Legionnaires disease, which is a severe form of. Colonization of Legionella Species in Hotel Water Systems in Turkey

Size: px

Start display at page:

Download "Legionnaires disease, which is a severe form of. Colonization of Legionella Species in Hotel Water Systems in Turkey"

Tamsyn Holt
5 years ago
Views:

1 369 Colonization of Legionella Species in Hotel Water Systems in Turkey Haluk Erdogan, MD * and Hande Arslan, MD * Department of Infectious Diseases and Clinical Microbiology, Baskent University Alanya Hospital, Antalya, Turkey ; Department of Infectious Diseases and Clinical Microbiology, Baskent University Faculty of Medicine, Ankara, Turkey DOI: /j x Background. The goal of this study was to evaluate the prevalence of Legionella species in hotel water distribution systems in Alanya, Turkey, which is an important tourism center. Methods. Water and swab samples were obtained from 52 Turkish hotels from August 2003 to September Water samples were collected in 100 ml sterile containers and were concentrated by membrane filters with a pore size of 0.45 m. Heat treatment was used to eliminate other microorganisms from the samples, which were then spread on buffered charcoal yeast extract agar plates and glycine, vancomycin, polymyxin, cycloheximide agar plates. Cysteinedependent colonies were identified by latex agglutination. Results. In all, 491 water and swab samples were analyzed. The results of all samples were negative for Legionella in 16 (30.8%) hotels. Legionella species (92.5% of which were Legionella pneumophila ) were detected in 93 (18.9%) of the samples. The most frequently isolated species were L pneumophila serogroups 6 (63.5%) and 1 (21.5%). Conclusions. Legionella pneumophila serogroup 6 was the most common isolate detected in Turkish hotel water systems in our study. The result of Legionella urinary antigen tests, which are the diagnostic tests most often used to identify legionnaires disease, may be negative in people infected with L pneumophila serogroup 6. We suggest that clinicians should apply the whole spectrum of laboratory methods for the detection of legionnaires disease in patients with pneumonia of unknown origin and history of travel to Alanya, Turkey. Legionnaires disease, which is a severe form of pneumonia caused by Legionella species, was first identified after a large outbreak in an American hotel in The infection is believed to be acquired by inhalation or aspirations from environmental sources and carries a mortality rate of 10% to 30%. 1 Legionella species are ubiquitous in both natural and man-made environments and are frequently found in domestic water systems. The preliminary findings of this report were presented at the 20th annual meeting of European Working Group for Legionella Infections, May 16 to 17, 2005, in Rome, Italy, as a poster. Corresponding Author: Haluk Erdogan, MD, Department of Infectious Diseases and Clinical Microbiology, Baskent University Alanya Hospital, Saray m, Kizlarpinari cd, No: 1, Alanya, Antalya, Turkey. erdoganhaluk@hotmail.com Clusters and sporadic legionnaires disease occur each year associated especially with hotels and hotels water systems contaminated by legionellae. European countries have had special disease surveillance programs protecting citizens against travel-associated legionnaires disease since According to reports by the European Working Group for Legionella Infection (EWGLI), Spain, Italy, France, and Turkey are the countries in which travel-associated legionnaires disease is most often diagnosed. 3,4 In Turkey, legionnaires disease was included in the national disease surveillance programme in A few studies in the literature have reported colonization by Legionella bacteria in hotel water systems in Turkey. 6,7 The goal of this study was to evaluate the frequency of colonization by Legionella species in hotel water systems in Alanya, Turkey, a town that accounted for 12.6% of the total number of tourists in Turkey in International Society of Travel Medicine, Journal of Travel Medicine, Volume 14, Issue 6, 2007,

