key: cord-1005238-tbceaxo4
authors: Quinaglia, Thiago; Shabani, Mahsima; Breder, Ikaro; Silber, Harry A.; Lima, João A.C.; Sposito, Andrei C.
title: Coronavirus disease-19: The multi-level, multi-faceted vasculopathy
date: 2021-02-15
journal: Atherosclerosis
DOI: 10.1016/j.atherosclerosis.2021.02.009
sha: 82d5838405a2024db8130401f1ad31734056bcfa
doc_id: 1005238
cord_uid: tbceaxo4

BACKGROUND AND AIMS: The new coronavirus disease (Covid-19) is a systemic disease. Mounting evidence depict signs and symptoms involving multiple organs, most of which supported by pathological data. A plausible link to these manifestations is vascular and endothelial dysfunction/damage. However, much of the current knowledge relies on opinion and incipient evidence. We aim to objectively appraise current evidence on the association between Covid-19 and vascular disease, specifically endotheliitis and vasculitis. METHODS: Two researchers independently entered the search terms Covid-19 OR SARS-CoV-2 AND vasculitis, endotheliitis OR endothelium in the following online platforms: MedRxiv and LitCovid (PubMed). The search period was set from November 1, 2019 to August 28, 2020. Manuscripts with unavailable full texts, not in English, mainly on pre-clinical data, presenting only study designs or not directly related to the topics of this review were excluded. Retrospective and prospective studies, especially longitudinal ones, were given priority to the purpose of this review. Since there was paucity of prospective controlled evidence, case reports/series were also considered. RESULTS: A total of 318 manuscripts were initially found. Sixty-seven (21%) were excluded: 59 (18.5%) met exclusion criteria and 8 (2.5%) were duplicates. One hundred and forty-two manuscripts (44,6%) did not provide original data and were also excluded: 35 (11%) were comments, 108 (33.9%) reviews; 1 (0.3%) position paper. One hundred and seven (33.6%) studies were considered for the present scoping review: 81 (25,5%) case reports/series; 18 (5.7%) prospective; 8 (2.5%) retrospective. Viral inclusions in endothelial cells, mononuclear cell infiltrates in the intima of small vessels and markers of endothelial cell apoptosis were demonstrated. Specificities of Covid-19 may lead to diverse vascular manifestations in different levels of the vascular bed. CONCLUSIONS: Evidence indicates that Covid-19 targets vasculature and endothelium. However, high quality data is still lacking and studies with prospective designs and appropriately matched controls are needed.

The new coronavirus disease-19 (Covid-19) is a major global issue and despite ubiquitous actions being employed to mitigate its effects, the disease still spreads relentlessly, and recrudescence waves of new cases seem to take place. At this moment, there is clear indication that the disease is systemic by design 1, 2 and involves not only the respiratory epithelium but also the cardiovascular endothelium 2 . Endothelial cells, like pneumocytes type II, abundantly express transmembrane angiotensin converting enzyme 2 (ACE-2) and host type 2 transmembrane serine protease (TMPRSS2). These are two fundamental locks through which the virus enters the cells. The disease, in fact, is primarily respiratory but may quickly gain the vascular network and hit gastrointestinal tract, kidneys, heart and nervous system, with associated platelet-endothelial dysfunction and abnormally rapid life-threatening blood clotting and inflammation.

In the lungs, endothelial cells of capillaries comprise up to 50% of the cellular mass of alveoli 3 and approximately one third of the alveolus' septal wall thickness is composed of a layer of endothelium 4 . Thus, it is not unexpected that these vascular cells could also be a target to the virus at least in the lungs. In parallel, the severe acute respiratory syndrome virus-2 (SARS-CoV-2) may also spread and affect endothelial cells in other sites, as it has been demonstrated. Evidence of pathology studies reveal viral inclusions within endothelial cells, inflammatory infiltrates in the intima layer of large and small arteries, and apoptosis-revealed by caspase-3 immunohistochemistry staining-or pyroptosis 5 of these cells in several organs and tissues' vessels, including lung, heart, kidney, and intestines 2, 6, 7 .

Although pathological findings have been conflicting in demonstrating viral particles within endothelial cells 1 , many mechanisms can concur for a vasculitis to happen. First, endothelial cell damage and apoptosis could lead to basal membrane exposure and a thrombotic macro-and/or microangiopathy 8 . Second, endothelial cell dysfunction due to unbalanced angiotensin II upregulation could drive local or diffuse vasoconstrictive responses whether in lungs, limbs, intestines or virtually any affected organ 9, 10 . Third, endothelial cells have been frequently recognized as part of the innate immune system and could be responsible for an exacerbated response in the inflammatory phase of the disease 11 . Lastly, perivascular inflammatory infiltrates can be implicated in a constrictive microangiopathy leading to chilblain lesions in extremities, a common symptom seen in Covid-19 mild presenting cases that could be frequently undiagnosed 12 .

In severe cases, cellular hypoxia may ultimately result from failure at different levels of the vascular bed-macrocirculation and microcirculation-and additionally from mitochondrial dysfunction 13 . At least macro-and microcirculation functions can be affected by vascular damage and/or endothelial dysfunction and may play a role in Covid-19. While macrocirculation failure emerges from impaired vascular tone control, microcirculation failure J o u r n a l P r e -p r o o f arises from shunting or maldistributed flow from disrupted perfusion of small arteries and capillaries. In this review, we systematically gather and appraise published data with the focus on vascular and endothelial function/damage and its relationship with clinical manifestations and disease severity.

Two researchers (TQ, IB) entered the search terms Covid-19 OR SARS-CoV-2 AND vasculitis, endotheliitis OR endothelium in the following online platforms: MedRxiv and LitCovid (PubMed). The search period was set from November 1 , 2019 to August 28, 2020. Manuscripts with unavailable full texts, that were not in English, had mainly pre-clinical data, presented only study designs or were not directly related to the topics of this review were excluded (figure 1). Retrospective and prospective studies (table 1), especially longitudinal ones, were given priority to the purpose of this review (figure 2). Since there was paucity of prospective controlled evidence, case reports/series were also considered. The terms OR SARS-CoV-2 AND endothelium OR vasculitis were also entered in the ClinicalTrials.gov platform (endotheliitis did not retrieve any studies in this specific platform), filtered for the same time period and same exclusion criteria, to retrieve ongoing studies investigating these aspects of the disease (supplementary table) . Additional relevant manuscripts were accessed by a manual search through the references of the initially retrieved studies in order to put data in context. All additional papers were included when there was agreement between researchers.

