UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Microbial biofilms in dental unit water lines Noce, Laura 1998

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_1998-0564.pdf [ 7.88MB ]
Metadata
JSON: 831-1.0088607.json
JSON-LD: 831-1.0088607-ld.json
RDF/XML (Pretty): 831-1.0088607-rdf.xml
RDF/JSON: 831-1.0088607-rdf.json
Turtle: 831-1.0088607-turtle.txt
N-Triples: 831-1.0088607-rdf-ntriples.txt
Original Record: 831-1.0088607-source.json
Full Text
831-1.0088607-fulltext.txt
Citation
831-1.0088607.ris

Full Text

MICROBIAL BIOFILMS IN DENTAL UNIT WATER LINES by LAURA NOCE DMD, University of Montreal, 1994 Multidisciplinary Residency Certificate, McGill University, 1995 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES (Department of Oral Biological and Medical Sciences) We accept this thesis as conforming to the required standard.  THE UNIVERSITY OF BRITISH COLUMBIA June 1998 © Laura Noce, 1998  In  presenting  degree freely  at  this  the  University  available  copying  of  department publication  for  this or of  thesis  this  partial  of  British  reference  thesis by  in  for  his thesis  and  of  scholarly  or for  §)^/UT/ST'fi  The University of British Vancouver, Canada  Date  DE-6  (2/88)  Columbia,  study.  her  of  I further  purposes  financial  V  Columbia  gain  the  requirements  I agree  that  agree  may  be  It  is  representatives.  permission.  Department  fulfilment  shall  not  that  the  Library  permission  granted  by  understood be  for  allowed  an  advanced  shall for  the that  without  make  it  extensive  head  of  my  copying  or  my  written  ABSTRACT Microbial biofilms located on the lumen surface of dental unit water lines have resulted in the persistent and widespread microbial contamination of the outflowing water. The aims of this study were to 1) evaluate the water quality delivered by dental units at the University of British Columbia 2) evaluate the effect of plating and incubation conditions on heterotrophic plate counts and 3) evaluate the effectiveness of marketed products in reducing microbial counts in water used for nonsurgical and surgical dental procedures.  Over a 12 week period, water  samples were collected weekly from the air/water lines, highspeed lines and sinks (controls) of 8 randomly chosen dental units used for nonsurgical dental procedures. Following serial dilutions, samples were plated on R A agar using the spread plate technique, and incubated aerobically at 2  35°C for 7 days. Individual colonies that were isolated were characterized based on colony morphology, Gram stain and oxidase test results. Representative samples of dental tubing water lines were subsequently processed for scanning electron microscopy (SEM) and evaluated for the presence of biofilm. In order to determine the effect of culture conditions on heterotrophic plate counts, different culture media, plating techniques, incubation time and temperature and chlorine neutralization were evaluated.  In addition, devices such as in-line filters, independent water  reservoirs, chemical disinfection and fully autoclavable systems were evaluated for their effectiveness at reducing the heterotrophic plate counts. Scanning electron microscopy was used to visualize the surface of in-line filters after several minutes of use, and to determine if 0.12% chlorhexidine injected in post filter tubing was able to prevent biofilm formation. The results of these studies revealed that while all water samples collected from the cold water sink taps met the standards for potable water, water collected from the warm water sink taps, the air/water ii  lines and the highspeed lines revealed extremely high heterotrophic plate counts and failed to meet the potable water standards in all but two instances.  Although the plating techniques  analyzed did not differ significantly, the incubation conditions had very significant effects on heterotrophic plate counts and dramatically affected the number of units that failed to meet potable water standards.  The evaluation of products which are currently being marketed as  effective ways to improve dental unit water line quality appear promising, however, aspects of their design may still leave concerns for some practitioners.  iii  TABLE OF CONTENTS ABSTRACT  ii  T A B L E OF CONTENTS  iv  LIST OF T A B L E S  vii  LIST OF FIGURES  viii  G L O S S A R Y OF T E R M S  ix  ACKNOWLEDGEMENTS  xi  C H A P T E R ONE - L I T E R A T U R E REVIEW 1.1 Microbial Biofilm Definition and Distribution of Microbial Biofilms Biofilm Formation and Structure Resistance of Biofilm Bacteria to Antibacterial Agents 1.2 Clinical And Environmental Significance of Microbial Biofilm Nosocomial Infections Sources of Dental Unit Water Contamination Characteristics of the Dental Units and their Water Lines which Enhance Biofilm Formation 1.3 Microbiology Of Dental Unit Water Lines Legionella Species Pseudomonas Species Mycobacterium Species Amoebas 1.4 Risk Factors For Disease Transmission From Dental Unit Water Lines Exposure to Pathogen Vs. Disease Transmission Risk Groups 1.5 Current Recommendations Potable Water Standards Recommendations for Non-Surgical Procedures in Dentistry Recommendations for Surgical Procedures in Dentistry 1.6 Methods Available To Reduce Dental Unit Water Contamination Autoclavable Systems Independent Water Reservoirs Flushing of Dental Unit Water Lines Chemical Disinfection Anti-Retraction Valves Microfiltration Devices Ultraviolet Irradiation  1 1 1 2 ...4 5 6 7 11 13 15 20 22 23 24 24 25 26 26 27 28 29 30 30 32 34 36 37 39 iv  Ethanol and Air Drying Rubber Dams Chair Design and Maintenance  39 39 40  CHAPTER TWO - AIMS OF T H E STUDY  41  CHAPTER THREE - MATERIALS AND METHODS 3.1 Water Collection Protocol Non-Surgical Units Surgical Units 3.2 Dilution, Plating and Culture Protocol Spread Plate Technique Pour Plate Technique 3.3 Effect of Sodium Thiosulfate on Heterotrophic Plate Counts 3.4 Evaluation of Water Flow Rate 3.5 Microbial Analysis 3.6 Scanning Electron Microscopy 3.7 Data Analysis  43 43 43 44 46 46 47 48 48 49 49 50  C H A P T E R FOUR - RESULTS 4.1 The Effect of Culture Conditions and Techniques on Heterotrophic Plate Counts Overview of the Dental Building Water Quality Spread Plates Vs. Pour Plates The Effect of Incubation Time and Temperature on Heterotrophic Plate Counts Sodium Thiosulfate Neutralization of Chlorine Microbial Analysis 4.2 Long Term Evaluation of Water Quality Used for Non-Surgical Procedures Scanning Electron Microscopy 4.3 Long Term Evaluation of Water Quality Delivered with Point of Use Filters Scanning Electron Microscopy 4.4 Evaluation of Water Quality Used for Surgical Periodontal Procedures Independent Water Reservoirs Scanning Electron Microscopy Fully Autoclavable Systems  52  C H A P T E R FIVE - DISCUSSION 5.1 The Effect of Culture Conditions and Plating Techniques on Heterotrophic Plate Counts 5.2 Long Term Evaluation of Water Quality Used for Non-Surgical Procedures 5.3 Long Term Evaluation of Water Quality Delivered with Point of Use Filters 5.4 Evaluation of Water Quality Used for Surgical Periodontal Procedures  52 52 53 56 59 60 63 66 68 71 73 73 75 77 78 78 82 85 86  C H A P T E R SIX - CONCLUSIONS  89  REFERENCES  92  APPENDICES Appendix 1  102  Appendix 2 Appendix 3 Appendix 4 Appendix 5 Appendix 6 Appendix 7 Appendix 8  Heterotrophic Plate Counts for Water Collected from Different Sources Heterotrophic Plate Counts on Spread Plates and Pour Plates Effect of Incubation Time and Temperature on Heterotrophic Plate Counts The Effect of Sodium Thiosulfate on Heterotrophic Plate Counts Heterotrophic Plate Counts for Water Collected from the Main Clinic Air/Water (AW) and Highspeed (HS) Lines Without Filters... Heterotrophic Plate Counts for Water Collected from the Main Cinic Air/Water (AW) and Highspeed (HS) Lines With Filters Water Flow Rate through Air/Water (AW) and Highspeed (HS) Heterotrophic Plate Counts for Surgical Units Supplied by Sterile Water Delivered by an Independent Water Reservoir  102 103 105 106 107 109 110 Ill  vi  LIST OF TABLES Table t - Percentage of Units that Meet Potable Water Standards (<500 CFU/mL) Based on Culture Conditions Table 2 - Effect of Sodium Thiosulfate on Heterotrophic Plate Counts Table 3 - Characterization of Bacteria Cultured from Highspeed Lines, Clinic and Building Sinks Table 4 - Long-Term Evaluation of Heterotrophic Plate Counts with and without the Use of In-Line Filters Table 5 - Heterotrophic Plate Counts for Surgical Units Supplied by Sterile Water Delivered by a Fully Autoclavable System  57 59  60 69  77  vii  LIST OF FIGURES Figure 1 - Water Collection Time-Table  45  Figure 2 - Heterotrophic Plate Counts for Water Collected from Different Sources  53  Figure 3 - Comparison of Heterotrophic Plate Counts Utilizing the Spread Plates Vs. Pour Plates Techniques. A) Air/Water Lines; B) Highspeed Lines Figure 4 - Effect of Incubation Time and Temperature on Heterotrophic Plate Counts  55 57  Figure 5 - Effect of Incubation Temperature on Colony Types Found on Heterotrophic Plates. A) 21°C, B) 28°C; C) 35°C Figure 6 - Percentage of Possible Pseudomonads in Water from Different Sources. A) Highspeed Lines (main clinic); B) Clinic Sinks (main clinic); C) Building Sinks (dentistry building)  62  Figure 7 - Heterotrophic Plate Counts for Water Collected from the Main Clinic A) Air/Water Lines; B) Highspeed Lines  64  Figure 8 - Heterotrophic Plate Counts for Water Collected from the Main Clinic  66  Figure 9 - S E M Analysis of Dental Unit Water Line Tubing Lumen. A) New Dental Unit Water Line; B,C) Air/Water Line from Main Clinic; D,E) Highspeed Line from Main Clinic  67  Figure 10 - Clinical Photograph of Two In-Line Filters. Filter on left is new. Filter on right has been used for 5 minutes in the Graduate Periodontics Clinic (prior to establishment of a water line disinfection protocol)  69  Figure 11 - Effect of Filters on Water Flow  70  58  Figure 12 - S E M Analysis of A,B) Surface of New In-Line Filter; C,D) Surface of Filter after 45 minutes of Continuous Use in the Main Clinic; E,F) Lumen of Highspeed Line, Downstream from the Filter 72 Figure 13 - Heterotrophic Plate Counts for Surgical Units Supplied by Sterile Water Delivered by an Independent Water Reservoir. A) Highspeed Lines; B) Cavitron Lines  74  Figure 14 - A , B ) S E M Analysis of the Surface of an In-Line Filter after 5 Minutes of Use in the Graduate Periodontics Clinic 76  GLOSSARY OF TERMS Biofilm  Composed of millions of microorganisms (bacteria, fungi, algae and protozoa) that accumulate on surfaces in aqueous environments. These film forming microbes excrete a glue-like substance that anchors them to materials such as metals, plastics, tissue and soil particles.  Biofouling  Contamination linked to microbial activity.  Coliform Bacteria  Gram negative, non-sporing facultative rods that ferment lactose with gas formation and belong to the genus Enterobacteriaceae. They are used as indicator organisms in water and in some foods.  Colony Forming Unit (CFU)  Minimum number of separable cells that give rise to a visible colony (progeny of the order of tens of millions of cells in number) on the surface or in solid or semi-solid agar. CFUs may consist of single cells, pairs, chains or clusters of cells.  Glycocalyx  Extracellular polymeric matrix produced by biofilm bacteria.  Heterotrophic Bacteria  Bacteria that require an organic carbon source for growth.  Heterotrophic Plate Count  Procedure for estimating the number of live heterotrophic bacteria in water. It may be determined by pour plate, spread plate, or membrane filter method.  Membrane Filter  Culturing technique which may be used to determine the heterotrophic plate count. It involves pouring the water sample through a sterile 0.45 pm, gridded membrane filter, under partial vacuum. The filter is then placed on the agar in a pertrf dish.  Mesophile  An organism that grows best at 20 to 55°C.  Nosocomial  Hospital aquired.  Opportunistic Microorganism One that uses the opportunities offered by weakened defense mechanisms to inflict damage on the host. A n opportunist may cause infection exclusively in compromised hosts, more frequently in compromised than in normal hosts or it may cause more severe infections in compromised than normal hosts.  Culturing technique which may be used to determine the heterotrophic plate count. It involves pouring the sterile melted medium over the water sample, and mixing the contents. This allows the colonies to form within the medium. An organism living upon dead or decaying organic matter. Culturing technique which may be used to determine the heterotrophic plate count. It involves spreading the water sample over the surface of the sterile solid or semi-solid medium. This allows the colonies to form on the surface of the medium.  ACKNOWLEDGEMENTS  I am very grateful to Dr. Edward Putnins, my thesis supervisor, for his excellent leadership, dedication and patience during the course of this project. I would also like to thank the members of my committee, Dr. Doug Waterfield, Dr. Michael Noble and Dr. Hannu Larjava for their suggestions and encouragement.  Finally, I would like to express my gratitude to Bruce  McCaughey, for his expertise and assistance with the photographic material included in this document.  xi  Chapter One - Literature Review 1.1 Microbial Biofilm  Definition and Distribution of Microbial Biofilms Biofilms are matrix-enclosed bacterial populations adherent to each other and/or to surfaces or interfaces. They are ubiquitous in nature and can be found in any nutrient-sufficient environment where moisture and a nonsterile solid surface exist. This includes the surfaces associated with natural water environments in streams, lakes and oceans, as well as those associated with "domestic-industrial" water environments such as water lines, sewer systems, drain pipes, wells, septic tanks, sewage treatment facilities, water storage containers, humidifiers and spray heads (Costerton et al., 1987; Costerton et al., 1995).  Planktonic and sessile microbes are two distinctly different communities which are present in aquatic ecosystems. Planktonic microbes are those found free floating in the water, while sessile microbes are those which may form biofilm by attaching to a surface (Miller, 1996).  In an  aquatic ecosystem, bacteria have a tendency to adhere to surfaces and the proportion of sessile bacteria far exceeds planktonic bacteria (Costerton et al., 1995). Biofilm cells are phenotypically, structurally and functionally distinct from their planktonic counterparts (Costerton et al., 1987; Costerton et al., 1995). Differences between sessile and planktonic bacteria include growth rate, composition and structure of cell walls, immunogenicity, enzymatic activity and susceptibility to antimicrobial agents and host defense mechanisms (Costerton et al., 1987; Costerton et al., 1995;  1  Khardori & Yassien, 1995). Planktonic cells are not attached to a substrate and are capable of colonizing new surfaces.  Sessile cells have a more active reproduction rate and general  metabolism (Costerton et a l , 1987; Costerton et a l , 1995). Microorganisms in biofilms adapt quickly to adverse environmental conditions, and this developmental strategy optimizes their probability of survival (Costerton et a l , 1987; Costerton et a l , 1981). The extent of biofilm accretion on surfaces in any aquatic system is dependent on nutrient availability which is required for cell replication and exopolysaccharide production (Costerton et al., 1995). In comparison to planktonic microorganisms, sessile microbes have several survival advantages: (1) they are retained on surfaces in a cooperative ecosystem; (2) they have a nutritional advantage, as organic and inorganic nutrients are bound by the biofilm matrix; and (3) they have a degree of resistance to antimicrobial substances, primarily as a result of the protection provided by the biofilm matrix (Costerton, 1995).  Biofilm Formation and Structure A biofilm is initiated when a solid surface is immersed in an aquatic environment, causing macromolecules and other low molecular weight hydrophobic molecules in the water to adsorb to the surface and form a conditioning film (Costerton et a l , 1987). This alteration in surface characteristics enhances bacterial adhesion.  The adhesion of bacteria and subsequent biofilm  formation is a rapid process, and is dependent on the nature of the surface and the microorganisms present in the fluid (Costerton et a l , 1995). Although bacteria initially adhere to surfaces in a reversible association, they eventually form an irreversible adhesion. A bacterial cell initiates the process of irreversible adhesion by binding to the surface using highly hydrated  2  exopolysaccharides, known as glycocalyx polymers. Following their adhesion to the surface, the bacteria surround themselves with additional glycocalyx and replicate within this matrix to form adherent microcolonies (Costerton et al., 1987; Costerton et al., 1995). These microcolonies, which represent the basic structural and functional unit of microbial biofilm, act as a nucleus for attracting other cells of the same or different bacterial strains (Khardori & Yassien, 1995). Recruitment of new bacteria from the planktonic phase and cell division within the microcolonies leads to the eventual coalescence of microcolonies which results in a continuous biofilm on the colonized surface.  While an individual bacterium is roughly one micron in size, biofilm is  typically thirty to fifty microns thick. The established gelatinous biofilm matrix is a substrate to which other microorganisms attach, it supplies the microorganisms with nutrients and protects them from the harrnful chemicals present in the overlying fluids (Costerton et a l , 1987). Bacterial populations living in this protected mode of biofilm growth produce planktonic cells, with much reduced chances of survival. These detached cells may colonize new surfaces or may burgeon to form large planktonic populations in those rare environments where nutrients are plentiful and bacterial antagonists are few (Costerton et al., 1987).  The architecture of biofilm is such that slime-enclosed microcolonies are interspersed between relatively open cell-free water channels, through which nutrients and microbial byproducts flow (Costerton, 1995).  Each biofilm bacterium lives in a customized microniche in a complex  microbial community that has primitive homeostasis, a primitive circulatory system, and metabolic cooperativity. Each of these sessile cells reacts to its special environment so that it differs fundamentally from a planktonic cell of the same species.  Microorganisms within a  3  biofilm community act as a consortium, with a variety of species working physiologically and metabolically together.  For instance, while one species may be responsible for gathering  nutrients, another may metabolize wastes, therefore eliminating them from the established biofilm (Costerton et a l , 1995).  Resistance of Biofilm Bacteria to Antibacterial Agents Biofilm microorganisms have a resistance to natural surfactants, phagocytosis, biocides and antibiotics, allowing them to survive long after all planktonic organisms have been killed by host defense factors and by antibacterial agents (Costerton et a l , 1987; Costerton et a l , 1995). The fact that bacteria from infections associated with permanent devices grow exclusively in the biofilm mode and are resistant to host defense factors (antibodies, surfactants, phagocytes) and vigorous antibiotic chemotherapy is strong evidence to support the benefits of biofilm growth for the survival of microorganisms (Costerton et a l , 1995). Costerton isolated bacteria from medical devices, and determined that 1500 times more antibacterial agent was required to kill the entrenched biofilm bacteria than their free-floating planktonic counterparts (Costerton, 1997).  The actual mechanisms of resistance still remain unclear, although three mechanisms are believed to be involved.  First, the thick extracellular matrix which envelops biofilm bacteria hinders  penetration of the antimicrobial agent, thus protecting the bacteria. It may do so by acting as a molecular sieve, by ionic interaction between the negatively charged matrix polymer and positively charged antimicrobial agents or by chemical reaction between the polymer and the agent leading to its inactivation (Blenkinshop et a l , 1992; Costerton et a l , 1987; Williams et a l ,  4  1993; Wilson, 1996). Second, nutrient availability within the biofilm may reduce the bacterial growth rate, thereby altering antibiotic sensitivity. Third, bacteria respond to changes in their environment by profound phenotypic variations in enzymatic activity, cell wall composition, and surface structure (Costerton et a l , 1987).  These phenotypic changes involve the target  molecules for biocides, antibiotics, antibodies and phagocytes, and also involve the external structures that control access of these agents to the targets.  Therefore, bacteria growing in  biofilms show alterations in susceptibility to these antibacterial agents (Costerton et a l , 1987).  Preliminary studies have shown that the bioelectric effect may aid in disrupting and killing biofilm. Combining a direct electrical current (DC) field with an antimicrobial agent enhances its penetration by charging its molecules and neutralizing repulsion, allowing it to penetrate deeper into the biofilms (Blenkinshop et a l , 1992; Costerton, 1997). Another study suggests enzyme degradation of biofilm may be used. However, due to the heterogeneity of biofilm extracellular polysaccharides, a mixture of enzymes may be necessary for a sufficient biofilm degradation (Johansen et a l , 1997).  