Water Safety & Public Health Reference
Legionella pneumophila
The threat, ecology, historical emergence, environmentally friendly control and standard laboratory detection of Legionnaires' disease.
Chapter 01
Background
Legionella pneumophila is a gram-negative, aerobic, rod-shaped bacterium (roughly 1–2 µm) and the causative agent of Legionnaires' disease — a severe and potentially fatal pneumonia — and of the milder, self-limiting Pontiac fever.
More than 50 Legionella species are recognised; L. pneumophila (especially serogroup 1) accounts for the large majority of human cases. Legionella is naturally present at low concentrations in surface water and moist soil, but it becomes a public-health threat when it colonises man-made water systems where warmth, stagnation, scale, and biofilm create conditions favourable for explosive growth. The Philadelphia epidemic of July 1976 first brought the bacterium to public attention.
↑ ContentsChapter 02
The threat: disease, dangers and negative effects
Two distinct illnesses. Infection with Legionella (legionellosis) presents either as Legionnaires' disease — a pneumonia with fever, cough, breathlessness, muscle aches, and sometimes confusion or gastrointestinal symptoms — or as Pontiac fever, a milder flu-like illness without pneumonia that usually resolves in a few days. Legionnaires' disease needs prompt antibiotic treatment; Pontiac fever generally does not.
Mortality. Case-fatality is commonly cited around 5–10% for community-acquired disease (about 9% of reported cases are fatal), but rises sharply in vulnerable and healthcare settings. Healthcare-associated infection has reported case-fatality rates around 25% and historically as high as ~40–46%, and mortality in severe cases among immunocompromised patients can approach 40%. Hospital-acquired infection carries a higher death rate than community-acquired infection.
Burden and trend. In the EU/EEA, 29 countries reported 15,362 cases in 2024 (a notification rate of 3.4 per 100,000), continuing a rising trend; about 9% of cases with a known outcome were fatal. Reported figures substantially understate true incidence — estimated to be roughly 1.8–2.7× higher — because symptoms are non-specific and testing is inconsistent. Cases skew strongly toward men aged 65 and older.
Wider impact. Beyond illness and death, outbreaks impose heavy costs: emergency shutdowns of buildings and industrial plants, remediation, litigation, and reputational damage. The 1976 outbreak famously forced the closure of the Bellevue-Stratford Hotel, and modern cooling-tower outbreaks routinely trigger regulatory action.
↑ ContentsChapter 03
Growth characteristics
| Optimal growth temperature | 32–35 °C |
| Growth range | 20–45 °C (proliferation slows sharply below 20 °C and above ~50 °C) |
| Inactivation temperature | > 60 °C (rapid kill; effectively immediate above 70 °C) |
| Doubling time | ~2 hours under optimal conditions |
| Primary infection route | Inhalation (or aspiration) of contaminated aerosols; not spread person-to-person |
| High-risk groups | Adults ≥50 years, immunocompromised and transplant patients, chronic lung disease, current/former smokers |
| Survival strategy | Protected within biofilms and inside free-living amoebae (intracellular replication) |
Danger zone 20–45 °C. Control keeps water out of the danger zone: keep hot water hot (≥ 60 °C) and cold water cold (≤ 20 °C).
↑ ContentsChapter 04
Where Legionella grows — ecology and water systems
Legionella persists through two linked survival strategies that also explain why it is so hard to eradicate from plumbing:
- Biofilms. Legionella grows within the slime layers coating pipes, tanks, and fittings. Biofilm provides nutrients and physically shields the bacteria from heat and disinfectants, and continuously sheds cells back into the water.
- Amoebae (intracellular replication). Legionella is engulfed by free-living amoebae (e.g. Acanthamoeba) but resists digestion and multiplies inside them. Amoeba-grown Legionella — and cells released in amoebal vesicles — are dramatically more resistant to chlorination and other biocides, and amoebal cysts survive desiccation and temperature extremes. Cells may also enter a viable-but-non-culturable (VBNC) state that evades both disinfection and standard testing.
Because both mechanisms depend on biofilm, biofilm is the master variable in engineered-system risk. Sediment, scale, rust, and materials such as iron, zinc and aluminium all favour colonisation.
High-risk water systems and environments
Colonisation plus aerosol generation together create risk. Common high-risk sources include cooling towers, evaporative condensers, showers and showerheads, taps and faucets, hospitals, hotels and resorts, office buildings, whirlpools and jacuzzis, public swimming pools, nursing homes, schools and universities, and water storage tanks. Other recognised aerosol sources include decorative fountains, misters and humidifiers, vehicle windscreen-washer fluid, and respiratory or hydrotherapy equipment.
