**1. Introduction**

Hot and cold water systems (e.g., tap water installations, distribution systems, and cooling towers) are important sources of nosocomial and community-acquired infections caused by opportunistic

waterborne pathogens. Among them, *Legionella* spp. are water-based organisms that cause lung infections when inhaled in an aerosol form [1].

Several national standards have been established to ensure a high water quality using disinfection techniques that control and prevent the colonization of water systems by *Legionella* [2]. A wide variety of disinfection techniques, including chemical disinfection, ultraviolet (UV) light, and high temperature, have been employed worldwide to reduce the risk of legionellosis [3,4].

In Italy, Legionnaires' disease (LD) is a class II statutorily notifiable disease [5]; since 1983, it has also been subject to a reporting system designed to collect detailed information about contamination cases, which is held in a national register at the Istituto Superiore di Sanità (ISS), Italy. However, according to ISS annual reports, the number of LD cases is under-diagnosed and under-reported, leading to a significant underestimation of the real incidence of LD. In 2017, the incidence rate was 33.2 cases per million persons [6].

Following publication of the new Italian Guidelines for the Control and Prevention of legionellosis in May 2015 [7], the importance of a surveillance program encompassing all facilities at risk of LD (hospitals, healthcare facilities, dental units, hotels, tourist facilities, and spas) has been acknowledged, and this program has been implemented. The guidelines support the development of a risk assessment plan based on an evaluation of "risk" and also emphasize the need for an adequate environmental surveillance plan that includes an appropriate number of sites that are potential sources of *Legionella*.

A recent multicenter study performed by Montagna et al. [8] has demonstrated, as the main methods to perform *Legionella* prevention and control for the water network, were shock treatment and chlorination.

The shock treatment consists of a thermal disinfection of hot-water distribution systems performed at a temperature between 70–80 ◦C starting from the hot water storage heater. The temperature must be maintained in all outlets, faucets, and shower heads at least 30 min at 60–65 ◦C, for three consecutive days [7,9,10].

Several studies showed as the main disadvantage of shock treatment is its transitorily e ffect on bacterial community structure, e.g., biofilm, that was not removed preserving pathogenic *Legionella* niche [11–13].

Chlorine is the most common chemical disinfectant used in water (including drinking water), acts as an oxidizing agent, and reacts with several cellular constituents including the cell membrane of microbes. To perform *Legionella* control, plumbing water systems can be treated using chlorine as a shock hyperchlorination (residual chlorine concentration at distal outlets of 20–50 mg/L) or as continuous treatment using a concentration of 1–2 mg/<sup>L</sup> [10]. Although di fferent studies have shown good performance using these methods to assess *Legionella* contamination, a reduction of e ffectiveness over a long-term period was consistently demonstrated [10,14–16]. However, increasing evidence suggests that humans are exposed to residual byproducts of water chlorination such as disinfection byproducts (DBPs) through drinking-water, oral, dermal, and inhalational contact. During the chlorination, especially by hypochlorous acid and hypobromous acid, the reaction with naturally occurring organic matter present in raw water supplies, create many water DBPs, including the four primary trihalomethanes: chloroform (CHCl3), bromodichloromethane (CHCl2Br), dibromochloromethane (CHClBr2), and bromoform (CHBr3), that can have adverse e ffects on human health [17–20].

Disinfection methods other than chlorination have been suggested for *Legionella* control in water, such as ozone treatment, copper and silver (Ag+) ionization, monochloramine, point-of-use filters, and UV light. These measures have been tested over the last 30 years and are e ffective at controlling the growth of *Legionella*, all of them presented advantages and disadvantages that must be carefully considered [10,16].

Di fferent studies have focused in the last years on the role of oxidizing agents, notably hydrogen peroxide (H2O2), as disinfection treatments. The use of H2O2 as a biocide is widespread, and it is increasingly used as a general surface disinfectant in the medical, food, and industrial fields, as well as for water treatment [21,22]. H2O2 is completely soluble in water and is stabilized in commercial

formulation for disinfection treatment. It is compatible with different pipeline materials, and does not react with the organic constituents in the water to form dangerous residues with respect to chlorine, sodium hypochlorite (NaOCl), and monochloramine treatment. H2O2 decomposes rapidly in different environmental conditions due to microbial catalase and peroxidase, and other than abiotic action, the decomposition is promoted by heavy metal, oxidative, and reductive reactions. It shows a broad antimicrobial spectrum and has been shown to be active against bacteria, yeast, fungi, viruses, spores, proto-, and metazoans [23–25].

