This website is intended for healthcare professionals only.

Hospital Healthcare Europe
Hospital Pharmacy Europe     Newsletter    Login            

Nosocomial-acquired Legionella infections

To reduce the risk of Legionella for hospitalised patients, guidelines and disinfecting procedures have been proposed; we should use those consistent with the level of contamination and the building and water network characteristics

Paola Borella
Full Professor of Hygiene and Public Health, University of Modena and Reggio Emilia
Legionella is a bacterium responsible for opportunistic bacterial pneumonia. It was isolated for the first time in 1976 in the air conditioning of a hotel in Philadelphia where 221 cases of pneumonia and 34 deaths occurred during a meeting of the American Legion.
Currently, the genus Legionella has 58 species and over 70 distinct serogroups; about 20 species are opportunistic pathogens ( Molecular studies show considerable genetic variability, with genotypic variations even within phenotypically homogeneous strains. L. pneumophila has 16 different serogroups, and is the main cause of Legionnaires’ disease (85% of cases), particularly serogroups 1, 3, and 6, followed by L. micdadei. Other species isolated from patients are L. dumoffii, L. bozemanii, L. gormanii, L. anisa and L. longbeachae (the most common in Australia). Legionella spp are difficult to cultivate, require specific culture media and long times for the growth in the laboratory.
Clinical signs and diagnostic tests
The legionellae enter through the mucous membranes of the respiratory tract by inhalation of contaminated aerosols, reach the lungs, where they are phagocytised by alveolar macrophages. Here, they multiply to cause lysis, with the release of new bacteria that can infect other cells.
Legionnaires’ disease is the most severe form of the infection, with a case fatality of 10%, but up to 30–50% in hospital-acquired infections. It is an acute pneumonia difficult to distinguish from other forms of acute infections of the lower respiratory airways. 
The disease occurs after an incubation period of 2–10 days, with high fever, non-productive cough, wheezing and symptoms common to other pneumonia. Sometimes complications, neurological manifestations and kidney and gastrointestinal disturbances may occur. The Pontiac fever is a flu-like, self-limiting acute disease that does not affect the lung. Subclinical forms are detectable with the appearance of antibodies to Legionella spp in the absence of pneumonia.1
Legionella can be isolated from all tissues and lung biopsy, but respiratory secretions are the standard of choice. The culture identifies all strains and allows the comparison of clinical isolates with the environmental strains, tracing the source of infection, but has the disadvantage of requiring a long time (from three to up 10 days). 
For this reason, the diagnosis is generally performed by searching for soluble antigens in a spot urine sample, having a response in a few hours, although it does not allow for the identification of all species. For the diagnosis of both clinical and environmental samples, molecular methods using DNA probes for in situ hybridisation and methods of DNA amplification by polymerase chain reaction (PCR) are currently in use and on sale.
Habitat of the microorganism
Legionella spp are widespread in nature, mainly associated with the presence of water (lake and river surfaces, hot springs, ground water and moist environments in general). From these sources, the bacteria move to artificial water environments: urban networks of distribution of drinking water, water systems of individual buildings, air conditioning, swimming pools, fountains, etc.2
They grow between 25 and 42°C, but are able to survive between 5.7 and 63°C. To prevent the infection, the World Health Organization (WHO) recommends that hot water should be heated and stored at 60°C and should reach the tap at a temperature of at least 50°C; the recommended temperature for storage and distribution of cold water is below 20°C.3
Man-made hot and cold water systems are the main sources of infection. Within the built environment, water temperature, configuration and age of the water distribution systems, physicochemical constituents of the water, and plumbing materials encourage their growth. Old components of the pipeline system, areas of stagnation or low flow, dead-legs and storage tanks allow their survival and development. The ability to enter in a viable but not culturable (VBNC) state and the presence of biofilm and protozoa are additional important factors for Legionella growth.4,5
Infections can be attributed to inhalation of contaminated aerosols produced from water outlets (showers and taps), hot tubs and pools (both cold water and heated pools), storage tanks, condensation trays in air conditioners and fan coils, evaporative coolers, irrigation systems, ornamental fountains, humidifiers, and dental chair unit waterlines. In hospitals, equipment for assisted ventilation and apparatus for aerosol and oxygen therapy may also be the source of nosocomial Legionnaires’ disease.
