Plumbing Africa May 2017 | Page 27

Health and sanitation • • Mòdol et al. (2007) described how a large hospital experienced success in decreasing the number of positive legionella samples after initiating CSI. The hospital had discontinued the use of chlorine in the hot water system due to difficulties in maintaining minimum concentrations at distal points in the building’s premise plumbing system. It also discontinued frequent superheating due to pipe damage and poor compliance with heating and flushing procedures. Initially, the copper and silver levels were maintained at 0.1– 0.3mg/L and 0.01–0.03mg/L, respectively. While the treatment system was under repair for two months, the percentage of positive samples for legionella increased from 20% to 65%. Following the interruption in treatment, hospital staff increased copper and silver concentrations to 0.4 and 0.04mg/L. Legionella samples taken after the increase were 16% positive. Blanc et al. (2005) reported that no significant difference was observed in the percentage of water and biofilm samples positive for legionella spp. after CSI treatment was installed in 1999. The CSI system electrodes were composed of 8% silver and 92% copper, and the copper concentration in the water was 0.3mg/L. A significant reduction in legionella isolates was observed after the hot water system temperature was increased from 50 to 65 degrees C (122 to 149 degrees F) in the year 2000. Survey results by Stout and Yu (2003) showed that of 13 hospitals reporting at least 30% legionella-positive samples before CSI treatment began, nine hospitals reported a sustainable (over a period of 6–9 years) decrease in the number of legionella-positive samples; five hospitals reported no positive samples after treatment. This survey also showed that all www.plumbingafrica.co.za • • • of the hospitals reported cases of hospital- associated legionnaires’ disease before CSI treatment, and all but one reported no cases after treatment. Lin et al. (2002) studied the effects of pH and other water quality parameters on CSI treatment for legionella control using water from a hospital hot water system. At pHs of 7.0 and 9.0, copper ions achieved a 6-log and 1-log (99.9999 and 90%) reduction, respectively, in the number of L. pneumophila in 24 hours. Silver ions achieved a 6-log reduction in 24 hours at all ranges of water quality parameters tested. Based on four years of monitoring data, Kusnetsov et al. (2001) reported that legionellae were no longer detected in the circulating warm water of a hospital water system after CSI treatment was employed and silver concentrations were increased to levels greater than 3µg/L. However, water samples collected from taps and showers that were not used on a regular basis showed that even a high silver concentration (55µg/L) did not prevent growth of legionella. Rohr et al. (1999) indicated that CSI had an initial impact on legionella occurrence in the hot water system of a German university hospital, where 100% of sampling sites were positive for legionella before treatment and 55% of sampling sites had positive results one year after treatment was initiated. Over the next three years, 75–78% of samples were positive for legionella. However, Lin (2000) disagreed with Rohr et al.’s (1999) results that the CSI system did not effectively control legionella in this system. Lin (2000) pointed out that the CSI system was not effective because it did not maintain an adequate concentration of copper and silver ions in the treated premise plumbing system (200–400µg /L of copper and 20– 40µg /L of silver was crucial). Lin (2000) also noted that Rohr et al. did not provide evidence for the development of legionella resistance to copper or silver. Rohr (2000) responded to Lin’s (2000) comments, and explained that the purpose of the study was to evaluate control of legionella using a CSI system that met German regulations limiting silver ion concentrations to a maximum of 10µg/L. States et al. (1998) reported that CSI treatment was successful in reducing the percentage of samples testing positive for legionella from 100% to less than 17% on average over a two-year period. Lin et al. (1996) found that L. pneumophila serogroup 1 was completely inactivated (6-log (99.9999%) reduction) within 2.5 hours using copper ions at concentrations of 0.1mg/L without silver ions, and more than 24 hours was needed to achieve a similar reduction using silver ions at concentrations up to 0.08mg/L without copper ions. When both copper and silver ions were used, inactivation was achieved at copper and silver concentrations of 0.04mg/L. Based on laboratory studies with filtered well water (pH 7.3), Landeen et al. (1989) determined that copper ions (at 0.4mg/L) and silver ions (at 0.04mg/L) can achieve a 3-log (99.9%) reduction in L. pneumophila at room temperature when the contact time is at least 24 hours. PA • • • The National Environmental Methods Index (https://www.nemi.gov/home/) “is a searchable database that allows scientists and managers to find and compare analytical and field methods for all phases of environmental monitoring.” 6 • 25 May 2017 Volume 23 I Number 3