The Efficacy of Residual Chlorine Content on the Control of Legionella Spp. In Hospital Water Systems.

BACKGROUND: Outbreaks of legionellosis may be a side effect of institution-water treatment. However, the long-term outcomes and the predictive factors of Legionella prevalence in such systems have still not been fully studied. This study was therefore conducted to investigate the prevalence of Legionella spp. and to evaluate the role of bacteriological water quality parameters on its prevalence and removal in hospital water systems.
METHODS: A total of 45 samples were collected from distinct sites at seven hospitals in Tehran, Iran. The prevalence of this bacterium was assayed through a sensitive and specific technique for DNA detection using PCR. Multivariable stepwise regression analysis was used to explore the independent effects of the baseline factors on the incidence of Legionella. Two positive samples were also identified for species by DNA sequencing.
RESULTS: Legionella were detected in 31.1% of samples. Showerheads and cold-water taps were the most and the least contaminated sources with 55.3 and 9 percent positive samples, respectively. Total mean of residual chlorine was 0.38 mg/L, with the peak value of 1.7 mg/L. Legionella detection was proportional to the residual chlorine content of water and the results indicated that residual chlorine content is a critical factor in the incidence and proliferation of Legionella (r=-0.33). The prevalence of Legionella also coincided with the prevalence of HPC and amoeba cysts.
CONCLUSION: The high positive rate of Legionella colonization shows that hospital-acquired legionellosis might be under diagnosed in studied hospitals. Further, Legionella colonization is independent of the type of water, system characteristics and of preventive maintenance measures.

Introduction

Legionella are gram-negative, aerobic, and sporeless bacteria which some of their species like Legionella pneumophila are implicated in severe pulmonary nosocomial infections (Legionnaire’s Disease) and Pontiac fever, especially in immunocompromised patients, as well as in the elderly (1, 2). Indeed twenty one species of Legionella are pathogens for humans, especially in patients with the chronic pulmonary disease (3). Inhalation or micro aspiration of Legionella from contaminated environmental sources such as hot water systems and cooling towers water is the most frequently route of transmission. Transmission has also been reported via nebulizer and showers in contaminated water as used (4, 5). Although, Legionella occurrence as a seasonal pattern is very common, its negative impact on health and hygiene requires a specific treatment cycle that is quite often a combination of management and sanitary stages. Hospitals are common habitats for the bacterium, where the bacterial niches are amply found. Hospitals provide the most likely places for susceptible people to contract the diseases. Outbreaks of legionellosis have been reported from hospital patients in many countries with an incidence range of 0 to 47% (1, 5-7). Furthermore, there has been a steady increase in the incidence of sporadic cases reported. Consequently, national Legionella surveillance programs have been established for regular monitoring of environmental samples in these countries (6, 8-9). In Iran, however, hospital-acquired Legionnaire’s Disease has rarely been reported and environmental surveillance for Legionella in hospital water systems to provide useful data for risk assessment and prevention of hospital-acquired Legionnaire’s Disease has never been systematically performed.

DNA-based techniques are innovative tools for routine quality control assessment in environmental water samples and are thought to be valid alternatives for culture methods. The PCR is considered the most adaptable and prevalent DNA-based assay technique, which is highly specific and sensitive alternative method to standard culture isolation, especially when rapid results are needed. This method is especially favorable when the samples contain abundant and diverse microbiota and when fastidious and slow-growing bacteria like Legionella are to be detected. Despite the fact that culture method for isolation of Legionella is approved by International standard Organization (ISO) and many other national standards for water quality determination; over the past few years, molecular techniques based on 16S rRNA gene beside other genetic markers have been developed to analyze bacterial communities in environmental samples (10, 11).

It is believed that the presence of Legionella is related to the physicochemical characteristics of water. The correlation between the occurrences of Legionella and water quality parameters is important to identify the main factors promoting the growth of Legionella in hospital water systems and to optimize the facility maintenance and operation. However, it appeared that results concerning the physicochemical characteristics of water and Legionella occurrence were often inconsistent or even contradictory and strong disparities in relationship have also been reported. In the context of this view, this study was conducted to investigate the presence of bacteria belonging to the Legionella genus in water supplies of some hospitals in Tehran, the capital city of Iran. The impact of water quality on Legionella existence was also determined. In spite of the large amount of the data available for the various indicator species in water samples, to our knowledge, this is the first attempt to gather information to monitor the presence of this bacterium in Iranian hospital water systems and there is no systematic study to assess the efficiency of chlorine disinfection. For more confirmation, randomly two of isolates determined as Legionella spp. were sequenced as well.

