigilant monitoring of the thrombotic complication of COVID-19 may be life-saving.