Before the session opened for general discussion, several speakers addressed a number of issues related to the topic including whether adding additional engineering controls always increases safety, the reality that regulations and guidelines rarely keep up with technology, and the benefits of making risk assessments more quantitative and pathogen-specific. Speakers also commented on the need for and difficulties in obtaining quantitative data and offered suggestions to address this problem. Following the presentations, Michael Callahan (Defense Advanced Research Projects Agency, United States) led a discussion.
BREAKOUT SESSION PRESENTATIONS
Evidence-Based Biosafety: Ensuring Precautions are Adequate and Appropriate
Allan Bennett (Health Protection Agency, U.K.) discussed the main causes of laboratory-acquired infections (LAI) and argued that the combination of building engineering, equipment, and practices commonly used today are neither economical nor maximally effective. While the best direct evidence that labs are not serving as sources of infection and that a set of precautions is effective would be accurate counts of the number of LAIs, lack of governmental reporting requirements make such data scarce. In the absence of direct evidence, he suggested that applied biosafety data could serve as indirect evidence.
Mr. Bennett started by pointing out that of the three main routes of laboratory infection—inhalation of aerosols, surface contact with the agent, and punctures from needles or other sharps—aerosols have historically received the most attention. Starting in the early 1980s, regulations began requiring a number of expensive technologies to reduce aerosol exposure including HEPA filters, directional air flow, multiple air exchanges per hour (ACH), and biological safety cabinets (BSC). Since then, materials and practices have evolved to offer additional ways to minimize the creation of aerosols, such as the use of sealed centrifugation rotors and the substitution of plastic flasks and bottles for glass ones. He believes that many research spaces, as a result, are now over-engineered. This can cause both unnecessary expense and a reduction in overall safety. As an example, he pointed out that many of the precautions used to reduce aerosol exposure (e.g., flexible film isolators, half suit isolators, respirators, and Class II BSCs) reduce vision and manual dexterity, which can increase splashes (Sawyer et al., 2006). As such, he believes that when evaluating protocols the whole set of precautions and their side effects should be considered collectively.
Glove usage and hand hygiene constitute another set of practices that Mr. Bennett believes should be altered. For example, the thick household gloves used in Class III BSCs and in BSL-4 conditions significantly decrease both gross dexterity of the hand as well as fine finger dexterity. Similarly, although latex gloves cause no loss in manual dexterity (Sawyer et al., 2006) and are critical for preventing direct contact infections, many BSL-2 lab workers, even in high resource countries, do not use latex gloves consistently. He cited a study conducted at the University of Utah and presented at the 2010 American Biological Safety Association (ABSA) Conference where James Johnston4 (University of Utah) found that only 46 percent of staff removed gloves on leaving a BSL-2 lab, hand hygiene compliance before exiting a lab was 10 percent, and 72 percent of individuals touched their face while working. In the study, compliance varied widely between labs and could not be predicted by training.
Mr. Bennett also stated that in the developed world, workers might become overly reliant on engineering and let down their guard with respect to biosafety procedures and good
4 James Johnston, Ph.D., C.I.H.: Hand Hygiene in the Biosafety Level-2 Lab: Is it a Matter of Training? (Tuesday, October 5, 2010) ABSA 53rd Annual Biological Safety Conference, September 30-October 6, 2010 Denver, CO. Available: www.absaconference.org/pdf53/Session8-Johnston.pdf.