ILAR Journal Online, Volume 39(4) 1998: Opportunistic Infections in Laboratory Rats and Mice

Online Issues

<< All Back-issues

<< This Issue's Table of Contents

ILAR Journal V39(4) 1998
Opportunistic Infections in Laboratory Rats and Mice

Current Strategies for Controlling/Eliminating Opportunistic Microorganisms
William J. White, Lynn C. Anderson, James Geistfeld, and Dale Martin

William J. White, V.M.D., M.S., is Senior Director of Professional Services at Charles River Laboratories, Wilmington, Massachusetts; Lynn C. Anderson, D.V.M., is Senior Director of Comparative Medicine/Laboratory Animal Resources at Merck Research Laboratories, Rahway, New Jersey; James Geisfeld, D.V.M., M.B.A, is Technical Director at Taconic Farms, Germantown, New York; and Dale Martin, D.V.M., Ph.D., is Director of the Division of Veterinary Medicine at Walter Reed Army Institute of Research, Washington, DC.

The strategies used to control or eliminate opportunistic microorganisms from an animal colony depend on these organisms' risk to institutional research programs. The research benefits of controlling or eliminating the microorganisms must be balanced against the control measures' cost, complexity, and probability of success. A nonessential control strategy may be so complicated, expensive, and time consuming that it is circumvented (intentionally or unintentionally) by those expected to use it. In this paper, we outline an approach for analyzing the risk associated with each organism and for developing a control strategy that includes consideration of the research requirements, numbers of animals at risk, and available facility resources.

RISK ASSESSMENT

For this discussion, risk assessment is the process of analyzing the nature and relative importance of a microorganism to a research program. Certainly, all risks are relative. For example, a virus such as sialiodacryoadenitis virus (SDAV¹) produces generalized clinical signs in naïve animals and can affect various physiological and metabolic functions (Bhatt and Jacoby 1985). Such a virus is likely to be of concern for more research programs than a bacterium such as Corynebacterium bovis (otherwise known as hyperkeratosis-associated coryneform species), which may cause generalized disease and research effects only in glabrous mice (Clifford and others 1995).

In very general terms, microorganisms can be ranked according to their established potential to cause disease and research interactions. This ranking should be based on data in the peer-reviewed literature coupled with the organism's ability to cause disease or research interactions in the absence of the predisposing factors. A simple ranking scheme is provided in Figure 1. To evaluate the risk imposed by a particular microorganism more thoroughly, the following, more complete set of factors should be applied to each species and group within the total research population at a particular institution: prevalence; species specificity; research/disease effects; manipulation and access; replaceability/availability; type of research; movement/transportation; husbandry supplies, research equipment, and other materials; transmissibility of microorganisms; zoonotic potential; and concurrent use of hazardous agents.

Prevalence

Some organisms are much more common than others. The prevalence of Staphylococcus aureus is greater in a research environment populated by people whose normal skin flora often contains this opportunistic organism than Hantaan virus, which requires contact with an animal host. Thus, if S. aureus is of concern, then it poses a greater risk by its greater prevalence (Weisbroth 1996).

Species Specificity

Some organisms, such as Citrobacter rodentia (Barthold and others 1976), have a relatively limited host range. Although this organism may be transferred from mice to other hosts, it is unlikely to cause disease or have research effects in any species other than mice. Other organisms may be shared across a number of host species and may produce clinical effects in multiple species, posing a qualitatively greater risk.

Research/Disease Effects

The effects of many naturally occurring infections of laboratory animals on biomedical research are unknown and require systematic scrutiny. For this reason, efforts to exclude or eliminate the causative organisms are encouraged. It is important to recognize that animals can become infected by an organism without developing disease. When disease does develop, it may have a range of effects. The presence of disease, however, does not always have a negative impact on research. For example, a bite wound that becomes infected with an opportunistic bacterium may undergo a brief clinical course and resolve itself with no measurable effect on research.

Some organisms cause problems only under special conditions. For example, clinical disease associated with beta-hemolytic Streptococcus sp. infection is more commonly seen in immunocompromised than in immunocompetent animals (Geistfeld and others 1998; Schenkman and others 1994). Similarly, animals bearing cutaneous tumors are more likely to develop localized skin infections from opportunistic organisms than non-tumor bearing controls.

It is also important to consider that within any given species of organism, there may be numerous clonal variants; however, only some may cause disease or research effects. For example, several serotypes of Streptococcus pneumoniae (Koneman and others 1988) have been associated with rodent disease although the majority of S. pneumoniae serotypes are not known to cause disease.

Manipulation and Access

The extent of access to animals within a facility can also affect the level of risk. Risk can be minimized by housing the animals in a barrier facility, isolators, or microisolation cages that exclude microorganisms. However, as the number of individuals who have access to the animals increases, the opportunity for breaks in techniques and introduction of organisms also increases. Similarly, as the frequency and intensity of animal handling increases, the opportunity for exposure to microorganisms increases. For example, daily removal of animals from their primary enclosures for oral gavaging, testing in specialized apparatuses, or weighing may increase their exposure to microorganisms.

Replaceability/Availability

Some types of biomedical research require the use of unique animals. Animals that have undergone extensive surgical modifications, treatments, or genetic manipulations (such as transgenic animals) may be very difficult or impossible to replace. Intensive, expensive control measures may be justified to prevent these unique animals from exposure to microorganisms. However, these same measures may not be justifiable for animals that can be replaced readily.

Type of Research

Opportunistic microorganisms do not affect research uniformly. In fact, relatively few opportunistic organisms have been implicated in consistently causing clinical disease or interfering with research following infection. Those that have disease or research interactions are usually associated with unique conditions or have involved every specific types of research. More commonly, infection occurs without disease or research effects. An individual who conducts risk assessment may identify 1 or 2 documented research interactions and make the unwarranted assumption that the implicated microorganism causes more generalized research interactions. If no studies substantiate these far-reaching effects, then strategies to exclude the organism from an entire facility may not be justified.

