Diagnosis_and_Risk_Prediction_of_Dental_Caries

DIAGNOSIS AND RISK PREDICTION OF DENTAL CARIES, VOL 2 (2000) Front Matter THE AXELSSON SERIES ON PREVENTIVE DENTISTRY The world-renowned authority on preventive and community dentistry presents his life's work in this five-volume series of clinical atlases focusing on risk prediction of dental caries and periodontal disease and on needs-related preventive and maintenance programs. Volume 1 An Introduction to Risk Prediction and Preventive Dentistry Provides a general overview of current and future trends in risk prediction, control, and nonaggressive management of caries and periodontal disease; preventive dentistry methods and programs; and quality control. Volume 2 Diagnosis and Risk Prediction of Dental Caries Includes a comprehensive discussion of the etiology, pathogenesis, diagnosis, risk indicators and factors, individual risk profiles, and epidemiology of caries. Volume 3 Diagnosis and Risk Prediction of Periodontal Diseases Presents a comprehensive discussion of the etiology, pathogenesis, diagnosis, risk indicators and factors, individual risk profiles, and epidemiology of periodontal diseases. Considers periodontal diseases as a possible risk factor for systemic diseases and presents current and future trends in the management of periodontal diseases, including nonaggressive debridement and preservation of the root cementum. Volume 4 Preventive Materials, Methods, and Programs Discusses self-care and professional methods of mechanical and chemical plaque control, use of fluorides and fissure sealants, and integrated caries prevention. Addresses needs-related preventive programs based on risk prediction and computeraided epidemiology analysis for quality control and outcome. Volume 5 Nonaggressive Treatment, Arrest, and Control of Periodontal Diseases and Dental Caries Details current and future trends in nonaggressive treatment methods that seek to preserve the root cementum; surgical versus nonsurgical periodontal therapy; repair and regeneration of periodontal support; management of furcation-involved teeth; restricted use of antibiotics; arrest of noncavitated enamel, dentin, and root carious lesions; nonaggressive mini-preparations; esthetic and hygienic aspects of restorations; and management of erosions. Focuses on needs-related maintenance programs to ensure the long-term success of treatment and to prevent recurrence of periodontal disease and dental caries. TITLE PAGE DIAGNOSIS AND RISK PREDICTION OF DENTAL CARIES, VOL 2 Per Axelsson, DDS, Odont Dr

Professor and Chairman Department of Preventive Dentistry Public Dental Health Service Karlstad, Sweden

Quintessence Publishing Co, Inc Chicago, Berlin, London, Tokyo, Paris, Barcelona, Sao Paulo, Moscow, Prague, and Warsaw DEDICATION To my wife Ingrid, my daughter Eva, and my son Torbjorn COPYRIGHT PAGE Library of Congress Cataloging-in-Publication Data Axelsson, Per, D.D.S., Odont. Dr. Diagnosis and risk prediction of dental caries / Per Axelsson. p. cm.  (The Axelsson series on preventive dentistry; vol. 2) Includes bibliographical references and index. ISBN 0-86715-362-8 1. Preventive dentistry. 2. Dental cariesPrevention. 3. Periodontal diseasePrevention. I. Title. II. Title: Risk prediction and preventive dentistry. III. Series: Axelsson, Per, D.D.S. Axelsson series on preventive dentistry; vol. 2. [DNLM: 1. Dental Cariesprevention & control. 2. Periodontal Diseasesprevention & control. 3. Preventive Dentistry. 4. Risk Factors. WU 270 A969i 2000] RK60.7.A94 2000 617.6dc21 DNLM/DLC for Library of Congress 99-16511 CIP