1.2 Clinical and Environmental Significance of Microbial Biofilms  The microbial biofilm acts as a reservoir from which bacteria detach and contaminate the surrounding medium (Khardori & Yassien, 1995). Numerous environmental sites such as dental units, hospitals, industrial equipment, community water supplies and public recreational facilities are susceptible to this problem (Costerton et a l , 1987; Costerton et a l , 1995; Favero, 1984; 5  Hayes et al., 1989, Johansen et al., 1997; Keene et al., 1994, Khardori & Yassien, 1995; Lowry et al., 1991; MacKenzie et al., 1994).  Progressive biofilm accumulation causes increased  frictional resistance (biofouling) of fluid-carrying lines, biocorrosion of metallic substrata, energy loss, plugging of filters and porous media, generation harrnful metabolites such as H S , flow 2  modification in linear systems, food contamination and a variety of health concerns (Costerton et al., 1987; Costerton et al., 1995; Johansen et al., 1997). Of particular interest and, perhaps, relevance to the dental office are the documented reports of waterborne infection and disease in hospital settings.  Nosocomial Infections The use of various medical devices has greatly facilitated the management of serious medical and surgical problems. Materials such as plastic, rubber and metal are used to construct the myriad of medical devices and prostheses (Costerton et al., 1987).  However, the successful  development and clinical use of these synthetic materials in various body systems is accompanied by microbial biofilm formation. The ability of microorganisms to adhere to these devices and form biofilms which protects them from antimicrobial agents and the host defense mechanisms is a significant clinical concern.  Extensive bacterial biofilms have been found on a variety of  medical devices and prostheses such as sutures, dressings, wound drainage tubes, endoscopes, hemodialysis buttons,  hemasite  access devices, catheters, central venous lines, cardiac  pacemakers, bioprosthetic and mechanical heart valves, urine collection systems, nephrostomy tubes, urethral and biliary stents, penile prostheses, intrauterine  contraceptive devices,  endotracheal tubes, prosthetic hip joints and contact lenses (Costerton et al., 1987; Johansen et  6  a l , 1997; Khardori & Yassien, 1995).  Biofilms may spread up to 100 cm along colonized  biomaterials surfaces in as few as 3 days, in spite of active host defense systems and prophylactic doses of antibiotics (Costerton et a l , 1987).  Although biofilms on the biomaterial surfaces do not usually cause overt infection or even detectable inflammation, they do constitute a nidus of infection when the host defense system fails to contain them (Costerton et a l , 1987). Biofilm formation on intravascular catheters are the leading cause of nosocomial bacteremias. They lead to infections at the insertion site, such as septic thrombophlebitis, septicemia, endocarditis and metastatic abscesses.  Other common  nosocomial biofilm-related infections include osteomyelitis, cystic fibrosis and hospital-acquired pneumonia (Costerton et a l , 1987; Khardori & Yassien, 1995, Williams et a l , 1996c).  Sources of Dental Unit Water Contamination Dental unit water lines are the small gauge pipes or tubes that deliver water (used for cooling, rinsing and flushing during patient treatment) from the dental unit to the dentist's handpiece, air/water syringe and ultrasonic scaler. In the past, it was assumed that the water delivered by the dental unit water lines would be of comparable quality to the source water entering the dental unit. Thus, if the water entering the dental unit met the quality standards for drinking water, it was assumed that the water delivered by dental unit water lines would also meet these standards. However, a number of studies have determined that the water delivered by dental unit water lines usually contains a higher number of microorganisms than is acceptable for water to be considered potable, or safe, according to the current Canadian standards for water quality (Atlas et a l , 1995;  7  Barbeau et a l , 1996; Blake, 1963; Fitzgibbon et a l , 1984; Kelstrup et a l , 1977; McEntegart & Clark, 1973, Miller, 1996; Sciaky & Sulitzeanu, 1962; Williams et a l , 1993, Williams et a l . , 1996c).  Over twenty years ago, Kelstrup linked the microbial biofilm found on dental unit water line lumen surfaces with the occurrence of high microbial levels in dental unit water (Kelstrup et a l , 1977).  There is now little question that the most common source of dental unit water  contamination is the microbial biofilm found along the walls of the long, narrow-bore tubing that furnishes cooling and irrigating water to dental hand instruments (Blake, 1963; Gross et a l , 1976; Kelstrup et a l , 1977; Martin, 1987; Mayo et a l , 1990; Pankhurst & Philpott-Howard, 1993, Whitehouse et a l , 1991; Williams et a l , 1993; Williams et a l , 1995b; Williams et a l , 1996c; Williams et a l , 1994b). Stretching and bending of the water lines during use may lead to sloughing of dislodged sheets of biofilm, creating not only an increased turnover of adherent populations, but also gross contamination of downstream water as clumps of slime and bacteria enter into the water flow (Pankhurst & Philpott-Howard, 1993; Santiago et a l , 1994). Biofilms therefore  serve  as a reservoir,  significantly amplifying  the  numbers  of free-floating  microorganisms in the water exiting the water lines through dental handpieces, air-water syringes and ultrasonic scalers (Kelstrup et a l , 1977; Mayo et a l , 1990; McEntegart & Clark, 1973; Whitehouse et a l , 1991). Microbial counts in the range of one million microorganisms per milliliter of water have been reported (Atlas et a l , 1995; Challacombe et a l , 1995; Mayo et a l , 1990; Pankhurst & Philpott-Howard, 1993; Whitehouse et a l , 1991; Williams et a l , 1993; Williams & Kelley, 1994). This is at least 1000 times greater than the number recovered from  8  potable water from fountains and sink faucets in the same treatment room (Gross et al., 1976; Williams et al., 1993). These microbial accumulations can contribute to occasional objectionable odors and visible particles of biofilm material exiting the system. Furthermore, these clumps of biofilm that become dislodged may enter the patient's mouth, be aspirated or absorbed into a vascular lesion, or may be aerosolized into the environment.  There is increasing agreement  among dental authorities that this represents an unacceptable bacterial load to introduce into the mouths of patients.  Most of the microorganisms that contribute to the formation of biofilms in dental units are derived from the regrowth of the low numbers of residual viable cells that survive municipal treatment. However, other sources include oral microbes which are sucked into the water lines during patient treatment, plumbing systems, and microbes derived from improper handling of sterile water bottles when an independent water system is used.  Municipal Water Supply The majority of microorganisms contributing to biofilm formation in dental unit water lines are derived from regrowth of municipal water contaminants.  In fact, a recent study has  demonstrated that water samples taken from dental units in clinical use contain similar concentrations and types of microorganisms as water samples collected from units that have never been exposed to patients (Whitehouse et al., 1991).  9  Retraction of Patient Fluids The passage of oral microbes into dental water lines through retraction of patient fluids or tissue fragments was first acknowledged with concern by the American Dental Association in 1978 (Association Reports, 1978). Oral microbes, blood, saliva and other materials may be drawn into the handpiece and, possibly into the attached water line, as a result of the transient suction that develops in the dental unit water line when the handpiece is turned off in the mouth (Kono, 1996; Miller, 1996). Up to 0.9 mL of contaminated oral fluids and other detritus generated during a procedure could be drawn up to several feet into the water line by this "suck-back" effect (Bagga et a l , 1984). This mechanism has been identified as a cross-contamination risk for patients, as well as a major source of dental unit water contamination (Fitzgibbon et a l , 1984).  Improper Handling of Water Reservoirs Environmental and human-derived contaminants, most commonly from sloppy handling of the water in the refilling process of independent water reservoirs, may lead to biofilm formation in the reservoirs and water lines. In a 24 hour period, these bacteria are capable of reaching concentrations exceeding lOVmL in distilled water (Favero et al, 1971).  Plumbing System Studies have demonstrated that microbial contamination of water may come directly from connecting plumbing in large dental clinics (Fiehn & Henriksen, 1988; Kelstrup et a l , 1977). In some plumbing systems, factors such as the plumbing construction, age, water quality and other  10  non-specified environmental conditions could further generate or sustain biofilms to cause a more generalized contamination (Whitehouse et a l , 1991).  Characteristics of the Dental Units and their Water Lines which Enhance Biofilm Formation The dental unit water lines provide an optimal environment for biofilm formation. produce a foul taste and odor, and jeopardize its safety during clinical use.  This may  The following  reasons should help explain why dental units are susceptible to biofilm formation.  Water Stagnation Approximately 40 mL of water can be retained within the plastic microbore tubing of the dental unit water lines (Oppenheim et a l , 1987). The inherent pattern of intermittent-use of dental lines leads to stagnation of the entire water column for over 99% of the time (Williams et a l , 1995c). Furthermore, water stagnation overnight, on weekends and holidays provides an optimal conditions for biofilm formation (Pankhurst & Philpott-Howard, 1993).  Laminar Flow Water movement down dental water lines exhibit a phenomenon called laminar flow. This means that the water flow rate within the tubing is greatest at the center, and this flow decreases as it encounters increasing ffictional drag at the margins. water/slime tubing interface.  The velocity approximates zero at the  This phenomenon results in virtually stagnant conditions at the  11  tubing wall interface even when water is flowing, ensuring minimal disruption of the biofilm layer (Santiago et al., 1994; Williams et al., 1995a; Williams et al., 1993).  Tubing Material Dental unit water line tubing is composed of hydrophobic polymeric materials, such as polyvinyl chloride and polyurethane (Merritt & Chang, 1991; Williams et a l , 1995c). The composition of these materials, along with the microscopic imperfections in the internal surfaces of the water lines, present a surface substrate on which bacteria will adhere and colonize (Santiago et a l , 1994; Williams et a l , 1995a; Williams et a l , 1993).  Tubing Geometry and Internal Bore Diameter Biofilm formation on the wall of small bore tubing is well documented (Santiago et a l , 1994; Williams et a l , 1995a; Williams et a l , 1993). As the diameter of a cylinder (the water line) decreases, the surface area available for biofilm growth increases geometrically in relation to a fixed volume of water (Mills & Bednarsh, 1996).  As a result of this vast increase in surface  area, the number of bacteria and other microorganisms living inside dental units can become highly concentrated in water from dental water lines versus the water mains and pipes that deliver water to the dental unit (Mills & Bednarsh, 1996).  Water Temperature For patient comfort, dental units are often equipped with warming units, which ensure that the water is kept at 35-37°C (Pankhurst & Philpott-Howard, 1993). Heating water to near body  12  temperature for patient comfort is strongly discouraged, as it may enhance the total number of microorganisms and their pathogenicity, by selecting organisms better adapted to growth inside a human host (Mills and Bednarsh, 1996; Office Sterilization & Asespsis Procedures Research Foundation, 1997).  Water Source A problem that arises in large dental practices, particularly dental hospitals, is that they obtain their water from intermediate storage tanks within the building. Such tanks have been found to harbor high total viable counts and increased Legionellae species (Pankhurst & Philpott-Howard, 1993).  Pre-Filters Pre-filters, intended to remove particulate material from the municipal water as it enters the dental unit, have no effect on particles the size of bacteria and may provide additional surface area for microbial colonization (Mills and Bednarsh, 1996).  3.1 Microbiology of Dental Unit Water Lines  Although long lists of bacteria (Williams et al., 1993), protozoa (Williams et al., 1995; Atlas et al., 1995; Michael et al., 1989), fungi (Williams et al., 1993), and occasional nematodes (Santiago, 1994; Michael et al., 1989) have been isolated from dental unit water lines, most of the microbes detected are of very low pathogenicity or are opportunistic pathogens that may 13  cause infections only under special conditions or in immunocompromised individuals (Miller, 1996).  The majority of the microbial residents of dental unit water lines are saprophytic, heterotrophic, gram-negative, aerobic or facultative bacilli (Kelstrup et a l , 1977; Mayo et a l , 1990; Whitehouse et a l , 1991, Williams et a l , 1993). Most of these are municipal water contaminants, such as Pseudomonas, Klebsiellae and Caulobacter (Blake, 1963; Fitzgibbon et a l , 1984; Gross et a l , 1976). Primary human pathogens, such as Staphylococcus, Streptococcus and Enterococcus, have also been identified and may originate from retraction of patient fluids during patient treatment and/or handling of self-contained water reservoirs with bare hands (Atlas et a l , 1995; Blake, 1963; Lewis et a l , 1992; Sciaky & Sulitzeanu, 1962; Williams et a l , 1993; Williams et a l , 1996c).  Viruses have never been identified in dental unit water, although there are no records to indicate that they have been sought (Williams et a l , 1996c). In experimental situations, evidence of the presence of human viral genomic sequences in dental water has been adduced and intact virions of bacteriophages have been found to enter the water line when high-speed handpieces are operated with the bur in suspensions of phage particles (Lewis et a l , 1992). Further work is needed to clarify the significance of these observations to routine patient care (Williams et a l , 1996c).  However, viruses cannot multiply in the dental unit water line, as they require  mammalian cells for replication and most cannot survive long outside of the host organism.  14  An opportunistic microorganism is one that uses the opportunities offered by weakened defense mechanisms to colonize a host. Opportunistic pathogens may account for more than 30 per cent of the total bacterial population in water distribution systems (LeChevalier et al., 1980).  In  polymicrobial biofilms, pathogens like Legionella (Atlas et al., 1995; Mackenzie et al., 1994; Oppenheim et al., 1987; Pankhurst et al., 1990; Reinthaler & Mascher, 1987; Williams et a l , 1993), Pseudomonas (Clark, 1974; Costerton et a l , 1987; Martin, 1987; McEntegart & Clark, 1973; Mills et a l , 1986; Santiago et a l , 1994; Williams et a l , 1995a; Williams et a l , 1993,) and non-tuberculous Mycobacteria (Schulze-Robbecke et a l , 1992; Schulze-Robbecke et a l , 1995) not only survive but proliferate.  This is due to the fact that, in addition to being intrinsically  resistant to high temperatures and biocides (Best et a l , 1983), they are protected by the biofilm environment (Costerton et a l , 1995). identified in dental unit water.  Free-living, non-pathogenic amoebae have also been  These organisms may have significance as host cells in which  Legionella species can proliferate and possibly gain enhanced virulence (Atlas et a l , 1995; Michael et a l , 1989; Williams et a l , 1995c). These genera contain opportunistic pathogens that have  been  implicated  in  nosocomial  waterborne  infections  in  hospitalized  and  immunocompromised patients (Shearer, 1996), and will be discussed in further detail.  Legionella Species  Medical Significance Legionella species are fastidious Gram-negative bacteria that are ubiquitous in water and soil environments. While Legionella pneumophila is the principal pathogen in this genus, thirty other species exist and may cause up to a third of all Legionella infections (Miller, 1996).  The  15  bacterium is usually transmitted by inhaling aerosols or possibly by aspirating contaminated water (Miller, 1996; Stout et a l , 1982). It enters the respiratory tract by means of fine aerosols (Bolin et a l , 1985) and any aerosol generating water system that contains Legionellae may represent a potential source of infection. Legionella species are also capable of invading surgical incision sites exposed to contaminated water (Lowry et a l , 1991). All identified environmental sources of Legionella infection have been linked to contaminated water. While air-conditioning cooling towers are considered the likeliest sources of transmission, heat-exchange apparatuses, potable water supplies, humidifiers, showerheads, whirlpools and even vegetable moisturizers in produce markets have been implicated in disease outbreaks due to infection with Legionella species (Bolin et a l , 1985; Stout et a l , 1982; Centers for Disease Control and Prevention, 1990). Since dental instruments form aerosols, dental-unit water is a potential source of exposure to species of Legionella (Williams et a l , 1993).  Legionella species is well known as the etiologic agent of Legionnaires' disease, which manifests clinically as potentially fatal pneumonia (Fang et a l , 1989).  Epidemiological analyses have  identified cigarette smoking (Fraser et a l , 1977; Meyer, 1989), advanced age (England et a l , 1981), chronic lung disease (England et a l , 1981), and immunosuppression (England & Fraser, 1977; Haley et a l , 1979) as important risk factors for the development of Legionnaires' disease. Legionella pneumophila may also cause Pontiac fever, which is a non-pneumonic, self-limiting flu-like illness (Fang et a l , 1989), and wound infections following irrigation with Legionellacontaminated water (Lowry et a l , 1991). Nosocomial outbreaks of Legionnaires' disease has been described in several reports (Bolin et a l , 1985; Center for Disease Control and Prevention,  16  1990; Cordes et al., 1981; O'Mahony et al., 1990, Palmer et al., 1986; Snyder et al., 1990; Stout et al., 1982, Wadowsky et al., 1982). Hospital-acquired Legionella pneumophila infections are emerging as a major problem, and eradicating the organism from its reservoir is of utmost importance in prevention of Legionella-induced nosocomial illnesses (Muraca et al., 1990).  Until recently, studies have relied on culture methods for detecting Legionella, the results of which have yielded poor and unreliable estimates of the organisms present (Fitzgibbon et al., 1984; Furuhashi & Miyamae, 1985; McEntegart & Clark, 1973). Legionella species are often difficult to isolate because of overgrowth by other microorganisms and because they are often harbored within protozoa (Miller et al., 1993).  For these reasons, plate counts often  underestimate the presence of Legionella species in water samples (Miller et al., 1993). In 1994, Paszko-Kolva and colleagues (Paszko-Kolva et al., 1994) reported the first use of a semiquantitative polymerase chain reaction procedure along with culture and direct fluorescent antibody methods to detect and estimate the numbers of Legionella from the dental unit water supply. The use of molecular and immunological methods has the dual advantage of increasing detection and providing rapid results.  Since both polymerase chain reaction and direct  fluorescent counts can detect viable nonculturable bacteria which are not counted by plating procedures, these are presently the techniques recommended for epidemiological investigations of Legionella species (Miller et al., 1993).  17  Significance of Legionella in Dentistry The presence of Legionella species in dental unit water is well documented (Atlas et a l , 1995; Challacombe & Fernandes, 1995; Oppenheim et a l , 1987; Pankhurst et a l , 1990; Reinthaler & Mascher, 1987; Reinthaler et a l , 1988; Williams et a l , 1996a). This may be a cause for concern, since the well-documented mode of infection for Legionella through fine aerosols offers an obvious parallel to dental systems, which routinely generate aerosols. These fine aerosols could be inhaled by patients, dentists and dental personnel (Clark, 1974).  