A total of 318 manuscripts were initially found. Sixty-seven were excluded: 59 (18.5%) met exclusion criteria and 8 (2.5%) were duplicates. One hundred and forty-two manuscripts (44,6%) did not provide original data and were also excluded: 35 (11%) were comments, 108 (33.9%) reviews; 1 (0.3%) position paper. One hundred and seven (33.6%) studies were considered for the present scoping review: 81 (25,5%) case reports/series; 18 (5.7%) prospective; 8 (2.5%) retrospective. None of the studies found in the ClinicalTrials.gov platform had results available. Evidence of viral inclusions in endothelial cells, mononuclear cell infiltrates in the intima of small vessels and markers of endothelial cell apoptosis was demonstrated. Specificities of the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) seem to lead to diverse vascular manifestations and involvement of different levels of the vascular bed mainly driven by pro-inflammatory, pro-thrombotic effects and possibly increased vascular reactivity. We outline below the main findings of the included articles discussing the vascular and endothelial crosstalk with the common features of Covid-19 pathophysiology.

J o u r n a l P r e -p r o o f

The SARS-CoV-2 enters the cells through a transmembrane angiotensin-converting enzyme 2 (ACE-2) 14 . Through such mechanism, ACE-2 activity is potentially downregulated.

Type II pneumocytes abundantly express the receptor, but also does the endothelium 15 . A host type 2 transmembrane serine protease (TMPRSS2) facilitates entry by cleaving ACE-2 and is also necessary for the process to occur 14 . The affinity of the spike protein of the virus for the transmembrane ACE-2 enzyme is 10-to 20-fold higher than that of the SARS coronavirus-1 (SARS-CoV1), the causal agent of SARS, which is one possible explanation for the virus high infectivity 16 . Thus, SARS-CoV-2 would target the endothelium, and impairment and loss of the vascular barrier integrity could promote arterial reactivity, and a pro-coagulative and proinflammatory state, clinically expressed, in the critical disease, as disseminated intravascular coagulation (DIC) and systemic inflammatory response syndrome (SIRS) [17] [18] [19] .

The advent of diffuse endotheliitis could lead to a cytokine storm frequently described in the severe disease 20 . An evidence of endothelial injury has been an elevation of detectable circulating endothelial cells 21, 22 and endothelial cells' progenitors 23 . Systemically, the endothelium and surrounding pericytes (mural cells) act as a barrier preventing immune cell and platelet interactions with tissue factor and collagen of the basement membrane. Thus, endothelial cells operate as a gatekeeper of the innate immune system and have a somewhat similar function to that of macrophages, which would include antigen presentation, cytokine secretion and scavenger activity 11, 24 . Reports of cardiac involvement, hypoxemia with apparently healthy lungs, and severe thrombotic complications 25 could be a consequence of diffuse endotheliitis expressed as localized or disseminated vasoconstriction, thrombotic macroand microangiopathy and a pro-inflammatory systemic response. Clinically, venous and arterial thromboembolic events have been reported in 10 to 25% of admitted patients 26 (figure 2).

Integrins seem to be increased, as unveiled by elevated levels of vascular cell adhesion molecule-1 (VCAM-1), intercellular adhesion molecule 1 (ICAM-1), and vascular adhesion protein-1 (VAP-1) 27 . However, an imbalance favouring the loss of vascular endothelial integrity is being indicated. Soluble fms-like tyrosine kinase 1 (sFlt-1) 28 -an inhibitor of the vascular endothelial growth factor pathway and demonstrated factor to promote endothelial dysfunction-is elevated in severe cases. This is also true for angiopoietin-2 29 , also a marker of endothelial damage/dysfunction 30 , and could be related to activated angiogenesis associated with extensive vascular damage.

J o u r n a l P r e -p r o o f

In the lung, damaged endothelium could trigger and propagate acute respiratory distress syndrome (ARDS), recognized as the leading cause of mortality in Covid-19 patients 18 .

Augmented concentration of angiotensin II, consequent to reduced levels of ACE-2, may lead to vasoconstriction and hypo-perfused lung areas (dead spaces) producing arterial hypoxemia exceeding abnormalities in pulmonary mechanics 31 . Another possible source of hypo-perfusion could derive from micro-thrombosis and congestion of small capillaries intermediated by inflammatory cells, as well as, thrombosis of medium to large vessels. After critical damage of pneumocytes, the pathogenesis of the hypoxemia changes. Alveolar flooding and impaired alveolar-arterial gas exchange give place to a ventilation-perfusion mismatch or intrapulmonary shunt. Endothelial cells present interleukin-6 (IL-6) and interleukin-2 receptors and stimulation of these induce capillary leakage and vascular fluid loss to interstitium and alveoli.

In turn, endothelial cells would intensify IL-6 secretion and perpetuate this response 32, 33 . Severe damage and denudation of the pulmonary endothelium could also deflagrate complement system activation which would elicit migration of neutrophils and monocytes further amplifying the inflammatory response 8 . At this point the exacerbated inflammation generates vasodilatation and increased vascular permeability which contributes to the interstitial and progressive alveolar edema. Receptor ACE-2 downregulation can play a role also by releasing the kallikreinbradykinin pathway which enhances vascular permeability 8 . Pneumonia has been reported to occur in 75% of hospitalized patients and ARDS in 15% 34 .

