253
Laboratory Animals (1979) 13,253-256
Evaluation of room cleaning procedures in a laboratory animal facility ROBERT
P. SHIELDS,
BARBARA
SCHRAMM·
& NORMAN
E. BRAUNEt
Department of Laboratory Animal and Wildlife Medicine, College of Veterinary Medicine, University of Florida, Gainesville, Florida 32610, United States of America Summary Animal room cleaning procedures were developed that could be used routinely and economically in this animal facility. Bacterial samples from the floors of rooms housing rabbits, rats and mice provided a useful way to evaluate the effectiveness of the cleaning procedures, and to determine the in-use effectiveness of disinfectant solutions. Infection control is important in the operation of a laboratory animal facility (ILAR, 1972). However, the complexities of preventing the spread of infectious agents between animals confined in close proximity are enormous. The problems inherent in the control of. nosocomial infections have provided some insight (LaForce, 1977), but in a laboratory animal facility these problems are even more complex because of the variety of animal species involved. To develop an infection control program in any animal facility one must consider the animals, the equipment, the physical plant or housing, and the personnel. Guidelines to minimize the spread of infection from each of these sources are available (ILAR, 1972), but their implementation may vary from one facility to another. This is particularly true of room cleaning procedures. Providing clean rooms for research animals and the regular maintenance of these rooms to minimize the spread of infectious agents are basic and essential parts of any program. A significant portion of the funds allocated for laboratory animal care is used to comply with the National Institutes of Health recommendation of 'regular schedule of sanitary maintenance' (ILAR, 1972). It seemed worthwhile, therefore, to determine whether cleaning procedures actually accomplish their purpose. During the 1960s routine microbial sampling as a means of determining the bioload in hospitals attained a degree of popularity (Pryor, Vesley & Shaffer, 1967). Later studies have indicated that to be of practical value in disease control, microbial sampling • Present address: Route I, Box 80-H, SI. Augustine, Florida 32084, U.S.A. t Director of Animal Resources, Duke University Medical Center, PO Box 3180, Durham, North Carolina 27710, USA. Received 15 August 1978. Accepted 8 March 1979.
must be done with specific goals in mind rather than as a routine (Meshelany, 1977; Hammond, 1976; AHA, 1974: APHA, 1975). As a result, only specific goaloriented microbial monitoring techniques are now recommended for infection control programs in hospitals. The work presented in this report was initiated to determine if microbial sampling of floors in a laboratory animal facility would be of practical value in determining the effectiveness of room cleaning procedures. Materials and methods The animal facility in which this work was conducted was about 3 years old. The average animal room was 3·7 x 4·3 m, with a concrete floor sealed with a mixture of polyester resins (Barrier, Type B; National Chemical Corporation, 950 Watertown Street, West Newton, Massachusetts 02165, USA). Room temperature was maintained at 20-22°C and the humidity averaged 50%. Room air was not recirculated and . ventilation was at the rate of 12-15 changes per hour. Rabbits were housed individually in standard stainless-steel rabbit cages with aspen-wood shavings in the dropping pans. Rats were housed individually in stainless-steel hanging rat cages with heavy asphaltimpregnated paper beneath to catch droppings. Mice were maintained on aspen-wood shavings in plastic boxes with filter tops.