Legionella pneumophilia, the most significant Legionella pathogen, is implicated in about eighty five percent of cases of Legionnaire's disease (Reingold et a l , 1984). Several investigators have detected Legionella pneumophila in their dental unit water lines, with a prevalence ranging from four to sixty percent of the dental units analyzed (Reinthaler & Mascher, 1987; Oppenheim et a l , 1986; Challacombe & Fernandes, 1995; Luck et a l , 1993; Atlas et a l , 1995). Although other non-pneumophilia Legionella species have been identified in approximately similar percentages of dental units and control potable water samples (68% and 61% respectively), concentrations of these species in dental unit water are often much greater (Atlas, 1995). Although Legionella pneumophila was never detected at concentrations above 1 000 CFU/mL, other species of Legionella were found at levels exceeding 10 000 CFU/mL in 19 percent of the samples from dental unit water lines (Atlas et a l , 1995). The dose of Legionella required to cause infection is extremely variable, making it very difficult to establish a "safe" level of the bacteria (Challacombe & Fernandes, 1995).  18  Although Legionella pneumophilia is not the dominant component in dental unit water, chronic exposure to this bacterium should be investigated as a potential health risk for dental personnel and patients (Atlas et al., 1995). No cases of Legionellosis were identified in studies that specifically assessed chronically exposed dental groups for infection (Oppenheim et al., 1987; Pankhurst et al., 1990). However, this can be explained by a variety of reasons. Firstly, most cases of community-acquired pneumonic legionellosis are never identified, so the potential implication of dental exposure may represent an unrecognized element of the medical history of some current clinical cases (Stout et al., 1992). Furthermore, nonpneumonic legionellosis (i.e. Pontiac fever) may occur in dental personnel and patients and cause seroconversion, but may be clinically indistinguishablefromaflu-likeepisode (Atlas et al, 1995).  Several retrospective studies have demonstrated higher seroprevalence rates for Legionella antibodies among dental personnel than among nonmedical control groups (Fotos et al., 1985; Reinthaler et al., 1988; Luck et al., 1993), and have suggested that aerosols generated in dental operatories are a source of exposure to Legionella.  One serological study analyzed dentists,  dental assistants and dental technicians for antibody titers to several Legionella species (Reinthaler et al., 1988). While 34% of the dental personnel showed a positive reaction (titers of > 1:32) to the Legionella pneumophila antigen, only 5% from a control group of nonmedical workers tested positive.  A strong correlation was found between Legionella-seropositive  individuals and the length of time individuals were exposed to aerosolsfromhigh-speed drills and dental syringes. In an analogous study, 20% of the dental students and employees at the dental clinic studied were seropositive for Legionella  (titers of >1:16) (Fotos et al, 1985). 19  Seroconversion rates increased proportionately with increased clinic exposure time (Fotos et a l , 1985). Higher seroprevalence rates for Legionella antibodies among dental personnel have not been directly correlated with higher rates of disease.  Investigators have speculated that the  higher serum antibody titers may reflect continuous exposure to small numbers of the organism, resulting in mild infections, such as Pontiac fever (Luck et a l , 1993).  Two fatal cases of  Legionnaires' pneumonia in dentists have been described in the literature, and there is speculation that the source of the Legionella may have been due to aerosols generated by their dental units (Atlas et a l , 1995; Mackenzie et a l , 1994).  Pseudomonas Species Medical Significance Pseudomonas species are common inhabitants of soil and natural waters.  Many strains can  survive and even multiply in water of very low nutrient content such as distilled water (Favero et a l , 1971). Thus, it is not unusual to find Pseudomonas species in almost any domestic water r  supply, storage tanks or drain lines.  Pseudomonas aeruginosa, an opportunistic pathogen, is responsible for nine to eleven percent of all nosocomial infections reported each year in the United States (Deretic et a l , 1995). Pseudomonas species are often the causative agents in urinary tract infections, wound infections, pneumonia and septicemia.  Contact-lens-associated keratitis is frequently associated with  Pseudomonas aeruginosa (Elder et al, 1995). These opportunistic pathogens are the cause of significant morbidity and mortality in compromised subjects, such as those with burns, cystic  20  fibrosis, chronic bronchitis, bronchiectasis and cancer (Deretic et al., 1995). Pseudomonas aeruginosa and Pseudomonas cepacia usually have a higher degree of resistance than many bacteria to disinfectants and antibiotics (Miller, 1996). Microbial contamination of iodophor antiseptic solutions with Pseudomonas during their manufacture has been reported and has been associated with subsequent outbreaks of nosocomial infections (Berkelman et a l , 1984; Craven et a l , 1981, Parrott et a l , 1982).  Importance in Dentistry Several investigators have reported that the most common bacterial types isolated from dental unit water lines were gram-negative Pseudomonad (Williams et a l , 1993; Clark, 1974; Fitzgibbon et a l , 1984; Martin, 1987). Studies have shown that Pseudomonas aeruginosa can be isolated from 2.9 to 50% of dental unit water lines (Barbeau et a l , 1996; Jensen et a l , 1997).  The only scientific reports that have directly implicated any microbe from dental unit water as a health risk has involved Pseudomonas.  The two case reports from England implicated  Pseudomonas aeruginosa from dental unit water as the cause of oral infections in two medically compromised dental patients (Martin, 1987). These reports described the placement of large amalgam restorations using matrix bands in two patients being treated for cancer (Martin, 1987). Three to five days after the amalgam was placed, the patients developed pain and swelling corresponding to the area where the matrix band had been used. Microbiologic culture of the infected sites recovered Pseudomonas aeruginosa. The same pyocine type of Pseudomonas aeruginosa was subsequently isolated from the dental unit water lines in both case reports. The  21  author speculated that both infections were a result of direct inoculation of the traumatized tissue with contaminated dental water (Martin, 1987). In the only prospective study of acquisition of microorganisms from dental unit water, the investigator subsequently showed that all of seventy eight immunocompetent patients treated in dental operatories with Pseudomonas aeruginosacontaminated water lines became transiently colonized for four to ten weeks with Pseudomonad from the dental water, during which time the strains were regularly reisolated from the patients' oral cavities (Martin, 1987). Microflora of oronasal mucosa of dentists and their staff reflect the prevalence of Pseudomonas and other bacterial genera in the dental unit water lines of the operatories in which these persons work (Clark, 1974).  Pseudomonas aeruginosa acquired  during dental treatment, by inhalation of aerosolized water droplets, could be a significant source of infection for cystic fibrosis patients (Jensen et a l , 1997). One study found two genetically identical strains of Pseudomonas aeruginosa present both in water from a dental office's equipment and in a cystic fibrosis patient treated in the clinic (Jensen et a l , 1997).  Mycobacterium Species  Non-tuberculous Mycobacteria are considered to be emerging pathogens since their presence in the general population has dramatically increased over the last two decades. Biofilms may be important sites for the growth of aquatic Mycobacteria, which have been found to be four hundred times more concentrated in dental unit water samples than in tap water (SchulzeRobbecke et a l , 1992; Schulze-Robbecke et a l , 1995).  Several species of non-tuberculous  Mycobacteria (Mycobacterium avium complex, M. chelonei, M. fortuitum, M. gordonae, M.  dansaii, M terrae, and M. xenopi) have been isolated from hospital water systems, and some  22  have been associated with outbreaks of nosocomial infection (Costrini et al., 1981, Lowry et al., 1990; Lowry et al., 1988; Weinberger et al., 1992). Disease transmission occurred through the formation of infectious aerosols, and consequent colonization of the patients through inhalation or ingestion of these infectious droplets.  Infections may remain asymptomatic, or involve  symptoms such as weight loss, diarrhea, fever, anemia and malaise (Peters et al., 1995). AIDS sufferers are particularly vulnerable to infections with non-tuberculous Mycobacteria, and between 15 and 23 percent of these patients get infected with these pathogens (Peters et al., 1995). Cystic fibrosis patients can develop life-threatening respiratory infections with nontuberculous Mycobacteria (Deretic et al., 1995).  Amoebas  Amoebas are unicellular microorganisms which are frequently observed in fresh water habitats. Amoebas feed on bacteria, and thus biofilm constitutes an ideal niche for these microorganisms. Most dental lines contain a variety of amoebae, and several isolates from this source have the capacity to invade mucosa (Michael et al., 1989). While free-living amoebas can be isolated from 96 to 100 percent of dental unit water lines, Acanthamoeba species can be recovered in 12.2 percent of water samples (Barbeau et al., 1996; Michael & Just, 1984). Acanthamoeba has been shown to cause a variety of eye infections, especially in contact lens wearers (Martinez & Janitschke, 1985; Seal, 1994). Through the aerosolization of water or splatters from dental unit water lines, amoebas could potentially represent a risk of ocular infections (Barbeau et a l , 1997). Aerosolized amoebae proteins are also potentially allergenic and may cause typical signs  23  of upper respiratory hypersensitivity (Rowbotham, 1986). Furthermore, amoebas are the natural environmental hosts for Legionella bacteria, in which they may multiply (Rowbotham, 1986).  1.4 Risk Factors for Disease TransmissionfromDental Unit Water Lines  The presence of opportunistic pathogens in dental unit water lines has been demonstrated in multiple studies conducted over the past 30 years, and suggests reason for cautious concern (Mills & Bednarsh, 1996). Despite the high levels of organisms often reported, there is no scientific documentation establishing that biofilm in dental unit water lines represents a definable public health risk. No outbreaks of disease and few clinical case reports have been associated with contamination of dental unit water lines. This lack of evidence may reflect an absence or very low rate of disease transmission. However, it may also reflect the difficulty of estabhshing epidemiological links between infections with long incubation times and antecedent dental procedures (Shearer, 1996). Therefore, the low documented incidence of infections resulting from exposure to dental unit water may be misleading.  Exposure to Pathogen Vs. Disease Transmission There is a considerable difference between the "theoretical potential" for risk and "evidence of disease" being caused by dental unit water lines. The key point is that exposure to a possible pathogen and transmission of disease are distinct from one another. In order to induce a disease, a given microorganism must first attach to a host cell tissue, express virulence factors enabling it to colonize and invade tissues or cells and produce cytopathic effects.  As the following 24  microbiological formula clearly expresses, the virulence of the microbe, the amount of exposure to the microbe and host resistance are all vital factors which interact and determine if infection will occur (Canadian Dental Association, 1997).  Risk of infection = Virulence of Microbe X Dose of Microbe Host Resistance  The risk associated with dental unit water lines is not a simple, clear cut issue, but rather a sliding scale with possible risk for the immunocompromised patients at one end, and minimal risk for the healthy patient at the other. Exposure to a pathogen is accordingly clearly not synonymous with infection by the pathogen. Examining the expression of selected virulence factors by the bacterial populations in water samples has been proposed as a way to detect bacteria with pathogenic potential in water, and is currently being tested as an indicator of potential health risks (Edberg et al., 1996).  Risk Groups Contaminated dental unit water could theoretically cause infection by directly inoculating bacteria into damaged tissues, or through the aspiration of aerosols generated from contaminated dental unit water  Where contaminated treatment water has been involved as a source of infection,  most nosocomially infected patients presented with prior immunosuppressive disorders (Centers for Disease Control, 1979; Lowry et al., 1991). The number of dental patients with diminished resistance to opportunistic pathogens is increasing.  Patients at high risk to opportunistic  25  infections include those suffering from acquired immune deficiency syndrome (AIDS) and other autoimmune disorders, cystic fibrosis, diabetes, organ transplant and blood transfusion recipients, persons who are undergoing radiotherapy or chemotherapy, persons with chronic organic disorders, patients who are receiving massive doses of broad spectrum antibiotics or corticosteroids, alcoholics and drug addicts, smokers, elderly people and pregnant females (Barbeau et a l , 1997; Challacombe & Fernandes, 1995; Mayo et a l , 1990).  Although the biofilm-associated water contaminants are generally regarded as opportunistic, they may have demonstrable effects in immunocompetent individuals if they are chronically exposed to an immune challenge.  Serological evidence of occupational risk, through aerosolization of  waterborne microorganisms, is accumulating (Williams et a l , 1995c; Williams et a l , 1996c). Studies suggested that there is seroconversion of dental professionals, which means that exposure to Legionella may be at a high enough level and of sufficient duration to elicit an immune response (Fotos et a l , 1985; Luck et a l , 1991; Oppenheim et a l , 1987; Reinthaler et a l , 1988).  1.5 Current Recommendations  Potable Water Standards The most widely used measure of water quality is the fecal contamination test, which measures the number of Escherichia coli found in a water sample. Escherichia coli is a Gram-negative facultative bacillus found in human or animal gastrointestinal tracts.  Water is generally 26  determined to be potable or safe for drinking if the Escherichia coli count is less than one colony per 100 mL, (Code of Federal Regulations, 1992; Health Canada, 1996). Another measure of water quality is called the heterotrophic plate count test, which may be carried out using the pour plate, spread plate, or membrane filter method. Heterotrophs are microorganisms that cannot synthesize their own food, and are dependent on complex organic substances for survival. The heterotrophic plate count, which is measured in colony forming units per milliliter (CFU/mL), determines the number of colonies of heterotrophic microorganisms per milliliter of water. The colony forming unit represents the minimum number of separable cells, on the surface of or in semi-solid agar medium. Water is considered to be potable if the heterotrophic plate count is less than or equal to 500 CFU/mL (Geldreich, 1986). Since the detection of coliforms is impaired by the presence of high bacterial loads, it has been argued that total bacterial counts higher than 500 CFU/mL may suppress or mask specific pathogenic microorganisms (Environmental Protection Agency, 1989). Furthermore, the counts obtained depend on which culture medium is used (low-nutrient vs. high nutrient) as well as the incubation time and temperature (American Public Health Association et a l , 1995).  Recommendations for Non-Surgical Procedures in Dentistry Counting colony forming units from a heterotrophic plate count is a simple way of determining the quality of water used in dentistry. Water samples from dental units routinely yield heterotrophic microbial growths far in excess of the level that is regarded as acceptable for drinking water (500 CFU/mL).  Studies have found that bacterial counts may reach 3X10  CFU/mL (Santiago et a l , 1994; Williams et a l , 1993).  6  The American Dental Association  27  challenged dental equipment manufacturers to produce equipment that will reduce the level of aerobic mesophilic heterotrophic bacteria at any point along the unfiltered line for nonsurgical procedures to no more than 200 CFU/mL by the year 2000. This is equivalent to an existing assurance standard for hemodialysis units. This helps to ensure that fluids used in hemodialysis units have not been colonized by indigenous waterborne organisms (Shearer, 1996). Water used in dental treatment must be compatible with dental restorative materials and free of potentially toxic and/or carcinogenic chemicals. To aid in achieving this goal, manufacturers of dental equipment are encouraged to develop accessory components that can be retrofitted to dental units currently in use, whatever the water source (public or independent). The A D A also urges the industry to ensure that all dental units manufactured and marketed in the United Sates in the future have the capability to be equipped with a separate water reservoir which is independent of the public water supply (Shearer, 1996).  Recommendations for Surgical Procedures in Dentistry Sterile saline or sterile water which meet the standards for sterility described in the United States Pharmacopoeia (USP) should be used for all dental procedures which involve the intentional penetration, incision, excision, abrasion or ablation of intact, non-sulcular oral mucosa, and which results in exposure of normally uncontaminated bone or soft tissue (Center for Disease Control and Prevention, 1993; Office Sterilization & Asespsis Procedures Research Foundation, 1997; Shearer, 1996). The USP standards for sterile water assures that they are free not only of viable microorganisms, but of bacterial endotoxins, pyrogens, and other potentially harmful chemicals.  Devices which are intended for surgical irrigation must provide a sterile, non28  pyrogenic pathway for coolant or irrigants which will enter the surgical site. All components of this pathway must be single-use disposable or heat sterilizable (Office Sterilization & Asespsis Procedures Research Foundation, 1997).  While clinicians are not bound to follow these recommendations, the dental profession should acknowledge these agencies' review of the research, the social and moral imperative, and other compelling reasons upon which they have established recommendations.  1.6 Methods Available to Reduce Dental Unit Water Contamination Commercial devices which are marketed for the control of microbial contamination in dental unit water systems should be cleared to market by the US Food and Drug Administration (FDA). The F D A classifies dental water treatment and delivery systems as medical devices which are subject to pre-market clearance requirements under Section 510(k) of the Federal Food, Drug and Cosmetic Act. Dental Health care workers should be aware that dental units are subject to F D A pre-market clearance requirements, and any retrofittable devices for use with dental unit water lines also require 510(k) clearance. Products marketed in Canada should be registered as a medical device with Health Canada.  Although there is no universally accepted procedure available to eliminate dental unit water line contamination, several approaches to reduce the level of contamination are available:  29  Autoclavable Systems Fully autoclavable systems represent the gold standard in control of microbial contamination of water delivered through dental handpieces and syringes, and in eliminating the potential for patient-to-patient cross contamination (Williams et a l , 1996b).  Fully autoclavable systems  include water reservoirs, silicon dental unit water line tubings, and fittings that are designed to be sterilized between patients. These units require sterile saline or water to be used in conjunction with the system. The disadvantage of such a system is that a sufficient number of units must be purchased for between-patient steam sterilization, and support staff must adopt a different pattern of work for reprocessing and operatory setup between patients (Williams et a l , 1996b).  Independent Water Reservoirs Independent water reservoirs are often referred to as "self-contained" or "clean" water systems. They consist of a bag or bottle which serves as a water reservoir. The water is then driven into the dental unit water lines by air pressure or a peristaltic pump. By using a water source other than the municipal water, the contamination in the dental unit water lines may be reduced. One study shows independent water reservoirs may lead to a 2-3 log™ reduction in the numbers of organisms in the dental unit water (Pankhurst & Philpott-Howard, 1993). The ideal alternative to tap water is sterile deionized water, which is free of microorganisms and other harmful compounds (Mills & Bednarsh, 1996).  There is a misconception that simply using a water source other than the municipal supply solves the problem of dental unit water contamination.  