Novel approaches for risk management of legionella bacteria in domestic water systems

Abstract

Legionella pneumophila, the causative agent of Legionnaires’ disease, is a water born pathogenic bacteria commonly found in natural and manmade water systems such as rivers, lakes, wet soil, hot and cold water storage systems (being able to survive at temperatures between 6-63 ͦC, and proliferating between 20-45 ͦC), showerheads, cooling towers and spa pools. The main pathway of exposure to Legionella is by inhaling the aerosols containing the microorganism. Legionnaires’ disease can be fatal if not diagnosed and treated at the right time. Practical Legionella control starts with a risk assessment of the water system and followed by the regular monitoring and water sampling. UK Health and Safety Executive (HSE) have implemented strict legislations to protect the public from Legionnaires’ disease. This research highlights and addresses three major data gaps identified in Legionella control and management strategy employed in the UK and worldwide; namely, (i) the underestimation of microbiological threat in current cold water storage sampling strategy, (ii) the inability of current qPCR diagnostic methods to detect live Legionella in water samples, and (iii) the lack of predictive ‘risk management system’ for Legionella control in domestic water systems. During my PhD, 15 relevant cold water storage tanks (selected from more than 6000 tanks surveyed at different sites located in different London Boroughs) were used to investigate the risk factors that contribute towards Legionella proliferation, and revealed serious shortcomings in the appropriateness of the water sample taken for regulatory testing. Secondly, molecular biology research was carried out to develop an accurate, reliable and rapid testing method for the detection and quantification of live Legionella using qPCR techniques. This was successfully achieved by extracting RNA from a Legionella lenticule, converting the RNA into cDNA and amplifying the cDNA using qPCR techniques. Finally, regular monitoring data from 120 London buildings (60 known to be Legionella positive and 60 known to be Legionella negative) was used to identify the possible risk factors contributing towards Legionella outbreaks. Data for these factors was then used to develop a predictive risk model for Legionella contamination using Principal Component Analysis (PCA). The model was validated with 66 new London buildings and 9 out of London buildings. The model showed 100% accuracy in predicting the risk of Legionella by distinguishing infected and non-infected sites in London as well as for the sites in out of London.