- Email: [email protected]
Investigation of gaseous ozone for MRSA decontamination of hospital side-rooms
Journal
of Hospital
Infection
( I 998)
M
40: 6 I-65
Investigation of gaseous ozone for MRSA decontamination of hospital side-rooms A. W. Berrington
and S. J. Pedler
Department of Medical Microbiology, Royal Victoria Infirmary Newcastle upon Tyne NE/ 4LFj UK
Queen Victoria Road,
Summary: A domestic, gaseous ozone generator was investigated for use in the decontamination of hospital side-rooms that have housed patients colonized with methicillin-resistant Staphylococcus ~UY~US(MRSA). Three models of bacterial contamination were used. These were exposed to ozone generation in a standard hospital side-room for 4 and 7 h. A methicillin-sensitive and a methicillinresistant strain of S. aureus were compared. Ozone concentrations of 0.14 ppm were reached, levels which are sufficient to cause mild pulmonary toxicity. Bacterial counts were reduced in the vicinity of the gas generator in most instances, but the effect elsewhere in the room was, at best, limited. MRSA appeared more resistant to the effects of ozone than methicillin-sensitive S. uureus. We conclude that the device tested would be inadequate for the decontamination of such hospital siderooms. Keywords:
ozone; disinfection; S. aureus;
MRSA.
Introduction Environmental disinfection with chemical agents removes a greater proportion of contaminating organisms than does physical cleaning alone.’ Although the extent to which either contributes to a reduction in the transmission of infection in hospitals is usually unknown, the use of disinfectants is standard practice in a variety of common clinical situations. For example, current guidelines for controlling the spread of methicillin-resistant Staphylococcus aureus (MRSA) recommend the use of a phenolic disinfectant when environmental contamination is likely.* It is our practice, following
Received 2 December 25 February 1998.
0 195-670
I /98/09006
1997;
I + 05 $12.00/O
revised
manuscript
accepted
of colonized patients, to dedischarge contaminate hospital side-rooms using chlorinereleasing agents. Such methods have inherent drawbacks including cost, labour and environmental concerns. Chlorine-releasing agents have the added disadvantage of potential toxicity, and we delay the use of a cleaned area until chlorine is no longer detectable by smell. This policy can result in clean side-rooms remaining empty for 24 h. Ozone is known to have antibacterial activity, is cheap to generate and although toxic, rapidly dissociates to oxygen. As a decontamination agent, gaseous ozone therefore offers potential advantages over chlorine-releasing agents and other disinfectants. We have investigated whether a commercially available ozone-generation system could be used to reduce environmental contamination with MRSA in hospitals.
0
I998
The
Hospital
Infection
Society
62
A. W. Berrington
0.16
Methods
.
0.14 -
A domestic ozone generator (Biosan 2000, Enright Industries Ltd., UK) was sited in the corner of a standard hospital side-room of volume 33.1 m3. All investigations were performed with the doors and windows closed. In a preliminary experiment, ozone concentrations were measured during gas generation using a handheld detector (Accuro Gas Detector Pump, Dra.. gerwerk A.G., Germany). Three models of contamination were investigated. A single clinical strain each of methicillin-sensitive S. UUY~US(MSSA) and of MRSA (EMRSA 15) was used, and suspensions made in sterile 0.9% saline. In model A, a defined volume (0.05 mL) of each suspension was dropped onto blood agar plates and dried by evaporation. In model B the suspensions were dropped onto squares of sterile filter paper and dried. In model C the suspensions were dropped and dried onto sterile glass squares. Plates, filter papers and glass squares were then exposed in the experimental field and ozone generation started. Ozone dissociates rapidly and previous studies have demonstrated a concentrated gradient with increasing distance from the machine.3 For this reason the organisms were exposed in two positions; close to the generator (S-l 5 cm) to provide the highest concentrations of ozone, and 300 cm from it to provide the lowest. Controls comprised inoculated specimens placed in a neighbouring side-room without ozone, and uninoculated agar plates, filter papers and glass squares. Samples were removed at 4 and 7 h and surviving organisms quantified immediately. In model A survivors were quantified by counting colonies after incubation at 37°C in aerobic conditions for 18-24 h. In model B survivors were quantified by laying the filter paper squares onto blood agar plates, incubating as above and counting colonies. In model C the glass squares were vortexed with a defined volume (3 mL) of sterile saline for 60 s, and 0.05 mL dropped and dried onto blood agar plates. For each model an appropriate suspension density was determined by prior experimentation. Colony
and S. J. Pedler
2 ,a
. . 0.12 . . . . . o.l E:‘::::::~:::.:.:::i:~:~~:~:~:~:::::i:i::~:::::::::::~:~:l
!j
0.08 -
:
0.06
-
.