However, there are limits to the degree to  30  which altering the source of water may have on the quality of water delivered at the output. Biofilm already present in the water lines (if these are not replaced), bacteria derived from water sucked back from the mouth and improper handling of water reservoirs during refilling can all contribute to the degradation of water quality on the output of a dental unit water line. Regardless of the water source, storage considerations during inactive periods in a dental clinic are of utmost importance. Since both glass and plastic containers can quickly develop biofilms, water should not be stored for long periods of time. Furthermore, water reservoirs and water lines are usually nonsterilizable and must be decontaminated routinely with biocidal solutions, such as concentrated hypochlorite or povidone iodine. These decontamination solutions must be pressurized to drive them through the system, creating occupational hazards such as personnel being sprayed with bleach due to leaking tubing or fittings, and explosive rupture of pressurized reservoirs containing corrosive bleach or other decontaminant solutions (Williams et al., 1996b). Another problem is that the biocidal chemical rinses are not reliably effective in eliminating established biofilm, and repeated exposures tend to select biocide resistant organisms (Mills et a l , 1986). Inadequate decontamination eventually leads to biofilm formation in the reservoirs (Mills & Bednarsh, 1996).  Evidence shows that use of separate water reservoir systems with periodic or continuous chemical treatment can improve the quality of dental treatment water so that it meets the A D A recommendations for acceptable quality (Mills et a l , 1986; Williams et a l , 1994a; Kelstrup et a l , 1977). Careful compliance with treatment protocols appears to be a critical factor for long-term success (Williams et a l , 1994a).  Using inadequate disinfection protocols may cause water  31  delivered to be of significantly poorer quality than units drawing from the municipal water supply, since they are unchecked by the residual chlorine or chloramine present in municipal supplies, opening the way for colonization by more chemically resistant and potentially pathogenic bacteria (Barbeau et a l , 1997; Williams et a l , 1994a).  Flushing of Dental Unit Water Lines The concept of flushing water through dental unit water lines had its origins in a report more than 15 years ago, which showed that important reductions in bacterial concentrations were achieved by flushing water through the lines continuously for a few minutes (Scheid et a l , 1982). The flow rate during flushing is slow, about 85 mL min "* (range 20-180 mL m i n ) , and it would take several minutes to fill the system with fresh water (Pankhurst & Philpott-Howard, 1993). The Center for Disease Control and Prevention has issued recommendations for the control of dental unit water line contamination, indicating that water lines should be flushed for "several minutes" at the beginning of each clinic day, and high-speed handpieces should be flushed for 20-30 seconds between patients (Centers for Disease Control and Prevention, 1993). Such flushing helps remove any microbes that may have entered the handpiece during patient care, and it also brings into the unit water system a small amount of chlorine (0.1-0.5 PPM) from the main water lines (Miller, 1996; Pankhurst & Philpott-Howard, 1993).  Several studies recognize that  although these guidelines are somewhat vague, flushing does produce substantial reductions in dental unit water microbial contaminants (Blake, 1963; Mayo et a l , 1990; Scheid et a l , 1982; Scheid et a l , 1990; Whitehouse et a l , 1991). However, the effectiveness of flushing has been challenged by some authors, who report that bacterial clearance is unreliable, often minimal, and  32  sometimes even increased following this procedure (Gross et al., 1976, Santiago et al., 1994, Williams et al., 1995c; Williams, Quimby et a l , 1994).  Electron microscopy studies have  confirmed that biofilms are not completely removed by flushing the water lines (Williams et a l , 1994a). Thus, although flushing may temporarily reduce the number of microbes in the water, it cannot be expected to yield bacteria-free water.  The optimal duration of flushing is somewhat controversial, based on widely disparate results that have ranged from 2 minutes (Gross et a l , 1976) to 20 minutes (Whitehouse et a l , 1991). Flushing the water lines for two minutes generally lowers the bacterial counts by 85 percent to 90 percent (Prevost et a l , 1995; Williams et a l , 1995b). However, despite this reduction in counts, they are still about 45 times higher than the maximum acceptable limit. Studies of flushing for 2 to 15 minutes have shown that there is usually a 1-2 logio reduction in the numbers of organisms and the worst of the malodorous water is removed (Gross et a l , 1976). However, this reduction is transient, as subsequent water stagnation from non-use of the unit results in recontamination within 0.5 to 24 hours (Whitehouse et a l , 1991).  It is not clear why the different studies have shown inconsistent results. Different water sampling protocols or different culture protocols may account for some of the differences (Williams et al., 1995c; Williams et a l , 1994b). The contribution of the connecting plumbing system, such as factors related to the plumbing's construction or age, as well as the water quality and other nonspecified environmental conditions, could also affect the results by generating or sustaining biofilms (Fiehn & Henriksen, 1988; Kelstrup et a l , 1977). This would cause a more generalized  33  contamination that would be resistant to flushing water through dental units. Furthermore, other dental parts, such as filters of fittings that trap debris, could contribute to the contamination. Laminar flow, the adherent nature of biofilms and the continuous re-seeding of water with clusters of cells dislodged from the tubing walls by the stretching and pulling that accompanies normal use of dental unit water lines could also explain why flushing techniques are often inadequate (Williams et a l , 1996b).  Thus, running water through dental units to control dental  unit water line contamination is an approach which should be applied with discretion, reflecting the variability specific to different units or clinical settings.  Chemical Disinfection Biocides used in dental practice must be of low toxicity and palatable, as the water coolant is both inhaled and ingested (Pankhurst & Philpott-Howard, 1993).  Antibacterial chemical  preparations containing povidone-iodine (McEntegart & Clark, 1973; Mills et al. 1986), Tween 80 (a laboratory surfactant) (Kelstrup et a l , 1977), chlorhexidine (Blake, 1963; McEntegart & Clark, 1973), Stericol (McEntegart & Clark, 1973), hydrogen peroxide (Kelstrup et a l , 1977; Santiago et a l , 1994) and sodium hypochlorite (Abel et a l , 1971; Fiehn et a l , 1988; McEntegart & Clark, 1973) have all been investigated with varying degrees of success. Chlorine compounds are the most widely studied and accepted for the control of biofilms in potable water systems (Van der Wende et a l , 1989). Sodium hypochlorite is a disinfectant generically accepted in the health care setting, and may be used with a variety of concentrations and delivery methods. The most common method is weekly use of a 1:10 dilution of sodium hypochlorite, which has been shown to be effective in reducing the numbers of bacteria in treatment water (Plamondon et a l .  34  1996; Williams et a l , 1994a).  Hyperchlorination refers to a method of delivery, whereby  chlorine is used in high doses at 50 P P M every six months.  Alternatively, a continuous  chlorination system can be installed, with an automatic dosing mechanism which provides 1 P P M of available chlorine at the chair (Pankhurst et a l , 1990). Although the latter two chlorination methods appear to be effective in maintaining drinking water standards in the storage tanks and distribution pipes (Muraca et a l , 1990), their effect is short-lived in the low pressure system of the dental chair (Pankhurst & Philpott-Howard, 1993). Sodium hypochlorite has a relatively low toxicity in humans but is highly corrosive and has the potential to damage dental unit components and shorten their useful life (Williams et a l , 1996b; Pankhurst & Philpott-Howard, 1993). One dental chair design incorporates an integral automatic flushing system that uses glutaraldehyde as a disinfectant. While it is undoubtedly effective, glutaraldehyde is known as a health hazard and is irritant, sensitizing and expensive (Pankhurst & Philpott-Howard, 1993,  Putnins, 1998).  Although bromide-containing biocides are less corrosive and less pH-dependent than chlorinebased ones, their use has not yet been evaluated in the dental setting (Pankhurst & PhilpottHoward, 1993).  Antibacterial agents can eradicate planktonic organisms in the tubing lumen, but usually spare the bacterial populations within the biofilm (Whitehouse et a l , 1991). The difficulty of removing dental unit water line biofilms comes from the resistance of established biofilm to biocides (Costerton, 1984). Disinfection of dental unit water lines was studied in units with (Blake, 1963; Kelstrup et a l , 1977; Mills et a l , 1986; Williams et a l , 1995a; Williams et a l , 1994a) or without  35  (Pankhurst et al., 1990) the use of separate water reservoirs, but neither of the proposed methods appears to permanently eradicate existing biofilms or their formation:  Incorporation of biocides into polymers during the production of indwelling catheters and drainage tubes for medical applications has been standard practice for several years. Incorporating biocidal agents directly into the substance of the dental unit water line wall is an attractive alternative to periodical disinfection. Any bacteria trying to adhere would be exposed to the biocide embedded in the surface, thereby preventing colonization. However, while medical devices such as catheters and drainage tubes place only short-term demands on biocide availability, dental unit water lines will require efficacy that extends over extended periods of time, to avoid costly and frequent line changes (Williams et al., 1996b). Currently, formulations of "medicated" dental unit water lines do not have satisfactory long-term potential (Williams et al., 1996b).  Anti-Retraction Valves Retraction of patient oral fluids into the water line through "suck-back" leads to contamination of the water with microbes from the oral cavity flora of patients that have been treated (Bagga et al., 1984). Some of these organisms are effective biofilm makers (i.e. dental plaque), and they become established in dental unit water lines, contaminating the water even if the lines have never been connected to municipal sources (Santiago et al., 1994). The risk of contamination can be minimized by the insertion of check valves to prevent the suction effect, and by sterilizing handpieces. A joint report written by the American Dental Association Councils on Therapeutics  36  and on Materials and Devices in 1978 firmly recommended the adoption of antiretraction measures (Association Reports, 1978). Once installed, the valves must be tested regularly and periodically replaced.  While check-valves all start out providing important benefits, their  working surfaces are susceptible to bacterial colonization and eventually become coated with a biofilm, which interferes with effective sealing. Furthermore, failure or fatigue of moving parts, deterioration of O-rings and metal corrosion may be responsible for antiretraction devices no longer functioning as they were intended (Williams et a l , 1996c). Although no microbiological performance criteria have been established for assessing the efficacy and longevity of antiretraction, current recommendations include the installation and proper maintenance of antiretraction valves and thorough flushing of the dental unit water lines after treatment of each patient (American Dental Association, 1992; Centers for Disease Control and Prevention, 1993) In the absence of microbial evaluation of check-valve function, installing antiretraction systems may give users a false sense of security.  Microfiltration Devices In-Line Filters In-line microfiltration provides a physical barrier to the efflux of microorganisms in treatment fluid. These polysulfone membranes are installed as close as possible to the instrument coupling device and remove bacteria by entrapment in the pores of the polymers. As certain bacteria may shrink under starvation conditions into a size range much lower than their normal vegetative forms, pores of 0.22 pm are required for effective removal of all contaminants (Williams et a l , 1996b).  Provided that the prescribed conditions of challenge and pressure are met (107  37  organisms per sq. cm at 30 psi), certification of the filtrate as "filter-sterilized" is permitted by the F D A and the USP (Williams et al., 1996b).  The principle of microfiltration to remove particulate contaminants is well established in the medical field, where it is used to trap microbes that may be present in intravenous fluids, epidural anesthetics, and ophthalmic irrigant solutions. Using a filter to remove bacteria from dental unit water was first proposed in 1978 (Dayoub et al., 1978). Although microfiltration systems can effectively reduce the level of microorganisms that exit the water lines, this effect is temporary and limited to 24 hours, after which time water recontamination occurs and the flow is reduced. Thus, these filters must be changed at least daily. Furthermore, microfiltration has no effect on the formation of biofilm (and its exotoxins), or its potential to obstruct and corrode the water delivery systems.  Other Microfiltration Devices Microfiltration can be achieved in a variety of other settings, including reverse osmosis water purification systems, ceramic candle filters, or pleated membrane filters. All of these are designed for long-term installations of weeks to months. These kinds of units, placed in-line before the long dental unit water line hoses, have a limited effect on the quality of water coming out through the handpiece or syringe. This is especially true if the dental unit water lines already have biofilm in them. Claims of control of infectious disease transmission by these kinds of installations have not been substantiated by any microbiological evidence, nor have any of these devices been reviewed by the F D A for dental use (Williams et al., 1996b).  38  Ultraviolet Irradiation Sterilization of municipal water could theoretically be achieved at the point-of-entry into the dental operatory, using ultraviolet (UV) irradiation devices (McEntegart & Clark, 1973). However, the disease control claims for U V devices used in the dental setting have no basis in scientific evidence, and they are not F D A cleared for this application. There is no evidence that U V irradiation alone has any impact on the microbial content of water delivered to the patient and, like point-of-use filters, U V irradiation of input water has no effect on biofilms (Williams et a l , 1996b).  Ethanol and Air Drying The dental unit is stagnant over 99 percent of the time, as water only flows through it for short periods when the handpieces or air-water syringes are in use. Potentially, stagnation could be eliminated if the chair was to be disconnected from the water system and the water drained off upstream from the junction box allowing the unit to be maintained in a "dry" state. A n in-line plumbing circuit was designed to permit isolation of the dental unit, air drying and ethanol disinfection (Furuhashi & Miyamae, 1985). Although the apparatus was shown to be effective, the procedure was cumbersome and could only be used for a single unit.  Rubber Dams A dam is a sheet of rubber with a small opening which is passed over the tooth, after which fluids that collect in the sheet are aspirated. The rubber dam serves as an effective barrier between the mouth of the patient and dental unit water which does not totally eliminate exposure, but greatly  39  reduces direct contact (Pankhurst & Philpott-Howard, 1993). The dam also greatly reduces the aerosolizing and spattering of patient microbes onto the dental team, but does not reduce exposure of the dental team to dental unit water (Miller, 1996).  Chair Design and Maintenance Many features of the dental chair design may contribute to water stagnation and amplification of organisms. Improvements in materials, for example using rubber or plastic fittings that do not encourage the growth of Legionellae and Pseudomonad, may be important. Easier access to the interior tubing for cleaning, disinfection and maintenance by dental surgery staff would significantly reduce the problem, and a high-temperature disinfection circuit could be considered. All dental units should be regularly serviced and maintained by a competent engineer, as neglect in this area may contribute to poor water quality in a number of cases (Pankhurst & PhilpottHoward, 1993).  40  Chapter 2 - Aims of the Study  1. Evaluate what Effect Culture Conditions have on Modifying Heterotrophic Plate Counts It is hypothesized that modifying incubation and plating protocols will have a significant influence on the heterotrophic plate counts. The modifying effect of different culture media, plating techniques, time and temperature of plate incubation, and the neutralizaton of chlorine will all be evaluated for their effects on heterotrophic plate counts.  2. Evaluate the Water Quality Delivered by Dental Units Connected to a Municipal Water Supply It is hypothesized that the water delivered by the dental units in the dental clinics at the University of British Columbia fails to meet potable water standards. Water samples will be collected from the air/water and highspeed lines of randomly selected dental units, and the water quality will be analyzed with the heterotrophic plate count technique.  Individual  colonies will be characterized based on colony morphology, Gram stain and oxidase test results. Representative samples of dental water line tubing will be processed for scanning electron microscopy and evaluated for the presence of biofilm.  3. Evaluate the Effectiveness of Marketed Products to Reduce Microbial Counts in Water Used for Nonsurgical and Surgical Dental Procedures.  It is hypothesized that marketed products are effective in reducing the number of heterotrophic organisms in water exiting dental units.  We will evaluate an in-line filter 41  system for its reduction in heterotrophic plate counts and determine its effect on water flow rates.  Scanning electron microscopy will be used to determine if 0.12% chlorhexidine  injected in the post filter tubing is able to prevent biofilm formation. Products designed to improve water quality used for surgical procedures, namely chemical disinfection of water lines, independent water reservoirs and autoclavable systems, will also be evaluated.  42  Chapter 3 - Materials and Methods  3.1 Water Collection Protocol  Non-Surgical Units The main clinic at the University of British Columbia consists of 80 operatories, each containing its own dental unit and sink. This clinic is used for the clinical training of dental students in nonsurgical dental procedures.  These units are connected to the municipal water supply, and the  water is heated and stored at 36°C in a central water reservoir.  Nine units were randomly  chosen, and the majority of samplings took place on these units. In-line 0.2 pm filters (SciTech Dental, Seattle, Washington) were installed on the highspeed lines and air/water syringes of seven of these units. In accordance with the manufacturers' instructions, the filters were changed daily, and the downstream tubing was filled with a 0.12% chlorhexidine solution on evenings and weekends.  On each collection day, water was collected both upstream (pre-filter) and  downstream (post-filter) from these filters, according to the following protocol.  A l l water  samples were collected at the beginning of the clinic day, after a 2 minute purge.  25 mL of  water was aseptically collected in sterile 50 mL tubes from the outlets of the water lines' of the air/water syringe tips and the highspeed handpieces.  Control water samples were aseptically  collected in the same manner from sinks in the operatories and other sinks within the university.  43  Surgical Units Independent Water Reservoirs The graduate periodontics clinic at the University of British Columbia consists of two operatories, each containing its own dental unit and sink. These operatories are used in training periodontal specialists in surgical periodontal procedures. Both units were supplied with bagged sterile water (Baxter, Toronto, Ontario), which was delivered to the unit via independent water reservoirs (Aseptico, Kirkland, Washington). All water samples were collected at the beginning of the clinic day, after a 2 minute purge. 25 mL of water was aseptically collected in sterile 50 mL tubes from the outlets of the water lines of the highspeed handpiece and the ultrasonic scaler. Control water samples were aseptically collected with a sterile syringe directly from the sterile water container.  After the above experiments were completed for the surgical units, the following weekly disinfection protocol was initiated: (1) air purge all the water lines; (2) fill all water lines with a 7500 P P M solution of sodium hypochlorite, and allow to stand for 10 minutes; (3)  discard all  the bleach and air purge the water lines; (4) flush all the lines with 500 mL of sterile water; (5) air purge and leave all lines dry until the next clinical use. The water collection time-table shown in Figure 1 was then executed on two occasions, namely 2 weeks and 4 months after the disinfection protocol was initiated. On each of the 10 collection time points (Figure 1), water was sampled from the water bag, the highspeed lines and the cavitron lines. Prior to sampling, the water bags were lightly agitated, a sterile needle was used to aseptically collect a 10 mL  44  sample of water from each bag.  