Legionella on board trains: effectiveness of environmental surveillance and decontamination

Legionella pneumophila is increasingly recognized as a significant cause of sporadic and epidemic community-acquired and nosocomial pneumonia.

In recent years, our understanding of Legionnaires’ Disease (LD) has improved substantially and new diagnostic and treatment strategies have been introduced [1].

To share knowledge and monitor trends of LD across Europe, the European Working Group for Legionella Infections (EWGLI) was established in 1986 to better protect the health of travellers by improving the detection and control of infection sources in European countries [2].

Members of EWGLI established a European surveillance scheme for travel-associated infections in 1987; this surveillance scheme, named EWGLINET in 2002, is the EU’s surveillance network dedicated to collecting data on cases of LD in the EU, including travel-associated LD (TALD) [3, 4].

Since April 2010, the programme has also been coordinated by European Centre for Disease Prevention and Control (ECDC), and the name of the scheme has changed to the European Legionnaires’ Disease Surveillance Network (ELDSNet) [4].

In the beginning of July 2002, the European Guidelines for Control and Prevention of Travel Associated LD [5] were introduced in an attempt to standardise the investigation of clusters in EWGLINET countries and their Ministers of Health were asked to support the use of these guidelines.

In 2009, the EWGLINET surveillance scheme reported a total of 818 cases (794 confirmed and 24 presumptive) of travel-associated LD with onset of illness, which is a decrease of 52 cases compared with 2008 and is 129 fewer than in 2007. The countries that reported the most cases by country of residence were the UK (n = 173), Italy (n = 169) and the Netherlands (n = 109). These three countries have consistently reported the greatest number of cases to EWGLINET for several years [4].

The Italian National Surveillance System has reported that, in Italy, the incidence of Legionellosis has been increasing in the last ten years despite a slightly decreasing number of notified cases in the time periods from 2002–2004 and 2005–2006. In 2009, 1,200 cases of Legionellosis were diagnosed and reported (1,146 confirmed and 54 presumptive), with an incidence proportion of twenty cases per million (Figure 1) [6].

In Italy, in 2009, a total of 281 TALD cases, including both Italian and foreign tourists, were reported to the National Health Institute [6]. Among these, 178 cases of legionellosis concern to Italian Tourists. The 86% of them stayed at the hotels, the 5.6% stayed at camping and the 8,4% in other structures. The 10% of 178 cases have travelled in foreign locations.

Many studies describe the frequency and strength of Legionella spp. contamination in spa pools, natural pools, hotels and ships [7–17], but no study has been conducted that addresses the environmental monitoring of Legionella on board trains.

To protect the health of travellers and workers, the Health Service of Italian Railways and the Hygiene Institute of the Catholic University in Rome established for the first time an environmental surveillance programme to evaluate the degree of contamination on trains and the effectiveness and length of water tank decontamination.

We have examined air conditioning plants and toilet water tanks, which are the primary sources of biological agents. The air conditioning plants on board Trenitalia trains are planned and built to avoid condensed water stagnation. Water is directly drained out of the trains by canalisations that prevent microorganism proliferation. Passenger trains contain water tanks made of steel or resin that supply hygienic services by falling water and they may have a varied capacity (from 200 to 1,800 litres) [18].

A comparative pre-post ecological study was conducted from September 2006 to January 2011.

In the pre-decontamination period, 58% of water samples were positive for Legionella and only L. pneumophila was identified; 55.84% contained serogroup 1, 19.03% contained serogroups 2–14 and 25.13% contained both serogroups.

The mean bacterial load value was 2.14 × 10³ CFU/L; 17.51% of positive samples had a load value < 100, 39.34% ranged from 100 to 999, 35.03% ranged from 1,000 to 9,999 and 8.12% were > 10,000 CFU/L.

During the post-decontamination period, 566 water samples were collected and analysed: 42.75% were positive for Legionella and Legionella pneumophila was identified in 98.76% of the positive samples; 74.06% contained serogroup 1, 16.32% contained serogroups 2–14 and 9.62% contained both serogroups.

The mean bacterial load of positive samples in the post-decontamination period was 1.72 × 10³ CFU/L: 5.79% had a load value < 100, 50% were between 100 and 999, 35.12% were between 1,000, and 9,999 and 9.09% were > 10,000 CFU/L.

Several studies describe the frequency of Legionella spp. contamination in spa pools, natural pools and hotels [7–9]. Cases of LD are also well documented among cruise ship passengers [9–17]. To date, this is the first study concerning the environmental monitoring of Legionella on board trains.

