Legionella control in hospital water systems

Harijeet Singh BSc, Business Development manager and Vikram Bamra Beng, Technical Services manager at Goodwater, present results from a three-month evaluation of monochloramine disinfection across two oncology hospitals within the same NHS Trust, both of which had experienced persistent Legionella pneumophila contamination despite conventional control measures. The study explores the operational and microbiological impact of introducing a controlled monochloramine dosing system to reduce contamination in healthcare water systems.

Waterborne pathogens such as Legionella pneumophila present an ongoing infection risk in water systems within healthcare facilities, particularly those with large and complex plumbing systems. Oncology hospitals are considered especially higher risk, owing to the immunosuppressed nature of their patient population and their dependence on consistent, microbiologically safe water.

Traditional Legionella control methods — such as thermal disinfection, localised disinfection of affected outlets using biocides such as chlorine and hydrogen peroxide, and point-of-use filtration — can offer temporary or partial efficacy but often fail to address entrenched biofilm contamination. Aged infrastructure, temperature fluctuations, and variable water usage patterns can further promote recolonisation, leading to repeated positive samples and high operational burden for estates teams.

Monochloramine (NH₂Cl) has recently emerged as a promising alternative secondary disinfectant within the building services sector in the UK. Its stability, ability to penetrate biofilms, and capacity to maintain residual effectiveness throughout distribution systems make it particularly well-suited to healthcare settings.

Materials and methods

The three-month evaluation was conducted by Goodwater at the following oncology hospitals:

• London site: 113 beds, seven operating theatres, and 16 critical-care beds.
• Surrey site: 106 beds, two operating theatres, and 16 critical-care beds.

Both hospitals had a long-standing history of
L. pneumophila positivity within blended outlets and showers, despite frequent disinfections using silver stabilised hydrogen peroxide (6 hours contact time) and temperature management regimes.

In conjunction with the Trust, a monochloramine generator (Sanikill Lite, manufactured by Sanipur SpA, Italy) was installed by Goodwater at each site within the primarily affected blocks. The generators were situated on a side-stream installation on the mains cold-water service feeding the cold water storage tanks within each block. At both sites, hot water services are generated via plate heat exchangers, fed from these cold-water storage tanks. At the London site, a water softener is also present on the feed to the hot water system. This setup meant that both the hot and cold water systems received monochloramine treatment via the single dosing station.

The Sanikill system generates monochloramine in situ through the controlled combination of pre-cursor chemicals; sodium hypochlorite and ammonium sulphate. The dosing rate was monitored and adjusted to ensure residual concentrations remained at ≤3.0 mg/L at any drinking water outlets, compliant with international drinking water standards.

Following the start-up of the Sanikill systems, no other supplementary disinfection, thermal interventions, or additional flushing (over and above what was already in place) was conducted during the trial periods. Any previously undertaken localised disinfections were ceased, meaning there was a complete reliance on traditional flushing and disinfection using monochloramine only.

Following initial start-up of the generators at each site, Goodwater’s Sanikill engineers attended site on an initial weekly basis over the first month, to ensure adequate levels of monochloramine were being detected throughout the systems, especially at those points known to have had presence of Legionella bacteria.

Residuals were also monitored by the hospitals’ site maintenance teams using a handheld colorimeter test-kit. Monthly inspection and calibration of the Sanikill generators were performed by Goodwater to maintain accuracy and safety.

Unlike other commonly used biocides such as chlorine dioxide or hydrogen peroxide which are known to struggle to achieve good residuals at the extremities of systems, or within the hot water circuits, monochloramine residuals (>1.0ppm and up to 3.0ppm) were seen throughout, including in the hot water systems and at the furthest outlets, demonstrating its stability.

Microbiological sampling was undertaken on a monthly basis by a third party organisation, as a means to validate the efficacy of monochloramine treatment. Over 25 outlets at each hospital were identified as having had Legionella contamination prior to the introduction of monochloramine. Majority of positive outlets were fed by thermostatic mixing valves/taps.

At the London site, during the previous 12 months prior to monochloramine, on average 63 Legionella samples were taken monthly, and an average failure of 10 per month, or 18 per cent. Prior to the start of the study, 25 outlets were still positive with counts of Legionella ranging between 50 cfu/litre to 1300 cfu/litre. 19 of these outlets had counts of 100 cfu/litre or higher.

At the Surrey site, during the previous 12 months prior to monochloramine, on average 71 Legionella samples were taken monthly, and an average failure of 9 per month, or 11%. Prior to the start of the study, 33 outlets were still positive with counts of Legionella ranging between 50 cfu/litre to 2850 cfu/litre. 23 of these outlets had counts of 100 cfu/litre or higher.

Legionella samples were taken monthly over the three-month study, and each location had a pre-flush (sample collected immediately after outlet opened) and post-flush (outlet flushed for a minimum of 2 minutes before collection) sample collected at every visit.

Microbiological analyses were performed by a UKAS-accredited laboratory in accordance with BS EN ISO 11731, using BCYE (GVPC) agar with L-cysteine confirmation and serogrouping by latex agglutination.

Results and discussion

Implementation of monochloramine dosing resulted in a substantial and rapid decrease in L. pneumophila detection across both sites.

