The role of chemical products at low doses in preventing the proliferation of bacteria in dental unit waterlines: the ICX^® experience

Dental units are a reservoir for potential pathogens of human or environmental origin, and dental instruments are believed to be responsible for the transmission of microorganisms by direct contact or by spreading through aerosol sprays created by handpieces (high-speed drills, scalers, air and water syringes). A dental unit is furnished with a system of thin plastic tubes, called dental unit waterlines (DUWLs), which deliver water to the different handpieces.

Contamination occurs when oral organisms (Abel et al. 1971; Fitzgibbon et al. 1984) enter the unit's waterlines through back siphonage when the handpiece is momentarily turned off (Bagga et al. 1984; Crawford & Broderius 1990). Dental units can also become contaminated from the main water supply, which, although potable, still carries bacteria producing an adherent heterogeneous microbial accumulation called a biofilm. Once formed, the biofilm protects the organisms from desiccation, chemical insult and predation, and it serves as a reservoir that significantly changes the number of free-floating microorganisms in the water exiting the waterlines (Whitehouse et al. 1991; Barbeau et al. 1998). Most of the bacterial species found in DUWL output water are Gram-negative aerobic heterotrophic environmental bacterial species that exhibit very low pathogenicity, although they may be of concern in the treatment of vulnerable patients, such as immunocompromised and medically compromised individuals (Martin 1987) and dental staff (Fotos et al. 1985; Reinthaler & Mascher 1986). Because of these contaminants, it is important to establish control methods for cleaning and disinfecting dental water systems and for providing quality irrigant/dental treatment water. Depending on the nature of various germicidal agents and the various devices or systems provided with the dental unit, chemical treatment protocols could be used intermittently as a ‘shock’ treatment and/or continuously introduced into waterlines in small quantities.

Intermittently used cleaners are not intended for patient contact and require that the waterlines be flushed after each chemical treatment, which eliminates any potential adverse effects the chemical may have on the bond strength of dental adhesive materials. However, one possible drawback is that intermittent cleaning may allow for reformation of the biofilm between uses (Karpay et al. 1999; Tuttlebee et al. 2002; Montebugnoli et al. 2004; Schel et al. 2006; Puttaiah et al. 2012). In contrast, cleaners intended for continuous use are less concentrated and do not require flushing of waterlines. The main benefit of continuous chemical use is that it reduces the potential for recolonization of waterlines between treatments, although the potential still exists and water quality should be monitored to ensure that it is of acceptable quality (O'Donnell et al. 2005; Schel et al. 2006; Szymańska 2006; Bansal et al. 2011). In a study conducted in 2015 (Ditommaso et al. 2016), we found that although a significant proportion of the DUWLs tested were disinfected, samples taken from them were more heavily contaminated with bacteria and that their bacterial levels were much higher than the Italian (G.U. 52 2001) and European quality standards (European Union 1998) set for drinking water (≤100 colony forming units per millilitre (CFU ml⁻¹) of heterotrophic bacteria at 22 °C and ≤20 CFU ml⁻¹ of mesophilic bacteria at 37 °C). Furthermore, the rate of recovery of Legionella from the DUWLs observed in our study greatly varied depending on the analytical method used. When we used the culture method to detect Legionella in the DUWL samples, we observed a low rate of contamination (6.6%); however, when we used propidium monoazide quantitative polymerase chain reaction (PMA-qPCR) for the same purpose, we found that 100% of the samples were positive. Cultivation has been the principal approach employed in the past, but the culture may result in false negative data or underestimated bacterial counts. In contrast, PCR detection methods also include viable but non-culturable (VBNC) legionellae, and legionella living within amoeba (Ng et al. 1997; Bates et al. 2000).

During the collection of samples (Ditommaso et al. 2016), information was also collected relating to waterline disinfection. We found that the most commonly used product for bacterial contamination control was hydrogen peroxide at various concentrations. Only one dental office had adopted a continuous disinfection system called ICX^® (A-dec Inc., Newberg, OR, USA), but none of the dental units treated with ICX^® tablets delivered potable water and were legionella free (negative by culture method and positive by PMA-qPCR).

From this finding, we decided to: (1) evaluate the chemical treatment of waterlines using a protocol that combines shock treatment with hydrogen peroxide and continuous treatment with ICX^® tablets for effective control of microbial contamination; and (2) investigate the in vitro antimicrobial activity of ICX^® tablets on collection strains and wild strains isolated from dental chair unit (DCU) output water.

In this study, we monitored two 3-year-old DCUs (DCU n°1 and n°2) and one 17-year-old chair (DCU n°3) as a control. The water source that fed all three DCUs was tap water (municipal water). All chairs had been in daily use; DCU n°1 and n°2 for conservative dentistry and dental surgery, and DCU n°3 for dental hygiene.