↑ ContentsChapter 05
Historical view: how the threat came to attention
In July 1976, during the United States bicentennial, the Pennsylvania American Legion held its annual convention at the Bellevue-Stratford Hotel in Philadelphia. Within days of the convention (21–24 July), attendees began falling ill with a severe pneumonia. The first death, on 27 July, was initially attributed to a heart attack; within a week a cluster of sudden deaths among Legionnaires had emerged. A Bloomsburg physician, Dr Ernest Campbell, noticed that three sick patients shared only one thing — they had all attended the convention — and alerted the Pennsylvania Department of Health.
The episode became a national medical mystery and one of the most publicised epidemics the CDC had ever investigated. In all, about 221 people fell ill and roughly 29–34 died (sources differ). The cause eluded investigators for months because the organism does not grow on standard media and stains poorly. In January 1977, CDC scientists Joseph McDade and Charles Shepard isolated the bacterium, named Legionella pneumophila after the Legionnaires. The hotel's rooftop cooling tower was implicated as the source of the contaminated aerosol.
With the organism in hand, investigators retrospectively solved earlier mysteries: a 1968 outbreak of a milder illness in Pontiac, Michigan ('Pontiac fever'), a 1965 hospital outbreak in Washington DC, and cases dating to the late 1950s. The discovery established waterborne aerosol transmission from building systems as a recognised hazard and drove the cooling-tower, water-safety-plan and building-water-management regulations in force today. Later landmarks — the 1999 Bovenkarspel flower-show outbreak in the Netherlands and the 2015 South Bronx cooling-tower outbreak in New York — reinforced regulation.
| Year | Location | Cases / deaths | Significance |
|---|---|---|---|
| 1957 | Austin, Minnesota (meat plant) | — | Retrospectively identified outbreak |
| 1965 | St. Elizabeths Hospital, Washington DC | ~81 / 14 | Later attributed to Legionella |
| 1968 | Pontiac, Michigan (health dept.) | ~144 / 0 | Milder form later named 'Pontiac fever' |
| Jul 1976 | Bellevue-Stratford Hotel, Philadelphia | 221 / ~29–34 | First recognised cluster; brought the disease to world attention |
| Jan 1977 | CDC laboratory | — | Causative bacterium isolated and named L. pneumophila |
| 1999 | Flower show, Bovenkarspel, Netherlands | ~188+ / 21+ | Whirlpool-spa display; landmark European outbreak |
| 2015 | South Bronx, New York City | 138 / 16 | Cooling-tower outbreak; drove NYC cooling-tower legislation |
Chapter 06
Conventional control methods and their limitations
Traditional controls reduce Legionella but each has well-documented limitations — most rooted in the same problem: biofilm and amoebal hosts shield the bacteria.
| Method | Principle and key limitations |
|---|---|
| Thermal shock (65–70 °C) | Raises water to lethal temperature. Energy-intensive; hard to deliver to dead-legs/distal outlets; biofilm and amoebal hosts insulate bacteria; scald risk. |
| Continuous chlorination (3–4 mg/L) | Maintains a disinfectant residual. Suppresses rather than eliminates; poor biofilm penetration; forms chlorinated by-products; corrosive to some materials. |
| UV-C irradiation (254 nm) | Inactivates planktonic bacteria at the install point. No downstream residual; cannot penetrate biofilm or amoebal cysts; effectiveness drops with turbidity/scaling. |
| Chemical shock (NaOCl, H₂O₂, ClO₂) | Periodic high-dose dosing. Costly and disruptive; disinfection by-products; corrosive at effective concentrations; biofilm/amoeba-protected cells survive and re-seed. |
Chapter 07
Physical treatment approach: targeting the biofilm
Biofilm formation is the primary mechanism enabling Legionella colonisation in engineered water systems: it provides nutrients, shelters bacteria from heat and chemicals, harbours the amoebal hosts in which Legionella multiplies, and continuously re-seeds the water. Removing the biofilm substrate therefore addresses the root cause rather than only the free-floating (planktonic) cells that conventional methods reach most easily.
By eliminating the biofilm substrate and combining several innovative, low-chemical techniques — controlled hydraulic/mechanical biofilm removal, temperature and flow management, and targeted physical disinfection — the environment that lets Legionella survive can be disrupted without heavy reliance on corrosive biocides and their by-products. This aligns with the modern water-safety-plan philosophy promoted by the ECDC and WHO: manage the system continuously rather than react to positive samples.