Room cleaning procedures A weekly system of physical and chemical methods was developed after consideration of personnel, cost, and the physical structure of the equipment and building. Routine mechanical procedures were employed to remove visible debris as thoroughly and economically as possible. Cages were changed and bedding or paper beneath them replaced. Rooms were swept with a nylon bristle push broom; adherent material was loosened by scraping. They were then mopped with a clean autoclaved synthetic-blend mophead which had been immersed in freshly-prepared disinfectant solution. A different mophead and disinfectant solution were used for each room. Dirty mopheads were soaked in a vat of disinfectant solution after use. Each afternoon these mopheads were removed from
254 the vat, washed in a tunnel cage washer, left to drain, and autoclaved. It was established (Litskey, 1966) that the phenolic disinfectant selected was effective against organisms isolated from this facility (Wellstood-Nuesse & Shields, 1976). The final pH of the solution was 3·6, within the reported effective pH range of the disinfectant. In order to determine the amount of disinfectant needed in the mop water to kill vegetative bacteria when employed with the mechanical procedures as given above, the following tests were conducted. Mop water was prepared with different amounts of disinfectant. Samples were taken for bacteriological culture immediately after mixing, immediately after the clean mop had been immersed in the solution and wrung out, 2 minutes after the mop had been immersed, and after the room had been mopped and the mop rinsed in solution. Rabbit rooms were used for these tests. Based on the results of these tests (see Results) the disinfectant concentration used was 15 ml per litre tap water. Surface sampling The Rodac plate method of surface sampling was employed. Rodac plates were filled with 16·5 trypticase soy agar (TSA) with lecithin and polysorbate 80 (BBL, Division of Becton, Dickinson & Co., PO Box 243, Cockeysville, Maryland 21030, USA) as neutralizers. They were incubated at 37°C for 24 h. A detailed discussion of the preparation of these plates and the sampling technique is described elsewhere (Fincher, 1965; Hall & Hartnett, 1964). Briefly, a raised agar surface was pressed firmly against a test site. Coordinates of floor test sites were determined from a table of random numbers and marked with numbered tape. 15 sites were sampled in each room after cleaning (after the floor dried and before traffic resumed). When sampling, the investigators wore clean, disposable shoe covers. . Colonies were counted with an electronic colony counter (Fisher Scientific Co., 711 Forbes Avenue, Pittsburgh, Pennsylvania 15219, USA). The maximum number of colonies that can be counted on a Rodac plate is about 200 (Pryor et al., 1967). Counts of 200 colonies or more are, therefore, indicated as too numerous to count (TNTC), and averaged in the data as 200. Spreading, coalescing growth of some Grampositive bacilli occasionally made colony counts more difficult. Representative colonies were picked for Gram staining, and biochemical studies were conducted to identify the genera of bacteria. Fungi and yeasts were identified microscopically. Results After the routine mechanical cleaning procedures had been established, it was necessary to determine the appropriate concentration of disinfectant. In the initial
Shields, Schramm & Braune tests, 8 ml disinfectant per litre of water was used. 3 rabbit rooms were tested and vegetative coliforms cultured from the mop bucket water in each case after the rooms had been mopped. No vegetative organisms were detected in any of the other samples. When the test was repeated on 3 other rabbit rooms using 15 ml disinfectant per litre of water, no vegetative organisms were cultured in any of the samples: it was concluded that 15 ml disinfectant per litre was an adequate concentration for our purposes. In later phases of this work, vegetative bacteria were occasionally isolated from cleaned animal-room floors. This occurred, however, in only 5 rooms (4 rabbit, 1 rat) of the 80 sampled (6·2%), and in only 15 of 1200 Rodac plates exposed (1.2%). Vegetative bacteria found in these specimens included Escherichia coli, a pseudomonad, and faecal streptococci. The principal organisms isolated were Gram-positive spore-forming rods (Bacillus sp.). The effectiveness of the cleaning procedures was evaluated by bacterial colony counts of samples taken from the floors. Baseline control (before cleaning) samples were collected from 3 rabbit rooms, 3 rat rooms and 3 mouse rooms. In each case most plates (> 50%) contained more than 200 colonies per plate-recorded as too numerous to count (TNTC). These plates ranged from 124-TNTC and averaged about 190 colonies per plate for each species (Table 1). After the rooms were cleaned, a total of 10 rabbit rooms were sampled 27 times, 6 mouse rooms were sampled 20 times, and 9 rat rooms were sampled 23 times. Each time a room was sampled, 15 Rodac plates were used for a total of 1228 samples. The rooms housing rabbits had the highest mean bacterial colony counts of 133 ± 45·8 sd. The lowest mean counts of 27 ± 20·9 sd were found in rat rooms, whereas mouse rooms averaged 74 ± 59·6 sd. It became apparent that with repeated cleanings the counts were lower; mean colony counts as well as the standard deviations of the means and the ranges decreased with successive samplings (Table 1). Rooms were cleaned once per week, but in some instances there was more than one cleaning between sampling periods. The 1st samples taken from rabbit rooms (after 1st cleaning) had a mean colony count of 146 ± 45·5 sd. All samples taken after that initial sampling averaged 118 ± 43·7 sd, but after the rooms had been cleaned at least twice samples averaged only 92 ± 15·4 sd: similar data are presented for mouse and rat rooms (Table 1). Discussion An effective cleaning procedure must significantly reduce the microbial load. Complete elimination of microorganisms (including spores) as a routine procedure is not practicable. In achieving what is, cost of equipment as well as personnel must be considered.