Both DCUs n°1 and n°2 were equipped with a self-contained dental unit water system that included a 2-litre water bottle that received ICX^® (A-dec Inc., Newberg, OR, USA) tablets added to the water reservoir bottle at each refill. These DCUs had been sanitized with ICX^® tablets from the date of installation in the year 2012. From 2012, these DCUs had never had biofilm removed or planktonic bacteria reduced using shock treatment. DCU n°3 had never been cleaned and used only municipal water as the irrigant.

ICX^® are effervescent tablets (sodium percarbonate 6.96%, dimethyl benzyl ammonium chloride 0.85%, dimethyl ethylbenzyl ammonium chloride 0.85%, silver nitrate 0.14%, and other ingredients 91.2%) specially formulated to maintain DUWLs and prevent accumulation of odours and foul-tasting bacteria and maintain DUWLs effluent ≤10 CFU ml⁻¹, as long as proper infection control is followed. The self-contained reservoir bottle was filled with 2 L of fresh tap water and one ICX^® effervescent tablet (212.4 mg). A tablet placed in the self-contained water delivery system remains effective in the waterlines for up to 2 weeks. Additionally, the manufacturer reports that it ‘eliminates the need to purge waterlines at night,’ and the tablet remains ‘continuously present in the water system and provides a preventive, proactive solution rather than a reactive one’ (A-dec Inc. 2015).

This product is intended for use with potable water and in conjunction with regular dental unit water testing, following shock treatment with antimicrobial products recommended by the manufacturer.

From March to September 2015, 18 water effluents were collected monthly in a private dental office. Samples were taken in the morning from the air-water syringes and turbines and were mixed together. Each sample was collected in a sterile, 1-litre plastic bottle containing thiosulfate sodium (10% w/v).

Samples were also taken initially from the sink faucet to compare the CFU counts of the water supplied to the building with CFU counts of DUWLs. Only potable water provided for drinking, bathing or culinary purposes was supplied to plumbing fixtures of the building in which the dental office was located.

Before sampling, the taps were disinfected according to the following procedure: (1) removal of the water flow regulator, (2) internal disinfection of the tap with a solution of sodium hypochlorite at (10% w/v) for 2–3 min, (3) disinfection of the whole faucet with a Bunsen flame, and (4) water flow for 5 min. Finally, the water samples were collected in a sterile, 2-litre plastic bottle containing thiosulphate sodium (10% w/v).

To assess the effectiveness of shock treatment, the bacteria total viable counts (TVCs) in DUWL water samples were investigated according to ISO 6222 (International Standards Organization 1999). One millilitre of the water samples and 1 ml of a decimal dilution of the samples were tested using the pour plate method on yeast extract agar. The calculation of the number of CFU ml⁻¹ of the sample was performed after 7 days of incubation at 22 °C and 5 days at 36 °C according to the US standard method (Eaton et al. 1995). If there were more than 300 colonies on the plates inoculated with the decimal dilution, the results were expressed as >3,000 CFU ml⁻¹.

The DCUs underwent intensive disinfection using 3% hydrogen peroxide, which was kept present in all waterline elements for 40 min due to the continual flow of water from the reservoir to the handpieces. Subsequently, fresh water is added to the bottle container, and from there, is used to flush the instrument waterlines and cup fill line. A total of 8 L of fresh main water is used to remove all hydrogen peroxide residue. After a rinsing cycle, the DCUs are ready for use. To assess the effectiveness of shock treatment, water samples were collected immediately following the first disinfection cycle and after 30 days of shock treatment; for each sample, the TVCs were measured. Later, throughout the study period (5 months of monthly shocks), samples were obtained 30 days after shock treatment. During the 6-month follow-up, chairs n°1 and n°2 received the tablet in addition to the treatments, while chair n°3 received only the shock treatment with hydrogen peroxide. The follow-up for DCUs n°2 and n°3 was extended for another 4 months.

To evaluate the occurrence of microbial resistance to ICX^® we compare bactericidal efficacy of ICX^® on planktonic population living in DUWLs of DCU n°1, n°2 and n°3.

Output water samples from DCUs were collected from chairs n°1 and n°2 after the lack of ICX^® in waterlines for 72 hours. We enumerated the background bacteria (N₀) according to ISO 6222 and also assessed the bactericidal efficacy of ICX^® on planktonic population living in DUWLs (bacteria that are released from the biofilm and circulating in the waterline) according EN 1040.

