Read our COVID-19 research and news.

A man receives the AstraZeneca COVID-19 vaccine in Pará state in Brazil on 17 April. The vaccine is the cornerstone of a global plan to end the pandemic.


What’s the future of vaccines linked to rare clotting disorders? Science breaks down the latest

Science’s COVID-19 reporting is supported by the Heising-Simons Foundation.

Vaccine regulators have delivered a clear verdict: In most settings, the benefits of the COVID-19 vaccines made by AstraZeneca and Johnson & Johnson (J&J) far outweigh the small risk they will cause an unusual and sometimes deadly clotting disorder. But many questions remain about who is most at risk, how the risk-benefit calculus changes when cases fall, and what the side effects mean for the future of these vaccines, which use adenoviruses to ferry the gene for SARS-CoV-2’s spike protein into human cells.

A major concern is how the rest of the world will respond to some European countries’ moves to limit the use of the AstraZeneca and J&J vaccines, and the brief suspension of the J&J shot in the United States. The AstraZeneca vaccine—named Vaxzevria, or Covishield when it’s produced by the company’s Indian partner, the Serum Institute of India—is the cornerstone of the COVID-19 Vaccines Global Access (COVAX) Facility, a scheme aimed at vaccinating billions in the developing world. J&J is expected to provide hundreds of millions of doses of its one-shot vaccine to COVAX this year.

“Once there are clear policies from the West regarding in what age groups to use these vaccines, it’s going be very hard to recommend anything different here,” says John Amuasi, an epidemiologist at the Kwame Nkrumah University of Science and Technology in Ghana, one of the first countries in the world to receive the AstraZeneca vaccine through COVAX. But if no other vaccines are available, limiting use of these could ultimately cause many more COVID-19 cases—and deaths.

How big is the risk, and who is most vulnerable?

Signs the AstraZeneca vaccine could lead to an unusual reaction that causes clots throughout the body, accompanied by low levels of platelets, first surfaced 2 months ago. Many of the first cases of what scientists now call vaccine-induced immune thrombotic thrombocytopenia or thrombosis with thrombocytopenia syndrome (TTS), were in women under the age of 60. But that may just be because many European countries used the shots in health care workers and educators, most of whom are women and under age 65. Indeed, the gender imbalance has started to even out as more cases came to light. Among 209 people affected in the United Kingdom, 87 were men and 120 women; 139 cases were in people younger than 60. Overall, one in roughly 120,000 AstraZeneca shots has triggered the side effect in the country.

In Sri Lanka, the health minister told Parliament last month that at least six people had developed the clotting disorder among nearly 925,000 who received the vaccine, or one in 150,000 recipients. Germany has reported cerebral venous thrombosis (CVT), an unusual type of stroke that is characteristic of TTS, in roughly one in 76,000 recipients of the vaccine.

The rate was higher in Norway and Denmark, where roughly one in every 40,000 AstraZeneca vaccine recipients developed CVT, with the frequency of other clotting events possibly even higher. Whereas most European countries have recommended using the vaccine in older recipients, Norway and Denmark have recommended against using the vaccine at all for now. Denmark announced today it will not use J&J’s vaccine either.

The symptoms of TTS closely resemble a condition called heparin-induced thrombocytopenia (HIT), a rare autoimmune reaction that is triggered by the blood thinner heparin. Despite decades of research, doctors can’t predict who’s at risk of HIT; it seems to affect men and women, old and young alike.

The same may be true for TTS. “We cannot identify any predisposing factors,” Beverley Hunt, a hematologist at King’s College London, said on a 29 April webinar sponsored by the Danish Health Authority. There’s no sign that a history of blood clots or other clotting risk factors—such as taking birth control pills—increases the risk of TTS; even people who have previously had HIT don’t seem at a higher risk. And it is not clear whether the risk differs between the first and second doses of the AstraZeneca vaccine.

Across countries, roughly one in five patients with the clotting disorder has died. Health authorities hope publicizing the early signs of TTS and how to treat it can help prevent fatalities. But the severe clotting is difficult to treat outside of a well-equipped hospital, so many recipients in rural areas or regions with limited health infrastructure will have little recourse.

Do other vaccines using an adenovirus vector have the same problem?