 2000 Quintessence Publishing Co, Inc Quintessence Publishing Co, Inc

4350 Chandler Drive Hanover Park, IL 60133 www.quintpub.com All rights reserved. This book or any part thereof may not be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, or otherwise, without the written permission of the publisher. Editor: Cheryl Anderson-Wiedenbeck Production: Gerda Steinmeyer Printed in Germany CONTENTS Preface vii Chapter 1 Etiologic Factors Involved in Dental Caries 1 Role of Plaque 2 Role of the Oral Environment 14 Role of Specific Cariogenic Microflora 18 Prediction of Caries Risk 29 Conclusions 40 Chapter 2 External Modifying Factors Involved in Dental Caries 43 Role of Dietary Factors 43 Role of Socioeconomic and Behavioral Factors 77 Conclusions 86 Chapter 3 Internal Modifying Factors Involved in Dental Caries 91 Role of Saliva 91 Role of Chronic Systemic Diseases and Impaired Host Factors 134 Role of Tooth Size, Morphology, and Composition 139 Conclusions 146 Chapter 4 Prediction of Caries Risk and Risk Profiles 151 Risk Groups 152 Individual Risk 155 Key-Risk Teeth and Surfaces 161 Risk Profiles 165 Cariogram Model 172 Conclusions 174 Chapter 5 Development and Diagnosis of Carious Lesions 179 Development of Carious Lesions 181

Diagnosis and Registration of Carious Lesions 208 Conclusions 245 Chapter 6 Epidemiology of Dental Caries 249 Limitations of Epidemiologic Surveys 249 Prevalence of Caries 253 Incidence of Caries 269 Caries Treatment Needs 272 Reasons for Changes in Caries Prevalence 275 Conclusions 278 References 281 List of Abbreviations 297 Index 299 PREFACE The etiology of dental caries and periodontal diseases is well understood, and we have now developed efficient methods of preventing and controlling these diseases. Over the last 25 years in County of Varmland, Sweden, large-scale implementation of our prevention programs has led to an increase in the percentage of caries-free 3 year olds, from 30% to 97%, while reducing caries in 12 year olds from an average of 25 DFS to less than 1 (0.6). In the last 10 years, we have increased the number of remaining teeth in 65 year olds by more than 15% and reduced their loss of periodontal support by more than 20% at the same time reducing the percentage who are edentulous from 17% to 7%. According to the principles of lege artis, all members of our profession are obliged to offer treatment based on the most current scientific and clinical knowledge available. As we enter the new millennium, we must therefore continue to concentrate our efforts on preventing, controlling, and arresting dental caries and periodontal diseases. However, needs-related preventive and maintenance programs must be cost effective and should be based on information derived from comprehensive diagnoses, histories, and risk predictions at group, individual, and tooth surface levels. For quality control and evaluation of such programs, computer-aided analytical epidemiology, using relevant variables, should be introduced. The aim of this clinical textbookthe second volume in a series of fiveis to provide up-to-date knowledge about the etiology, modifying factors, risk evaluation, development, diagnosis, and epidemiology of dental caries. A detailed scientific background is presented for each topic discussed, as well as an illustrated guide to implementing the state-of-the-art and conclusions and future recommendations. Thus this book will be useful not only for dentists and dental hygienists, but also for undergraduate and postgraduate students and teachers. This project could not have been completed without the assistance and support of my family, friends, and colleagues. I offer my deepest thanks to my wife Ingrid and my daughter Eva and my son Torbjorn and their families, as well as to all my other relatives and friends, for their patience and understanding throughout the last 5 years

in which I spent almost every night, weekend, and vacation preparing the material for these five volumes. I also wish to thank my wonderful staff at the Department of Preventive Dentistry, Public Dental Health Service, County of Varmland, for all their service, and particularly my assistant, Pia Hird, who typed most of my manuscript. I owe special thanks to Art Director Fredrik Persson, Dr Jorgen Paulander, and the Dumex Company for their excellent support with computer-aided illustrations, and to Associate Professor Joan Bevenius for her work in checking my English manuscript. I am very grateful to all my colleagues and friends around the world and to several publishers (Munksgaard International, the American Academy of Periodontology, S. Karger Medical and Scientific Publishers, FDI World Dental Press, WHO Oral Health Unit), who have generously permitted me to use their illustrations (about 20% of the total). Last but not least, the excellent cooperation of the publisher is gratefully acknowledged.

