The Detection Limits of Legionella According to the EU Directive 2020/2184. Could That Be Too Permissive?

INTRODUCTION: The problem of detecting legionella after a case of legionellosis from the source of environmental contamination has been known since a long time ago. Legionella is a bacterium present in various natural and artificial habitats and especially in surface fresh waters. It is found in greater concentration in warm waters, at temperatures between 20 °C and 42 °C. The greatest risk factor for humans is represented by the presence of Legionella in water distribution systems in hospitals, medical equipment (e.g. respirators, dialyzers, inhalers, humidifiers, water, massage equipment used in balneotherapy) and turbines used in dental practices, especially for hospitalized individuals. In the EU directive 2020/2184, issued by the European parliament on 16/12/2020, the concentration of Legionella was added to the parameters to be determined in assessing the quality of drinking water intended for human consumption. The objectives were to improve the quality standard of drinking water, reduce the consumption of bottled water and consequently reduce plastic waste. The WHO notes that Legionella causes the greatest burden from a health point of view and it is included among the parameters that require careful monitoring with a limit of less than 1000 CFU/L. The aim of this report was to evaluate the new EU directive 2020/2184 on the light of our laboratory experience. MATERIALS AND METHODS: A total of 459 samples were processed at our Hygiene of food Laboratory - Department of Medical Sciences and Public Health. All statistical analyses were conducted using the SPSS statistical package (version 23 for Windows. SPSS, Inc. Chicago, Ill). RESULTS AND DISCUSSION: Of the 67 structures examined where the cases occurred, 35 showed samples with at least one over-threshold value considering the reference value of 100 CFU/L, whereas using the new limit of 1000 CFU/L, only 25 structures resulted as having at least one sample above the threshold. In our experience as a regional reference laboratory for Legionella research, the increase from 100 CFU/L to 1,000 CFU/L could lead to a lower alert level. In fact, in the period between October 2017 and October 2021, the median value of CFU/L in presence of a case was 0 (0-100). Despite the large amount of studies on Legionella only a few relate the withdrawals and the consequent CFU/L with the confirmed cases of legionellosis, as in our analysis. The 75° percentile values of the Legionella concentration equal to 100 CFU/L in all samples associated with cases and clusters leads us to hypothesize that the limit equal to 1000 CFU/L that will be introduced for environmental monitoring as per recent European regulations may not be sufficiently protective for minimizing risk in the population, especially in healthcare facilities where fragile patients are assisted.

Acute Pancreatitis Associated with Atypical Bacterial Pneumonia: Systematic Literature Review.

BACKGROUND: Extra-pulmonary features sometimes occur in association with atypical bacterial pneumonia and include neurologic manifestations, diarrhea, rashes, altered liver enzymes, or kidney injury, among other conditions. Acute pancreatitis has been associated with atypical pneumonias since 1973. METHODS: We performed a systematic review of the literature in the Excerpta Medica, National Library of Medicine, and Web of Science databases. We retained 27 reports published between 1973 and 2022 describing subjects with an otherwise unexplained pancreatitis temporally associated with an atypical pneumonia. RESULTS: The reports included 33 subjects (19 males, and 14 females; 8 children and 25 adults) with acute pancreatitis temporally associated with atypical pneumonia caused by Mycoplasma pneumoniae (n = 18), Legionella species (n = 14), or Coxiella burnetii (n = 1). Approximately 90% of patients (n = 29) concurrently presented with respiratory and pancreatic diseases. No cases associated with Chlamydophila pneumoniae, Chlamydophila psittaci, or Francisella species were found. CONCLUSIONS: Acute pancreatitis has been associated with various infectious agents. The present review documents the association with atypical pneumonia induced by Mycoplasma pneumoniae, Legionella species, and Coxiella burnetii.

Atypical pneumonia: Pathophysiology, diagnosis, and treatment.

