ABOVE: © ISTOCK.COM, DESIGN CELLS

A COVID-19 diagnosis is associated with a 39-in-1-million chance of developing a rare blood clot condition, compared with about a 4-in-1-million chance after receiving the Pfizer or Moderna mRNA vaccines against the disease, according to a data analysis led by researchers at the University of Oxford. The study, posted Wednesday (April 14), has not yet been peer reviewed. The findings add weight to concerns that suspending the use of other vaccines, namely, AstraZeneca’s and Johnson & Johnson’s adenovirus vector–based shots, might not be worth the tradeoff of leaving people without protection against SARS-CoV-2. 

Last month, some European countries halted use of the AstraZeneca jab due to blood clot concerns. The European Medicines Agency later determined cerebral venous sinus thrombosis (CVST, also known as cerebral venous thrombosis or CVT) is an extremely rare side effect of the shot. Then, this week, US health authorities recommended a pause in Johnson & Johnson vaccinations after a handful of blood clots were diagnosed in recipients. Federal health agencies are investigating whether they are related to the vaccine. 

For the new study, researchers used a database of electronic medical records of patients based primarily in the US to determine the risk of blood clots after vaccination versus infection. Of the 513,284 COVID-19 patients identified in the database, 20 developed CVST within two weeks after their diagnosis. Two out of 489,871 patients who received an mRNA vaccine developed CVST. As a control, the researchers also analyzed the risk of developing CVST within two weeks of a flu diagnosis, which was zero in 1 million.  

The study notes that the European Medicines Agency’s latest estimate for the risk of CVST associated with the AstraZeneca vaccine is 5 per 1 million. The study did not address the Johnson & Johnson vaccine.  

“We’ve reached two important conclusions,” study coauthor Paul Harrison tells CBS News. “Firstly, COVID-19 markedly increases the risk of CVT, adding to the list of blood clotting problems this infection causes. Secondly, the COVID-19 risk is higher than seen with the current vaccines, even for those under 30; something that should be taken into account when considering the balances between risks and benefits for vaccination.” 

The study is not the first to link COVID-19 to an elevated risk of blood clots. Aaron Petrey, an immunologist at the University of Utah who was not involved in the Oxford study, writes in an email to The Scientist that for both COVID-19 and the vaccines against it, the risk appears to be tied to antibodies against a protein called anti-platelet factor 4. “I think the broad conclusion from this article is correct, and the risk of clots . . . is significantly higher in COVID because you have multiple mechanisms that all feed together which can result in coagulopathy—from endothelial damage to platelet hyperreactivity and the cytokine storm associated with COVID-19 among others,” he writes. “There is a rich body of study that shows COVID is a highly prothrombotic state, and as a consequence, thrombosis and stroke occur in even young healthy individuals who have mild symptoms after infection.” 

See “The Specter of Endothelial Injury in COVID-19” 

“I would say the risk/benefit tradeoff is still vastly in favor of vaccination, though as the authors point out in that article they did not compare the different risk/benefit between vaccines,” Petrey adds. 

Pfizer disputes the study’s finding that mRNA vaccines were associated with a small risk of CVST, saying in a statement emailed to The Scientist, “With over 200 million doses having been administered globally, Pfizer has conducted a comprehensive assessment of ongoing aggregate safety data for the Pfizer/BioNTech BNT162b2 vaccine which provided no evidence to conclude that arterial or venous thromboembolic events, with or without thrombocytopenia, are a risk associated with the use of our COVID-19 vaccine.” The statement notes that recent reviews by the US Centers for Disease Control and Prevention and the UK’s Medicines and Healthcare products Regulatory Agency have also found no association between the Pfizer vaccine and blood clots.

Mary Cushman, a professor of medicine at the University of Vermont who specializes in hematology and was not involved in the study, says its findings are in line with the known connection between COVID-19 and abnormal blood clotting disorders. But, she cautions in an email to The Scientist, “The authors of this work did not apply adjustment for important risk factors for CVT or [venous thromboembolism] like age, that might differ across the groups of interest (COVID, influenza, covid vaccine), so it is difficult to fully interpret the results.”