- Email: [email protected]
Non-touch fittings in hospitals: a possible source of Pseudomonas aeruginosa and Legionella spp.
Journal of Hospital Infection (2001) 49: 117±121 doi:10.1053/jhin.2001.1060, available online at http://www.idealibrary.com on
Non-touch fittings in hospitals: a possible source of Pseudomonas aeruginosa and Legionella spp. M. Halabi*y, M. Wiesholzer-Pittly, J. SchoÈberly and H. Mittermayerz *Department of Pathology and Microbiology, yInfection Control Team, Krankenhaus der Barmherzigen Schwestern Ried im Innkreis, A-4910 Ried im Innkreis, zDepartment of Hygiene, Microbiology and Tropical Medicine, Krankenhaus der Elisabethinen Linz, Fadingerstrasse 1, A-4020 Linz, Austria Summary: Non-touch fittings are gradually becoming very common in the bathrooms and toilets of public facilities and restaurants. Hospitals and other healthcare facilities have recently started to install these types of water taps to lower water consumption, thus saving costs, and to prevent healthcare workers from touching the tap, thus promoting hygiene. This study analysed the bacteriological water quality of 38 non-touch water taps in different settings in a 450-bed secondary-care hospital in Upper Austria. Two different tap types were installed: 23 taps were without temperature selection and 15 were with temperature selection (cold and warm). A membrane filtration method was used, and the authors screened for both indicator organisms and Pseudomonas aeruginosa in 100 ml water samples. In 10 non-touch taps without temperature selection, the authors also screened for Legionella spp. in 500 ml water samples. Seventy four percent of the taps without temperature selection and 7% of the taps with temperature selection showed contamination with P. aeruginosa (P < 0.001). None of the taps showed contamination with indicator organisms. Detailed analysis of the source of contamination revealed that the magnetic valve and the outlet itself were heavily contaminated, whereas the junction from the central pipe system was free of contamination. All 10 analysed taps showed contamination with Legionella spp. It was concluded that the local contamination of non-touch fittings is a result of the low amount of water that flows through the outlet, the low water pressure and the column of water, which is `still-standing' and has a temperature of about 35 C, thus providing nearly ideal growth conditions for P. aeruginosa. Additionally, the presence of materials such as rubber, PVC, etc. in the fittings enhances the adhesion of P. aeruginosa and thus the production of biofilms. In conclusion, the authors wish to encourage infection control teams to evaluate the use of non-touch fittings in hospitals, especially when they are installed in risk areas. & 2001 The Hospital Infection Society
Keywords: Non-touch fittings; hospital; nosocomial infection; Pseudomonas aeruginosa; Legionella spp.
Introduction Tap water as a source of nosocomial infections has been discussed extensively in the literature and is still Received 28 February 2001; accepted 24 July 2001. Author for correspondence: Dr M. Halabi, Department of Pathology, Infection Control Team, Krankenhaus Ried im Innkreis, Schlossberg 1, A-4910 Ried im Innkreis, Austria. Fax: 0043-7752-602-6520; E-mail: [email protected]
0195-6701/01/020117 + 05 $35.00/0
a major issue in hospital epidemiology.1±13 New technologies in the manufacturing of taps and fittings have opened up a new discussion concerning tap water as a possible source of infections, especially when non-touch water taps are used in hospitals and other healthcare facilities. Non-touch water taps are used in public facilities and restaurants mainly to prevent customers from touching `contaminated' taps and thus acquiring an & 2001 The Hospital Infection Society