Atypical pneumonia is caused by atypical pathogens that are not detectable with Gram stain and cannot be cultured using standard methods. The most common causative organisms of atypical pneumonia are Mycoplasma pneumoniae, Chlamydia pneumoniae, and Legionella species. The therapeutic approach for atypical pneumonias is different than that for typical pneumonia. Typical bacterial pathogens classically respond to ß-lactam antimicrobial therapy because they have a cell wall amenable to ß-lactam disruption. On the contrary, most atypical pathogens do not have a bacterial cell wall, some are intracellular (e.g., Legionella), and some are paracellular (e.g., M. pneumoniae). To prevent an increase in the number of antimicrobial-resistant strains, the Japanese pneumonia guidelines have proposed a differential diagnosis for typical bacterial pneumonia and atypical pneumonia to select an appropriate antibiotic for the management of mild-to-moderate pneumonia. The guidelines have set up six parameters and criteria based on the clinical symptoms, physical signs, and laboratory data. However, in the elderly individuals and patients with underlying diseases, the differential diagnosis may be difficult or a mixed infection may be latent. Therefore, in these individuals, the administration of a ß-lactam drug plus a macrolide or tetracycline, or only fluoroquinolone should be considered from the beginning to cover bacterial and atypical pneumonia.

Detection of Legionella species, the influence of precipitation on the amount of Legionella DNA, and bacterial microbiome in aerosols from outdoor sites near asphalt roads in Toyama Prefecture, Japan.

BACKGROUND: Legionellosis is caused by the inhalation of aerosolized water contaminated with Legionella bacteria. In this study, we investigated the prevalence of Legionella species in aerosols collected from outdoor sites near asphalt roads, bathrooms in public bath facilities, and other indoor sites, such as buildings and private homes, using amoebic co-culture, quantitative PCR, and 16S rRNA gene amplicon sequencing. RESULTS: Legionella species were not detected by amoebic co-culture. However, Legionella DNA was detected in 114/151 (75.5%) air samples collected near roads (geometric mean ± standard deviation: 1.80 ± 0.52 log10 copies/m3), which was comparable to the numbers collected from bathrooms [15/21 (71.4%), 1.82 ± 0.50] but higher than those collected from other indoor sites [11/30 (36.7%), 0.88 ± 0.56] (P < 0.05). The amount of Legionella DNA was correlated with the monthly total precipitation (r = 0.56, P < 0.01). It was also directly and inversely correlated with the daily total precipitation for seven days (r = 0.21, P = 0.01) and one day (r = - 0.29, P < 0.01) before the sampling day, respectively. 16S rRNA gene amplicon sequencing revealed that Legionella species were detected in 9/30 samples collected near roads (mean proportion of reads, 0.11%). At the species level, L. pneumophila was detected in 2/30 samples collected near roads (the proportion of reads, 0.09 and 0.11% of the total reads number in each positive sample). The three most abundant bacterial genera in the samples collected near roads were Sphingomonas, Streptococcus, and Methylobacterium (mean proportion of reads; 21.1%, 14.6%, and 1.6%, respectively). In addition, the bacterial diversity in outdoor environment was comparable to that in indoor environment which contains aerosol-generating features and higher than that in indoor environment without the features. CONCLUSIONS: DNA from Legionella species was widely present in aerosols collected from outdoor sites near asphalt roads, especially during the rainy season. Our findings suggest that there may be a risk of exposure to Legionella species not only in bathrooms but also in the areas surrounding asphalt roads. Therefore, the possibility of contracting legionellosis in daily life should be considered.

Microbial Contamination of Rubber Ducks Floating in Bathtubs of Bathing Facilities, and an Evaluation of Their Washing Methods.

Microbiological contamination inside rubber ducks floating in the bathtub at a "duck bath" of a bathing facility was analyzed by examining bacterial and amoebic counts. The results of microbial tests, such as standard plate count, heterotrophic plate count and Legionella-LAMP (Loopmediated isothermal amplification) , showed that microbial contamination increased in the rubber ducks over time. When the ducks were washed with sodium hypochlorite, those microbial contaminations were not detected; but when the ducks were washed with an electrolyzed water, the standard plate counts and the heterotrophic plate counts were detected in the amount of 103 per duck in the wipe samples. Without proper washing of rubber ducks, bacteria and free-living amoeba can grow and colonize biofilms, and can thereby cause infection in humans. Also, microbial contamination inside ducks may reduce chlorination of the entire bathtub and cause bacterial infection such as Legionellosis from the bathtub water.

How Molecular Typing Can Support Legionella Environmental Surveillance in Hot Water Distribution Systems: A Hospital Experience.