Prevention is the key to an effective solution. A reactive response to Legionella is almost always more costly and disruptive than a continuous physical-prevention strategy built around biofilm control.
Supporting research: peer-reviewed studies on biofilm and disinfection resistance (e.g. PubMed identifiers 19854466 and 19962336) and the European ECDC Legionnaires' disease guidance provide the evidence base for a biofilm-centred, prevention-first approach.
↑ ContentsChapter 08
Environmentally friendly physical-prevention techniques in detail
These techniques share one philosophy: reduce reliance on corrosive, by-product-forming biocides, and instead deny Legionella its habitat (biofilm, stagnation, and the 20–45 °C danger zone). In practice they are layered within a Water Safety Plan, and the operators' rule of thumb still governs: keep the hot water hot (≥ 60 °C), the cold water cold (≤ 20 °C), and the water moving.
1. Biofilm prevention and removal (the foundation)
Because biofilm both shelters Legionella and harbours the amoebae in which it multiplies, preventing biofilm is the most effective long-term lever. Eco-friendly measures remove the conditions biofilm needs rather than dosing poisons:
- Flow and stagnation management — routine flushing of little-used outlets, removing redundant pipework and dead-legs, and right-sizing tanks. Flushing measurably lowers cell counts, deposits, and Legionella at taps; stagnation (as in buildings closed during COVID-19) predictably raises risk.
- Temperature discipline — storing hot water ≥ 60 °C, delivering it above ~55 °C, and keeping cold water below 20 °C holds the system outside the growth range and suppresses biofilm without chemicals.
- Scale, sediment and corrosion control — removing calcium scale, rust and sludge from tanks, calorifiers and heat exchangers eliminates sheltered micro-habitats and the iron/zinc surfaces that favour colonisation.
- Material selection and design — choosing materials that support less biofilm (e.g. PEX/PVC rated for hot water; avoiding nutrient-leaching hoses) and designing for turbulent flow rather than dead zones.
- Hydraulic / mechanical biofilm removal — periodic high-velocity flushing, pigging or engineered scouring physically strips the biofilm layer that chemical dosing cannot penetrate.
2. Copper–silver ionisation (CSI)
How it works. Water passes through a chamber with copper and silver electrodes across which a low-voltage current is applied (electrolysis). Positively charged copper ions bind to the negatively charged bacterial cell wall and disrupt permeability and nutrient uptake, letting silver ions enter and denature proteins and nucleic acids, causing cell death. Generated in-situ and leaving a metal-ion residual rather than chlorinated by-products, CSI sits among the lower-impact, longer-acting techniques.
Dosing & evidence. Typical targets are ~0.2–0.4 mg/L copper and 0.02–0.04 mg/L silver. A five-year study of four complex drinking-water systems and a cooling tower found dosing near 400 µg/L copper and 40 µg/L silver controlled Legionella at all sites, with only occasional temporary re-occurrence (~3.8% of samples). Reviews conclude CSI is highly effective when systems are correctly designed, operated and maintained.
Advantages. Easy to install; persistent residual that reaches distal outlets and resists early re-colonisation; unlike thermal and UV methods, efficacy is not lost at high water temperature; no chlorination by-products.
Limitations. Efficacy depends on maintaining ion levels and on water chemistry — high pH (above ~8) markedly reduces the effect, and conductivity, hardness, chloride and organic carbon all matter. Silver is capped by drinking-water regulations in many countries (e.g. historically ≤ 10 µg/L in Germany); WHO tolerances are ~2 mg/L copper and 0.1 mg/L silver. Silver resistance and re-growth can develop, so CSI works best within a layered, monitored programme.
3. Point-of-use (POU) filtration
Sterile-grade 0.2 µm filters on taps and showers physically remove Legionella at the last few centimetres before the outlet, protecting vulnerable users even when the wider system is colonised. They add no chemicals and act as an immediate barrier during outbreaks — especially at dead-legs and distal points disinfectants struggle to reach. Filters have a limited life and must be changed on schedule, so they complement rather than replace system-wide biofilm control.
4. UV-C and other point barriers
UV-C (254 nm) at strategic points inactivates planktonic Legionella passing through, with no residual chemical and no by-products. Its intrinsic weakness — no downstream residual and no biofilm penetration — means it is deployed as a barrier on incoming or recirculated water in combination with biofilm control, not as a standalone cure.