255
Evaluation of animal room cleaning
Table I. Bacterial colony counts offtoors in rooms housing laboratory animals
Rabbit rooms (10) Control (before cleaning) All test samples First sampling All after first sampling All after second sampling Mouse rooms (6) Control (before cleaning) All test samples First sampling All after first sampling All after second sampling All after third sampling Rat rooms (9) Control (before cleaning) All test samples First sampling All after first sampling All after second sampling
Range
No. roOms sampled
No. of samples
Colony count (mean ± scI)
3 27 10 17 9
45 405 150 255 135
197 (80%TNTC) 133 ± 45·8 146 ± 45·5 118±43·7 92 ± 15·4
161-TNTC* 70-TNTC 100-TNTC 70-TNTC 70-t21
3 30 6 24 18 12
45 450 90 360 270 180
189 (67%TNTC) 74 ± 59·6 108 ± 75·1 64 ± 49·7 54±35·7 42 ± 10·7
124-TNTC 32-TNTC 32-TNTC 33-191 33-182 33-66
3 23 9 14 6
45 345 135 210 90
191 (64%TNTC) 27 ± 20.9 39 ± 23·7 19 ± 15·1 10 ± 0·8
136-TNTC 9-77 13-77 9-66 9-11
*TNTC colonies too numerous to count, averaged as 200. Equipment should be readily available and easy to operate. Synthetic fibre mops were used because they were more durable and more easily cleaned than cotton mopheads. Mopping was preferred to hosing to avoid excessive moisture and additional steps in applying the disinfectant. Most cleaning methods include the use of a disinfectant to kill vegetative bacteria. Specific methods for testing the 'killing-power' of disinfectant solutions (Litskey, 1966; Wellstood-Nuesse & Shields, 1976) should always be employed to determine if the disinfectant is effective against the organisms present in a particular facility. Most disinfectants have an optimum pH range and it is essential to know that the water used does not render the solution ineffective. An 'in-use' test of the mop bucket before and after the rooms were mopped should illso be performed. The disinfectant was judged effective if no vegetative forms of bacteria were detected on the culture plate (spore forming bacteria were exceptions). Failure to kill vegetative organisms was corrected in this study by increasing the concentration of disinfectant above that recommended by the manufacturer. The room cleaning procedures adopted were judged effective when evaluated by the Rodac method of bacterial floor sampling. Not only were the bacterial counts lower after the rooms were cleaned (Table 1), l?ut with repeated cleaning they fell progressively, the ranges became narrower and fewer plates were TNTC. After the 2nd samplings the counts were essentially the same as those obtained by WellstoodNuesse & Shields (1976), work performed by a
different microbiologist and with a different type of disi!1fectant (chlorhexidine). This reproducibility lends support to the effectiveness of this evaluation system. After this study was completed, there was an outbreak of Streptococcus zooepidemicus dermatitis in one of the rat rooms. The organism was isolated from the floor of this rat room by the sampling techniques described. This was the only time this organism was isolated from the floors in this facility and indicates not only the effectiveness of these sampling procedures but the ability of such bacteria to spread in the environment. Rabbits consistently had higher bacterial counts than were found in rat or mouse rooms. Dropping pans were emptied or scraped clean in the rabbit rooms before sampling. Faecal material was often found on the floor near the cages in