In this study, we aimed to associate the molecular typing of Legionella isolates with a culture technique during routine Legionella hospital environmental surveillance in hot water distribution systems (HWDSs) to develop a risk map able to be used to prevent nosocomial infections and formulate appropriate preventive measures. Hot water samples were cultured according to ISO 11731:2017. The isolates were serotyped using an agglutination test and genotyped by sequence-based typing (SBT) for Legionella pneumophila or macrophage infectivity potentiator (mip) gene sequencing for non-pneumophila Legionella species. The isolates' relationship was phylogenetically analyzed. The Legionella distribution and level of contamination were studied in relation to temperature and disinfectant residues. The culture technique detected 62.21% of Legionella positive samples, characterized by L. pneumophila serogroup 1, Legionella non-pneumophila, or both simultaneously. The SBT assigned two sequence types (STs): ST1, the most prevalent in Italy, and ST104, which had never been isolated before. The mip gene sequencing detected L. anisa and L. rubrilucens. The phylogenetic analysis showed distinct clusters for each species. The distribution of Legionella isolates showed significant differences between buildings, with a negative correlation between the measured level of contamination, disinfectant, and temperature. The Legionella molecular approach introduced in HWDSs environmental surveillance permits (i) a risk map to be outlined that can help formulate appropriate disinfection strategies and (ii) rapid epidemiological investigations to quickly identify the source of Legionella infections.

Epidemiology of Legionnaires' Disease, Hong Kong, China, 2005-2015.

We reviewed findings of clinical, epidemiologic, and environmental investigations for 288 confirmed case-patients with Legionnaires' disease reported in Hong Kong, China, during January 2005-December 2015. We found that chronic renal failure/impairment (adjusted odds ratio [aOR] 4.09), chronic pulmonary diseases (aOR 3.22), malignancy (aOR 3.04), and heart diseases (aOR 2.15) were independently associated with a higher risk for severe Legionnaires' disease. However, patients with hyperlipidemia had a lower risk for severe outcome (aOR 0.17). Legionella positivity rate was 22% for 1,904 water samples collected. We found a higher positivity rate in summer months (28%-30%), which corroborated with months of highest rainfalls. Our novel finding that Legionnaires' disease patients with hyperlipidemia had a lower risk for severe outcome deserves further study to confirm the observation and ascertain the underlying reason.

Investigations on Contamination of Environmental Water Samples by Legionella using Real-Time Quantitative PCR Combined with Amoebic Co-Culturing.

We analyzed the contamination of environmental water samples with Legionella spp. using a conventional culture method, real-time quantitative PCR (qPCR), and real-time qPCR combined with an amoebic co-culture method. Samples (n = 110) were collected from 19 cooling towers, 31 amenity water facilities, and 60 river water sources of tap water in Japan. Legionella was detected in only three samples (3/110, 2.7%) using the culture method. The rate of Legionella detection using amoebic co-culture followed by qPCR was 74.5%, while that using qPCR without amoebic co-culture was 75.5%. A higher than 10-fold bacterial count was observed in 19 samples (19/110, 17.3%) using real-time qPCR subsequent to amoebic co-culture, compared with identical samples analyzed without co-culture. Of these 19 samples, 13 were identified as Legionella spp., including L. pneumophila and L. anisa, and the non-culturable species were identified as L. lytica and L. rowbothamii. This study showed that the detection of Legionella spp., even in those samples where they were not detected by the culture method, was possible using real-time qPCR and an amoebic co-culture method. In addition, this analytical test combination is a useful tool to detect viable and virulent Legionella spp..

Legionella pneumophila and Other Legionella Species Isolated from Legionellosis Patients in Japan between 2008 and 2016.