Another target to SARS-CoV-2 appears to be the heart. Cardiomyocytes present ACE-2 and TMPRSS2 and are susceptible to viral entry. Interstitial infiltration by inflammatory cells and viral particles were identified within the myocardium but viral inclusions were not found in cardiomyocytes or cardiac endothelial cells 35 . Cardiac injury reported as troponin elevation arises in 7 to 17% in hospitalized patients. Acute heart failure, myocarditis and arrhythmias have been reported as well 36 . In the SARS disease, the SARS-CoV viral RNA was detected in 7 out of 20 (35%) of autopsied human heart samples from a Toronto outbreak. The authors demonstrated that myocardial damage was associated with macrophage infiltration. ACE-2 expression was markedly reduced in the tissue and left ventricular hypertrophy was the most prominent attribute of the cardiac phenotype 37 . SARS-CoV and SARS-CoV-2 share very alike structures (79% similarity in the nucleotide sequence) 38 , thus, these findings suggest that the heart is highly susceptible to Covid-19 injury and attention should be paid to signs of cardiac injury. Another piece of evidence indicates that the incidence of Kawasaki disease, a systemic vasculitis of small-and medium-sized arteries that can affect coronary arteries, has been J o u r n a l P r e -p r o o f occurring in children and is apparently related to the current coronavirus pandemics 39, 40 .

Moreover, it has been suggested that the multisystem inflammatory syndrome seen in children can have even more diffuse endothelial involvement than Kawasaki disease 41 . However, this association may not be specific to Covid-19. A similar raise in Kawasaki disease cases has also occurred in a previous influenza pandemic 39 . From a mechanistic standpoint, SARS-CoV-2 could affect the heart through direct cell invasion, induction of atherosclerotic plaque rupturefor instance, via activated macrophage-induced collagen degradation of its fibrous cap-, coronary spasm, microthrombi obstruction of a coronary or cardiomyocyte dysfunction owing to inflammation, or even other still unknown pathways 42 .

As for the kidneys, renal epithelial cells, the podocytes, express ACE-2 and TMPRSS2, but, thus far, little expression was demonstrated in human endothelial cells within the kidney 43 . In spite of that, viral particles have been described in endothelial cells of glomerular capillaries 2 .

In a case series of deceased patients presenting multiple organ dysfunction, obstruction of capillaries and inclusion of viral particles within tubular epithelium and podocytes was shown without evident vasculitis or overt interstitial inflammation. Acute kidney failure may occur, as well, in consequence of exacerbated inflammation and compromised blood flow. This condition has been reported in around 9% of hospitalized patients. Enhanced coagulation and stagnant flow contribute to systemic hypoxia and can damage oxygen-sensitive organs, such as the kidneys 44 . Mechanistically, kidney injury may occur due to direct invasion of proximal tubular cells or podocytes. Indirect injury may result from reduced blood flow following a dysregulated ACE-2 pathway and thus, acute tubular necrosis, protein leakage in Bowman's capsule, collapsing glomerulopathy and mitochondrial impairment. Exacerbated inflammation and the cytokine storm may also lead to kidney malfunction 45 .

High level-expression of ACE-2 has been documented too in brain endothelial cells, neurons and glial cells 46, 47 . Particles of the virus were recently identified in brain capillary endothelial cells 46 and in brain tissue of 36% of a Covid-19 series 48 . Infection of the olfactory tract and the consequent anosmia were supported by increased MRI signal of the olfactory cortex 49 . SARS-CoV-2 may reach the central nervous system by retrograde transneuronal routes or hematologic route by crossing a compromised blood-brain barrier due to endothelial injury/endotheliitis 50 , direct infection of endothelial cells or through macrophages carrying the virus. Interleukin(IL)-6 and tumor necrosis factor have been previously demonstrated to disrupt the blood-brain barrier 51 . These cytokines are also powerful activators of the hypothalamicpituitary-adrenocortical axis, a pivotal regulator of the systemic immune activity 52 . Indirectly, J o u r n a l P r e -p r o o f the nervous system can also be affected via unbalanced ACE/ACE-2 pathways, which leads to excessive vasoconstriction, ischemia and a possible disturbance of cerebral blood flow regulation 10 . Cytokine storm may contribute to both blood-barrier disruption and neurotoxicity 50 although the mechanisms are not completely understood.

Intestinal glandular cells present transmembrane ACE-2 and could be a direct target to the virus 53 . The expression ACE-2 appears to be more frequent in ileum and colon mainly in the absorptive enterocytes which could explain the manifestation of diarrhea in some patients 54 .

Viral RNA has been isolated in stools 55 . Diffuse endothelial inflammation and mesenteric ischemia in submucosal vessels of small bowel has been described and can be related to local microvascular injury 2 .

Cutaneous manifestations have been frequently reported in highly probable cases, but with a low positivity of real-time reverse transcriptase polymerase chain reaction (RT-PCR).

This may happen because potential mechanisms can relate to immune response to SARS-CoV-2 rather than direct virus infection 56 . For this reason, these cutaneous manifestations may not occur in the early phase of the disease, but patients commonly report previous respiratory symptoms 56 . Evidence to support this theory has come from the observation of superficial perivascular dermatitis evidenced as diffuse and dense lymphoid infiltrates accompanied by endothelial inflammation on biopsies of acrocutaneous lesions 57 . Eventually, small thrombi have been also shown in vessel of the dermis 58 . In a survey, 20% of dermatologists have reported having recognized cutaneous manifestations of Covid-19 and 16% of these lesions to be of a vasculitis nature 59 . Although still largely unknown, the most documented cutaneous lesion has been chilblains. One study has shown a likely association between endothelial infection and this manifestation. Variable degrees of lymphocytic vasculitis, from endothelial swelling to necrosis, and perivascular lymphocytic inflammation were reported 60 .