Materials and Methods

Sample collection and preparation

Forty five samples were collected from distinct sites at seven hospitals underlying Tehran University of Medical Sciences. This cross-sectional study was conducted from June 2011 to January 2012. Hospital water facilities sampled included tap cold and hot water, shower, cold water tank, hot water tank, and cooling tower water. Samples were collected in 1-litre sterile bottles directly from the outlet. Before sampling, a sterile swab was inserted into faucet outlets to dislodge the sediment. At the moment of sampling the electric conductivity (EC), temperature, pH, and free chlorine (DPD method) of each sample were recorded. Concentrations of hardness, calcium, magnesium (calculation method), iron and manganese (atomic absorption method) were determined in laboratory. Furthermore, samples were analyzed for microbiological quality using the standard plate method to determine the total heterotrophic counts per milliliter at 37 °C (HPC) and the presence of amoeba cyst in water. HPC values were determined with the pour-plate method using R2A medium. Volumes of 1 ml were spread on plates and colonies were counted after 10 days of incubation at 37 °C. Analysis of the presence of amoeba cysts in the water was based on morphological characteristics of cysts. After filtration of samples through a 0.22 μm mixed cellulose ester membrane filters (Schleicher & Schuell), the filter was separated and directly placed on the non-nutrient agar medium prepared with Amoeba Page Saline. Plates were then kept at room temperature for two weeks and were monitored daily for the growth of amoeba during this period. Reversed contrast phase microscope with a 10x objective was used to investigate the presence of amoeba. All of the experiments were measured according to standard procedures reported in Standard methods (12).

DNA extraction, PCR assay, gel electrophoresis and DNA sequencing

One liter of samples was filtered through 0.22 μm mixed cellulose ester membrane filters in a stainless-steel filter holder with a water aspirator. Each membrane was aseptically scraped, cut into smaller pieces and placed into sterile containers with 10 ml of the original sample. The samples were then sonicated for 5 min (Bandelin Sonorex), and shaken for 15 min to dislodge bacterial cells from the membranes. The eluate was transferred into a 15 ml conical centrifuge tube and centrifuged (2000 g, 20 min) to remove cell debris. Total DNA was extracted from concentrated water samples using freeze & thaw and/or phenol & chloroform methods. DNA extraction using freeze & thaw method was conducted by placing 1mL of each concentrated water sample within 1.5mL microtubes and alternating application of freezing the samples in liquid nitrogen and their incubation in water bath in the temperature of 100 °C for three times. The suspension was then centrifuged again (18000 g, 10 min) and an aliquot of 20 μL from the bottom of the tubes was transferred to new microtubes.

Extracted DNA was stored at -20 °C until PCR. Amplification reactions were preformed according to what described earlier by Hsu (10). The PCR primers LEG 225 (5’-AAGATTAGCCTGCGTCCGAT-3’) and LEG 858 (5’-GTCAACTTATCGCGTTTGCT-3’) were used to amplify a 650 bp fragment of the 16S rRNA gene of Legionella species. Each 25 μl of reaction contained 20 ng genomic DNA, 1.5 mM MgCl2, 0.2 mM dNTP, 20 pmol of each primer, and 1u of Tag Polymerase (Roche Biotech) in the PCR buffer. The cycling conditions were 94 °C for 5 min, followed by 30 cycles at 95°C (30 sec), 64 °C and 74 °C for 20 sec each, and 1 cycle of 72 °C for 5 min in Thermocycler (Techne USA). PCR products were loaded onto a 2% agarose gel containing ethidium bromide. To confirm the obtained results, DNA sequencing was prepared by the 16S rRNA. The PCR products of two Legionella isolates were sequenced at MWG (mwg_biotech.com, Germany).

