Seventy five lab workers may have been exposed to anthrax at the Centers for Disease Control and Prevention (CDC) in Atlanta, Georgia, the leading US lab for tracking infectious disease. The incident has not yet made anyone ill, and poses negligible risk for the public, but it raises concerns about work with deadly pathogens.
According to the CDC, culture dishes of anthrax bacteria kept in a level 3 high-containment lab, were subjected to a treatment that should have killed them, then sent to a level 2 lower containment lab in the same building on 6 June, to help develop detectors for environmental anthrax.
Workers there did not use heavy protective gear for procedures that could have spritzed the bacteria into the air as the anthrax was supposed to be dead. But on 13 June when the dishes were being gathered for disposal, a live anthrax culture was growing on one.
People thought to have been exposed have been given the anthrax vaccine and ciprofloxacin, the antibiotic that prevented illness when anthrax was sent to US government and media offices in 2001. Heads, the CDC says, will roll.
The danger from the anthrax is minimal, however. The CDC has not said whether the bacteria were in their growth phase or the more dangerous spore form, or whether they were the virulent Ames strain used in US biodefence work and the 2001 attack.
But even if they are, no one has shown symptoms yet, and antibiotics should protect anyone who inhaled bacteria. No one else is at risk, as anthrax cannot pass from person to person.
The incident has however rattled critics of "gain of function" experiments with flu, so called because they can turn flu that does not pose a pandemic threat into a virus that could. Unlike anthrax such flu is highly contagious, and harder to kill with drugs.
"If this type of problem can happen at a CDC lab, it points out the issues regarding gain-of-function lab accidents," says Michael Osterholm of the University of Minnesota, Minneapolis. In reply, flu virologists have insisted that they use very stringent safety procedures.
Accidents do happen, however. In 2011 the US National Research Council reported 196 "loss of containment" incidents involving dangerous pathogens between 2003 and 2009 at US government labs.
Today Mark Lipsitch of Harvard University and Alison Galvani of Yale University published a critique of gain-of-function flu work that emphasised the risk of lab escapes.
"While such a release is unlikely in a specific laboratory conducting research under strict biosafety procedures," they write, "even a low likelihood should be taken seriously, given the scale of destruction if such an unlikely event were to occur."
They say that such work is unethical if whatever public health benefit it provides can be obtained with less-dangerous experiments – or just by investing in better vaccines.