Early data from the United States suggest J&J’s vaccine does. So far, U.S. regulatory agencies have reported 15 cases of TTS in about 7 million vaccinees. That’s a lower frequency than the one seen in Europe with the AstraZeneca vaccine, says Klaus Cichutek, president of Germany’s regulatory agency, the Paul Ehrlich Institute. But J&J’s rollout is just beginning in Europe, he cautions: “With vaccinations starting here, we will see whether these numbers hold true.”

Less data exist on the other two adenovirus-based vaccines, Russia’s Sputnik V and Convidecia, made by the Chinese company CanSino Biologics. CanSino said in mid-April it is watching for similar clotting disorders and had not received any reports. Fewer than 1 million shots had been given then; the vaccine rollout is just beginning in Malaysia, Pakistan, Mexico, and other countries this month. The makers of Sputnik V say there have been no cases of clotting disorders among recipients. But it is unclear how many people have received the vaccine, and many of the countries where it has been distributed may be hard-pressed to diagnose the syndrome.

Before COVID-19, the only adenovirus-based vaccines in use were Ebola vaccines developed by J&J and by CanSino. There were no reports they caused the rare side effect, but it is unclear how many people received the Chinese vaccine. The J&J shot has been given to about 200,000 people, according to the company. Stanley Plotkin, a veteran vaccine developer and an emeritus professor at the University of Pennsylvania (UPenn), notes that low platelet counts have been described after adenovirus infections, which typically cause the common cold, but can occasionally trigger severe infections. So it’s possible, he says, that the problems with AstraZeneca and J&J may be related to their adenoviral vectors. “But that has to be settled in the laboratory.”

How do the AstraZeneca vaccine’s risks and benefits stack up?

For an older person in an area with lots of infections, the benefits vastly outweigh the risks. For a young person in a place where the pandemic is ebbing, they may not.

Guidance issued by the European Medicines Agency on 23 April showed vaccinating 100,000 people ages 80 and older in an area with high infection rates—886 infections per 100,000 people per month, the level seen in Europe in January—would prevent 1239 hospitalizations and 733 deaths over a period of just 4 months (see tables, below). At the low infection rate seen in September 2020—55 per 100,000 per month—151 hospitalizations and 90 deaths would be prevented. In both scenarios, only 0.4 cases of TTS would be expected in those 100,000 people.

By contrast, vaccinating 100,000 people age 20 to 29 would lead, on average, to 1.9 cases of the blood clotting disorder. But it would not prevent any deaths from COVID-19, although it would prevent 64 hospitalizations in an area with high infection rates.

There are other things to consider, however, says Jeremy Farrar, an infectious disease expert who heads the Wellcome Trust. “Don’t underestimate the impact of Long COVID,” he says. “These vaccines do appear to protect against that as well.” And vaccinating younger people not only protects them, but also helps keep them from spreading the virus to more vulnerable people in a community.

Weighing risks and benefits

The benefits of the AstraZeneca COVID-19 vaccine—in hospitalizations and deaths prevented over a period of 4 months—far outweigh the risks of a rare clotting disorder for most age groups, both when infections in a region are high (top) or low (bottom). All data are per 100,000 people vaccinated.

High infection rate
AgeHospitalizations preventedDeaths preventedCases of blood clots with low platelets
Low infection rate
AgeHospitalizations preventedDeaths preventedCases of blood clots with low platelets
European Medicines Agency

In many places, the decision is not whether to vaccinate, but whether to use AstraZeneca now or wait for another vaccine to become available. Denmark based its decision to stop using AstraZeneca and J&J altogether in part on the availability of alternatives such as the Pfizer vaccine. Hong Kong, which canceled its order of AstraZeneca, said it already had enough doses of other vaccines.

Even a long wait for an alternative may be worth it if infections are low. In Norway, the risk of dying from TTS for women age 45 to 49 is equivalent to the risk of dying from COVID-19 over the next 79 weeks, assuming the infection rate stays the same, Camilla Stoltenberg, director general of the Norwegian Institute of Public Health, told the 29 April webinar. “That group will have alternative vaccines, maybe with a small delay but nothing close to 79 weeks,” she said.

Is the calculus different in low- and middle-income countries?

In the coming, crucial months, the AstraZeneca vaccine is one of the great hopes in the struggle toward vaccine equity—ensuring that poorer countries will be able to vaccinate their populations as well. Many of them can’t afford the messenger RNA vaccines produced by Pfizer and Moderna, which are also more difficult to store and distribute because they need to be kept at a very low temperature.

But restrictions on the use of the AstraZeneca vaccine in wealthy countries could tarnish its reputation globally—and slow or even block the plan to vaccinate the world. “I think we’ve got ourselves into a really difficult position,” where it looks like there’s “one vaccine for the rich world and a different vaccine for the poor world,” Farrar says.

“Information travels, and regulators in other countries feel pressure to say, ‘We’re not giving our population a second-class or a poor-quality product,’” adds Ashish Jha, dean of the Brown University School of Public Health. The fact that 70% of Africa’s population is under 30, an age group that no longer gets AstraZeneca in Europe, only serves to highlight the inequity, Jha says. Yet for African countries to shun the vaccine would be a terrible decision, he says, because it is very effective and safe in the vast majority of cases.

The European pause has already slowed vaccinations in the Democratic Republic of the Congo (DRC). The country received 1.7 million doses from COVAX in early March but delayed its rollout when European countries hit pause. Vaccinations started on 19 April, but the government said last week it would return 1.3 million doses because it can’t use them before they expire in June. (The vaccine will be redistributed to other countries.)

The situation reminds some observers of their experience with RotaShield, a vaccine protecting children from rotavirus. The vaccine in rare cases led to intussusception, a condition in which a part of the intestine folds in on itself, leading to a potentially fatal bowel obstruction. In the United States, where rotavirus infections can easily be treated, government authorities in 1999 recommended stopping use of Rotashield. (Wyeth, its manufacturer, had already withdrawn the vaccine from the U.S. market by then.)

At some point, we’re going to need a global strategy that shifts away from the AstraZeneca vaccine towards others.

Ashish Jha, Brown University School of Public Health

The decision made many developing countries wary of the vaccine, says pediatrician and vaccinologist Paul Offit of UPenn’s Perelman School of Medicine, even though in their case the benefit vastly outweighed the risk. Offit recalls a 2000 World Health Organization (WHO) meeting about Rotashield: “Country after country stood up at the end of that meeting, and said, and I quote, ‘If it’s not safe for America’s children, then it’s not safe for our children.’” A new vaccine became available 7 years later. “But 2000 children died a day for 7 years,” Offit says.

“I think RotaShield traumatized a generation of vaccine and immunization people,” says WHO epidemiologist Kate O’Brien. But so far, there has been little sign of countries refusing the AstraZeneca vaccine because of other countries’ decisions, she says. And since the Rotashield drama, lower income countries have become more confident and experienced in making regulatory decisions for their own context, O’Brien says. “I think we’re in a really different place now.”

Of course, the public also has to want a vaccine. Gagandeep Kang, a virologist at the Christian Medical College in Vellore, India, says many patients are calling her about the clotting disorder. “So it is affecting vaccine confidence to some extent,” she says. “But I also don’t think that we really have a choice, because we don’t have the infrastructure to distribute the Pfizer and Moderna vaccines.”

In Ghana, “It’s difficult to tell how people think about this vaccine, because it’s not here,” Amuasi says: The country has run out of doses to administer, a far bigger problem than safety worries. Amuasi received his first shot earlier this year, but his appointment for the second shot had to be canceled because the country has no more doses. Ghana is scheduled to receive 350,000 of the doses going unused in the DRC in the coming days.

What is the future of the adenovirus vector vaccines?

Once the global pandemic begins to retreat, the relative importance of even a small vaccine risk will increase. As other inexpensive, easy-to-distribute shots based on different technologies are developed, adenovirus-based vaccines will play a smaller role. “At some point, we’re going to need a global strategy that shifts away from the AstraZeneca vaccine towards others,” Jha says.

Such vaccines are already being developed in Cuba, Kazakhstan, Mexico, and elsewhere, notes Hilda Bastian, an independent scholar of evidence-based medicine. “A lot of these countries are coming up with their own answers. It’s just been slower,” she says. If these new vaccines are approved and gain trust, “then this time next year could look very different,” she says.

*Update, 4 May, 10:45 a.m.: This story has been updated to include the fact that Ghana is scheduled to receive vaccine doses from the DRC. A quote from Klaus Cichutek has been added.

*Correction, 5 May 2021, 6.30 a.m.: EMA calculated the risk of COVID-19 deaths and hospitalizations over a period of 4 months, not 4 weeks as stated in a previous version of this story.