Laboratory tests were conducted by dissolving ICX^® tablets according to the manufacturer's instructions in output water sampled from each DCU. After 60 minutes and 24 hours of contact time, 1 ml of the mixture was neutralized using Dey-Engley, serially diluted tenfold and inoculated in duplicate using the pour plate technique. The numbers of surviving bacteria (N_a) were enumerated and lgR was calculated (Equation (1)).

We decided to investigate the diversity and distribution of aerobic bacterial species in DUWLs' output water from dental chairs to determine whether the relative abundance of individual species is associated with the failure of waterline disinfection observed in the first part of the study described above.

At the same time, we identified the microorganisms present in the output water samples collected from DCUs before and after the in vitro contact with ICX^®. Selected examples of the colony types present on each plate were purified by subculture and stored on nutrient agar slopes in the dark for subsequent identification; the colonies were Gram-stained, and then identified using Matrix Assisted Laser Desorption Ionization Time-Of-Flight (MALDI-TOF) (Carbonnelle et al. 2011; Biswas & Rolain 2013) and API 20 NE (BioMérieux, Marcy l'Etoile, France).

Differences between the microbial loads detected in the three dental units were tested using standard one-way analysis of variance (ANOVA) with software package SPSS. A p value <0.05 was considered statistically significant.

At baseline, samples from DCUs showed massive contamination with a high bacterial count at both 22 °C and 37 °C. Only samples from the taps were found to satisfy the Italian standards for drinking water.

Post-intervention monitoring samples (taken the same day as the first shock) achieved the lowest levels of TVCs at 37 °C and 22 °C, conforming to the standard. However, after 30 days of shock treatment, chairs began to show an increase in TVCs. During the microbiological surveillance, the post-intervention monitoring (samples taken after 30 days from each treatment) produced the results reported in Table 1.

We performed one-way ANOVA on data obtained during the 6 months of monitoring; the samples collected from DCU n°1 and n°2 and samples collected from DCU n°3 did not differ significantly on the basis of TVCs at 37 °C and 22 °C (37 °C p = 0.86; f between treatment = 0.15; 22 °C p = 0.52; f between treatment = 0.67).

Additionally, in the following 4 months, there was no significant difference between the results obtained from DCU n°2 and DCU n°3 (37 °C p = 0.63; f between treatment = 0.24; 22 °C p = 0.78; f between treatment = 0.09).

At the concentration recommended by the manufacturer, the product caused a 4.0 log reduction of S. aureus and a 1.3 log reduction of P. aeruginosa within 60 min of contact time. However, the reduction in number of bacterial cells of S. aureus strains was approximately 7.6 log after 24 hours of contact time, while in the P. aeruginosa strain, the reduction in the number of bacterial cells after 24 hours remained unchanged (Table 2).

Bacterial loads of output water from all DCUs were evaluated quantitatively and qualitatively before and after disinfection. Before in vitro treatment, all baseline TVCs measured in samples from the three chairs were above the Italian drinking water standard. After in vitro exposure to ICX^®, only planktonic bacteria from chair n°3 were completely killed (lgR = 4.3), while in DCU n°1 and n°2, the bacteria survived (Table 3).

The bacterial species recovered from the samples taken from the dental chairs were all Gram-negative and belonged to the families of aquatic and soil bacteria. Among our isolates, Delftia acidovorans, Stenotrophomonas maltophilia, Mycobacterium flavescent, Listeria innocua, Comamonas testosteroni, Pseudomonas putida, Pseudomonas fluorescens, Vibrio metschnikovii, Sphingomonas adhaesiva, Vibrio alginolyticus, and Aeromonas salmonicida were identified, and some of these are opportunistic human pathogens (Hsueh et al. 1998; Martins et al. 2010; Brooke 2012; Kim et al. 2012; Orsini et al. 2014). After in vitro exposure to ICX^®, only planktonic bacteria from DCU n°3 were completely killed, while Delftia acidovorans, Mycobacterium flavescent, Stenotrophomonas maltophilia, and Listeria innocua survived in DCU n°1 and n°2 (Figure 1).

As previously described in the literature, the only method to lower bacterial counts in dental water to acceptable levels is to permanently eliminate the existing biofilm inside the tubes and prevent biofilm formation in new units. Most of the cleaners and disinfectants do not effectively remove biofilms: Walker & Marsh (2007) appraised a range of chemical DUWL treatment agents and reported that only a few disinfectants successfully remove biofilm and consistently reduce the microbial load of DUWL output water to <200 CFU ml⁻¹. However determining the bacterial load using culture does not predict the exact size of the problem because bacteria in a VBNC state cannot grow on standard growth media.

Different methods have been developed to avoid contamination of DUWLs and dental chairs by pathogens and environmental organisms, with one being chemical disinfection but frequent application may result in the development of disinfectant/biocides resistance contributing to resistance to antibiotics by co-selection of antibiotic resistance genes (Russell 2004).

