From how often one thinks about and tests for Legionella, one might have expected it to appear in the exam papers sooner. Certainly other rarer (tetanus) and more exotic or tropical (leptospirosis) have been the subject of SAQs, and so it was only a matter of time before this classical ICU organism had made its way into the CICM exams as the subject of Question 13 from the second paper of 2023. The college wanted to know about both L.pneumphila and L.longbeachae, which seemed like a lot (two organisms in one SAQ?).

Microbiology of Legionella species

These rods are

  • Gram-negative
  • Strictly aerobic
  • Non-spore forming
  • Unencapsulated
  • Catalase positive
  • Urease negative

Legionella are legion. Jomehzadeh et al (2019) mention that there are about thirty species that have been cultured from patients, let alone in the environment, and so it suddenly no longer seems like such a burden to have to focus on two common organisms (L.pneumophila and L.longbeachae). Water is their main reservoir, and only L.longbeachae is an exception, as it likes to live in organic potting mix where the heat of decomposition sustains a temperature and moisture content favourable for growth. So, from the perspective of Legionella, your lungs are basically compost, or some kind of welcoming warm pond. Under normal circumstances, these rods cannot reproduce in the extracellular environment, and would usually make their way into a protozoan where they live intracellularly as parasites.  The preferred host is probably some kind of freshwater amoeba like a Hartmanella or Naegleria , as that is where free-range Legionella tend to hang out, though if pushed they will also colonise some kind of random ciliate. The relationship with the protozoan is strictly parasitic and the bacterial cells end up in a phagosome which they disable and subvert for their own personal needs, continuing to reproduce until they ultimately kill the host. And where the amoeba content of the normal human lung is thankfully very small, Legionella gladly invade and parasitise human macrophages, as the next most amoeba-like thing. 

Is there much difference between Legionella pneumophila, the beast that got its name from the convention of the American Legion in Philadelphia (1976), and Legionella longbeachae, discovered a year later from with pneumonia in Long Beach, California, USA?

Diagnosis of Legionella

The way this is likely to play out in CICM exams is via the vehicle of the "outline your assessment" question, which typically requires a "history, examination, investigations" sort of structure. The "history" would surely contain some of the risk factors and the modes of transmission, and the investigations would yield some characteristic biochemistry findings like hyponatremia, but the examination would be fairly vanilla, as nothing about the physical features shouts "Legionella" at the examiner. The money is in the history and bloods.

Risk factors for Legionella

Risk factors (from Jomehzadeh et al, as well as the CDC website) include:

  • Age ≥50 years
  • Smoking
  • Alcoholism
  • COPD
  • Immune suppression, eg. organ transplant
  • Malignancy
  • Diabetes, renal failure, or hepatic failure
  • Recent travel with an overnight stay outside of the home
  • Recent care at a healthcare facility
  • Exposure to hot tubs

That's right, the CDC website specifically implicates hot tubs. In the absence of hot tubs, Legionella will reluctantly continue to party in other bodies of water, including manmade ones, such as showerheads, decorative water features, aquariums, cooling systems, and water heaters. The main reason the healthcare facilities or "travel with overnight stay" are implicated is because these facilities typically have that magical combination of lax maintenance and large complex water distribution/cooling/heating systems. Those water systems also help these bacteria gain access to your lungs by facilitating the generation of aerosols, which is the main mode of transmission. Inhalation of aerosolised water is the main mechanism, but you can also aspirate the Legionella-rich water directly into your lungs if you happen to be immersed in it, and there is at least one case of person-to-person transmission (involving a lot of coughing and a small poorly ventilated room). 

This watery route is mostly a reference to the transmission of L.pneumophila. The other major species in Australia, L.longbeachaeis mostly acquired from compost, potting mix and manure, where there are plenty of amoebae to parasitise. Those soil amoebae often also have the delightful ability to form protective cysts, which therefore also protect the Legionella, and allow both organisms to survive dessication and storage. Thus, as most normal people do not aerosolise their compost, they can still inhale these bacteria while gardening, as dried dusty particles of potting mix are an essential part of the hobby.  Thus,

Modes of transmission of Legionella

for L.pneumophila:

  • Aerosol transmission
  • Mostly from contaminated water sources
  • Sources in the built environment include:
    • Water heaters/coolers
    • Humidifiers
    • Air conditioners and their ducts
    • Showers, plumbing fixtures, water features

for L.longbeachae:

  • Aerosol/dust transmission
  • Usually not found in water sources
  • Most commonly associated with exposure to
    • Potting mix
    • Garden soil
    • Compost

It is apparently a rather reluctant human pathogen and the CDC website mentions that only about 5% of people who are exposed to a source will become infected.

Legionellosis and Pontiac Fever

There are two main syndromes which can present with distinct patterns of illness:

  • Legionellosis is an atypical pneumonia hanging off the side of a much larger multiorgan system failure problem, with a mortality of around 10% (about the same as severe community acquired pneumonia)
  • Pontiac fever is a short mild flu-like illness with little respiratory involvement and similarly little ICU involvement

Legionellosis proper takes about 14 days to incubate, whereas Pontiac fever only takes about 60 hours to get started.