Surveillance and epidemiology of Legionnaires’ disease
Legionella spp are opportunistic bacteria that attack people who have an underlying illness or weakened immune system. Risk factors for Legionnaires’ disease include male gender, heavy smoking and alcohol abuse, chronic debilitating illness (chronic heart/lung disease, renal/liver failure and diabetes), haematologic malignancies, lung cancer, steroid therapy and other immunosuppressive treatments. In healthcare systems, additional risk factors include recent surgery, intubation or other ventilation assistance, respiratory therapy, and use of aerosol generators. There is no evidence of person-to-person transmission.
Cases or outbreaks have been reported in hospitals, nursing homes, dental offices, hotels, camp grounds, recreational and spa facilities, cruise ships, homes, etc. The surveillance of Legionnaires’ disease is carried out by the European Legionnaires’ Disease Surveillance Network (ELDSNet) coordinated by the ECDC.6 In 2012, 5852 cases were reported by 29 countries. The three largest reporting countries were France, Italy and Spain followed by Germany, the Netherlands and the United Kingdom. 
Among 5136 cases in which the setting of infection was reported, 3553 (69%) were community-acquired, whereas travel and healthcare-associated cases accounted for 20% and 8%, respectively. The median age was 62 years, and the disease was more common in males, with an overall male-to-female ratio of 2.5. The mortality rate was 0.8 per million inhabitants. The infections caused by L. pneumophila serogroup 1 accounted for 76.5% of the total cases.
Prevention and control
The control of Legionella spp contamination is relevant in healthcare settings where patients, mostly with compromised immune systems, are at increased risk of contracting the disease and having a fatal outcome.7 For this reason, national and international guidelines advocate the adoption of preventive measures to control Legionella water contamination. 
Legionella prevention should start from the correct design and construction of water supply, in order to make colonisation and multiplication of the germ in hot water distribution systems and air conditioning systems unlikely. During renovations or new construction, avoid installing pipes with end pieces blind, preferring the systems with instantaneous production of hot water to those with storage tanks and install the air conditioning systems so that the exhaust air from the cooling towers and evaporative condensers does not enter the building. 
In large buildings (hospitals, hotels, recreational facilities, etc.) as well as in small rooms (apartments, dentists, etc.) periodic maintenance can effectively contribute to the prevention of colonisation and limit the bacterium growth and dissemination.
The available methods for controlling the contamination of Legionella spp in water networks are numerous, all effective in the short term, but not long term.8 The choice of the most appropriate method depends on the characteristics of the building (that is, departments at risk in a hospital have different problems than a spa or a hotel), the water type and the characteristics of water distribution systems (for example pH, temperature and water turbidity, the complexity and the material of construction). 
Furthermore, healthcare facilities should follow these recommendations: water that is used to rinse and to clean respiratory apparatus should be sterile; birthing pools should be designed for the purpose, and should be physically cleaned and disinfected both before and after birth; in high risk areas, such as transplant centres and intensive care units, point-of-use filters are needed at the outlets.
For disinfection, we can use physical means such as temperatures of more than 60°C, ultraviolet radiation placed at the entrance of the water distribution network, filters applied to the points of use (taps, showers) especially in a hospital environment for the protection of patients at greater risk. Among the chemicals, the most popular are continuous systems involving the use of hydrogen peroxide, metal ions (copper, silver), and chlorine-based disinfectants such as chlorine dioxide, hypochlorite and monochloramine. 
Each modality differs in its design and application, and to choose an appropriate cost effective measure requires careful analysis and planning. The selection of the vendor for installation of a systemic disinfection method requires careful consideration with intense scrutiny. The WHO recommends performing cultures for Legionella every three months to verify the efficacy of procedures adopted. Monitoring points should be identified throughout the system on the basis of system design, operating parameters and high risk areas. 
Particular attention should be given to areas where control is most difficult to achieve, and areas where Legionella is most likely to grow. The suggested number of outlets to be sampled for a 500-bed hospital is a minimum of 10 distal sites plus the hot water storage tanks. 
In Table 1, we present the effectiveness of different measures adopted in our hospital to control Legionella contamination, both in terms of reduction of positive points and of points exceeding 104cfu/l compared to pre-treatment, with information on their cost. 
