Posted by Guest Blogger Steven Niemi, director of the Center for Comparative Medicine at Massachusetts General Hospital and PRIM&R Board Member

To the Select Committee Appointed to Update the Guide for the Care and Use of Laboratory Animals:

I am writing to urge emphasis on performance standards over engineering standards in the next edition of the Guide, for the reasons provided below. For the purpose of this commentary, an engineering standard is any guideline or directive that includes a measurable unit (expressed as a number or range of numbers for time, space, mass, velocity, etc.) versus a performance standard that involves a desired outcome without quantified parameters.

First, some engineering standards commonly in use today are unnecessary or wrong. For example, the engineering standard of 180°F water for washing cages is a waste of energy and money if those cages are going to be sterilized before their use. We routinely use cold water (and no detergent) to sanitize rodent microisolator cages that will be autoclaved prior to their return to animal rooms; visual inspection and testing for microbial residue post-wash/pre-sterilization confirm those cages to be clean. Another example of inappropriate engineering standards involves maximum cage change intervals of 14 days. If ventilated mouse cage changing is performed faithfully on a 14-day cycle but too soon after a new litter of mice is born, there is risk of increased pup mortality (Reeb-Whitaker, et al., Laboratory Animals 35:58­ 73, 2001). A third example involves the ambient temperature in which laboratory rodents are routinely maintained, expressed as an engineering standard in the Guide of 18-26°C. While this temperature is comfortable for humans because it corresponds to our thermoneutral zone, the thermoneutral zone of mice and rats is 28-31°C. Maintaining laboratory rats and mice at or below 26°C significantly increases their blood pressure, heart rate, and pulse pressure because these animals feel cold (Swope, et. al., Am J Physiol Regulatory Integrative Comp Physiol 287:391-396, 2004). This is not to suggest that rodent rooms should be maintained at higher temperatures; that would result in higher energy costs and a workplace uncomfortably warm for personnel. However, like the other examples above, it does point out the risks involved with using engineering standards that may be easy to follow, but which avoid the need for justification.

Second, engineering standards discourage innovation. One could argue that every engineering standard is always accompanied with an allowance for effective alternatives or IACUC exemptions, thereby providing sufficient flexibility. But it’s been my experience in laboratory animal care over the past 35 years that engineering standards are almost always adopted verbatim because “they’re in the Guide” and become embedded in our policies and practices. This ingrained culture may forestall adoption of new modalities incorporating engineering and design, construction, and operational processes that could provide a more environmentally benign, less expensive and safer workplace without risk to animal health and welfare. For example, it is conceivable that the technology of conditioning air may advance to the point where 100% recycled air is not only cheaper but cleaner than 100% fresh air. This should render the engineering standard of 10-15 single-pass air changes per hour obsolete, but changing to the new technology would be conceptually difficult as long as that standard remains in the Guide. On the other hand, an updated Guide that promotes performance standards may encourage more creative thinking about how to improve rather than prolong established methods and constraints.

Third, the advent of genomics, in vivo imaging, nanotechnology, and synthetic biology may create novel situations in which laboratory animals will be used for research, testing, and education. Engineering standards may not be compatible with how animals are used with such technologies even though animal welfare would not be impaired in specific protocols. This could result in missed opportunities to advance knowledge and medicine. Furthermore, these same technologies may be used to better define animal comfort as well as detect pain or distress more accurately. Consider that someday we will be able to detect expression of particular genes in situ that indicate in real time when an animal is occasionally stressed (a positive state) versus chronically distressed (a negative state). We should employ such tools and signals to identify unmet animal needs and to challenge engineering standards, with the objective of continuously improving how animals are cared for and used. But sustained promulgation of engineering standards conflicts with the practice of continuous improvement and makes it less likely that these new technologies will be leveraged for advancing animal welfare.

In conclusion, laboratory animal needs and protections are necessary. The Guide provides standards describing those needs and protections with unmatched credibility and authority to an increasingly global audience. Animal needs and protections that are entrenched as engineering standards may be insufficient or inappropriate today or tomorrow. By contrast, performance standards that are evidence-based reaffirm past improvements as laboratory animal care and use has evolved and are more likely to facilitate similar progress in the future.

Respectfully submitted,

Steven M. Niemi, DVM, DACLAM
Director, Center for Comparative Medicine
Massachusetts General Hospital
149 Thirteenth Street, Room 5249
Charlestown, MA 02129 USA
e-mail: sniemi@partners.org