Molecular epidemiology of Legionnaires disease in Israel

Transcription

1 ORIGINAL ARTICLE EPIDEMIOLOGY Molecular epidemiology of Legionnaires disease in Israel J. Moran-Gilad 1,2, M. Mentasti 3, T. Lazarovitch 4, Z. Huberman 5, T. Stocki 3, C. Sadik 6, T. Shahar 6, E. Anis 6, L. Valinsky 7, T. G. Harrison 3, I. Grotto 2 and on behalf of ESCMID Study Group for Legionella Infections (ESGLI) 1) National Programme for Legionellosis Control, Ministry of Health, 2) Public Health Services, Ministry of Health, Jerusalem, Israel, 3) Reference Microbiology Services, Public Health England, London, UK, 4) Department of Clinical Microbiology, Assaf Harofeh Medical Centre, Zerifin, 5) National Public Health Laboratory, Ministry of Health, Tel-Aviv, Israel, 6) Division of Epidemiology, Ministry of Health and 7) Central Government Laboratories, Ministry of Health, Jerusalem, Israel Abstract National surveillance of Legionnaires disease (LD) is important to inform control measures and facilitate international networking for timely reporting. This study is the first to describe the molecular epidemiology of LD in Israel. Case notifications for , collated through mandatory reporting, were identified and demographic, clinical and laboratory data were extracted. Unrelated clinical and environmental Legionella pneumophila strains were characterized using standard procedures, Dresden panel of monoclonal antibodies and the ESCMID Study Group for Legionella Infections (ESGLI) Sequence-Based Typing scheme. In all, 294 cases were reported (crude incidence 0.67 cases/ ; age-standardized incidence 1/ ). LD epidemiological trends and features largely resembled those of the EU, except for a larger proportion of nosocomial cases. Of 28 clinical and 23 environmental strains analysed, 71.4% and 21.7% were serogroup (sg) 1 and the most common immunological subgroup was OLDA/Oxford (64%). Of the clinical strains, OLDA/Oxford, ST1 was the most common (43%) followed by Allentown/France, ST40 (14%). The unusual sg 3 ST338 was found in 17.4% of environmental strains. Novel STs were detected amongst 23.5% of strains. These findings warrant further molecular investigation. Molecular epidemiology data generated from neighbouring countries newly adopting the ESGLI typing scheme for L. pneumophila contribute to understanding of regional strain diversity. Keywords: Legionella, Legionnaires disease, molecular epidemiology, typing Original Submission: 6 August 2013; Revised Submission: 26 September 2013; Accepted: 5 October 2013 Editor: S. Cutler Article published online: 9 December 2013 Clin Microbiol Infect 2014; 20: / Corresponding author: J. Moran-Gilad, National Microbiology Focal Point & Head of Microbiology Cooperation, Public Health Services, Ministry of Health, 39 Yermiyahu St., Jerusalem , Israel giladko@post.bgu.ac.il This work has been presented in part at the 23rd European Conference on Clinical Microbiology and Infectious Diseases (ECCMID), Berlin, April Background Legionellae are Gram-negative bacteria ubiquitously found in aqueous environments [1]. Humans can be infected through exposure to aerosol particles generated from contaminated man-made water systems, such as spa pools, cooling towers or showering facilities. Clinical manifestation varies from a mild influenza-like illness called Pontiac fever to pneumonia of varying severity known as Legionnaires disease (LD) [2]. Amongst the 60 or more Legionella species, Legionella pneumophila causes the vast majority of LD and out of 16 known serogroups, L. pneumophila serogroup 1 accounts for over 80% of LD cases [3,4]. Legionnaires disease is a reportable communicable disease in many European countries, with an overall incidence of 1.16 per EU population [4]. EU surveillance of LD is coordinated by the European Legionnaires Disease Surveillance Network (ELDSNet) of the European Centre for Disease Prevention and Control (ECDC). Surveillance data demonstrated a mild increase in the incidence of legionellosis over the last decade and excess morbidity in recent years that may be associated with climate change [5,6]. Clinical Microbiology and Infection ª2013 European Society of Clinical Microbiology and Infectious Diseases

2 CMI Moran-Gilad et al. Molecular epidemiology of Legionnaires disease 691 The standardized sequence-based typing (SBT) scheme for L. pneumophila developed by the European Working Group for Legionella Infections (EWGLI, now European Society for Clinical Microbiology Study Group on Legionella Infections, ESGLI) marked an important advancement in the study of the molecular epidemiology of LD [7,8]. Implementation of SBT has yielded useful and comparable data at a regional, national and international level [9] and has also been shown to have an important role in the investigation of unusual LD cases [10] and LD outbreaks [11]. Currently, there are only limited data on the epidemiology of LD in Israel and no molecular epidemiology data. The aim of this report is to present national surveillance data from Israel collated in recent years and results of molecular typing following the implementation of SBT in this country. Methods The study was conducted by the Israeli Ministry of Health. LD is a reportable disease in Israel and LD case data are passively reported to the MOH Division of Epidemiology by District Health