"Clearly, this is classical Legionella"

It is generally said that one cannot make distinction between Legionella and any other sort of pneumonia, but a few general points can be made:

  • It's an "atypical" pneumonia, so the constitutional symptoms are dominant
  • The cough is usually unproductive
  • Diarrhoea vomiting and myalgias make the presentation almost viral-sounding
  • Hyponatremia, weirdly, and more so than other causes of pneumonia
  • Deranged LFTs and impaired renal function

Which brings us to:

Biochemistry and serology for Legionella

Agulló-Ortuño et al (2006) took about 170 patients with Legionella infection and compared them to non-Legionella pneumonias. The biochemical differences were:

  • LFT derangement
  • Hyponatremia
  • Hypoalbuminaemia
  • A modestly elevated WCC

But these are generic and one would not commit to any diagnosis on the basis of a suspicious history and vaguely supportive biochemistry. Proper laboratory diagnosis of Legionella consists of:

  • Urinary antigen is cheap and effective. The molecule being tested is the specific lipopolysaccharide of L.pneumophila which is apparent in the urine in the first days of infection and which has 100% specificity in some studies. Problem is, L.longbeachae is not detected by this method.
  • PCR is the next best option, as both L.pneumophila and L.longbeachae will both be detectable this way, and the sensitivity/specificity attributed to this test is usually very high. It can also be performed on tissue samples and various random body fluids.  
  • Serology. Sick and convalescent plasma may be compared for maximum effect, but even a single wildly elevated antibody titer can be informative. Paired serology has a 99% specificity and 90% sensitivity.
  • Culture. Legionella grows slowly, and the results may take up to seven days, but it can be cultured - on special media impregnated with antibiotics that would wipe out competing organisms. The specificity is 100%, but sensitivity can be as low as 20%, because this fragile organism may simply refused to grow.

Question 13 from the second paper of 2023 called for "four methods for diagnosis" which was a simple list-em-up for only two marks, meaning that all the extra detail was probably not necessary. It is not inconceivable that at some point in the future, somebody may ask about the advantages and disadvantages of these tests, and then they would be able to refer to this excellent table in the CDC page. For most people it is enough to remember that the urinary antigen does not detect fully 50% of the total Legionella burden in the Australian population.

Antimicrobial therapy for Legionella

So, what might you treat this with? Normal thinking would trend towards something that treats intracellular pathogens, like a macrolide or a fluoroquinolone. That is indeed what eTG recommends for the Australian Legionella sufferer:

  • Azithromycin
  • Ciprofloxacin
  • Doxycycline

And these recommendations mostly just upgrade from oral to IV for disease of increased severity, with a 7-10 day duration of therapy. The Sanford Guide and UpToDate do something similar, though UpToDate authors recommend not to rely on tetracyclines for L.longbeachae as it is apparently resistant, and Sanford people stand out with the range of alternatives and helpful tidbits (eg. where they point out that there is no additive effect from using dual therapy). 

Prevention of Legionella infection

Well. This is not a transmissible disease that requires isolation, as you literally need to be breathing the same uncirculated air as the patient for a number of whole days before you get infected. However, as the usual sources of infection are highly visited public places, notification of the authorities becomes an essential part of the management process. Thus:

  • Prevention of transmission to staff
    • No need for isolation
    • Standard contact precautions apply
  • Prevention of transmission in the community
    • Notification to public health officials to help identify source and trace other potentially exposed individuals
  • Prevention of future infection or outbreaks
    • For L.longbeachae, instructions re. safe handling of potting mix/gardening soil (gloves, handwashing, masks, taking care not to inhale dust and to minimise exposure to aerosol)
    • For L.pneumophila, control of water supply contamination is the most important element, which moves well into the territory of civil engineering and well out of the domain of CICM, eg:
      • Chlorination of water
      • Regular audit and regulation for maintenance of water systems
      • Copper-silver ionisation
    • Information for susceptible individuals who have greater risk (eg. people on long term steroids) to help them mitigate their personal exposure risk


Edelstein, Paul H., and Christian Lück. "Legionella." Manual of clinical microbiology (2015): 887-904.

Chambers, Stephen T., et al. "Legionellosis caused by non-Legionella pneumophila species, with a focus on Legionella longbeachae." Microorganisms 9.2 (2021): 291.

Jomehzadeh, Nabi, et al. "Legionella and legionnaires' disease: An overview." Journal of Acute Disease 8.6 (2019): 221-232.

Lindsay, D. S. J., et al. "Legionella longbeachae serogroup 1 infections linked to potting compost." Journal of medical microbiology 61.2 (2012): 218-222.

Nisar, Muhammad Atif, et al. "Legionella pneumophila and protozoan hosts: implications for the control of hospital and potable water systems.Pathogens 9.4 (2020): 286.

McKINNEY, ROGER M., et al. "Legionella longbeachae species nova, another etiologic agent of human pneumonia." Annals of internal medicine 94.6 (1981): 739-743.

Agulló-Ortuño, Ma Teresa, et al. "Biochemical and immunologic features of an outbreak of Legionnaires disease: comparative study between community-acquired pneumonias." Diagnostic microbiology and infectious disease 56.1 (2006): 7-11.