Legionellosis: Global Epidemiology and Current Perspectives on Diagnosis and Treatment

Introduction

Legionella, a Gram-negative intracellular parasite, exists in freshwater habitats. It was initially isolated and characterized as a “rickettsia-like” microorganism in 1947. However, it was not until 1977 that it was recognized as the causative pathogen, and the same serogroup, responsible for a serious pneumonia outbreak that occurred in Philadelphia, USA, in 1976.^1,2 This outbreak resulted in a total of 182 cases and 29 fatalities (a case fatality rate of 16%).³ Legionellosis, an acute human illness primarily caused by L. pneumophila, presents in two distinct clinical forms: Legionnaires’ disease (a severe, pneumonia-associated systemic illness) and Pontiac fever (a milder, self-limiting, non-pneumonic febrile illness). Legionnaires’ disease is characterized by high fever, cough, dyspnea, and radiographic evidence of pneumonia; it may progress rapidly to respiratory failure and multiorgan dysfunction. With appropriate antimicrobial therapy, the case fatality rate ranges from 10% to 15%.⁴ Pontiac fever resembles influenza-like illness and is characterized primarily by fever, myalgia, and headache. Pontiac fever is a nonlethal and self-limiting condition.⁵ To date, 58 species and 3 subspecies within the genus Legionella have been identified. Among them, L. pneumophila serotype 1 is the predominant pathogen responsible for Legionellosis, accounting for over 80% of reported cases.⁶ Since the 1976 outbreak in Philadelphia, an increasing number of Legionella infection cases associated with communities, healthcare facilities, and travel have been reported worldwide.⁷ According to the 2021 Global Burden of Disease (GBD) study, the global age-standardized DALY rate and mortality ratefor Legionella infections were 24.74 and 0.86 per 100,000 population, respectively; among adults aged 70 years and older, these rates exceeded 100 and 8 per 100,000, confirming a disproportionately high disease burden in the elderly.⁸ Meanwhile, antimicrobial resistance has been detected in environmental Legionella isolated from China and other countries. Given that Legionnaires’ disease frequently progresses to severe pneumonia, delayed or inadequate treatment poses a life-threatening risk, particularly among immunocompromised individuals.⁹ This paper presents a narrative review of current literature on Legionellosis. In light of its evolving epidemiology and the emerging threat of antimicrobial resistance, we integrate recent findings on disease burden, diagnostic challenges, and resistance mechanisms to inform clinical practice, guide public health interventions, and identify priorities for future research.

Microbiology and Pathogenicity

Legionella is an environmental pathogen. It is a small, Gram-negative, aerobic bacillus that does not form spores or capsules.¹⁰ Legionella proliferates in freshwater environments within a temperature range of 25 to 45 °C, with an optimal growth temperature of approximately 37 °C. When the temperature exceeds 42 °C, colony counts of all strains decline.¹¹ L. pneumophila, an aquatic bacterium, primarily replicates within protozoa (amoebae). Cysteine auxotrophy is a key phenotypic characteristics of Legionella pneumophila.^12,13 Additionally, L. pneumophila adheres to biotic and abiotic surfaces and forms structured, matrix-enclosed microbial communities known as biofilms, which enhance its environmental persistence.¹⁴ Accumulating evidence indicates that biofilms formation is strongly associated with Legionella transmission and outbreak occurrence. Biofilms augment bacterial virulence and impair host immune defenses through multiple mechanisms.^15–17

Legionella infections are primarily transmitted via inhalation of aerosols contaminated with the bacteria; person-to-person transmission has not been convincingly documented.^18,19 The pathogenesis of Legionellosis begins with bacterial attachment to host phagocytic cells, particularly alveolar macrophages, and subsequent intracellular replication. Upon entering the human respiratory tract, Legionella activates key virulence mechanisms: it establishes a specialized replicative niche known as the Legionella-containing vacuoles (LCV) and expresses genes encoding the type IV secretion system (T4SS), thereby evading lysosomal fusion and degradation, which facilitates persistent intracellular infection.²⁰