Movement/Transportation

Animals are moved within an institution's facilities for a variety of reasons. For example, animals may be transported to an investigator's laboratory for specialized testing and then returned to the animal facility for continued housing. The stress associated with transport and testing results in reduced immune function that predisposes the animals to disease (Doerning and Cruze 1997; FBR 1987). Similarly, animals may be moved from an institution's breeding colony to various animal housing locations throughout an institution. Animals are also shipped to or received from other institutions. The process of transportation always entails some risk even when specialized housing devices or containers are used to transport the animals. Microbiological control is often lost during the unpacking process because the outside of the transport container is contaminated. Decontamination procedures cannot completely eliminate this risk.

Husbandry Supplies, Research Equipment, and Other Materials

Unsterilized feed, bedding, water, research equipment, multiple-use instruments, ear punching devices, and dosing needles can introduce or transfer microorganisms between animals. It is often difficult to ensure that the supplies or equipment are effectively sterilized or disinfected.

Sterilization equipment must be load calibrated and validated to ensure complete sterilization (Oxborrow and Berule 1991). The effectiveness of specific disinfectants against specific microorganisms of concern in animal facilities is often poorly understood. The effective concentrations, contact times, and interfering conditions for many agents are based largely, and perhaps incorrectly, on extrapolation from limited testing (Block 1991). The risk of exposure to unwanted microorganisms increases as the animals' exposure to potentially contaminated supplies and materials increases.

Transmissibility of Microorganisms

A variety of factors governs the transmission of microorganisms between animals. The routes by which microorganisms are shed, their concentrations in secretions and excretions, their survival in the environment, the dosages required for infection, and their susceptibility to sanitation and disinfection procedures vary significantly. For example, SDAV in rats appears to be transmitted very rapidly (Bhatt and Jacoby 1985), whereas viruses such as pneumonia virus of mice and many bacteria are transmitted much more slowly (NRC 1991).

Zoonotic Potential

Zoonotic microorganisms are capable of infecting both people and animals. A small subset of these agents may cause serious disease or mortality in people but do not produce severe disease in animal hosts and may not interfere with research results. Nevertheless, the human risk associated with such organisms in an animal facility should be of great concern and may justify elimination of infected animals from a research program (NRC 1997).

Concurrent Use of Hazardous Agents

Infectious, toxic, carcinogenic, and radioactive agents are employed in some research studies. These hazardous agents may affect the risk assessment and control strategy for animals exposed to opportunistic microorganisms. Strategies used to contain hazardous agents may contradict and potentially jeopardize strategies to exclude microorganisms from animals. For example, systems used to contain hazardous agents to which animals have been exposed may use low air flows to minimize the amount of potentially contaminated air that must be treated. In contrast, housing methods used to exclude opportunistic organisms often use high flows of sterilized air to dilute contaminants. Similarly, biocontainment strategies often specify infrequent cage cleaning and disinfection to minimize the amount of waste that needs to be treated. Conversely, bioexclusion methodologies often involve frequent cage cleaning and disinfection that produce large amounts of waste.
Summarizing Risk

When all of these factors have been considered, it is possible to determine the relative risk imposed by each organism to each species and group of animals. This is best expressed as a relative weight, such as "low," "medium," or "high" risk. Figure 2 is an example of risk assessment using all the parameters. We recommend that the veterinarian or other appropriate staff person complete such a risk analysis on any organism of concern to the animal, staff, or research project. Although numerical weighting systems can be considered for the various risk factors, these systems generally do not yield consistent results when manipulated by addition, multiplication, or other transformations to develop an overall score that can be used for comparison.

It is important to determine whether the risk is real (would occur absolutely) or is theoretical (may occur). The rationale used to rank the level of risk should be based on literature citations (Waggie and others 1994). Caution should be used in extrapolating limited data to make risk assessments. To do so may result in erroneous conclusions that all opportunistic organisms pose a risk and must be eliminated from an animal facility. The control strategies required for such a risk assessment may not be realistic or achievable for most institutions.

When the level of risk has been determined, the size of the population at risk and the nature and number of studies potentially affected must be considered. It may be appropriate to apply 1 control strategy for a small number of studies or a few animals and a different strategy for multiple projects involving many animals.

CONTROL STRATEGIES FOR MICROORGANISMS

Control strategies are based on exclusion of specific microorganisms. The control strategy should include the overall program of acquiring, housing, transporting, and utilizing animals in a biomedical research environment. The strategy should also identify the microorganisms to be excluded and the testing program and methodologies used to determine whether specific organisms are present within the animal population. The testing program must utilize adequate sample numbers and sampling frequencies to provide the level of confidence required to determine the presence of microorganisms identified by the risk assessment process. Sampling small numbers of animals infrequently may be inadequate to detect the presence of unwanted microorganisms and hence pose an unacceptable risk to research programs. The testing program design may be limited by economic constraints. When limited resources are available, it is often advisable to concentrate on controlling and monitoring for several prevalent organisms that pose the greatest risk. In these instances, trying to control and monitor for a large variety of organisms, which would require more comprehensive but less frequent testing of fewer animals, may be counterproductive.

Three basic control strategies may be applied to any microorganism. One strategy is to accept its presence within the facility and make no changes in operating procedures to address its presence. This may be an acceptable strategy if, under the conditions specific to the facility, the organism has no known impact on the research. A second strategy is to exclude the organism completely. Depending on the factors previously outlined, this may be impossible to achieve or may be prohibitively expensive. Such a course of action should be undertaken only after assessing the likelihood of success, based on the facility and its resources. A third control strategy is to exclude the organism only from designated animal populations or research programs within the facility while tolerating the organism in the remainder of the animal facilities.

It is also important to clearly define the course of action if a microorganism of concern gains entrance to the facility or project area.. Although such breaks are inevitable, their frequency can be minimized and rationally addressed if reasonable bioexclusion practices are in place.

AVAILABLE TOOLS FOR BIOEXCLUSION

A bioexclusion system refers to the facilities, equipment, and procedures used to exclude microorganisms from groups of animals within an institution. It is not a complete control strategy that encompasses the entire program of acquiring, housing, transporting, and utilizing animals. Instead, it is designed only to prevent the migration of organisms across a barrier and into an enclosure. The system can operate at the primary enclosure (cage) level or at the secondary enclosure (room) level. The physical characteristics and biology of each microorganism and its concentration in the environment limit the effectiveness of any bioexclusion method. Most bioexclusion methods focus on controlling airborne transmission by using filters. Although the characteristics of the filter material and the method and frequency of removing microorganisms from it can affect its effectiveness, other forms of transmission (including fomite transmission) can be equally important.