At the London site, the results were as follows:

• Following start-up of the Sanikill system, initial monochloramine residuals averaged at 1.25mg/L across outlets tested.
• Month 1: 50 samples taken in total (pre and post flush), no reported failures. The corresponding monochloramine residuals at these locations were at an average of 1.74mg/L.
• Month 2: 50 samples taken in total (pre and post flush). 3 positive counts reported (3%), ranging from 20 to 400cfu/litre. The corresponding monochloramine residuals at these locations were at an average of 1.24mg/L.
• Month 3: 48 samples taken in total (pre and post flush), no reported failures. The corresponding monochloramine residuals at these locations were at an average of 1.70mg/L.

All sampled locations were previously positive as discussed above, therefore demonstrating a complete reduction from 100% to 0% within four weeks, with full eradication remaining until month three.

Monochloramine residuals were seen to gradually increase within the system as the disinfectant took control.

At the Surrey site, the results were as follows:

• Following start-up of the Sanikill system, initial monochloramine residuals averaged at 1.35mg/L across outlets tested.
• Month 1: 54 samples taken in total (pre and post flush), 6 positive counts reported (11%), ranging from 20 to 3000cfu/litre. The corresponding monochloramine residuals at these locations were at an average of 1.45mg/L.
• Month 2: 58 samples taken in total (pre and post flush). 2 positive counts reported (3%), ranging from 80 to 200cfu/litre. The corresponding monochloramine residuals at these locations were at an average of 1.17mg/L.
• Month 3: 60 samples taken in total (pre and post flush), no reported failures. The corresponding monochloramine residuals at these locations were at an average of 1.94mg/L.

All sampled locations were previously positive as discussed above, therefore demonstrating a complete reduction 
from 100% to 0% over the three month study.

Monochloramine residuals were again seen to gradually increase within the system as the disinfectant took control. (See Figures 1 and 2).

As outlined above, average residual monochloramine concentrations increased during the study period, and stabilised between 1.5 and 3.0 mg/L, with consistent readings across distal outlets. Data indicated an inverse relationship between L. pneumophila detection and monochloramine residuals, suggesting effective biofilm penetration and persistence of the disinfectant throughout the water systems.

The introduction of monochloramine delivered several operational advantages:

• Elimination of routine local disinfections, resulting in significant timesaving for the site estates teams.
• Significant reduction in point-of-use filter 
replacements, lowering consumable costs. Those locations where Legionella has not been detected 
for 3 consecutive events are no longer fitted with 
filters.
• Cessation of daily manual flushing at previously positive outlets, once again resulting in considerable time saving. Any little used outlets are now being flushed twice a week only, or in accordance with the WSP.
• Hospital estates teams have reported fewer reactive maintenance interventions.
• A corresponding reduction is Pseudomonas aeruginosa contamination (the study for this is still ongoing).

These findings align with international evidence demonstrating that monochloramine not only suppresses Legionella but also reduces colonisation by other opportunistic pathogens such as non-tuberculosis mycobacterium (NTMs) and Pseudomonas spp.

Monochloramine’s lower reactivity with organic matter compared to free chlorine allowed it to persist for longer within the distribution systems, maintaining effective residuals at distal outlets. Residual levels throughout this trial remained within safe limits, with no reports of taste, odour, or aesthetic concerns from users.

The Sanikill Lite generator operated reliably throughout the evaluation, requiring only periodic replenishment of precursor chemicals and standard dosing pump calibration checks. This low-maintenance profile contrasts with more labour-intensive control methods such as regular thermal disinfections or manual disinfection, both of which can pose scalding or handling risks.

The study duration (three months) represents a limitation, as long-term stability and recontamination potential were not evaluated beyond this period. However, these short-term results are consistent with multi-year international case studies demonstrating sustained control with monochloramine systems in healthcare environments. Future studies incorporating six- and twelve-month datasets will strengthen confidence in long-term performance and resilience.

Conclusions

Monochloramine dosing achieved complete eradication of L. pneumophila within three months across these two complex oncology hospital water systems with a persistent history of contamination.

Key outcomes include:

• 100% reduction in L. pneumophila-positive outlets within the 12 weeks period.
• Stable monochloramine residuals maintained between 1.5—3.0 mg/L.
• Secondary reduction in Pseudomonas aeruginosa detections (study ongoing).
• Reduced operational workload and maintenance demand.

The Sanikill monochloramine system provided a practical, reliable, and scalable approach to water safety management. Its integration into healthcare water systems offers a sustainable and resource-efficient alternative to traditional control measures, enhancing patient safety and reducing infection risk across critical care environments.

Further reading

• NHS Estates. Technical Bulletin: Control of Nontuberculous Mycobacteria and Other Opportunistic Pathogens in Healthcare Water Systems. NHS England; 2024.
• World Health Organization. Guidelines for Drinking-Water Quality. 4th ed. WHO; 2023.
• Kool JL, et al. Hospital-acquired Legionnaires’ disease linked to potable water. Infect Control Hosp Epidemiol. 2000.
• Farina C, et al. Water Safety Plan, Monochloramine Disinfection and Extensive Environmental Sampling Effectively Control Legionella and Other Waterborne Pathogens in Nosocomial Settings: The Ten-Year Experience of an Italian Hospital. Microorganisms. 2023;11(7):1794.
• Lin YE, et al. The efficacy of monochloramine for controlling Legionella in hospital water systems. Am J Infect Control. 2011.
• Bartram J, Chartier Y. Legionella and the Prevention of Legionellosis. WHO Press; 2019.