Chapter 1. Etiologic Factors Involved in Dental Caries Introduction Dental caries is an infectious, transmissible disease. As early as 1954, Orland et al demonstrated that, although germ-free animals do not develop caries, even with frequent sugar intake, all animals in the group rapidly develop carious lesions when human cariogenic bacteria (mutans streptococci) are introduced in the mouth of one animal. Specific bacteria (acidogenic and aciduric) that colonize the tooth surfaces are recognized as etiologic factors in dental caries. Frequent intake of fermentable carbohydrates, such as sugar, is regarded only as an external (environmental) modifying risk factor or prognostic risk factor. In the presence of these and other external risk factors, the outcome may be modified by internal host factors, such as the quality of the teeth and the amount and quality of saliva (Fig 1): 1. Microflora: acidogenic bacteria that colonize the tooth surface. 2. Host: quantity and quality of saliva, the quality of the tooth, etc. 3. Diet: intake of fermentable carbohydrates, especially sucrose, but also starch. 4. Time: total exposure time to inorganic acids produced by the bacteria of the dental plaque. The development of a clinical carious lesion involves a complicated interplay among a number of factors in the oral environment and the dental hard tissues. A simplified explanatory model of the major events is illustrated in Fig 2. The carious process is initiated by bacterial fermentation of carbohydrates, leading to the formation of a variety of organic acids and a fall in pH. Initially, H+ will be taken up by buffers in plaque and saliva; when the pH continues to fall (H+ increases), however, the fluid medium will be depleted of OH- and PO34-, which react with H+ to form H2O and HPO24- On total depletion of these compounds, the pH can fall below the critical value of 5.5, at which point the aqueous phase becomes undersaturated with respect to

hydroxyapatite. Therefore, whenever surface enamel is covered by a microbial deposit, the ongoing metabolic processes within this biomass cause fluctuations in pH, and occasional steep falls in pH, which may result in dissolution of the mineralized surface. In their classic study of experimental caries in humans, von der Fehr et al (1970) showed that, in the absence of oral hygiene (ie, with free accumulation of plaque and rinsing nine times a day with a sucrose solution), clinical signs of enamel caries develop within 3 weeks. When the same research team repeated the study, but introduced chemical plaque control (rinsing twice a day with 0.2% chlorhexidine solution), the subjects did not develop caries, even though they rinsed with sucrose solution nine times a day for 3 weeks (Loe et al, 1972). In other words, when the etiologic factor was suppressed or eliminated, the precondition for caries did not exist, and no lesions developed, despite the subjects' very frequent exposure to sucrose. Like the inflammation induced in the gingival soft tissues adjacent to the gingival plaque, carious lesions of enamel, which develop on individual tooth surfaces beneath the undisturbed bacterial plaque, represent the net result of an extraordinarily complex interplay among "harmless" and "harmful" bacteria, antagonistic and synergistic bacterial species, their metabolic products, and their interaction with the many salivary and other host factors. This explains why combinations of different nonspecific plaque control programs have been so effective against caries, gingivitis, and periodontitis (for review, see Axelsson, 1994, 1998). However, more recently, there has been intense interest in the role plaque (amount, formation rate, and ecology) and specific cariogenic microflora play in the etiology of dental caries. Fig 1 Diagram of the development of dental caries. Interaction among etiologic risk factors (microflora), external modifying risk factors (diet), internal modifying risk factors (host), and time exposed. (Modified from Keyes, 1960.) Fig 2 Development of noncavitated enamel caries. (Modified from Fejerskov and Clarkson, 1996.)

Role of Plaque Development of plaque According to Dawes et al (1963), dental plaque is "the soft tenacious material found on tooth surfaces which is not readily removed by rinsing with water." It is estimated that 1 mm3 of dental plaque, weighing about 1 mg, will contain more than 200 million bacteria. Other microorganisms, such as mycoplasma, "yeasts," and protozoa, also occur in mature plaque; sticky polysaccharides and other products form the so-called plaque matrix, which constitutes 10% to 40% by volume of the supragingival plaque. The most readily discernible plaque on the smooth surfaces of the teeth, along the gingival margin, is termed dentogingival plaque. Dentogingival plaque on the approximal surfaces, apical to the contact points, is called approximal dental plaque.

