Monochloramine for the control of Legionella contamination of domestic hot water systems of health-care facilities. A six months experience from Sicily (Italy).

OBJECTIVES To assess the effectiveness of monochloramine for the control of Legionella colonization of two hospital water distribution systems (WDS) after using a combined control strategy for 1 year. METHODS In 2012, a sampling plan was implemented to assess Legionella contamination in two Sicilian hospitals’ WDS. The plan included the loops and one ward for each floor, reiterating the same sites every 4 months. High levels of L. pneumophila detected in the two hospitals led to the adoption of a combined control strategy: shock hyperchlorination, installation of point-of-use water filters in the most at risk wards and, in one hospital, continuous disinfection with H2O2. Due to the persistence of the contamination of the WDS, 3 experimental equipments continuously injecting monochloramine (Sanikill patented system, Sanipur Srl, Brescia, Italy) have been in place since February 2014. Legionella was isolated in accordance with ISO 11731. Serogrouping was performed by the agglutination Legionella latex test. Total Viable Count (TVC) at 22°C was isolated in accordance with ISO 6222. The total water hardness, pH, ORP, free chlorine, total chlorine, monochloramine, free ammonia, NO3-, NO2- were measured before the injection of the monochloramine (baseline) and every month for 6 months. Total THMs were measured at baseline and after 1, 3 and 6 months. RESULTS Between 2012 and 2013, in total 168 hot water samples were collected. Lp3 and Lp6 were isolated from 100% of the sampling points in the two hospitals (mean count from 103 to 105 CFU/L; TVC 22°C from >102 to >104 CFU/ml). Apart from the outlets with point-of-use filters, after each shock hyperchlorination or during continuous addition of H2O2 Legionella returned to the previous concentrations or remained stable, respectively. At the beginning of the disinfection with monochloramine, the disinfectant was injected at an average concentration of 3 mg/L for 30 days. Subsequently, the dosage was regulated within the range 2.0 – 2.5 mg/L. In the relatively new hospital, 1 week after the start of the monochloramination, the positive sample points decreased at 8% (Lp<103 CFU/L; TVC 0 UFC/ml). In the same hospital, however, as a consequence of a stop for around 15 days of the monochloramine generator device, Legionella was isolated in all the sampling sites (102-103 CFU/L). Anyway, all samples became negative as soon as the system got back to operation. No changes in water chemical parameters, included the production of disinfection by-products (DBP), were recorded. CONCLUSIONS Six-months application of monochloramine in the WDS of the two hospitals heavily contaminated by L. pneumophila gave satisfactory results both per se and compared with the other control contamination systems previously adopted. Continuous injection of monochloramine in hot WDS can control L. pneumophila and TVC in contaminated hospitals, without modifying water chemical composition and without releasing DBP.