118
M. Halabi et al.
infection. These taps work only when hands are put in front of an infra-red sensor, which causes water to flow out of the tap. When hands are removed, water flow stops. The major arguments of the manufacturers are the decrease of water consumption, which saves costs, and elimination of the need to touch the tap. The latter seems to be an obvious advantage, especially in hospitals, where hygienic measures are very important. The authors' secondary-care hospital in Upper Austria (450 beds) consumes 90 000 m3 water/year. According to data from the manufacturers, water consumption can be reduced by 50% if non-touch fittings are used, a measure which could save the hospital 100 000 D /year. In this hospital, 38 conventional water taps were replaced by 38 non-touch water taps in different settings, such as the operating room, wards, outpatient units and toilets. Two different types of taps were installed, one type with a fixed temperature, the other with temperature selection (cold and warm). After a couple of months of usage, the hospital infection control team evaluated the bacteriological water quality of all 38 taps. This paper presents the results of the analysis and the authors' recommendations. Materials and methods Table I shows the location of the 38 non-touch taps. After an average three-month period of use, 500 ml water samples from each of the 38 taps were taken. The samples were processed according to regulations of the Austrian Health Authorities `Verordnung uÈber die QualitaÈt von Wasser fuÈr den menschlichen Gebrauch' [Regulation concerning water intended for human use] Nr. 235, July 1998, based upon the guideline 98/93 of the European Union. To ensure authentic water samples in the wards and the operating room, the aerator was not removed and the water was not allowed to flow for 10 min, as is
usually done. Furthermore, the outlet was not flamed, and therefore not disinfected. One hundred millilitres of each water sample was processed by membrane filtration. The filters (Sartorius 0.45 mm) were placed on Cetrimide-Agar (Biotest) and EndoAgar (Becton Dickinson). Endo-Agar was used to screen for indicator organisms such as Escherichia coli. Cetrimide-Agar was used for the selective cultivation of Pseudomonas aeruginosa. According to the quoted regulation, P. aeruginosa should be absent in a 100 ml water sample. The incubation period was 48 h at 37 C. Total cell count was also performed by using the spread plate method, in which 1 ml of each water sample was inoculated on plate-count agar. Colony forming units (CFU) were counted after 48 and 72 h at incubation temperatures of 22 and 37 C, respectively. Colonies which grew on Cetrimide-Agar were identified biochemically as P. aeruginosa using the API-20 NE (BioMeÂrieux) identification system. After testing the 38 non-touch taps, 10 of the taps without temperature selection were compared with 10 conventional water taps, used as controls, that were situated adjacent to the non-touch taps and had the same water source from the central pipe system. Parts of the technical equipment of the 10 nontouch taps (Figure 1) were subsequently analysed, taking 500 ml samples from the outlet, the magnetic valve and both junctions of the central pipe system. These samples were processed using the membrane filtration method as described above. Finally, water samples from the 10 non-touch taps without temperature selection and the 10 adjacent conventional taps were analysed for the presence of Legionella spp., using a membrane filtration method without any treatment. After spreading 1 ml of native water Outlet
Table I Location of the non-touch fittings in the hospital Number 16 6 5 5 4 1 1
Location Toilets and sanitary facilities Hospital kitchen Intensive care unit Operating room (wardrobe and toilet) Wards Wardrobe Outpatients' area
Magnetic valve Pipe junction warm
Pipe junction cold Figure 1 Diagram of the non-touch fitting.
Non-touch fittings in hospitals
sample on Legionella Agar GVPC (Biotest), 500 ml of the water sample was filtered through a 0.2 mm membrane with 50 mm diameter (Cellulose nitrite, Sartorius). The filter was resuspended in 20 ml sterile distilled water using an ultrasound bath. After this procedure, 1 ml and 0.1 ml of the 20 ml sample was spread on Legionella Agar GVPC. The agar plates were incubated at 37 C with 5% CO2 for 14 days. GVPC agar selectively suppresses normal aquatic flora and therefore is an excellent media for the detection of Legionellae from water. Colony forming unit counts were read every day up to 14 days; candidate colonies were isolated on TSS Agar (BioMeÂrieux) and Legionella Agar GVPC for confirmation of the presence of Legionella spp. The temperature of the water from both the nontouch and the conventional taps was measured immediately after opening the tap. The temperature of the water from the non-touch taps ranged from 20.4 C to 28.8 C (median 27.9 C); the temperature range of the water from the conventional taps ranged from 18.7 C to 27.4 C (median 24.5 C). The central plumbing system of the hospital is constructed mainly of galvanised water pipes. The temperature of the circulating water leaving the main boiler is 65 C; its temperature before re-entering the boiler after circulation varies between 50 and 55 C. The cold water temperature is between 10 and 12 C at the take-over junction from the public water supplier. Within the hospital, there are no special precautions to prevent the growth of Legionellae. To test differences in proportions, the Fisher's exact test was used.