A delayed or halted adaptive immune response has been implicated in severe Covid-19 cases, particularly when a cytokine storm takes place. When uncompensated, unleashing the innate immunity could lead to deleterious exacerbated inflammation and ineffective clearance of the virus. Interferon type I production has been implicated in effective viral debacle in previous coronavirus diseases 61 . However, these viruses seem to present mechanisms to supress IFN-I action 62 . Interferon type I activates the adaptive immune system in mammals through T-cell stimulation. Accordingly, lymphopenia has been linked to worse prognosis particularly due to J o u r n a l P r e -p r o o f depletion of circulating CD4+ and CD8+ T-cell 63 . In pathology studies, CD4+ T-cells were more numerous in lungs from patients who died with Covid-19 than in patients with influenza, but the opposite was seen for CD8+ 64 . CD4+ T-cells interact with B-cells to create virusspecific antibody production and also with CD8+ T-cells. The latter, in their turn, recognize peptides in the virus surface and directly attack them through a cytotoxic response. Aside from immune evasion by the virus 62 , an early bypass of innate immunity could be due to submicrometer sized virus-laden aerosols reaching deep into alveoli without being undertaken by the defence mechanisms of the upper respiratory tract and the first antigen presenting cells 65 .

Another plausible explanation would be a swift replication of SARS-CoV-2, known to be at least 3 times faster than SARS-CoV-1 66 .

However, deleterious inflammation induced by the virus can only occur in conjunction with an inappropriate host response. Only around 20% of patients present severe to critical disease and clinical characteristics in these cases are consistently described as older age, male sex and presence of cardiovascular risk factors 67, 68 . These main clinical features are all related to baseline endothelial dysfunction as suggested previously 69 . Which could explain, at least in part, the susceptibility of these subgroups of patients. The endothelium is currently recognized as one of the sentinels of the innate immune system 11 . These cells express IFN-α, a cytokine of paramount importance in regulating innate immune responses against viruses 70 Moreover, it is important to differentiate these skin manifestations from acral ischemia, which is more related to vasoconstriction and thrombosis and unfold mostly in critically ill patients.

The local and systemic pro-inflammatory state per se promotes an environment prone to coagulation. This is true for influenza and other respiratory coronavirus diseases, in which thrombotic events appear to be increased. However, some particularities pertain to SARS-CoV-2 especially in regard to the vasculature. Pathology studies demonstrate severe endothelial damage associated with intracellular inclusions of the virus and disrupted cell membranes 64 . In a comparison of lung specimens from influenza patients matched for age, sex and disease severity, capillary microthrombi (<10 µm) were 9-fold more common in Covid-19. Although, influenza patients also present thrombi in pulmonary arteries from 1 to 2 mm of diameter and downstream, the greater difference was indeed observed at the capillary level. Also, microvascular structure was significantly distorted due to capillary elongation and angiogenesis (2.7 times more prevalent than in the influenza counterparts). These findings can possibly be consequence of the more common endothelial inflammation and injury and of the capillary thrombosis 64 . The data endorses that the pathogenesis of hypoxemia differs in Covid-19 and influenza disease at least in an initial stage of the disease 83 . In Covid-19 pulmonary dead space-non-perfused but ventilated alveoli-would play more of a role 21 , whereas in other viral disease pulmonary shunt-perfused but not ventilated alveoli-would have much of an importance. In accordance with this understanding, the same study shows that lungs from influenza patients were heavier in gross examination than Covid-19 ones indicating a greater amount of fluid accumulation.

Endothelial disruption and apoptosis of endothelial cells expose basement membrane, which in turn, activates thrombotic cascades. Interleukin-1β and Tumor Necrosis Factor J o u r n a l P r e -p r o o f stimulate endothelial cells to express P-selectin, von Willebrand factor, factor VIII and fibrinogen 84, 85 -ultimately binding substrates to platelets. The endothelial cells release additional cytokines that further increases platelet production, and, in a closed loop, platelets liberate vascular endothelial growth factor (VEGF) upregulating tissue factor expression by the endothelium, a primer to the thrombotic cascade 86 . While under a coronavirus infection, endothelial cells also increase plasminogen activator inhibitor 1 (PAI-1) 77 inhibiting conversion of plasminogen to plasmin, which normally degrades fibrin clots, further contributing to a coagulation state. An antagonizing homeostatic effort generates high levels of fibrin breakdown products (D-dimers) as a result of blood clots dissolution 87 . The ischemic tissue may induce endothelial cells hyperplasia additionally contributing to ischemia or to angiogenesis that could alleviate ischemia, but could also facilitate the spread of inflammatory cells 8 .

There has been controversy about the pathophysiology of the thrombotic manifestations in Covid-19 especially in severe cases. Opposition has been made between thrombotic microangiopathy (TMA) and sepsis-induced coagulopathy or DIC as the main pathophysiological process of the microvascular thrombosis. The distinction is very important because of therapeutic implications. TMA is a specific process characterized by microangiopathic hemolytic anemia thrombocytopenia and organ dysfunction and damage.

TMA occurs mainly at arterioles and results of increased platelet adhesion to the vascular endothelium in association with platelet aggregation and activation, driving to a consumptive thrombocytopenia 88 . Common laboratorial features in Covid-19 supporting this presentation are increased lactate dehydrogenase (LDH), low red blood cell count (RBC) and thrombocytopenia 89 . However, information on important aspects are still missing, such as the presence of schistocytes in peripheral blood smear to confirm microangiopathic hemolysis.

Complement system plays a pivotal role in TMA as part of the innate immune response 90 . A recent study demonstrated that N proteins of SARS-CoV-2 binds to a key serine protease in the lectin pathway of complement activation MASP-2 (Mannan-binding lectin serin protease 2) 91 .

Another study corroborates with the activation of the lectin and the alternative pathway of the complement system. The authors found deposits of C5b-9 (the membrane attack complex-MAC), C4d and MASP-2 in the microcirculation of purpuric skin lesions and also in normallyappearing skin, as well as, in the interalveoli septa 92 . Soluble C5a also appears to be increased in a proportionate fashion to disease severity in parallel to high expression of its receptor C5aR1 in blood and pulmonary cells 93 . Together with the fact that renal and cardiac dysfunction are common features of severe to critical Covid-19, it appears that a complement-mediated TMA is a part of Covid-19 pathophysiology.