2 Period of ozone generation l .
a
0
0.04 .
0.02 1 I 20
0
Figure
I
Ozone
1 40
I 60 Time
concentration generation.
I 80 (min) during
.
I 100
120
and
after
counts were analysed by the unpaired, tailed t-test.
1 140
gas
two-
Results Ozone measurement Ozone levels were measured in the centre of the room and were undetectable at baseline. With the generator operating, levels climbed during the first hour to reach a plateau of 0.13 ppm; when generation was discontinued levels fell below 0.05 ppm within 30 min (see Figure 1). When levels within the room were at their maximum, ozone was undetectable outside the closed door. Model A Countable colonies were achieved using a lo-’ dilution of a suspension of one colony of an overnight culture of S. UUY~USin 10 mL saline. For each organism, exposure category and exposure duration, 10 groups of colonies were counted (Table I). For MSSA the average colony count following 4 h exposure to room air was 62.8. After 4 h exposure to ozone within 15 cm of the generator the average colony count was 3.9 (P
Gaseous
Table
Model
ozone
I
Average
A
Model
B
Model
C
for
MRSA
survivor
63
decontamination
counts
(number
of observations
mode
in brackets) 7h
Duration:
4h
Exposure:
Air
High ozone
MSSA MRSA MSSA MRSA MSSA MRSA
62.8 (IO) 29.5 (IO) 40.2 (5) 50.0 (5) 68 (3) 62 (3)
3.9 31.7 7.0 0.0 33.7 48.7
(IO) (IO) (5) (5) (3) (3)
P-value
Low ozone
57.7
(IO)
29.9 8.4 28.4 37.7 65.3
(IO) (5) (5) (3) (3)
co.0 I ~0.01
7 h exposure, with colony counts averaging 61.7 after exposure to air, 4.2 after exposure to ozone in high concentration (93% reduction, P
Model
0
Countable colonies were achieved using a lop2 dilution of a suspension of one colony of an overnight culture of S. ~UY~USin 10 mL saline. For each organism, exposure category and exposure duration, five groups of colonies were counted (Table). For MSSA the average colony count following 4 h exposure to room air was 40.2. After 4 h exposure close to the generator the average colony count was 7.0 (83% reduction, P
P-value
Air
0.14 0.88
61.7 29.4 17.2 26.8 5.7 27.0
High ozone (IO) (IO) (5) (5) (3) (3)
4.2 18.4 0.6 0.0 3.7 36.7
(IO) (IO) (5) (5) (3) (3)
P-value
Low ozone
58.0 28.1 8.0 7.4 12.7
F-value
(IO) (IO) (5) (5) (3)
0.20 0.66 0.16 co.05
23 (3)
The average count following 7 h exposure near the generator was 0.6, a significant further reduction of 97% (P
Model
C
Countable colonies were achieved using a suspension of one colony of an overnight culture of S. uuwus in 2mL saline. Throughout the experimental period this model gave the least consistent results. Although for each organism, exposure category and exposure duration only three groups of colonies were counted, the
64
results ceding
A. W. Berrington
are in general agreement observations (Table).