Offices following reports generated by frontline clinicians. A case of LD was defined as an acute respiratory illness with laboratory evidence of infection with L. pneumophila by urinary antigen testing, serology, sputum culture or molecular detection in sputum by PCR based on modification of EU case definition [12]. All cases of LD reported to the MOH between 2006 and 2011 were identified in the registry of notifiable diseases and relevant available data were extracted and analysed, including demographic data, underlying risk factors, setting of exposure, laboratory diagnosis and outcome. A convenience sample of L. pneumophila strains recovered during the study period was chosen from the MOH strain repository. Clinical strains were recovered from respiratory samples of suspected LD patients and identified by standard laboratory procedures at the point of care, prior to referral to the reference laboratory. Environmental strains were recovered at MOH-certified public health laboratories from inspection samples obtained from water systems according to national regulations. Clinical and environmental strains were epidemiologically unrelated. Isolates were cultured on BCYEa plates (Oxoid, Basingstoke, UK) for h at 35 C before phenotypic and molecular tests were performed. Presumptive identification as L. pneumophila was confirmed using the MonoFluo Legionella pneumophila IFA Test (Biorad, Hemel Hempstead, UK), then the Dresden panel of monoclonal antibodies was used to determine the serogroup and, in the case of sg 1 strains, the subgroup [13]. Strains not readily confirmed as L. pneumophila were identified to species level by sequencing the mip gene as described by Ratcliff et al. [14] and comparing the sequence to the mip database [15]. Between two to three single colonies per L. pneumophila strain were pelleted and DNA extracted using the InstaGene Matrix (Biorad). The genotype of each strain was determined using the M13 modification of the ESGLI SBT method [16]. All alleles and sequence types (STs) were determined using the Legionella Sequence Quality Tool [17,18].The global frequency of specific sequence types for comparison with local findings was obtained from SBT database queries as of 1 July The strain diversity index was calculated according to the modified method of Hunter and Gaston [19]. Results Demographic data During the study period, 294 cases of LD were reported; an annual mean of 49.3 cases (range cases per year). These frequencies corresponded to an annual incidence rate of cases per population (mean incidence 0.67 per ). The mean age was years (median 65.5, range 2 months to 99 years). The vast majority of cases (86.4%) occurred amongst patients over 45 years of age, and 51.7% of cases occurred amongst the 65+ year age band. The age-specific mean annual incidence rates in the year and 65+ year age bands were 1.2 and 3.5 per population, respectively. Male patients accounted for 201 LD cases (68.3%), yielding a male-to-female ratio of 2.16:1. Seasonal distribution of LD cases showed a slightly increased occurrence during the warm season, with 51% of cases diagnosed between July and November (Fig. 1). The main geographical areas from which LD cases were reported included northern Israel (132 cases, 49%), central Israel (80 cases, 27.2%) and the Tel-Aviv region (71 cases, 24.15%). Only a few cases were reported from Jerusalem and the southern districts. The adjusted incidence per district was highest in the Haifa district of northern Israel (1.5/ ), followed by the Tel-Aviv and central districts (0.9 and 0.8/ , respectively). Clinical features The setting of exposure was known in 156 cases (53%), of which 45.5% were healthcare-associated acquisitions (24% overall), followed by community acquisitions in 39.6% (21% overall) and travel-associated acquisitions (TALD) in 15% (8% overall). Apart from male sex and age, the most common risk factors for LD found amongst patients were chronic renal

3 692 Clinical Microbiology and Infection, Volume 20 Number 7, July 2014 CMI FIG. 1. Seasonal distribution of Legionnaires disease cases, failure (19%), malignancy (19%), diabetes mellitus (16%), immune system compromise (7%), including bone marrow transplantation, solid organ transplantation, corticosteroid therapy and HIV, and chronic respiratory disease (4%). LD was diagnosed by urinary antigen test in 88% of patients, culture in 7%, serology in 3% and PCR in 2%. The case fatality rate was 12.6%. Phenotypic and molecular subtyping In all, 28 clinical strains and 23 environmental strains were characterized at both phenotypical and molecular level. Serogrouping results are summarized in Table 1. The majority of clinical strains (71.4%) were sg 1, whereas for environmental strains, sg 3 was most prevalent (52.2%). The distribution of subgroups amongst sg 1 strains is depicted in Table 1. Notably, TABLE 1. Clinical and environmental Legionella pneumophila strains by serogroup and subtype Source Serogroup N (%) Monoclonal subgroup N (% of sg.1 strain) Sequence type N (% of clinical strains) Healthcareassociated cases Global frequency of ST a N % Clinical (N = 28) 1 20 (71.4) OLDA/Oxford 11 (64) ST1 12 (42.85) d Allentown/France 4 (16) ST23 1 (3.6) ST40 4 e (14.3) ST1207 b 1 (3.6) NA f Benidorm 1 (4) ST1206 b 1 (3.6) 1 NA Knoxville 1 (4) ST345 1 (3.6) Not tested 3 (12) NA NA NA NA 3 4 (14.3) NA ST87 2 (7.15) ST338 1 (3.6) ST93 1 (3.6) (7.1) Untypeable c 2 (7.15) 1 NA 6 1 (3.6) ST1208 b 1 (3.6) NA 8 1 (3.6) ST1351 b 1 (3.6) NA Environmental (N = 23) 1 5 (21.7) OLDA/Oxford 5 (100) ST1 5 NA (52.2) NA ST ST ST <0.1 ST <0.1 ST1269 b 1 NA ST1271 b 4 NA 6 6 (26.1) ST ST <0.1 ST1208 b 1 NA ST1270 b 2 NA a As of 15 July 2013; combined clinical and environmental strains. b Novel ST. c Only a partial allelic profile has been generated due to failed amplification of certain loci. d One isolate was not mab typed. e Two isolates were not mab typed. f Not applicable.