The Legionella Reference Center in Japan collected 427 Legionella clinical isolates between 2008 and 2016, including 7 representative isolates from corresponding outbreaks. The collection included 419 Legionella pneumophila isolates, of which 372 belonged to serogroup 1 (SG1) (87%) and the others belonged to SG2 to SG15 except for SG7 and SG11, and 8 isolates of other Legionella species (Legionella bozemanae, Legionella dumoffii, Legionella feeleii, Legionella longbeachae, Legionella londiniensis, and Legionella rubrilucens). L. pneumophila isolates were genotyped by sequence-based typing (SBT) and represented 187 sequence types (STs), of which 126 occurred in a single isolate (index of discrimination of 0.984). These STs were analyzed using minimum spanning tree analysis, resulting in the formation of 18 groups. The pattern of overall ST distribution among L. pneumophila isolates was diverse. In particular, some STs were frequently isolated and were suggested to be related to the infection sources. The major STs were ST23 (35 isolates), ST120 (20 isolates), and ST138 (16 isolates). ST23 was the most prevalent and most causative ST for outbreaks in Japan and Europe. ST138 has been observed only in Japan, where it has caused small-scale outbreaks; 81% of those strains (13 isolates) were suspected or confirmed to infect humans through bath water sources. On the other hand, 11 ST23 strains (31%) and 5 ST120 strains (25%) were suspected or confirmed to infect humans through bath water. These findings suggest that some ST strains frequently cause legionellosis in Japan and are found under different environmental conditions.IMPORTANCELegionella pneumophila serogroup 1 (SG1) is the most frequent cause of legionellosis. Our previous genetic analysis indicated that SG1 environmental isolates represented 8 major clonal complexes, consisting of 3 B groups, 2 C groups, and 3 S groups, which included major environmental isolates derived from bath water, cooling towers, and soil and puddles, respectively. Here, we surveyed clinical isolates collected from patients with legionellosis in Japan between 2008 and 2016. Most strains belonging to the B group were isolated from patients for whom bath water was the suspected or confirmed source of infection. Among the isolates derived from patients whose suspected infection source was soil or dust, most belonged to the S1 group and none belonged to the B or C groups. Additionally, the U group was discovered as a new group, which mainly included clinical isolates with unknown infection sources.

In vitro activity of various antibiotics against clinical strains of Legionella species isolated in Japan.

The activities of various antibiotics against 58 clinical isolates of Legionella species were evaluated using two methods, extracellular activity (minimum inhibitory concentration [MIC]) and intracellular activity. Susceptibility testing was performed using BSYEα agar. The minimum extracellular concentration inhibiting intracellular multiplication (MIEC) was determined using a human monocyte-derived cell line, THP-1. The most potent drugs in terms of MICs against clinical isolates were levofloxacin, garenoxacin, and rifampicin with MIC90 values of 0.015 µg/ml. The activities of ciprofloxacin, pazufloxacin, moxifloxacin, clarithromycin, and azithromycin were slightly higher than those of levofloxacin, garenoxacin, and rifampicin with an MIC90 of 0.03-0.06 µg/ml. Minocycline showed the highest activity, with an MIC90 of 1 µg/ml. No resistance against the antibiotics tested was detected. No difference was detected in the MIC distributions of the antibiotics tested between L. pneumophila serogroup 1 and L. pneumophila non-serogroup 1. The MIECs of ciprofloxacin, pazufloxacin, levofloxacin, moxifloxacin, garenoxacin, clarithromycin, and azithromycin were almost the same as their MICs, with MIEC90 values of 0.015-0.06 µg/ml, although the MIEC of minocycline was relatively lower and that of rifampicin was higher than their respective MICs. No difference was detected in the MIEC distributions of the antibiotics tested between L. pneumophila serogroup 1 and L. pneumophila non-serogroup 1. The ratios of MIEC:MIC for rifampicin (8) and pazufloxacin (2) were higher than those for levofloxacin (1), ciprofloxacin (1), moxifloxacin (1), garenoxacin (1), clarithromycin (1), and azithromycin (1). Our study showed that quinolones and macrolides had potent antimicrobial activity against both extracellular and intracellular Legionella species. The present data suggested the possible efficacy of these drugs in treatment of Legionella infections.

Evaluation of timing of re-appearance of VBNC Legionella for risk assessment in hospital water distribution systems.

BACKGROUND: In this study we estimated the presence of Legionella species, viable but non-culturable (VBNC), in hospital water networks. We also evaluated the time and load of Legionella appearance in samples found negative using the standard culture method. METHODS: A total of 42 samples was obtained from the tap water of five hospital buildings. The samples were tested for Legionella by the standard culture method and were monitored for up to 12 months for the appearance of VBNC Legionella. RESULTS: All the 42 samples were negative at the time of collection. Seven of the 42 samples (17.0%) became positive for Legionella at different times of monitoring. The time to the appearance of VBNC Legionella was extremely variable, from 15 days to 9 months from sampling. The most frequent Legionella species observed were Legionella spp and L. anisa and only in one sample L. pneumophila srg.1. CONCLUSIONS: Our study confirms the presence of VBNC Legionella in samples resulting negative using the standard culture method and highlights the different time to its appearance that can occur several months after sampling. The results are important for risk assessment and risk management of engineered water systems.

Prevalence of Legionella species isolated from shower water in public bath facilities in Toyama Prefecture, Japan.

