Effect of Common Variables on Autoclave Best Practices in Rodent Barrier Programs

Introduction

To maintain rodent colonies free from harmful infectious agents, laboratory animal care programs frequently employ the use of sterilized caging and supplies.¹² The Guide for the Care and Use of Laboratory Animals states that sterilization can be used to prevent opportunistic infections in the colony and to ensure that any hazardous experimental agents are killed before cleaning.¹⁴ Methods of sterilization for caging in the laboratory setting include steam sterilization, convective heat sterilization, and irradiation.^7,10,15 Laboratory animal programs should regularly confirm that their sterilization methods are working¹⁴; however, there have been no widescale studies assessing sterilization effectiveness under the most common conditions seen in laboratory animal programs.

Biologic indicators (BIs) represent a quick and affordable test to ensure quality in the sterilization process.²² BIs for steam autoclave validation contain a defined amount of live, bacterial Geobacillus stearothermophilus spores.¹⁶ These spores are incubated after steam sterilization to ensure that specific sterilization conditions are effectively met.² If there is bacterial growth in the BI media, represented by a color change, the sterilization process was not effective.¹³

Outbreaks can bring research to a halt, leading to lost research time and huge expenses in decontaminating facilities of infectious agents.⁵ In many cases, animals need to be euthanized and lines rederived when agents are discovered in a colony.¹² In a study looking at mouse parvovirus, all animals experimentally exposed to caging components contaminated with the pathogen became seropositive.¹¹ A mouse rotavirus outbreak occurred due to an unsterilized diet and bedding in shipping containers.¹⁸ Another study demonstrated that viral particles of mouse parvovirus and mouse norovirus remain infectious after pelleting feed but not after steam autoclaving.¹ As these studies show, contaminated fomites and caging can lead to outbreaks in a nontrivial manner. There is relatively limited literature on the steam autoclaving of reusable rodent microisolation cages. Clearly, sterilization of components that animals come into contact with is an important part of a biosecurity plan and issues can arise when sterilization failures occur.¹⁴

Beyond infection in animals, some pathogens used for research can cause illness in humans who encounter fomites. For example, lymphocytic choriomeningitis virus (LCMV), is a commonly used rodent pathogen to study the immune system. Infected animals are housed under animal biosafety level 2 (ABSL2) conditions because LCMV can cause disease in developing human fetuses.^24,26 Pathogens can be found both in living and dead animals. In recent years, several studies have evaluated autoclaving procedures for carcasses to prevent spread of infectious agents and identified autoclaving procedures to kill infectious organisms without the need for incineration.^6,17,21 Without effective autoclave practices, breaks in containment of these agents can occur, leading to illness and possible death.²⁵

The purpose of this study was to evaluate the most common processes encountered that may cause failures of steam sterilization of cages. We investigated effects on both clean cages and cages with soiled bedding. For clean cages, we hypothesized that cages closest to the autoclave floor drain and cages with more dense bedding types would have the highest probability of sterilization failures as measured by BI when stacked atop each other. For soiled cages, we investigated effects of bedding type, bagging method of cages, length of cage change interval, and presence or absence of diet and water on steam penetration and time bedding spent above 121 °C. We hypothesized that more dense bedding, cages bagged in plastic, long cage change interval, and cages with food and water present would lead to more sterilization failures and less heat exposure to the bedding as measured by biologic indicator and a load temperature probe.

Materials and Methods

Cages.

New Super Mouse 750™ mouse cages (model 75031-GAM; Lab Products, Seaford, DE) were used for all experiments. The cages were made of polysulfone. The cage dimensions were 34 cm long × 20 cm wide × 16 cm deep. The cages had stacking lugs 0.635 cm wide on each corner to allow for steam penetration between cages and to prevent the cages from sticking together. The cages had 1.5 mm of clearance for steam at the front and back of each cage and 1.3 mm of clearance on each of the sides. Cages were evaluated for integrity before each trial. Any cage with cracked or missing stacking lugs was excluded.