Deposition of MAC on cell membranes induce calcium influx and activate intracellular signals 94 . Through this mechanism, deposits of MAC can activate platelets and enhance aggregation 95 presence of NETs indicate that thrombosis is immuno-related and may affect many organs, including lung, kidney and heart and is associated with disease severity 99, 100 . Also, that viable virus can be responsible for the direct induction of NETs release by healthy neutrophils indicating inhibition of its mechanism may be a potential therapeutic target 101 .

On the other end, increased D-dimer levels, thrombocytopenia and prolonged coagulation tests could resemble a DIC pattern often seen in severe sepsis. However, a hallmark of DIC is abnormal thrombin generation and no evidence has been presented towards this direction 102 . Tissue factor is often expressed by mononuclear cells and endothelial cells, which in turn accrues from proinflammatory cytokines' discharge, particularly IL-1β and IL-6 102, 103 .

DIC is a consumptive coagulopathy often followed by hemorrhagic phenomenon, which is not frequently reported in Covid-19. Although elevation of D-dimer is common (40-60% of cases)

in Covid-19, this is possibly due to localized exacerbated coagulation and fibrinolysis within the lung 104 . On top of that, damaged endothelial cells change their properties to a procoagulant mode following disruption of the glycocalyx and its binding protein, antithrombin 105, 106 ( figure   2 ). Thrombin-antithrombin complex were elevated in ICU patients. Interestingly, thrombomodulin levels-a specific marker of endothelial injury-were associated with reduced hospital discharge. The normally preserved antithrombin, protein C and protein S activities support the idea of a different mechanism than that of DIC 85 . A recent report stated that 8.7% of a series of hospitalized Covid-19 cases met criteria for DIC as established by international guidelines, whereas the prevalence of DIC in sepsis reaches around 30% 17 . As the disease advances in severity other factors come into play and DIC may arise. 

Endothelial function may be assessed in a variety of ways and many techniques are being applied to assess possible effects and chronic consequences of Covid-19 in the vasculature (supplementary table) . One of such methods is flow mediated dilation (FMD) evaluated by ultrasound. Its parameters associate with vasomotor endothelial function and may be a proxy to cardiovascular health. Macrocirculation can be assessed by the dilation response of a large artery (diameter > 400 µm), mainly driven by acute nitric oxide production, and microcirculation can be assessed by hyperemic flow increase, due to accumulation of metabolites in arterioles (diameter < 100 µm). Ultrasound-based FMD has been used to assess vasomotor endothelial function in patients with sepsis and septic shock due to primarily respiratory infections and correlates with adverse outcomes and survival 120 . In Covid-19, the correlation could be greater due to the intertwined relationship of the virus cycle and the endothelium/arterial wall. FMD has also been previously validated by magnetic resonance imaging with reduced variability of results 121 . To specifically evaluate microvascular rarefaction J o u r n a l P r e -p r o o f nailbed capillaroscopy is being used, a technique that allows direct visualization and count of peripheral capillaries (<10µm).

Many interventions targeting the vasculature and specifically the endothelium are being tested. Drugs acting on the thrombotic and inflammatory aspects of the endothelium are presented (supplemental table). The effects of these drugs can vary from endothelial glycocalyx function restoration with glycosaminoglycans to reduction of vascular permeability with anti VEGF recombinant humanized monoclonal antibody. Actions on the renin-angiotensinaldosterone system are of particular interest because this seems to be the mainstay of Covid-19 pathophysiology. Angiotensin-converting enzyme (ACE) inhibitors, angiotensin (AT) II receptor blockers and statins are among the most cited. The first trial seem to show no difference in mortality in the first 30 days after discharge 122 for patients who had the treatment discontinued versus those who did not. The effect of ACE inhibitors or AT II receptor blockers is still unknown if initiated in naïve-treated patients admitted for Covid-19. In this specific scenario, these drugs could have an effect on rebalancing the renin-angiotensin pathway and regulating vascular reactivity as well as potentially reducing cell entry by the virus, although the effect in angiotensin II activity is controversial 123 . Recombinant ACE-2 may have a potential benefit, however, because it would directly increase angiotensin II cleavage 124 . Therapy with statins can improve peripheral NO-mediated arterial relaxation and offset the effects of SARS-CoV-2 125 . This class of drugs may also reduce macrophage chemotaxis and downregulate MCP-1 activity 126 . Other therapies being investigated that could act upon vascular reactivity are inhaled nitric oxide, prostacyclin and phosphodiesterase-5 inhibitors, such as sildenafil.

This study has limitations which are mostly associated with the nature of the manuscripts used for this review. Covid-19 is a newly discovered disease and although much progress was done towards building the knowledge on its different aspects, high quality evidence is still lacking in some specific aspects of the disease. The vast amount of data derives from case reports and case series. Only a few retrospective and prospective studies were published with the specific focus of understanding the association with Covid-19 and endotheliitis and/or vasculitis. Moreover, the majority of them did not have any or had control groups of healthy subjects which helps shed light on the pathophysiology of the disease but may introduce some bias in understanding some of its features. Having appropriate controls, for instance other viral pneumonias or viral diseases, would allow identifying specific biomarkers of Covid-19. At this moment, there is no certainty of the specific features of Covid-19. We also did not exclude case report or series from the present analysis due to the paucity of prospective J o u r n a l P r e -p r o o f controlled evidence. Another limitation of the present study is that a meta-analysis to assess the incidence of vascular events was hindered by the heterogeneity of control groups across studies and this greatly limits more precise conclusions about the topic. NETs still need to be addressed. Furthermore, the specific aspects of induced-thrombosis also need further understanding. Whether complement-related or sepsis-like the management of thrombosis can be quite different. Lastly, the proposed vascular reactivity is still mostly in the hypothesis field. Clinical trials targeting this specific aspect of the disease will clarify how important this is to its pathophysiology. Further larger studies with prospective designs are needed to better delineate the disease pathophysiology and potential targets for treatments.

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Drs Quinaglia, Silber, Lima and Sposito designed the study. Drs Quinaglia, Shabani and Breder strategized and extracted the data. Drs. Quinaglia and Breder drafted the manuscript.