No single bioexclusion system is perfect; each has some inherent risks associated with maintaining the animals and conducting research. Many of these risk factors to bioexclusion systems are illustrated in Figure 3. Although not all systems have all risks, the risks include the following: (1) movement of people, materials, and animals across the barrier; (2) incomplete disinfection/sterilization of materials; (3) inappropriate decontamination/gowning of personnel; (4) inadequate training and lack of understanding of the rationale behind the system, which can lead to inappropriate actions when unusual circumstances occur; (5) disregard of procedures; (6) failure of equipment; (7) failure to provide for emergencies including power failure, escaped animals, and breaks in the integrity of the physical barrier or in personnel clothing/gloves; (8) inadequate maintenance; and (9) failure to evaluate the performance of the system regularly. The following examples of bioexclusion systems are described below: barrier facilities or barrier rooms, isolators, microisolation cages, cubicles, venilated cabinets, and mass air displacement and laminar flow rooms.

Barrier Facilities or Barrier Rooms

The first and most common bioexclusion systems are barrier rooms or barrier facilities that contain many rooms. A barrier room is designed to prevent undesirable microorganisms from entering on any materials, equipment, animals, or personnel (Figure 4). To be effective, the walls, ceilings, and floor must be constructed of impervious materials, and all junctions and seams must be sealed. Air to the room should be high efficiency particulate air (HEPA¹) resistance filtered to exclude airborne particulates, including microorganisms. Air exhausted from the room does not have to be filtered since animals contained within the room are presumed to be free of unwanted microorganisms.

Barrier rooms must be designed to allow introduction of disinfected/decontaminated supplies and equipment. This is often accomplished by using a pass-through autoclave. The autoclave should be calibrated and validated before its initial use and then annually for each specific load configuration and cycle to help ensure that sterilization conditions are met. The placement of heat-sensitive, biological, or cumulative temperature indicators on a few locations within a load does not always ensure sterilization or disinfection.

Since some materials used in barrier facilities cannot be autoclaved, chemical agents must be provided to decontaminate their external surfaces. This is most often achieved by using the chemical disinfectant inside a spray port. The chemical agent must be active against the organisms that are to be excluded. The process of chemical disinfection must also be calibrated so that adequate concentrations, application procedures, and contact times are used. These procedures can be verified using test organisms.

Personnel working within a barrier room cannot be completely decontaminated, especially when common environmental opportunistic organisms are to be excluded. However, staff should cover all body surfaces with protective clothing to minimize the risk of humans introducing microorganisms through direct animal contact or aerosolization. Uniforms and associated attire must be sufficiently flexible and comfortable to allow performance of duties. Personnel also should sufficiently cover the neck and head as well as the junction of the uniform to the gloves and footwear to prevent exposure of skin surfaces to the animal's environment. These areas often become exposed during the normal workday, increasing the risk of contamination from personnel.

Adequate facilities must be provided for changing clothes or donning protective clothing. These facilities often include 1 of 2 types of entry lock systems: the wet-entry system, which requires personnel to take a water shower within the system, or the dry-entry system. These systems are divided into several components to minimize the potential transfer of organisms into the barrier.

Wet-entry system. The first component of the wet-entry system is an air lock, which inhibits the migration of insects and other vermin and controls air flow. Traps and other pest control devices are generally located in this area, where personnel commonly remove their shoes. In the second component, personnel remove their street clothing or dedicated facility uniforms, place them in an assigned locker, and then proceed to the third component of the system, where they shower. Upon exiting the shower, personnel enter the fourth area of the entry lock system where they dry off and put on a suitably disinfected or sterilized uniform. Often, a final hand washing is conducted in this area to ensure adequate disinfection of the hands before gloving. Upon leaving this area, personnel enter the barrier room or facility directly.

From an operational standpoint, the wet-entry system has several disadvantages. At best, the shower reduces but does not eliminate the microorganisms on skin surfaces. The high degree of moisture in the area may serve as a nidus of infection for opportunistic organisms if the area is not regularly and completely disinfected. The process is also time consuming and requires additional laundry and supplies for the barrier room operation.

Dry-entry system. In the alternative, dry-entry lock system, the first 2 components of the lock are the same as the wet-entry system. After removing their site uniform or street clothing in the second lock, the employee enters a clean gown area (the third component of the system) without taking a shower. The employee then puts on sterile clothing, including a 1-piece jumpsuit, cap, mask, dedicated shoes, socks, underwear, and gloves. Before gloving, the employee scrubs his/her hands using an antibacterial soap for at least 3 min and dries them thoroughly. After gloving, the employee enters the fourth component of the lock system, the air shower.

The air shower is a doubled-door chamber with multiple jets capable of delivering high-velocity HEPA-filtered air that is designed to wash particles from the surface of the protective clothing. It is assumed that the particles, which may consist of microorganisms or inanimate particulates carrying microorganisms, are washed to the floor of the chamber and exhausted out. The 30- to 45-sec air shower is interlocked in such a way that the air shower must be completed before the employee can proceed into the barrier. When the employee exits the barrier, a second air shower is taken, which serves to wash off hair, dander, and other particulates acquired while working within the barrier. This prevents the spread of these fomites throughout the lock system and minimizes the potential of exporting contamination.

Some barrier facilities provide break areas, restrooms, laboratories, and other areas oriented toward personnel rather than animal care. From a logistical standpoint, although these areas may be convenient and reduce operational costs, they pose a significant threat to transmission of microorganisms from personnel to animals. This is particularly important with respect to opportunistic organisms that are common environmental or human commensals.

Introduction of large pieces of equipment into an operating barrier room often poses a problem because few barrier rooms are equipped to handle them. Barrier operations often must be halted if large pieces of equipment are required. All necessary equipment items presumably will be in place when the barrier room is first set up and all surfaces are sterilized.

Introduction of animals into the barrier room is also problematic. The health status of animals to be introduced must be assured, which usually requires quarantining them in some form of containment system before they are transferred into the barrier. Quarantining animals within the barrier invites disaster even when secondary containment systems are used. Handling the potentially contaminated waste and other supplies poses risk to the microbiological status of the entire barrier.