Plaque occurring below the gingival margin, in the gingival sulcus or in the periodontal pocket, is known as subgingival plaque (Theilade and Theilade, 1976). Occlusal or fissure plaque may also be formed, particularly in erupting molars. Although there are more than 350 species of bacteria in the oral cavity, only a few have the ability to colonize a newly cleaned tooth surface. This initial association depends on the presentation and interaction of surface molecules on the bacteria and the pellicle-coated tooth surface. These molecules are vulnerable to alteration by chemical agents. Plaque adhesion is especially favored by high free energy of the tooth surfaces and the microorganisms. The initial bacteria are called pioneer colonizers, because they are hardy and successfully compete with the other members of the oral flora for a place on the tooth surface (Gibbons and Van Houte, 1980). These pioneer colonizers are mainly the streptococcal strains S oralis, S mitior, and S sanguis. The deposition of these pioneer species is not a chance occurrence, but the outcome of an exquisitely sensitive interaction between protein adhesions on the surface of the colonizing bacteria and carbohydrate receptors on the salivary components adsorbed to the tooth surface. After initial deposition, clones of pioneer colonizing bacteria, in particular Streptococcus sanguis, begin to expand away from the tooth surface to form columns that move outwardly in long chains of pallisading bacteria. These parallel columns of bacteria are separated by uniformly narrow spaces. Plaque growth proceeds by deposition of new species into these open spaces (Listgarten et al, 1975). Figure 3 illustrates a cross section of such columns and open spaces. These newly deposited species attach to pioneer species in a specific, molecular locking manner. Expansion of existing species in a lateral direction causes the interbacterial spaces to merge. It is hypothesized that, when the spaces are close enough, a starter substance is secreted by bacteria within the plaque matrix, signaling the surrounding bacteria to undergo a growth spurt. Within a short time, the tooth surface adjacent to the gingiva is covered by intermeshed bacteria. New bacteria derived from saliva or surrounding mucous membranes now sense only the bacterialaden landscape of the tooth surface and attach by a bonding interaction to bacteria already attached to the plaque. These associations, called intergeneric coaggregations, are mediated by specific attachment proteins that occur between two partner cells (Di Renzo et al, 1985; Kolenbrander, 1988). All this activity occurs within the first 2 days of plaque development and, for descriptive purposes, is called phase I of plaque formation (Theilade et al, 1976). After 24 to 48 hours, continuous plaque has formed along the gingival margin (Fig 4). The plaque is dominated by cocci and a few rods. In phase II of plaque development, the remaining interstices are occupied by increasing levels of gram-positive rods, such as Actinomyces viscosus, and gramnegative cocci, including Neisseria and Veillonella species (Fig 5). The outer surface of the gingival plaque is covered by tall rods. Figure 6 illustrates the thickness of freely accumulated gingival plaque after 2, 3, and 4 days. There is a dramatic increase in plaque thickness after 3 and 4 days compared to the first 2 days. Now the gingival plaque is mature, and so-called homeostasis is established among the different

microorganisms. In phase III, 5 to 7 days after initiation, plaque begins to migrate subgingivally, and bacteria and their products permeate and circulate in the pocket. In phase IV, 7 to 11 days after initiation, the diversity of the flora increases to comprise motile bacteria, including spirochetes and vibrios as well as fusiforms. Attached gingival plaque fills the gingival sulcus, while spirochetes and vibrios move around in the outer and more apical regions of the sulcus (Fig 7). Fig 3 Cross section of 2-day-old gingival plaque growth. (From Listgarten et al, 1975. Reprinted with permission.) Fig 4 Scanning electron micrograph showing 24 to 48 hours of continuous plaque formation along the gingival margin (arrow). Plaque is dominated by cocci (right). (Courtesy A. Saxton) Fig 5 Surface of 3- to 4-day-old gingival plaque. (Courtesy A. Saxton)

Fig 6 Thickness of gingival plaque: 2, 3, and 4 days old. (From Listgarten, et al, 1975. Reprinted with permission.) Fig 7 Gingival plaque and motile microflora filling the gingival sulcus. (From Listgarten, 1976. Reprinted with permission.)

