Under the Safe Drinking Water Act, the Environmental Protection Agency, or EPA, has set a standard for drinking and recreational water at 500 colony-forming units per milliliter, or CFU/mL, of noncoliform bacteria.⁶ The ADA set a goal that dental water for non-surgical procedures should contain no more than 200 CFU/mL by the year 2000.⁷ As a result of the challenge issued by the ADA to manufacturers, a number of proprietary compounds formulated to reduce bacterial counts now are on the market. The products can be categorized broadly as continuous or intermittent use, and the efficacy of a number of products has been tested.

The most widely tested product, household bleach, used in 1:10 dilution in the lines, has demonstrated efficacy. The main disadvantage of bleach is that it can corrode metal parts over an extended period. Other products tested include hydrogen peroxide, povidone-iodine and essential oils. Efficacy of products vary greatly.^8–12

In search of a practical solution to reduce microbial contamination of DUWLs at The University of Texas Health Science Center at San Antonio, or UTHSCSA, Dental School Student Clinic, we conducted this study to test the efficacy of a continuous-use, stabilized chlorine dioxide proprietary compound to decrease the number of bacteria in DUWLs.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
We used three dental units in an outpatient dental clinic to test the efficacy of a continuous-use, stabilized chlorine dioxide compound in the spring of 2000. We used three other units in the same clinic as controls. All of the units were equipped with self-contained water systems and each had five separate waterlines: two handpiece lines, an operator’s water syringe line, an assistant’s water syringe line and an ultrasonic line. We received approval from the Institutional Review Board at UTHSCSA to conduct the study.

One week before the study began, we flushed the waterlines according to ADA and Centers for Disease Control and Prevention, or CDC, recommendations. At the beginning of each day, we flushed all of the lines for two minutes; after use on each patient, we ran the high-speed hand-pieces for a minimum of 20 seconds to discharge water and air; and at the end of each day, we air-purged all of the lines by reattaching the empty reservoir bottle to the dental unit and flushing lines until the flow of water had stopped.¹³

For the duration of the study, we flushed the waterlines in the three control units daily, according to the ADA and CDC recommendations as described previously. We gave the three test units an initial, one-time, full-strength treatment, according to the manufacturer’s recommendations. We filled the self-contained bottle with 100 mL of the stabilized chlorine dioxide product, ran it through each of the five waterlines in all three units and left it in the lines overnight. The next morning, we removed the self-contained bottle and replaced it with a bottle containing the daily-use formula. This involved adding 75 mL of the product to the 750-mL–capacity, self-contained reservoir bottle and filling the remainder of the bottle with tap water that met EPA drinking water standards, creating a 1:10 dilution. We used this solution daily as a continuous-use product for the remainder of the study.

We took water samples from the six units (three test units and three control units) at baseline and once a week thereafter for the remainder of the study, according to validated sampling methods.¹⁴ First, we removed the reservoir bottle containing the continuous-use product from the unit and discarded the remaining liquid. We then rinsed out the bottle with tap water, filled it with tap water and reattached it to the unit. Before sampling, we flushed the lines for 20 to 30 seconds if the dental unit had been used that day and for two to three minutes if the unit had not. We wiped the end of each waterline with an alcohol-soaked pad before collecting 100 mL of water aseptically from each unit in a sterile collection bottle containing sodium thiosulfate to neutralize the residual chlorine present. We collected approximately 20 mL from each of the five lines of each unit. After we completed the sample collection on each unit, we immediately removed the reservoir bottle, filled it with the daily-use formula, reattached it to the unit and ran the product through the lines for 20 to 30 seconds. We then deemed the units to be ready for patient treatment.

Whenever possible, we took the water samples to the laboratory immediately; if that was not possible, we refrigerated and plated them within four hours of collection. We made tenfold serial dilutions of each unit sample in a phosphate buffer solution and vigorously agitated the sample by vortex for 15 seconds. We then plated one-tenth of a milliliter of each dilution on R2A agar in triplicate, using the spread plate method. We left the plates to incubate at 22 to 28 C for seven days.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
We recorded the number of CFU/mL for each unit each week, beginning with baseline collection. We entered data into a database, using Microsoft Excel 2001 (Microsoft, Redmond, Wash.) spreadsheet software. After we completed data collection, we interpreted the data. We calculated the mean CFU/mL found in the treatment and control units. Figure 1 shows the typical appearance of the isolates recovered. Since the purpose of the study was to enumerate bacterial load, we did not identify specific organisms.

View larger version (147K): [in this window] [in a new window]   Figure 1. Typical appearance of colonies of heterotrophic, mesophilic bacteria isolated from dental units grown on low nutrient R2A agar plates.

View larger version (149K): [in this window] [in a new window]   Figure 2. Growth on R2A agar plates after eight weeks of treatment with a continuous-use chlorine dioxide cleaner changed to a preponderance of small, dark, shiny colonies.

View larger version (155K): [in this window] [in a new window]   Figure 3. Similar black colonies were isolated from source tap water.

Laboratory investigation. We sent the isolates we recovered from unit A pooled waterline samples and handpiece lines to The Fungus Testing Laboratory at UTHSCSA to have the fungus identified. We subsequently sent culture isolates from the source tap water for unit A, pooled samples for unit B and a sample from unit B’s ultrasonic line for species confirmation.