5. Emerging biological / low-chemical methods
- Predatory bacteria (e.g. Bdellovibrio bacteriovorus) — a 'living antibiotic' that invades and lyses gram-negative bacteria and penetrates biofilms; lab studies report large reductions in preformed biofilm (~4-log).
- Bacteriophages and enzymes — host-specific viruses and biofilm-degrading enzymes that target Legionella or dissolve the biofilm matrix with minimal ecological footprint.
- Beneficial / competitive microbiomes — managing the wider microbial community (and amoebal hosts) so Legionella is out-competed rather than chemically suppressed.
Comparative overview
| Technique | Acts on | Residual / reach | Main limitation |
|---|---|---|---|
| Biofilm prevention (flow, temp., cleaning, materials) | Root cause — habitat, stagnation, scale | Whole system (design-led) | Needs sustained discipline & good design |
| Copper–silver ionisation | Planktonic + biofilm cells (ion contact) | Persistent residual to distal outlets | pH-sensitive; silver limits; possible resistance |
| Point-of-use filtration | Cells at the outlet | Local barrier only | Limited life; must be replaced on schedule |
| UV-C (254 nm) | Planktonic cells at install point | No downstream residual | No biofilm penetration; needs clear water |
| Predatory bacteria / phages (emerging) | Biofilm matrix & gram-negative cells | Potentially self-propagating | Not yet routine for potable systems |
No single technique suffices alone. The environmentally sound, cost-effective strategy layers biofilm prevention as the foundation, adds a persistent barrier such as copper–silver ionisation and point-of-use filtration where residual protection is needed, and reserves chemical shock treatment for emergencies.
↑ ContentsChapter 09
Standard detection & analysis of Legionella (WAC/V/A/005)
Prevention has to be verified. A water-safety programme is only as strong as the sampling and laboratory analysis that confirm whether Legionella is actually under control. In Flanders (Belgium) the accredited reference procedure for detecting and quantifying Legionella in water is WAC/V/A/005, built on the international standard ISO 11731:2017. This chapter summarises how a sample travels from bottle to reported result, so that facility managers can understand what a Legionella figure means and how it is produced. (The version summarised here is a draft dated April 2026; consult the current published method for authoritative detail.)
Scope and principle
The method covers two accreditation subdomains — drinking water and wastewater — and, importantly, cooling-tower water is always classified as wastewater regardless of its feed water. Samples are assigned to a water matrix: matrix A is drinking water; matrices B and C are cooling-tower waters of increasing interfering-flora load; wastewater from manure-processing plants (WZI/BB) follows a derivative of the matrix-C method. Every result must be serologically typed into three categories: Legionella pneumophila serotype 1; L. pneumophila serotype 2–14 (2–15); and Legionella species (non-pneumophila). Because the organism needs the amino acid L-cysteine to grow, cysteine dependence is the key confirmation test.
The procedure follows the four-step ISO 11731 decision matrix: Step 1 classify the water matrix (A/B/C); Step 2 choose the analysis scenario (direct plating and/or membrane filtration with glass-bead elution, or a filter-on-plate variant); Step 3 apply treatments that suppress competing flora (heat, acid, or both); and Step 4 culture on the prescribed selective media (BCYE, BCYE+AB, GVPC or MWY).
The A/B/C analysis scenarios
| Matrix / Scenario | Water type | Concentration & plating | Fractions | Media |
|---|---|---|---|---|
| A · Scenario A | Drinking water (low interfering flora) | Direct plating (R) and membrane filtration 0.2 µm + glass-bead elution, ≥250 ml (pref. 500–1000) | O · H · Z | BCYE + selective (BCYE+AB / MWY / GVPC) |
| B · Scenario B | Cooling-tower water (high flora) | Direct plating (R) + MF 0.2 µm + elution, target 250 ml | O · H · Z | MWY or GVPC |
| C · Scenario C | Cooling-tower water (extreme flora) | Direct plating (R) + MF 0.45 µm + elution, target 250 ml (≤2 filters) | H · Z · H&Z | MWY or GVPC |
| WZI/BB (§7) | Manure-plant wastewater | Direct plating only (no MF); ultrasonic pre-treatment | Z · TZ (+ 1:100, 1:1000) | ≥3 GVPC plates |
An alternative MPN method — Legiolert — is approved specifically for L. pneumophila in drinking water: an enzyme-substrate test in which a brown well is positive, read within 7 days via a Quanti-Tray/MPN table. Non-pneumophila species must still be determined by ISO 11731 culture. Exception A1 allows membrane filtration with the filter placed directly on the plate for very low counts and low interfering flora (and for heavily chlorinated pool water, or as an additional method).