Among Legionella species, L. pneumophila is the predominant cause of human infection, accounting for 80–90% of reported cases globally, mainly serogroup 1 (Sg 1). Notably, Legionella longbeachae accounts for approximately 1% of global cases; however, in Australia and New Zealand, it causes 50–60% of locally acquired infections. An increasing number of Legionella longbeachae infections have also been reported in Europe.^21,22 Due to their relative rarity and the lack of diagnostic reagents, other subtypes such as L. pneumophila Sg 3 and Sg 6, L. bozemanii, and L. micdadei may cause disease but are rarely reported.^23,24

Clinical Manifestations

Legionnaires’ disease (LD) is an atypical pneumonia with an incubation period of 2–10 days.²⁵ Clinical manifestations depend on inhaled bacterial load, virulence factors, and host immune status.²⁶ Clinically, it is difficult to distinguish pneumonia caused by Legionella from that caused by common respiratory pathogens. Mild cases may only present with cough, whereas severe cases can progress rapidly to life-threatening pneumonia.²⁷ Pulmonary manifestations of Legionellosis are nonspecific and often subtle in the early disease phase. Common symptoms include fever > 38.8°C (67–100%), cough (41–92%), dyspnea (36–56%), and radiographic findings such as pulmonary consolidation or pleural effusion.^28–31 Some immunosuppressed patients (such as organ transplant or SLE patients) may develop lung abscesses and empyema.³²

Extrapulmonary manifestations of Legionellosis carry clinical and diagnostic importance. Gastrointestinal symptoms, including watery diarrhea (with or without abdominal pain), may present as isolated features of the disease.³³ Among central nervous system manifestations, experience headache accompanied by confusion is the most common presentation in patients with L. pneumonia infection.³⁴ The typical cardiac manifestation is pulse-temperature dissociation (such as relative bradycardia, Faget sign), which is uncommon in typical bacterial pneumonias.³⁵ Legionella has been implicated in culture-negative endocarditis; however, this remains a rare complication.³⁶ Myalgia frequently accompanies fever and chills but is typically mild. Severe myalgia should raise suspicion for influenza-related diagnosis.³⁷ Additionally, non-specific laboratory abnormalities can be helpful, such as hyponatremia, mildly elevated liver enzyme levels, unexplained microscopic hematuria, and abnormally elevated serum creatinine.^38,39 Beekman et al validated a Legionella prediction scoring system based on six items admission criteria, which aids in the early identification of Legionella pneumonia in clinical practice.⁴⁰

Pontiac fever is a nonfatal, self-limiting, influenza-like illness associated with exposure to Legionella. The first outbreak caused by Legionella was reported in Pontiac, Michigan, in 1968, affecting at least 144 individuals. Its incubation period is 24–48 hours, and symptoms typically resolve within 2–5 days; the illness is characterized primarily by fever, myalgia, and headache. Antibiotic treatment is generally unnecessary.^41,42

Epidemiology

Legionella strains capable of causing human infection are widely distributed in natural water sources and anthropogenic environments. However, the global incidence of Legionellosis remains poorly characterized. Due to the large differences in disease recognition, diagnostic capacity, and surveillance infrastructure, reported case counts are subject to significant underascertainment. Consequently, epidemiological interpretations of Legionellosis incidence data require cautious interpretation.⁴³