these rooms, and the wood shavings used in the dropping pans must also be considered a possible source of microflora. Since mice were housed in filter-top boxes, it was surprising that colony counts were higher than in rat rooms. This was contrary to our earlier findings (Wellstood-Nuesse & Shields, 1976). Although no specific explanation can be offered, it was established that human traffic was greater in the mouse rooms than in the rat rooms during the tests. It has been recommended that bacterial surveillance programs should be directed toward specific goals to be economically justifiable (Hammond, 1976; AHA, 1974; APHA, 1975). From this study and previous work in this animal facility we concur. Nevertheless, such procedures can be economically used occasion-
Shields, Schramm & Braune
256
ally in an animal facility to evaluate cleaning procedures, to evaluate a new disinfectant, to aid in controlling specific disease outbreaks, and to monitor specific problem areas such as those used for surgery, nursing or quarantine.
Acknowledgements This work was supported in part by Grant RR0068505 from the National Institutes of Health. The authors thank Dora M. Norman and Jesse Williams for their assistance with the room cleaning procedures.
References AHA (1974). Statement on microbiologic sampling in the hospital. Committee on Infection Within Hospitals, American Hospital Association. Hospitals 48, 125-126. APHA (1975). Environmental microbiological sampling in the hospital. Committee on Microbial Contamination of Surfaces, American Public Health Association. Health Laboratory Science 12,234-235. Fincher, E. L. (1965). Surface sampling-Application Methods. Recommendations. U.S. Department of Health, Education and Welfare, publication 66, 592-599. Hall, L. B. & Hartnett, M. 1. (1964). Measurement of the bacterial contamination on surfaces in hospitals. Public Health Reports, Washington 79, 1021-1024. Hammond, J. B. (1976). Guide to the prevention and control of hospital-associated bifections. U.S- Army, Environmental Hygiene Agency, 84-85. ILAR (1972). Guide for the care and use of laboratory animals. Committee on Revision of the Guide for
Laboratory Animal Facilities and Care of the Institute of Laboratory Animal Resources, National Research Council. U.S. Department of Health, Education, and Welfare, publication (NIH) 78-23. LaForce, F. M. (1977). The hospital infection control committee: a personal view. Hospital Practice 12, 135-148. Litskey, N. Y. (1966). Hospital sanitation, p. 124. Chicago: Clissold. Meshelany, C. (1977). ltifection control manual. Oradell, New Jersey: Medical Economics. Pryor, A. K., Vesley, D. & Shaffer, J. G. (1967). Cooperative microbial surveys of surfaces in hospital patient rooms. Health Laboratory Science 4, 153-159. Wellstood-Nuesse, S. & Shields, R. P. (1976). Environmental monitoring in a laboratory animal facility. Laboratory Animal Science 26,592-599.
Beurteilung von Raum-Reinigungsverfahren in einem Versuchstierbereich R. P. SHIELDS,
B. SCHRAMM
& N. E. BRAUNE
Zusammenfassung Tierraum-Reinigungsverfahren, welche routinemasslg und wirtschaftlich in diesem Versuchstierbereich eingesetzt werden konnten, wurden entwickelt. Proben zur bakteriologischen Beurteilung von Boden aus Kaninchen-, Ratten-
und Mausriiumen bewiihrten sich als eine niitzliche MethOde zur Beurteilung der Wirksamkeit der Reinigungsverfahren und des Wirkungsgrades der Desinfektionslosungen im effektiven Einsatz.