The laboratory confirmed the existence of the genus Exophiala by microscopic identification. Since the fungus could not be identified to the species level by macroscopic and microscopic examination, we had ribosomal DNA, or rDNA, sequencing performed at Louisiana State University in Baton Rouge. We compared the resultant sequence with an rDNA database at the Centraalbureau voor Schimmelcultures in Utrecht, Netherlands, and found it to be 100 percent homologous with Exophiala mesophila. Details of the laboratory investigations regarding the species will be published elsewhere.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
Because the ADA recommendation is to reduce the number of heterotrophic, mesophilic bacteria to 200 CFU/mL, quantitative laboratory analysis of water samples rather than qualitative analysis is performed routinely. The purpose of our study was to test the efficacy of the continuous-use chemical cleaning product used to reduce the bacterial load to the ADA-recommended level. However, the unexpected and unusual finding of fungal growth in treated units during the course of the study prompted us to investigate, identify and report on the fungus.

The process of fungus identification proved to be arduous because Exophiala mesophila has been described only once in the literature; it was isolated from silicone shower seals in a hospital ward in Hamburg, Germany, in 1996. The most remarkable characteristics of this species, compared with other Exophiala species, are that it grows at a maximum temperature of 32 C and up to a pH of 9.5.¹⁵ When the identification of the species was complete in our study, we confirmed that the pH of the undiluted product was 6.3 and that the pH was 7 when the product was diluted with water for continuous use.

The fungus was barely detectable in the source tap water, but it seems to have proliferated in conditions that were more favorable for growth.

Since this Exophiala species is newly described, nothing is known about its clinical significance. It has not been described previously as a human pathogen. The genus Exophiala does include species with global distribution that are pathogenic to animals or humans. Exophiala moniliae is an agent of subcutaneous abscesses and skin lesions, and Exophiala pisciphila is pathogenic in fish.

Fungal infections are becoming more common in hospitalized patients.¹⁶ One report described more than 20 nosocomial cases of fungemia caused by infection with Exophiala jeanselmei alone or in combination with other dematiaceous, or dark, fungi.¹⁷ These fungi are considered to be of low virulence, because they can persist in the skin of the host for months without disseminating to other organs. All of the infected people were immunocompromised due to cancer or AIDS, and all, with the exception of one patient, responded to antifungal therapy.

A recent study reported that Exophiala was one of three fungal genera that was disseminated efficiently by public drinking water.¹⁸ In our study, the fungus was barely detectable in the source tap water, but it seems to have proliferated in conditions that were more favorable for growth. Chlorine dioxide is considered to be a viable alternative to chlorine as a primary disinfectant in drinking water. It is not known to react with organic substances to form trihalomethanes.¹⁹ Oral rinses that contain chlorine dioxide in proprietary formulations are used commonly because of their demonstrated ability to reduce levels of Streptococcus mutans, Lactobacilli and Candida albicans and to control halitosis.^20,21

The amount of reactive chlorine dioxide in the product used in our study is unknown because it is stabilized in a proprietary formulation. The concentration of the formulation is 1,000 parts per million, or ppm, full strength and 100 ppm when diluted for continuous treatment. Studies have tested the efficacy of different concentrations of reactive chlorine dioxide. One study concluded that intermittent treatment with 50 ppm resulted in a temporary decrease in bacterial counts, but it did not maintain a long-term reduction in levels.²² Consistent with these findings, our study showed a decline, but one that was not consistently at the ADA-recommended levels. Details of the efficacy of the product used in our study will be published elsewhere. Puttaiah and colleagues²³ demonstrated successful disruption and removal of mature biofilms at concentrations of 140 ppm of chlorine dioxide. They also found that water quality was safe for routine patient treatment when 2 to 3 ppm chlorine dioxide was added to municipal water samples.

The rationale for the initial, overnight, full-strength treatment we used in our study was to disrupt or remove existing biofilm. This was followed by a 1:10 dilution with tap water used daily thereafter to maintain low bacterial counts. The initial dose may not have been fungicidal or may not have removed existing biofilm completely, or the daily-use concentration may have altered the natural water flora sufficiently to provide critical environmental conditions for propagation of this unusual fungus. At a pH of 7, the environmental condition was favorable for fungal growth.

Although the finding of organisms growing in opportunistic conditions was unexpected in this case, the phenomenon is not new to the dental profession. The development of oral candidiasis in patients who are receiving broad-spectrum antibiotic therapy is common. While the presence of fungi in biofilms in contaminated waterlines has been confirmed using electron microscopy for a number of decades,²⁴ we report the first-known incidence of fungal growth in treated waterlines. DUWL cleaners, many of which have U.S. Food and Drug Administration clearance for marketing, undergo testing in laboratory conditions, in simulated clinical settings or in clinical settings for limited periods. Research on the long-term consequences of adding chemical waterline cleaners in clinical settings is lacking.

One action practitioners can take is to dilute continuous-use products with sterile water, though usually this is too costly for institutions with large numbers of operatories. Caution should be exercised when using this option because water stagnates during periods of inactivity in the nonsterile tubing and eventually biofilm will recur.^5,25 Since the findings from our study indicate that DUWL treatment with continuous-use chemicals may select for the growth of fungi in biofilms, practitioners should monitor DUWL quality regularly if they choose to treat waterlines with continuous-use products. Otherwise, practitioners may have a false sense of DUWL quality when they invest financial and personnel resources into using these products.

As previously mentioned, Exophiala organisms have been known to cause infection in people who are immunocompromised. With advances in medical treatment in recent decades, many ambulatory patients who are immunocompromised can be treated routinely in the dental operatory. Therefore, practitioners need to continually monitor the DUWL quality to avoid this potential source of infection.

	CONCLUSIONS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
We did not isolate Exophiala mesophila in the control units at any time during the study, suggesting that treatment of DUWLs with a continuous-use waterline cleaner may alter the natural water flora and promote the growth of a fungus that already may be present in small amounts in the municipal water supply. Further research on the long-term consequences of treating waterlines with continuous-use chemicals is needed.