Cleaners intended for continuous use are less concentrated, and these cleaners are introduced after an initial shock treatment that has acted upon the biofilm in the waterlines, or they are used with new waterlines where biofilms have not yet formed. The literature contains only two accounts (McDowell et al. 2004; Meiller et al. 2004) of the properties, germicidal effectiveness, and potential uses of the chemical waterline cleaner ICX^® tablets. These studies, performed in a simulated use in DUWLs demonstrate that in the presence of a bacterial challenge of 100 to 1,000 CFU ml⁻¹ in incoming water, ICX^® effectively prevented the development of biofilm and maintained the water quality at a level consistently well below 200 CFU ml⁻¹. Moreover, the treatment prevented biofilm build-up during daily use and over weekends during a period of inactivity.

A-dec recommends in their Waterline Maintenance Guide a ‘Maintain, Monitor, and Shock’ approach to waterline care. This approach involves daily maintenance with ICX^®, monitoring water quality regularly with devices such as the Millipore Heterotrophic Plate Counter (HPC) Sampler, and shocking the waterlines when ‘test results are greater than the water quality action level.’

As a part of laboratory procedure, in the first stage of our study, the number of heterotrophic microorganisms in each water sample collected after the first shock treatment with hydrogen peroxide and in 6 months of monitoring was evaluated. However, the effects of a continuous disinfecting system proposed by the manufacturer for a daily disinfection of waterlines were not demonstrated in this study; no significant differences were found between the bacterial loads (TVCs at 37 °C and 22 °C) detected in the effluent water of the two dental chairs that received the ICX^® tablet versus chair n°3. As a result, if dental offices are using disinfectant waterline tablets, shock treatment may be needed before 30 days. These results are similar to the finding of Bowen et al. (2015) who compared ICX^® and Citrisil disinfectants in one clinical setting. They reported that bacterial loads increased to unacceptably high levels within 1–2 weeks of treatment, with a gradual increase after disinfection treatments ceased.

To explain this poor performance, we tested the product versus the collection strains and environmental strains in vitro. For Staphylococcus aureus, ICX^® achieved a four decimal log reduction at 60 min and a seven decimal log reduction after overnight treatment (24 hours). This last contact time reflects a realistic period of inactivity in which the chemical product (although at low doses) could prevent the proliferation of bacteria. Instead, the product is ineffective in reducing the number of Pseudomonas aeruginosa cells after 60 min and 24 hours of contact time.

This inability to kill Gram-negative bacteria is a disadvantage because most of the bacterial species found in DUWL output water are Gram-negative. Furthermore, it has been recognized that in vivo bacteria are found predominantly attached to surfaces (biofilm), hence being more resistant to heat, dehydration, ultraviolet (UV) light, disinfectants, antibiotics, etc.

It is known that opportunistic pathogens may account for more than 30% of the total bacterial populations in water distribution systems (LeChevallier et al. 1980), and these bacteria have all been implicated in waterborne nosocomial infections (Barbeau et al. 1996).

A detailed analysis of the output water from these three DCUs showed that the most prevalent bacterial species recovered were Gram-negative, and when we treated DCU output samples in vitro with ICX^®, we obtained a partial reduction of counts in dental chairs continuously treated with ICX^® tablets, but the total killing of bacteria was observed from the output water of dental chair n°3.

Biofilms proved to be difficult to control despite all of the measures in this study. Our results show that the continuous introduction of ICX^®, used at low levels to minimize potential toxic effects, was not effective in maintaining the heterotrophic bacterial counts within the recommended standards in the output water of dental devices, and shock treatment may be needed more frequently than monthly treatments. Moreover, the prolonged use of ICX^® tablets was not always able to control microorganisms that are very resistant to disinfectants due to their intrinsic characteristics, because they are protected within the biofilm or because of the development of biofilm tolerance which we observed in the growth of Delftia acidovorans and Stenotrophomonas maltophilia.

The only way to lower the bacterial counts in the dental water to acceptable levels is to permanently eliminate the existing biofilms inside the tubes and to ensure that the frequency of the treatment is sufficient to successfully maintain a good quality of water when performing periodic water testing; therefore, routine monitoring is strongly recommended.

We wish to thank Drs T. Zaccaria and R. Cipriani for technical help in using Matrix Assisted Laser Desorption Ionization Time-Of-Flight (MALDI-TOF).

This work was supported by a grant from the Department fund ZOTC1PRE15, Department of Public Health and Pediatrics, University of Turin.

This research received no specific grant from commercial, or not-for-profit sectors.

None of the authors has a financial relationship or conflict of interest to disclose.