Bioterror drill goes awry, lab workers are exposed

Outmoded, dangerous practices discovered

An exercise designed to test laboratory readiness for a bioterrorism incident turned into a real-life disaster of another sort when specimen mislabeling and flagrant breaches in infection control resulted in several exposures to an attenuated vaccine strain of Brucella abortus RB51, the Centers for Disease Control and Prevention reports (CDC).1 Attenuated vaccines are live virus particles that grow in the recipient, but they don't cause disease because the vaccine virus has been altered.

"In any sort of bioterrorism event or large scale outbreak, you may have lots of clinical specimens coming through a laboratory, and laboratorians themselves can be exposed," says Lisa Rotz, MD, director of the CDC's Division of Bioterrorism and Response. "We saw the same thing in the SARS outbreaks and in other outbreaks when there is large number of specimens coming through. It is very important for laboratorians to be aware of what's going on and how they need to handle certain types of specimens."

In this case the specimens were an attenuated vaccine strain of B. abortus used to vaccinate cattle against brucellosis. Human infections with the vaccine strain have been documented in previously published reports. One study looked at accidental exposure to RB51 via needlestick in 21 people, conjunctival spray exposure in four, and spray exposure of an open wound in one individual.2 At least one systemic symptom was reported in 19 (73%) people, including three who reported persistent local reactions with systemic involvement. One case required surgery.

In 2006, the Laboratory Preparedness Survey (LPS), a voluntary proficiency test, was revised to include attenuated organisms such as RB51 that more closely mimic those on the CDC list of bioterrorism agents. The LPS is designed to simulate a scenario in which the presence of a bioterrorism agent is suspected in a clinical laboratory and the pathogen must be ruled out or referred for additional measures. Last fall, an LPS kit containing simulated or modified strains of pathogens identified as potential bioterrorism agents, including RB51 for the first time, was distributed to 1,316 laboratories throughout the United States and Canada. The RB51 vaccine strain was added after participating sentinel laboratories indicated a need for a more "realistic" exercise. No one could have imagined how realistic the situation was about to become.

In November 2007, the New York State Department of Health notified the CDC of potential exposures to the Brucella vaccine strain in many clinical laboratories that participated in the LPS test. The LPS kit included written instructions stating that all samples should be handled inside a Class II biological safety cabinet (BSC) with biosafety Level 3 (BSL-3) primary barriers and safety equipment. Basic laboratory procedures performed included preparing specimens for culture by reconstitution and inoculation onto appropriate media, preparing and performing a Gram stain, and possibly performing biochemical spot/slide tests (e.g., oxidase, indole, or catalase), the CDC reports.

Mislabeling leads to exposures

The exposures initially occurred after an RB51 specimen was mislabeled as a routine patient specimen and submitted by an LPS-participating laboratory to the New York state bacteriology laboratory. "It was human error," Rotz says.

As a result, routine benchtop procedures were used by unsuspecting laboratory personnel to handle the isolate, which resulted in 24 workers with potential exposure to RB51. However, further investigation determined that 16 LPS-participating laboratories in New York had not handled the RB51 samples properly, despite correct labeling of the samples. Thus, exposures occurred even when the samples were appropriately labeled.

"With this particular exercise, they were to already assume that these were [bioterrorism] agents and start handling them a certain way from the very beginning," Rotz says. "We are looking at whether the instructions were clear enough or whether we need to make any adjustments with that."

After those exposures were reported, the CDC urged all participating state health departments to review biosafety practices used by LPS labs in their states while working with the RB51 sample. The investigation eventually included 36 states, two cities, one county, and the District of Columbia. The sites reported 916 laboratory workers with exposure to RB51, including 679 (74%) with high-risk exposures and 237 (26%) with low-risk exposures. Post-exposure prophylaxis (PEP) was recommended for people with high-risk exposures and offered to those categorized as having low-risk exposures. Rotz said she was not aware of any infections that had occurred.

"As far as I know, we don't have any updated numbers from that [MMWR] report," she says.

'Sniffing' increases risk

As part of the investigation, the CDC developed a set of questions identifying the types of manipulations and aerosol-generating procedures that might result in exposure. RB51 exposure was deemed to have occurred if the specimen was handled in a manner other than the established recommended practice (i.e., working inside a Class II BSC using BSL-3 primary barriers and safety equipment). Some lab workers were deemed to have high-risk exposures after reporting inappropriate practices such as smelling or "sniffing" lab cultures to try and determine the agent.

"That is definitely not the usual practice," Rotz emphasizes. "It is a practice that some laboratorians were taught many years ago as one of the 'diagnostic' things you could do. Certain types of cultures smell a certain way. It is certainly not encouraged in this day and age with increased biosafety standards."

Lab workers also were considered at high risk of exposure if they were within 5 feet of any manipulation of RB51 on an open bench or if they were present in the laboratory during a widespread aerosol-generating event (e.g., vortexing) involving the specimen. The CDC investigation underscores the need for routine adherence to recommended biosafety practices when working with infectious organisms, particularly during widespread infectious-disease events.

"One of the reasons we do this type of study is to make sure they know the right protocols to be able to work up an isolate that could be a potential [bioterrorism] agent," Rotz says. "They also need to understand how they need to handle these types of specimens without putting themselves at risk. [We are looking at whether] people need to have refresher training courses."


1. Centers for Disease Control and Prevention. Update: potential exposures to attenuated vaccine strain Brucella abortus RB51 during a laboratory proficiency test — United States and Canada, 2007. MMWR 2008; 57:36-39.

2. Ashford D, di Pietra J, Lingappa J, et al. Adverse events in humans associated with accidental exposure to the livestock brucellosis vaccine RB51. Vaccine 2004; 22:3,435-3,439.