Results Comparing the two types of non-touch taps with and without temperature selection, 17 (74%) of the taps without temperature selection and one (7%) of the taps with temperature selection were found to be contaminated with P. aeruginosa. The difference between both types was highly significant (P < 0.001, Fisher's exact test). Comparison of the 10 non-touch taps without temperature selection with the 10 conventional taps adjacent to the non-touch taps showed that P. aeruginosa could be detected in none of the conventional taps, whereas all the non-touch taps showed contamination with P. aeruginosa. The total cell counts in some fittings were up to 100-fold higher in the non-touch taps than in conventional taps (Table II).
119 Table II Total colony forming units (CFUs) at 22 vs. 37 C in samples from non-touch fittings vs. conventional taps Outlet no.
Total CFU non-touch taps in 1 ml after 72 h 37 C 22 C
1 2 3 4 5 6 7 8 9 10
10 150 1500 15 100 200 34 No CFU 350 34
26 38 > 5000 300 2500 1500 16 2 46 7
Total CFU conventional taps in 1 ml after 72 h 22 C 37 C No CFU No CFU 5 No CFU No CFU 45 6 No CFU 10 70
5 No CFU 5 94 250 120 No CFU No CFU 54 200
Table III Evidence of P. aeruginosa in 100 ml water samples from 10 non-touch fittings at 37 C after 48 h Outlet no.
1 2 3 4 5 6 7 8 9 10
Sample 1: Outlet CFU/100 ml
Sample 2: Magnetic valve CFU/100 ml
2 58 12 180 13 5 220 3 8 2
9 200 13 100 200 7 300 34 200 6
Sample 3: Junction cold water CFU/100 ml No No No No No No No No No No
CFU CFU CFU CFU CFU CFU CFU CFU CFU CFU
Sample 4: Junction warm water CFU/100 ml No No No No No No No No No No
CFU CFU CFU CFU CFU CFU CFU CFU CFU CFU
The results of the water samples taken from the outlet, the magnetic valve and both junctions showed that none of the junction samples were contaminated with P. aeruginosa, but that the magnetic valve and the outlet were highly contaminated with P. aeruginosa (Table III). Legionella spp. were found in all of the non-touch water taps but only in three of the conventional taps. The CFU counts are shown in Table IV. Growth of P. aeruginosa was seen on some of the GVPC agar plates, whereas growth of other bacteria was suppressed. Discussion Non-touch fittings are gradually becoming very common in the bathrooms and toilets of public facilities. Hospitals and other healthcare facilities
120
M. Halabi et al.
Table IV Evidence of Legionella spp. in 1 ml water sample incubated at 37 C for 14 days Outlet no. 1 2 3 4 5 6 7 8 9 10
Non-touch fitting CFU/ml
Conventional tap CFU/ml
19 90 1 > 500 8 25 23 170 3 96
No CFU No CFU No CFU 2 No CFU No CFU No CFU No CFU 5 1
have recently started to install these types of water taps to lower the water consumption, thus saving costs, and also prevent healthcare workers from touching the tap, thus promoting hygiene. The results of the bacteriological water analysis of 38 non-touch water taps (with and without temperature selection) in the authors' hospital revealed that approximately 50% of the non-touch taps were heavily contaminated with P. aeruginosa. In comparison, the taps with temperature selection (cold and warm) were less contaminated than the taps without temperature selection (P < 0.001). Results of the analyses of the conventional taps adjacent to the non-touch taps indicated that the central pipe system was not the source of contamination, upon which various parts of the non-touch taps were analysed to look for local sources of contamination in the non-touch taps themselves. The magnetic valve, the mixing device and the outlet were found to be the most contaminated parts of the tap. The magnetic valve is the technical `heart' of the non-touch tap system. Cold and hot water from the junctions of the central pipe system is mixed to provide an acceptable temperature. After being mixed, the water stands still in a column up to the outlet at a temperature of 35 C, until released by the next person who puts his hands in front of the infra-red sensor. After rinsing and subsequent removal of the hands from the infra-red sensor, water flow ceases. The total water flow is very low, as is the water pressure. Both factors may benefit bacterial growth. The magnetic valve itself is constructed of material very likely to encourage bacterial growth, e.g. rubber, plastic and PVC membranes. P. aeruginosa biofilms may develop, which are very hard to remove with biocides and disinfectants.