For the steam penetration study, the cages were stacked 8 cages high in 16 stacks on 2 carts (Figure 1), the maximum number of cages that can fit on these carts. The cages were filled with bedding as described below. The top cage of each stack had a Super Mouse 750 lid.

View Figure

• Download as Powerpoint
• Download Figure

For the extended program experiment, 6 lidded cages in a 3 high by 2 wide stack were bagged in either 60-gallon Earthguard 100% paper bags (PB; model no. 514625; Portco Corporation, Vancouver, WA) or autoclavable 44-gallon biohazard plastic polypropylene (PP) bags 2 mm thick (model no. 01-828E; Fisher Scientific, Waltham, MA) inside of an Earthguard bag. PP bags were closed by folding the plastic over on itself but not sealed, such that no cage could be seen but steam penetration could occur. The PBs were closed by folding the top over on itself and then sealing it with 3 strips of autoclave tape. Two such bags for each treatment group were loaded into the autoclave on a 3-sided cart, with the cages placed over each of the 2 drains (Figure 1). The sample size was determined by identifying the maximum number of cages that could reasonably fit in the bags. Upon removal from the autoclave, each bag was assessed for integrity. All autoclavable red plastic bags were sealed shut from the effects of the heat and vacuum.

Results

Discussion

Effective autoclaving of laboratory rodent cages is dependent on many factors, including the cage style, design, and integrity, the autoclave functionality, the cycle design, and what is being autoclaved.⁷ The Guide for the Care and Use of Laboratory Animals makes it a point to discuss autoclaving cages and the rationale for it.¹⁴ Our study is the first to address whether common practices of autoclaving cages in laboratory animal programs are efficacious. The authors chose to evaluate parameters such as steam penetration and effect of common processes encountered such as bedding type, bagging style, and presence of diet and water in the cage.

The initial question that was answered and which informed all the other experiments, was whether steam can penetrate into stacked cages and if there is an effect of position of materials to be sterilized in the autoclave. We saw adequate steam penetration for all cages, regardless of position in the stack or position in the autoclave as evidenced by lack of growth in the BIs. Prevacuum cycles critically depend on adequate vacuum to ensure that steam can penetrate throughout the machine.²⁵ We found all cages, no matter the bedding type, cycle length, location in the autoclave, or location in the stack passed the BI test. The HTLT cycle took about 10 to 15 min less time than the LTHT cycles for adequate steam penetration. While the shorter exposure cycle would allow for more materials to be sterilized within a typical workday, it is also important to consider the impact on plastic longevity. Polysulfone can withstand more heat than polycarbonate caging.²⁰

We next evaluated the autoclaving of dirty cages. Institutions may need to autoclave soiled cages such as ABSL2 cages before they enter cage wash processing.¹² We first evaluated the effects of bagging the soiled cages on steam penetration and time the bedding spent above 121 °C. Our hypothesis was that the plastic bags acted as insulators and the cages in the polypropylene bags inside the paper bags would spend less time above 121 °C as compared with the paper-bagged only cages. We chose to test the 2 bagging options because plastic bagging represents a significant waste of materials, as this plastic cannot be recycled, while paper is recyclable and therefore more sustainable. Polypropylene is a semicrystalline thermoplastic with a very high melting point of 160 to 166 °C, making it a useful plastic for steam sterilization.¹⁹ Cages with PP spent more time on average at or above 121 °C than the PB cages, but this difference was not significant (P = 0.0765). Regardless of the time spent above 121 °C, cages in both treatment groups passed the BI test.

