Sloppy lab practices by an experienced but overworked scientist rushing to get to a noon meeting is likely how the Centers for Disease Control and Prevention cross-contaminated a specimen of a benign bird flu virus with a dangerous strain that can kill people, according to an internal agency investigation released Friday.
The CDC investigation found a wide range of serious lapses and revealed additional flu research that was jeopardized because of the contaminated samples. The CDC scientist may have even handled both the benign strain of bird flu and the dangerous H5N1 strain inside a biosafety cabinet at the same time, the report concludes, which would be a significant breach of basic procedures that carries risks of cross-contaminating specimens.
The CDC report revealed for the first time that contaminated bird flu specimens were not just sent to a U.S. Department of Agriculture lab in Athens, Ga., which discovered the contamination issue May 23 after chickens in their research became seriously ill and died — something that shouldn't have happened from the virus strain they thought they were working with.
A contaminated sample of the benign bird flu virus also was sent to another CDC lab, where it was still being used in experiments more than a month after the mistake was discovered because nobody alerted that lab to the problem. In addition, CDC had also planned to send a sample from the contaminated virus batch to the infectious disease department at St. Jude Children's Research Hospital, which has a prestigious influenza research facility.
The investigators' report provides the first detailed account the bird flu virus contamination incident that occurred earlier this year and the troubling six-week delay between its discovery and notification of top CDC officials. The case is one of several recent, serious safety and security lapses at federal research labs, which have also involved mishandling of anthrax by the CDC and the discovery of forgotten vials of deadly smallpox in a cold-storage room at the National Institutes of Health.
CDC officials emphasized that the bird flu incident did not appear to have posed a safety risk because all of those who worked with the contaminated virus samples handled them with the same precautions that would have been used if scientists would have knowingly been working with the dangerous H5N1 strain. The H5N1 flu strain, which primarily causes disease in birds, has the potential to significantly harm the U.S. poultry industry if it spreads to flocks outside the labs. In rare instances, the H5N1 virus has caused severe and fatal illnesses in people.
"We were very lucky, in this episode, there was no risk to human health," said Anne Schuchat, director of the CDC's National Center for Immunization and Respiratory Diseases, in an interview with USA TODAY. "But being lucky with that doesn't mean we're not taking this seriously."
Schuchat said the series of mistakes made in the incident by experienced lab scientists was "unacceptable," adding: "We just don't think shortcuts are permissible when working with these kinds of dangerous pathogens."
The report outlines a number of actions the CDC is taking to improve lab safety at the agency and within its influenza lab, including new quality-assurance procedures, better record-keeping, and prioritizing workloads for lab scientists. The agency had already closed the flu lab involved in the incident and has been reviewing safety and security procedures at all of the agency's labs that work with dangerous viruses, bacteria and toxins.
The CDC said it is testing all cultures and other preparations done in the past year by the scientist involved in the cross-contamination incident. If other contamination is discovered, additional testing of all preparations done by the scientist may be warranted, the report said. It's unclear how long this will take, Schuchat said, but it is important that all specimens leaving the agency only contain what they should.
The lab scientist is not named in the report, which gives no indication of whether the employee has been disciplined. Schuchat said the agency is not going to release names of staff or specifics of any personnel actions being taken in the bird flu and anthrax incidents. "There are a number of disciplinary actions in process for the two incidents," she said, including suspension, reassignment, written reprimand and counseling of staff.
The head of CDC's bioterrorism response lab at the agency's Atlanta headquarters resigned July 22 after being reassigned to other duties in the wake of his staff's mishandling live anthrax spores in June, prompting concerns that dozens of the agency's employees may have been exposed to a particularly lethal strain of the bacteria. None has shown any signs of infection, and CDC officials have said the risks are remote.
"We are holding individuals accountable," Schuchat said Friday. "But we also are focusing very much on the system. We want redundant systems so that when an individual makes a mistake, we have a second or third protective layer."
An investigation of CDC's bird flu bungling by lab regulators from the U.S. Department of Agriculture, sent to the CDC on July 29, found many of the same things found by the internal investigation, the CDC said. The CDC and USDA jointly run the Federal Select Agent Program, which that monitors labs working with potential bioterror agents. Neither CDC nor USDA lab staff made a required timely report to select agent regulators about the bird flu incident, the report said.