AIMS: We investigated the prevalence of Legionella spp. isolated from shower water in public bath facilities in Toyama Prefecture, Japan. In addition, we analyzed the genetic diversity among Legionella pneumophila isolates from shower water as well as the genetic relationship between isolates from shower water and from stock strains previously analyzed from sputum specimens. METHODS: The isolates were characterized using serogrouping, 16S rRNA gene sequencing, and sequence-based typing. RESULTS: Legionella spp. were isolated from 31/91 (34.1%) samples derived from 17/37 (45.9%) bath facilities. Isolates from shower water and bath water in each public bath facility were serologically or genetically different, indicating that we need to isolate several L. pneumophila colonies from both bath and shower water to identify public bath facilities as sources of legionellosis. The 61 L. pneumophila isolates from shower water were classified into 39 sequence types (STs) (index of discrimination = 0.974), including 19 new STs. Among the 39 STs, 12 STs match clinical isolates in the European Working Group for Legionella Infections database. Notably, ST505 L. pneumophila SG 1, a strain frequently isolated from patients with legionellosis and from bath water in this area, was isolated from shower water. CONCLUSIONS: Pathogenic L. pneumophila strains including ST505 strain were widely distributed in shower water in public bath facilities, with genetic diversity showing several different origins. This study highlights the need to isolate several L. pneumophila colonies from both bath water and shower water to identify public bath facilities as infection sources in legionellosis cases.

Molecular diagnosis of Legionella infections--Clinical utility of front-line screening as part of a pneumonia diagnostic algorithm.

OBJECTIVES: Urinary antigen testing for Legionella pneumophila serogroup 1 is the leading rapid diagnostic test for Legionnaires' Disease (LD); however other Legionella species and serogroups can also cause LD. The aim was to determine the utility of front-line L. pneumophila and Legionella species PCR in a severe respiratory infection algorithm. METHODS: L. pneumophila and Legionella species duplex real-time PCR was carried out on 1944 specimens from hospitalised patients over a 4 year period in Edinburgh, UK. RESULTS: L. pneumophila was detected by PCR in 49 (2.7%) specimens from 36 patients. During a LD outbreak, combined L. pneumophila respiratory PCR and urinary antigen testing had optimal sensitivity and specificity (92.6% and 98.3% respectively) for the detection of confirmed cases. Legionella species was detected by PCR in 16 (0.9%) specimens from 10 patients. The 5 confirmed and 1 probable cases of Legionella longbeachae LD were both PCR and antibody positive. CONCLUSIONS: Front-line L. pneumophila and Legionella species PCR is a valuable addition to urinary antigen testing as part of a well-defined algorithm. Cases of LD due to L. longbeachae might be considered laboratory-confirmed when there is a positive Legionella species PCR result and detection of L. longbeachae specific antibody response.

Multiplex real-time PCR assay for Legionella species.

Legionella pneumophila serogroup 1 (sg1) accounts for the majority of infections in humans, but other Legionella species are also associated with human disease. In this study, a new SYBR Green I-based multiplex real-time PCR assay in a single reaction was developed to allow the rapid detection and differentiation of Legionella species by targeting specific gene sequences. Candidate target genes were selected, and primer sets were designed by referring to comparative genomic hybridization data of Legionella species. The Legionella species-specific groES primer set successfully detected all 30 Legionella strains tested. The xcpX and rfbA primers specifically detected L. pneumophila sg1-15 and L. pneumophila sg1, respectively. In addition, this assay was validated by testing clinical samples and isolates. In conclusion, this novel multiplex real-time PCR assay might be a useful diagnostic tool for the rapid detection and differentiation of Legionella species in both clinical and epidemiological studies.

Assessment of relative potential for Legionella species or surrogates inhalation exposure from common water uses.