Following a 30 second flush, a 25 mL water sample was  aseptically collected into a sterile 50 mL tube from each cavitron and highspeed lines.  Bleach Lines & Insert New Water Bag  Bleach Lines & Insert New Water Bag  Days  Collection Times  A B  D  H  I  J  Figure 1 - Water Collection Time-Table  Autoclavable System The water delivery system of one of the surgical units described above was subsequently changed.  The replacement system (Lares Research, Chico, California) consisted of a fully  autoclavable water reservoir, tubing and connector, which was supplied by USP sterile bottled water (Baxter, Toronto, Ontario). This system was tested on two separate occasions, using the following protocol. Prior to inserting the bottled sterile water into its reservoir, a 25 mL sample was collected aseptically in a sterile 50 ml, tube (Ti). The water was then loaded into its autoclavable reservoir and a 25 mL sample was collected as the water exited the highspeed line (Ti).  The highspeed line was then run until the water reservoir was half empty, and 25 mL  samples were collected from both the exit of the highspeed line and the water reservoir (T ). 2  45  3.2 Dilution, Plating and Culture Protocol  Spread Plate Technique A low-nutrient agar, R A (Difco, Detroit, Michigan), was used for culturing with the spread plate 2  technique (American Public Health Association et a l , 1995). After the medium was rehydrated with distilled water and sterilized at 121°C for 15 minutes, a volume of approximately 15 mL was dispensed on each sterile disposable 100X15mm plastic petri dish (Fisher Scientific, Ottawa, Ontario). After the medium had solidified, the plates were inverted and stored in a refrigerated room until usage.  Immediately following water collection, 10-fold serial dilutions were prepared to 10" (if needed) 4  by addition of sterile phosphate buffered saline (pH 7.3). Prior to plating, all water samples were vigorously vortexed for 15 seconds in order to disrupt any planktonic biofilm matrix. Plating was done in a Class II safety hood (Canadian Cabinets, Ottawa, Ontario). Duplicate 1 mL drops of each undiluted sample were pipeted onto the medium, and a sterile bent glass rod was used to spread the sample over the surface of the medium. Duplicate 100 uL samples of each dilution were similarly spread plated on a half plate of R A agar. To aid in drying the plates, they were 2  placed on a plate warmer which was set to 35°C. Following absorption of the water sample into the medium, the plates were incubated aerobically at 21°C, 28°C and 35°C for 2, 4 or 7 days. For each water sample collected, the dilution plate containing between 30-300 colonies was counted  46  using a Quebec colony counter, and data tabulated in CFU/mL. The mean number for duplicate platings was calculated.  Pour Plate Technique A high-nutrient agar, Plate Count Agar (Difco, Detroit, Michigan), was used for culturing with the pour plate technique. After the medium was rehydrated with distilled water and sterilized at 121°C for 15 minutes, it was placed in a water bath and maintained at a temperature between 44 and 46 °C. Media was used within 3 hours of preparation.  Immediately following water collection, 10-fold serial dilutions were prepared to 10" (if needed) 4  by addition of sterile phosphate buffered saline (pH 7.3). Prior to plating, all water samples were vigorously vortexed. Plating was done in a Class II safety hood. Duplicate 1 mL drops of each sample dilution were dispensed on individual sterile disposable 100X15mm plastic petri dishes. A volume of approximately 10 mL of liquefied medium was then poured into each dish. As each plate was poured, the melted medium was mixed with the water sample in the petri dish by carefully rotating the dish. After the medium solidified, the plates were inverted and aerobically incubated at 35°C for 2 and 7 days.  For each water sample collected, the dilution plate  containing between 30-300 colonies was counted using a Quebec colony counter, and tabulated in CFU/mL. The mean number for duplicate platings was calculated.  47  3.3 Effect of Sodium Thiosulfate on Heterotrophic Plate Counts  Sodium thiosulfate is a dechlorinating agent that, at a concentration of 18 mg/mL, is capable of neutralizing up to 5 mg/mL of residual chlorine in water (American Public Health Association et al., 1995). Upstream (pre-filter) water was collected from air/water syringe tips in the main clinic, according to the previously described water collection protocol for non-surgical units. Each sample was vortexed and divided in two equal portions.  While filter sterilized sodium  thiosulfate (Sigma, Missisauga, Ontario) at a final concentration of 18mg/mL was added to one portion, the other portion served as a control. Dilutions were done as previously described, and samples were plated using the spread plate technique on R A media and incubated aerobically at 2  35°C for 7 days.  3.4 Evaluation of Water Flow Rate  The influence of in-line filters on the rate of water flow was evaluated in the main clinic, for the highspeed and air/water lines in all 7 units equipped with in-line filters. This was accomplished by measuring the volume of water generated by flushing the appropriate water line for 10 consecutive seconds.  This was done prior to filter installation, immediately after filter  installation, and after 15, 30 and 45 minutes of continuous flushing through the filter. The percentage decrease in flow rate over time was calculated. Pre-filters were then installed on two randomly chosen units in the main clinic, and the experiment was repeated.  This was done to  study the effect that prefiltration has on preventing premature plugging of the in-line filters. 48  3.5 Microbial Analysis  Three operatories were randomly selected in the main clinic, and water samples were collected from the highspeed lines and warm water sinks taps of these operatories. In addition, water was also collected from 3 randomly chosen sinks within the dentistry building (but not in the clinics). Samples were processed on R A media with the spread plate technique previously described, and 2  incubated at 35°C for 7 days. A representative of each colony type formed was then replated on a separate R A plate, and incubated aerobically at 35°C for 3 days. 2  After the morphological  characteristics of each colony were determined by light microscopy, Gram stains and oxidase testing was done. Gram-positive coccoid colonies were also tested for catalase reactivity.  3.6 Scanning Electron Microscopy  Two dental operatories were randomly chosen in the main clinic, and a section of both the upstream (pre-filter) and downstream (post-filter) tubing for air/water or highspeed lines was collected from each unit. A 1 cm long piece of each tube was aseptically sectioned with a sterile scalpel, and prepared for analysis by scanning electron microscopy (SEM). Filters used in the main clinic for 5 minutes and 45 minutes or in the graduate periodontics clinic for several minutes (while unit disinfection protocol in place), were also collected. These filters were removed from their plastic housing by using a fissure bur on a highspeed handpiece, and prepared for analysis by SEM. Controls included a section of new tubing and an unused filter.  49  Tubing specimens to be analyzed were cut longitudinally, to expose the tubing lumen. While on ice, tubing and filter specimens were fixed in 2.5% gluteraldehyde in 0.1M phosphate buffered saline, pH 7.2, and washed 3 times with the buffer. They were then dehydrated for 10 minutes , through a series of alcohol washes (50%, 70%, 80%, 95%, 95%, 100%, 100%) at room temperature, critical point dried with liquid C 0 , mounted with adhesive onto aluminum stubs 2  and sputter coated with gold/palladium in a 60:40 ratio. Samples were viewed on a scanning electron microscope (Cambridge 260, Cambridge, England) at the appropriate magnification and representative photographs taken.  3.7 Data Analysis  In our study, data will be reported as geometric means and graphed as log values. This is due to the fact that bacterial counts do not follow a symmetrical normal distribution, but rather a positively-skewed one. This leads to an arithmetic mean which is considerably larger than the median.  Furthermore, application of the most rigorous statistical techniques requires the  assumption of symmetrical distributions such as the normal curve.  Thus, it is necessary to  convert skewed data so that a symmetrical distribution resembling the normal distribution results. This can be done by converting numbers to their logarithms.  The geometric mean, which  corresponds to the antilog of the arithmetic mean of the logarithms, is the best estimate of central tendency of log-normal data (American Public Health Association et a l , 1995). In order to test the hypotheses that means from several samples are equal, the analysis of variance (Anova) was  50  performed on the log values. This statistical procedure was used to determine whether means from two or more samples are drawn from populations with the same mean.  51  Chapter Four - Results  4.1 The Effect of Culture Conditions and Plating Techniques on Heterotrophic Plate Counts Overview of the Dental Building Water Quality Three operatories in the main clinic were randomly chosen, and water samples were collected from the highspeed lines, air/water lines, cold and warm water sink taps. Samples were collected and processed using the spread plate technique protocol previously described.  The water  temperatures were measured with a manual thermometer, and determined to be 36°C for water exiting the dental unit water lines and the warm water sink tap, and 7°C for the water exiting the cold water sink tap. The geometric mean of the heterotrophic plate counts for all units tested was calculated. The highest counts were found in the highspeed lines with an average of 1.8X10  4  (log=4.26) CFU/mL, followed by the air/water lines at 1X10 (log=4.01) CFU/mL, the warm 4  water tap at 9.1X10 (log=3.96) CFU/mL and the cold sink at 30 (log=l .49) CFU/mL (Figure 2). 3  Analysis of variance revealed that the cold water sink had significantly lower counts (Anova p<0.006) than the highspeed lines, air/water lines and warm sink tap (Appendix 1).  52  6.00  Highspeed  Air/Water  Warm Sink  Cold Sink*  Water Source  Figure 2 Heterotrophic Plate Count for Water Collected from Different Sources. Mean ± standard deviation, n=3 dental units. Spread plate technique; R A media. Incubated aerobically at 35°C for 7 days. * Anova p<0.006. 2  Spread Plates Vs Pour Plates On two occasions, water samples were collected from the highspeed and air/water lines of six randomly chosen units in the main clinic, according to the previously described protocol. In order to determine the effect of the plating technique on heterotrophic plate counts, the water samples were processed using both the spread plate technique on R A agar and the pour plate 2  technique on Plate Count Agar (Figure 3).  Although not statistically significant, the results  revealed (with one exception) that samples plated on R A agar with the spread plate technique 2  had higher counts than those plated on Plate Count Agar with the pour plate technique (Appendix 2). Calculation of the geometric mean revealed that this trend was seen for both the 2 day counts and the 7 day counts, despite a very large standard deviation which was particularly  53  pronounced on the 2 day count (Figure 3). For the air/water lines, the 2 day counts revealed a geometric mean of 2.0X10 (log=2.30) CFU/mL for the spread plates and 1.2 X I 0 (log=2.09) 2  2  CFU/mL for the pour plates. The 7 day counts for the air/water lines revealed a geometric mean of 3.4X10 (log=4.53) CFU/mL for the spread plates and 2.2X10 (log=4.34) CFU/mL for the 4  4  pour plates. For the highspeed lines, the 2 day counts revealed a geometric mean of 3.0X10  2  (log=2.48) CFU/mL for the spread plates and 19 (log=l .27) CFU/mL for the pour plates. The 7 day counts for the air/water lines revealed a geometric mean of 4.7X10 (log=4 67) CFU/mL for 4  the spread plates and 3.6X10 (log=4.56) CFU/mL for the pour plates. 4  54  *  5  4.5  0 Spread Plates • Pour Plates  4  3.5  I  3  g .5 2  32 1.5 1  0.5 0 2 days Incubation Time  B  6  Q Spread Plates • Pour Plates  E O  9  3  2\ 1 \  2 days  7 days  Incubation Time  Figure 3 - Comparison of Heterotrophic Plate Counts Utilizing the Spread Plates Vs. Pour Plates Techniques. A ) Air/Water Lines and B) Highspeed Lines. Mean ± standard deviation, n=6 dental units, 2 replicate trials. Spread plate technique used on R A media; pour plate technique used on Plate Count Agar . Incubated aerobically at 35°C for 2 and 7 days. 2  55  The Effect of Incubation Time and Temperature on Heterotrophic Plate Counts Water samples were collected from the highspeed and air/water lines of nine randomly chosen units in the main clinic, according to previously described protocol. In order to determine the effect of incubation time and temperature on the heterotrophic plate counts, each dilution was plated in triplicate with the spread plate technique, and incubated aerobically at 21°C, 28°C and 35°C. Colony forming units were counted following 2, 4 and 7 days of incubation. Although the general trend was that higher incubation times and temperatures yielded higher heterotrophic plate counts, the highest counts were obtained when samples were incubated at 28°C for 7 days. A slight decrease was then seen when samples were incubated at 35°C for 7 days (Figure 4). At 21°C, the incubation time influenced plate counts to a lesser degree than at higher temperatures. For instance, while the geometric mean of the heterotrophic plate counts at 21°C varied from 2 (log=0.29) CFU/mL on day 2 to 2.8X10 (log=3.44) on day 7 (a difference of 2798 CFU/mL), 3  the counts varied at 35°C from 3X10 (log=2.47) CFU/mL on day 2 to 9.5X10 (log=3.98) 2  3  CFU/mL on day 7 (a difference of 9200 CFU/mL). Analysis of variance revealed that varying the incubation time (2, 4 and 7 days) had a highly significant (Anova p<0.001) effect on the heterotrophic plate counts obtained.  Furthermore, incubation at a temperature of 21°C yielded  significantly different (Anova p<0.0005) counts than incubation at 28 and 35°C (Appendix 3). At 7 days, the mean counts of organisms at the three temperatures were quite similar, however, evaluation of colony morphologies that were cultured showed quite clearly that different temperatures select for different bacteria (Figure 5).  56  The number of dental units with water exiting the lines that meets potable water standards (<500 CFU/mL) is a very important question. Our data showed that the time and temperature at which plates are incubated drastically affects the number of dental units whose water meets potable water standards (Table 1).  21 C  35 C  28 C Incubation Temperature  Figure 4 - The Effect of Incubation Time and Temperature on Heterotrophic Plate Counts. Mean, n=9 dental units. Spread plate technique; R A media. Incubated aerobically at 21, 28 and 35°C for 2, 4 and 7 days 2  Incubation Temperatur B 21°C 35°C 28°C 55% 33% 100% 2 Days 55% 0% 22% 4 Days 0% 0% 11% 7 Days Table 1 - Percentage of Units that Meet Potable Water Standards (<500CFU/mL) Based on Culture Conditions. n=9 dental units. Spread plate technique; R A media. Incubated aerobically at 21, 28 and 35°C for 2, 4 and 7 days. Incubation Time  2  57  Figure 5 - Effect of Incubation Temperature on Colony Types Found on Heterotrophic Plates. A ) 21°C B) 28°C C ) 35°C. Spread plate technique; R A media. Incubated aerobically for 7 days. 2  58  Sodium Thiosulfate Neutralization of Chlorine Water was collected from the air/water lines of eight randomly chosen dental units in the main clinic, according to the previously described protocol. Each sample was subsequently divided into two equal portions, and sodium thiosulfate at a final concentration of 18 mg/mL was added to half the samples, as previously described. Samples were plated on R A media with the spread 2  plate technique and incubated aerobically at 35°C for 7 days. For seven of the eight air/water lines tested, the heterotrophic plate counts were higher for the water samples in which chlorine was neutralized by the addition of sodium thiosulfate (Table 2). The geometric mean of the heterotrophic plate counts for the samples in which chlorine was neutralized was 2.2X10  5  (log=5.34) CFU/mL, nearly double that obtained for the chlorinated water samples, which had a mean count of 1.4X10 (log=5.14) CFU/mL.  Analysis of the variance revealed that these  5  differences were significant (Anova p<0.05) (Appendix 4).  Dental Unit 8-1 AW 8-2 AW 8-3 AW 8-4 AW 8-5 AW 8-6 AW 8-7 AW 8-8 AW  With Na S 0 (CFU/mL) 915 000 16 050 580 000 220 000 30 500 4 755 000 200 000 92 000 2  2  3  Without Na S 0 (CFU/mL) 900 000 10 550 221 000 185 500 32 500 1 000 000 199 500 48 000 2  2  3  Table 2 - Effect of Sodium Thiosulfate on Heterotrophic Plate Counts. Spread plate technique; R A media. Incubated aerobically at 35°C for 7 days. 2  59  Microbial Analysis The culture of isolated colonies from several water sources was undertaken as previously described. Gram stains, oxidase and catalase tests were subsequently performed (Table 3).  Colony 1  Diameter (mm) 9-10  Color  Opacity  Texture  Borders  Elevation  Yellow  Opaque  Smooth  Smooth  2  3  Coral  Opaque  Smooth  Smooth  Slightly Raised Raised  3  1  Pink  Opaque  Smooth  Smooth  4  2.5-3  White  Opaque  Smooth  Smooth  5  0.5-1  White  Opaque  Smooth  Smooth  6  0.9  Opaque  Smooth  Smooth  7  1  Light Yellow Cream  Opaque  Smooth  Rippled  8  6  Yellow  Opaque  Smooth  Smooth  9  3.5  White  Mixed  Lumpy  10  3  Lucent  11  1.5  Pearl White White  12  0.25-1  13  Gram Stain Cocci/-  Oxidase Test -  Rods/-  -  Raised  Rods/-  -  Slightly Raised Slightly Raised Slightly Raised Raised  Rods/-  -  Rods/-  +  Cocci/-  Smooth  Raised in center Raised  Smooth  Smooth  Very Raised  Rods/-  +  Opaque  Lumpy  Rippled  Rods/-  +  Grey  Mixed  Smooth  Smooth  3.5  Orange  Lucent  Smooth  Smooth  14  1  White  Opaque  Lumpy  Smooth  IS  1  Mat White  Lucent  Smooth  Rippled  Lightly Raised Slightly Raised Slightly Raised Slightly Raised Flat  16  1  White  Opaque  Rippled  Rippled  17  4  Yellow  Opaque  Smooth  18  0.5  Orange  Lucent  19  1-2.5  20  5.5  21  3  Milky White Egg Yellow Grey  22  0.5  23 24  Cocci/+  Catalase Test  -  Rods/-  Cocci/+  -  RodV-  -  -  Cocci/+  -  RooV-  -  Raised  Rods/-  +  Smooth  Flat  Cocci/+  Smooth  Smooth  Rods/-  Opaque  Smooth  Smooth  Opaque  Lumpy  Rippled  Lucent  Smooth  Sinuous  Coral  Opaque  Smooth  Smooth  1  Clear  Lucent  Lumpy  Rippled  Slightly Raised Slightly Raised Slightly Raised Slightly Raised Slightly Raised Very Raised  0.5  Mat White  Lucent  Smooth  Rippled  Flat  +  +  Cocci/+  +  Cocci/+  +  Rods/-  -  Filamento us/Rods/-  +  Cocci/+  Table 3 - Characterization of Bacteria Cultured from Highspeed Lines, Clinic and Building Sinks. Each colony (n=24) was isolated in pure culture by streaking onto R A media. Incubated aerobically at 35°C for 3 days. 2  60  Five of the 24 colony types identified were gram-/oxidase+ rods, and thus potentially pseudomonads. The percentage of potential pseudomonads isolated from different water sources was determined by computing the number Gram negative, oxidase positive rods that were found in plates from different water sources. Results showed that although potential pseudomonads can be isolated from all the water sources tested, they represent a much higher proportion of the organisms in water emerging from highspeed lines (52%), than in those of the clinic (6%) and building sinks (12%) (Figure 6).  61  [  1  Others 43%  Gram  \  Oxidase + Rods  j I  1 ^%^^J  Figure 6 - Percentage of Possible Pseudomonads in Water from Different Sources. A ) Highspeed Lines (main clinic) B) Clinic Sinks (main clinic) C ) Building Sinks (dentistry building). n=3 dental units. Pour plate technique, R A media. Incubated aerobically at 35°C for 3 days. 2  62  4.2 Long Term Evaluation of Water Quality Used for Non-Surgical Procedures Seven dental units in the main clinic were randomly chosen, and their air/water and highspeed lines were sampled once per week for 12 weeks according to the previously described protocol. Samples were plated on R A media with the spread plate technique previously described. For all 2  units sampled, the heterotrophic plate counts were (with 2 exceptions) greater than the maximum potable water standard of 500 CFU/mL (Figure 7).  For the highspeed lines, counts varied  between 1.2X10 (log 3.08) and 2.3X10 (log 5.36) CFU/mL, with a geometric mean of 1.5X10 3  5  4  (log=4.17) CFU/mL and a standard deviation of 3 (log=0.52) CFU/mL. For the air/water lines, counts varied between 4X10 (log 2.60) and 2.9X10 (log 5.46) CFU/mL, with a geometric mean 2  5  of 7.2X10 (log=3.86) CFU/mL and a standard deviation of 5 (log-0.68) CFU/mL. 3  63  6.00  5.00 4-  Potable Water Standard  2.00 +  Dental Unit  1.00 +  •8-1 =8-2 8-3 —g—8-4 +=" 8-5 - -Q - 8-6 • O- • -8-7 3  0.00 6  7  10  11  12  Week#  Figure 7 - Heterotrophic Plate Counts for Water Collected from M a i n Clinic A) Air/Water Lines and B) Highspeed Lines. Mean, n=7 dental units. Spread plate technique, R A media. Incubated aerobically at 35°C for 7 days. 2  64  Long-term analysis of each individual unit clearly showed that some units had consistently higher levels of contamination than others (Figure 7). For instance, during the entire length of the study, the highspeed line of unit 8-1 demonstrated consistently higher counts (geometric mean of 1.2X10 (log=5.07) CFU/mL) than that of unit 8-5 (geometric mean of 4.2X10 (log=3.62) 5  CFU/mL).  