4 CMI Moran-Gilad et al. Molecular epidemiology of Legionnaires disease 693 OLDA/Oxford was the most common subgroup, accounting for 64% of sg 1 strains (55% of sg 1 clinical strains and 100% of sg 1 environmental strains). The frequency of the obtained sequence types (STs) according to SBT results is summarized in Table 1. The most common was ST1 (accounting for 33.3% of strains, roughly half of clinical and a quarter of environmental strains) and all 17 ST1 strains belonged to the OLDA/Oxford Dresden subgroup. L. pneumophila sg 1 OLDA/Oxford ST1 is thus the leading strain found in LD in Israel. This phenon accounts for 39% of ST1s in the global Legionella database. Of clinical strains, the second most common was ST40, which accounted for 14.3% of strains in Israel but is relatively uncommon globally. All these strains belonged to the Allentown/France subgroup of sg 1, which is the typical phenon found in ST40. Of environmental strains, the second most common ST was ST338 (17.4%), an sg 3 previously reported only on rare instances. Interestingly, 12 (23.5%) of the L. pneumophila strains yielded novel STs. Most of those strains were sg 3 (ST1269, ST1271) or sg 6 (ST1208, ST1270, ST1438) and the remainder were sg 1 (ST1206, ST1207) or sg 8 (ST1351). The calculated FIG. 2. Minimum spanning tree showing the distribution of clinical (red) and environmental (blue) Legionella pneumophila strains with seven sequenced alleles in relation to sequence type (ST). ST number is shown and circle size indicates the number of strains assigned to ST. Thick lines connect single-locus variants, thin lines connect double-locus variants and dotted lines connect other variants (more than two allele difference). diversity index was for clinical isolates and for environmental isolates. The diversity of study strains is shown in Fig. 2. Discussion Until recently, data on LD in Israel have been limited to case reports or case series [10,20] as well as prevalence data based on serological testing for legionella amongst a large cohort of patients with respiratory infections in southern Israel, showing rates of up to 16.2% [21,22]. The increasing volume of activity following the revision of national regulations for legionella control in water systems in Israel over recent years, the challenges associated with linking LD cases to environmental sources and the growing need to enhance international collaboration with regard to travel-associated LD have played a key role in the inauguration of a National Programme for Legionellosis Control in Israel. This study, which is the first to outline the molecular epidemiology of LD in Israel, reflects the priorities set by the programme. The annual incidence of LD found in Israel is slightly lower than that reported on average in Europe [4]. Taking into account Israel s warm climate that renders cold water systems hot for most of the year, we hypothesized that LD incidence would be relatively high. Observed rates of notifications could be explained by under-diagnosis, under-reporting, efficiency of control measures, differences in distribution of age and host risk factors in the Israeli population or a combination of such factors. That almost no cases of LD were reported from certain geographical regions, such as the south of Israel, which has the warmest climate, strongly suggests that under-diagnosis may have indeed occurred. This is plausible, especially because the seasonal distribution of LD in Israel resembles that of the EU, with the exception of the proportion of cases occurring in November. Notably, the age distribution and sex ratio of LD in Israel were strikingly similar to those reported from the EU [5]. In order to account for differences in population composition and age as a risk factor, incidence rates of Israel were subject to standardization by the direct method taking the average EU population as standard [23]. The age-standardized incidence rates increased from a range of per to per and the mean annual incidence increased following standardization from 0.67 to 1 per population, which is very close to the EU average. The setting of exposure observed in Israel markedly differs from that of the EU [5]. The proportion of nosocomial LD (24%) is much higher than in the EU (5%) whereas the proportion of combined domestic and international TALD