The Legionella species have been identified as important waterborne pathogens in terms of disease morbidity and mortality. Microbial exposure assessment is a tool that can be utilized to assess the potential of Legionella species inhalation exposure from common water uses. The screening-level exposure assessment presented in this paper developed emission factors to model aerosolization, quantitatively assessed inhalation exposures of aerosolized Legionella species or Legionella species surrogates while evaluating two generalized levels of assumed water concentrations, and developed a relative ranking of six common in-home uses of water for potential Legionella species inhalation exposure. Considerable variability in the calculated exposure dose was identified between the six identified exposure pathways, with the doses differing by over five orders of magnitude in each of the evaluated exposure scenarios. The assessment of exposure pathways that have been epidemiologically associated with legionellosis transmission (ultrasonic and cool mist humidifiers) produced higher estimated inhalation exposure doses than pathways where epidemiological evidence of transmission has been less strong (faucet and shower) or absent (toilets and therapy pool). With consideration of the large uncertainties inherent in the exposure assessment process used, a relative ranking of exposure pathways from highest to lowest exposure doses was produced using culture-based measurement data and the assumption of constant water concentration across exposure pathways. In this ranking, the ultrasonic and cool mist humidifier exposure pathways were estimated to produce the highest exposure doses, followed by the shower and faucet exposure pathways, and then the toilet and therapy pool exposure pathways.

Identification of water pathogens by Raman microspectroscopy.

Legionella species can be found living in water mostly in a viable but nonculturable state or associated with protozoa and complex biofilm formations. Isolation and afterwards identification of these pathogens from environmental samples by using common identification procedures based on cultivation are extremely difficult and prolonged. The development of fast and sensitive method based on the cultivation free identification of bacteria is necessary. In this study Raman microspectroscopy combined with multiclass support vector machines have been used to discriminate between Legionella and other common aquatic bacteria, to distinguish among clinically relevant Legionella species and to classify unknown Raman spectra for a fast and reliable identification. Recorded Raman spectra of the twenty-two Legionella species as well as the Raman spectra of Escherichia coli, Klebsiella pneumoniae and Pseudomonas aeruginosa were utilized to build the classification model. Afterwards, independent Raman spectra of eleven species were used to identify them on the basis of the classification model that was created. The present study shows that Raman microspectroscopy can be used as a rapid and reliable method to distinguish between Legionella species recognized as human pathogens and to identify samples which are unknown to the model based on multiclass support vector machines (MC-SVM).

An Epidemiological Survey on Serological Diagnosis of Legionnella Infection in Seoul, Korea

Serological investigation of antibodies to Legionella species in 1,802 sera collected in seoul was conducted with indirect fluorescent antibody assay (IFA). With an antibody titer of > or =1:128 to be positive, 17 (0.9%) of these sera were positive and 6 (35.3%) of positive sera showed cross-reactions between Legionella species. The number of sera with antibody titers of > or =1:128 to L. pneumophila serogroup 1, L. pneumophila serogroup 4, L. pneumophila serogroup 5, L. bozemanii, L. micdadei, L. anisa were 6 (35.3%), 3 (17.6%), 3 (17.6%), 2 (11.8%), 1 (5.9%), 2 (11.8%) respectively. Among 17 positive sera, 10 (58.8%) sera were from male and 7 (41.2%) from female. An average age of them was 68.9 (+/-15.3; 27~89). Except for one serum, 16 (94.1%) of positive sera were from those older than 50 years old. The result suggests that the aged over 50 years old should be more careful of Legionella infection.

Clinical utility of a Legionella pneumophila urinary antigen test in a large university teaching hospital.

OBJECTIVE: To determine the clinical utility of diagnosing Legionella pneumonia by urinary antigen testing (LPUAT) in a low prevalence centre. DESIGN: The results of LPUATs were abstracted and analyzed from the authors' laboratory information system. Medical records were reviewed in detail for all positive tests and a random sample of 50 negative tests. SETTING: The Queen Elizabeth II Health Sciences Centre, a large university hospital complex. POPULATION STUDIED: Patients who were admitted from the emergency room with pneumonia or who had developed pneumonia in hospital and who had a LPUAT performed between April 1998 and October 2000. MAIN RESULTS: One thousand one hundred fifty-four tests were performed on 1007 patients. Seven patients had nine positive LPUATs. Three of these patients had confirmed Legionella pneumophila pneumonia. Three others had probable or possible L pneumophila pneumonia. There was one probable false positive. Six of the seven patients were already on empirical anti-L pneumophila therapy. Of the 50 negative tests reviewed in detail, 31 patients were on one of the antibiotics of choice for L pneumophila at the time the test was ordered; in 21 (68%) of these patients the negative result did not lead to a change in therapy. CONCLUSIONS: The cost to diagnose each case of Legionella pneumonia by LPUAT was approximately $5,770 and most patients were already on appropriate antibiotics. In patients with negative tests, antibiotics were often not changed in response to the test result. Rigorous screening of patients is required to increase pretest probability for LPUAT to be justified.