with
the pre-
Discussion Ozone, or triatomic oxygen, is generated naturally in the upper atmosphere by the action of solar ultraviolet light on oxygen, and may be produced in the lower atmosphere through the photochemical oxidation of pollutants. Background levels are usually of the order of 0.02-0.03 ppm. Ozone may be generated using ultraviolet radiation or electrical discharge, and is usually produced in situ, since its extreme oxidative reactivity precludes storage and transportation. Most studies into the antibacterial effects of ozone have been performed in aqueous solution, where its bactericidal concentration for vegetative bacteria is between O-1-0.2 ppm.4 Activity has also been demonstrated against bacterial spores,’ cryptosporidial oocysts,6 and viruses.’ Less is known about the bactericidal effects of gaseous ozone although there is evidence that it is enhanced by humidity.” It probably acts through oxidation of cell-wall targets such as fatty acids and peptides. Ozone has various industrial uses including wide application in water disinfection, for example in the control of legionellae in hot-water systems.’ Commercially available ozone generators are usually marketed for the ‘improvement of air Manufacturers may assert that this is quality’. achieved through the oxidation of airborne gases or particulates, but there is evidence to the contrary and the main effect may simply be through olfactory camouflage.” Ozone is toxic to man, and has been shown in controlled experiments to cause transient respiratory symptoms, impairment in respiratory function and inflammatory changes at levels below 0.2 ppm. *’ At the levels reached during our experiment in the middle of a hospital side-room, most people would be symptomatically unaffected while a few might experience cough and pain on deep breathing,
and S. J. Pedler
particularly if engaged in heavy exercise. However, concerns remain about the long-term implications of exposure to ozone, particularly its potential for genotoxicity.” The 15 min exposure limit currently recommended is 0.3 ppm. The threshold for smell is 0.1 ppm. The surface of a blood agar plate is a highly nutritious (and atypical) environment for a vegetative organism. Under these circumstances S. uuwus survived equally well in air and in the concentrations of ozone reached at 3 m from the gas generation equipment. After 4 h exposure close to the ozone generator counts of MSSA had been reduced by over SO%, although a further 3 h exposure did not markedly improve the antibacterial effect. MRSA appeared less susceptible, with a significant reduction appearing only at 7 h when counts were still more than 60% of their unexposed counterparts. Filter paper squares proved less amenable to bacterial survival, with significant reductions between 4 and 7 h when both organisms were exposed to room air. Ozone provided an additional bactericidal effect against each organism for both high and low inocula. This was marked close to the machine but in this model still significant at a greater distance, although the reductions are unlikely to be sufficient for room decontamination. The surface of glass squares appeared even more hostile based on the reductions in unexposed counts between 4 and 7 h. This model was the most difficult to use and yielded the least consistent colony counts. Although the numbers are limited they support the previous observations that the effect of the generator tested, if any, was small, confined to organisms exposed close to the equipment, and more apparent with MSSA than MRSA. These results suggest that gaseous ozone at the levels reached at the periphery of a standard hospital side-room using the equipment tested is not sufficient to decontaminate the hospital environment, although it may reduce counts slightly on some surfaces. Activity was greater within 15 cm of the equipment, which raises the possibility of success using a more powerful ozone generator, but the effect was limited in some models and the use of higher ozone concentrations would carry a greater risk of toxicity.
Gaseous
ozone
for
MRSA
65
decontamination
It was interesting that throughout the experiments the methicillin-resistant strain of S. UUY~USappeared less susceptible to the effects of ozone than the methicillin-sensitive strain. This has been observed previously with ozone in solution,13 and presumably reflects differences in the structure of the cell wall. Given the organism’s reputation for environmental persistence we were surprised by its apparently poor survival on filter paper and glass. The use of gaseous ozone in hospital infection control has been investigated before.14 These authors exposed inoculated agar plates to gaseous ozone for 4 h. Bactericidal activity was demonstrated at levels of 0.3-0.9 ppm reached in a cupboard, but their machine was insufficiently powerful to generate detectable ozone in a hospital room. Other investigators have reported success using ozone to decontaminate bioclean rooms,” although the levels achieved were also high enough to corrode rubber. In conclusion, a commercially available ozone generator generated 0.1-0.15 ppm ozone in a standard hospital side-room, levels at which some people might experience respiratory symptoms but which are below the recommended 15 min exposure limits. None was detectable outside the room. Over a period of between 4 and 7 h gaseous ozone showed activity against two strains of S. UUY~US on three different surfaces, but the effect appeared to be concentration dependent and at the periphery of the room was limited at best. The MRSA strain was less susceptible to ozone than the MSSA strain. These results demonstrate that the ozone generator investigated would not be suitable for use in the decontamination of hospital siderooms after discharge of patients colonized with MRSA.
Acknowledgement With thanks to Enright of the equipment.
Industries
for the loan
References 1. Ayliffe GIJ, Coates D, Hoffman Disinfection in Hospitals. London:
PN. Chemical PHLS 1994.