Globally, the reported incidence of Legionnaires’ disease continues to rise in countries with robust surveillance systems (Table 1). In the United States, the age-standardized average incidence increased from 0.48 cases per 100,000 population during 1992–2002 to 2.71 cases per 100,000 in 2018.⁴⁴ In Canada, 1401 cases of Legionellosis were reported between 1978 and 2006, corresponding to a crude average incidence rate of 0.41 cases per 100,000 person-years.⁴⁵ In all EU and European Economic Area countries, Legionellosis is a notifiable disease. The European Centre for Disease Prevention and Control (ECDC) established the European Legionellosis Surveillance Network (ELDSNet) in 2010. Reported Legionellosis incidence across Europe ranges from 0.1 to 30 cases per million population, reflecting substantial intercountry variation.⁴⁶ Most reported data on Legionella infections in Europe originate from Western European countries. France, Germany, Italy, and Spain collectively account for 70% of notified cases, despite representing only 50% of the total EU population. Spain ranks among the European countries with the highest reported incidence of Legionellosis. A study analyzing 13,472 samples from 465 hospitals in Spain reported a Legionella detection rate was 65.4%.⁴⁷ Notably, following the French government’s 1987 mandate requiring case notification, reported Legionellosis cases rose steadily, peaking in 2005 at 2.5 cases per 100,000 residents, which is twice the average incidence in Europe. Subsequently, France implemented comprehensive source-control measures, leading to a decline in incidence to 2.0 cases per 100,000 residents by 2008.^48,49 The Netherlands launched the National Legionella Outbreak Detection Project (NLODP) in 2002. Between 2002 and 2012, 1991 confirmed legionellosis cases were reported, and 1484 trace back investigations were conducted, yielding a positive rate of 24.7%.⁵⁰ In addition, the United Kingdom, Portugal, Switzerland, Norway, Greece, etc. have all reported cases of Legionella infection.^51–55 Legionellosis incidence in Australia (2.1 per 100,000) exceeds the national average (1.5 per 100,000);⁵⁶ New Zealand reports an annual incidence of 2.7 per 100,000.⁵⁷ In Asia, the incidence rates in Japan, South Korea and Singapore are 0.2–0.7, 0.74, 0.28 cases per 100,000 people, respectively.^58–60 In 1982, the first domestically acquired case of Legionella infection in China was discovered in Nanjing. Subsequently, numerous Legionella infection-related cases were reported, mostly sporadic.⁶¹ Since there is no mandatory requirement to report Legionella infection cases in China, data on Legionella infection remain limited. In Hong Kong, China, Legionellosis has been a notifiable infectious disease since 1994. The incidence of Legionella is 0.9 cases per 100,000 people, showing an upward trend year by year.⁶²

Table 1 Global Epidemiological Characteristics of Legionnaires’ Disease

The composition of the susceptible population for Legionellosis is evolving along with changes in the global population structure. High-risk factors for Legionnaires infection include men over 50 years old, smokers, and individuals with underlying diseases such as diabetes, cancer, or immunosuppression; however, anyone can develop Legionellosis.^63–65 With the accelerated global aging process and the increasing number of immunocompromised individuals, the size of the susceptible population for Legionellosis continues to expand, presenting new challenges for disease prevention and control. In addition, Legionellosis shows a seasonal pattern, with summer and early autumn being the most common times of year for infection. Research has shown that warm and humid climates are associated with an increased incidence of Legionellosis.^66,67 This feature suggests that global climate change may further affect the spatiotemporal distribution of this disease.

Legionellosis develops as sporadic cases or as outbreaks. According to the sources of infection, legionellosis is usually classified into community-acquired (CALD), travel-associated (TALD), or healthcare-associated (HALD). In recent years, the number of TALD cases, and ship-associated events, occur repeatedly, making them a key focus of monitoring.⁶⁸ Hotel-associated cases are typically associated with hotel cooling towers and/or potable water systems, while ship-related cases are most frequently linked to hot tubs.⁶⁹ Consistent with European data, most L. pneumophila strains isolated from German patients belong to clones that have emerged worldwide.⁷⁰ In Singapore, the proportion of imported cases has increased from 6.2% to 27.3%, usually associated with overnight stays in public places.⁶⁰ Differences exist in the distribution of infection sources across countries. In Japan, Public baths are the main source of infection.⁷¹ Australia and New Zealand have reported an increase in Long Beach Legionella infections associated with potted plants and compost.⁷²