Cage washing also poses significant problems in barrier room operation. Two alternatives must be considered. First, cages can be transferred from the barrier, washed in a central cage, sterilized, and then transferred back into the barrier. Besides being labor intensive, this procedure poses a significant cumulative risk with each passage of materials across the barrier. Moreover, cages designated to return to the barrier travel through hallways are processed in cage wash areas with other materials that are likely to be contaminated with opportunistic organisms. The cage washing procedure in itself does not usually produce totally disinfected materials. For this reason, great reliance is placed on postwashing sterilization.

The second possibility is to wash cages within the barrier facility either mechanically or by hand. This alternative eliminates the risk of passing materials across the barrier but increases the labor required within the barrier and requires significant, dedicated space. If automated equipment is used for cage washing, issues such as maintenance of the equipment and provision of detergents/disinfectants for use with this equipment become important logistical issues.

Overall, barrier rooms or barrier animal facilities are expensive and logistically difficult to operate. Because such facilities are so heavily technique dependent, they are more readily prone to microbiological contamination than some other bioexclusion systems. Unless additional bioexclusion systems are used, the entire population is at risk from a single breach of the barrier. Once a microorganism gains entrance into the barrier, there is little to stop it from spreading.

Personnel are generally the weakest link in a barrier facility. Greater numbers of personnel crossing into the barrier each day results in greater potential for inadvertent introduction of unwanted microorganisms. To minimize the risk, the number of personnel entering the barrier should be kept to a minimum, and each needs to be well trained in the appropriate procedures to safeguard barrier integrity. It is important to constantly renew such training to ensure that each individual understands the rationale behind the various measures to enter and sustain the barrier so that shortcuts are not taken.

Isolators

Although barrier facilities may be suitable for large groups of animals or very large animals, smaller groups of animals can be managed effectively using a second bioexclusion system---isolators (Coates and Gustafsson 1984). An isolator is an enclosure constructed of either flexible or rigid material (often plastic) that is used to surround a group of cages containing animals. Isolators have attached sleeves with gloves for manipulating animals inside the isolator and an entry port for passing decontaminated or sterilized supplies and equipment into the isolator and for removing waste materials. The surfaces of the material introduced into the isolator are also chemically disinfected or sterilized inside the entry port. Entering and exiting air is often filtered with a HEPA filter.

In many ways, isolators are analogous to small barrier rooms. They rely heavily on chemical and physical means of disinfection to ensure that materials entering the isolator are free of unwanted microorganisms. Food, bedding, water, and other supplies must be packaged within large metal containers fitted with filters to allow steam penetration during autoclaving. After the containers are autoclaved, they are held until biological indicators confirm that the load was sterilized effectively. If the quality control is satisfactory, the container is attached to the port with a flexible sleeve and the port, sleeve, and attached surface of the container are chemically disinfected or sterilized. More recently, vacuum-packed irradiated feed and bedding and filter-sterilized bagged water have become available, eliminating the need for transfer cylinders. Unlike the metal cylinders, these new packaging devices allow representative samples of the materials to be cultured before they are introduced into the isolators. Vacuum packing also provides visual evidence of the integrity of the packaging.

Isolators are normally kept under positive pressure. In some instances, such as when work with infectious disease is being conducted, isolators may be run under negative pressure. In this latter instance, the negative pressure prevents the infectious organisms from escaping if a hole occurs in the membrane of the isolator or the gloves.

The integrity of the isolator should be assessed regularly using both visual and leak detection methodologies. For example, helium gas may be introduced to the isolator, and the external surfaces of the isolator may be scanned with a helium gas detector. Other tracer gases and ultrasonic systems are also available for this purpose. Breaks in the integrity of isolators most often occur in the gloves and the floor of the isolator. Regular replacement of gloves during operations will help minimize breaks in the system.

Isolator sizes can vary significantly. The most common sizes will hold 2 to 70 mouse cages and can be accommodated in most animal facilities. Larger isolators are also available and often incorporate built-in half suits or complete suits with air supply for personnel to allow access to much larger numbers of cages. However, these larger isolators increase the risk of breaks and loss of microbiological integrity.

As a bioexclusion system, isolators provide a number of advantages. Compared with barrier rooms, the risk from personnel contact with animals is greatly reduced. Moreover, fewer animals are placed at risk in an isolator due to its size limitations. Unlike bioexclusion systems that focus on cage-level barriers, both isolators and barrier rooms allow sentinel animals to be housed within the same microbiological environment as research animals. This arrangement makes possible a much more accurate health assessment program. However, isolators do pose certain disadvantages. The space required for isolators decreases the total amount of animal housing space available, and manipulation of animals using the glove system is somewhat more cumbersome.

Microisolation Cages

A third bioexclusion system uses disinfected or sterilized filtered cages known as microisolation cages. Cage components include a molded plastic cage bottom and stainless steel wire bar lid covered by a fitted, plastic lid that contains a filter. The filter is commonly made from spun-bound materials and has a spectrum of pore sizes. Different filtration media can be used to achieve different degrees of particle exclusion and efficiency. Animal health and well-being must be considered when selecting filter material. Appropriate concentrations of gasses, heat, and water vapor must be sustained within the cage. Filter efficiency increases and air flow decreases as continued use soils the filter.

A microisolation cage must be adequately disinfected or sterilized before use and must be protected from contamination after disinfection or sterilization. The thoroughness of disinfection or sterilization of caging and supplies must match the intended level of bioexclusion. Complete sterility is required for total exclusion of opportunistic microorganisms.

If caging is autoclaved for decontamination, bedding, food, and in some cases water contained within water bottles may be processed with the cage as a single unit. Although this technique is acceptable, calibration and validation of the sterilization process must accommodate the very different nature of the various components being processed together. Conditions appropriate for sterilization or disinfection of caging may be quite different from those required for processing of water or food. Alternatively, cage components and supplies may be individually wrapped and sterilized and then assembled in a horizontal laminar flow workstation using aseptic technique. With this methodology, personnel must be careful to disinfect the wrapping materials before handling the sterilized components.