2 370 Materials and Methods After separating Alanya into geographic regions, we randomly asked hotels managers to participate in our study. Fifty-four hotels from different regions in Alanya were invited to this study, but two hotel managers refused to cooperate so 52 hotels were included in the study. Only three hotels were related to patients having legionnaires disease. During each summer season from August 2003 to September 2005, water samples were taken from hot water and cold-water tanks, air-conditioners, plumbing fixtures, and pipes via the procedure described in the Turkish travel-associated legionnaires disease guidelines. 5 The number of collected samples was determined by the size of each hotel. Immediately after the drain valve at the base of each heater or tank was opened, the first water sample was collected in 100 ml sterile containers. The second sample was collected after the water had drained for 15 to 30 seconds. The samples from showerheads and faucets were obtained in rooms that were located furthest from the hot water source. The rooms were selected to achieve an even distribution among all hot water distribution systems. The swab samples were taken with a sterile Dacron swab that was rotated approximately four times around the inner circumference of the faucet and showerhead. After the swab had been placed in a tube, a small amount of water was added from the collection site to the container to keep the swab moist during transport. Hot water samples from faucets were also collected in 100 ml sterile containers after a brief flow time to eliminate cold water inside the pipe. To neutralize residual free chlorine, sodium thiosulfate was added in sterile containers. Cellulose nitrate membrane filters with a pore size of 0.45 m (Sartorius AG, Goettingen, Germany) were used to concentrate each water sample and the membranes were then placed aseptically in screwcapped containers with 10 ml of the distilled water. These samples were vortexed to dislodge bacterial cells from the membranes. Heat treatment (incubation for 30 min in a water bath at 50 C) was used to reduce the number of bacteria that were not Legionella in the concentrated samples. Plates containing buffered charcoal yeast extract (BCYE- ) agar [charcoal yeast extract base agar, with BCYE growth supplement (ACES buffer/potassium hydroxide, l -cysteine, ferric pyrophosphate, alphaketoglutarate) (Oxoid, Hampshire, England), and glycine, vancomycin, polymyxin B, cycloheximide (GVPC) agar (BCYE- agar with added GVPC supplement) (Oxoid)] were inoculated with a Erdogan and Arslan 0.1 ml sample and were incubated at 35 C with 2.5% CO 2 in a humid atmosphere. Colonies suspected of being Legionella were subcultured on tryptone soya agar containing 5% sheep blood and BCYE- agar. Isolates that grew on BCYE- agar but failed to grow on 5% sheep blood agar with characteristic morphologic features were presumed to be Legionella. A commercially available Legionella latex test (DR0800M, Oxoid) was used for final confirmation as Legionella pneumophila serogroup 1, L pneumophila serogroups 2 to 14, and nonpneumophila Legionella species. Legionella pneumophila serogroup subclassifications 2, 3, 4, 5, and 6 were identified via reagents supplied in latex test kits (Denka Seiken Co., Ltd, Tokyo, Japan). Results A total of 491 samples (including 145 swab samples and 346 water samples) were collected from 52 Turkish hotels. The mean number of samples was 9.5 ± 2.6 (range 4 16 samples). The results of all samples were negative for Legionella only in 16 (30.8%) hotels. Legionella species were detected in 93 samples, 29 (20%) of which were from swabs and 64 (18.5%) of which were from water samples. Of a total 93 Legionella isolates, L pneumophila accounted for 86 (92.5%) and nonpneumophila Legionella species accounted for 7 (7.5%). The most common species identified were L pneumophila serogroups 6 (63.5%) and 1 (21.5%) ( Table 1 ). The hotels positive for L pneumophila serogroup 6 were distributed through all of the different geographical areas within the town. The colony count was lower than 100 colony-forming unit (CFU)/100 ml in 19 (29.7%) samples, between 100 and 1,000 CFU/100 ml in 26 (40.7%) samples, and higher than 1,000 CFU/100 ml in 19 (29.7%) samples ( Table 2 ). We found that in 16 (30.8%) hotels, the colony count exceeded 100 CFU/100 ml in at least one sample. Table 1 Distribution of Legionella species and serogroups from Turkish hotels by isolated swab and water samples Legionella species and serogroups n (%) Legionella pneumophila 86 (92.5) Serogroup 1 20/93 (21.5) Serogroup 3 4/93 (4.3) Serogroup 6 59/93 (63.5) Serogroups /93 (3.2) Nonpneumophila Legionella species 7 (7.5) Total 93 (100)