According to the CDC report, the contamination of the sample occurred on Jan. 17, when the CDC lab scientist worked on multiple tasks with cell cultures involving samples of the dangerous H5N1 bird flu strain that had recently arrived from Vietnam, as well as a sample of the benign H9N2 bird flu virus that had arrived from China.
The scientist initially told investigators that best practices and procedures had been followed that day in one of the agency's biosafety Level 3 labs — where scientists wear special protective gear, including respirators, and manipulations of dangerous viruses and bacteria are done inside of special biological safety cabinets. The surrounding lab space is also operated under negative air pressure to keep dangerous viruses and bacteria contained should a spill occur outside the safety cabinet.
The scientist said that work was first done with the benign strain sample, then the biosafety cabinet was decontaminated, and then the work was done with the dangerous H5N1 samples. The entire process done properly would have taken at least 90 minutes, the report said. At least 30 minutes would have been needed just to clean the biosafety cabinet.
But investigators' review of key card logs showed the work couldn't have been done in the manner described by the scientist. Just 51 minutes elapsed between the time the scientist accessed a specimen freezer to obtain the virus samples and when the scientist exited the biosafety Level 3 lab suite. The scientist manually signed out of the suite at 11:45 a.m.
"During the 51 minutes, the Team 1 laboratory scientist also would have been required to shower out of the suite and change into street clothes. Therefore, the time that this staff member performed the cell culture work was substantially less than the 1.5 hours that would have been required if the protocol had been followed," the report said.
Because there were no lab notebooks or other documentation of what the scientist did in the lab, the CDC investigation concluded that it's not possible to say conclusively what actions occurred and in what order. "When interviewed, the Team 1 laboratory scientist acknowledged being rushed to attend a laboratory meeting at noon." The scientist said it was difficult to remember specifics of what had happened almost six months earlier.
At the time of the incident, the flu division had a heavy workload because of an approaching World Health Organization vaccine meeting, the scientist and others in the division told investigators.
The report said possible explanations for how the contamination occurred include that the scientist handled both virus types in the biosafety cabinet at the same time, or that the scientist handled the dangerous H5N1 samples first and then didn't follow appropriate decontamination procedures before working with the benign strain.
"You don't want to have these multiple strains in the same place at the same time," Schuchat said, noting that what happened wasn't a result of a lack of training, as the scientist had "great experience."
CDC investigators said that failures of the scientist and the scientist's supervisor to notify those above them of the incident for weeks after it was discovered May 23 was partially a result of "ambiguous" guidance on what kind of incident requires reporting and is considered a notifiable release. Still, the report said, "sound professional judgment would have warranted that the Team 1 lead report it immediately to a supervisor."
Instead, higher-ranking CDC officials not involved in the mistake weren't notified until June 23. On that day, a CDC researcher who had been unwittingly using a specimen from the contaminated batch became concerned that there was a problem with the virus sample after it produced unexpected results in the experiment. This researcher contacted the scientist and supervisor whose lab had provided the contaminated sample and raised concerns that a lab error had occurred.
Only then did the scientist and supervisor report the mistake to their branch chief. Still, as the branch chief gathered more information, it took until July 9 before CDC's senior leadership was notified, the report said. CDC disclosed the incident to the public on July 11.
The CDC's investigation found that the delays were a result of "poor judgment" and confusion about what needed to be reported to others. "There's absolutely no cover-up," said Schuchat,
The CDC closed the flu and anthrax labs involved in the recent mistakes and, on July 11, imposed a moratorium on all transfers of biological specimens to and from CDC's high-containment labs until each lab's safety practices are assessed. High-containment labs work with the most dangerous viruses, bacteria and toxins.
On July 24, an internal CDC lab safety work group had given the OK for CDC's Clinical Tuberculosis Laboratory to resume transfers of certain TB samples used in tests to quickly tell doctors whether their patients carry multi-drug-resistant TB strains and which drugs will be most effective in their treatment. But the moratorium remained in place for the agency's other high-containment labs, the CDC said. The agency is conducting lab-by-lab reviews of safety procedures, giving priority to those directly involved in patient care. So far, specimen shipments have resumed involving 12 of the 23 labs covered by the moratorium, including the Ebola lab, Schuchat said.
Follow USA TODAY investigative reporter Alison Young on Twitter: @alisonannyoung