Laboratory Animals (1979) 13,253-256
Evaluation of room cleaning procedures in a laboratory animal facility ROBERT
P. SHIELDS,
BARBARA
SCHRAMM·
& NORMAN
E. BRAUNEt
Department of Laboratory Animal and Wildlife Medicine, College of Veterinary Medicine, University of Florida, Gainesville, Florida 32610, United States of America Summary Animal room cleaning procedures were developed that could be used routinely and economically in this animal facility. Bacterial samples from the floors of rooms housing rabbits, rats and mice provided a useful way to evaluate the effectiveness of the cleaning procedures, and to determine the in-use effectiveness of disinfectant solutions. Infection control is important in the operation of a laboratory animal facility (ILAR, 1972). However, the complexities of preventing the spread of infectious agents between animals confined in close proximity are enormous. The problems inherent in the control of. nosocomial infections have provided some insight (LaForce, 1977), but in a laboratory animal facility these problems are even more complex because of the variety of animal species involved. To develop an infection control program in any animal facility one must consider the animals, the equipment, the physical plant or housing, and the personnel. Guidelines to minimize the spread of infection from each of these sources are available (ILAR, 1972), but their implementation may vary from one facility to another. This is particularly true of room cleaning procedures. Providing clean rooms for research animals and the regular maintenance of these rooms to minimize the spread of infectious agents are basic and essential parts of any program. A significant portion of the funds allocated for laboratory animal care is used to comply with the National Institutes of Health recommendation of 'regular schedule of sanitary maintenance' (ILAR, 1972). It seemed worthwhile, therefore, to determine whether cleaning procedures actually accomplish their purpose. During the 1960s routine microbial sampling as a means of determining the bioload in hospitals attained a degree of popularity (Pryor, Vesley & Shaffer, 1967). Later studies have indicated that to be of practical value in disease control, microbial sampling • Present address: Route I, Box 80-H, SI. Augustine, Florida 32084, U.S.A. t Director of Animal Resources, Duke University Medical Center, PO Box 3180, Durham, North Carolina 27710, USA. Received 15 August 1978. Accepted 8 March 1979.
must be done with specific goals in mind rather than as a routine (Meshelany, 1977; Hammond, 1976; AHA, 1974: APHA, 1975). As a result, only specific goaloriented microbial monitoring techniques are now recommended for infection control programs in hospitals. The work presented in this report was initiated to determine if microbial sampling of floors in a laboratory animal facility would be of practical value in determining the effectiveness of room cleaning procedures. Materials and methods The animal facility in which this work was conducted was about 3 years old. The average animal room was 3·7 x 4·3 m, with a concrete floor sealed with a mixture of polyester resins (Barrier, Type B; National Chemical Corporation, 950 Watertown Street, West Newton, Massachusetts 02165, USA). Room temperature was maintained at 20-22°C and the humidity averaged 50%. Room air was not recirculated and . ventilation was at the rate of 12-15 changes per hour. Rabbits were housed individually in standard stainless-steel rabbit cages with aspen-wood shavings in the dropping pans. Rats were housed individually in stainless-steel hanging rat cages with heavy asphaltimpregnated paper beneath to catch droppings. Mice were maintained on aspen-wood shavings in plastic boxes with filter tops.
