Eradication is the wrong word, and chasing it quietly burns a lot of budget. You can knock Legionella back hard inside a water system, and there are moments you must. What you cannot do is sterilise a building’s pipework and expect it to stay that way. The reason coats the inside wall of almost every pipe you own: biofilm.

Biofilm is the thin, living layer of bacteria, slime and trapped debris that forms on any wetted surface. Legionella does not drift about in open water waiting to be flushed away. It shelters inside that film, frequently inside other microorganisms living there, and the film is what survives a treatment and reseeds the system a few weeks later. So talk of Legionella eradication is really talk of how deeply and how durably you can suppress it, not whether you can wipe it out. If you have ever watched a system come back positive after a clean disinfection report, this is the explanation.

Why the film wins

Free-floating Legionella cells are easy to kill. The ones living in mature biofilm are not, and the gap between those two situations is where most disappointment lives.

Three things make the film stubborn. First, the slime itself: the bacteria sit in a self-made matrix of sticky polymers that slows the diffusion of disinfectant and dilutes it before it reaches the cells deepest in the layer, so by the time chlorine has worked through the outer slime and the scale packed into it, much of its strength is spent. Second, Legionella is rarely alone in there; it survives and multiplies inside protozoa and amoebae grazing in the film, armoured hosts that ride out conditions a lone bacterium could not. Third, some cells deep in a thick film sit slow-growing and barely metabolising, giving treatments that rely on hitting active bacteria little to grab.

Layer onto that the conditions that grow the film in the first place. CDC points to biofilm, warm water, stagnation and low disinfectant as the cluster of conditions that let Legionella multiply [1]. As a general guide reflected in HSE guidance, the bacteria grow fastest in roughly the 20-45C band and are progressively killed above about 60C [2]. A long, lukewarm, low-use pipe run is therefore not just a place Legionella might appear; it is a place biofilm is being actively cultivated, complete with the scale and sediment that anchor it. HSE’s own list of higher-risk features reads like a list of good biofilm habitat: dead legs, infrequently used outlets, stored water and components that hold debris [3].

Picture the pipe wall in cross-section

If an image would normally sit here, this is what it would show. Sketch it as you read.

  • Start at the pipe wall and move inward toward the water.
  • Against the metal or plastic sits a thin conditioning layer of organic material, deposited within hours of first flow. Nothing grows without it; it is the primer coat.
  • On top of that, raised mounds of bacteria embed themselves in a clear, gel-like matrix of polymers they secrete. These are the microcolonies, and the matrix is the “slime” people picture, except it is structured rather than uniform.
  • Threading between the mounds are open water channels. The film is not a solid skin but more like a city with streets, carrying nutrients in and waste out, which is part of why it is so durable.
  • Tucked inside the mounds are the amoebae and protozoa, and inside some of them, Legionella. This is the protected core: disinfectant and heat must cross the matrix, then the host’s outer membrane, to reach it.
  • At the surface, the film constantly sheds. Clumps slough off and travel downstream, seeding fresh growth at the next quiet bend, fitting or shower hose, the shower head being a classic destination with its warmth and dead-end geometry.

Hold that picture and the failures of one-shot treatment stop being a mystery. You are not cleaning a surface. You are trying to reach the protected core of a structured, self-repairing layer that regrows from whatever you leave behind.

Why disinfection knocks it back but doesn’t clear it

Shock disinfection and thermal disinfection both work in a limited sense: they cut the count sharply and buy time, but neither reliably removes the film. A high chlorine dose strips the surface and kills exposed cells, yet penetration into the deep matrix is incomplete and survivors regrow from the channels and the protected hosts. Thermal pasteurisation hits the same wall: heat must reach every fitting at temperature for long enough, and in a real building the far outlet, the insulated dead leg and the scaled hose are where it underperforms.

This is why a negative sample taken days after a disinfection can be genuine and still mislead: it describes the planktonic water at that outlet at that moment, not the state of the film. HSE is clear that testing frequency should follow the system and the risk assessment rather than being treated as proof in itself [4]. Many duty holders eventually recognise the rebound pattern, clean at four weeks and positive again by the next review, because the conditions that built the film were never changed. Why bacteria in biofilm shrug off doses that would kill them in open water is worth its own read in Chlorine tolerance and disinfectant limits in Legionella control.