2. Report of a combined working party of the Hospital Infection Society and British Society for Antimicrobial Chemotherapy. Revised guidelines for the control of epidemic methicillin-resistant Staphylococcus aweus. J Hosp Iflfect 1990; 16: 351-377. 3. Company data. 4. Broadwater WT, Hoehn RC, King PH. Sensitivity of three selected bacterial species to ozone. Appl Micvobiol 1973; 26: 391-393. 5. Ishizaki K, Shinriki N, Matsuyama H. Inactivation of Bacillus spores by gaseous ozone. J Appl Bacterial 1986; 60: 67-72. 6. Peeters JE, Mazas EA, Masschelein WJ, Martinez de Maturana IV, Debacker E. Effect of disinfection of drinking water with ozone or chlorine dioxide on survival of Cryptospovidium pavvum oocysts. Appl Environ Microbial 1989; 55: 1519-1522. 7. Vaughn JM, Chen YS, Lindburg K, Morales D. Inactivation of human and simian rotaviruses by ozone. Appl Environ Microbial 1987; 53: 2218-2221. 8. Doroszkiewicz V, Sikorska I, Jankowski S. Studies on the influence of ozone on complementmediated killing of bacteria. FEMS Immunol Med Microbial 1994; 9: 281-286. 9. Hart CA, Makin T. Legionella in hospitals: a review. J Hosp Infect 1991; 18 (Suppl. A): 481489. 10. Boeniger MF. Use of ozone generating devices to improve indoor air quality. Am Ind Hyg Assoc J 1995; 56: 590-598. 11. Krishna MT, Mudway I, Kelly FJ, Frew AJ, Holgate ST. Ozone, airways and allergic airways disease. Clin Exp Allergy 1995; 25: 1150-1158. 12. Victorin K. Review of the genotoxicity of ozone. Mutat Res 1992; 277: 221-238. 13. Yamayoshi T, Tatsumi N. Microbicidal effects of ozone solution on methicillin resistant Staphylococcus aweus. Drugs Exp Clin Res 1993; 19: 59-64. 14. Dyas A, Boughton BJ, Das BC. Ozone killing action against bacterial and fungal species; microbiological testing of a domestic ozone generator. J Clin Path01 1983; 36: 1102-1104. 15. Masaoka T, Kubota Y, Namiuchi S, Takubo T, Ueda T, Shibata H, Nakamura H, Yoshitake J, Yamayoshi T, Doi H, Kamiki T. Ozone decontamination of bioclean rooms. Appl Environ Microbial 1982; 43: 509-513.
of Hospital
Infection
( I 998)
M
40: 6 I-65
Investigation of gaseous ozone for MRSA decontamination of hospital side-rooms A. W. Berrington
and S. J. Pedler
Department of Medical Microbiology, Royal Victoria Infirmary Newcastle upon Tyne NE/ 4LFj UK
Queen Victoria Road,
Summary: A domestic, gaseous ozone generator was investigated for use in the decontamination of hospital side-rooms that have housed patients colonized with methicillin-resistant Staphylococcus ~UY~US(MRSA). Three models of bacterial contamination were used. These were exposed to ozone generation in a standard hospital side-room for 4 and 7 h. A methicillin-sensitive and a methicillinresistant strain of S. aureus were compared. Ozone concentrations of 0.14 ppm were reached, levels which are sufficient to cause mild pulmonary toxicity. Bacterial counts were reduced in the vicinity of the gas generator in most instances, but the effect elsewhere in the room was, at best, limited. MRSA appeared more resistant to the effects of ozone than methicillin-sensitive S. uureus. We conclude that the device tested would be inadequate for the decontamination of such hospital siderooms. Keywords:
ozone; disinfection; S. aureus;
MRSA.
Introduction Environmental disinfection with chemical agents removes a greater proportion of contaminating organisms than does physical cleaning alone.’ Although the extent to which either contributes to a reduction in the transmission of infection in hospitals is usually unknown, the use of disinfectants is standard practice in a variety of common clinical situations. For example, current guidelines for controlling the spread of methicillin-resistant Staphylococcus aureus (MRSA) recommend the use of a phenolic disinfectant when environmental contamination is likely.* It is our practice, following
Received 2 December 25 February 1998.
0 195-670
I /98/09006
1997;
I + 05 $12.00/O
revised
manuscript
accepted
of colonized patients, to dedischarge contaminate hospital side-rooms using chlorinereleasing agents. Such methods have inherent drawbacks including cost, labour and environmental concerns. Chlorine-releasing agents have the added disadvantage of potential toxicity, and we delay the use of a cleaned area until chlorine is no longer detectable by smell. This policy can result in clean side-rooms remaining empty for 24 h. Ozone is known to have antibacterial activity, is cheap to generate and although toxic, rapidly dissociates to oxygen. As a decontamination agent, gaseous ozone therefore offers potential advantages over chlorine-releasing agents and other disinfectants. We have investigated whether a commercially available ozone-generation system could be used to reduce environmental contamination with MRSA in hospitals.
