fb-pixel Skip to main content

US closes CDC flu, anthrax labs for errors

After potentially serious back-to-back laboratory accidents, federal health officials announced Friday that they had closed the flu and anthrax laboratories at the Centers for Disease Control and Prevention and have halted shipments of all infectious agents from the agency’s highest-security labs.

The accidents, and the CDC’s emphatic response to them, could have important consequences for other laboratories engaged in efforts to produce dangerous viruses and bacteria.

If the CDC — which the agency’s director, Dr. Thomas Frieden, called “the reference laboratory to the world” — had multiple accidents that could, in theory, have killed staff members and people outside, there will undoubtedly be calls for stricter controls on other university, military, and private laboratories that handle pathogens.


In the second accident, disclosed Friday, a CDC lab accidentally contaminated a relatively benign flu sample with a dangerous H5N1 bird flu strain that has infected 650 people since 2003, killing 386 of them. Fortunately, an Agriculture Department laboratory realized the strain was more dangerous than expected and alerted the CDC.

The first accident occurred last month. As many as 75 CDC employees may have been exposed to live anthrax bacteria after potentially infectious samples were sent to laboratories unequipped to handle them. Workers who were not wearing protective gear ended up moving and experimenting with samples of the highly infectious bacteria that were supposed to have been deactivated, the agency said.

In addition to those mistakes, Frieden also said Friday that two of six vials of smallpox recently found stored in a National Institutes of Health laboratory since 1954 contained live virus capable of infecting people.

All the samples will be destroyed as soon as the genomes of the virus in them can be sequenced. The NIH will scour freezers and storerooms for more dangerous material, he said. “These events revealed totally unacceptable behavior,” Frieden said. “They should never have happened. I’m upset, I’m angry, I’ve lost sleep over this, and I’m working on it until the issue is resolved.”


Frieden said that the accidents had implications for labs beyond his agency, arguing that the world needs to reduce to absolute minimums the number of labs handling dangerous agents, the number of staff involved, and the number of agents circulating.

Scientists doing the most controversial work — efforts to make pathogens more lethal or more transmissible — say the research helps predict mutations that might arise in nature so vaccines can be designed. But other scientists feel that creating superstrains is unacceptably dangerous because lab accidents are more common than is acknowledged, as Frieden’s statement indicated.

At the CDC itself, Frieden said, staff who knowingly failed to follow procedures or who failed to report dangerous incidents would be disciplined. A panel of experts will be convened to revise procedures.

In the flu-related incident, a CDC lab accidentally contaminated a sample of less-dangerous H9N2 bird flu that it was preparing for shipment to an Agriculture Department laboratory with the H5N1 bird flu strain.

Although the contamination was discovered May 23, Frieden said he was dismayed to find that senior CDC officials were not informed until July 7, and he was told only 48 hours ago.

Nonetheless, he said, “we have a high degree of confidence that no one was exposed.” The flu material was handled in high-biosafety-level labs in both agencies and the workers wore breathing apparatuses.


In theory, the flu-related accident could have been much worse than the anthrax one.

Anthrax can kill people who inhale it, but is not normally transmitted between humans, so an infected laboratory worker presumably could not have gone home, fallen ill, and passed it on. Although H5N1 flu is not easily transmitted between humans, it is believed to have done so in a few instances, and is fatal in about 60 percent of known cases.

The CDC released a report Friday on its inquiry into the anthrax episode, which occurred June 5 in the agency’s bioterrorism rapid response lab in testing a new mass spectrometry method. The intent was to see whether the technique was adequate for state and local laboratories testing mystery samples for anthrax.

The report found several errors: A scientist used a dangerous anthrax strain when a safer one would have sufficed, had not read relevant studies, and used an unapproved chemical killing method.

The error was found by accident. A door to an autoclave that would have sterilized samples taken for safety tests was stuck, so they were left in an incubator days longer than usual. Only then did a technician see that bacteria believed dead were growing.

Later tests done at the CDC and at a Michigan state health department lab as part of the investigation confirmed that the chemical method would have killed any live, growing anthrax in the samples that were sent out, but might not have killed all spores, which can also be lethal.


All the workers were offered vaccine and antibiotics.

For the flu lab, that will be finished in time for vaccine preparation for next winter’s flu season, he said.