Room cleaning procedures A weekly system of physical and chemical methods was developed after consideration of personnel, cost, and the physical structure of the equipment and building. Routine mechanical procedures were employed to remove visible debris as thoroughly and economically as possible. Cages were changed and bedding or paper beneath them replaced. Rooms were swept with a nylon bristle push broom; adherent material was loosened by scraping. They were then mopped with a clean autoclaved synthetic-blend mophead which had been immersed in freshly-prepared disinfectant solution. A different mophead and disinfectant solution were used for each room. Dirty mopheads were soaked in a vat of disinfectant solution after use. Each afternoon these mopheads were removed from
254 the vat, washed in a tunnel cage washer, left to drain, and autoclaved. It was established (Litskey, 1966) that the phenolic disinfectant selected was effective against organisms isolated from this facility (Wellstood-Nuesse & Shields, 1976). The final pH of the solution was 3·6, within the reported effective pH range of the disinfectant. In order to determine the amount of disinfectant needed in the mop water to kill vegetative bacteria when employed with the mechanical procedures as given above, the following tests were conducted. Mop water was prepared with different amounts of disinfectant. Samples were taken for bacteriological culture immediately after mixing, immediately after the clean mop had been immersed in the solution and wrung out, 2 minutes after the mop had been immersed, and after the room had been mopped and the mop rinsed in solution. Rabbit rooms were used for these tests. Based on the results of these tests (see Results) the disinfectant concentration used was 15 ml per litre tap water. Surface sampling The Rodac plate method of surface sampling was employed. Rodac plates were filled with 16·5 trypticase soy agar (TSA) with lecithin and polysorbate 80 (BBL, Division of Becton, Dickinson & Co., PO Box 243, Cockeysville, Maryland 21030, USA) as neutralizers. They were incubated at 37°C for 24 h. A detailed discussion of the preparation of these plates and the sampling technique is described elsewhere (Fincher, 1965; Hall & Hartnett, 1964). Briefly, a raised agar surface was pressed firmly against a test site. Coordinates of floor test sites were determined from a table of random numbers and marked with numbered tape. 15 sites were sampled in each room after cleaning (after the floor dried and before traffic resumed). When sampling, the investigators wore clean, disposable shoe covers. . Colonies were counted with an electronic colony counter (Fisher Scientific Co., 711 Forbes Avenue, Pittsburgh, Pennsylvania 15219, USA). The maximum number of colonies that can be counted on a Rodac plate is about 200 (Pryor et al., 1967). Counts of 200 colonies or more are, therefore, indicated as too numerous to count (TNTC), and averaged in the data as 200. Spreading, coalescing growth of some Grampositive bacilli occasionally made colony counts more difficult. Representative colonies were picked for Gram staining, and biochemical studies were conducted to identify the genera of bacteria. Fungi and yeasts were identified microscopically. Results After the routine mechanical cleaning procedures had been established, it was necessary to determine the appropriate concentration of disinfectant. In the initial
Shields, Schramm & Braune tests, 8 ml disinfectant per litre of water was used. 3 rabbit rooms were tested and vegetative coliforms cultured from the mop bucket water in each case after the rooms had been mopped. No vegetative organisms were detected in any of the other samples. When the test was repeated on 3 other rabbit rooms using 15 ml disinfectant per litre of water, no vegetative organisms were cultured in any of the samples: it was concluded that 15 ml disinfectant per litre was an adequate concentration for our purposes. In later phases of this work, vegetative bacteria were occasionally isolated from cleaned animal-room floors. This occurred, however, in only 5 rooms (4 rabbit, 1 rat) of the 80 sampled (6·2%), and in only 15 of 1200 Rodac plates exposed (1.2%). Vegetative bacteria found in these specimens included Escherichia coli, a pseudomonad, and faecal streptococci. The principal organisms isolated were Gram-positive spore-forming rods (Bacillus sp.). The effectiveness of the cleaning procedures was evaluated by bacterial colony counts of samples taken from the floors. Baseline control (before cleaning) samples were collected from 3 rabbit rooms, 3 rat rooms and 3 mouse rooms. In each case most plates (> 50%) contained more than 200 colonies per plate-recorded as too numerous to count (TNTC). These plates ranged from 124-TNTC and averaged about 190 colonies per plate for each species (Table 1). After the rooms were cleaned, a total of 10 rabbit rooms were sampled 27 times, 6 mouse rooms were sampled 20 times, and 9 rat rooms were sampled 23 times. Each time a room was sampled, 15 Rodac plates were used for a total of 1228 samples. The rooms housing rabbits had the highest mean bacterial colony counts of 133 ± 45·8 sd. The lowest mean counts of 27 ± 20·9 sd were found in rat rooms, whereas mouse rooms averaged 74 ± 59·6 sd. It became apparent that with repeated cleanings the counts were lower; mean colony counts as well as the standard deviations of the means and the ranges decreased with successive samplings (Table 1). Rooms were cleaned once per week, but in some instances there was more than one cleaning between sampling periods. The 1st samples taken from rabbit rooms (after 1st cleaning) had a mean colony count of 146 ± 45·5 sd. All samples taken after that initial sampling averaged 118 ± 43·7 sd, but after the rooms had been cleaned at least twice samples averaged only 92 ± 15·4 sd: similar data are presented for mouse and rat rooms (Table 1). Discussion An effective cleaning procedure must significantly reduce the microbial load. Complete elimination of microorganisms (including spores) as a routine procedure is not practicable. In achieving what is, cost of equipment as well as personnel must be considered.