A disadvantage of using H2O2 is that its potency is influenced by several factors: pH, temperature, or the presence of substances that hamper its reactivity [26]. Since H2O2 is a renowned disinfectant, legislation [27] allows its use for the disinfection of water and in food; additionally, this compound is generally considered to have low eco-toxicity, as well as no odor or color [23,28].

To enhance its activity, H2O2 is sometimes used in combination with other oxidants such as ozone, Ag<sup>+</sup>, or UV radiation [24]. Silver, a biologically non-essential metal, has been investigated and used as a biocide for many years [29], and multiple strategies have been proposed for its use to treat drinking water [30–32]. Indeed, the World Health Organization (WHO) allows its use in drinking water. It is thought that concentrations up to 50 μg/<sup>L</sup> (ppb) in drinking water pose no risk to health [33].

The literature contains several accounts of the properties, germicidal effectiveness, and potential uses for stabilized H2O2 in healthcare facilities [34–37]. In 2015, Martin et al. [24] have demonstrated that Huwa-San peroxide (HSP), a new generation peroxide stabilized with ionic silver and suitable for continuous disinfection of potable water, preferentially interacts with the bacterial cell surface in a mechanism likely mediated by silver. Furthermore, treatment of hospital hot water systems with various formulations of H2O2/Ag+ compounds prevents contamination by *Legionella* and other microorganisms because of its bactericidal properties [38–40].

The H2O2/Ag+ formulation is stable at high temperatures, and its disinfection power increases significantly as water temperature increases. In a hot water system, a temperature range of 40–50 ◦C and a residual disinfectant concentration of 20–25 mg/L, seems to be able to induce a *Legionella* control [41,42]. Casini et al. suggested therefore, how a continuous feed rate of approximately 25 mg/L, was able to control the planktonic population, and silver can be deposited on the piping system, promoting a bacteriostatic effect [42].

Different commercial formulations based on H2O2/Ag+ are available to control *Legionella* contamination, but many studies lack data about the hospital settings and long-term applications.

Our study evaluated the effectiveness of a new disinfectant, Water Team Process 828 (WTP 828), based on H2O2 and Ag+ salts in the hot water distribution networks at Maria Cecilia Hospital (MCH), Cotignola (RA), Italy, controlling *Legionella* contamination.

The hospital is comprised of three buildings connected to each other but were built and submitted to renovation works at different times. The plumbing system comprises a single cold water supply and three different hot water return lines. These characteristics permitted us to study the activity of WTP 828 as three separate hot water networks (HWNs), modulating the dosage with respect to the level of *Legionella* contamination found. *Legionella* level in response to disinfection treatment was also studied by taking into account the following water network characteristics: building area, annual water consumption, hospital activities involving the use of water, that can influence the *Legionella* contamination and the disinfectant exposure to distal outlets [16]. The isolates were typed using an agglutination and genotyping approach to assess the distribution of strains in the buildings. The effect of WTP 828 was also tested on *Pseudomonas aeruginosa* (*P. aeruginosa*), one of the main components of biofilm [43], and HPC at 36 ◦C, commonly used as an indicator of water quality, and to monitor the effectiveness of disinfection treatment [33,44].

The physical and chemical parameters were also measured during implementation of WTP 828 treatment in all buildings in order to maintain the water quality characteristics [45,46] and preserve the plumping system materials.

The purpose of the study is to perform an extended investigation on the e ffect of H2O2/Ag+ treatment in a complex hospital water network system. The goal is to control *Legionella* infections throughout a risk assessment model based on the use of a low-cost disinfectant, easy to dose, and less aggressive on the material pipelines, with quick and safe monitoring of residual concentrations at distal outlets. This model associated to ordinary and extraordinary maintenance procedures (e.g., flushing, temperature control, and cleaning activities) could be extended to other hospitals, companies, and leisure facilities, where water represents a risk for public health.