3 Legionella and Hotel Water Systems 371 Table 2 Distribution of Legionella species isolated by colony count from Turkish hotel water samples Colony-forming Unit/100 ml Legionella pneumophila ( n ) SG 1 SG 3 SG 6 SGs 7 14 Nonpneumophila Legionella species Total, n (%) < (29.7) 100 to <1, (40.6) 1,000 to <10, (21.9) 10, (7.8) Total (100) SG, serogroup. During our study, three cases of legionnaires disease were diagnosed primarily by means of Legionella urinary antigen test (Binax-NOW Legionella Urinary Antigen Test; Binax Inc., Scarborough, ME, USA). A single patient was diagnosed with Legionella from each of three different study hotels. Legionella pneumophila serogroup 1 was isolated in sputum culture in one patient, and the associated hotel water system had a high concentration (>100 CFU/100 ml) and an extensive colonization (8/12 samples) of L pneumophila serogroup 1. In one of the other two hotels, L pneumophila serogroup 1 and nonpneumophila Legionellla species were isolated in only 2 of 15 samples in low concentrations (<100 CFU/100 ml). One of 11 samples from the third hotel yielded nonpneumophila Legionella species. We were not able to determine DNA patterns for the patient and environmental strains. Discussion The colonization of water distribution systems by Legionella species depends on a combination of several factors, including water temperature, sediment accumulation, and commensal microflora. 9,10 In European hotel water systems, Legionnella colonization ranged from 27% to 75% in several studies Akbas and colleagues 6 and Uzel and col leagues 7 reported Legionella colonization rates as 61 and 92%, respectively, in Turkish hotel water systems. In our study, the colonization of Legionella in hotel water systems was found to be 69.2%. The colony count of Legionella exceeded 100 CFU/100 ml in two thirds of water samples with positive results for that genus of bacteria. The correlation between the extent of colonization and the risk of legionnaires disease may be more important than the concentration of bacteria. 14 However, in many European countries, guidelines concerning the control of legionellosis suggest that bacterial concentrations exceeding 1,000 CFU/L are the threshold for considering preventive measures. 2,11 Our study showed that in 30.8% of the hotels, at least one water sample concentration was equal or more than 100 CFU/100 ml, suggesting that the risk for legionnaires disease in contaminated hotels are worrisome. Extensive colonization or high concentration of Legionella was detected only in one of three hotels in which the three patients with legionnaires disease had stayed. There is uncertainty about the infective dose of organism required to produce infection in human beings. Animal experiments suggest that high doses are required for infection and person-toperson transmission has never been documented. However, like our study, the number of bacteria isolated from the source of infection of legionnaires disease might be very low or undetectable. 15 There are several possible explanations for this circumstance. First, environmental concentration of Legionella might be underestimated because of technical obstacles to detection and also might change at the sampling time. Second, viable nonculturable but pathogenic form of Legionella or legionella-like amoebic pathogens can occur in water systems. Third, Legionella requires an intracellular environment of protozoa for its replication and growth within protozoa enhances the pathogenicity (virulence) of Legionella. Protozoa have been shown to release vesicles of respirable size that contain nu merous Legionella. These vesicles or a few legionellaladen deeply inhaled protozoa can cause infection. 9,15 However, the source of infection in the last two patients might also be associated with another site or hotel. Bacterial transmission to humans is most commonly accepted to be via droplets generated from an environmental source such as cooling towers, showerheads, whirlpools, and other human-made devices that generate aerosols. The epidemiological evidence suggests that the aerosol travels and remains infective from several hundreds meters to a few kilometers. 16 Therefore, combination of national

