March 2021
The LCA requires its Members to have in place management procedures and appropriately competent staff, that give them the capability of responding appropriately to positive legionella results. Positive legionella results following sampling are one area where it’s possible to misinterpret the meaning and end up with what could at best be inappropriate expense and at worst, a dangerous situation and risk to life.
Sample Planning
Interpretation of results starts with a suitable sample plan. If inappropriate samples are taken, results will be difficult to interpret meaningfully. Samples can be taken pre or post flush and with or without disinfection of the outlet. The selection of location, type of sample and method of analysis chosen forms part of the sample plan.
For populations with elevated susceptibility, such as in some healthcare settings, or for evaporative cooling systems there is some specific guidance on what to take and where. In many other situations there is no specific guidance that indicates samples are required and the need must be driven by risk assessment.
Where risk assessment identifies a risk of legionella growth within a water system and the need to monitor using legionella samples, the first priority should be to consider measures to eliminate or reduce the risk before designing monitoring measures. HSG274 Part 2 paragraphs 2.120 to 2.125 give useful advice on where and why to sample. Where it is impossible or not reasonably practicable to reduce the risk and the risk assessment recommends monitoring for the presence of legionella by sampling, there are a number of factors to consider in a sample plan. The points below give some further considerations for hot and cold water systems sample plans:
- How often should samples be taken?
o Legionella is relatively slow growing but may develop to hazardous levels within days or weeks in ideal conditions
o The HSE guidance in HSG274 Part 2 details weekly samples for legionella where controls, such as temperature or biocide, are not being consistently achieved
- Where should samples be taken?
o Samples should be checking areas where legionella is likely to grow – deadlegs, underused outlets, areas where controls have failed or areas where risk is identified for some other reason
- How should samples be taken?
o For systems otherwise under control this method should generally be pre-flush without disinfection to examine the risk that exists to the users of the system from the outlets
o In some cases, it would also be appropriate to sample post-flush, with disinfection of the outlet to examine how the system is performing as a whole
o There is limited value in sampling post-flush downstream of mixing valves or sampling post-flush without disinfecting the outlet although sometimes this is useful in investigation work
- How should samples be analysed:
o There are many options available, and all have advantages and disadvantages
o Culture, MPN, PCR, IMS, ICA, rapid protein ID, etc. will all give either a count or a yes/no
o For the purposes of the laboratory section below we are assuming culture is to be used
o The option selected must be appropriate for the question being asked by the sample plan
The Laboratory Process
Positive legionella results occur in around 10 – 20% of laboratory water samples that are analysed using the ISO11731 culture method. The method takes 10 days to complete and may take a further 2 days to confirm with traditional confirmation techniques. Rapid confirmations techniques can remove the confirmation step through Protein identification, such as MALDI-TOF.
The ISO states that samples must be read at several points throughout the 10 days; most laboratories do this around days 4 and 7 before a final read on day 10. Any of these reads can result in a presumptive positive. This is where Legionella bacteria are suspected of being present on the agar plate. The laboratory will run their confirmation process to confirm a colony is the suspect bacteria, in this case Legionella. This confirmation can be rapid in the case of MALDI-TOF, or it can take up-to 2 days using the traditional approach.
The laboratory will normally send a positive notification once the result in confirmed. At this point the laboratory has confirmed that the bacteria present on the culture plate is Legionella; they may also offer you an early count in Colony Forming Units (CFU) in the volume of sample submitted; normally a litre; giving results in CFU/L. If this confirmation is sent on an interim read, on days 4 or 7 in the example above, then you should be aware that the count could increase in the final read.
What does the LCA expect of Registered Members when a Positive Result is Confirmed?
This section relates to ANY species of Legionella, the HSE and DoH guidance documents do not discriminate between sero-groups or species of Legionella and nor should your actions.
The LCA would expect that any registered member would have a procedure to notify their client, log the result, track actions, note resolution and close out a positive legionella result. Each of those stages may have numerous steps.
Notifying a Positive Legionella result.
The result should be notified to your client at the earliest opportunity, and this includes interim positive results. Where you are registered with the LCA under section 7.4 the notification must include an interpretation of what the results mean and advice on what to do to resolve the underlying issue.
HTM0401 Part B gives good advice on how to interpret positive legionella results in healthcare settings. For other settings the general principles are the same, but the susceptibility of the population is likely to be lower.
The interpretation of sample results will follow directly from a well-designed sample plan. A sample is a small part designed to represent a larger population. In this case you’ve selected small volumes of water to represent the risk in the water system as a whole. When addressing issues, you must consider the whole system and not just the sample locations if the results indicate widespread contamination.
Positive results from pre-flush samples at limited locations with no positive results on any post-flush samples could be interpreted as localised colonisation at outlets only and actions recommended to address outlets such as localised disinfection, flushing, etc.
Widespread positive results from pre-flush samples with no positive results on any post-flush samples may indicate common issues and widespread colonisation at outlets. Actions should address all outlets represented by the sampling and not just the ones actually sampled.
Widespread positive results from pre-flush samples with positive results on post-flush samples indicate system wide issues. Actions should address the whole system with all its outlets, not just the ones sampled.
An LCA Member’s procedure should detail how they tell their client the information (is it an email, a phone call, a form to fill out and send, etc.) and how they arrive at the advice they give. This procedure should then be followed in practice and at audit the LCA will expect to see an audit trail for each positive result.
For legionella to grow there must be suitable conditions and there will be a root cause for these conditions. Actions following positive samples must address the root cause rather than just the symptom. Common factors include:
- Stagnation, areas being less used than previously thought
- Thermal gain/loss, meaning systems normally reside at growth temperature despite running below 20 degrees after two minutes or above 50 degrees within one minute
- Stagnation due to failed HWS circulation
- Unknown deadlegs
- Materials or contamination in the system that support growth
Reviewing the assessment of the risk and addressing whatever is found must be the primary recommendation but unless very simple and straightforward to achieve, there should also be interim recommendations. Processes like taking outlets out of use, fitting POU legionella filters or carrying our cleaning and disinfection work are typical interim actions that address the immediate symptoms but may not address the root cause.
Recommended actions should be appropriate for the results. This can be a difficult balance to strike but if the results indicate limited localised issues, take limited localised action, if results indicate more widespread issues, take system level actions.
Logging a Positive Legionella result.
The result should be logged on your systems as a non-conformance like any other loss of control non-conformance. Your interpretation, recommended actions and the reasons behind these should also be logged and recorded as part of the process.
Tracking a Positive Result
Once the result has been logged in your system the consequent actions need to be tracked and auditable. If actions are required by you or your customer then ensure these are completed and recorded in your logging process.
Resolve the Positive Result and Close Out
Once the root cause has been identified and resolved this action should be noted on your log and the positive result closed out. In many cases it is normal to carry out follow up sampling after remedial action to verify success.
Summary
Sampling for legionella can be an important part of the monitoring of control measures in a written scheme and competent interpretation of positive results is essential. Appropriate planning, technique and interpretation is vital to ensure the information obtained by sampling is valid and useful. The LCA requires members to have in place management procedures that detail how samples are planned, carried out, non-conformances logged and dealt with and actions closed out including communication with their clients. Please see the LCA service delivery standard for further details on what the LCA requires of its members.