0
I998
The
Hospital
Infection
Society
62
A. W. Berrington
0.16
Methods
.
0.14 -
A domestic ozone generator (Biosan 2000, Enright Industries Ltd., UK) was sited in the corner of a standard hospital side-room of volume 33.1 m3. All investigations were performed with the doors and windows closed. In a preliminary experiment, ozone concentrations were measured during gas generation using a handheld detector (Accuro Gas Detector Pump, Dra.. gerwerk A.G., Germany). Three models of contamination were investigated. A single clinical strain each of methicillin-sensitive S. UUY~US(MSSA) and of MRSA (EMRSA 15) was used, and suspensions made in sterile 0.9% saline. In model A, a defined volume (0.05 mL) of each suspension was dropped onto blood agar plates and dried by evaporation. In model B the suspensions were dropped onto squares of sterile filter paper and dried. In model C the suspensions were dropped and dried onto sterile glass squares. Plates, filter papers and glass squares were then exposed in the experimental field and ozone generation started. Ozone dissociates rapidly and previous studies have demonstrated a concentrated gradient with increasing distance from the machine.3 For this reason the organisms were exposed in two positions; close to the generator (S-l 5 cm) to provide the highest concentrations of ozone, and 300 cm from it to provide the lowest. Controls comprised inoculated specimens placed in a neighbouring side-room without ozone, and uninoculated agar plates, filter papers and glass squares. Samples were removed at 4 and 7 h and surviving organisms quantified immediately. In model A survivors were quantified by counting colonies after incubation at 37°C in aerobic conditions for 18-24 h. In model B survivors were quantified by laying the filter paper squares onto blood agar plates, incubating as above and counting colonies. In model C the glass squares were vortexed with a defined volume (3 mL) of sterile saline for 60 s, and 0.05 mL dropped and dried onto blood agar plates. For each model an appropriate suspension density was determined by prior experimentation. Colony
and S. J. Pedler
2 ,a
. . 0.12 . . . . . o.l E:‘::::::~:::.:.:::i:~:~~:~:~:~:::::i:i::~:::::::::::~:~:l
!j
0.08 -
:
0.06
-
.
2 Period of ozone generation l .
a
0
0.04 .
0.02 1 I 20
0
Figure
I
Ozone
1 40
I 60 Time
concentration generation.
I 80 (min) during
.
I 100
120
and
after
counts were analysed by the unpaired, tailed t-test.
1 140
gas
two-
Results Ozone measurement Ozone levels were measured in the centre of the room and were undetectable at baseline. With the generator operating, levels climbed during the first hour to reach a plateau of 0.13 ppm; when generation was discontinued levels fell below 0.05 ppm within 30 min (see Figure 1). When levels within the room were at their maximum, ozone was undetectable outside the closed door. Model A Countable colonies were achieved using a lo-’ dilution of a suspension of one colony of an overnight culture of S. UUY~USin 10 mL saline. For each organism, exposure category and exposure duration, 10 groups of colonies were counted (Table I). For MSSA the average colony count following 4 h exposure to room air was 62.8. After 4 h exposure to ozone within 15 cm of the generator the average colony count was 3.9 (P
Gaseous
Table
Model
ozone
I
Average
A
Model
B
Model
C
for
MRSA
survivor
63
decontamination
counts
(number
of observations
mode
in brackets) 7h
Duration:
4h
Exposure:
Air
High ozone
MSSA MRSA MSSA MRSA MSSA MRSA
62.8 (IO) 29.5 (IO) 40.2 (5) 50.0 (5) 68 (3) 62 (3)
3.9 31.7 7.0 0.0 33.7 48.7
(IO) (IO) (5) (5) (3) (3)
P-value
Low ozone
57.7
(IO)
29.9 8.4 28.4 37.7 65.3
(IO) (5) (5) (3) (3)
co.0 I ~0.01
7 h exposure, with colony counts averaging 61.7 after exposure to air, 4.2 after exposure to ozone in high concentration (93% reduction, P
Model
0
Countable colonies were achieved using a lop2 dilution of a suspension of one colony of an overnight culture of S. ~UY~USin 10 mL saline. For each organism, exposure category and exposure duration, five groups of colonies were counted (Table). For MSSA the average colony count following 4 h exposure to room air was 40.2. After 4 h exposure close to the generator the average colony count was 7.0 (83% reduction, P
P-value
Air
0.14 0.88
61.7 29.4 17.2 26.8 5.7 27.0
High ozone (IO) (IO) (5) (5) (3) (3)
4.2 18.4 0.6 0.0 3.7 36.7
(IO) (IO) (5) (5) (3) (3)
P-value
Low ozone
58.0 28.1 8.0 7.4 12.7
F-value
(IO) (IO) (5) (5) (3)
0.20 0.66 0.16 co.05
23 (3)
The average count following 7 h exposure near the generator was 0.6, a significant further reduction of 97% (P
Model
C
Countable colonies were achieved using a suspension of one colony of an overnight culture of S. uuwus in 2mL saline. Throughout the experimental period this model gave the least consistent results. Although for each organism, exposure category and exposure duration only three groups of colonies were counted, the
64
results ceding
A. W. Berrington
are in general agreement observations (Table).
