Endotheliopathy: A Marker of Progression to Critical Illness in COVID-19

Kristi Rosa
Kristi Rosa

Managing Editor, OncLive®
Kristi Rosa joined MJH Life Sciences in 2016 and has since held several positions within the company. She helped launch the rapidly growing infectious disease news resource Contagion, strengthened the Rare Disease Report, of HCPLive, and now serves as the main digital news writer for OncLive. Prior to working at the company, she served as lead copywriter and marketing coordinator at The Strand Theater. Email: krosa@onclive.com

Endotheliopathy represents a marker of progression to critical illness in patients with COVID-19 infection and soluble thrombomodulin segregates with mortality.

Endotheliopathy represents a marker of progression to critical illness in patients with COVID-19 infection and soluble thrombomodulin segregates with mortality, according to results from an exploratory analysis virtually presented during the 2020 European Hematology Association Annual Congress.1

“There is evidence of significant endotheliopathy that we show for the first time, and it’s characterized by both elevated Von Willebrand factor (VWF) antigen levels that are sky high, and factor VIII (FVIII) levels, soluble thrombomodulin, among others,” lead study author George Goshua, MD, a hematology fellow in the Hematology and Oncology Fellowship Program at the Yale University School of Medicine said during the oral presentation.

A prominent feature of SARS-CoV-2 pathogenesis is novel coronavirus-19 (COVID-19)–associated coagulopathy (CAC), which is described by increases in thrombotic and microvascular complications. In an effort to better understand the pathophysiology of CAC, investigators evaluated markers of endothelial cell and platelet activation, PAl1, and coagulation factors in both stable and critically ill patients with COVID-19 who had been hospitalized for their infection.

The research started with the first patient admittance to Yale New Haven Hospital on March 13, 2020 who had tested positive for the virus. The hospital saw its peak in cases a little more than a month later, on April 21, 2020, when 450 patients were simultaneously hospitalized for COVID-19 infection. Since that time, COVID-19–related admissions have been declining, according to Goshua. The first death related to the virus was reported in mid-March 2020.

“In reviewing through the deaths at our hospital, the very first death may have actually been due to COVID-19–associated coagulopathy,” said Goshua. “Within the right atrium was a large mobile hyperechoic mass with fibrinous strands extending through the tricuspid valve into the right ventricle. It was felt that this finding represented clot in transit, resulting in massive pulmonary embolism.”

The patient was in his 40s, and his only comorbidity was hypertension, noted Goshua; he was on no medications otherwise and had been doing well.

“For researchers and clinicians, this is not a surprise. We have known about the hypercoagulable nature of this coagulopathy, both from our Chinese colleagues and their publications,” said Goshua. “As this virus has spread around the world, however, the general public has gotten to know this coagulopathy, as well.”

On May 26, 2020, the Associated Press published an article examining a patient with COVID-19 who also had blood clots. It was noted that these blood clots that can cause strokes, heart attacks, and dangerous blockages, are being detected more often in patients of all ages who have been infected with the virus.2 BBC News also covered the topic, touting that up to 30% of patients who have fallen seriously ill with the virus are developing such clots.3

The laboratory derangements that are often observed include a significantly elevated D-dimer, significantly elevated fibrinogen, mostly normal or slightly increased prothrombin time, and mostly normal or slightly decreased platelet count, according to Goshua. Clinically, venous thromboembolism is seen in 17% to 69% of COVID-19 cases, arterial thrombosis is reported in 3.6% to 4.0% of cases, and microvascular thrombosis is observed in as high as 45% to 87% of cases.

Investigators wanted to understand the mechanism of coagulopathy in an effort to tailor treatment to patients. To this end, investigators looked at a cross-sectional cohort of patients who had been hospitalized with COVID-19; this included 48 patients in the ICU and 20 non-ICU patients. Most ICU patients were intubated, while non-ICU patients were on 3 or fewer liters of oxygen via nasal cannula.

With regard to general demographics, there was a significance difference in gender, with more men than women in the ICU patient population. Of the entire population of 68 patients, 60% were men and 40% were women; of the ICU and non-ICU patients, 69% and 40% were male, respectively, and 31% and 60% were female, respectively.

In their analysis, investigators evaluated hemostatic markers, which were broken up into 4 groups. D-dimer and thrombin–antithrombin (TAT) levels were significantly higher in ICU patients versus non-ICU patients, and these 2 markers were also found to be significantly correlated. Investigators hypothesized that the source of the high D-dimer in CAC is augmentation of the coagulation cascade, explained Goshua.