All changing of microisolation cages or manipulation of animals housed within microisolation cages must be done using aseptic technique within an environment that is adequately disinfected or sterilized. This is accomplished most commonly in a horizontal laminar flow workstation, where HEPA-filtered air is blown across the caging and work surface. However, when the laminar flow air strikes the external surfaces of the cage, it can move microorganisms from the external surface into the open cage. Therefore, before the cages are manipulated within the hood, their external surfaces should be decontaminated. If the animals have been inoculated with hazardous agents, personnel manipulating the cages and animals must exercise safety precautions to prevent their exposure to those agents. It is also important to minimize the amount of materials placed within the laminar flow hood since air striking objects tends to eddy for distances up to 3 times the diameter of the object encountered. Hence, uncontrolled air currents can be set up within the hood and contaminate the animals and equipment. Personnel manipulating the animals should wear clean protective clothing to minimize introduction of microorganisms, and their gloves should be disinfected regularly. For routine cage changing, disinfected forceps should be used to transfer animals between cages.

Microisolation cages may be ventilated in 2 different ways. The first, a static microisolation cage, has no mechanical means of ventilation (Keller and others 1989). Heat, humidity, and the concentration of particulates and waste gases are increased within the static microisolation cage relative to the room in which they are maintained (Corning and Lipman 1991). For this reason, frequent cage changing is often required to keep the cage environment within acceptable limits.

The second type of ventilation system forces air into each microisolation cage (Keller and others 1983). These ventilated units interface with a racking system that supplies air into each cage through a sterilizable coupling or blows air across the cage filter and into the cage. Some systems also provide individual exhaust from the microisolation cages. The air supply and the exhaust (when present) are HEPA filtered and can be powered either by individual blowers mounted on the rack or through a central air supply system. With these systems, the ventilation to the room in which these units will be housed must also be considered. This is especially important when individual blowers are utilized because they produce additional room heat load. Moreover, consideration should be given to sound levels and to the provision of emergency power for these units. In the absence of forced ventilation, heat builds up quickly with these units.

Ventilated microisolation caging provides a number of advantages compared with static microisolation cages (Choi and others 1994). The consistent supply of fresh filtered air provides more appropriate environmental conditions. This increased ventilation also helps to maintain a drier environment that inhibits bacterial growth and may minimize particulate and gaseous concentrations within the environment.

In both systems, it is important for the junction between the cage and plastic lid to be well sealed to prevent air from leaking and potentially contaminating the external surfaces of the cages. This may not be an issue if the microisolation cage technique is used perfectly, including external disinfection of cages.

Cubicles

Use of cubicles constitutes a fourth method of bioexclusion (Hessler and Moreland 1984). A cubicle is a small room within a room used to prevent airborne cross-contamination between adjacent groups of animals. Air may be supplied to the cubicle from a separate air supply within the cubicle or from the room containing the cubicle, in which case an exhaust system located inside the cubicle creates negative pressure and pulls air into the cubicle through a space left between the floor and the bottom of the cubicle door. Depending on the height of the space, its area, and the room ventilation, large volumes of air pass under the door and face velocities will often exceed 100 ft per min. Once the cubicle door is opened, however, control of airborne cross-contamination ceases. The system does not provide for disinfection of equipment, materials, or supplies entering the cubicle, and it may allow particulates in the room to enter the cubicle through the air stream.

Cubicles by themselves do not represent a very comprehensive bioexclusion system. They can be used to slightly decrease the risk of airborne transmission of agents under certain conditions and may be useful in certain quarantine applications. Their use is best combined with other bioexclusion systems.

Ventilated Cabinets

Use of ventilated cabinets constitutes a fifth, albeit somewhat limited method of bioexclusion (Corning and Lipman 1992). These enclosures have 1 or more doors and a series of shelves on which animals in open cages may be placed. Each cabinet is supplied with HEPA-filtered air by a small blower attached to the cabinet. Generally, the air is not supplied in a laminar flow fashion and can either be exhausted directly to the outside of the cabinet or into a dedicated exhaust system with or without filtration. Like cubicles, the ventilated cabinets address only airborne cross-contamination. They do not provide a method for the following: husbandry or manipulation of animals within a controlled environment, controlled movement of organisms between cages in the same cabinet, or prevention of microorganisms' fomite transmission. When the doors to the cabinet are opened, the system does not provide bioexclusion. Like cubicles, ventilated cabinets may only have use in conjunction with other bioexclusion systems.

Mass Air Displacement and Laminar Flow Rooms

Mass air displacement (Hessler and Moreland 1984) and laminar flow rooms (Beall and others 1971) have been used as bioexclusion systems for many years. Like cubicles and ventilated cabinets, these systems are designed to control airborne movement of microorganisms between groups of animals. In mass air displacement rooms, large volumes of air are blown into the room, usually from ceiling-mounted supply diffusers. In theory, the air washes out particulates by sweeping them to floor-mounted exhaust vents or, in the case of "portable rooms," through a 1- to 3-in gap between the bottom of a plastic curtain and the floor. When large quantities of air are blown into the room, particulates and microorganisms are swept away from cages containing animals and are diluted out in the air stream.

In contrast to mass air displacement rooms that use several supply diffusers, laminar flow rooms use HEPA-filtered air that is delivered to a space above the ceiling. Thousands of tiny holes in the ceiling are designed to generate a laminar air stream as the pressurized HEPA-filtered air passes through them and into the room. The air is exhausted at floor level. Unfortunately, large objects such as racks, cages, and personnel can disturb the air stream and cause eddying. Depending on the object, its orientation, and the length of time an object is in the air stream, varying levels of airborne cross-contamination may occur (Thigpen and Ross 1983). The principal benefit of this system is dilution of contaminants with HEPA-filtered air.

Neither mass air displacement nor laminar flow rooms prevent fomite transmission of organisms. Because these systems do not provide for disinfection or sterilization of equipment and supplies, movement of such items in and out of the area and the manipulation of animals within the area may introduce organisms. These rooms are useful adjuncts to other bioexclusion systems but are rarely successful by themselves.

Effectiveness of Bioexclusion Systems

No single bioexclusion system is perfect for all applications, and certain features of all systems require regular checking and calibration. It is important to test the integrity of systems regularly---especially those that rely heavily on personnel training. This can be done using marker compounds such as fluorescein-labeled microspheres, fluorescent dyes, or tracer gasses such as helium, sulfurhexaflouride, or nitrous oxide. Microbiological assessment methods such as rodac plates, slit to agar sampling, liquid impingement, and other similar techniques can also be used. The importance of training and retraining personnel cannot be overemphasized, especially for technique-intensive systems such as barrier rooms and microisolation cages.

BIOEXCLUSION CONTROL STRATEGIES

To develop an effective bioexclusion control strategy, it may be necessary to use multiple bioexclusion systems either separately or in combination. In addition, the strategy should include ensuring the health status of animals obtained by the facility, requirements for the health status of existing animal colonies, routine animal health surveillance, implementation of appropriate personnel procedures, and disaster planning.

Assessment of Animal Health Status

Criteria for selecting animals to be purchased or transferred into the facility must be based on the bioexclusion list formed during the risk assessment. Since the list may vary depending on the specific research program or other factors, it is important to ensure that errors do not occur when ordering or approving the transfer of animals for specific projects.

Current and detailed health monitoring information should be requested from commercial suppliers or institutions providing animals. This information should be reviewed critically to assess the health status of all incoming animals. Sample size, testing frequency, and methodology should also be considered to help assess the reliability of the heath monitoring information. If the health monitoring information is not current or complete, it may be necessary to test animals on arrival. Additional animals from the same colony should be ordered for this purpose.

The method of transportation and type of shipping container can significantly influence the health status of animals. Animals provided by commercial suppliers are commonly shipped in filtered containers on trucks dedicated to that supplier or to specific colonies within the supplier's facility. Animals shipped by air freight may be exposed to animals and freight of unknown microbiological status. Such exposure is unlikely in a dedicated truck, although some potential always exists for microbiological contamination of the external surfaces of a shipping container. Contamination is much more likely to occur with air shipments. Regardless of the mode of transportation, it is good policy to disinfect the external surfaces of all shipping containers (Orcutt 1987) with a chlorine-based agent and, when possible, to unpack them before bringing the animals into the facility.

Because there is always some risk involved with acquiring animals from any source, it is prudent to hold newly arrived animals in an area separate from research colonies. They should be housed in systems that prevent the introduction of unwanted microorganisms and provide biocontainment in case the animals harbor undesirable organisms. The decision to quarantine any or all animals is often tempered by practical issues, including the availability of sufficient quarantine space, length of quarantine, and investigators' need for access to the animals.

If animals are quarantined, health monitoring should be conducted during this period in addition to regular clinical observations to assess the animals' health (Small 1984). Serological, bacteriological, and parasitological examinations are often conducted on representative animals from a shipment of rodents. Sentinel animals may be used, but this often lengthens the process since adequate exposure time is essential.

Many rodents provided by commercial suppliers are maintained under rigorous bioexclusion conditions, and the colonies are extensively monitored for the presence of certain microorganisms. Depending on the institution's risk assessment, animals acquired from these sources may not be quarantined or may require only short quarantine periods.

Quarantine may also be imposed on a group of animals within a facility when there is some evidence that an unwanted microorganism has been introduced in 1 or more animals within the group. The animals and the materials, personnel, and equipment that have contact with the animals must be separated from other animals housed within the same facility. Bioexclusion practices, including disinfection methods, and the protection and movement of personnel within the facility may be changed to ensure that the agent is contained. In some facilities, it may be more appropriate not to quarantine the affected group of animals but instead, to cull them and carry out extensive decontamination procedures.

Personnel Procedures

The role of personnel in acquiring and transmitting organisms of concern must be placed in perspective. Humans and animals share a number of opportunistic organisms; however, most of the organisms of concern to research facilities are animal- and often species-specific. Although personnel may have contact with such animal-specific organisms during or after work, their ability to sustain and transfer such organisms is limited.

Personnel who work in barrier rooms or barrier facilities should be screened for potential animal contact outside the work environment. This may be accomplished by regular surveys or interviews. Pet animals, secondary occupations including farming, jobs in other research facilities, pet stores, animal breeders, or other tasks that involve the handling of live or dead animals may increase the risk that an employee will introduce unwanted microorganisms into the barrier facility. This information may be used to develop specialized employee procedures or practices that will allow them to continue to have pets or participate in certain activities while minimizing the microbiological risk to an institution's animal colony. In some instances, however, the risk may be deemed too great to allow continued access to the barrier.

Clothing

Clothing worn by personnel working within an animal facility should be based on the functions they perform and the function of the clothing. The clothing should not exceed what is necessary to accomplish the bioexclusion task. If different types of clothing are worn by different types of personnel in the same area, the rationale for this difference must be well thought out and justified. Convenience is seldom a good reason.

Laboratory coats and scrub suits. Investigators and technicians often wear laboratory coats over street clothing in animal facilities. Although possibly protecting the employees' clothes from soiling with animal wastes and dander, laboratory coats leave too many areas exposed and offer little protection against transmission of opportunistic organisms. The same is true for often-worn short-sleeved surgical scrub suits, which also provide limited coverage of personnel although they at least help to ensure that street clothing is not worn into animal areas. Laboratory coats and scrub suits must be laundered regularly and should not be worn outside the animal facilities or stored on hangers within the animal facility or animal rooms for extended periods of time. Jumpsuits, which provide more complete coverage, should have elastic or other closures at the wrist and ankles to ensure that the interface with gloves and footgear provides a total barrier to transmission of organisms from the skin of personnel to animals.

Shoes. Shoes have traditionally been considered vehicles for moving microorganisms between locations. Since most microorganisms are carried on fomites, which often settle to the floor, it is reasonable to expect that shoes will become contaminated. Shoe covers may provide a temporary means of minimizing transmission but are often not made of durable materials and do not accommodate large feet. The use of shoe covers in an animal facility should include the establishment of a convenient system for placing them on the feet and removing them. Their removal and replacement with new shoe covers between different areas of risk should be well thought out and justified. Dedicated shoes that are frequently cleaned can provide a reasonable alternative for personnel who spend significant amounts of time within the facility. Dedicated, frequently disinfected shoes are essential in barrier rooms/facilities.

Gloves. Gloves are often used in animal facilities to minimize transmission of microorganisms and to protect employees from exposure to animal secretions, excretions, and dander. The gloves must be regularly disinfected or sterilized between operations to help control opportunistic organisms. When sterilized gloves are required, employees should be trained to put on the gloves correctly and to be observant for breeches in aseptic technique that could contaminate the gloves.

Face masks. Face masks (surgical type) are worn in many animal facilities to protect employees against exposure to aerosols and allergens and to protect the animals from respiratory and oral microorganisms being shed by the employees. However, due to the design of these masks, protection is provided mainly for the animals (Mullan 1996). These masks are not adequately designed for protection against aerosol exposures but will help reduce the quantity of allergen exposure and provide protection from particulate antigens when worn properly. When aerosol protection is required (such as for infectious diseases and toxic compounds), a respirator certified by the National Institute of Occupational Safety and Health and designed to protect the employee should be used. To provide complete protection from microorganisms and allergens for both the animals and employees, a whole body helmeted uniform that has a self-contained ventilation system is required. This uniform covers the entire body and filters exhaled air. Such a uniform is expensive, which limits its use in most animal facilities.

Paper or molded fiber surgical masks are usually used to help reduce but not to eliminate the release of opportunistic microorganisms into the animal facility environment. If the face mask becomes wet, opportunistic bacteria can multiply on the inside of the mask and grow through it rapidly. Then the organisms will be expelled into the environment by normal breathing, creating the potential for contamination. For this reason, personnel should change face masks frequently during the day. This may not be important in most areas of a facility but will have an impact where common human commensals can endanger the research being conducted. Caps, as well as other head gear to cover beards and other facial hair, are also necessary to provide protection from opportunistic microorganisms.

Training

Training of personnel is the key to maintaining bioexclusion successfully. Decisions regarding strategies and types of equipment must be made based on the full understanding and cooperation of all who will use them. The program of use should be outlined completely in standard operating procedures, and each individual using the animal facility must be trained in the proper techniques. Such training should be documented, and retraining should be provided especially in critical techniques or for individuals who use the animal facility infrequently. Some bioexclusion systems such as barriers, isolators, and microisolation cages are highly technique dependent and can easily fail if appropriate practices are not followed.

Changing the Health Status of Animals

Invariably, animals infected with 1 or more unwanted microorganisms will be encountered. If alternatives to the use of these animals are not available, it may be necessary to alter their health status.

The process of rederivation is the most common means of changing the health status of animals (Trexler 1983). This process can be accomplished by removing young by caesarean section and cross-fostering them onto mothers of the appropriate health status or by transferring embryos to mothers of the appropriate health status (Hogan and others 1986). Both of these techniques require technical proficiency and extensive health monitoring after rederivation to assure that the organism of concern has been eliminated. Repeated sampling and adequate sample size of the rederived population are critical to assure this. These processes are expensive and time consuming, especially for outbred animals, which require large numbers of rederivations to ensure genetic heterogeneity.

Another strategy used to change health status is to discontinue breeding and introduction of naïve animals that serve as hosts. This technique works well with nonpersistent organisms such as mouse hepatitis virus. A period of 6 to 8 wk is usually sufficient to eradicate the organism from a colony. If the environment is adequately disinfected, breeding can be resumed or naïve animals can be introduced after this period. Newly introduced naïve sentinel animals must be assayed after exposure to the colony for a period of 6 to 8 wk to confirm successful elimination of the agent. It should be noted that offspring from breeders in which the cycle has been successfully broken may still possess detectable maternal antibodies. This procedure is not effective against many agents, particularly those passed through the placenta.

Medication may also be used to eliminate certain bacterial infections (Goelz and others 1996); however, the treatment may pose a greater risk to certain studies than the presence of the microorganisms. The decision to medicate must be made jointly with the investigators using the animals. To be effective, the medication must be coupled with rigorous environmental decontamination and must be continued long enough to ensure that all animals have received a therapeutic dosage. Repeated testing of the animals over a prolonged time may be required after treatment to ensure that the animals are free of the organism of concern.

Culling infected populations is another, albeit drastic strategy, which may be the only way to limit the spread of certain infectious agents within an animal facility. Although testing and selective culling of animals may be a more modest approach to culling an entire group of animals, it is seldom effective. Most organisms spread throughout a population of animals faster than detection methods are able to identify infected animals. If bioexclusion housing methods are in place, it may be possible to utilize this strategy.

Disaster Plan

Eventually, an unwanted microorganism will inevitably gain entrance to an animal colony despite bioexclusion practices. To help limit the research damage that may occur, it is important to have a planned course of action with which all potentially affected parties agree. This requires consideration of the bioexclusion systems and techniques previously discussed. The plan should include the types and extent of communication that will take place once a microorganism is found and should be updated periodically as research directions change.

To avoid unnecessary alarm, it is important to determine what constitutes evidence of infection. Since false-positive results can occur (Smith 1986), the types and numbers of additional samples needed to confirm positive findings and the extent of infection within the facility should be established. Diagnostic testing is expensive and should be applied prudently.

In many cases, a committee composed of animal facility personnel and investigators may assist in managing contamination and ensure that all affected parties are notified appropriately. The facts needed to make reasonable decisions must be conveyed to investigators in a timely fashion, but unconfirmed results may set off unwarranted actions. A substantive review of all microorganisms of concern, including appropriate literature citations, should be made available to help investigators make informed decisions regarding their research.

After a contamination, it is important to investigate its possible cause and to make necessary changes in the program and procedures to try to prevent recurrence. It is not always possible to determine the cause of contamination, and it is important to resist the temptation to jump to conclusions that cannot be supported by fact. For example, it may be assumed that the microorganism gained entrance into the facility in contaminated biological materials such as tumor cells. However, unless mouse antibody production testing of the tumor line verifies the presence of the organism, this assumption may be unwarranted.

SUMMARY

Exclusion of microorganisms from any research facility is a dynamic process prone to occasional failures. Many available techniques can be assembled into a control strategy for any animal facility, but their selection and application must be carefully considered. Key to any control strategy is an assessment of risk with respect to specific microorganisms. The decision to exclude microorganisms from 1 or more groups of animals within a research facility must be carefully considered and based on facts, not assumptions. As the list of selected organisms becomes more extensive and the organisms become more ubiquitous in the environment, the likelihood of success with bioexclusion strategies decreases. Planning must be done well in advance of any potential contamination to limit the damage caused by it. In the end, the selection of appropriate bioexclusion systems and control strategies must be balanced between the level of risk assessment and the resources available to accomplish the objective.

¹Abbreviations used in this paper: HEPA, high-efficiency particulate air; SDAV, sialiodacryoadenitis virus.