Globally, L. pneumophila is the primary pathogen, with regional variations in serotypes and genotypes. In Valencia, Spain, ST1, ST578, and ST23 were the most prevalent sequence types among 1088 samples.⁷³ In France, L. pneumophila serogroup 1 accounts for 95.4% of clinical isolates but only 28.2% of environmental isolates. The most common non-L. pneumophila species, L. anisa, is more prevalent in the environment (13.8%) than in clinical samples (0.8%).⁷⁴ A 2002–2017 study in northeastern Italy tested 18,104 water samples and reported a 19.8% Legionella positivity rate; 50.8% of 63 medical institutions yielded positive results. L. pneumophila Sg 2–15 dominated hospitals (>74%), whereas Sg 1 predominated in swimming pools (82%). ST23 was the most common genotype in Italy.^75,76 Among 76,200 warm-water samples from 24 southern German regions, 15,300 (20.07%) were Legionella-positive. L. pneumophila constituted 84% of isolates, with marked regional variations.⁷⁷ Furthermore, the disease burden caused by Legionella longbeachae is particularly prominent in countries in the Southern Hemisphere.⁷² L. longbeachae is prominent: 59% in Australia,⁷⁸ 51.0% in New Zealand,⁵⁷ exceeding L. pneumophila and showing marked regional difference, suggesting significant differences in the composition of pathogen spectra across regions. In Japan, L. pneumophila is predominant, with common sequence types including ST23, ST120 and ST138.⁷¹ In South Korea, L. pneumophila is predominant, accounting for 85%.⁵⁹ In Thailand, L. pneumophila serogroup 1 accounts for 2–5% of urban cases. By contrast, it is rare in rural areas, where Legionella longbeachae is more common.⁷⁹ In China, a study detected Legionella in 22.43% (129/575) of water samples and 9.28% (41/442) of soil samples, with L. pneumophila accounting for 75%.⁸⁰

The global epidemiology of Legionellosis shows significant regional heterogeneity, and its true disease burden is shaped by the interplay of multiple factors, including surveillance systems, diagnostic capacity, demographic structure, climatic conditions, and pathogen distribution. While current data reflect the epidemiological trends in some regions, their comparability and completeness at a global level still require systematic improvement. Despite revealing the epidemiological characteristics of certain areas, existing data face substantial challenges in global comparability and completeness, highlighting an urgent need for further integration and optimization.

Diagnosis

The diagnosis of Legionellosis is based on a combination of clinical and radiological features as well as laboratory tests. Currently, the detection methods for Legionella include serology and antibody testing, bacterial culture, urinary antigen testing, and nucleic acid amplification testing.^81,82 Every method has its own merits and drawbacks (Table 2).

Table 2 Legionella Diagnostic Methods and Limitations

Serological testing for Legionella infection is a valuable epidemiological tool. A variety of serological detection techniques, such as indirect immunofluorescence assay (IFA), enzyme immunoassay (EIA), and microagglutination testing, are used to detect Legionella pathogens, with a sensitivity of 78% −90%.^83–85 Since the antibodies identified in serological diagnosis are usually a mixture of immunoglobulin A (IgA), M (IgM), and G (IgG), testing for all three antibody classes is recommended to maximize sensitivity.⁸⁶ Specific IgM antibodies are unreliable biomarkers for acute infection. Because IgM antibodies may persist for a long time and seroconversion often take several weeks, this delayed kinetics represents the main limitation of serological testing. In most cases, a four-fold increase in antibody titer is observed within 3–4 weeks.⁸⁷ Obtaining samples too early can lead to false-negative results. In practice, clinicians are encouraged to simultaneously test serum samples during the acute phase and 3 weeks after the onset of the disease.⁸⁸ Data suggest that even the best commercial assays have a positive predictive value of only about 50%, especially in regions where L. pneumophila serogroup 1 infections are less common.^89,90