The Graphical Abstract was crafted using elements provided by Servier Medical Art under the Baseline biomarkers compared with healthy controls (n=36); patients with sepsis (n=37), ARDS (n=19), or burns (n=35)

Baseline serum levels of IL-6, MCP-1, and IL-10 in the severe COVID-19 patients were higher than in healthy controls but lower than in CRS patients (sepsis, ARDS, and burns). Only PAI-1 levels were similar to that of CRS patients. An IL-6 receptor antagonist use associated with a decrease in PAI-1, IL-10 and C-reactive protein (CRP) levels. Levels of MCP-1 were unaffected and IL-6 increased Stahl et al. 106 Injury to the Endothelial Glycocalyx in Critically Ill COVID-19 Patients Prospective (observational) 39 Healthy controls (n=10)

Endothelial markers Angpt-1 and -2 were similar between COVID-19 patients and controls. In contrast, an increase in the soluble form of the Tie2 receptor and in Syndecan-1 indicates shedding of transmembrane proteins of the glycocalyx structure and processing. The endothelial glycocalyx protective molecule Hpa-2 was reduced in Covid-19 patients Perrin et al. 50 Cytokine release syndrome-associated encephalopathy in patients with COVID-19 Prospective (observational) 5 No control group Neurologic manifestations occurred simultaneously with the peak in CRS serum markers (CRP, IL-6 and LDH). Serum levels of the astroglial marker, S100B protein, were increased at the time of CRS reflecting an increased permeability of the blood-brain barrier and returned to their reference range when neurological symptoms and signs of hyperinflammation regressed Ladikou et al. 84 Von Willebrand factor (vWF): marker of endothelial damage and thrombotic risk in COVID-19?

No control group

The INR, APTT ratio and platelet count were all within the normal range. Fibrinogen and D-dimer were elevated. Factor VIII and von Willebrand factor antigen were proportionally higher in deceased patients.

Carvelli et al. 93 Association of COVID-19 inflammation with activation of the C5a-C5aR1 axis Prospective (observational) 82 Healthy controls (n=10)

Increase in plasma complement factor C5a is proportional to COVID-19 severity. High expression of its receptor C5aR1 in blood and pulmonary myeloid cells, as well as, obliterating endarteritis associated with accumulation of C5aR1+ macrophages around the arteries and within thrombus were observed. Findings support a role for the C5a-C5aR1 axis Dupont et al. 28 Excess soluble fms-like tyrosine kinase 1 correlates with endothelial dysfunction and organ failure in critically ill COVID-19 patients Prospective (observational) 56 No control group

High circulating sFlt-1, a variant of the VEGF-A receptor 1 that impairs endothelial cell homeostasis, is associated with severe COVID-19 and correlates with the endothelial dysfunction biomarker sVCAM-1 and with adverse outcomes during the follow-up of these patients J o u r n a l P r e -p r o o f An incidence of 1.4% of cerebrovascular disease was seen in patients with COVID-19, mostly ischemic (17 cases). Of these, major etiology was of undetermined origin (9 cases, as in TOAST classification). Age was the only independent predictor of poor prognosis. D-dimer levels were high in most patients J o u r n a l P r e -p r o o f Dp-ucMGP, an inactivated form of MPG, was elevated in COVID-19 patients compared to controls and associated with disease severity. Vitamin K activates MPG, which protects against pulmonary and vascular elastic fiber damage Guervilly et al. 22 Circulating Endothelial Cells as a Marker of Endothelial Injury in Severe COVID -19

Retrospective (cross-sectional) 99 No control group CEC count was above the upper normal limit in 55% of Covid-19 patients and it was significantly higher in ICU patients than in non-ICU. CEC count correlated with sVCAM-1 and IP-10 concentrations but not with E-selectin and IL-6 concentrations Herman et al. 82 Evaluation Consiglio et al. 41 The Immunology of Multisystem Inflammatory Syndrome in Children with COVID-19

Prospective (cross-sectional) 73

Historical control of Kawasaki disease children (n=28); healthy controls (n=12)

Differences in T-cell subsets and cytokine mediators place MIS-C between Kawasaki disease and acute Covid-19 immune states. Results suggest a more diffuse endothelial involvement and immunopathology in MIS-C than in Kawasaki disease Chang et al. 40 Serum responses of children with Kawasaki Disease against SARS-CoV-2 proteins Retrospective (cross-sectional) 87

Historical febrile (n=36) and Kawasaki disease (n=39) controls

There is not a non-specific global cross-reactivity from previous coronaviruses contributing to these pediatric COVID-19 related inflammatory disorder cases Shanshal et al. 59 The Impact of COVID-19 on Medical Practice: A Nationwide Survey of Dermatologists and Healthcare Providers Prospective (cross-sectional) 300

No control group COVID-19-related cutaneous manifestations were recognized by 20% of dermatologists, including maculo-papular rash (41.67%), urticaria (37.50%), chilblain (25%) and vasculitis (16.67%) For the prospective interventional studies, a label of randomized or non-randomized was added to the cell. Angpt: angiopoietin; APTT: activated partial thromboplastin time; ARDS: acute respiratory distress syndrome; CEC: circulating endothelial cells; CRP: C-reactive protein; CRS: cytokine release syndrome; dp-ucMGP inactive vitamin K-dependent matrix Gla protein, FiO2: fraction of inspired oxygen; FKN: fractalkine; Hpa-2: heparanase-2; ICAM-1: intercellular adhesion molecule-1; ICU: intensive care unit; IFNγ: interferonγ; IL-6: interleukin-6; IL-10: interleukin-10; IL-18: interleukin-18; IP-10: interferon-induced protein; INR: international normalized ratio; LDH: lactase dehydrogenase; MCP-1: monocyte chemotactic protein-1; MIS-C: multisystem inflammatory syndrome in children; NETs: neutrophil extracellular traps; PAI-1: plasminogen activator inhibitor-1; PaO2: partial pressure of oxygen; RANTES: regulated on activation, normal T-cell expressed and secreted; sFlt-1: soluble fms-like tyrosine kinase 1; TNFα: tumor necrosis factorα; VAP-1: vascular adhesion protein-1; VCAM-1: vascular cell adhesion molecule-1; sVCAM-1: soluble vascular cell adhesion molecule-1.