The evidence of growth of Legionella spp. in the non-touch fittings, compared with the very low CFU counts of the conventional taps, indicates that besides the problem of contamination with P. aeruginosa, non-touch water taps seem to be a growth promoter of Legionellae, which is an emerging issue in water hygiene.9±13 To the authors' knowledge, there were no clinical cases of Legionellosis or infections with P. aeruginosa during the time the non-touch taps were in use. The results of this investigation prompted the hospital authorities to remove all of the non-touch taps and replace them with conventional taps. In conclusion, the authors wish to encourage infection control teams to evaluate the use of non-touch fittings in hospitals, especially when they are installed in risk areas. It is necessary to question technical issues, e.g. the manufacturing materials, as well as the construction itself, and to be aware of this possible source of nosocomial infections with micro-organisms such as P. aeruginosa and Legionella spp.
Acknowledgements The authors would like to thank Diane Thompson for proof-reading the paper, and Mr Gruber for technical assistance with the plumbing.
References 1. Cross AS. Evolving epidemiology of Pseudomonas aeruginosa infections. Eur J Clin Microbiol 1985; 2: 156± 159. 2. Rusin PA, Rose JB, Haas CN, Gerba CP. Risk assessment of opportunistic bacterial pathogens in drinking water. Rev Environ Contam Toxicol 1997; 152: 59±61. 3. Ferroni A, Nguyen L, Pron B, Quesne G, Brusset MC, Berche P. Outbreak of nosocomial urinary tract infection due to Pseudomonas aeruginosa in a paediatric surgical unit associated with tap-water contamination. J Hosp Inf 1998; 39: 301±307. 4. Bert E, Maubec E, Bruneau B, Berry P, LambertZechovsky. Multi-resistant Pseudomonas aeruginosa outbreak associated with contaminated tap water in a neurosurgery intensive care unit. J Hosp Inf 1998; 39: 53±62. 5. Schulze-RoÈbbecke R, Pleischl S. Verminderung von Infektionsrisiken durch die mikrobielle Besiedlung von Krankenhaus-Leitungswassersystemen. Krh Hyg u Inf verh 1997; 19: 150±155. 6. Grobe S, Wingender J, Truper HG. Characterization of mucoid Pseudomonas aeruginosa strains isolated from technical water systems. J Appl Bacteriol 1995; 79: 94±102.
Non-touch fittings in hospitals 7. Barbeau J, Gauthier C, Payment P. Biofilms, infectious agents, and dental unit waterlines: a review. Can J Microbiol 1998; 44: 1019±1028. 8. Bollin GE, Plouffe JF, Para MF, Hackman B. Aerosols containing Legionella pneumophila generated by shower heads and hot-water faucets. Appl Environ Microbiol 1985; 50: 1128±1131. 9. Luck PC, Leupold I, Hlawitschka M et al. Prevalence of Legionella species, serogroups, and monoclonal subgroups in hot water systems in southeastern Germany. Zentralbl Hyg Umweltmed 1993; 193: 450±460. 10. Ribiero CD, Burge SH, Palmer SR, Tobin JO, Watkins ID. Legionella pneumophila in a hospital water system following a nosocomial outbreak: prevalence, monoclonal antibody subgrouping and effect of control measures. Epidemiol Infect 1987; 3: 253±262.
121 11. Kool JL, Bergmire-Sweat D, Butler JC et al. Hospital characteristics associated with colonization of water systems by Legionella and risk of nosocomial legionnaires' disease: a cohort study of 15 hospitals. Infect Control Hosp Epidemiol 1999; 20: 798±805. 12. Visca P, Goldoni P, Luck PC et al. Multiple types of Legionella pneumophila serogroup 6 in a hospital heated-water system associated with sporadic infections. J Clin Microbiol 1999; 37: 2189±2196. 13. Goetz AM, Stout JE, Jacobs SL et al. Nosocomial legionnaires' disease discovered in community hospitals following cultures of the water system: seek and ye shall find. Am J Infect Control 1998; 26: 8±11.