In my view, the single most useful mental shift is to stop scoring a disinfection on the post-treatment sample and start scoring it on whether the conditions that grew the film have changed.

What actually keeps the film thin

Since you cannot win by treatment alone, the realistic goal is a system that gives biofilm a poor home. CDC frames control around exactly that: deny the warmth, stagnation and low disinfectant the bacteria need [5], and the supporting routine is the familiar L8 and HSG274 chain of risk assessment, written scheme, monitoring and review [2][6]. Translated to the surfaces inside your pipes:

  • Keep hot water hot and cold water cold, so most of the system sits outside the growth band. Temperature has the broadest reach; the practical side of hitting it is in Temperature control basics.
  • Keep water moving. Turnover starves the film; stagnation does the opposite, which is why low-use outlets reward attention out of all proportion to their size.
  • Remove the habitat by design, not by repeated flushing. Cutting out a genuine dead leg once beats flushing it forever and permanently lowers the biofilm load. The stagnation case is made in full in Neglected water systems.
  • Clean and descale the components that harbour debris: scale and sediment are both food and anchor for the film, and they shield it from whatever you dose.

Do those consistently and disinfection becomes what it should be: an occasional reset on a system already held in check, not a recurring rescue.

Where this guidance stops

Biofilm behaves differently depending on your pipe materials, water chemistry, temperatures and how the building is used, so treat everything above as background for a competent, site-specific risk assessment, not a treatment recipe. Disinfection methods, doses, contact times and pasteurisation regimes carry real safety and material-compatibility risks, and must be specified by someone competent for your system. Nothing here is clinical advice, and none of it replaces the judgement of your responsible person.

Common questions

If we can’t fully eradicate it, what is disinfection actually for?

It resets a system that has drifted, cutting the bacterial count sharply and buying time. Treat it as managing an acute problem, not curing a chronic one: the cure, such as it is, is the ongoing control that keeps biofilm thin between resets. Without that follow-through, a disinfection buys only a few quiet weeks before the film regrows.

Does a clean sample after disinfection mean the biofilm is gone?

No. A negative result describes the free water at one outlet at one moment; it says little about the film clinging to the pipe wall or hiding inside protozoa. Rebound positives weeks or months later are common precisely because the structure survived. Use sampling as supporting evidence for the conditions sampled, not as a clearance certificate [4].

Would a stronger dose or a hotter flush finally clear it for good?

Rarely, and there is a cost. The matrix slows penetration, the protected hosts shield the cells inside them, and slow-growing bacteria deep in the film give heat and chemicals little to act on. Pushing doses or temperatures ever higher tends to damage seals, washers and pipework, shedding debris that biofilm then recolonises. Sustained control of temperature, flow and cleanliness does more than any heroic single treatment.

Next step

Pull your most recent disinfection report and your outlet temperature and usage records, and lay them side by side. Mark every outlet that is warm, low-use, scaled or on a long run; those are where the film has the best home and where a clean post-treatment sample means the least. For each one, decide whether the honest fix is another treatment or a change to the conditions. That short exercise usually shows the money is better spent on design and turnover than on the next round of dosing.

Sources

[1] CDC, “How Legionella Spreads”. https://www.cdc.gov/legionella/causes/index.html [2] HSE, “Legionnaires’ disease. The control of legionella bacteria in water systems - Approved Code of Practice and guidance (L8)”. https://www.hse.gov.uk/pubns/books/l8.htm [3] HSE, “Systems most likely to create legionella risk”. https://www.hse.gov.uk/legionnaires/risk-systems.htm [4] HSE, “Testing and monitoring your water system for legionella”. https://www.hse.gov.uk/legionnaires/testing-monitoring-water-system.htm [5] CDC, “Controlling Legionella”. https://www.cdc.gov/control-legionella/index.html [6] HSE, “Legionnaires’ disease: Technical guidance (HSG274)”. https://www.hse.gov.uk/pubns/books/hsg274.htm