3  Analysis of variance revealed that there were statistically significant differences  between dental units. For instance, units 8-1 and 8-3 had significantly higher counts (Anova p<0.0005) than all other units analyzed (with one exception) (Appendix 5). Calculation of the geometric mean for all units studied during the 12-week experiment demonstrated that different types of water lines may yield different heterotrophic plate counts.  For instance, while the  highspeed lines had a geometric mean heterotrophic plate count of 1.5X10 (log=4.17), the 4  air/water lines for the same units had a comparatively low geometric mean count of 7X10  3  (log=3.86) CFU/mL. This trend can be seen in all units tested, with the exception of unit 8-1 and 8-4 (Figure 8). Analysis of variance demonstrated that the highspeed lines had significantly higher counts (Anova p<0.0005) than the air/water lines (Appendix 5).  65  B Highspeed O Air/Water  T ffl .E  O  8>  8-2  8-3  8-4 Dental Unit  8-5  8-6  8-7  Figure 8 - Heterotrophic Plate Counts for Water Collected from the Main Clinic. Geometric mean + standard deviation, n=12 replicate trials. Spread plate technique; R A medium. Incubated aerobically at 35°C for 7 days. 2  Scanning Electron Microscopy Tubing specimens from the air/water and highspeed lines of two randomly selected dental units, as well as a new water line tubing, were collected and prepared for analysis by S E M , according to the previously described protocol.  Scanning electron microscopy revealed that while the  unused dental unit water line tubing surface was free of biofilm, those of the air/water and highspeed units in the main clinic were entirely covered by a layer of mature biofilm (Figure 9).  66  Figure 9 - S E M Analysis of Dental Unit Water Line Tubing Lumen A) New Dental Unit Water Line; B,C) Air/Water Line from Main Clinic; D,E) Highspeed Line from Main Clinic.  67  4.3 Long Term Evaluation of Water Quality Delivered with Point of Use Filters  In order to reduce the level of contamination delivered by the non-surgical units in the main clinic, seven dental units were randomly chosen and their air/water and highspeed lines were equipped with in-line  filters.  On each sampling day, water was collected upstream and  downstream from the filter, as previously described. Samples were plated on R A media using 2  the spread plate technique, and incubated at 35°C for 7 days. For the entire length of this 12 week study, all water samples collected downstream from the filters (post-filter) yielded water which was within the acceptable limits for potable water (<500CFU/mL), with heterotrophic plate counts varying between 0 and 300 CFU/mL with a geometric mean of 2 CFU/mL (Appendix 6). When samples were collected upstream from the filters (pre-filter) for these same units, counts varied between 4X10  2  (log=2.6) and 2.9X10  5  (log=5.46) CFU/mL, with a  geometric mean of 10 CFU/mL (Appendix 5). Calculation of the reduction in heterotrophic 4  plate counts which accompanies the use of filters revealed that this value varied between 99.2 and 100%, with a geometric mean of 99.9% (Table 4). Visual inspection of in-filters before and after use clearly shows all the material that has been trapped by the filter (Figure 10).  68  Dental Unit  Heterotrophic Plate Counts (C FU/mL) Without Filter With Filter % Count Reduction 117 500 8-1 air/water 2 99.9% 63 100 8-1 highspeed 2 99.9% 1 700 8-2 air/water 3 99.9% 7 800 8-2 highspeed 1 99.8% 11 500 8-3 air/water 1 99.9% 38 900 5 8-3 highspeed 99.9% 12 600 8-4 air/water 2 99.9% 9 600 8-4 highspeed 2 99.9% 4 200 8-5 air/water 2 99.9% 6 300 8-5 highspeed 4 99.9% 8-6 air/water 3 600 2 99.9% 13 800 8-6 highspeed 2 99.9% 8-7 air/water 2 500 1 99.9% 8-7 highspeed 8 700 3 99.9% Table 4 - Long-Term Evaluation of Heterotrophic Plate Counts with and without the Use of In-Line Filters. Geometric Mean, n= 12 replicate trials. Spread plate technique, R A media. Incubated aerobically at 35°C for 7 days. 2  Figure 10 - Clinical Photograph of Two In-Line Filters . Filter on the left is new Filter on the right has been used for 5 minutes in the Graduate Periodontics Clinic (prior to establishment of a water line disinfection protocol).  69  Flow rates for the in-line filters used alone, and in conjunction with pre-filters were evaluated as previously described, and expressed as percent flow reduction when compared to the flow rate in those same units without the filters. The results showed that the in-line filters caused a 28 to 37% flow reduction immediately after insertion, which increased to a 92% flow reduction after 15 minutes of continuous use, a 95 % reduction after 30 minutes and a 97 % flow reduction after 45 minutes.  Pre-filters were then added to the units and left in place for 1 month prior to  retesting. Experiments revealed that the use of pre-filters in conjunction with the in-line filters caused a flow reduction of 69 to 76% immediately after filter insertion, and resulted in 100 % reduction leading to filter plugging within 10 minutes of continuous use (Figure 11, Appendix 7).  o-i  0  15  30  45  Minutes of Continuous Filter Use  Figure 11 - Effect of Filters on Water Flow. Mean % flow reduction. N=7 dental units for in-line filters; n=2 dental units for pre-filters.  70  Scanning Electron Microscopy Several specimens were collected from the main clinic, including 3 filters (one new, one after 5 minutes of use, one after 45 minutes of use) and a post-filter tubing from the highspeed line in the main clinic.  These samples were prepared for analysis by S E M , according to the previously  described protocol. Analysis by S E M revealed that while the surface of the new in-line filter was free of microorganisms (Figures 12 A and B), that of a filter after 45 minutes of use had its pores plugged with trapped bacteria and particles (Figures 12 C and D). The section of dental unit water line tubing, which was located downstream from the filter and was filled with 0.12% chlorhexidine when not in use, revealed no bacterial biofilm, but rather a layer of crystalline deposits (Figures 12 E and F). .  71  Figure 12 - S E M Analysis A , B ) Surface of New In-Line Filter, C,D) Surface of Filter after 45 Minutes of Continuous Use in the Main Clinic; E,F) Lumen of Highspeed Line, Downstream from Filter.  72  4.4 Evaluation of Water Quality Used for Surgical Periodontal Procedures  Independent Water Reservoirs Water emerging from the highspeed and cavitron lines of both surgical units equipped with independent water reservoirs in the periodontal clinic were sampled according to the previously described protocol. Samples were spread plated on R A media and incubated aerobically at 35°C 2  for 7 days. When the independent water reservoirs were used with sterile water, but without an adjunctive disinfection protocol, the heterotrophic plate counts for water emerging from the highspeed and cavitron lines were found to be extremely high (samples were not sufficiently diluted, and a precise microbial count is not available). However, initiation of a weekly water line disinfection protocol with a 7500 P P M sodium hypochlorite solution produced almost a 100% decrease in the heterotrophic plate counts, which were maintained at low levels for the entire period between weekly disinfections (Figure 13, Appendix 8).  73  1100  • --  - •  900  •  " No disinfection " Bleach weekly X 2 weeks  •6— Bleach weekly X 4 months  700 -4  1500 300  A  100 -100 E Lines Bleached  F Lines Bleached  Collection Dates  1100  - -• --  B  900  •  • No disinfection  "O " Bleach weekly X 2 weeks  700  u.  -A—  Bleach weekly X 4 months  500 | -  u  300 100  iii -100 A  X  B  Lines Bleached  i E  i F  Collection Dates  Lines Bleached  Figure 13 - Heterotrophic Plate Counts for Surgical Units Supplied by Sterile Water Delivered by an Independent Water Reservoir A) Highspeed Lines and B) Cavitron Lines. Mean; n=2 dental units. Water samples were collected prior to initiation of disinfection protocol (no disinfection), after 2 weekly sodium hypochlorite treatments (X2 weeks) and after 4 months of weekly sodium hypochlorite disinfections (X4 months). Spread plate technique, R A media. Incubated aerobically at 35°C for 7 days. 2  74  Scanning Electron Microscopy During the course of surgical water quality evaluation, in-line filters were installed in the surgical units of the graduate periodontics clinic.  Despite the initiation of the weekly disinfection  protocol and the extremely low microbial counts, the in-line filters would plug very rapidly. This was found to occur even after 4 months of weekly disinfection with sodium hypochlorite. A filter which had plugged after approximately 5 minutes of use in the periodontal clinic was prepared for S E M according to the previously described protocol. S E M analysis revealed that the filter's pores were plugged with a non-bacterial, unrecognizable amorphous substance (Figure 14).  75  Figure 14 - A,B) SEM Analysis of the Surface of an In-Line Filter after 5 Minutes of Use in Graduate Periodontal Clinic.  76  Fully Autoclavable System. Following installation of a fully autoclavable system in one of the surgical units in the periodontal clinic, water was collected from both the sterile water bottle and the highspeed line.  This  experiment was repeated twice and done according to the protocol previously described. Samples were spread plated on R A media and incubated aerobically at 35°C for 7 days. This 2  experiment revealed that the autoclavable system delivers water which is sterile (or with minimal counts) (Table 5). Water Source  Heterotrophic Plate Counts (CFU/mL) Week A WeekB 0 0 Bottle T i 0 2 Highspeed T i 1 0 Bottle T 0 0 Highspeed Tubing T Table 5 - Heterotrophic Plate Counts for Surgical Units Supplied by Sterile Water Delivered by a Fully Autoclavable System. Bottle=Sterile Water in Container. Ti=Collection time immediately after autoclavable system is set up. T =Collection time after handpiece has been run for several minutes and half the bottled water is finished. Experiment was repeated twice, on Weeks A and B. Spread plate technique, R A media. Incubated aerobically at 35°C for 7 days. 2  2  2  2  77  Chapter Five - Discussion  Biofilms are matrix-enclosed bacterial populations adherent to each other and/or surfaces or interfaces, which can be found in any nutrient-sufficient environment where moisture and a nonsterile surface exist (Costerton et al., 1987, Costerton et a l , 1995).  Most of the  microorganisms that contribute to the formation of biofilms in dental units are derived from the regrowth of the low numbers of residual viable cells that survive municipal treatment. However, other sources include oral microbes which are sucked into the water lines during patient treatment, plumbing systems, and microbes derived from improper handling of sterile water bottles when an independent water system is used. Biofilms serve as a reservoir, significantly amplifying the numbers of free-floating microorganisms in the water exiting the water lines through dental handpieces, air-water syringes and ultrasonic scalers (Kelstrup et a l , 1977; Mayo et a l , 1990; McEntegart & Clark, 1973, Whitehouse et a l , 1991). There is increasing agreement among dental authorities that this represents an unacceptable bacterial load to introduce into the mouths of patients.  5.1 The Effect of Culture Conditions and Plating Techniques on Heterotrophic Plate Counts Although numerous studies have been conducted and published on the issue of dental unit water contamination, the methodologies and materials used by different investigators have not been standardized. An analysis of the literature revealed that a variety of culture media and incubation  78  conditions have been used in different studies to quantitate the microorganisms. Different types of media used for the recovery of bacteria from dental unit water supply include dilute peptone medium (Williams et al., 1995b; Williams et al., 1994a; Williams et al., 1994b), brain heart infusion agar (Prevost et al., 1995), glucose broth (Blake, 1963), horse blood agar (Fitzgibbon et al., 1984; Scheid et al., 1982), trypticase soy agar (Gross et al., 1976; Kelstrup et al., 1977; Williams et al., 1993), and tryptose blood agar base with yeast extract (Whitehouse et al., 1991). Incubation temperatures used in different studies have varied from 20°C to 37°C and incubation times have varied from 2 to 28 days (Fitzgibbon et a l , 1984; Gross et a l , 1976; Kelstrup et a l , 1977; Prevost et a l , 1995; Scheid et a l , 1982; Whitehouse et a l , 1991; Williams et a l , 1995b; Williams et a l , 1994a; Williams et a l , 1993). The incubation time and temperature, along with the culture medium used, are factors which may greatly influence both the number and type of organisms recovered (American Public Health Association et a l , 1995; Williams et a l , 1994b).  In order to demonstrate the effect that culture conditions may have on heterotrophic plate counts, we conducted a number of experiments using a variety of culture media, plating techniques, times and temperatures of plate incubation and neutralization of chlorine. In one such experiment, water samples were collected and then plated on R A agar with the spread plate 2  technique and on Plate Count Agar with the pour plate technique (American Public Health Association et a l , 1995). For 83% of the samples, those plated with the spread plate technique on R A media yielded higher heterotrophic plate counts than those plated with the pour plate 2  technique on plate count agar (these differences were not statistically significant). Williams and co-workers conducted a similar study, whereby they compared the effect of different culture  79  media (dilute peptone, trypticase soy agar and 5% sheep blood agar) on the number colony forming units recovered from water samples collected from dental unit water lines. They found that significantly greater numbers of bacterial colony-forming units were recovered on the nutrient-poor medium (dilute peptone) than on the enriched media (blood agar and trypticase soy agar) (Williams et a l , 1994b). The effect of incubation time and temperature on heterotrophic plate counts was also evaluated. Samples were collected and incubated at each of 21°C, 28°C and 35°C for 2, 4 and 7 days.  As previous investigators have also noted, macroscopic  examination of the colonies in our study revealed that different colony types are selected at different incubation temperatures (Williams et a l , 1994b). Since the water supplying the dental units in our study is heated to 36°C, we anticipated that this would encourage proliferation of microorganisms which are adapted to survival at high temperatures. Although the general trend was that higher incubation times and temperatures yielded higher heterotrophic plate counts, the highest counts were obtained when samples were incubated at 28°C for 7 days. A slight decrease was then seen when samples were incubated at 35°C for 7 days. These findings are similar to those of the American Public Health Association, who found that the highest counts were typically obtained from 5 to 7 day incubation at a temperature of 20 to 28°C (American Public Health Association et a l , 1995). Williams and co-workers conducted a similar study, whereby they compared the effect of different incubation temperatures (25°C and 37°C) on the number colony forming units recovered from water samples collected from dental unit water lines. In contrast to our results, they found that lower incubation temperatures yielded greater numbers of colony-forming units, regardless of the type of media used (Williams et a l , 1994b). They explain  80  these results by claiming that since most of the bacteria present in dental unit water lines are environmental, these are better adapted to growth at low temperatures (Williams et al., 1994b).  Current water testing standards include the dechlorination of water samples prior to culturing, in order to prevent the continuation of bactericidal action (American Public Health Association et al., 1995). We therefore evaluated what effect chlorine neutralization by sodium thiosulfate had on the heterotrophic plate counts.  Our study demonstrated that the geometric mean of the  heterotrophic plate counts for the samples in which chlorine was neutralized was 2.2X10  s  (log=5.34) CFU/mL, nearly double that obtained for the chlorinated water samples, which had a mean count of 1 4X10 (log=5.14) CFU/mL. This may help explain why the use of unchlorinated 5  water in units which are not adequately disinfected may cause water delivered to be of significantly poorer quality than units drawing from the municipal water supply. Such units are unchecked by the residual chlorine or chlorarhine present in municipal supplies, and open the way for colonization by more or different bacteria (Barbeau et al., 1997; Williams et al., 1994a).  In the health sciences, the use of blood agar and other enriched media and an incubation temperature representative of that of the human body is the accepted standard procedure for the recovery of bacteria (Williams et al., 1994b). In contrast, low nutrient media and an incubation temperature of 35°C for 48 hours is the standard protocol used for the recovery of bacteria in the water industry and in environmental microbiology (American Public Health Association et al., 1995). In order to establish a standard protocol for testing dental unit water lines, we must first determine what kind of isolates we wish to sample. If the objective is to recover only bacteria  capable of growth at body temperature, then the use of enriched media and high incubation temperatures (35-37°C) is the appropriate technique. However, if the objective is to quantitate the bacterial population as a measure of the level of microbial contamination, then the use of a low-nutrient medium and an incubation temperature of 35°C is most appropriate. The use of different culture media to study bacteria in dental unit water lines by independent investigators over a period of many years has resulted in a lack of standardization and data that cannot readily be compared with resultsfromother studies. For these reasons, there is a need to standardize the materials and procedures used in the study of bacteria in dental unit water lines.  5.2 Long Term Evaluation of Water Quality Used for Non-Surgical Procedures  Dental units in the main clinic at the University of British Columbia are supplied by municipal water, which is heated to 36°C and stored in a central water reservoir. Our study was able to establish that while all water samples collected from the cold water sink taps met the standards for potable water, water collected from the warm water sink taps, the air/water lines and the highspeed lines revealed extremely high heterotrophic plate counts and failed to meet the potable water standards in all but two instances.  This indicates that while the municipal water is of  satisfactory quality, contamination occurs in the hot water reservoir that supplies the dental units (air/water syringes, highspeed lines and warm water sink taps). Biofilm formation may not be limited only to the dental unit water lines, but most likely coats the surfaces of the warm water reservoir and the tubings connecting it to the dental units. This most likely contributes to the extremely high levels of contamination in the water exiting the dental water lines. Our findings 82  are in agreement with those of several other investigators, who strongly discourage heating water to near body temperature, as it may enhance the total number of microorganisms and their pathogenicity, by selecting organisms better adapted to growth in or on a human host (Mills and Bednarsh, 1996; Office Sterilization & Asespsis Procedures Research Foundation, 1997).  In our 12-week study assessing the water quality delivered by the highspeed and air/water lines of dental units in the main clinic, 99% of water samples revealed heterotrophic plate counts which were greater than the maximum potable water standard of 500 CFU/mL. Heterotrophic plate counts varied between 4X10 and 2.3X10 CFU/mL, with a geometric mean of 1X10 and 2  standard deviation of 4 CFU/mL.  s  4  These results are consistent with those of numerous other  studies, which have determined that the water delivered by dental unit water lines usually contains a higher number of microorganisms than is acceptable for water to be considered potable, or safe, according to the current Canadian standards for water quality (Atlas et a l , 1995; Barbeau et a l , 1996; Blake, 1963; Fitzgibbon et a l , 1984; Kelstrup et a l , 1977; McEntegart & Clark, 1973; Miller, 1996; Sciaky & Sulitzeanu, 1962; Williams et a l , 1993; Williams et a l , 1996c). These high heterotrophic plate counts are of particular concern, due to the fact that a bacterial concentration greater than 500 CFU/mL can mask the presence of coliforms by bacterial competition during the growth of the cultures (Geldreich, 1986).  