Statistical analysis

Statistical analysis was worked out using SPSS software. Quantitative variables were expressed as mean ± standard deviation when the data was normally distributed, while variables were expressed as a median (range) when the data was not in a normal distribution. Quantitative variables were compared by Student’s test. The X test was used where applicable to compare categorical variables. Univariable logistic regression analysis and multivariable logistic stepwise regression analysis were used to explore the independent effect of the baseline factors on the incidence of Legionella. Differences were considered significant at a P-value of 0.05 or less.

Results

Identification of Legionella specie by PCR in samples was based on the presence of an amplified product of 654 bp. Analysis of PCR results revealed a great diversity with regard to the sources from which samples were taken. Table 1 demonstrates the results of Legionella monitoring in seven hospitals by the source. Quality characteristics of hospital water sources on leginella detection are shown in table 2.

Table 1

Legionella prevalence by the source

Sampling source	Legionella positivity, % (No. of positive/total No.)
Cold water tap	9.1(1/11)
Hot water tap	25.0(2/8)
Showerhead	55.5(5/9)
Cooling tower	30.8(4/13)
Hot water tank	25(1/4)
Total*	28.9(13/45)

Table 2

Quality characteristics of hospital water sources (routine monitoring data of a 1-year period)

According to the table the Cooling towers water showed considerably higher mean HPC values than other water environments. As shown in this table, HPC values in the sampled waters from cooling towers have been reported as TNTC (too numerous to count), and the median values for other samples were 72.1 CFU/ml. This finding is expected and is in accordance with the nature of such systems. Cooling towers were therefore excluded in the further analysis of the HPC data and a moderate positive correlation (r=0.37) was found between HPC values and Legionella presence.

Amoeba was detected in 21 out of 45 (i.e., 46.7%) investigated hospital water samples. Excluding cooling tower samples, all polluted with amoeba, there were two hospitals with no positive sample taken from water sources.

Discussion

Legionella prevalence and species identification

In general, showerheads were the most contaminated source with 55.3 percent positive samples. Similarly, the samples from cooling waters, hot water taps and cold-water tanks yielded fairly comparable results (37.5%, 25%, and 20% respectively). However, this was not the case with the cold-water tap samples and only 9% were positive for Legionella and no Legionella was detected in hot water tank samples.

Two positive samples were identified for species by DNA sequencing. DNA for sequencing was prepared by the 16S rRNA.

The PCR products of two Legionella isolates were sequenced at MWG (mwg_biotech.com, Germany), DNA sequence was used to search the Gene Bank database, and the database entry with the highest percentage similarity was taken to identify the species. Nucleotide sequences data have been submitted to the Gene Bank database with accession No. FJ480932 for L. pneumophila and FJ48093. Detection of Legionella in aquatic environments has been demonstrated in other researches (10, 13, 14). Although PCR inhibitor may interfere with the results obtained by PCR, it has shown higher sensitivity than culture as demonstrated by Lye et al., and Morio et al. (1, 15). This may be derived from the relatively low concentrations of Legionella and supported by the fact that Legionella bacteria are commonly present in aquatic environments in the viable but non-culturable status which cannot be detected by culture (16, 17).

In our study, although no statistical difference among Legionella positive rates in various sources was found (P>0.05), the positive rate itself showed the severity of contamination. Therefore, even though the results obtained by PCR are not a valid determinant of Legionella viability in the environmental samples, it should be seriously considered as a potential public health threat. Owing to the lack of epidemiological and ecological studies, no Legionella outbreaks have been reported in environmental water samples in Tehran or other cities of Iran till now. However, considering patients’ complaint about their acquired pulmonary diseases at hospitals, the results of this study showed that sporadic or even a fairly high incidence of Le-gionella might have occurred but neglectfully distinguished as other pulmonary diseases.

Relationship between the occurrence of Legionella and water quality

Previous studies have shown that the incidence of pulmonary diseases at hospitals associated with Legionella differs markedly with physicochemical properties of water.