255
Evaluation of animal room cleaning
Table I. Bacterial colony counts offtoors in rooms housing laboratory animals
Rabbit rooms (10) Control (before cleaning) All test samples First sampling All after first sampling All after second sampling Mouse rooms (6) Control (before cleaning) All test samples First sampling All after first sampling All after second sampling All after third sampling Rat rooms (9) Control (before cleaning) All test samples First sampling All after first sampling All after second sampling
Range
No. roOms sampled
No. of samples
Colony count (mean ± scI)
3 27 10 17 9
45 405 150 255 135
197 (80%TNTC) 133 ± 45·8 146 ± 45·5 118±43·7 92 ± 15·4
161-TNTC* 70-TNTC 100-TNTC 70-TNTC 70-t21
3 30 6 24 18 12
45 450 90 360 270 180
189 (67%TNTC) 74 ± 59·6 108 ± 75·1 64 ± 49·7 54±35·7 42 ± 10·7
124-TNTC 32-TNTC 32-TNTC 33-191 33-182 33-66
3 23 9 14 6
45 345 135 210 90
191 (64%TNTC) 27 ± 20.9 39 ± 23·7 19 ± 15·1 10 ± 0·8
136-TNTC 9-77 13-77 9-66 9-11
*TNTC colonies too numerous to count, averaged as 200. Equipment should be readily available and easy to operate. Synthetic fibre mops were used because they were more durable and more easily cleaned than cotton mopheads. Mopping was preferred to hosing to avoid excessive moisture and additional steps in applying the disinfectant. Most cleaning methods include the use of a disinfectant to kill vegetative bacteria. Specific methods for testing the 'killing-power' of disinfectant solutions (Litskey, 1966; Wellstood-Nuesse & Shields, 1976) should always be employed to determine if the disinfectant is effective against the organisms present in a particular facility. Most disinfectants have an optimum pH range and it is essential to know that the water used does not render the solution ineffective. An 'in-use' test of the mop bucket before and after the rooms were mopped should illso be performed. The disinfectant was judged effective if no vegetative forms of bacteria were detected on the culture plate (spore forming bacteria were exceptions). Failure to kill vegetative organisms was corrected in this study by increasing the concentration of disinfectant above that recommended by the manufacturer. The room cleaning procedures adopted were judged effective when evaluated by the Rodac method of bacterial floor sampling. Not only were the bacterial counts lower after the rooms were cleaned (Table 1), l?ut with repeated cleaning they fell progressively, the ranges became narrower and fewer plates were TNTC. After the 2nd samplings the counts were essentially the same as those obtained by WellstoodNuesse & Shields (1976), work performed by a
different microbiologist and with a different type of disi!1fectant (chlorhexidine). This reproducibility lends support to the effectiveness of this evaluation system. After this study was completed, there was an outbreak of Streptococcus zooepidemicus dermatitis in one of the rat rooms. The organism was isolated from the floor of this rat room by the sampling techniques described. This was the only time this organism was isolated from the floors in this facility and indicates not only the effectiveness of these sampling procedures but the ability of such bacteria to spread in the environment. Rabbits consistently had higher bacterial counts than were found in rat or mouse rooms. Dropping pans were emptied or scraped clean in the rabbit rooms before sampling. Faecal material was often found on the floor near the cages in