Detecting soluble Legionella antigen in urine samples is a rapid method for diagnosing early Legionella infections and is also a useful tool for investigating Legionellosis outbreaks. This method was first reported shortly after the 1976 epidemic but was not widely accepted as a routine diagnostic method or incorporated into the international case definition for confirmed cases until the mid-1990s.⁹¹ It detects the lipopolysaccharide component of the cell wall of L. pneumophila. Antigen is typically detectable in urine within a few days of disease onset and persists for several days to weeks. Currently, urinary antigen testing accounts for 70–80% of diagnosed cases in Europe and the United States, with a sensitivity of up to 80%-90%. However, This method may miss up to 40% of cases caused by non-serogroup 1 L. pneumophila.^92,93 Japan’s Ribotest Legionella (2019) enables broader detection and may improve early diagnosis and prognosis for non-serogroup 1 infections, though its efficacy requires further validation. Notably, prolonged antigen excretion in immunocompromised or critically ill patients can yield false-positive results, whereas approximately 8% of patients may test false negative due to absence of urinary antigen excretion. A negative result does not exclude Legionella infection, and repeat testing is recommended when clinically indicated.^84,94,95

The cultivation and isolation of Legionella from clinical specimens remains the gold standard for diagnosis.⁹⁶ Studies have shown that Legionella can be isolated from 66% of cases. If samples are collected within 2 days of admission, Legionella can be isolated from 80% of cases.⁹⁷ However, many factors limit the sensitivity of culture. First, experienced laboratories are required for the isolation and cultivation of Legionella. A survey by the College of American Pathologists showed that up to two-thirds of microbiology laboratories in the United States are unable to cultivate pure L. pneumophila isolates. Similarly, in China, the number of such laboratories is extremely limited.⁹⁸ Second, the use of selective agar and sample pretreatment (heat or acid) is not straightforward. Given the relatively low prevalence of Legionellosis, these methods lack sufficient cost-effectiveness, and culture requires 3–5 days.⁹⁹ In addition, to improve the specificity of pneumonia diagnosis caused by pyogenic bacteria, laboratories usually reject sputum specimens containing squamous epithelial cells or few polymorphonuclear leukocytes. Some patients with Legionellosis produce little sputum or non-purulent sputum, and it is crucial to culture such specimens immediately.^100,101

Polymerase chain reaction (PCR) is a molecular method capable of detecting all known Legionella infections. It is characterized by high specificity, sensitivity, and rapid turnaround time. The application of PCR overcomes limitations related to pathogen viability and species identification, with a specificity of 95%-100%. However, PCR is not ideal for detecting non-respiratory samples such as urine and serum. In addition, this method makes it difficult to evaluate the activity of pathogenic bacteria (after exposure to antibiotics).^102,103 Moreover, metagenomic next-generation sequencing (mNGS), as a revolutionary technology, is applicable to critically ill patients with rapidly progressive Legionella infections and concomitant infections, especially providing clinical guidance for the management of immunosuppressed patients. However, due to its high cost and the lack of a standardized interpretation standard, mNGS has not yet been adopted as a routine clinical diagnostic tool.¹⁰⁴

Advances in detection technology have improved Legionella diagnostics. Novel biomarkers such as ribosomal proteins L7/L12 and interleukin-17A (IL-17A) show strong diagnostic potential.¹⁰⁵ CRISPR-Cas-based detection enables strip-based readout within 30 minutes with single-copy sensitivity, making it highly suitable for emergency and primary care settings.¹⁰⁶ Loop-mediated isothermal amplification (LAMP) offers excellent sensitivity and holds promising potential for portable applications.¹⁰⁷ Targeted next-generation sequencing (tNGS) combines high sensitivity, high throughput, and low cost and may further enhance diagnostic accuracy through integration with dynamic primers.¹⁰⁸

Antibacterial Treatment

The high mortality rate of Legionellosis without timely treatment underscores the clinical priority principle: early diagnosis and prompt, effective antibiotic therapy, along with management of complications such as respiratory failure, hepatic and renal dysfunction, and neurological involvement.¹⁰⁹ Legionella is an intracellular pathogen residing in tissues and alveolar macrophages, which means that anti-Legionella drugs should accumulate and be biologically active within cells.¹¹⁰ The IDSA/ATS guidelines recommend that, when treating bacterial community-acquired pneumonia, antibiotics covering atypical pathogens should be considered. For Legionella, macrolides (such as azithromycin, clarithromycin) and quinolones (such as levofloxacin, moxifloxacin) are recommended as first-line treatments, and other drugs such as doxycycline, rifampicin, and compound sulfamethoxazole.^111,112 New-generation antibiotics —including sitafloxacin (a novel fluoroquinolone), omadacycline (a new-generation tetracycline), and lefamulin (a pleuromutilin)—exhibit potent antibacterial activity and broaden the available therapeutic options¹¹³ (Table 3).