Measurement of plaque Amount of accumulation Several indices for recording supragingival plaque have been developed. The two most frequently used are the Plaque Index (PI), developed by Silness and Loe (1964), and O'Leary's Plaque Index (O'Leary et al, 1972). The Silness and Loe Plaque Index has a four-point scale: • Score 0 = The tooth surface is clean. • Score 1 = The tooth surface appears clean, but dental plaque can be removed from the gingival third with a sharp explorer. • Score 2 = Plaque is visible along the gingival margin. • Score 3 = The tooth surface is covered with abundant plaque. O'Leary's Plaque Index is based on the visible continuous plaque along the gingival margin after staining. Four or six sites per tooth are examined, and the percentage of tooth surfaces exhibiting stained plaque is calculated. Unlike Silness and Loe's PI, no

attempts are made to evaluate the area of tooth surface covered by plaque. O'Leary's Plaque Index is most commonly used for evaluation of the oral hygiene standard of the individual patient and for patient motivation, based on self-diagnosis. Used in dental practice, the Plaque Index is capable only of revealing areas that the patient has failed to clean effectively, even though he or she may have made a special effort on the day of the dental appointment; it does not indicate the rate at which plaque forms in the individual or the oral hygiene status 1 week before or after the dental appointment. This accounts for the failure by clinicians as well as examiners in clinical studies to observe the correlation between on one hand, the amount and location of plaque and, on the other, the sites of carious lesions. Despite these limitations, disclosure of plaque by staining is the fastest and most efficient method for self-diagnosis by the patient. The technique also allows the clinician to locate remaining plaque and to demonstrate the close relationship between the localization of plaque and the presence of gingivitis and dental caries (Figs 8 and 9). Prevention of dental caries and gingivitis must, therefore, be based on plaque control. The telemetric method, developed by Graf and Muhlemann (1966), allows in vivo measurement of the "true" pH on the tooth surface beneath the undisturbed plaque. The importance of the age, amount, and composition of plaque, as well as different concentrations of sugar, can thereby be evaluated. Using the telemetric method, Imfeld (1978a) showed that rinsing with a 10% sucrose solution causes a dramatic drop in pH to below 4 in 3-day-old interdental plaque. Such plaque is typical for the approximal surfaces of the molars and premolars in a toothbrushing population. In contrast, the fall in pH in immature lingual plaque (12 hours old) is very limited (Fig 10). Firestone et al (1987) used the same telemetric test in vivo, measuring the pH drop after subjects rinsed with a 10% sucrose solution. Four different sites on molars with approximal plaque (Fig 11) were compared to plaque-free approximal surfaces. The authors concluded: "removing plaque from interdental surfaces significantly reduced the exposure of the surfaces to plaque acids following sucrose rinse. This further supports mechanical removal of plaque from interdental surfaces as a means of reducing dental caries." In toothbrushing populations, that is, those who have an established habit of using a toothbrush and fluoride toothpaste daily, dental plaque more than 2 days old is located mainly on the approximal surfaces of the molars and premolars, partly subgingivally. Access with a toothbrush to the wide approximal surfaces is limited by the buccal and lingual papillae. At least in European countries, although daily toothbrushing with a fluoride dentifrice is an established oral hygiene habit, special aids to approximal oral hygienesuch as dental floss, dental tape, toothpicks, and interdental brushesare used daily by fewer than 10% of the population. These conditions explain why caries, gingivitis, and marginal periodontitis are much more prevalent on the approximal surfaces of the molars and premolars than on the buccal and lingual surfaces of the dentition. Rate of accumulation (Plaque Formation Rate Index)