5 694 Clinical Microbiology and Infection, Volume 20 Number 7, July 2014 CMI (8%) is much lower (24%). Possible explanations for that observation include the lack of historical data for 47% of cases and suboptimal TALD surveillance before Specifically, it is possible that nosocomial cases are more likely to be diagnosed and reported and thus our findings represent a surveillance bias. As a result, these factors have been rectified in order to ensure adequate data for future surveillance. It is noteworthy that none of the reported LD cases in Israel were associated with disease clusters whereas in the EU 7% of cases with known status are related to clusters [5]. Because all cases of LD in Israel are investigated by local health authorities we believe that under-diagnosis does not explain lack of clusters. Even in Europe, no clusters were noted in 16 out of 27 countries, and most clusters were reported from as few as four countries. This suggests differences in epidemiology and surveillance between countries. In Israel, only a few LD clusters have been reported over the last two decades. These include a nosocomial outbreak involving a bone marrow transplant unit [24] and a well publicized cluster involving a hotel in Eilat ( disease.2003.html). Whether this is associated with enforcement of control measures or virulence of circulating strains remains to be determined. The case fatality rate observed in our cohort is slightly higher than that reported from the EU (12.6% vs. 9%). This may reflect the larger proportion of nosocomial cases in our cohort, which may involve patients at high risk of mortality due to their underlying health conditions, such as immunocompromised patients, who comprised over a quarter of our patient sample. Reliance on urinary antigen testing for LD diagnosis was evident in our patient series as reported from most countries, for example Italy [25]. In 2011, 77%, 12% and 4% of EU patients were diagnosed by means of urinary antigen, culture and PCR, respectively [5]. The use of such tests in Israeli hospitals was similar and in light of the increasing availability of frontline molecular diagnostics, the fraction of PCR-positive cases is expected to be higher in the forthcoming years. This approach may be expected to increase the detection of LD cases caused by non-sg. 1 strains, which may be overlooked by urinary antigen testing. Our molecular analysis revealed 19 distinct STs found amongst the 51 characterized strains obtained over a 6-year period. This corresponded to an index of diversity of and for clinical and environmental strains, respectively. Other nationwide studies have yielded 86 STs amongst 214 isolates in Canada over a 29-year period [26] or 53 STs amongst 86 isolates in Japan over a 28-year period [27]. Both studies found an index of diversity >0.9 for clinical strains. Similarly high diversity indices were also found in the United Kingdom [9]. One probable explanation for the relatively low level of diversity found in our clinical strain series is the abundance of ST1, which accounted for over 40% of strains. Whilst the number of study years and country size may negatively contribute to strain diversity, the high index of diversity found in our environmental strain collection suggests that these factors cannot solely account for such findings. ST1 has been shown to be the most common strain in many countries [28,29] and is commonly found in the environment even when the number of ST1-associated LD cases is low [9]. ST1 has also been reported as the most common ST causing infection in Germany in a study that included a large proportion (45.7%) of nosocomial cases [30]. In our molecular analysis, only five clinical strains (18%) were of nosocomial origin and thus it is difficult to draw conclusions on the role of ST1 as a cause of nosocomial LD in Israel. While one of those involved OLDA/Oxford, ST1, this should be addressed in future studies. Notably, ST1 is found more commonly in man-made as opposed to natural water sources [26]. Mindful of the arid conditions found in many parts of Israel, STs typical of natural water sources are less likely to be encountered, thus increasing the proportion of ST1. ST47 is very common in many EU countries, such as the United Kingdom [9], Belgium [31], France [32] and the Netherlands [33]. No ST47 strain was found in our study, either in the environment or in clinical samples. We cannot rule out the possibility that ST47 was not detected due to the small sample size and therefore, the presence of this important strain should be assessed prospectively. In addition to ST1 and ST47, several other types are relatively common from a global perspective. These include in decreasing order of frequency ST23, ST62, ST42, ST9 and ST37. Together with ST47, these types account for 21% of SBT database entries. Except for one ST23 strain, none of those have been detected in our study and this further emphasizes the unique ST distribution found in Israel. Our study has several limitations, including the limited number of years and relatively small sample size of studied strains. Moreover, due to under-diagnosis and under-reporting as well as missing meta-data, our data could have been biased towards inclusion of nosocomial LD and thus may not adequately reflect community-acquired LD. Conclusions This is the first study on the molecular epidemiology of LD in Israel. Our data demonstrate many similarities to the epidemiological trends found in EU countries, especially in relation to incidence rates, seasonality and methods used for laboratory diagnosis. Notably, our data include a larger proportion of nosocomial cases than that reported in many countries.