with
the pre-
Discussion Ozone, or triatomic oxygen, is generated naturally in the upper atmosphere by the action of solar ultraviolet light on oxygen, and may be produced in the lower atmosphere through the photochemical oxidation of pollutants. Background levels are usually of the order of 0.02-0.03 ppm. Ozone may be generated using ultraviolet radiation or electrical discharge, and is usually produced in situ, since its extreme oxidative reactivity precludes storage and transportation. Most studies into the antibacterial effects of ozone have been performed in aqueous solution, where its bactericidal concentration for vegetative bacteria is between O-1-0.2 ppm.4 Activity has also been demonstrated against bacterial spores,’ cryptosporidial oocysts,6 and viruses.’ Less is known about the bactericidal effects of gaseous ozone although there is evidence that it is enhanced by humidity.” It probably acts through oxidation of cell-wall targets such as fatty acids and peptides. Ozone has various industrial uses including wide application in water disinfection, for example in the control of legionellae in hot-water systems.’ Commercially available ozone generators are usually marketed for the ‘improvement of air Manufacturers may assert that this is quality’. achieved through the oxidation of airborne gases or particulates, but there is evidence to the contrary and the main effect may simply be through olfactory camouflage.” Ozone is toxic to man, and has been shown in controlled experiments to cause transient respiratory symptoms, impairment in respiratory function and inflammatory changes at levels below 0.2 ppm. *’ At the levels reached during our experiment in the middle of a hospital side-room, most people would be symptomatically unaffected while a few might experience cough and pain on deep breathing,
and S. J. Pedler
particularly if engaged in heavy exercise. However, concerns remain about the long-term implications of exposure to ozone, particularly its potential for genotoxicity.” The 15 min exposure limit currently recommended is 0.3 ppm. The threshold for smell is 0.1 ppm. The surface of a blood agar plate is a highly nutritious (and atypical) environment for a vegetative organism. Under these circumstances S. uuwus survived equally well in air and in the concentrations of ozone reached at 3 m from the gas generation equipment. After 4 h exposure close to the ozone generator counts of MSSA had been reduced by over SO%, although a further 3 h exposure did not markedly improve the antibacterial effect. MRSA appeared less susceptible, with a significant reduction appearing only at 7 h when counts were still more than 60% of their unexposed counterparts. Filter paper squares proved less amenable to bacterial survival, with significant reductions between 4 and 7 h when both organisms were exposed to room air. Ozone provided an additional bactericidal effect against each organism for both high and low inocula. This was marked close to the machine but in this model still significant at a greater distance, although the reductions are unlikely to be sufficient for room decontamination. The surface of glass squares appeared even more hostile based on the reductions in unexposed counts between 4 and 7 h. This model was the most difficult to use and yielded the least consistent colony counts. Although the numbers are limited they support the previous observations that the effect of the generator tested, if any, was small, confined to organisms exposed close to the equipment, and more apparent with MSSA than MRSA. These results suggest that gaseous ozone at the levels reached at the periphery of a standard hospital side-room using the equipment tested is not sufficient to decontaminate the hospital environment, although it may reduce counts slightly on some surfaces. Activity was greater within 15 cm of the equipment, which raises the possibility of success using a more powerful ozone generator, but the effect was limited in some models and the use of higher ozone concentrations would carry a greater risk of toxicity.