In the present study of seven hospitals in Tehran, total mean of residual chlorine was 0.38 mg/L, with the peak value of 1.7 mg/L. Legionella detection was proportional to the residual chlorine content of water and the results indicated that residual chlorine content is a critical factor in incidence and proliferation of Legionella (r=-0.33). However, even a concentration of 0.4 mg/l of residual chlorine showed to have no effect on disinfecting the bacterium. This study was generally conducted during the warm seasons of the year, when the municipal drinking water was super chlorinated due to water shortage and subsequent loss of water pressure in the distribution system at some hours of the day. Although Legionella is susceptible to disinfection, and it is believed to reduce the potential risk of contracting infections (18), the characteristics of cooling tower facilities make it impossible to maintain adequate disinfectant residuals. Although many guidelines recommend conscientious maintenance and though such recommendations seem reasonable, to the best of our knowledge, there is not any documented data on the minimization of Legionella colonization in cooling towers by maintenance measures and also on the usefulness of control measures in preventing outbreaks of Legionnaires’ disease from cooling towers. Routine maintenance measures have been recommended for general quality of hospital water, not on Legionella colonization. On the other hand, some species of Legionella have shown an ability to survive in high free chlorine concentration, the condition that inhibits other bacterial groups (19). Many microorganisms of aquatic origin, such as Pseudomonas, produce bacteriocins that may actually be inhibitory to Legionella spp. Thus, chlorine levels that eliminate these microorganisms could increase the population of indigenous Legionella. Furthermore, Legionella spp. are more resistant to chlorine than other bacteria because they can attain a viable but nonculturable state, be protected by amoebae, and/or survive in pipe biofilms (19). It has therefore been deliberated that the scientific method has been less successful in the prevention of legionellosis contracted from cooling towers than from drinking water systems (7). Floating and attached biofilms are common in such aquatic environments. The presence of Legionella and some of its amoeba hosts has been proved in these biofilms from both anthropogenic and natural aquatic systems (20), which explain why the sampled cooling towers in this study generally demonstrated higher positive results for Legionella than other sources. Many studies have documented that Legionella need an amoeba host for replication in the biofilms and also amoebae are able to resist different treatments and survive under stressful environmental conditions, such as high chlorine levels. Therefore, they probably act as reservoirs for Legionella, allowing quick recolonization of the system once the treatments are interrupted (20–22). To assist with confirmation, it has been recently shown that Legionella may be necrotrophic, allowing bacterial growth despite water treatment (23). Another hypothesis that may validate this finding is that cooling towers are generally operated at temperatures close to the optimal growth temperature of Le-gionella (35°C), which may lead to high bacterial concentrations (4). It should be also mentioned that cooling towers are more liable to colonization owing to their aged-old and low hydraulic circuit.

Besides old and partially somewhat damaged water piping in most of the hospitals under study, survival and extension of amoeba in the nature and especially in water support its high presence in hospital water samples. Nevertheless, based on table 2, the chlorine concentration in most of the samples (excluding cooling towers) was greater than 0.2 mg/L, confirming that water quality standards, both cold and hot tap water samples, especially in old hospitals had amoeba contamination. This can be rooted from the fact that amoeba have high resistance against water chlorination, as the minimum concentration of 1.5 mg/L of free residual chlorine has been reported to be effective against Acanthamoeba cyst (a well-known Legionella amoeba host) (24). A variety of microorganisms such as Legionella spp., which nest in the form of endosymbiont in the amoeba as amoeba associated bacteria, can survive after chlorination and applying other disinfectants (25). A fairly strong and significant correlation (r=0.46) was found between the presence of amoeba and Le-gionella in the present study (P<0.01).

Conclusion

Molecular techniques based on PCR assay offer a rapid, practical, cost-effective and sensitive alternative for detection of Legionella. Although the concentration of Legionella in the sampled hospital water systems was not determined, given the high positive rate of Legionella colonization, hospital-acquired legionellosis might be under diagnosed in Tehran. It calls for urgent control measures to minimize the transmission rate of Legionella from the source to the host and to prevent an outbreak. In light of these data, it can be concluded that Le-gionella colonization in environmental waters such as hospital water systems is a deep-rooted phenomenon. Moreover, Legionella are able to persist and increase with time, independent of the type of water, system characteristics and of preventive maintenance measures.

Ethical considerations

Ethical issues (Including plagiarism, Informed Consent, misconduct, data fabrication and/or falsification, double publication and/or submission, redundancy, etc) have been completely observed by the authors.