General

According to FELASA recommendations at least 10 animals of each hygienic unit should be submitted to necropsy and microbiological routine tests in at least quarterly intervals (1). This applies to barrier and conventional environments of populations exceeding 100 animals if certain additional presuppositions are given (3). Very often the number of animals in hygienic units, using compartments intensively (e.g. filter cages), is much lower. This recommendation is intended to assist in the microbiological monitoring of such animal populations.

Hygienic unit

One hygienic unit consists of the area of a laboratory animal facility that is cared for by the same personnel (animal care personnel, research scientists) without in-between hygienic measures (i.e. showering, changing of clothes, hand-disinfection).

Examples could be:

• a barrier unit with several rooms, accessible through locks (shower, changing of clothes, disinfecting hands)
• several conventional animal locations in one building
• several animal locations on different floors or in different buildings, which are entered by the same persons without any preventive measures
• an isolator
• a microisolator cage (filter cage). It may act as an efficient barrier if the cage is only opened in ventilated workstations by using an appropriate sterilisation agent for the hands (gloves).

Choice of animals

The choice of the animals used for routine examination depends on several factors. Most suitable are animals directly from the animal facility, which were born there or which have spent a long time at this place. If those animals are not available, sentinels should be used

Sample size

The necessary sample size depends on the size of the animal stock, the prevalence rate of an eventual infection and the examination interval. With an inventory of more than 100 animals, at random sampling by routine examination of 10 animals results in a 95% chance of detecting an infection if at least 30% of a population is infected. Theoretically, with 4 examinations per year, an infection with a prevalence of 10% can be detected with nearly 99% certainty (2). If the risk of introducing agents into an experimental unit is very high (e.g. introduction of animals more than one time per month, introduction of animals from different breeders, frequent entry of research personnel in addition to animal care staff) more frequent monitoring is necessary. In such cases, a sample size of at least 3-5 animals is recommended to be monitored at least per month.

Animals

Preferably old breeding animals (retired breeders) should be used, as these, due to their long presence in the facility and their decreasing immunocompetence, very often harbour the broadest spectrum of agents. Except for serological and certain parasitological (e.g. helminth) examinations, freshly weaned young animals also are very suitable. The animals are preferably taken from different locations and cages.

Sick animals

Valuable additional information is obtained through sick animals. Generally, all sick animals should be submitted to examination in addition to scheduled testing so that current infections can be detected at an early stage. The research personnel should be instructed to deliver all animals to microbiological and pathological examination which show any sign of sickness or appear moribund or dead.

Sentinel animals

Definition and use

Sentinel animals act as surveillance substitutes in order to determine whether or not infectious agents are existing. They will be used when insufficient numbers of animals are available for health monitoring (e.g. animals involved in an experiment, small populations) or when it is inappropriate to carry out health monitoring on animals from the hygienic unit to be examined (e.g. serological testing of immunodeficient mice may be misleading). If sentinels are not bred within the colony that is being monitored, they must be obtained from a colony with known microbiological status (see “selection criteria”).

Selection criteria for sentinel animals

• animals free from all agents to be monitored and free from antibodies to these agents
• the same animal species like the one of the population to be examined
• young adult animals, which show a good immune reaction (serology)

There are different opinions concerning the genetics of the animals to be used:

• Outbred animals are usually more robust and fertile and are therefore relatively cheap. They are usually sensitive to a broad spectrum of agents but are generally considered more resistant to clinical disease.
• Inbred strains (e.g. DBA/2, BALB/c) are more difficult to breed and are therefore more expensive as compared to outbred animals. Susceptibility to infections and to clinical disease may vary between strains. For example, infection of A/J mice by MHV does usually not result in clinical signs whereas the morbidity and mortality rate may be very high in other mouse strains (e.g. C57BL/6). In contrast, infection of C57BL/6 (and other highly resistant strains such as C57BL/10) with ectromelia virus is frequently inapparent whereas infection of susceptible strains (e.g. A, DBA/2, BALB/c, C3H) may result in sudden morbidity and mortality affecting 80-90 % of a colony.
• Immunodeficient animals are usually very expensive. They are more susceptible to clinical disease than immunocompetent animals and therefore require specific housing conditions to avoid disease caused by environmental organisms and other agents that are not relevant for the population to be tested. They may not respond to an infection by seroconversion so that infections may not be detectable by serology. For these reasons they are less suitable as sentinels. However, while most infections are self-limiting in immunocompetent animals, immunodeficient animals are frequently persistently infected which may facilitate agent isolation.

As a general rule, one may use outbred animals as sentinels. The use of inbred animals or mutants as sentinels is to be favoured in case special agents, for which these animals are known to be suscebtible, shall be detected or excluded.

In experimental animal facilities, usually animals of the same species as that to be examined, free from all agents to be monitored and free from antibodies to these agents will be used as sentinels. Whenever possible, they should be derived from the same breeding area as the experimental animals and should be ordered at the same time. It is important that by thorough choice of the breeding areas an introduction of agents via added sentinel animals is excluded. This is of special importance for populations of immunodeficient animals.

Acquisition of sentinels should depend upon the type of experiments.

In hygienic units where re-occupation occurs (short-term experiments, multipurpose units), at least (10)-20 animals should always be available at short notice. It is recommended to order the demand of sentinels for a longer period, e.g. for 6 months. Thus, an adequate long exposition period in the area to be examined is ensured.