Non-touch fittings in hospitals: a possible source of Pseudomonas aeruginosa and Legionella spp. M. Halabi*y, M. Wiesholzer-Pittly, J. SchoÈberly and H. Mittermayerz *Department of Pathology and Microbiology, yInfection Control Team, Krankenhaus der Barmherzigen Schwestern Ried im Innkreis, A-4910 Ried im Innkreis, zDepartment of Hygiene, Microbiology and Tropical Medicine, Krankenhaus der Elisabethinen Linz, Fadingerstrasse 1, A-4020 Linz, Austria Summary: Non-touch fittings are gradually becoming very common in the bathrooms and toilets of public facilities and restaurants. Hospitals and other healthcare facilities have recently started to install these types of water taps to lower water consumption, thus saving costs, and to prevent healthcare workers from touching the tap, thus promoting hygiene. This study analysed the bacteriological water quality of 38 non-touch water taps in different settings in a 450-bed secondary-care hospital in Upper Austria. Two different tap types were installed: 23 taps were without temperature selection and 15 were with temperature selection (cold and warm). A membrane filtration method was used, and the authors screened for both indicator organisms and Pseudomonas aeruginosa in 100 ml water samples. In 10 non-touch taps without temperature selection, the authors also screened for Legionella spp. in 500 ml water samples. Seventy four percent of the taps without temperature selection and 7% of the taps with temperature selection showed contamination with P. aeruginosa (P < 0.001). None of the taps showed contamination with indicator organisms. Detailed analysis of the source of contamination revealed that the magnetic valve and the outlet itself were heavily contaminated, whereas the junction from the central pipe system was free of contamination. All 10 analysed taps showed contamination with Legionella spp. It was concluded that the local contamination of non-touch fittings is a result of the low amount of water that flows through the outlet, the low water pressure and the column of water, which is `still-standing' and has a temperature of about 35 C, thus providing nearly ideal growth conditions for P. aeruginosa. Additionally, the presence of materials such as rubber, PVC, etc. in the fittings enhances the adhesion of P. aeruginosa and thus the production of biofilms. In conclusion, the authors wish to encourage infection control teams to evaluate the use of non-touch fittings in hospitals, especially when they are installed in risk areas. & 2001 The Hospital Infection Society
Keywords: Non-touch fittings; hospital; nosocomial infection; Pseudomonas aeruginosa; Legionella spp.
Introduction Tap water as a source of nosocomial infections has been discussed extensively in the literature and is still Received 28 February 2001; accepted 24 July 2001. Author for correspondence: Dr M. Halabi, Department of Pathology, Infection Control Team, Krankenhaus Ried im Innkreis, Schlossberg 1, A-4910 Ried im Innkreis, Austria. Fax: 0043-7752-602-6520; E-mail: [email protected]
0195-6701/01/020117 + 05 $35.00/0
a major issue in hospital epidemiology.1±13 New technologies in the manufacturing of taps and fittings have opened up a new discussion concerning tap water as a possible source of infections, especially when non-touch water taps are used in hospitals and other healthcare facilities. Non-touch water taps are used in public facilities and restaurants mainly to prevent customers from touching `contaminated' taps and thus acquiring an & 2001 The Hospital Infection Society