6 CMI Moran-Gilad et al. Molecular epidemiology of Legionnaires disease 695 The molecular characterization of our L. pneumophila strains shows several unusual features, including an unusually high percentage of the sg 1 OLDA/Oxford, ST1 phenon leading to reduced clinical strain diversity, a notable number of clinical cases involving the sg 1 Allentown/France, ST40 phenon and the presence of the rare sg 3 ST338 strain in the environment. In addition, a substantial percentage of our strains yielded novel STs, which were added to the global Legionella SBT database. Accumulation of molecular epidemiology data from neighbouring countries newly adopting the ESGLI SBT scheme for L. pneumophila contributes to understanding of regional strain diversity and may facilitate international activities such as TALD surveillance. Our findings deserve further molecular investigation by expanding the clinical and environmental strain sample as well as harnessing of genomic technologies to dissect the molecular epidemiology of ST1, which appears to be the dominant L. pneumophila strain in Israel. Transparency Declaration No conflict of interest to declare. References 1. Fields BS, Benson RF, Besser RE. Legionella and Legionnaires disease: 25 years of investigation. Clin Microbiol Rev 2002; 15: Diederen BM. Legionella spp. and Legionnaires disease. J Infect 2008; 56: 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: European Centre for Disease Prevention and Control. Annual Epidemiological Report Reporting on 2010 surveillance data and 2011 epidemic intelligence data. Stockholm: ECDC, European Centre for Disease Prevention and Control. Legionnaires disease in Europe, Stockholm: ECDC, Available from (last accessed 15 July 2013). 6. Beaute J, Zucs P, de Jong B; on behalf of the European Legionnaires Disease Surveillance Network. Legionnaires disease in Europe, Euro Surveill 2013; 18: Gaia V, Fry NK, Afshar B et al. A consensus sequence-based epidemiological typing scheme for clinical and environmental isolates of Legionella pneumophila. J Clin Microbiol 2005; 43: Ratzow S, Gaia V, Helbig JH, Fry NK, L uck PC. Addition of neua, the gene encoding N-acylneuraminate cytidylyl transferase, increases the discriminatory ability of the consensus sequence-based scheme for typing Legionella pneumophila serogroup 1 strains. J Clin Microbiol 2007; 45: Harrison TG, Afshar B, Doshi N, Fry NK, Lee JV. Distribution of Legionella pneumophila serogroups, monoclonal antibody subgroups and DNA sequence types in recent clinical and environmental isolates from England and Wales ( ). Eur J Clin Microbiol Infect Dis 2009; 28: Moran-Gilad J, Lazarovitch T, Mentasti M et al. Humidifier-associated paediatric Legionnaires disease, Israel, February Euro Surveill 2012; 17: Coetzee N, Duggal H, Hawker J et al. An outbreak of Legionnaires disease associated with a display spa pool in retail premises, Stoke-on-Trent, United Kingdom, July Euro Surveill 2012; 17: Commission Decision of 28 April 2008 amending Decision 2002/253/ EC laying down case definitions for reporting communicable diseases to the Community network under Decision No 2119/98/EC of the European Parliament and of the Council. [Internet]. Available from: :01:EN:HTML (last accessed 6 December 2011). 13. Helbig JH, Bernander S, Castellani Pastoris M, Etienne J, Gaia V 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: Ratcliff RM, Lanser JA, Manning PA, Heuzenroeder MW. Sequencebased classification scheme for the genus Legionella targeting the mip gene. J Clin Microbiol 1998; 36: Health Protection Agency. mip gene sequence database. Available from: web_c/ (last accessed 15 July 2013). 