In units where no re-occupation occurs (long-term experiments, only one type of experiment, all-in all-out practices), the number of sentinels is based upon the estimated time of the experiment. If possible, an adequate number of sentinels should be available during the complete time period of the experiment. Thus, the risk of agent introduction by added sentinels can be eliminated.

With immunodeficient animals (e.g. thymus-aplastic Foxn1^nu mice or Whn^rnu rats, B- and T-cell-deficient Prkdc^scid or Rag^tm1Mom mice) immunocompetent animals will be kept as sentinels in the same unit (it is best to use heterozygous littermates with nude mice). It is important that the sentinel animals are free of opportunistic agents (e.g. Staphylococcus aureus, Pneumocystis carinii) that can cause diseases in immunodeficient animals.

Husbandry of sentinel animals

• same husbandry conditions as for the rest of the stock.
• In order to enhance transmission of an infection, the sentinel animals should receive soiled bedding from other cages.
• exposition period in the area to be examined: 4 weeks at least
• cages preferably well beneath on the rack, evenly distributed in the room
• cages with sentinel animals should always be handled and changed at last

Use of sentinel animals

Sentinel animals are not randomly selected from a population. They rather are animals which were consciously kept in a way that they will face a higher infection risk. Their husbandry is exclusively oriented towards the fact that in a certain area existing infectious agents will be transmitted to them with the highest possibility. With a correct animal husbandry (on soiled bedding, housing preferably well beneath on the cage shelves), reliable results can be obtained for most agents, even with smaller sample sizes (animal numbers). Especially in small populations (e.g. individually ventilated cages, isolators) an even higher likelihood of agent transmission to sentinels is given if contact sentinels are used which are housed in the same cage as the animals to be monitored (5).

Health monitoring in conventional or barrier units

(see also “sample size”)

The FELASA recommendations can be used with the mostly practised conventional or barrier housing of rodents (in open cages).

Health monitoring in isolators

Mostly, very small populations are kept in isolators and there is not enough space to keep an adequate number of animals for health monitoring as recommended for open housing. Appropriate monitoring is therefore only possible by keeping a realistic number of animals in an isolator consequently on soiled bedding (and with soiled feed) from preferably many (all) cages. Use of contact sentinels may also be appropriate. According to the size of the isolator one or several cages with sentinel animals should be scheduled and for infection controls approximately 3-5 animals are recommended per sampling together with an increased monitoring frequency (1).

Health monitoring in filtered cabinets

If properly handled, filter cabinets represent a hygienic unit, wherein smaller animal populations can be isolated from other animal populations, which are kept outside a cabinet or in other cabinets. Within a cabinet it is possible – in case of housing in open cages – that microorganisms spread between cages or will be transmitted through the air. Due to the small number of animals kept in such cabinets and the risk or extent of a transfer of microorganisms from cage to cage (which are both difficult to estimate), a sentinel programme similar to animal husbandry in isolators or microisolators is reasonable.

Health monitoring in microisolator cages (IVCs)

(see also “Handling of microisolator cages”)

When microisolator cages are used, the hygienic unit is not the room but rather every single cage. This principle very efficiently inhibits the transfer of infections between the cages if proper husbandry techniques are used. At the same time it hampers the use of sentinel animals for health monitoring, as a homogeneous spread of infectious agents within the population will be reduced or even prevented. This refers to static microisolator cages (i.e. filter lid cages) as well as to individually ventilated microisolator cages (IVCs) on special racks. It is now generally accepted that filter lid cages are not recommended for regular husbandry because ammonia concentrations may reach harmful levels rapidly. As for practical reasons not from all cages a sample can be taken, one has to make a compromise. The following procedure is based on soiled-bedding sentinels and enables monitoring without the need for breaching the microisolator principle:

• Sentinel animals are also kept in filter cages.
• After changing of the cages in a sterile workbench, bedding samples are taken from as many dirty cages as possible and put in a separate cage. The sentinel animals will then be transferred into this cage.
• Weekly changing of the bedding donors results in a good profile throughout the colony.

Literature:

1. FELASA working group on health monitoring of rodent and rabbit colonies (2002). Recommendations for the health monitoring of rodent and rabbit colonies in breeding and experimental units. Lab. Anim. 36: 20-42.
2. GV-SOLAS  (1989). Mikrobiologische Diagnostik bei Laboratoriumstieren. GV-SOLAS Veröffentlichung Nr. 11.
3. ILAR (1976). Long term holding of laboratory rodents. ILAR News 19, L1-L25.
4. Lipman, N. S. and F. R. Homberger (2003). Rodent quality assurance testing: use of sentinel animal systems. Lab Anim. (NY) 32 (5): 36-43.
5. Compton, S. R., F. R. Homberger, (2004). Microbiological monitoring in individually ventilated cage systems. Lab. Anim. (NY) 33: 36-41.
6. Compton, S. R., F. R. Homberger, F. X. Paturzo, and J. M. Clark (2004). Efficacy of three microbiological monitoring methods in a ventilated cage rack. Comp. Med. 54: 382-392.
7. Brielmeier, M., E. Mahabir, J. R. Needham, C. Lengger, P. Wilhelm, and J. Schmidt (2006). Microbiological monitoring of laboratory mice and biocontainment in individually ventilated cages: a field study. Lab. Anim. 40: 247-260.

Authors: Felix R. Homberger, Yale University; Werner Nicklas, DKFZ Heidelberg

Revised by Bettina Kränzlin, Universität Heidelberg

Date: 10/10/06