4 372 and international collaboration is very important in the identification of the sources of legionnaires disease and in the prevention of its dissemination. Forty-two Legionella species with 64 serogroups have been classified to date. Although the distribution of serotypes causing community-acquired or hospital-acquired legionnaires disease varies by geographic region, L pneumophila serogroup 1 is the most common agent of legionnaires disease and accounts for about 70% to 90% of the reported cases of that infection. 17,18 Legionella pneumophila serogroup 6 is more common in nosocomial strains and more likely to be associated with poor outcome. 19 In contrast to clinical specimens, L pneumophila serogroups 2 to 14 accounted for more than 50% of the isolates obtained from environmental water systems in several studies In our study, L pneumophila serogroup 6 accounted for 63.5% of Legionella isolates. Akbas and colleagues 6 from Turkish National Legionella Reference Laboratory have reported L pneumophila serogroup 5 (29.5%), serogroup 6 (24.2%), and serogroup 1 (22.1%) as the most common serogroups in Turkish hotel water systems, but Uzel and colleagues 7 have reported L pneumophila serogroup 1 (85.9%) as the most common serogroup in Izmir province of Turkey. Until November , 598 cases of legionnaires disease, which were mainly diagnosed by urinary antigen test, were reported with a history of travel to Turkey by EWGLI. 4 In Turkey, individual case reports and small case series have shown that L pneumophila serogroup 1 was the most common causative serogroup of patients with legionnaires disease. 4,23,24 However, there is not enough data to compare the environmental isolates with the clinical isolates. The incidence of legionnaires disease depends on the degree of contamination of the aquatic reservoir, the susceptibility of people exposed to that water, differences in virulence among Legionella species, and the intensity of the exposure. 19 However, the discovery of this infection also depends on the availability of specialized laboratory tests and their use in the infected patient. The US Centers for Disease Control and Prevention estimates that overall, more than 25,000 cases of communityacquired legionnaires disease may be occurring in the United States every year, and perhaps more than 95% of those cases are undiagnosed. 1,25 Legionnaires disease is clinically indistinguishable from other causes of pneumonia, so the diagnosis of legionnaires disease requires specific diagnostic tests. The definitive diagnosis of legionnaires disease is made via culture, Legionella urinary antigenuria, or serologic methods. Isolation of Legionella Erdogan and Arslan species is the gold standard for diagnosis of legionnaires disease. The major limitation of sputum culture is that fewer than one half of patients with legionnaires disease produce sputum. Serological testing for Legionella infection is a valuable epidemiological tool but has little impact on making clinical decision because of the time delay before a result is available. Legionella urinary antigen tests, which are easy and useful for early diagnosis of legionnaires disease, have a sensitivity ranging from 70% to 90% and a specificity approaching 100% but only useful for the diagnosis of L pneumophila serogroup 1 infection. 26 According to EWGLI data in 2004, the main diagnostic methods of legionaires disease were Legionella urinary antigen (84.9%), serology (8.7%), culture (5.6%), and other methods (0.8%). 27 It is obvious that legionnaires disease caused by L pneumophila nonserogroup 1 might easily be overlooked. In conclusion, the usefulness of diagnostic tests is influenced by local Legionella epidemiology so international collaboration is very important not only for epidemiological evidence but also microbiological proof of contamination of hotels. We found that L pneumophila serogroup 6 was the most common isolate detected in Turkish hotel water systems in our study. The result of Legionella urinary antigen tests, which are the diagnostic tests most often used to identify legionnaires disease, may be negative in people infected with L pneumophila serogroup 6. For patients with pneumonia of unknown origin and history of travel to Alanya, Turkey, we suggest that clinicians should apply the whole spectrum of laboratory methods for the diagnosis of legionnaires disease and take samples for culture wherever possible. The availability of clinical isolates from culture can also be important in subsequent epidemiologic investigations. Acknowledgments We thank Durmus Ali Akilli, Sibel Karadas, and Burcu Kiras for their assistance in collecting and working samples. Declaration of Interests The authors state that they have no conflicts of interest. References 1. Akbas E, Yu VL. Legionnaires disease and pneumonia. Beware the temptation to underestimate this exotic cause of infection. Postgrad Med 2001 ; 109 :