these rooms, and the wood shavings used in the dropping pans must also be considered a possible source of microflora. Since mice were housed in filter-top boxes, it was surprising that colony counts were higher than in rat rooms. This was contrary to our earlier findings (Wellstood-Nuesse & Shields, 1976). Although no specific explanation can be offered, it was established that human traffic was greater in the mouse rooms than in the rat rooms during the tests. It has been recommended that bacterial surveillance programs should be directed toward specific goals to be economically justifiable (Hammond, 1976; AHA, 1974; APHA, 1975). From this study and previous work in this animal facility we concur. Nevertheless, such procedures can be economically used occasion-
Shields, Schramm & Braune
256
ally in an animal facility to evaluate cleaning procedures, to evaluate a new disinfectant, to aid in controlling specific disease outbreaks, and to monitor specific problem areas such as those used for surgery, nursing or quarantine.
Acknowledgements This work was supported in part by Grant RR0068505 from the National Institutes of Health. The authors thank Dora M. Norman and Jesse Williams for their assistance with the room cleaning procedures.
References AHA (1974). Statement on microbiologic sampling in the hospital. Committee on Infection Within Hospitals, American Hospital Association. Hospitals 48, 125-126. APHA (1975). Environmental microbiological sampling in the hospital. Committee on Microbial Contamination of Surfaces, American Public Health Association. Health Laboratory Science 12,234-235. Fincher, E. L. (1965). Surface sampling-Application Methods. Recommendations. U.S. Department of Health, Education and Welfare, publication 66, 592-599. Hall, L. B. & Hartnett, M. 1. (1964). Measurement of the bacterial contamination on surfaces in hospitals. Public Health Reports, Washington 79, 1021-1024. Hammond, J. B. (1976). Guide to the prevention and control of hospital-associated bifections. U.S- Army, Environmental Hygiene Agency, 84-85. ILAR (1972). Guide for the care and use of laboratory animals. Committee on Revision of the Guide for
Laboratory Animal Facilities and Care of the Institute of Laboratory Animal Resources, National Research Council. U.S. Department of Health, Education, and Welfare, publication (NIH) 78-23. LaForce, F. M. (1977). The hospital infection control committee: a personal view. Hospital Practice 12, 135-148. Litskey, N. Y. (1966). Hospital sanitation, p. 124. Chicago: Clissold. Meshelany, C. (1977). ltifection control manual. Oradell, New Jersey: Medical Economics. Pryor, A. K., Vesley, D. & Shaffer, J. G. (1967). Cooperative microbial surveys of surfaces in hospital patient rooms. Health Laboratory Science 4, 153-159. Wellstood-Nuesse, S. & Shields, R. P. (1976). Environmental monitoring in a laboratory animal facility. Laboratory Animal Science 26,592-599.
Beurteilung von Raum-Reinigungsverfahren in einem Versuchstierbereich R. P. SHIELDS,
B. SCHRAMM
& N. E. BRAUNE
Zusammenfassung Tierraum-Reinigungsverfahren, welche routinemasslg und wirtschaftlich in diesem Versuchstierbereich eingesetzt werden konnten, wurden entwickelt. Proben zur bakteriologischen Beurteilung von Boden aus Kaninchen-, Ratten-
und Mausriiumen bewiihrten sich als eine niitzliche MethOde zur Beurteilung der Wirksamkeit der Reinigungsverfahren und des Wirkungsgrades der Desinfektionslosungen im effektiven Einsatz.
