People who had COVID-19 were about eight times more likely to end up with a blood clot on their brain than somebody who took the AstraZeneca vaccine, according to a new study.
Clots are extremely rare in both cases, but occured significantly more in people who got sick from COVID-19, according to a comparison of two large data sets by experts at the University of Oxford.
The study, released on Thursday, has not yet been peer-reviewed, the formal process by which scientific findings are scrutinized and challenged.
Nonetheless, it provides an insight into the fraught decisions being made by regulators around the world when deciding how to respond to reports of rare occurrences of blood clots in people who have taken the vaccine.
Many nations in Europe paused use of the AstraZeneca vaccine, and later restricted who could receive it, based on the reports of clots, and US officials made a similar decision this week for the Johnson & Johnson shot. Both use similar technology.
The study looked at instances of cerebral venous sinous thrombosis (CVST), particular type of clot, in people who had the AstraZeneca vaccine and also in people who had no vaccine and caught COVID-19.
The COVID patient data was 500,000 records from TryNetX, which collects electronic health records in the US. The AstraZeneca patient data was from figures compiled by the European Medicines Agency.
It concluded that the risk of getting a CVST within two weeks of a COVID-19 diagnosis was 39 in a million.
This compared to a 5-in-a-million risk after one dose of the AstraZeneca vaccine, around eight times smaller.
The comparison complicates the decision-making involved in pausing the vaccines, raising the prospect the pauses could actually lead to a net increase in blood clots if more people get COVID-19 as a result of fewer vaccinations.
The data did not discuss the Johnson & Johnson vaccine, but a rough estimate for the risk there is around one in a million, far smaller than with AstraZeneca.
Professor Paul Harrison, one of the experts involved in the study, said: "COVID-19 markedly increases the risk of CVST, adding to the list of blood clotting problems this infection causes.
"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 signal that COVID-19 is linked to the CVST "is clear, and one we should take note of," Dr. Maxime Taquet, another Oxford expert.
The scientists said that the numbers should be taken with some caution, because they are comparing risks found in a US database to the risk for the AstraZeneca vaccine that was taken from European data.
Another limitation to the study is that the scientists could only look at the risk of the cerebral blood clot, whereas the safety signal that was picked up by the European Medicines agency was linked to the CVST with low platelets.
This news came after the European Medicines agency on April 7 the unusual blood clots with low platelets as a potentially lethal, but very rare, severe adverse event to the AstraZeneca vaccine
The Johnson & Johnson vaccine was temporarily paused in the US on Tuesday while health agencies investigate cases of very rare blood clots that were reported after taking the jab.
Experts said that there is currently no definite link between the vaccines and these rare types of blood clots — establishing causality is significantly harder than noting a correlation.