5 Legionella and Hotel Water Systems Joseph C, Lee J, van Wijngaarden J, et al. European guidelines for control and prevention of travel associated legionnaires disease. Available at : ewgli.org/data/european_guidelines/european_ guidelines_jan05.pdf. ( Accessed 2006 Nov 5 ) 3. Ricketts KD, Joseph CA. The distribution of travelassociated Legionnaires disease within selected European countries, and a comparison with tourist patterns. Epidemiol Infect 2005 ; 22 : The European Working Group for Legionella Infections Website. tables.htm. ( Accessed 2006 Nov 5 ) 5. Guidelines for control of travel-associated legionnaire s disease. Vol. 34. Ankara, Turkey: The Ministry of Health of Turkey, Akbas E, Dalk ı l ı nc I, Gozalan A, Guvener E. Legionella spp in plumbing systems of hotels: a study in Aegean and Mediterranean coasts. Flora 1999 ; 4 : Uzel A, Ucar F, Hames-Kocabas EE. Prevalence of Legionella pneumophila serogroup 1 in water distribution systems in Izmir province of Turkey. APMIS 2005 ; 113 : Department of Statistics, Turkish Ministry of Tourism. Number of arrivals and nights spent, average length of stay and occupancy rate in establishments by provinces, Ankara, Turkey. 9. Abu Kwaik Y, Gao LY, Stone BJ, et al. Invasion of protozoa by Legionella pneumophila and its role in bacterial ecology and pathogenesis. Appl Environ Microbiol 1998 ; 64 : Stout JE, Yu VL, Best MG. Ecology of Legionella pneumophila within water distribution systems. Appl Environ Microbiol 1985 ; 49 : Borella P, Montagna MT, Stampi S, et al. Legionella contamination in hot water of Italian hotels. Appl Environ Microbiol 2005 ; 71 : Alexiou SD, Antoniadis A, Papapaganagiotou J, Stefanou T. Isolation of Legionella pneumophila from hotels of Greece. Eur J Epidemiol 1989 ; 5 : Leoni E, De Luca G, Legnani PP, et al. Legionella waterline colonization: detection of Legionella species in domestic, hotel and hospital hot water systems. J Appl Microbiol 2005 ; 98 : Yu VL. Resolving the controversy on environmental cultures for Legionella: a modest proposal. Infect Control Hosp Epidemiol 1998 ; 19 : Erratum in: Infect Control Hosp Epidemiol 1999; 20: Obrien SJ, Bhopal RS. Legionnaires-disease the infective dose parodoxe. Lancet 1993 ; 342 : Nguyen TM, Ilef D, Jarraud S, et al. A communitywide outbreak of legionnaires disease linked to industrial cooling towers how far can contaminated aerosols spread? J Infect Dis 2006 ; 193 : Helbig JH, Bernander S, Castellani Pastoris M, et al. Pan-European study on culture-proven Legionnaires disease: distribution of Legionella pneumophila serogroups and monoclonal subgroups. Eur J Clin Microbiol Infect Dis 2002 ; 21 : Yu VL, Plouffe JF, Pastoris MC, et al. Distribution of Legionella species and serogroups isolated by culture in patients with sporadic community-acquired legionellosis: an international collaborative survey. J Infect Dis 2002 ; 186 : Yu VL. Legionella pneumophila (legionnaires disease). In : Mandell GL, Bennett JE, Dolin R, eds. Principles and practice of infectious diseases. 5th Ed. Philadelphia, PA : Churchill Livingstone Inc., 2000 : Doleans A, Aurell H, Reyrolle M, et al. Clinical and environmental distributions of Legionella strains in France are different. J Clin Microbiol 2004 ; 42 : Marinella F, Michela G, Lucio M, et al. Environmental investigations and search for Legionella pneumophila in water. In : Mandarino G, ed. 20th Annual meeting European Working Group for Legionella Infections. Rome : ISTISAN Congressi 05/C2, 2005 : Montagna MT, Napoli C, Tato D, et al. Clinicalenvironmental surveillance of legionellosis: an experience in Southern Italy. Eur J Epidemiol 2006 ; 21 : Ozerol IH, Bayraktar M, Cizmeci Z, et al. Legionnaire s disease: a nosocomial outbreak in Turkey. J Hosp Infect 2006 ; 62 : Erdogan H, Lakamdayali H, Y ı lmaz A, et al. Seven cases of travel-associated legionnaire s disease in Turkey. In : Mandarino G, ed. 20th Annual Meeting European Working Group for Legionella Infections. Rome : ISTISAN Congressi 05/C2, 2005 : Marston BJ, Plouffe JF, File T, et al. Incidence of community-acquired pneumonia requiring hospitalization results of a population-based active surveillance study in Ohio. Arch Intern Med 1997 ; 157 : Murdoch DR. Diagnosis of Legionella infection. Clin Infect Dis 2003 ; 36 : Ricketts KD, McNaught B, Joseph CA, European Working Group for Legionella Infections. Travelassociated legionnaires disease in Europe: Euro Surveill 2006 ; 11 :

Community acquired pneumonia due to Legionella pneumophila in a tertiary care hospital

101 Research article Community acquired pneumonia due to Legionella pneumophila in a tertiary care hospital Abstract BN Dissanayake 1,, DE Jayawardena 2, CG Senevirathna 1, TM Gamage 1 Sri Lankan Journal