In cases of outbreaks, superheat-and-flush can be quickly implemented. The water temperature is elevated to above 70°C, and distal sites are flushed for 30 minutes at temperature greater than 60°C. This technique is not suitable for large buildings where temperatures >60°C at each outlet cannot be maintained; from our experience, this was the most unreliable method, and re-colonisation rapidly occurred because disinfection is only temporary. Another thermal disinfection strategy was the installation of small electric boilers serving 1–2 adjacent rooms; the absence of contamination is guaranteed when the temperature is maintained at >58°C. 
Point-of-use filters (0.2μm pore size) have been used for prevention of nosocomial infections due to Legionella and Pseudomonas aeruginosa, particularly in high risk areas, such as intensive care units and transplant units. They achieve 100% negative samples, but must be replaced regularly. Retrograde contamination of the filtration membrane is a critical issue that may occur by either splash water from the water basin during use or by direct contact with contaminated hands. Therefore, regular education of personnel and patients is recommended.
Chlorine can be added using chlorine gas or hypochlorite salts. The shock hyperchlorination is performed injecting elevated chlorine dosage (20–50mg/l) and after 1–2 hours, replacing the water in the system with fresh water, and maintaining around 1mg/l of chlorine concentration. This is the most expensive quick disinfection modality, due to the fact that many persons are required to monitor chlorine concentration at the distal sites, and reduction of positive points is limited. After 1–2 months the re-colonisation occurs, sometimes at levels higher than before.
In our experience, chlorine dioxide is produced in situ by using hydrochloric acid and sodium chlorite (Sanipur S.r.l., Italy) and injected in continuous to reach 0.3–0.5mg/l at distal points. After three years, 60 out of 201 samples were found still positives (29.8%) although at low levels (mean 3.0x102cfu/l). This is due to the inadequate penetration of the biocide into biofilms, and the resistance of the bacteria to chlorine. 
Furthermore, problems with corrosion of pipelines may appear, and the introduction of disinfection by-products (DBPs) into the drinking water poses a health concern. A guideline value has not been established, but the taste and odour threshold for this compound is 0.4mg/l and the maximum residual disinfectant level suggested by EPA is 0.8mg/l. 
We recently experimented with monochloramine generated in situ by reaction between a stabilised chlorine-based precursor and an ammonium salt (Sanipur S.r.l., Brescia, Italy), and used at concentration between 2.0 and 3.0mg/l at distal sites, according to WHO and EPA guidelines. Monochloramine was effective in reducing Legionella counts from 97.0% to 8.3% in the first month of its application. 
After three years, only nine out of 95 samples (9.5%) were positive, at a mean level of 2.2×102 cfu/l, and no sample exceeded 104cfu/l versus 59.4% at baseline.9 The increase of other microorganisms (Mycobacterium species), and the formation of nitrogen by-products with carcinogenic activity have been suggested as possible adverse effects, but we could not find any increase of nitrosamines during a six months surveillance.
We highlight that in healthcare structures the continuous disinfection of hot water may be an effective tool to reduce Legionella contamination, but we emphasise that all systems must be continuously monitored and adequate levels of the biocides should be selected in order to obtain the best effectiveness with the minimum risk of pipe damages.
  1. Borella P et al. Prevalence of anti-legionella antibodies among Italian hospital workers. J Hosp Infect 2008;69(2):148–55.
  2. Borges A et al. Detection of Legionella spp. in Natural and Man-made Water Systems Using Standard Guidelines. J Microbiol Research 2012;2(4):95–102.
  3. Bartram J et al. Legionella and the prevention of legionellosis. 2007;ISBN 92 4 156297 8.
  4. Mansi A et al. Legionella spp. survival after different disinfection procedures: comparison between conventional culture, qPCR and EMA–qPCR. Microchem J 2014;112:65–69. 
  5. Borella P et al. Water ecology of Legionella and protozoan: environmental and public health perspectives. Biotechnol Ann Rev 2005;11:355–80.
  6. European Centre for Disease Prevention and Control. Legionnaires’ disease in Europe, 2012. 2014;ISBN 978-92-9193-565-9.
  7. Ferranti G et al. Etiology, source and prevention of waterborne healthcare-associated infections: a review. J Med Microbiol 2014;63(10):1247–59.
  8. Marchesi I et al. Effectiveness of different methods to control Legionella in the water supply: ten-year experience in an Italian university hospital. J Hosp Infect 2011;77(1):47–51.
  9. Marchesi I et al. Control of Legionella contamination in a hospital water distribution system by monochloramine. Am J Infect Control 2012;40(3):279–81.