16. Health Protection Agency. Legionella pneumophila Sequence based typing. Available from: legionella/legionella_sbt/php/sbt_homepage.php (last accessed 15 July 2013). 17. Underwood AP, Bellamy W, Afshar B, Fry NK, Harrison TG. Development of an online tool for the European Working Group for Legionella Infections sequence-based typing, including automatic quality assessment and data submission. In: Cianciotto NP, Abu Kwaik Y, Edelstein PH, Fields BS, Geary DF, Harrison TG, Joseph CA, Ratcliff RM, Stout JE, Swanson MS, eds, Legionella: state of the art 30 years after its recognition, Chapter 44. Washington: ASM Press, 2006; Health Protection Agency. Legionella SBT quality assessment. Available from: assemble_legionella1.cgi (last accessed 15 July 2013). 19. Hunter PR, Gaston MA. Numerical index of the discriminatory ability of typing systems: an application of Simpson s index of diversity. J Clin Microbiol 1988; 26: Shachor-Meyouhas Y, Kassis I, Bamberger E et al. Fatal hospital-acquired Legionella pneumonia in a neonate. Pediatr Infect Dis J 2010; 29: Lieberman D, Porath A, Schlaeffer F, Lieberman D, Boldur I. Legionella species community-acquired pneumonia: a review of 56 hospitalized adult patients. Chest 1996; 109: Lieberman D, Schlaeffer F, Boldur I et al. Multiple pathogens in adult patients admitted with community-acquired pneumonia: a one year prospective study of 346 consecutive patients. Thorax 1996; 51: Eurostat. Population on 1 January by five years age groups and sex ( ). Luxembourg: Eurostat. Available from: eurostat.ec.europa.eu/nui/show.do?dataset=demo_pjangroup&lang=en (last accessed 20 August 2012). 24. Oren I, Zuckerman T, Avivi I, Finkelstein R, Yigla M, Rowe JM. Nosocomial outbreak of Legionella pneumophila serogroup 3 pneumonia in a new bone marrow transplant unit: evaluation, treatment and control. Bone Marrow Transplant 2002; 30: Rota MC, Caporali MG, Bella A, Ricci ML, Napoli C. Legionnaires disease in Italy: results of the epidemiological surveillance from 2000 to Euro Surveill 2013; 18:

7 696 Clinical Microbiology and Infection, Volume 20 Number 7, July 2014 CMI 26. Reimer AR, Au S, Schindle S, Bernard KA. Legionella pneumophila monoclonal antibody subgroups and DNA sequence types isolated in Canada between 1981 and 2009: laboratory Component of National Surveillance. Eur J Clin Microbiol Infect Dis 2010; 29: Amemura-Maekawa J, Kura F, Helbig JH et al. Characterization of Legionella pneumophila isolates from patients in Japan according to serogroups, monoclonal antibody subgroups and sequence types. J Med Microbiol 2010; 59: Cazalet C, Jarraud S, Ghavi-Helm Y et al. Multigenome analysis identifies a worldwide distributed epidemic Legionella pneumophila clone that emerged within a highly diverse species. Genome Res 2008; 18: Tijet N, Tang P, Romilowych M et al. New endemic Legionella pneumophila serogroup I clones, Ontario, Canada. Emerg Infect Dis 2010; 16: Borchardt J, Helbig JH, L uck PC. Occurrence and distribution of sequence types among Legionella pneumophila strains isolated from patients in Germany: common features and differences to other regions of the world. Eur J Clin Microbiol Infect Dis 2008; 27: Vekens E, Soetens O, De Mendonna R et al. Sequence-based typing of Legionella pneumophila serogroup 1 clinical isolates from Belgium between 2000 and Euro Surveill 2012; 17: Ginevra C, Forey F, Campese C et al. Lorraine strain of Legionella pneumophila serogroup 1, France. Emerg Infect Dis 2008; 14: Den Boer JW, Bruin JP, Verhoef LPB, Van der Zwaluw K, Jansen R, Yzerman EP Genotypic comparison of clinical Legionella isolates and patient-related environmental isolates in The Netherlands, Clin Microbiol Infect 2008; 14: