When things get out of hand at the lab: Hundreds of accidents expose the ‘catastrophic’ risk of dangerous pathogen leaks

The virologist Xavier Abad reflected on the source of the COVID-19 pandemic when the virus had already killed nearly three million people. “Is it unlikely that SARS-CoV-2 escaped from a laboratory? Yes, but it’s damned possible,” he warned on his blog in March 2021. Abad is the head of the biocontainment unit at the Animal Health Research Center (CReSA), the Barcelona institution that was searched this Thursday by the Spanish Civil Guard and the Mossos d’Esquadra (Catalan regional police) looking for evidence of a suspected leak of the African swine fever virus. “I’ve read about dozens of incidents and accidents in laboratories, out of the hundreds recorded worldwide, which are just the tip of the iceberg of those that ACTUALLY occur,” the virologist cautioned.

All hypotheses remain open, but the focus is now on CReSA, a bunker containing dangerous pathogens located on the Bellaterra campus of the Autonomous University of Barcelona, in northeastern Spain. The laboratory, essential for the search for new vaccines and treatments, was undergoing some construction work and experimenting with the virus in late November, when the first wild boar infected with a very similar strain was found just a few hundred meters from the facility, which lacks double fencing. Most Spanish experts on African swine fever insist that this virus is not easily transmitted through the air, so they cannot imagine how it could have escaped from a biosafety level 3 laboratory—BSL3, the second highest—like CReSA. Abad himself offered an explanation regarding a possible leak of the coronavirus from a Chinese laboratory: “Biocontainment units can be considered impenetrable, aseptic, extremely controlled, and restricted-access fortresses, but they are NOT free from errors and unfortunate coincidences.”

The virologist wasn’t exaggerating when he spoke of hundreds of known cases worldwide. In the last half-century, at least 435 incidents of laboratory-acquired infections have been recorded, according to a review published in March by Costa Rican researcher Esteban Zavaleta. Accidents occur even in facilities with the highest biosafety conditions, BSL4 labs, as happened at the Bernhard Nocht Institute for Tropical Medicine in Hamburg, Germany, on March 12, 2009. That day, a virologist had just injected the deadly Ebola virus into a mouse when the needle pierced his three gloves and punctured his skin, though he didn’t bleed. He did not develop the disease.

Opacity surrounds these incidents in many countries. A decade ago, an international team of scientists attempted to determine how many accidental infections had occurred in high-security laboratories, those considered level 3 or 4. The researchers sent a questionnaire with 15 simple questions to some 120 institutions, but only 23 responded. One of them acknowledged two people infected with the bacteria that causes Q fever, a disease of ruminants that can jump to humans. Another center admitted two cases with the microbe that causes brucellosis, an undulant fever that can last for years. Among the authors of the survey was the virologist Núria Busquets, who works at CReSA.

The signatories denounced at the time that “some laboratories are reluctant to disclose their accidents.” The real risk, they lamented, “is difficult to quantify because there is no systematic reporting system.” Their review of published cases detected 220 people infected with highly dangerous pathogens in laboratories between 1980 and 2015. Busquets’ team at CReSA underscored that there were also leaks affecting livestock, as happened in the English town of Pirbright, where a damaged pipe in two BSL3 laboratories caused a leak of the foot-and-mouth disease virus in 2007. Construction trucks facilitated the spread of the pathogen to farms in the region, resulting in millions of euros in losses. Dump trucks are now constantly entering and leaving CReSA, which has been undergoing expansion work for the past three months, but the facility maintains that it has not registered any biosafety breaches.

The American epidemiologist Marc Lipsitch prefers not to comment on the specific case of African swine fever in Spain, but he emphasizes that, on other occasions, “infectious material has escaped from very high containment labs, in some cases higher than BSL3.” Lipsitch, director of the Center for Communicable Disease Dynamics at Harvard University, cites three examples that have occurred since 2014: a mix-up of vials that resulted in the release of an Ebola virus sample from a level 4 laboratory in Atlanta; the failure of anthrax spore inactivation protocols at a military facility in Utah; and the shipment from a level 3 laboratory to the U.S. Department of Agriculture of material inadvertently contaminated with the highly pathogenic H5N1 influenza virus.

Cases are occurring constantly. Stuart Blacksell’s team at the University of Oxford has described 16 episodes of infection in BSL3 laboratories and another five in BSL4 facilities between 2000 and 2024. Their figures are staggering. If research laboratories of all types are taken into account, 276 infections and eight deaths have been recorded in the same period: two deaths from mad cow disease, one from Ebola, and one each from hantavirus, monkey herpes simplex virus type B, bacterial meningitis, severe acute respiratory syndrome (SARS), and the plague. In Spain, a biochemist from the Bellvitge Biomedical Research Institute died in 2022 after experiencing symptoms consistent with Creutzfeldt-Jakob disease, the lethal illness he was studying at the University of Barcelona and at CReSA itself.

A report by the British organization Chatham House warned two years ago of the “potentially catastrophic consequences” of a laboratory accident. A particularly serious example could be the 1977 flu pandemic, which killed approximately 700,000 people. The virus, very similar to others decades old that were being experimented on in laboratories at the time, may have leaked from a Soviet facility, according to biologist Michelle Rozo, a biotechnology risk advisor to the U.S. government, and immunologist Gigi Kwik Gronvall, a biosafety expert at Johns Hopkins University.

The source of the COVID-19 pandemic, which has claimed more than seven million lives, remains unclear. A group of independent experts established by the World Health Organization (WHO) declared on June 27 that “the available evidence suggests a jump from animals, either directly from bats or through an intermediate host,” but the possibility that the virus escaped from a laboratory in Wuhan has not been ruled out. The Chinese government is refusing to share essential information, according to the WHO.

Chatham House is one of the world’s most influential think tanks. Its report from two years ago warned that “the true scale of laboratory accidents is opaque.” The authors—analyst Emma Ross and microbiologist David Harper, both former WHO collaborators—documented 309 people infected in laboratories in nearly 100 incidents involving 51 different pathogens between 2000 and 2021.

Chatham House specialists recorded 16 leaks from scientific facilities during the same period, including the airborne release of Brucella bacteria from a vaccine factory in Lanzhou, China, which caused more than 10,000 cases of brucellosis in the surrounding area in 2019. Most of the incidents were due to preventable human error. Expired disinfectants were being used in Lanzhou. The biosafety profession is in its infancy in many parts of the world, the Chatham House report warned.

Alexandra Peters, from the Institute of Global Health at the University of Geneva (Switzerland), has been warning for years about another threat to the world: the number of high-security laboratories worldwide is unknown. Peters points out that BSL3 facilities have strict measures in place, such as double filtration of exhaust air, mandatory showers for scientists, chemical decontamination of effluents, and waste incineration. “However, a BSL3 lab is not always as secure as one would expect,” she cautions. “Personally I don’t think it’s impossible for any virus to escape because, ultimately, a lot of how laboratories keep pathogens safe is through human behavior,” she reflects. “And even if it’s not human error directly, it can be issues in the built environment or just bad luck.”

Sign up for our weekly newsletter to get more English-language news coverage from EL PAÍS USA Edition