Despite the small sample size, there was a significant difference in the time bedding spent above 121 °C for cages autoclaved with diet and water versus those without diet and water. The rationale for adding water to the cages was to simulate water being poured directly into the cage by staff under conditions where water cannot be disposed of directly into the drain. We hypothesize that the observed difference may be due to the high specific heat of water and the extra mass in the cage. Despite the fact that some cages spent up to 4 min less than what is theoretically required to kill Geobacillus stearothermophilus, all such cages passed BI evaluation. We hypothesize that this phenomenon is due to a cumulative heat effect. Geobacillus stearothermophilus is a spore with a known Z value, or the temperature increase necessary to kill 1 log unit of organism; this value is 10 °C.³ Heat over time, even below 121 °C, has an effect on lethality of the spore.⁸ This can be modeled using ${\text{F}}_0=\mathrm{\Delta }t{\displaystyle \sum 10\frac{\left(T-121 \text{°C}\right)}{Z}}$ where Δt = the time interval between measurements, T= temperature at time t, and Z = 1 °C. F₀ shows that heat is additive over time and represents the equivalent lethality in minutes of any heat exposure at 121 °C.

We did not see any difference between the time the bedding spent above 121 °C of cages at 2- compared with 4-wk time points, supporting the idea that cage change schedule does not affect the temperature that the substrate reached or the steam penetration. The autoclaves ran several degrees warmer than the set point, which may have contributed to the fact that the materials easily all passed BI evaluation. All the bedding did eventually reach 121 °C and surpassed it by several degrees Celsius, presumably because the machine often ran 2 to 3 °C hotter than it was set. While this is a confounding variable, we do not plan to reprogram the machine because of state law requirements that the machine be set to at least 121 °C.⁹ There was a single positive BI result in the steam penetration study. We hypothesize this may be a false positive due to contamination with external bacteria. When cracking the inner glass ampule, the cap can lift because of the increased pressure in the tube. It is possible that the cap partially came off and bacteria entered the vial.

The authors do not want to overstate these results. The results found for these experiments apply only to the autoclaves at our institution and the Super Mouse 750 mouse cages (model 75031-GAM; Lab Products, Seaford, DE). It is essential that each institution validate its own autoclave practices to confirm efficacy. This is an aspect of cage component sterilization that is frequently overlooked but is critical for animal welfare and research integrity.

California regulations state that medical waste must be autoclaved at 121 °C for at least 30 min.⁹ The “at least” in the regulations is important. This guidance notes a minimum amount of time. There may be situations where even cumulative heat exposure may not lead to properly autoclaved cages when the machine is set to 121 °C for the bare minimum of 30 min. Very dense loads or loads that are very wet may need more than 30 min to be adequately autoclaved. It is up to the institution to confirm adequate decontamination of cage materials, and our results demonstrate a possible method for doing so.

Regular preventive maintenance is critical to effective and efficient autoclave operation.⁷ Bowie-Dick tests should be run often using a Bowie-Dick cycle according to manufacturer instructions. The Bowie-Dick tests used in these experiments showed that the machines had fully functional vacuum pumps at the time of experiment. BIs should be placed in the locations where steam is most unlikely to penetrate.² Our experience has identified numerous autoclaves at our institution that were in operation and had successfully passed BI evaluation but were unable to pass Bowie Dick tests. When all of the air is not removed from the autoclave, there may be air pockets which cause failures.⁴ We speculate that not all cages would be appropriately sanitized if there are zones with air pockets, but this was not investigated as part of the present study, because the authors wanted to focus on properly operating autoclaves. When using only a single BI per load, a passing Bowie-Dick test is even more important.

Future directions include testing different caging systems as well as testing autoclaves that consistently fail Bowie-Dick tests, and evaluation of cages with cracked stacking lugs. Another area of possible study is the environmental and economic impact of short- compared with long-duration autoclave cycles. In addition, testing the soiled bedding for bacterial spores and pinworm eggs would be ideal to confirm that the bedding had no infectious particles remaining. We observed some results for the extended cycle bagging experiment that approached significance, and it is possible that with a greater sample size, the results may show significant differences for bagging style.

Overall, our data demonstrate that our current practices, which are common to the field, are effective, although there are nuances to the procedures being performed. Cages experience steam penetration regardless of where they are located in the autoclave and the number of stacked cages. We recommend that institutions closely evaluate practies related to sterilization of cage components and confirm that those practices are robust and complete in terms of confirming steam penetration and time spent above the desired temperature set point.