Jul 3, 2012 (CIDRAP News) – News reports about possible airflow problems at the US Centers for Disease Control and Prevention (CDC) biosafety level 3 (BSL-3) lab in Atlanta are fueling perceptions that the agency isn't following its own rules, but the reported conditions at the lab don't appear to have posed a public health threat, according to some expert observers.
An investigative series by USA Today over the past few weeks has cast a spotlight on concerns that CDC employees raised internally about an incident in February in which air from a lab blew into a corridor where visitors weren't wearing respiratory protection and about engineering changes in the building that apparently reversed airflow. The reports caught the attention of a congressional committee, which recently launched an investigation with a request for CDC documents, according to a Jun 22 CNN report.
The latest USA Today report on the lab raised security concerns, citing internal agency e-mails that cited repeated incidents of security doors left unlocked where experiments on pathogens were underway. In one instance, an unauthorized employee was found inside a restricted area, according to the Jun 27 story.
CDC officials have responded with assurances that the labs are safe and haven't been linked to any contamination events or threats. However, CDC spokesman Tom Skinner told USA Today that the CDC is considering having outside biosafety experts, such as those from an Army biodefense lab or the Canadian government, review safety and biosecurity at the lab. The CDC conducts the safety reviews of its own labs, as well as outside research labs that work with select agents. Skinner also said the unsecured doors are not an acceptable practice at the CDC.
Questions surrounding safety and security at the CDC lab come in the wake of controversy that has swirled around what level of threat is posed by experiments with airborne-transmissible H5N1 viruses and at a time when the security community, the public, and even some scientists are still jittery about the role played by government microbiologist Bruce Ivins in the 2001 anthrax letter attacks.
CIDRAP News asked two biosafety experts to review the news reports and offer their analyses.
Joseph Kanabrocki, PhD, is assistant dean for biosafety and associate professor of microbiology at the University of Chicago. He has overseen select agent safety at several institutions and serves on two biosafety advisory groups for the National Institutes of Health (NIH). John Keene, DrPH, is managing partner of Global Biohazard Technologies and president of Biohaztec Associates, based in Midlothian, Va. He was manager of corporate biosafety at Abbott Laboratories, oversaw biosafety at a variety of other institutions, and is past president of the American Biological Safety Association.
Both experts said they are not familiar with the layout of the CDC labs and don't have data or measurements on airflow at the facility in question, but added that there was enough information in the stories to make some observations and offer a few insights.
Kanabrocki and Keene said they worried that the stories leave the public with the impression that the lab conditions are posing a biosafety hazard. Kanabrocki said BSL-3 labs such as the one at CDC have redundant safety systems in place to protect workers and the public, and that pathogens are always handled inside primary containment cabinets when working at BSL-3. He said the air blowing into the corridor in the February incident would have been a hazard only if that airflow reversal happened in conjunction with a major spill that accompanied a break in back-up systems.
Keene said, "I think the first thing to get across to the public and Congress is that containment does not mean contaminated." He added that BSL-3 and BSL-4 labs aren't considered contaminated unless a massive spill occurs outside of the primary containment device, such as the very small chance of a centrifuge exploding at the same instant an exhaust system fails.
He also said the majority of organisms for which BSL-3 conditions are required, such as Francisella tularensis and Bacillus anthracis, are found normally in nature, and scientists are taking them out of their natural environment and growing them in high numbers that result in a higher risk. "That's why we build the biosafety labs," he added.
Concerns about the reported air leak into the corridor appear to be somewhat of an overreaction, Kanabrocki said. Though the puffs of air in the corridor suggest some sort of positive airflow in the labs, which should have negative air flow, "Based upon the information presented in the article, I'm not convinced that the rooms are running under positive pressure all the time," he said.
Kanabrocki said he senses some confusion about airflow requirements that might stem from a recent subtle language change in the latest edition of Biosafety in Microbiological and Biomedical Laboratories (BMBL), a set of scientist-written biosafety guidelines that federal officials require for labs under select agent rules.