REFERENCES

Barthold SW, Osbaldiston GW, Jonas AM. 1977. Dietary, bacterial and host genetic interactions in the pathogenesis of transmissible murine colonic hyperplasia. Lab Anim Sci 27:938-945.

Beall J, Torning F, Runkle R. 1971. A laminar flow system for animal maintenance. Lab Anim Sci 21:206-212.

Bhatt PN, Jacoby R.O. 1985. Epizootiological observations of natural and experimental infection with sialoacryoadentitis virus in rats. Lab Anim Sci 35:129-134.

Block SS. 1991. Disinfection, sterilization and preservation. 4th ed. Malvern PA: Lea and Febiger.

Choi GC, McQuinn JS, Jennings BL, Hassett DJ, Michaels SE. 1994. Effect of population size on humidity and ammonia levels in individually ventilated microisolation rodent caging. Contemp Topics 33:77-81.

Clifford C, Walton BJ, Reed TH, Coyle MB, White WJ, Amyx HL. 1995. Hyperkeratosis in athymic nude mice caused by a coryneform bacterium: Microbiology, transmission, clinical signs and pathology. Lab Anim Sci 45:131-139.

Coates ME, Gustafsson BE, editors. 1984. The germ-free animal in biomedical research. London: Laboratory Animals LTD. p 11-32.

Corning B, Lipman NS. 1992. The effects of a mass air displacement unit on the microenvironmental parameters within isolator cages. Lab Anim Sci 42:91-93.

Corning BF, Lipman NS. 1991. A comparison of rodent caging systems based on mircoenviromental parameters. Lab Anim Sci 41:498-503.

Doerning B, Cruze C. 1997. The effects of routine animal husbandry and experimental procedures upon physiological parameters of rats. . Contemp Topics 36:42-43.

FBR [Foundation for Biomedical Research]. 1987. The Biomedical Investigator's Handbook. Washington DC. p 11-18.

Geistfeld J, Weisbroth S, Jansen E, Kumpfmiller D. 1998. Epizootic of group B Streptocooccus agalactiae serotype V in DBA/2 mice. Lab Anim Sci 48:29-33.

Goelz MF, Thigpen JE, Mahler J, Rogers WP, Locklear J, Weigler BJ, Forsythe DB. 1996. Efficacy of various therapeutic regimens in eliminating Pasteurella pneumotropica from the mouse. Lab Anim Sci 46:280-285.

Hessler J, Moreland A. 1984. Design and management of animal facilities. In: Fox J, Cohen B, Loew F, editors. Laboratory Animal Medicine. New York: Academic Press. p 505-526.

Hogan B, Constantini F, Lacy E. 1986. Manipulating the mouse embryo. Cold Spring Harbor NY: Cold Spring Harbor Laboratory.

Keller LSF, White WJ, Snider MT, Lang CM. 1989. An evaluation of intracage ventilation in three animal caging systems. Lab Anim Sci 39:237-242.

Keller GL, Mattingly SF, Knapke FB. 1983. A forced-air individually ventilated caging system for rodents. Lab Anim Sci 33:580-582.

Koneman EW, Allen SD, Dowell VR, Janda W, Schreckenberger P , Winn W Jr. 1988. Color Atlas and Textbook of Diagnostic Microbiology. 3rd ed. Philadelphia: JB Lippincott Co. p 333-353.

Mullan R. 1996. 4th Symposium on Biosafety. Atlanta GA: Centers for Disease Control and Prevention.

NRC [National Research Council]. 1991. Infectious Diseases of Rats and Mice. Washington DC: National Academy Press. p 55, 97.

NRC [National Research Council]. 1997. Zoonoses. In: Occupational Health and Safety in the Care and Use of Research Animals. Washington DC: National Academy Press. p 65-105.

Orcutt RP. 1987. How to conduct rodent research free of MHV, Sendai virus and other pathogens. Lab Anim 16:31-40.

Oxborrow G, Berule R. 1991. Sterility testing---Validation of sterilization processes and sporicide testing. In: Black S, editor. Disinfection, Sterilization, and Preservation. 4th ed. Philadelphia: Lea and Febiger. p 1047-1057.

Schenkman D, Rahijia R, Klingenberger K, Elliott J, Richter C. 1994. Outbreak of group B streptococcal meningoencephalitis in athymic mice. Lab Anim Sci 44:639-641.

Small JD. 1984. Rodent and lagomorph health surveillance---Quality assurance. In: Fox J, Cohen B, Loew F, editors. Laboratory Animal Medicine. New York: Academic Press. p 709-723.

Smith AL. 1986. Serologic tests for detection of antibody to rodent viruses. In: Bhatt PN, et al. , editors. Viral and Mycoplasmal Infections of Laboratory Rodents: Effects on Biomedical Research. New York: Academic Press. p 731-739.

Thigpen JE, Ross PW. 1983. Viral cross contamination of rats maintained in a fabric-walled mass air flow system. Lab Anim Sci 33:446-450.

Trexler P. 1983. Gnotobiotics. In: Foster H, Small JD, Fox J, editors. The Mouse in Biomedical Research. Vol III. New York: Academic Press. p 1-15.

Waggie K, Kagiyama N, Allen A, Nomura T, editors. 1994. Manual of Mircobiologic Monitoring of Laboratory Animals. 2nd ed. (National Institutes of Health (NIH) publication no. 94-2498). Washington DC: GPO.

Weisbroth SH. 1996. Post-indigenous disease: Changing concepts of disease in laboratory rodents. Lab Anim 25:25-33.

FIGURE 1 Relationship of pathogenic potential and research interactions. Reprinted with permission from Charles River Laboratories, Wilmington, Massachusetts.

FIGURE 2 (A) Example of a risk assessment completed for SHR rat colony with positive skin cultures of Staphylococcus aureus. (B) Blank risk assessment form.

FIGURE 3 Movement of items across bioexclusion systems for animals that entail risk to the systems' ability to exclude unwanted microorganisms. Reprinted with permission from Charles River Laboratories, Wilmington, Massachusetts.

FIGURE 4 Barrier production room for rodents illustrates typical safeguards for preventing the introduction of unwanted microorganisms. Reprinted with permission from Charles River Laboratories, Wilmington, Massachusetts.