Sample concentration and treatments
Membrane filtration (polycarbonate or PES, 0.2 or 0.45 µm) is followed by glass-bead elution (5–10 ml diluent, vortex ≥2 min or ultrasonic), after which the eluate is split into three portions. Heat treatment is 50 ± 1 °C for 30 ± 2 min; acid treatment mixes one volume of sample with nine volumes of acid buffer for 5.0 ± 0.5 min — a built-in 1:10 dilution that must be accounted for when calculating the result. The direct filter-on-plate route uses a cellulose-nitrate/ester filter with a minimum filtered volume of 10 ml.
Culture, confirmation and identification
Each fraction is inoculated at 0.1–0.2 ml onto the prescribed media and incubated at 36 ± 2 °C for 10 ± 1 days in a humid atmosphere. Plates are read first on day 3, 4 or 5 (to catch overgrowth) and again at the end of incubation, preferably under a stereomicroscope with oblique lighting. Legionella colonies are typically white-grey with an entire edge and a characteristic ground-glass appearance; L. pneumophila often shows a green-pink edge.
Under UV light many species autofluoresce bright blue-white (and may be reported directly as non-pneumophila species), whereas L. pneumophila is dull green and does not fluoresce. Cysteine dependence is confirmed by subculturing at least six presumptive colonies (≥2 per treatment and morphology) onto BCYE(+cys) and BCYE−cys: growth on the cysteine plate but not on the cysteine-free plate confirms Legionella. Serotyping then follows by agglutination or immunochromatographic assay, and MALDI-TOF MS may be used for identification (verified per ISO 16140).
Special cases
- WZI/BB wastewater (Section 7). Based on matrix C but by direct plating only, because membrane filtration is unusable. It requires ultrasonic pre-treatment (40–45 kHz, 9 min), thorough homogenisation before each step, acid (Z) and heat-then-acid (TZ) treatments, dilutions to 1:100 and 1:1000, and at least three GVPC plates.
- Real-time PCR / qPCR (Section 8). Approved only as a screening tool for L. pneumophila in drinking water within a management plan, per ISO/TS 12869, targeting the mip gene. PCR-positive samples must then be quantified by culture; after thermal-shock disinfection, sampling is done at least 24 h later. It does not replace culture.
- Outbreaks (Section 9). An early indicative result is reported to Departement Zorg (mandatory notification), then incubation continues to 10 ± 1 days with re-confirmation. Screening limited to L. pneumophila is not applicable here, since non-pneumophila species can also cause disease.
Reporting
Results are expressed in CFU/L (kve/l) for each of the three categories, reporting the highest converted value and accounting for dilution. Rules govern when to report from direct plating versus membrane filtration (a plate with ≥3 colonies per species — a converted value ≥30,000 CFU/L — is reported from direct plating), how to handle just 1–2 colonies, and the reporting limit (ISO 11731 Table J.1). For cooling-tower water the total Legionella is the sum of the three categories. Overgrown plates may never be reported as “not detected”.
Abbreviations used in this chapter
| Abbrev. | Meaning | Abbrev. | Meaning |
|---|---|---|---|
| R | Direct plating (NL: rechtstreekse uitplating) | MWY | Modified Wadowsky–Yee agar (selective) |
| MF | Membrane filtration (+ glass-bead elution) | MPN | Most Probable Number (Legiolert) |
| O / H / Z | Untreated / heat-treated / acid-treated fraction | CFU/L | Colony-forming units per litre (kve/l) |
| H&Z / TZ | Combined heat-then-acid treatment | UV | Ultraviolet autofluorescence screening |
| Lp | Legionella pneumophila | MALDI-TOF MS | Mass-spectrometry identification |
| BCYE | Buffered Charcoal Yeast Extract agar | qPCR | Quantitative real-time PCR |
| BCYE(+cys)/−cys | BCYE with / without L-cysteine | WZI/BB | Wastewater-treatment plant / biological basin |
| BCYE+AB | BCYE with antibiotics (selective) | dw / aw | Drinking water / wastewater subdomains |
| GVPC | Glycine/Vancomycin/Polymyxin B/Cycloheximide agar | ISO / WAC | Int'l Standards Org. / Flemish Water Analysis Compendium |
Source & standards. Departement Omgeving (Vlaamse overheid). WAC/V/A/005 — Bepaling van Legionella in drinkwater en afvalwater (“Determination of Legionella in drinking water and wastewater”), in the Compendium voor de monsterneming, meting en analyse van water (WAC), draft version April 2026, based on ISO 11731:2017.