This study demonstrates the necessity for periodic evaluation on trains to assess the potential contamination of Legionella spp., with the aim to drive decontamination efforts in the absence of specific guidelines. We hypothesized that an environmental surveillance programme, followed by decontamination procedures when necessary, would decrease the total bacterial load, thereby protecting the health of travellers and workers.

This study demonstrated a statistically significant decrease in total bacterial load after decontamination procedures. The t-test indicated a decrease in bacterial load since October 2008, with statistical significance from January 2009 to March 2010, suggesting that a disinfection treatment performed every one and a half years could reduce environmental contamination.

Moreover, this study demonstrated a decrease in Legionella-positive samples after decontamination procedures: 58% in the pre-decontamination period and 42.75% in post one were positive for Legionella. Although there was a decrease in the total Legionella-positive samples, we observed an increase of Legionella serogroup 1 (55.84% pre- vs. 74.06% post-decontamination), which is responsible for approximately 70% of Legionella infections in Europe [21]. This increase could have been due to the development of a protective biofilm, especially when load values were high. As demonstrated in this study, the disinfection treatments were more effective when the bacterial load was lower (17.51% of positive samples with a load value < 100 CFU/L in pre- vs. 5.79% in post-decontamination).

Even if hyper-chlorination of the water is consistently performed, this approach is particularly appropriate for the treatment and removal of planktonic cultures of L. pneumophila, but remains ineffective against sessile communities of the bacterium [22].

This study suggests that the resistance to chlorination of L. pneumophila serogroup 1 could be due to the biofilm that facilitates a higher tolerance to the disinfectant. In fact, we observed limestone deposits in different water tanks in which the biofilm could have been developed. Environmental strains could develop a resistance mechanism to chlorine-based disinfection treatments, so these decontamination procedures are not completely appropriate for removing the bacterium from plumbing and tanks. As these procedures do not lead to a definitive solution to the problem, we suggest that mechanical cleaning could increase the effectiveness and length of chemical decontamination procedures, thereby reducing bacterial load values and increasing the amount of time that the water can be considered safe.

We aimed to highlight that environmental surveillance represents an evidence-based approach for evaluating the contamination of the water supply on board trains. This work stresses the effectiveness of decontamination and, above all, provides information about the required frequency of disinfection procedures to achieve a statistically significant decrease in total bacterial load.

Periodic environmental surveillance could be a useful approach for assessing the risk of exposure to Legionella bacteria, which still poses a threat to public health. Legionella spp. are opportunists par excellence and are therefore especially dangerous for vulnerable individuals (e.g., older people, current tobacco smokers, immune-compromised or immune-suppressed persons, patients with chronic degenerative diseases and transplanted people).

Our purpose is to ensure the safety of the passengers and the workers on board trains by defining proper preventive measures to be implemented with an established frequency.

The primary limitation of this study concerns the use of the t-test; a paired t-test was not performed because this study focused on the general trend of contamination/decontamination on board trains as opposed to a specific sampling site.

Moreover in this study we analyzed only environmental data because we were given the responsibility of doing only the environmental surveillance; in fact, clinical surveillance was carried out by other institutions and, up to date, we have no information about cases of Legionellosis linked to our environmental data.

The lengthy time period of the environmental surveillance and the large number of water samples collected and analysed represent a point of strength in this study.

Public health programmes should focus on reducing the risk for LD among travellers.

In addition to cruise ships and ferries [16], the routine monitoring of water on board trains must be considered an essential measure of primary prevention for guaranteeing water safety and, consequently, to avoid cases of LD. For these reasons, the drafting of guidelines for LD prevention on public transportation may be necessary to protect the health of travellers and workers. According to our experience, it is necessary to pay close attention to water quality on board trains, especially after decontamination procedures to prevent bacterial colonization.

Although the eradication of Legionella spp. in plumbing and water tanks is difficult to achieve, our approach can be considered an evidence-based method for assessing and managing Legionella infection risk on trains.

The air conditioning plants on board Trenitalia trains are designed to avoid air conditioner condensation drip. The particular characteristics of water supply facilities on trains, which work by gravity at a low pressure, and the lack of showers reduce the possibility of the formation of aerosol, which is potentially dangerous for Legionellosis. Nevertheless, Italian Railways have carried out the present study in collaboration with Catholic University in Rome and the positive results may improve future train sanitation following the practical measures described in detail in this article.