Gaseous
ozone
for
MRSA
65
decontamination
It was interesting that throughout the experiments the methicillin-resistant strain of S. UUY~USappeared less susceptible to the effects of ozone than the methicillin-sensitive strain. This has been observed previously with ozone in solution,13 and presumably reflects differences in the structure of the cell wall. Given the organism’s reputation for environmental persistence we were surprised by its apparently poor survival on filter paper and glass. The use of gaseous ozone in hospital infection control has been investigated before.14 These authors exposed inoculated agar plates to gaseous ozone for 4 h. Bactericidal activity was demonstrated at levels of 0.3-0.9 ppm reached in a cupboard, but their machine was insufficiently powerful to generate detectable ozone in a hospital room. Other investigators have reported success using ozone to decontaminate bioclean rooms,” although the levels achieved were also high enough to corrode rubber. In conclusion, a commercially available ozone generator generated 0.1-0.15 ppm ozone in a standard hospital side-room, levels at which some people might experience respiratory symptoms but which are below the recommended 15 min exposure limits. None was detectable outside the room. Over a period of between 4 and 7 h gaseous ozone showed activity against two strains of S. UUY~US on three different surfaces, but the effect appeared to be concentration dependent and at the periphery of the room was limited at best. The MRSA strain was less susceptible to ozone than the MSSA strain. These results demonstrate that the ozone generator investigated would not be suitable for use in the decontamination of hospital siderooms after discharge of patients colonized with MRSA.
Acknowledgement With thanks to Enright of the equipment.
Industries
for the loan
References 1. Ayliffe GIJ, Coates D, Hoffman Disinfection in Hospitals. London:
PN. Chemical PHLS 1994.
2. Report of a combined working party of the Hospital Infection Society and British Society for Antimicrobial Chemotherapy. Revised guidelines for the control of epidemic methicillin-resistant Staphylococcus aweus. J Hosp Iflfect 1990; 16: 351-377. 3. Company data. 4. Broadwater WT, Hoehn RC, King PH. Sensitivity of three selected bacterial species to ozone. Appl Micvobiol 1973; 26: 391-393. 5. Ishizaki K, Shinriki N, Matsuyama H. Inactivation of Bacillus spores by gaseous ozone. J Appl Bacterial 1986; 60: 67-72. 6. Peeters JE, Mazas EA, Masschelein WJ, Martinez de Maturana IV, Debacker E. Effect of disinfection of drinking water with ozone or chlorine dioxide on survival of Cryptospovidium pavvum oocysts. Appl Environ Microbial 1989; 55: 1519-1522. 7. Vaughn JM, Chen YS, Lindburg K, Morales D. Inactivation of human and simian rotaviruses by ozone. Appl Environ Microbial 1987; 53: 2218-2221. 8. Doroszkiewicz V, Sikorska I, Jankowski S. Studies on the influence of ozone on complementmediated killing of bacteria. FEMS Immunol Med Microbial 1994; 9: 281-286. 9. Hart CA, Makin T. Legionella in hospitals: a review. J Hosp Infect 1991; 18 (Suppl. A): 481489. 10. Boeniger MF. Use of ozone generating devices to improve indoor air quality. Am Ind Hyg Assoc J 1995; 56: 590-598. 11. Krishna MT, Mudway I, Kelly FJ, Frew AJ, Holgate ST. Ozone, airways and allergic airways disease. Clin Exp Allergy 1995; 25: 1150-1158. 12. Victorin K. Review of the genotoxicity of ozone. Mutat Res 1992; 277: 221-238. 13. Yamayoshi T, Tatsumi N. Microbicidal effects of ozone solution on methicillin resistant Staphylococcus aweus. Drugs Exp Clin Res 1993; 19: 59-64. 14. Dyas A, Boughton BJ, Das BC. Ozone killing action against bacterial and fungal species; microbiological testing of a domestic ozone generator. J Clin Path01 1983; 36: 1102-1104. 15. Masaoka T, Kubota Y, Namiuchi S, Takubo T, Ueda T, Shibata H, Nakamura H, Yoshitake J, Yamayoshi T, Doi H, Kamiki T. Ozone decontamination of bioclean rooms. Appl Environ Microbial 1982; 43: 509-513.