In our study, long-term  analysis of each individual dental unit (Figures 7 and 8) clearly showed that some units yielded water with consistently higher heterotrophic plate counts than others.  Another trend which  became apparent for 5 of the 7 units studied, was that water collected from the highspeed lines yielded higher heterotrophic plate counts than water samples collected from the air/water lines.  83  Similar trends were also noted by previous investigators, and may be due to the lower level of utilization of the highspeed lines and a slower flow rate, leading to increased stagnation in the highspeed lines (Prevost et al., 1995). These hypotheses will need to be verified by further investigations. S E M analysis of the lumen of the air/water and highspeed line tubings confirmed these surfaces are in fact covered with microbial biofilm. We therefore concur with the findings of many other investigators, who concluded that the most common source of dental unit water contamination is the microbial biofilm found along the walls of the long, narrow-bore tubing that fAirnishes cooling and irrigating water to dental hand instruments (Blake, 1963; Gross et al., 1976; Kelstrup et al., 1977; Martin, 1987; Mayo et al., 1990; Pankhurst & Philpott-Howard, 1993; Whitehouse et a l , 1991; Williams et a l , 1993; Williams et a l , 1995b; Williams et a l , 1996c; Williams et a l , 1994b).  Bacteria from the genus Pseudomonas are well-known in the medical field as important opportunistic pathogens in cases of nosocomial infections. Their potential to cause disease in dental patients has been reported (Martin, 1987), and concern for the safety of patients would dictate that their introduction to the human body via dental unit water lines be minimized. Our study has demonstrated that oxidase positive, Gram negative rods (i.e. possible pseudomonads) make up 52% of the bacterial colonies isolated from the clinic highspeed lines.  This in  accordance with previous the results of previous studies of dental unit water lines, which demonstrated that Gram negative pseudomonads were the most common bacterial types seen (Prevost et a l , 1995; Williams et al, 1993; Clark, 1974; Fitzgibbon et a l , 1984; Martin, 1987).  84  5.3 Long Term Evaluation of Water Quality Delivered with Point of Use Filters  In-line microfiltration is an F D A approved technique for delivery of "filter sterile" water. In-line filters provide a physical barrier (0.22 pm pores) to the efflux of microorganisms in the treatment fluid, by trapping bacteria (Williams et al., 1996b). Nonetheless, it is important to remember that microfiltration has no effect on the formation of biofilm (and its exotoxins), or its potential to obstruct the water delivery systems.  During our 12-week study, water samples collected  upstream from the in-line filters (pre-filter) were compared to water samples collected downstream from the filters (post-filters) for these same units. We determined that in-line filters caused a 99-100% reduction in heterotrophic plate counts, and produced water which was within the acceptable limit for potable water (<500 CFU/mL). Our results are in agreement with those of previous investigators, who determined that the use of in-line filters can produce high quality water for coolant and irrigation purposes (Santiago et al., 1994). During the course of our study, questionnaires were handed out to the student-dentists working in the units equipped with in-line filters. The main complaint about the in-line filters was that these were causing a noticeable reduction in water flow and occasional premature plugging of the filters which necessitated their replacement several times during the course of a clinical day. We therefore carried out a study to assess the extent of flow reduction caused by the in-line filters. Results revealed that these caused a 28 to 37% flow reduction immediately after insertion, which increased to a 97% flow reduction after 45 minutes of continuous use. In order to remove particulate material from the municipal water as it entered the dental unit and attempt to alleviate the water flow problems associated with the in-line filters becoming prematurely plugged, pre-filters (5 pm pore size) 85  were added to the units. Water flow tests were repeated on the units equipped with both in-line filters and pre-filters. Results revealed that the addition of pre-filters worsened the water flow problem, causing a flow reduction of 69 to 76% immediately after filter insertion, and resulting in plugging of the in-line filter within 10 minutes of continuous use.  As noted by previous  investigators, this is most likely due to the fact that pre-filters have no effect on particles the size of bacteria and may provide additional surface area for microbial colonization (Mills & Bednarsh, 1996).  5.4 Evaluation of Water Quality Used for Surgical Periodontal Procedures  Sterile saline or sterile water which meet the standards for sterility described in the United States Pharmacopeia (USP) should be used for all dental procedures which involve the intentional penetration, incision, excision, abrasion or ablation of intact, non-sulcular oral mucosa, and which results in exposure of normally uncontaminated bone or soft tissue (Center for Disease Control and Prevention, 1993; Office Sterilization & Asespsis Procedures Research Foundation, 1997; Shearer, 1996). The USP standards for sterile water assures that they are free not only of viable microorganisms, but of bacterial endotoxins, pyrogens, and other potentially harmful chemicals.  Devices which are intended for surgical irrigation must provide a sterile, non-  pyrogenic pathway for coolant or irrigants which will enter the surgical site. All components of this pathway must be single-use disposable or heat sterilizable (Office Sterilization & Asespsis Procedures Research Foundation, 1997).  86  Independent water reservoirs consist of a bag or bottle which serves as a water reservoir. The water is then driven into the dental unit water lines by air pressure or a peristaltic pump. By using a water source other than the municipal water, the contamination in the dental unit water lines may be reduced. However, there are limits to the degree to which altering the source of water may have on the quality of water delivered at the output. Biofilm already present in the water lines (if these are not replaced), bacteria derived from water sucked back from the mouth and improper handling of water reservoirs during refilling can all contribute to the degradation of water quality on the output of a dental unit water line. To ensure little or no contamination of water delivered by clean water units, it is necessary to disinfect the units every week according to the manufacturer's instructions and to use sterile water in the bottle reservoir.  In our study, we tested the effectiveness of an independent water reservoir supplied by sterile water.  Results revealed heterotrophic plate counts which were extremely high, and at the  dilutions tested were uncountable. These high counts are most likely due to the fact that the water lines were not changed and no disinfection regimen was in place. Thus, biofilm almost certainly coated the lumen of the water lines, causing contamination of the sterile water as it flowed through the lines. Furthermore, the fact that the sterile water did not contain any chlorine may have further contributed to its degree of contamination as it flowed through the water lines. Similar results were also shown in a previous study, where the investigators examined clean water units that were supplied by sterile water, but disinfection protocols were not adhered to (Williams et a l , 1994a). Regardless of the fact that a precise bacterial count was not obtained, a disinfection regimen consisting of a weekly water line disinfection with a 7500 P P M sodium  87  hypochlorite solution was instituted immediately. Water was re-sampled 2 weeks and 4 months after the weekly disinfection regimen had begun, and results revealed extremely low heterotrophic plate counts at all time periods between weekly disinfections. Our results concur with those of prior studies, which showed that the use of separate water reservoir systems with periodic or continuous chemical treatment can improve the quality of dental treatment water so that it meets the A D A recommendations for acceptable quality (Kelstrup et al., 1977; Mills et al., 1986; Murdoch-Kinch et al., 1997; Williams et al., 1994a). Filters were also installed on these units, and despite the consistently low bacterial counts, water flow was greatly reduced and the filters were plugging within minutes of use. S E M analysis of these filters revealed that the filter's pores were plugged with a non-bacterial, unrecognizable amorphous substance. We suspect that this substance is most likely sloughing of the tubing material, resulting from the disinfection process. We therefore suspect that, in the long run, disinfection of the water lines may shorten their useful life, and these may need to be replaced regularly.  Fully autoclavable systems include water reservoirs, silicon dental unit water line tubings, and fittings that are designed to be sterilized between patients. These units require sterile saline or i  water to be used in conjunction with the system. We tested one such system, which was used in conjunction with USP sterile water. Results revealed heterotrophic plate counts of 0 in all but 2 instances. We suspect that the plates which revealed counts of > 1 CFU/mL most likely were the result of contamination during the process of water collection or plating.  88  Chapter Six - Conclusions  Although the problem of dental unit water contamination has been extensively studied, no standard testing procedures for dental unit water exist. In our study, we showed that culture conditions had a significant effect on heterotrophic plate counts. Based on a thorough literature review, the heterotrophic plate count testing standards set out in "Standard Methods for the Examination of Water and Wastewater" (American Public Health Association et a l , 1995) and our test results, we suggest the following testing protocol for water delivered by dental units: (1) flush water lines for 2 minutes; (2) collect 25mL of water in a sterile 50mL tube and proceed to plating samples immediately; (3)  neutralize chlorine with an 18mg/mL solution of sodium  thiosulfate; (4) perform serial dilutions as needed, by addition of PBS (lOOmM, p H 7.3); (5) plate samples on R A media with the spread plate technique or into Plate Count Agar using the 2  pour plate technique; (6) Incubate samples aerobically at 35°C for 2 days. Implementation of such a standard procedure would create consistency in the data and allow comparison of the results of different studies.  The water delivered by all the dental units sampled in our study contained high levels of heterotrophic microorganisms, some of which were believed to be opportunistic pathogens. The following trends were noted: (1) heterotrophic plate counts within a unit fluctuated on different testing days; (2) although all units are connected to the same central water pipe, significant variations in heterotrophic plate counts were found between units; (3) highspeed lines yielded significantly higher heterotrophic plate counts than air/water lines.  In order to achieve an 89  accurate estimate of the level of contamination within a dental clinic, we recommend that testing be repeated at several time points and involve all dental units and water lines within a clinic (e.g. highspeed, air/water and cavitron lines).  Eliminating water of sub-standard quality should be an integral part of a complete infectioncontrol program in dental offices. The effectiveness of in-line filters, which have F D A approval for providing "filter-sterile water", was assessed.  Our study showed that although the water  delivered with these filters met the potable water standards, these filters were disliked by dentists due to the dramatic decrease in flow rate that accompanied their usage.  In our surgical dental units, independent water reservoirs used in conjunction with sterile water were tested. Results revealed that when units were properly disinfected (weekly with 7500 P P M sodium hypochlorite) and supplied with sterile water, they delivered water that was clean or not contaminated for at least one week. However, the chemical disinfection process likely causes sloughing of the existing biofilm and/or the erosion of the inner lining of the water lines, which may lead to decreased useful life of these tubes.  Nonetheless, if the chemical disinfection  regimen was not strictly followed, the benefits of clean water systems were negated, leading to water contamination that was similar to or worse than that of dental units supplied with municipal water. A fully autoclavable system which was used in conjunction with sterile water was also assessed. Despite the extra time required in the operatory preparation and sterilization, this system was found to be extremely effective in providing sterile water for dental treatment.  90  While the true scope of the hazards associated with dental water contamination remains unclear, cases of patients who claimed they became extremely ill from contaminated water in dental units have already been the subject of lawsuits in the United States (Clappison, 1997).  Sufficient  evidence of potential health risks exists to justify the voluntary implementation of scientifically valid treatment protocols.  Dentists should adhere to water line maintenance protocols and  ensure that risk is reduced for all patients and personnel.  91  REFERENCES American Dental Association Council on Dental Materials, Instruments and Equipment; Council on Dental Therapeutics; Council on Dental Research; and Council on Dental Practice. Infection control recommendations for the dental office and the dental laboratory. J Am Dent Assoc 123: supplement, 1992. American Public Heath Association, American Water Works Association, Water Environment Federation. Standard methods for the examination of water and wastewater. American Public Health Association, 19th ed, 1995. Association Reports. Infection control in the dental office. J Am Dent Assoc 97. 673-674, 1978. Atlas R . M , Williams J . F , Huntington M . K . Legionella contamination of dental unit waters. Appl Environ Microbiol 61: 1208-1213, 1995. Bagga B . S . R , Murphy R A , Anderson A W , Punwani I. Contamination of dental unit cooling water with oral microorganisms and its preventions. J Am Dent Assoc 109: 712-716, 1984. Barbeau J, Nadeau C. Dental unit water line microbiology: a cautionary tale. J Can Dent Assoc 63(10): 775-779, 1997. Barbeau J , Tanguay R , Faucher E , Avezard C , Trudel L , Cote L , Prevost A.P. Multiparametric analysis of water line contamination of dental units. Appl Environ Microbiol 62: 3954-3959, 1996. Berkelman R L , Anderson R L , Davis F.J. Intrinsic bacterial contamination of a commercial iodophor solution: Investigation of the implicated manufacturing plant. Appl Environ Microbiol 47: 752-756, 1984. Best M , Y u V . L , Stout J , Goetz A , Muder R . R , Taylor F. Legionellaceae in the hospital water supply. Lancet!. 307-310, 1983. Blake G.C. The incidence and control of bacterial infection in dental spray reservoirs. Br Dent J 115:413-416,1963. Blenkinsop S . A , Khoury A . E , Costerton J.W. Electrical enhancement of biocide efficacy against Pseudomonas Aeruginosa biofilms. Appl Environ Microbiol 58: 3770-3773, 1992. Bolin G . E , Ploutte J . F , Para M . F , Hackman F. Aerosols containing Legionella pneumophilia generated by shower heads and hot water faucets. Appl Environ Microbiol 50: 1128-1131, 1985.  92  Canadian Dental Association. Dental unit water lines. CDA Member Information Bulletin. 1-5, April 1997. Centers for Disease Control and Prevention. MMWR 28:289-290, 1979.  Nosocomial pseudomonas cepacia infection.  Centers for Disease Control and Prevention. Recommended infection-control procedures for dentistry-1993. MMWR 41(RR-8): 1-12, 1993. Centers for Disease Control and Prevention. Legionnaries' disease outbreak associated with a grocery store mist machine—Louisiana 1989. MMWR39(7): 108-109, 1990. Challacombe S.J., Fernandes L . L . Detecting Legionella Pneumophila in water systems: comparison of various dental units. J Am Dent Assoc 126: 603-608, 1995.  a  Christensen G.J. Infection control: Some significant loopholes. J Am Dent Assoc 122: 99-100, 1991. Clappison R. A. Priority one: decontamination of dental unit water lines. Oral Health 87(1): 1115, 1997. Clark A. Bacterial colonization of dental units and the nasal flora of dental personnel. Proc Roy Soc Med67:29-30,19'/ 4. r  Code of Federal Regulations. Bottled water quality standard. Fed. Regist. 21: 62-64, 1992. Cordes L.G., Wiesental A . M . , Gorman G.W. Isolation of Legionella pneumophila from hospital shower heads. Ann Intern Med 94: 195-197, 1981. Costerton J.W. The formation of biocide-resistant biofilms in industrial, natural, and medical systems. Dev IndMicrobiol 25: 263-372, 1984. Costerton J.W. Cleaning techniques for medical devices: biofilms. Biomed Instrum Techno/ 31(3): 222-226, 247-249, 1997. Costerton J.W. Overview of microbial biofilms. J Indust Microbiol 15(3): 137-140,1995. Costerton J.W., Cheng K.-J., Geesy G G , Ladd T.I., Nickel J.C., Dasgupta M . , Marrie T.J. Bacterial biofilms in nature and disease. Annu Rev Microbiol 41:435-464, 1987. Costerton J.W., Irvin R.T., Cheng K.J. The bacterial glycolax in nature and disease. Annu Rev Microbiol 35: 299-324, 1981.  93  Costerton J . W , Lewandowski Z , Caldwell D E , Korber D R . , Lappin-Scott H . M . biofilms. Annu Rev Microbiol 49: 711 -745, 1995.  Microbial  Costrini A . M . , Mahler D A , Gross W . M . Clinical roentgenographic features of nosocomial pulmonary disease due to Mycobacterium xenopi. Am Rev Resp Dis 123: 104-109, 1981. Craven D E , Moody B , Connolly M G . Pseudobacteria caused by povidone-iodine solution contaminated with Pseudomomas cepacia. NEngl JMed 305: 621-623, 1981. Dayoub M B , Rusilko D . J , Gross A. A method of decontamination of ultrasonic scalers and high speed handpieces. JPeriodontol 49:261-265, 1978. Deretic V , Schurr M . J , Y u H . Pseudomonas aeruginosa, mucoidy and the chronic infection phenotype in cystic fibrosis. Trends Microbiol 3(9): 351-356, 1995. Edberg S C . , Gallo P, Kontnick C. Analysis of the virulence characteristics of bacteria isolated from bottled, water cooler, and tap water. Microbial Ecol Health & Dis 9(2): 67-77, 1996. Elder M . J , Stapleton F , Evans E , Dart J.K.G. Eye 9 (Ptl): 102-109, 1995  Biofilm-related infections in ophthalmology.  Elves A . W . S , Feneley R.C.L. Long-term urethral catheterization and the urine-biomaterial interface. Br J Urol 80(1): 1-5, 1997. England A C . I l l , Fraser D . W , Plikaytis B.D. Sporadic legionellosis in the United States: the first thousand cases. Ann Intern Med 94: 164-170, 1981. England A.C. I l l , Fraser D.W. Sporadic and epidemic nosocomial legionellosis in the United States: epidemiologic features. Am JMed 70: 707-711, 1981. Environmental Protection Agency. Drinking water; national primary drinking water regulations; filtration, disinfection; turbidity, Giardia lambila, viruses, Legionella, and heterotrophic bacteria. FedRegist 54(124): 27486-27540, 1989. Epstein J . B , Rea G , Sherlock C . H , Mathias R.G. Continuing investigation and controversy regarding risk of transmission of infection via dental handpieces. J Can Dent Assoc 62(6): 485490, 1996. Fang G . J , Y u V . L , Vickers R . M . Disease due to the Legionellaceae. Medicine 68: 116-132, 1989. Favero M.S. Whirlpool spa-associated infections: are we really in hot water? Am J Public Health 74: 653-654, 1984.  94  Favero M S , Carson L A , Bond W W , Petersen N.J. Pseudomonas aeruginosa: distilled water from hospitals. Science 173: 836-838, 1971.  Growth in  Fiehn N . E , Henriksen K. Methods of disinfection of the water system of dental units by water chlorination. J Dent Res 67: 1499-1504, 1988. Fitzgibbon E . J , Bartzokas C A , Martin M V . The source, frequency and extent of bacterial contamination of dental unit water systems. Br Dent J157: 98-101, 1984. Fotos P . G , Westfall H . N , Snyder I S , Miller R . W , Mutchler B . M . Prevalence of Legionellaspecific IgG and IgM antibody in a dental clinic population. J Dent Res 64(12): 1382-1385, 1985. Fraser D . W , Tsai T . R , Orenstein W. Legionnaires' disease: pneumonia. N Engl J Med 297: 1189-1197, 1977.  description of an epidemic of  Furuhashi M , Miyamae T. Prevention of bacteria contamination of water in detnal units. JHosp Infect 6: 81-88, 1985. Galal-Gorchev H , Ozolins G , Bonnefoy X. Revision of the WHO guidelines for drinking water quality. Ann 1st Super Sanita 29(2): 335-345, 1993. Gander S. Bacterial biofilms: 37(6): 1047-1050, 1996.  resistance to antimicrobial agents. J Antimicrob Chemother  Geldreich E E . Potable water: new directions in microbial regulations. Am Soc Microbiol News 52: 530-534, 1986. Gross A , Devine M . J , Cutright D E . Microbial contamination of dental units and ultrasonic scalers. J Periodontal 41. 670-673, 1976. Haley C . E , Cohen M L , Halter J , Meyer R D . Nosocomial Legionnaires' disesase; a continuing common-source epidemic at Wadsworth Medical Center. Ann Intern Med 90: 583586, 1979. Hayes E . B , Matte T . D , O'Brien T.R. Large community outbreadof cryptosporidiosis due to contamination of a filtered public water supply. N Engl J Med 320: 1372-1376, 1989. Health Canada. Guidelines for Canadian drinking water quality. 