J o u r n a l P r e -p r o o f 

Histopathology and ultrastructural findings of fatal COVID-19 infections in Washington State: a case series

Endothelial cell infection and endotheliitis in COVID-19

Structure and composition of pulmonary arteries, capillaries, and veins

Cell number and cell characteristics of the normal human lung

Clinical and pathological investigation of patients with severe COVID-19

SARS-CoV2 vertical transmission with adverse effects on the newborn revealed through integrated immunohistochemical, electron microscopy and molecular analyses of

Evolving functions of endothelial cells in inflammation

COVID-19 Presenting as Acute Limb Ischaemia

Cerebrovascular disease in patients with COVID-19: neuroimaging, histological and clinical description

Coronavirus (COVID-19) infection-induced chilblains: A case report with histopathologic findings. JAAD Case Rep. 2020. 13. Nencioni A, Trzeciak S and Shapiro NI. The microcirculation as a diagnostic and therapeutic target in sepsis

SARS-CoV-2 Cell Entry Depends on ACE2 and TMPRSS2 and Is Blocked by a Clinically Proven Protease Inhibitor

Single-cell RNA-seq data analysis on the receptor ACE2 expression reveals the potential risk of different human organs vulnerable to 2019-nCoV infection

Cryo-EM structure of the 2019-nCoV spike in the prefusion conformation

Abnormal coagulation parameters are associated with poor prognosis in patients with novel coronavirus pneumonia

Characteristics of and Important Lessons From the Coronavirus Disease 2019 (COVID-19) Outbreak in China: Summary of a Report of 72314 Cases From the Chinese Center for Disease Control and Prevention. JAMA. 2020. 19. Fan BE

Hlh Across Speciality Collaboration UK. COVID-19: consider cytokine storm syndromes and immunosuppression

Respiratory mechanics and gas exchanges in the early course of COVID-19 ARDS: a hypothesis-generating study

Circulating Endothelial Cells as a Marker of Endothelial Injury in Severe COVID -19

An Integrated Gene Expression Landscape Profiling Approach to Identify Lung Tumor Endothelial Cell Heterogeneity and Angiogenic Candidates

Incidence of venous thromboembolism in hospitalized patients with COVID-19

Excess soluble fms-like tyrosine kinase 1 correlates with endothelial dysfunction and organ failure in critically ill COVID-19 patients

Angiopoietin-2 as a marker of endothelial activation is a good predictor factor for intensive care unit admission of COVID-19 patients

Plasma vascular endothelial growth factor, angiopoietin-1, and angiopoietin-2 in diabetes: implications for cardiovascular risk and effects of multifactorial intervention

Basing Respiratory Management of Coronavirus on Physiological Principles

Interleukin-6 causes endothelial barrier dysfunction via the protein kinase C pathway

Improved IL-2 immunotherapy by selective stimulation of IL-2 receptors on lymphocytes and endothelial cells

Sah R and Latin American Network of Coronavirus Disease C-REahwlo. Clinical, laboratory and imaging features of COVID-19: A systematic review and meta-analysis

Myocardial localization of coronavirus in COVID-19 cardiogenic shock

Cardiovascular complications in COVID-19

SARS-coronavirus modulation of myocardial ACE2 expression and inflammation in patients with SARS

Genomic characterisation and epidemiology of 2019 novel coronavirus: implications for virus origins and receptor binding

Emergence of Kawasaki disease related J o u r n a l P r e -p r o o f to SARS-CoV-2 infection in an epicentre of the French COVID-19 epidemic: a timeseries analysis

Serum responses of children with Kawasaki Disease against SARS-CoV-2 proteins

COVID-19 and cardiovascular disease: from basic mechanisms to clinical perspectives

Reduced vitamin K status as a potentially modifiable risk factor of severe COVID-19

Kidney disease is associated with in-hospital death of patients with COVID-19

Central nervous system involvement by severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2)

Tissue distribution of ACE2 protein, the functional receptor for SARS coronavirus. A first step in understanding SARS pathogenesis

Multiorgan and Renal Tropism of SARS-CoV-2

Magnetic Resonance Imaging Alteration of the Brain in a Patient With Coronavirus Disease 2019 (COVID-19) and Anosmia

Cytokine release syndrome-associated encephalopathy in patients with COVID-19

Neuroimmune Axes of the Blood-Brain Barriers and Blood-Brain Interfaces: Bases for Physiological Regulation, Disease States, and Pharmacological Interventions

Neuroimmune Interactions: From the Brain to the Immune System and Vice Versa

SARS-CoV-2 productively infects human gut enterocytes

The digestive system is a potential route of 2019-nCov infection: a bioinformatics analysis based on single-cell transcriptomes

Manifestations and prognosis of gastrointestinal and liver involvement in patients with COVID-19: a systematic review and meta-analysis

Cutaneous manifestations in COVID-19: a first perspective

Cutaneous Clinico-Pathological Findings in three COVID-19-Positive Patients Observed in the Metropolitan Area of

The Impact of COVID-19 on Medical Practice: A Nationwide Survey of Dermatologists and Healthcare Providers

SARS-CoV-2 endothelial infection causes COVID-19 chilblains: histopathological, immunohistochemical and ultrastructural study of seven paediatric cases

High secretion of interferons by human plasmacytoid dendritic cells upon recognition of Middle East respiratory syndrome coronavirus

SARS and MERS: recent insights into emerging coronaviruses

Lymphopenia predicts disease severity of COVID-19: a descriptive and predictive study

Pulmonary Vascular Endothelialitis, Thrombosis, and Angiogenesis in Covid-19

Recognition of aerosol transmission of infectious agents: a commentary

Comparative replication and immune activation profiles of SARS-CoV-2 and SARS-CoV in human lungs: an ex vivo study with implications for the pathogenesis of COVID-19