118
M. Halabi et al.
infection. These taps work only when hands are put in front of an infra-red sensor, which causes water to flow out of the tap. When hands are removed, water flow stops. The major arguments of the manufacturers are the decrease of water consumption, which saves costs, and elimination of the need to touch the tap. The latter seems to be an obvious advantage, especially in hospitals, where hygienic measures are very important. The authors' secondary-care hospital in Upper Austria (450 beds) consumes 90 000 m3 water/year. According to data from the manufacturers, water consumption can be reduced by 50% if non-touch fittings are used, a measure which could save the hospital 100 000 D /year. In this hospital, 38 conventional water taps were replaced by 38 non-touch water taps in different settings, such as the operating room, wards, outpatient units and toilets. Two different types of taps were installed, one type with a fixed temperature, the other with temperature selection (cold and warm). After a couple of months of usage, the hospital infection control team evaluated the bacteriological water quality of all 38 taps. This paper presents the results of the analysis and the authors' recommendations. Materials and methods Table I shows the location of the 38 non-touch taps. After an average three-month period of use, 500 ml water samples from each of the 38 taps were taken. The samples were processed according to regulations of the Austrian Health Authorities `Verordnung uÈber die QualitaÈt von Wasser fuÈr den menschlichen Gebrauch' [Regulation concerning water intended for human use] Nr. 235, July 1998, based upon the guideline 98/93 of the European Union. To ensure authentic water samples in the wards and the operating room, the aerator was not removed and the water was not allowed to flow for 10 min, as is
usually done. Furthermore, the outlet was not flamed, and therefore not disinfected. One hundred millilitres of each water sample was processed by membrane filtration. The filters (Sartorius 0.45 mm) were placed on Cetrimide-Agar (Biotest) and EndoAgar (Becton Dickinson). Endo-Agar was used to screen for indicator organisms such as Escherichia coli. Cetrimide-Agar was used for the selective cultivation of Pseudomonas aeruginosa. According to the quoted regulation, P. aeruginosa should be absent in a 100 ml water sample. The incubation period was 48 h at 37 C. Total cell count was also performed by using the spread plate method, in which 1 ml of each water sample was inoculated on plate-count agar. Colony forming units (CFU) were counted after 48 and 72 h at incubation temperatures of 22 and 37 C, respectively. Colonies which grew on Cetrimide-Agar were identified biochemically as P. aeruginosa using the API-20 NE (BioMeÂrieux) identification system. After testing the 38 non-touch taps, 10 of the taps without temperature selection were compared with 10 conventional water taps, used as controls, that were situated adjacent to the non-touch taps and had the same water source from the central pipe system. Parts of the technical equipment of the 10 nontouch taps (Figure 1) were subsequently analysed, taking 500 ml samples from the outlet, the magnetic valve and both junctions of the central pipe system. These samples were processed using the membrane filtration method as described above. Finally, water samples from the 10 non-touch taps without temperature selection and the 10 adjacent conventional taps were analysed for the presence of Legionella spp., using a membrane filtration method without any treatment. After spreading 1 ml of native water Outlet
Table I Location of the non-touch fittings in the hospital Number 16 6 5 5 4 1 1
Location Toilets and sanitary facilities Hospital kitchen Intensive care unit Operating room (wardrobe and toilet) Wards Wardrobe Outpatients' area
Magnetic valve Pipe junction warm
Pipe junction cold Figure 1 Diagram of the non-touch fitting.
Non-touch fittings in hospitals
sample on Legionella Agar GVPC (Biotest), 500 ml of the water sample was filtered through a 0.2 mm membrane with 50 mm diameter (Cellulose nitrite, Sartorius). The filter was resuspended in 20 ml sterile distilled water using an ultrasound bath. After this procedure, 1 ml and 0.1 ml of the 20 ml sample was spread on Legionella Agar GVPC. The agar plates were incubated at 37 C with 5% CO2 for 14 days. GVPC agar selectively suppresses normal aquatic flora and therefore is an excellent media for the detection of Legionellae from water. Colony forming unit counts were read every day up to 14 days; candidate colonies were isolated on TSS Agar (BioMeÂrieux) and Legionella Agar GVPC for confirmation of the presence of Legionella spp. The temperature of the water from both the nontouch and the conventional taps was measured immediately after opening the tap. The temperature of the water from the non-touch taps ranged from 20.4 C to 28.8 C (median 27.9 C); the temperature range of the water from the conventional taps ranged from 18.7 C to 27.4 C (median 24.5 C). The central plumbing system of the hospital is constructed mainly of galvanised water pipes. The temperature of the circulating water leaving the main boiler is 65 C; its temperature before re-entering the boiler after circulation varies between 50 and 55 C. The cold water temperature is between 10 and 12 C at the take-over junction from the public water supplier. Within the hospital, there are no special precautions to prevent the growth of Legionellae. To test differences in proportions, the Fisher's exact test was used.