Related standards: ISO 11731:2017 · ISO 8199:2018 · ISO 16140-1/-2:2016 · ISO/TS 12869:2019 · NF T90-471 · Legiolert (IDEXX) NF cert. IDX 33/06-06/19 · ASTM D8429-21.
Chapter 10
Key message
Identify risks. Control water. Protect people.
Legionella thrives across many water systems; the effective strategy is to identify every potential source, monitor it, keep water out of the 20–45 °C danger zone, and — above all — deny the bacteria the biofilm habitat on which their survival depends.
Keep hot water ≥ 60 °C · keep cold water ≤ 20 °C · eliminate stagnation and biofilm.
Chapter 11
References
All sources retrieved July 2026. Some journal links may require institutional access for full text; abstracts and cited figures are publicly visible. The standard-method sources for Chapter 09 are listed in that chapter's source box.
- CDC — About Legionnaires' disease. cdc.gov/legionella/about
- CDC — Legionellosis surveillance and trends. cdc.gov/legionella/php/surveillance
- Soda, E. A. et al. (2017). Vital Signs: Health Care–Associated Legionnaires' Disease. MMWR 66(22):584–589. cdc.gov/mmwr
- ECDC — Legionnaires' disease, Annual Epidemiological Report 2024. ecdc.europa.eu
- ECDC — Legionnaires' disease AER 2024 (PDF). ecdc.europa.eu (PDF)
- The Lancet Microbe (2024). Global surge of Legionnaires' disease in 2024. thelancet.com
- Beauté, J. et al. (2020). Healthcare-Associated Legionnaires' Disease, Europe, 2008–2017. Emerg. Infect. Dis. 26(10). wwwnc.cdc.gov/eid
- 1976 Philadelphia Legionnaires' disease outbreak (timeline). en.wikipedia.org
- The Conversation (2025). Legionnaires' disease outbreak in Philadelphia in 1976 — 50 years later. theconversation.com
- Foege, W. H. (CDC). The Changing Priorities of the CDC (early Legionella outbreaks 1965–1978). stacks.cdc.gov (PDF)
- Abdel-Nour, M. et al. / Frontiers (2018). Factors Mediating Environmental Biofilm Formation by L. pneumophila. frontiersin.org
- Frontiers (2017). From Many Hosts, One Accidental Pathogen: Protozoan Hosts of Legionella. frontiersin.org
- PMC (2021). Intracellular Behaviour of Legionella within amoebae (Willaertia magna). pmc.ncbi.nlm.nih.gov
- WHO — Legionellosis fact sheet. who.int
- ECDC — Legionnaires' disease topic hub. ecdc.europa.eu
- van der Kooij, D. et al. (2016). Efficacy of copper-silver ionisation in complex water systems and a cooling tower. sciencedirect.com
- LeChevallier, M. W. (2023). Examining the efficacy of copper-silver ionization for management of Legionella. AWWA Water Science e1327. awwa.onlinelibrary.wiley.com
- Lin, Y. E. et al. (2002). Negative effect of high pH on biocidal efficacy of copper and silver ions. Appl. Environ. Microbiol. pmc.ncbi.nlm.nih.gov
- Cachafeiro, S. P. et al. (2007). Is copper–silver ionisation safe and effective in controlling legionella? J. Hosp. Infect. sciencedirect.com
- U.S. EPA (2016). Technologies for Legionella Control in Premise Plumbing Systems. epa.gov (PDF)
- National Academies (2019). Management of Legionella in Water Systems. ncbi.nlm.nih.gov/books
- CDC — About Water Management Programs. cdc.gov/control-legionella
- Kadouri, D. & O'Toole, G. A. (2005). Susceptibility of biofilms to Bdellovibrio bacteriovorus attack. Appl. Environ. Microbiol. pmc.ncbi.nlm.nih.gov
- Departement Omgeving (Vlaamse overheid). WAC/V/A/005 — Bepaling van Legionella in drinkwater en afvalwater. Compendium voor de monsterneming, meting en analyse van water (WAC), draft, April 2026; based on ISO 11731:2017.