6th ed. 12-13,1996. Jensen E T , Giwercman B , Ojeniyi B , Bangsborg J . M , Hansen A , Koch C , Fiehn N . E , Hoiby N . Epidemiology of Pseudomonas aeruginosa in cystic fibrosis and the possible role of contamination by dental equipment. JHosp Infect 36(2): 117-122, 1997.  95  Johansen C, Falholt P, Gram L. Enzymatic removal and disinfection of bacterial biofilms. Appl Environ Microbiol 63(9): 3724-3728, 1997. Keene W E , McAnulty J . M , Hoesly F.C. A swimming-associated outbreak of hemorrhagic colitis caused by Escherichia coli 0157:H7 and Shigella sonnei. N Engl J Med 331: 579-584, 1994. Kelstrup J , Funder-Nielson T . D , Theilade J. Microbial aggragate contamination of water lines in dental equipment and its control. Acta Path Microbiol Scand %5. 177-183, 1977. Khardori N , Yassien M . Biofilms in device-related infections. JIndMicrobiol 1995.  15 (3): 141-147.  Kono G. The dental unit water line controversy. Dentistry Today 15(8): 82-85, 1996. Kono G. Dental unit water lines: taking the high road.now. Dent Today 16(8): 32-41, 1997. LeChevalier M . W , Seidler R . J , Evans T M . Enumeration and characterization of standard plate count bacteria in chlorinated and raw water supplies. Appl Environ Microbiol 40: 922-930, 1980. Lewis D . L , Arens M , Appleton S.S, Nakashima K , Ryu J , Boe R . K , Patrick B P , Watanabe D . T , Suzuki M . Cross-contamination potential with dental equipment. Lancet 340:1252-1254, 1992. Lewis W . J , Sherertz R.J. Microbial interactions with catheter material. Nutrition 13(suppl): 5S9S, 1997. Lowry P . W , Beck-Sague C M , Bland L A . Mycobacterium chelonae infection among aptients receiving high-flux dialysis in a hemodialysis clinic in California. J Infect Dis 161: 85-90, 1990. Lowry P . W , Blankenship R . J , Gridley W , Troup N . J , Tompkins L.S. A cluster of Legionella sternal wound infections due to post-operative topical exposure to contaminated tap water. N Engl J Med 324 (2): 109-113, 1991. Lowry P . W , Jarvis W . R , Oberle A D . Myobacterium chlonae causing otitis media in an earnose-and-throat practice. N Engl J Med 319: 978-982, 1988. Luck P C , Bender L , Ott M . Analysis of Legionella pneumophilia serogroup 6 strains isolated from dental units. In: Barnaree J . M , Breiman R . F , Dufour A.P. eds. Legionella: current status and emerging perspectives. Washington, D C : American Society of Microbiology: 240, 1993.  96  MacKenzie W . R , Hoxie N . J , Proctor M E . A massive outbreak in Milwaukee of Cryptosporidium infection transmitted through the public water supply. N Engl JMed 331: 161167, 1994. Mackenzie C D , Huntington M . K , Williams J . F , Costanza D . J , Detsch S . G , Atlas R . M . Legionella contamination in dental unit water lines: an associated fatality in a dentist (Abstract). FASEBJX994. Marsh P . D , Bradshaw D.J. Dental plaque as a biofilm. JIndMicrobiol  15(3): 169-175,1995.  Martin M . V . The significance of the bacterial contamination of dental unit water systems. Br Dent J163: 152-154, 1987. Martinez A . J , Janitschke K. Infection 13: 251-256, 1985.  Acanthamoeba, an opportunistic microorganism:  A review.  Mayo J . A , Oertling K M , Andrieu S C . Bacterial biofilm; A source of contamination in dental air-water syringes. Clin Prev Dent 12: 13-20, 1990. McEntegart M G . and Clark A. Colonization of dental units by water bacteria. Br Dent J 134: 140-142, 1973. Merritt K , Chang C.C. Factors influencing bacterial adherence to biomaterials. JBiomater Appl 5: 185-203, 1991. Michel R , Borneff M . Significance of amoebae and other protozoa in water conduit systems of dental units. ZblBaktHygB 187: 312-323, 1989. Michel R , Just H . M . Acanthamoebae, Naegleria and other free-living Ameobae in cooling and rinsing water of detnal treatment units. ZblBaktHygB 179: 56-72, 1984. Miller C.H. Microbes in dental unit water. J Can Dent Assoc 24 (1): 47-52, 1996. Miller L A , Beebe J.L., Butler N . C , Martin H , Benson R , Hoffman R E , Fields B.S. Use of polymerase chain reaction in an epidemiologic investigation of Pontiac fever. J Infect Dis 198: 769-772, 1993. Mills S , Bednarsh H. Dental water lines and biofilms. Implications for clinical practice. Dental Teamwork. 15-21, 1996. Mills S . E , Lauderdale P . W , Mayhew R.B. Reduction of microbial contamination in dental units with Povidone-Iodone 10%. J Am Dent Assoc 113: 280-284, 1986.  97  Muraca P . W , Victor L , Goetz A. Disinfection of water distribution systems for Legionella: A review of application procedures and methodologies. Infect Control Hosp Epidemiol 11: 79-88, 1990. Murdoch-Kinch C A , Andrews N . L , Atwan S , Jude R , Gleason M . J , Molinari J.A. Comparison of dental water quality management procedures. J Am Dent Assoc 128(9): 12351243, 1997. Noble Office Sterilization & Asepsis Procedures Research Foundation, Position Paper, January 1997. O'Mahony M . C , Stanwell-Smith R E , Tillett H E . The Stafford outbreak of Legionnaires' disease. Epidemiol Infect 104: 361-380, 1990. Oppenheim B . A , Sefton A . M . , Gill O.N. Widespread Legionella pneumophila contamination of dental stations in a dental school without apparent human infection. Epidemiol Infect 99: 159166, 1987. Palmer S R . , Zamiri I , Ribeiro C D , Gajewstra A . Legionnaires' disease and reduction in hospital hot water temperatures. Br Med J 292: 1494-1495, 1986. Pankhurst C . L , Philpott-Howard J.N. The microbiological quality of water in dental chair units. JHosp Infect 23(3): 167-174, 1993. Pankhurst C . L , Philpott-Howard J . N , Hewitt J.H. The efficacy of chlorination andfiltrationin the control and eradication of Legionella from dental chair water systems. J Hosp Infect 16: 918, 1990. Parrott P L , Terry P.M., Whitworth E M . Pseudomonas aeruginosa peritonitis associated with contaminated poloxamer-iodone solution. Lancet 2: 683-685, 1982. Paszko-Kolva C. Risk of infection from detnal handpieces. Am Soc Microbiol News 57: 287, 1991. Paszko-Kolva C , Shahamat M , Williams H , Palmer C , Petties C. Examination of detnal units for the presence of Legionella species by the EnvitoAmp polymerase chanin reaction (PCR). (Abstract no Q-471). 93rd General Meeting of the American Society for Microbiology 471, 1994. Peters M , Muller C , Rusch-Gerdes S , Seidel C , Gobel U , Pohle H . D , Ruf B. Isolation of atypical mycobacteria from tap water in hospitals and homes: is this a possible source of disseminated M A C infection in AIDS patients? J Infect 31: 39-44, 1995.  98  Prevost A . P , Robert M , Charland R , Barbeau J. Doctor, would you drink water from your dental unit? NY State Dent 761(10): 22-28, 1995. Plamondon T P , Mills S . E , Sherman L . R , Nemeth J , Puttaiah R. Effect of bleach on mature biofilm in dental unit water lines. J Dent Res (IADR Abstracts) 75: 414, 1996. Putnins, E. Surgical Dental Procedures and Water line Sterility. Good Impressions. 4-5, Winter 1998. Reingold A L , Thomason B . M , Brake B.J. Legionella pneumonia in the United States: the distribution of serogroups and species causing human illness. J Infect Dis 149: 819, 1984. Reinthaler F . F , Mascher F , Stunzner D. Serological examinations for antibodies against Legionella species in dental personnel. J Dent Res 67(6): 942-943, 1988. Reinthaler F , Mascher F , Stunzer D. Legionella pneumophila: seroepidemiologic studies of dentists and dental personnel in Austria. Zol Bakt Hyg 185: 164-170,1987. Rowbotham T.J. Current views on the relationships between amoebae, Legionellae and man. Isr J Med Sci 22: 678-689, 1986. Santiago J.I, Huntington M . K , Johnston M . A . Microbial contamination of dental unit water lines: Short and long-term effects of flushing. Gen Dent 42:528-535, 1994. Scheid R . C , Kim D . K , Bright J.S. Reduction of microbes in handpieces by flushing before use. J Am Dent Assoc 105: 658-660, 1982. Scheid R . C , Rosen S , Beck F . M . Reduction of CFUs in high-speed handpiece water lines over time. Clin Prev Dent 12: 9-12, 1990. Schulze-Robbecke R , Janning B , Fischeder R. Occurrence of mycobacteria in biofilm samples. Tuber Lung Dis 73: 141-144, 1992. Schulze-Robbecke R , Feldmann C , Fischeder R , Janning B , Exner M , Wahl G. Dental units: an environmental study of wourcews of potentially pathogenic mycobacteria. Tuber Lung Dis 76:318-323, 1995. Sciaky I , Sulitzeanu A. Importance of dental units in the mechanical transfer of oral bacteria. J Dent Res 41: 714, 1962. SealD.V. Acanthamoeba keratitis. Br Med J10%: 1116-1117, 1994. Shearer B . G Biofilm and the dental office. J Am Dent Assoc 127: 181-189, 1996.  99  Stout J , Y u V . L , Vickers R . M . Ubiquitousness of Legionella pneumophila in the water supply of a hospital with endemic Legionnaires disease. N Engl JMed 306: 466-468, 1982. Van der Wende E , Characklis W . G , Smith D.B. Biofilms and bacterial drinking water quality. Wat Res 23(10): 1313-1322, 1989. Wadowsky R . M , Yee R . B , Mezmar L. Hot water systems as sources of Legionella pneumophila in hospital and nonhospital plumbing fixtures. Appl Environ Microbiol 43: 11041110, 1982. Weinberger M , Berg S . L , Feurstein I . M , Pizzo P A , Witebsky F.G. Disseminated infection with Mycobacterium gordonae: report of a case and critical reveiw of the literature. Clin Infec Dis 14: 1229-1239, 1992. Whitehouse R . L . S , Peters E , Lizotte J , Lilge C. Influence of biofilms on microbial contamination in dental unit water. J Dent 19: 290-329, 1991. Williams H . N , Baer M L , Kelley J.I. Contribution of biofilm bacteria to the contamination of the dental unit water supply. J Am Dent Assoc 126: 1255-1260, 1995a. Williams H . N , Johnson A , Kelley J.I, Baer M L , King T . S , Mitchell B , Hasler J.F. Bacterial contamination of the water supply in newly installed dental units. Quintessence International 26(5): 331-337, 1995b. Williams H . N , Kelley J , Folineo D , Williams G . C , Hawley C . L , Sibiski J. Assessing microbial contamination in clean water dental units and compliance with disinfection protocol. J Am Dent Assoc 125(9): 1205-1211, 1994a. Williams H . N , Paszko-Kolva C , Shahamat M , Palmer C , Pettis C , Kelley J. Molecular techniques reveal high prevalence of Legionella in dental units. J Am Dent Assoc 127: 11881193, 1996a. Williams H . N , Quinby H , Romberg E. Evaluation and use of a low nutrient medium and reduced incubation temperature to study bacterial contamination in the water supply of dental units. Can J Microbiol 40: 127-131, 1994b. Williams J.F, Andrews N , Santiago J.I. Microbial contamination of dental unit water lines: current preventive measures and emerging options. Compendium 17(7): 691-708, 1996b. Williams J.F, Huntington M . K , Johnston M . Dental water lines. A source of contamination. Infect Control Steril Technol 14-20, 1995c.  100  Williams J . F , Johnston A . M . , Johnson B , Huntington M . K , Mackenzie C D . Microbial contamination of dental unit water lines: prevalence, intensity and microbiological characteristics. J Am Dent Assoc 124(10): 59-65, 1993. Williams J . F , Molinari J . A , Andrews N . Microbial contamination of dental unit water lines: origins and characteristics. Compendium 17(6): 538-558, 1996c. Wilson I . G , Hogg G . M , Barr J.G. Microbiological quality of ice in hospital and community. J Hosp Infect 36(3). 171-180, 1997. Wilson M . Susceptibility of oral bacterial biofilms to antimicrobial agents. J Med Microbiol 44(2): 79-87, 1996.  101  Appendix 1 Heterotrophic Plate Counts for Water Collected from Different Sources  Data (CFU/mL): Dental Unit  Highspeed  Air/Water  Warm Sink  Cold Sink  8-1  219500  162000  10950  8-6  5700  2000  7900  42 12  8-7  4800  3350  8600  58  Anova Table (Two-Factor Without Replication): Sumof-Squares Water Source  DF  Mean-Square  F-Ratio  P  15.221  3  5.074  17.144  0.002  Dental Unit  2.432  1.216  4.109  0.075  Error  1.776  2 6  Tukev HSD  0.296  Multiple Comparisons.  Matrix of Pairwise Comparison Probabilities: Air/Water  Cold Sink  Cold Sink  0.005  Highspeed  0.941  0.003  Warm Sink  0.999  0.006  Highspeed  0.901  102  Appendix 2 Heterotrophic Plate Counts on Spread Plates and Pour Plates  Data: 2-Days Incubation Time (CFU/mL) Dental Unit  8-1 AW 8-2 AW 8-4 AW 8-5 AW 8-6 AW 8-7 AW 8-1 HS 8-2 HS 8-4 HS 8-5 HS 8-6 HS 8-7 HS  Spread Plate week 1 week 2  0 0 200 50 0 83500 150 0 350 1800 0 72000  Pour Plate week 1  1000 27500 250 500 150 8500 2050 0 11500 5450 1200 550  0 0 27 133 0 62700 0 0 289 275 50 60000  week 2  255 0 352 175 369 8650 880 1 6750 2650 360 322  Anova Table (Two-Factor With Replication): Sum-of-Squares Plating Technique (PT) Water Source (WS) Dental Unit (DU) PT&WS PT4DU WS&DU Error  1.565 0.968 58.821 0.122 2.771 8.741 45.565  DF  Mean-Square  1 1 5 1 5 5 29  1.565 0.968 11.764 0.122 0.554 1.748 1.571  F-Ratio  0.996 0.616 7.487 0.078 0.353 1.113  P  0.327 0.439 0.000 0.782 0.876 0.375  103  Heterotrophic Plate Counts on Spread Plates and Pour Plates  Data: 7-Days Incubation Time (CFU/mL) Dental Unit  8-1 AW 8-2 AW 8-4 AW 8-5 AW 8-6 AW 8-7 AW 8-1 HS 8-2 HS 8-4 HS 8-5 HS 8-6 HS 8-7 HS  Spread Plate week 1 week 2  25000 137000 22500 52000 27000 136500 18500 37000 12350 56000 23000 143000  79500 97000 11800 6500 13000 18500 80000 8000 101500 143500 75000 107000  Pour Plate week 1  20150 34050 6700 35450 23800 94200 15800 36500 43150 46000 24600 72400  week 2  46900 66300 26000 3650 6500 15700 38600 7900 48500 90100 58300 37400  Anova Table (Two-Factor With Replication): Sum-of-Squares Plating Technique (PT) Water Source (WS) Dental Unit (DU) PT&WS PT&DU WS&DU Error  0.266 0.417 0.800 0.020 0.080 2.309 3.878  DF  1.000 1.000 5.000 1.000 5.000 5.000 29.000  Mean-Square  0.266 0.417 0.160 0.020 0.160 0.462 0.134  F-Ratio  1.991 3.120 1.196 0.152 0.119 3.453  P  0.169 0.088 0.335 0.700 0.987 0.014  104  m o  C  o  o  <D # -—< ro  o  \J If)  2 -#-» 2  c o  (D  3 U  Q) X  c >> IS  c o  CO  O  00  M  o  0)  T—  o o o o CM oo CO X —  o o CO lO  o o o o m co m oo  o o o o o o o o o o m o o r-- o oo o> T— r-~ T— T— o>  o o oo CM CM  o o o o o o CO o ^— CO m 00  oo o o "* co o o CM oo oo co CM  o o o o o o in  t—  T—  E  *r> CO  CD  I-  g  **  "D C CO (D  JO O o C  o o o o o o o o o o o o o o h- o co o o o o o o •* o oo CO CO a> o oo co CO ^ - X — CM CM o o o CO  o o  00  o o o co o o o o o> o o O) o CO a> co co  days  <D Q.  o oo o CM •* uCM  o o o  o o o  o o o  days  V—  |  "co  b b b  a.  10 >>  E  o o o  o o o  b b  CM  IS  3  c  p10  O o o o o O o o o CO o CO t o CO CO  o  o o CM  CO o> CM T—  o  •5  LU Q)  U  c >»  o o o o o u- o o m— , o o o o p  00  m  oo CD co CM  o  CO co  CM o o CM  o o o o o o o  E  u. o  5 (0 Q  1  c IS  c  i  °?  en 1  L_  vi o o  o  1  o  1 oo  oo  lO 1 co  CO  1 oo  O  o o  o o  o 8 o o oo CM CM b  S i, 0)1  N.  o in o co o> CM oo co -<tco CM co b  to  Ol  S> CO  O o o> CO oo  00 CM CM CO m  CO CO  •«t  c  .2 Q <5 a E E  o ocu o o  Q 0) *3 JO "5 I  c  £  ca  <<<<<<<<< •  •  e  o  T—  IS  o M  LL  CO  03 +*[ 3  CO  c o  o  c o o  •* o o CD co o co ^> oo co in CM co o  CM CO  days  o  days  o  T—  CM  O  o  iso on  co  O o o o o O o o o oo o o CM 00 lO  c  3 *-»  E  CM  uCO CM  # «  CM  b O o O  o o  o b b  Q (0 z o > X  Pa  c o  |Dental |Tem p« JTime [Error  o  -Ral  Q. Q. <  T—  o o o m  Meai Squa  c Q>  co  o o o o o CO oo uco  DF  x  T—  o o o o o T— CM CM  SumSquai  CO  o  o o o  0.001|  o o  Q_  cu  3 co H s  u u  CM IB CO 00  Appendix 4 The Effect of Sodium Thiosulfate on Heterotrophic Plate Counts  Data (CFU/mL): Dental Unit  With Na S 0 2  8-1 AW 8-2 AW 8-3 AW 8-4 AW 8-5 AW 8-6 AW 8-7 AW 8-8 AW  2  3  915000 16050 580000 220000 30500 4755000 200000 92000  Without Na S 0 2  2  3  900000 10550 221000 185500 32500 1000000 199500 48000  Anova Table (Two-Factor Without Replication): Sum-of-Squares Na S 0 2  2  3  Dental Unit Error  0.193 7.670 0.183  DF  Mean-Square  1 8 6  0.193 0.959 0.031  F-Ratio  6.306 31.354  P  0.046 0.000  r--  o  CO  cu c o o o o o m o m t- o r-~ i n C O r-~ o> oo J! i n  CO X  o m CM CM  i  "O (D Q) CL CO  o o o o  i  sz O)  o to CO  o  "O  2  cz CO  $  3  CD  o o o o m OJ CO CN 00  o o o r— oo  o o o o o m o o co ^ — o m -* m O J •* CM CN  o o O o O o o o O o O o o o O m m o m o O m m o CO o m • i n m m oo CM i n <r 00 CM o — CO 00 o CD T,— CO C O T — o T— T— CM CO OJ CM  o o o o o o lO C O m CM ^ — T" a CN T—  X  o o o o o To— o o o o in oo CN r-~ oo m oo  O o  o O o o O o in m o m O co o m o m CM CM •* o 00 CN T — oo CM  O O  o o O m o o cn C M o m r - t^T— T}-  o o m co TT  o O o m o o oi CM  o m -«* oo  o o m o oo co CM <(—  O o o o m in i n oo CM  O o o o o OJ o>  O O m oo CM  o o •* CM  o o o o r-- co oo CO  o o o CO o  o o o O o o o m m o O J oo O J o O J CO o> o •«}• •*— CM CM  T—  o o o o o OJ CN CM  o o 00 r-  CO  o o o o m ^— m % -<* CM  00  £  o  ~ CO  O o m m m ^— o T — oo OJ CM CM  CO  s> 3  CD ±i  O  £$  O  $  1  CO  **>  $  1  CO  ••—'  c  Z3  o O CD  S  1  •*—<  JO Q_  O  4)  SZ Q.  CD CD  X  3 U.  o  3(d Q  il Unit  S  c  2  CM  o m  o  m  CO  ^—  o m  o m CN •xf CM oo J—  T—  o o CM co  o O o o O o o • f cn o> T T O J o CN  o o o o o T— m in  o O o o o m m o m o o 00 o CM CM CD CM oo m o> O J oo Ti-  CM  T—  O o o o m m o o oo o o r» CM m oo CM  O o co CO CM  O  O  r» r-~  o o o m foo m  CO  CM  o o o O  O o o es o o o O m m m o m m CN oo o> CM CM O J m co oo t ^ CM T—  o o o O o o o o o o m O m m o o m T — CO O o CO f~- o •* oo oo co •* o T— o>  T—  o m CN CO  iri  T—  o o o o o o m o m o CM T — O J CO oo CO o • t f h- CM  r-OJ s  T—  T—  o o m co r~  o o CM CM  o o o o in o m m CO o CM CO m CO CM ^—  o o o O o m o in O m CM a> CM m co m  o m m o  o o in o oo o CM O J  II  c o  co'l  m oo  I co  *3 (0  o  U  CU o o o o o>  o o o O o m in in m o m o •* oo o o 00 m o o> co CN o O) T — CN T — CM * — oo O o  o o o o m o o oo CO co  m  AW  •8 5s  O  o  O O  T -  O o  fi  * - —<  o  CM CO  in m m  i  o  w O i n CM  E £ 2 £  O  o m CM m  O m  O O  £  o O o m m CM O) T—  O  1  oo  o o O o m m m o T— m m oo CM oo  1i i  CM IO «? 1 OO CO CO oo  1i oo  o o CO CN  o o cn CM  *<—  o m o co  i  00  CM  co  o O o o o o in o in m  1—O J O J cn 00 00 oo  9 £ E 3  t — CN  o o CN o CN  o o o o o m m o r - 00 oo CM oo r-T—  HS HS HS HS HS HS HS  o  T -  o o CM l~-  r—  AW  <  o o co oo  1 i •? eo oo oo  t 00  up  •  co eo oo  o ro u.  3 cri  OJ CM  m to  CO d CM  I  o .5 a, CO  lce > o  3  Q a  c 3  cl  UJJ  oo  to  o  c CO X "O <D <D Q. CO  sz CT X "O  cr  0) -I—« CO  < o  o c  0.540|  CO <?  0.747 1.000  aa *s e 3  £ o  >?  "O <D  00  7J  c 3  « CO  -•—*  c o O  Q)  D_ o  !c Q. O v_  2  CD  -«—<  Q) X  to o c o <> / ro a E a o Q o E a> o O a> (A  «?  c  O O O d  c* c 3  c 3  o o  oo o o o  o  d  d  ^—  OS  •  OJ  o o o d  o o o o  oo co CD 00 o d d d  5> X O O  o d  o o  o o  o o  d  d  d  o o o  o  o o  o 8 d d  10  a a. *to o X x > £ 0) J£  3  I-  2  00 *>»  c 3  •?  t  ™~ O fa. oo d  i  «? «? T 00  O S OS ±s c c c 3 3 3  00  00  £ *: C c 3 3  — ispeed  to  si c 3  «*• CN O d  | Air;  -«Q) —<  O o CO T — o 00 d d  ispeed 1  o  o  o> I  CD O  Week 12 1  CD C  CO  I •o CD CD Q. CO JC CO  o  •t o  CO  CM  T—  -**  CM o  o  "*  CM o  T—  O  CM CM CM o  CO O  CM  O  o  T—  O  o  CO  1 o  c  o  o  o  CM co O  o  o  O  co o  oo o  o  co  CM  x—  i  CO  CM m  o  CM o  CM CM o  CA  je  9  i  CD CO  T—  <  o  CD  £ -c *^  -t-«  l  g  CD  TJ  o  O  o CD  CO  CO  oo o  CO O  o  m  CO  •*  O  o  m  o  CM  o  •«-  o  o  <*  CO  o  CM  O  *- o  o  CO o o CM  co  T—  CM  T -  o  CM CM  1 o  CO CM co o CO  lO  o  m  o>  o  -t—'  c o  t—  o  • -CD —<  m  o>  co O  CM  O  O  Q.  o !c Q.  CD  co o  CM  CO  X  o  T—  Week  0)  CD  CM  i  "D  2  o  o  u>  4—  2  CO  Wee  0 LL.  -4—'  O  , —  CO  Week 2  <  «  •*  CO o  1  .E co  Weekl  C CD Q. Q.  O  CM  r»  3  U.  o  5 Q Q  1 8-1 AW  TD '  o  9 5  o X  o  3  o  CO  CO O  CO J£  co  111111  CM CM m  o  o  CO CO CO CO CO CO CO X X X X X X X  i  co  CM  to  1 *r 1 «? T l O <? CO CO oo co 00 oo co co CO co co 00  CM co  Appendix 7 Water Flow Rate through Air/Water (AW) and Highspeed (HS) Lines  Data (mL/10 seconds): Dental Unit  No Filter 0 min  8-1 AW  8-2 AW 8-3 AW 8-4 AW 8-5 AW 8-6 AW 8-7 AW 8-1 HS 8-2 HS 8-3 HS 8-5 HS 8-6 HS 8-7 HS  Dental Unit  32 14 30 38.3 30 42.5 46.2 29 43.1 25 20.4 31 29  Pre-Fitter  19 13.2 17.5 24 19 24 30 20 31 15.8 15.2 21.4 24  With In-Line Filter 15 min 30 min  1.9 4.8 2.2 2.7 2 2.2 2.3 2.6 2.6 1.8 1.8 1.9 4.8  45 min  1.7 1.8 2 2.2 1.5 1.5 0 1.8 2.3 1.1 1 1 1.5  Pre-Fitter + In-Line Filter 10 min 5 min  0 min  8-2 AW 8-3 AW 8-2 HS 8-3 HS  23 38 12 20  7 7.5 6 4  0 1 0.5 0  0 0 0 0  1.7 1.8 0.8 0.4 1.3 1 0 1 1.8 0.8 0.8 0.9 0.8  Appendix 8 Heterotrophic Plate Counts for Surgical Units Supplied by Sterile Water Delivered by an Independent Water Reservoir (CFU/mL)  Disinfection Pr otocol (Bleacl i)  Sampling Days  Weekly fc r 2 weeks  No ne Unit A  Unit A  UnltB  A B C 0 E F G H 1 J  Weekly fo r 4 months UnitB Unit A  UnitB  50  0 0 1 1  0 0 0 0 10  Water Bags Sampling No ne  Days Unit A  A B C D E F G H 1 J  UnitB  »1000 »1000 »1000 »1000 »1000  0 0  3  0  0  0  0  1 0  0 0  0  Disinfection Pr otocol (Bleac ^) Weekly fo r 2 weeks Weekly fo r 4 months UnitB UnltB Unit A Unit A  »1000 »1000 »1000 »1000 »1000  0 4 12 9 29  1 0 14 1 1  3 0  2 0  0  0  0  0  0 1  0 0  Highspeed Lines Sampling No ne  Days  A B C D E F G H 1 J  Disinfection Pr otocol (Bleacl i) Weekly fo r 2 weeks Weekly fo r 4 months UnitB UnitB Unit A Unit A  Unit A  UnitB  »1000 »1000 »1000 »1000 »1000  »1000 »1000 »1000 »1000 »1000  1 1 5 1 177  1 0 12 39 22  0 0  0 0  0  0  1  0  0 0  0 0  Cavitron Lines  111  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0088607/manifest

Comment

Related Items