Results Comparing the two types of non-touch taps with and without temperature selection, 17 (74%) of the taps without temperature selection and one (7%) of the taps with temperature selection were found to be contaminated with P. aeruginosa. The difference between both types was highly significant (P < 0.001, Fisher's exact test). Comparison of the 10 non-touch taps without temperature selection with the 10 conventional taps adjacent to the non-touch taps showed that P. aeruginosa could be detected in none of the conventional taps, whereas all the non-touch taps showed contamination with P. aeruginosa. The total cell counts in some fittings were up to 100-fold higher in the non-touch taps than in conventional taps (Table II).
119 Table II Total colony forming units (CFUs) at 22 vs. 37 C in samples from non-touch fittings vs. conventional taps Outlet no.
Total CFU non-touch taps in 1 ml after 72 h 37 C 22 C
1 2 3 4 5 6 7 8 9 10
10 150 1500 15 100 200 34 No CFU 350 34
26 38 > 5000 300 2500 1500 16 2 46 7
Total CFU conventional taps in 1 ml after 72 h 22 C 37 C No CFU No CFU 5 No CFU No CFU 45 6 No CFU 10 70
5 No CFU 5 94 250 120 No CFU No CFU 54 200
Table III Evidence of P. aeruginosa in 100 ml water samples from 10 non-touch fittings at 37 C after 48 h Outlet no.
1 2 3 4 5 6 7 8 9 10
Sample 1: Outlet CFU/100 ml
Sample 2: Magnetic valve CFU/100 ml
2 58 12 180 13 5 220 3 8 2
9 200 13 100 200 7 300 34 200 6
Sample 3: Junction cold water CFU/100 ml No No No No No No No No No No
CFU CFU CFU CFU CFU CFU CFU CFU CFU CFU
Sample 4: Junction warm water CFU/100 ml No No No No No No No No No No
CFU CFU CFU CFU CFU CFU CFU CFU CFU CFU
The results of the water samples taken from the outlet, the magnetic valve and both junctions showed that none of the junction samples were contaminated with P. aeruginosa, but that the magnetic valve and the outlet were highly contaminated with P. aeruginosa (Table III). Legionella spp. were found in all of the non-touch water taps but only in three of the conventional taps. The CFU counts are shown in Table IV. Growth of P. aeruginosa was seen on some of the GVPC agar plates, whereas growth of other bacteria was suppressed. Discussion Non-touch fittings are gradually becoming very common in the bathrooms and toilets of public facilities. Hospitals and other healthcare facilities
120
M. Halabi et al.
Table IV Evidence of Legionella spp. in 1 ml water sample incubated at 37 C for 14 days Outlet no. 1 2 3 4 5 6 7 8 9 10
Non-touch fitting CFU/ml
Conventional tap CFU/ml
19 90 1 > 500 8 25 23 170 3 96
No CFU No CFU No CFU 2 No CFU No CFU No CFU No CFU 5 1
have recently started to install these types of water taps to lower the water consumption, thus saving costs, and also prevent healthcare workers from touching the tap, thus promoting hygiene. The results of the bacteriological water analysis of 38 non-touch water taps (with and without temperature selection) in the authors' hospital revealed that approximately 50% of the non-touch taps were heavily contaminated with P. aeruginosa. In comparison, the taps with temperature selection (cold and warm) were less contaminated than the taps without temperature selection (P < 0.001). Results of the analyses of the conventional taps adjacent to the non-touch taps indicated that the central pipe system was not the source of contamination, upon which various parts of the non-touch taps were analysed to look for local sources of contamination in the non-touch taps themselves. The magnetic valve, the mixing device and the outlet were found to be the most contaminated parts of the tap. The magnetic valve is the technical `heart' of the non-touch tap system. Cold and hot water from the junctions of the central pipe system is mixed to provide an acceptable temperature. After being mixed, the water stands still in a column up to the outlet at a temperature of 35 C, until released by the next person who puts his hands in front of the infra-red sensor. After rinsing and subsequent removal of the hands from the infra-red sensor, water flow ceases. The total water flow is very low, as is the water pressure. Both factors may benefit bacterial growth. The magnetic valve itself is constructed of material very likely to encourage bacterial growth, e.g. rubber, plastic and PVC membranes. P. aeruginosa biofilms may develop, which are very hard to remove with biocides and disinfectants.