Table 3 Stratified Severity‑Based Treatment for Legionnaires’ Disease

However, there are no prospective randomized trials comparing the clinical outcomes of patients after treatment with levofloxacin and azithromycin.¹¹⁴ A meta-analysis compared the efficacy of fluoroquinolones and macrolides as monotherapy for L. pneumonia. A total of 21 studies involving 3,525 patients were included. No significant differences were observed between the two groups in clinical cure rate, time to defervescence, length of hospital stay, and recovery from complications. Moreover, no significant difference in mortality reduction was found between patients treated with fluoroquinolones and macrolides (6.9% vs 7.4%; pooled OR = 0.94, 95% CI: 0.71–1.25; P = 0.66). Owing to patient heterogeneity, subgroup analyses stratified by patient population, disease severity, and ICU admission status could not be performed.¹¹⁵ Kato et al’s meta-analysis compared the effectiveness and safety of the two treatments. The results showed that the cure rates of the two groups were comparable, but the overall and 30-day mortality rates of macrolides were higher than those of fluoroquinolones. Compared with macrolides, fluoroquinolones significantly shortened hospital stay in patients with Legionella pneumonia.¹¹⁶ In addition, when a meta-analysis subgroup pooled data on severe Legionella pneumonia, it was found that mortality was significantly lower among patients treated with fluoroquinolones alone than among those treated with macrolides (72.8% vs 30.8%, p = 0.027), whereas length of hospital stay and complication rates were comparable.¹¹⁷ Currently, the included studies are mainly observational and lack randomized controlled trials; moreover, clinical cure is not among the reported outcomes. This suggests that the available evidence cannot robustly determine whether fluoroquinolones are superior to macrolides, and meaningful comparisons are limited to mortality outcomes. Therefore, in the absence of contraindications, fluoroquinolones may be a preferred choice for reducing mortality, particularly in critically ill patients.¹¹⁸

The evidence supporting the benefits of combination therapy for Legionella infection is limited. Monotherapy with azithromycin or fluoroquinolones remains the standard treatment for patients with Legionella infection¹¹⁹ Chahin et al recommend combination therapy with fluoroquinolones and azithromycin in patients with severe pneumonia, especially those with significant comorbidities and immunocompromised hosts who have failed conventional treatment.¹²⁰ It should be noted that combination therapy increases the risk of adverse reactions, such as Q-T interval prolongation.^121,122 For confirmed cases of moderate to severe Legionella pneumonia, treatment with levofloxacin or azithromycin for 7 to 10 days is recommended. For immunocompromised patients, the treatment course should be extended to 21 days.^123,124