Presenting Characteristics, Comorbidities, and Outcomes Among 5700 Patients Hospitalized With COVID-19 in the New York City Area. JAMA

The assessment of endothelial function: from research into clinical practice

IFN-alpha enhances TLR3-mediated antiviral cytokine expression in human endothelial and epithelial cells by up-regulating TLR3 expression

Immune response in COVID-19: addressing a pharmacological challenge by targeting pathways triggered by SARS-CoV-2

Dysregulation of immune response in patients with COVID-19 in Wuhan, China

Targeting Endothelial Dysfunction in Eight Extreme-Critically Ill Patients with COVID-19 Using the Anti-Adrenomedullin Antibody Adrecizumab (HAM8101)

IL-6 in diabetes and cardiovascular complications

Endothelial cells are central orchestrators of cytokine amplification during influenza virus infection

Mapping the innate signaling cascade essential for cytokine storm during influenza virus infection

IL-6 trans-signaling induces plasminogen activator inhibitor-1 from vascular endothelial cells in cytokine release syndrome

Two cases of COVID-19 presenting with a clinical picture resembling chilblains: first report from the Middle East

A clinical, histopathological and laboratory study of 19 consecutive Italian paediatric patients with chilblain-like lesions: lights and shadows on the relationship with COVID-19 infection

SARS-CoV-2 launches a unique transcriptional signature from in vitro, ex vivo, and in vivo systems

Heightened Innate Immune Responses in the Respiratory Tract of COVID-19 Patients

Evaluation of Chilblains as a Manifestation of the COVID-19 Pandemic

Enhanced platelet inhibition treatment improves hypoxemia in patients with severe Covid-19 and hypercoagulability. A case control, proof of concept study

Von Willebrand factor (vWF): marker of endothelial damage and thrombotic risk in COVID-19

Endotheliopathy in COVID-19-associated coagulopathy: evidence from a single-centre, cross-sectional study

Platelets, petechiae, and preservation of the vascular wall

Basic and clinical aspects of fibrinolysis and thrombolysis

Syndromes of thrombotic microangiopathy

Coagulation abnormalities and thrombosis in patients with COVID-19

Complement in Thrombotic Microangiopathies: Unraveling Ariadne's Thread Into the Labyrinth of

Highly pathogenic coronavirus N protein aggravates lung injury by MASP-2-mediated complement over-activation

Complement associated microvascular injury and thrombosis in the pathogenesis of severe COVID-19 infection: A report of five cases

Human complement in thrombin-mediated platelet function: uptake of the C5b-9 complex

Complement proteins C5b-9 induce secretion of high molecular weight multimers of endothelial von Willebrand factor and translocation of granule membrane protein GMP-140 to the cell surface

Simultaneous activation of complement and coagulation by MBL-associated serine protease 2

Vascular occlusion by neutrophil extracellular traps in COVID-19

Viral presence and immunopathology in patients with lethal COVID-19: a prospective autopsy cohort study

SARS-CoV-2-triggered neutrophil extracellular traps mediate COVID-19 pathology

Coagulopathy: Disseminated Intravascular Coagulation and Thrombotic Microangiopathy-Either, Neither, or Both

Bidirectional relation between inflammation and coagulation

Curative anticoagulation prevents endothelial lesion in COVID-19 patients

The unique characteristics of COVID-19 coagulopathy

Injury to the Endothelial Glycocalyx in Critically Ill COVID-19 Patients

Influenza and SARS-coronavirus activating proteases TMPRSS2 and HAT are expressed at multiple sites in human respiratory and gastrointestinal tracts

Evaluation of angiotensin-converting enzyme (ACE), its homologue ACE2 and neprilysin in angiotensin peptide metabolism

Hydrolysis of biological peptides by human angiotensin-converting enzyme-related carboxypeptidase

Angiotensin-(1-7) through receptor Mas mediates endothelial nitric oxide synthase activation via Akt-dependent pathways

Nitric oxide and vascular reactivity in sepsis

COVID-19 With Limb Ischemic Necrosis

Characteristics of Large-Vessel Occlusion Associated with COVID-19 and Ischemic Stroke

Nitric oxide and virus infection

Nitric oxide and pathogenic mechanisms involved in the development of vascular diseases

The glycocalyx: a novel diagnostic and therapeutic target in sepsis

Sources of Vascular Nitric Oxide and Reactive Oxygen Species and Their Regulation

Cytokine responses in severe acute respiratory syndrome coronavirus-infected macrophages in vitro: possible relevance to pathogenesis

Dual effect of nitric oxide on SARS-CoV replication: viral RNA production and palmitoylation of the S protein are affected

Endothelial dysfunction assessed by brachial artery ultrasound in severe sepsis and septic shock

The relationship between vascular wall shear stress and flow-mediated dilation: endothelial function assessed by phase-contrast magnetic resonance angiography

Effect of Discontinuing vs Continuing Angiotensin-Converting Enzyme Inhibitors and Angiotensin II Receptor Blockers on Days Alive and Out of the Hospital in Patients Admitted With COVID-19: A Randomized Clinical Trial

Renin-Angiotensin-Aldosterone System Inhibitors in Patients with Covid

Human recombinant soluble ACE2 in severe COVID-19

Simvastatin, an HMG-coenzyme A reductase inhibitor, improves endothelial function within 1 month

of SARS-CoV-2-related vasculitis/endotheliitis. Covid-19 may affect the endothelium and smooth muscle cells of the arterial and venous wall at different levels and through multiple ways, namely: inflammation/immune response, hypercoagulability and increased vascular reactivity. ACE-2: angiotensin-converting enzyme 2; Angpt: angiopoietin; CEC: circulating endothelial cells; CRP: C-reactive protein; DIC: disseminated intravascular coagulopathy

NETs: neutrophil extracellular traps; NO: nitric oxide; sVCAM-1: soluble vascular cell adhesion molecule-1

sVCAM-1: soluble vascular cell adhesion molecule-1; TMA: thrombotic microangiopathy; TNF-α: tumor necrosis factor -α

VAP-1: vascular adhesion protein-1