The evidence of growth of Legionella spp. in the non-touch fittings, compared with the very low CFU counts of the conventional taps, indicates that besides the problem of contamination with P. aeruginosa, non-touch water taps seem to be a growth promoter of Legionellae, which is an emerging issue in water hygiene.9±13 To the authors' knowledge, there were no clinical cases of Legionellosis or infections with P. aeruginosa during the time the non-touch taps were in use. The results of this investigation prompted the hospital authorities to remove all of the non-touch taps and replace them with conventional taps. In conclusion, the authors wish to encourage infection control teams to evaluate the use of non-touch fittings in hospitals, especially when they are installed in risk areas. It is necessary to question technical issues, e.g. the manufacturing materials, as well as the construction itself, and to be aware of this possible source of nosocomial infections with micro-organisms such as P. aeruginosa and Legionella spp.
Acknowledgements The authors would like to thank Diane Thompson for proof-reading the paper, and Mr Gruber for technical assistance with the plumbing.
References 1. Cross AS. Evolving epidemiology of Pseudomonas aeruginosa infections. Eur J Clin Microbiol 1985; 2: 156± 159. 2. Rusin PA, Rose JB, Haas CN, Gerba CP. Risk assessment of opportunistic bacterial pathogens in drinking water. Rev Environ Contam Toxicol 1997; 152: 59±61. 3. Ferroni A, Nguyen L, Pron B, Quesne G, Brusset MC, Berche P. Outbreak of nosocomial urinary tract infection due to Pseudomonas aeruginosa in a paediatric surgical unit associated with tap-water contamination. J Hosp Inf 1998; 39: 301±307. 4. Bert E, Maubec E, Bruneau B, Berry P, LambertZechovsky. Multi-resistant Pseudomonas aeruginosa outbreak associated with contaminated tap water in a neurosurgery intensive care unit. J Hosp Inf 1998; 39: 53±62. 5. Schulze-RoÈbbecke R, Pleischl S. Verminderung von Infektionsrisiken durch die mikrobielle Besiedlung von Krankenhaus-Leitungswassersystemen. Krh Hyg u Inf verh 1997; 19: 150±155. 6. Grobe S, Wingender J, Truper HG. Characterization of mucoid Pseudomonas aeruginosa strains isolated from technical water systems. J Appl Bacteriol 1995; 79: 94±102.
Non-touch fittings in hospitals 7. Barbeau J, Gauthier C, Payment P. Biofilms, infectious agents, and dental unit waterlines: a review. Can J Microbiol 1998; 44: 1019±1028. 8. Bollin GE, Plouffe JF, Para MF, Hackman B. Aerosols containing Legionella pneumophila generated by shower heads and hot-water faucets. Appl Environ Microbiol 1985; 50: 1128±1131. 9. Luck PC, Leupold I, Hlawitschka M et al. Prevalence of Legionella species, serogroups, and monoclonal subgroups in hot water systems in southeastern Germany. Zentralbl Hyg Umweltmed 1993; 193: 450±460. 10. Ribiero CD, Burge SH, Palmer SR, Tobin JO, Watkins ID. Legionella pneumophila in a hospital water system following a nosocomial outbreak: prevalence, monoclonal antibody subgrouping and effect of control measures. Epidemiol Infect 1987; 3: 253±262.
121 11. Kool JL, Bergmire-Sweat D, Butler JC et al. Hospital characteristics associated with colonization of water systems by Legionella and risk of nosocomial legionnaires' disease: a cohort study of 15 hospitals. Infect Control Hosp Epidemiol 1999; 20: 798±805. 12. Visca P, Goldoni P, Luck PC et al. Multiple types of Legionella pneumophila serogroup 6 in a hospital heated-water system associated with sporadic infections. J Clin Microbiol 1999; 37: 2189±2196. 13. Goetz AM, Stout JE, Jacobs SL et al. Nosocomial legionnaires' disease discovered in community hospitals following cultures of the water system: seek and ye shall find. Am J Infect Control 1998; 26: 8±11.