The second group included endogenous anticoagulants—anti-thrombopoietin, protein C, protein S—and fibrinolytic enzymes—alpha 2-antiplasmin. These markers were largely preserved in both ICU and non-ICU patients with COVID-19. The hypothesis was that CAC is distinct from disseminated intravascular coagulopathy.

The third group includes plasminogen activator inhibitor-1, which was found to be universally high in both non-ICU and ICU patients. Investigators were also able to measure LY30, which was found to be normal in 93% of ICU patients. The hypothesis is that the classic fibrinolysis is inhibited, said Goshua.

The last category included VWF and FVIII, which were found to be high in non-ICU patients and significantly elevated, “to sky-high levels” in ICU patients, stressed Goshua.

Investigators partnered with our vascular endothelial cell biologists and platelet biologists to run 3 different research assays. Investigators hypothesized that endothelial cells and platelet activation markers would be elevated in ICU patients with the virus.

“Because these are research assays and there are no well-defined upper and lower limits of normal for them, we obtained blood from 13 healthy, nonhospitalized individuals without COVID-19 to compare it to,” said Goshua.

Soluble CD40 ligand, which is found in both platelets and lymphocytes as a source, was found to be significantly elevated in the ICU patients with the virus versus the controls. When the maximum amplitude on thromboelastography was measured in these patients, 65% of ICU patients were found to have elevated maximum amplitude. “This is notable, of course, because the contribution of platelet to maximum amplitude is significant,” noted Goshua.

Investigators noted that soluble P-selectin, the source of which is endothelial cells and platelets, has been found to be significantly elevated in ICU patients with COVID-19 versus the controls; this was found to be true in non-ICU patients with the virus compared with the controls, as well.

Soluble thrombomodulin, the source of which are endothelial cells, was another marker of significance that the investigators evaluated. Because of the sample size, however, the results did not reach significance in the ICU patients with COVID-19 versus the controls. “That being said, given the fact that thrombomodulin has a significant role in the coagulation cascade, we set out to [see if this can be used to] predict mortality, and indeed, at a level of 3.26 of soluble thrombomodulin, we plotted Kaplan Meier curves and saw that the curves in all patients and in ICU patients separate quite nicely,” said Goshua.

Goshua added that at the hospital, investigators instituted an escalated intensity anticoagulation regimen as part of the hospital algorithm. This effort was based on D-dimer and on clinical suspicion. Based on those data, the hospital system looked into the utilization of aspirin for all patients who had been hospitalized with the virus.

If D-dimer is lower than 5 mg/L, all patients should be given standard prophylactic anticoagulation and aspirin 81 mg daily unless contraindicated. Furthermore, if D-dimer is 5 mg/L or higher, or the patient is receiving convalescent plasma, weight-based intermediate prophylactic anticoagulation and aspirin given at 81 mg daily should be given, unless contraindicated. If the patient has confirmed venous thromboembolism or high clinical suspicion, a therapeutic dose of anticoagulation and aspirin at 81 mg daily should be started, unless contraindicated.

Other therapeutic agents, such as defibrotide, dipyridamole, and eculizumab (Soliris) are under investigation in several clinical trials being conducted at Yale University School of Medicine.

“We show for the first-time evidence of platelet activation with elevation of soluble CD40 ligand and elevated maximum amplitude, and augmented coagulation cascade in both non-ICU and ICU patients with COVID-19 as evidenced by the thrombin antithrombin levels,” said Goshua. “Preserved natural anticoagulants and an inhibition of classical fibrinolysis as shown by universally elevated PAl-1 levels as well as normal LY30. As such, endotheliopathy is a marker of progression to critical illness in COVID-19 and soluble thrombomodulin segregates with mortality in an exploratory analysis."


  1. Goshua G, Pine A, Meizlish M, et al. Endotheliopathy is essential in COVID-19 associated coagulopathy. Presented at: 2020 European Hematology Association Congress; June 11-21, 2020; Virtual. Abstract LB2605. bit.ly/2YC8HF9
  2. Tanner L. Dangerous blood clots pose a perplexing coronavirus threat [news release]. Associated Press. Published May 26,2020. Accessed June 14, 2020. bit.ly/2MZeFKF
  3. Galpin R. Coronavirus: a third of hospital patients develop dangerous blood clots [news release]. BBC News. Published May 16, 2020. Accessed June 14, 2020. bbc.in/2YNrQ7h