Antimicrobial resistance in Legionella represents an emerging yet underappreciated threat to clinical management and public health. Although traditionally considered uniformly susceptible to macrolides and fluoroquinolones, evidence from environmental surveillance and clinical isolates reveals a more complex resistance landscape.¹²⁵ In China, Jia et al found that among 149 L. pneumophila serogroup 1 strains, 25 environmental isolates were resistant to azithromycin, yielding a resistance rate of 16.78% (25/149). Upregulation of the efflux pump lpeAB gene was associated with reduced azithromycin susceptibility in these 25 strains.¹²⁶ The resistance-nodulation-division family (RND family) efflux pump LpeAB, the most well-studied resistance determinant in L. pneumophila, actively extrudes azithromycin, leading to a 2–4-fold MIC increase, and has been widely investigated in environmental isolates.¹²⁷ Polish researchers isolated a non-serogroup 1 L. pneumophila strain resistant to azithromycin from the water supply system of a sanatorium; it also showed reduced sensitivity to ciprofloxacin and rifampicin.¹²⁸ Minetti et al studied 107 L. pneumophila serogroup 1 strains from clinical and related environmental sources in Portugal. Among them, 12 isolates had an azithromycin minimum inhibitory concentration (MIC) exceeding the provisional highest wild-type (WT) MIC by the European Committee on Antimicrobial Susceptibility Testing (EUCAST). Nine of these isolates exhibited negative lpeAB expression, implying the existence of other uncharacterized mechanisms underlying drug resistance. Some isolates showed a reduced sensitivity to fluoroquinolones, including 4 isolates with MIC values above the breakpoints of all tested antibiotics; however, sequencing of the gyrA and parC gene revealed no mutations in the quinolone resistance-determining regions.¹²⁹ Environmental Legionella exhibit diverse resistance mechanisms. Although LpeAB-mediated low-level resistance is common, it does not explain elevated MICs in clinical strains, implying unknown links in resistance evolution from environment to clinical settings. Water environment may be an important reservoir of drug resistance genes of Legionella, posing a potential public health risk. Monitoring antimicrobial resistance in environmental isolates has significant predictive value for forecasting clinical risks. Meanwhile, pathogens can develop resistance during clinical treatment, and antibiotic exposure can accelerate the emergence of resistance. In 2014, Bruin et al first isolated ciprofloxacin-resistant L. pneumophila from clinical specimens. The resistance mechanism may be related to a mutation in the gyrA gene (amino acid position 83). Although no fluoroquinolone (FQ) resistance breakpoint has yet been established for Legionella, this isolate was regarded as FQ-resistant due to the poor clinical response to antibiotic therapy and prolonged hospital stay.¹³⁰ Shadoud et al used next-generation sequencing to monitor the proportion of gyrA(83) mutant strains in two Legionnaires’ disease patients treated with fluoroquinolones. The mutant proportion increased markedly from 1.05% at baseline to 94% post-treatment. Fluoroquinolone resistance in Legionella is primarily mediated by mutations in the quinolone resistance-determining regions (QRDR) of gyrA/gyrB (DNA gyrase) and parC/parE (topoisomerase IV). The gyrA T83I mutation (248C→T) is the most clinically relevant, conferring an 8‑fold increase in ciprofloxacin MIC.¹³¹ This suggests that extremely low-frequency resistant mutants preexist in wild-type populations, which rapidly expand to become dominant under antibiotic selection pressure, ultimately resulting in treatment failure. We reported 3 patients with confirmed severe L. pneumophila pneumonia who did not respond to initial treatment with moxifloxacin. We attempted to culture L. pneumophila from the patients to evaluate its resistance to conventional antibiotics such as moxifloxacin, but this was unsuccessful due to their critical condition and laboratory limitations. We speculated that the occurrence of bacterial resistance may be related to the repeated moxifloxacin use in patients with recurrent infections or with subtherapeutic moxifloxacin concentrations resulting from renal insufficiency and hypoalbuminemia.¹³² Conventional susceptibility testing relies on culture, which is often unavailable due to clinical or laboratory constraints, leading to underdiagnosed resistance. An integrated environment–clinical network is thus needed to clarify resistance gene transmission, guiding regional empirical therapy, delaying resistance, and improving outcomes.

Conclusions and Outlook

Over the past 50 years, substantial progress has been made in understanding Legionella biology and Legionellosis; however, critical knowledge gaps remain. Future priorities include epidemiological studies to better define disease burden across populations, standardized diagnostics capable of detecting non-L. pneumophila species, and the use of lung microbiota sequencing to inform novel prevention and control strategies. Optimal treatment regimens remain uncertain, with insufficient evidence to guide monotherapy versus combination therapy. However, conducting such trials is challenging due to limitations including clinical operations and inadequate sample size. There is an urgent need to establish an integrated environment–clinical network to monitor resistant Legionella strains and their role in therapeutic failure, with direct implications for antimicrobial stewardship and resistance monitoring. Addressing these priorities will be essential to advancing both clinical management and public health strategies for Legionellosis.