Many people who have long COVID—a condition in which health issues persist months after infection—report struggling with “brain fog,” recurring memory and concentration lapses that make it difficult to function in everyday life. Now a new study has found these cognitive problems could result from blood clots triggered by infection, possibly through mechanisms like those that cause some types of dementia. These clots leave telltale protein signatures in blood, suggesting that testing for them could help predict, diagnose and possibly even treat long COVID.
The findings, published on Thursday in Nature Medicine, suggest that existing blood tests for detecting these proteins could help physicians recognize long COVID (although some experts caution that long COVID’s symptoms and causes likely vary among individuals). Up to 15 percent of people who contract SARS-CoV-2, the virus that causes COVID, go on to develop long COVID symptoms that can last for months or years. The condition has proved difficult to treat—or even diagnose—because of the mishmash of reported symptoms such as brain fog, fatigue, respiratory problems and numerous other effects. It’s still unclear whether the virus sticks around in the body or the initial infection triggers another reaction, such as an autoimmune response, that leads to the continuing symptoms.
To look into this question, lead study author and psychiatrist Maxime Taquet of the University of Oxford followed more than 1,800 people in the U.K. who were hospitalized for COVID between 2020 and 2021. Taquet and his colleagues checked in with the patients six and 12 months after their initial infection to monitor any prolonged symptoms and give them a cognitive test designed to diagnose disorders such as Alzheimer’s.
When the researchers examined blood tests taken while the patients were hospitalized, they found that blood from people who still had brain fog six or 12 months after infection tended to have elevated levels of at least one of two proteins in their blood. The first protein, called D-dimer, is produced when blood clots break down in the body. Although patients with high levels of D-dimer reported memory problems, they did not score lower on cognitive tests. They were more likely to have shortness of breath and fatigue than other patients, however. Taquet suspects these effects could have been caused by blood clots in the lungs, which can also lead to low oxygen levels in the brain.
The second protein, fibrinogen, is produced in the liver and causes clotting to stop bleeding. Patients who had elevated fibrinogen levels during active COVID infection not only reported memory impairment but also scored poorly on the cognitive test. On average, people in this group scored under 86.7 percent six months after infection—a test result that could indicate dementia. Taquet says these findings suggest that either fibrinogen caused blood clots in the brain or that clots elsewhere in the body affected the brain, causing more severe cognitive symptoms.
Blood tests for both D-dimer and fibrinogen are already standard around the world, so researchers had plenty of available data to check whether the same phenomenon occurred in other groups of patients. In a separate analysis, the team examined health records from nearly 50,000 people in the U.S.—including some who had taken blood tests for fibrinogen or D-dimer before the pandemic started. Brain fog with elevated levels of D-dimer, they found, only occurred in people who had had COVID. But elevated fibrinogen seemed to correlate with brain fog whether or not a person previously had COVID, which indicated that cognitive problems caused by other conditions may also involve fibrinogen.
The new findings are “very exciting,” says Resia Pretorius, a physiologist at Stellenbosch University in South Africa. Her own research has found links between brain fog in long COVID and “microclots” in the blood. These clots often contain misfolded fibrinogen proteins that prevent the clots from breaking down, and this effect can clog blood vessels and block oxygen flow to the brain and other organs. Pretorius suspects that SARS-CoV-2’s infamous spike protein interacts with the fibrinogen protein and causes it to change its shape.
Fibrinogen has previously been linked to cognitive difficulties, particularly vascular dementia. Studies have also shown that mice injected with fibrinogen can develop cognitive problems. It’s unclear whether dissolving clots could reverse existing damage to the brain or whether the lack of oxygen kills neurons permanently, says Sidney Strickland, an Alzheimer's researcher at Rockefeller University in New York City.
Strickland says it’s possible that clot-dissolving anticoagulant medications (often called blood thinners) could help alleviate long COVID symptoms, yet this has not been directly tested in clinical trials. He adds that these drugs need to be used cautiously because they can increase the risk of fatal blood loss. “You definitely want blood clots when you need them,” Strickland says, “but not after the fact.”
Blood thinners are a blunt instrument that affects several different ways in which the body can clot blood. Strickland’s lab is developing antibodies that target just one of these pathways: the one that involves fibrinogen. These treatments shouldn’t cause hemorrhages like generalized anticoagulants do, he says. Such treatments could be tested in those with long COVID, he adds, but his lab is still gearing up for clinical trials in people with dementia.
In the meantime, Strickland says, researchers still don’t know what the exact relationship between COVID and fibrinogen is or what the protein does during long COVID. It may cause lung problems that affect blood flow or leaks in the blood-brain barrier. “We need to have a lot more information,” Strickland says.
Avindra Nath, a neuroimmunologist at the National Institute of Neurological Disorders and Stroke, says the new results are interesting. He is skeptical about the blood clotting proteins’ role in long COVID, however. Nath notes that the study only included people whose COVID infection was so severe that they were hospitalized and likely had organ damage. Long COVID can also occur in people who experienced few symptoms during active infection, he says, and their long-term conditions may have different causes. Nath’s group has found blood clots in the brains of people who died of COVID, but he says more research is needed to determine if these clots caused their symptoms.
In the meantime, some physicians have noticed that blood-thinning medications may help certain long COVID patients. David Joffe, a respiratory physician at Royal North Shore Hospital in Sydney, Australia, says he prescribes anticoagulants because they tend to work better at restoring brain function in some long COVID patients than other treatments—although he also says that they need to be used carefully because of their inherent risks.
Joffe adds that the condition is difficult to study and treat because its symptoms vary so much. “Long COVID is lots of things,” he says. “As a physician, I don’t see two patients who are identical.”
The variety of symptoms could be caused by different systems that the virus triggers, Joffe says. COVID infection has been linked to metabolic problems caused by cell damage and inflammation in the brain, as well as autoimmune conditions in which the immune system attacks the body. But the virus can also suppress the immune system, allowing underlying conditions to become active and cause their own array of symptoms.
Long COVID “is so much more complicated than people imagine,” Joffe says.
Taquet says his team’s results don’t prove blood clots and vascular problems are the root of long COVID. “There is room for other hypotheses as well,” he says. His group is now working with brain-imaging data from people with long COVID that could reveal more precisely where blood clots occur, as well as a more sensitive cognitive test that could show how brain impairment happens over time.