The previous edition of BMBL, in effect until 2 years ago, advised against "sustained" positive airflow in labs. However, in the most recent version of the BMBL (5th edition), the emphasis was changed from preventing a "sustained positive pressurization of the laboratory" to providing a "sustained directional airflow by drawing air into the laboratory from 'clean' areas toward 'potentially contaminated' areas" and ensuring that the "laboratory shall be designed such that under failure conditions the airflow will not be reversed." He added that the article appears to reflect an interpretation of the new language literally to mean that positive airflow in the lab shouldn't occur, not even for a second.
Ventilation systems in labs can be complicated from an engineering perspective, and even the best systems that are designed to maintain negative pressure in the event of a power outage or fan failure may take a few seconds to reverse the airflow, Kanabrocki said. Those transient pressure changes aren't dangerous, he said. "What it translates to is not a break in containment," he said, adding that a containment breach would require a power failure to occur simultaneously with a lab spill.
The earlier versions of the BMBL prohibited positive excursions for "a sustained period." Keene said some people wanted a definition of the "sustained period" of time for positive airflow. "The fact is that there is no definition and the decision would have to be made on an individual basis, based on a risk assessment of what was being done in the laboratory," he said.
More recent versions of the BMBL stated a zero tolerance for positive airflow, but "There's no such thing as absolute," he said. "There is going to have to be a reasonable determination made based on the probability of the actual release of agents at a level that could be potentially hazardous."
He agreed that there are some flaws in the way the BMBL is written and used. The document was written by scientists and was not designed as a regulatory document. Therefore, there has never been a true regulatory process with public comments, and the "framers" of the document have not had to justify the information in the document to the public. Technically, neither the CDC nor the NIH is a regulatory agency, Keene said.
Both experts agreed that the BMBL has always been a guidance tool, not a regulatory document. However, the Select Agent Act incorporated the BMBL as a "regulatory requirement," and as such it has become a "performance based" standard for biocontainment.
The two men said they're puzzled by the allegation in USA Today's initial story that the BSL-3 lab was directly accessible from the "clean" corridor, given rules that stipulate there must be an anteroom between the public area and the BSL-3 work area.
The story said CDC records suggest it never built the anterooms, because of cost concerns and delays. Kanabrocki said the buffer areas, typically used for donning and doffing personal protective gear, are routinely required for select agent labs, and it would be a breach of design criteria and protocol if one of the CDC's own labs didn't have them.
Tom Skinner, a spokesman for the CDC, told CIDRAP News that the anteroom issues are under discussion at the agency. "After this evaluation is complete, the agency will take appropriate action if necessary to further address the situation," he said.
Keene said design requirements from the NIH and CDC require at least two doors between the containment space and public areas, and if the BSL-3 laboratories lacked the airlocks that they knew were required, that would represent a break from the CDC's own rules. "If they didn't provide that separation, it doesn't make sense," he said.
The CDC could ease public misperceptions and address concerns about the agency inspecting its own biosafety labs by being more open, such as by explaining clearly what protection systems are in place, he said. Responding after reports of problems surface, especially ones from disgruntled employees, often ends up sounding like making excuses, he added.
Keene also said bringing in outside experts to review the labs could help reduce the appearance of a conflict of interest and help to suggest solutions to any problems that are found. Reviewers should include experts who specialize in biocontainment, not just lab design or commissioning agents, he added.
The commissioning step provides proof that the building or facility was built to specifications, and an evaluation or certification step is meant to demonstrate that the process works to contain spills and other hazards, Keene said. Sometimes people use the terms synonymously, he said, emphasizing that the reviews represent two different processes. He added that certification also covers facility management plans and safety standard operating procedures to ensure that lab operations don't override engineering safety features.
"Properly designed and operated biocontainment laboratories are safe. The fact is, they're safer if they're reviewed on a regular basis by people, inside or outside the organization, who really know what to look for," Keene said. "It's not rocket science, and it's not overly expensive, but it is important."
Jun 13 USA Today story
Jun 22 CNN report
Jun 25 USA Today story
Jun 27 USA Today story
Dec 2009 BMBL, 5th edition