In the early days of the New York Covid-19 outbreak, as March turned to April, hematologist Jeffrey Laurence was called to consult on the case of a 32-year-old bodybuilder. Nurses had noticed an odd rash on his buttocks, “as if you had kind of peeled away the skin layer and were seeing what blood vessels look like on his bottom,” recalls Laurence, who works at Weill Cornell Medicine in New York City. The vessels were outlined so clearly because the blood inside was coagulating, almost jelly-like.
Clotting is normally a good thing. When a blood vessel is injured, cell fragments called platelets rush to plug the leak. Proteins in the blood called clotting factors switch from dormant to active states in a chain reaction, and build a fibrous mesh. “It’s sort of a domino effect,” says Hanny Al-Samkari, a hematologist at Massachusetts General Hospital in Boston.
Clotting in uninjured blood vessels is a common occurrence in hospital patients, especially those in the intensive care unit. Being bedridden encourages clotting, especially in the legs and pelvis, and the clots may migrate to the lungs where they impede the organs’ ability to load the blood with oxygen. Depending on their location, clots can lead to problems such as breathing difficulties, heart attack, stroke and death.
Inflammation due to infection can also tip those clotting-factor dominoes. But as Covid-19 patients filled hospital wards, it became apparent that their clotting was more frequent, more widespread and more severe than in other infections. The clots filled needles used to draw blood, or the tubing connecting patients to medication drips and machines. “Everything was clotting,” Al-Samkari says.
The consequences can be devastating. In a July report in the journal Blood, Al-Samkari and colleagues found that nearly 10 percent of 400 people hospitalized for Covid-19 developed clots. In a February report by researchers in China, about 70 percent of people who died of Covid-19 had widespread clotting, while few survivors did. And in a July article in the New England Journal of Medicine, autopsies revealed that the lungs of people who died of Covid-19 were nine times as likely to be speckled with tiny clots as those of people who died of influenza. Major risk factors for severe Covid-19 — such as diabetes, obesity and advanced age — are linked to worn-out blood vessels that make clotting more likely, says John Atkinson, an immunologist and rheumatologist at Washington University School of Medicine in St. Louis.
What Laurence finds downright “spooky” is that all this clotting happens in spite of the common US practice of prescribing blood thinners, such as heparin, to hospital patients to ward off clotting.
Why does clotting go overboard in some people with Covid-19? Theories abound. One possibility, Al-Samkari speculates, is that the virus activates one of the clotting factors and jump-starts the domino effect — but there’s no specific evidence that this is happening.
Another idea is that because SARS-CoV-2 infects and damages the cells lining blood vessels, it could expose the tissue underneath. That tissue makes proteins that promote clotting and normally perform a vital function, Al-Samkari says: If blood vessels are injured, the proteins get into the blood and induce clotting to plug any leak.
A third possibility is that clotting results from inflammation. And here, many experts are eyeing a set of proteins called the complement system. These proteins, known collectively as complement, attack invaders and call in other parts of the immune system to assist. They also can activate platelets and promote clotting.
Like the clotting cascade, the proteins of the complement system are activated in sequence, and scientists now know that SARS-CoV-2 can directly activate one of them, Laurence says. So can damaged body tissues, which build up during the virus’s attack.
Clinicians have observed that the complement cascade appears to get out of hand in many people with severe Covid-19, says immunologist and complement expert Claudia Kemper of the National Heart, Lung, and Blood Institute, who coauthored an article about complement and immune cells in the Annual Review of Immunology. She and her colleagues found signs of complement activity in the lungs and livers of people who died from Covid-19, for example, and Laurence found several active complement proteins in the skin and blood vessels of his early Covid-19 clotting cases. “There is currently not super-super-hard evidence, but many complementologists think that this is a massive part of the disease,” Kemper says.
In another study of 11,000 people who had Covid-19 published in August in Nature Medicine, a New York team found that patients were more likely to become very ill and die if they had a history of clotting or bleeding, or if they had macular degeneration, which can indicate complement problems. The team also found that genes involved in complement and clotting responses were more active when the virus was present in patients’ nasal swabs.
Not only that, but the researchers also reported that people with certain variants of genes involved in the complement and clotting systems had a higher risk of severe Covid-19 disease.
In addition to complement, another immune element may promote clotting in severe Covid-19 cases: an overreaction called a cytokine storm, in which the body releases an excess of inflammation-promoting cytokine molecules. “Your whole system gets revved up,” Atkinson says. “When it’s revved up, your clotting system gets revved up, because it senses danger.”
As they treat their Covid-19 patients, physicians seek to hit the brakes on these clotting, complement and cytokine effects. “What you try to do is calm the trigger,” says Atkinson, who cowrote an overview of abnormal complement control in macular degeneration and a childhood disorder for the Annual Review of Pathology: Mechanisms of Disease.
Early in the course of infection, that trigger is the virus itself, so doctors reach for antivirals such as remdesivir. But later on, says Laurence, the body’s response is the biggest problem. “The virus, you might as well forget about it,” he says. “You’ve got to control the clotting, you’ve got to control the inflammation, you’ve got to control the complement pathway — and that’s easier said than done.”
For clotting, there are blood thinners like heparin. Hematologists are hotly debating how much to use for Covid-19 patients, Al-Samkari says, because doctors must balance the risk of clotting with the danger of bleeding. Al-Samkari has most often observed bleeds into the digestive system for these patients, but they may also hemorrhage in the lungs, brain or spots where medical devices pierce the skin.
Many hospitals are discharging Covid-19 patients with a prescription for blood thinners in case the risk of clotting remains high at home, though there are currently no solid data to back up this practice, Al-Samkari says. More than a dozen clinical trials aim to identify the right course of action to manage clotting alongside Covid-19.
Al-Samkari stresses that there is no evidence that people with less severe Covid-19, who do not require hospitalization, should take blood thinners or aspirin to ward off clots.
For some patients, stifling inflammation may help. Steroids such as dexamethasone calm the immune system, and other medications specifically block cytokines or individual proteins in the clotting and complement cascades. Argatroban, for example, is a Food and Drug Administration-approved anticoagulant that interferes with thrombin, an element of the clotting cascade. And eculizumab, which blocks one of the complement proteins, is approved for certain inflammatory conditions.
Again, physicians await better guidance from trials. “Right now,” says Al-Samkari, “we use clinical judgment as best we can, and just do our best.”
Editor's note: This story was updated on September 18, 2020, to correct an erroneous hyperlink and to clarify that a study originally published on a preprint server before peer review had in fact completed peer review and been published in Nature Medicine in early August. We have updated the link.