The quantity of plaque that forms on clean tooth surfaces during a given time represents the net result of interactions among etiologic factors, many internal and external risk indicators and risk factors, and protective factors: • The total oral bacterial population • The quality of the oral bacterial flora • The anatomy and surface morphology of the dentition • The wettability and surface tension of the tooth surfaces • The salivary secretion rate and other properties of saliva • The intake of fermentable carbohydrates • The mobility of the tongue and lips • The exposure to chewing forces and abrasion from foods • The eruption stage of the teeth • The degree of gingival inflammation and volume of gingival exudate • The individual oral hygiene habits • The use of fluorides and other preventive products, such as chemical plaque control agents This observation has been the rationale for the development of the Plaque Formation Rate Index (PFRI) by Axelsson (1989, 1991). The index includes all but the occlusal tooth surfaces and is based on the amount of disclosed plaque freely accumulated (de novo) in the 24 hours following professional mechanical toothcleaning (PMTC), during which period subjects refrain from all oral hygiene practices. In a pilot study on 50 adult subjects, adherent plaque was disclosed on 5% to 65 % of the total number of tooth surfaces (for details on materials and methods, see Axelsson, 1989, 1991). On the basis of this study, the following five-point scale was constructed for the PFRI. • Score 1 = 1% to 10% of surfaces affected: very low • Score 2 = 11% to 20% of surfaces affected: low • Score 3 = 21% to 30% of surfaces affected: moderate • Score 4 = 31% to 40% of surfaces affected: high • Score 5 = More than 40% of surfaces affected: very high

The PFRI was evaluated in a large-scale cross-sectional study of 14-year-old schoolchildren (n = 667) in the city of Karlstad, Sweden, in 1984. The subjects were followed over a 5-year period, up to the age of 19 years (Axelsson, 1989, 1991). Many indicators and factors possibly related to PFRI were also evaluated, including (1) caries prevalence and caries incidence; (2) gingival inflammation; (3) Plaque Index; (4) dietary intake during the 24 hours of free plaque accumulation; (5) salivary levels of Streptococcus mutans and glucosyl transferase; (6) agglutinin levels in resting saliva; and (7) oral hygiene, dietary, and fluoride habits. Figure 12 shows the frequency distribution of PFRI scores 1 to 5 among the 14-yearold schoolchildren. The majority were low (score 2 = 48%) or moderate (score 3 = 27%) plaque formers. However, the standard of oral hygiene is very high among schoolchildren in Karlstad, and, as a consequence, the caries prevalence is low. The relationship between caries prevalence (the mean number of decayed or filled surfaces) and different scores is presented in Fig 13. These results indicated a threshold for caries risk between PFRI scores 2 and 3, and this was subsequently confirmed in the longitudinal part of the study over 5 years (Axelsson, 1989, 1991). Among other observations from the study were: 1. Individuals with a PFRI score of 4 or 5 had considerably higher scores for gingival bleeding than did individuals with a PFRI score of 1 or 2. 2. An initially high Plaque Index usually correlated with PFRI scores 3 to 5. 3. There was no significant correlation between different salivary S mutans levels and PFRI scores. 4. The level of salivary glucosyl transferase was lower in individuals with a PFRI score of 4 or 5 than in those with a score of 1 or 2, probably because glucosyl transferase had already accumulated in the matrix of the plaque in the high and very high plaque formers. 5. The scores for individuals with a very low and low PFRI (scores 1 and 2, respectively) tended to remain constant over the 5-year period, while the scores of some individuals with PFRI scores of 3 to 5 tended to vary, increasing or decreasing by 1 unit. This final observation indicates that plaque formation rates in individuals with a PFRI score of 4 or 5 can be reduced; such individuals should, therefore, be thoroughly evaluated to identify the factors that contribute to their rapid plaque formation. Needsrelated preventive measures could then be introduced. For example, there is a strong correlation among plaque formation rate, the severity of gingival inflammation, and the volume of gingival exudate (Axelsson, 1989; Quirynen et al, 1986a; Ramberg et al, 1994a,b, 1995; Saxton 1973, 1975). Initially intensive and frequent mechanical and chemical plaque control, both self-care and

professional, is indicated in individuals with PFRI scores of 4 and 5 and high gingival index scores to heal all inflamed sites as quickly as possible and thereby reduce the plaque formation rate. If the high plaque formation rate is associated with inadequate salivary secretion, frequent plaque control measures (before every meal) should be supplemented with salivary stimulation, provided by the use of fluoridated chewing gum immediately after every meal. A high intake of fermentable carbohydratesparticularly sucrosewill result in sticky plaque, rich in polysaccharides, and an increased plaque formation rate (Carlsson and Egelberg, 1965). Needs-related prevention for individuals with a PFRI score of 4 or 5 and a frequent intake of sugar-containing products should, therefore, emphasize not only frequent plaque control but also a reduction in the frequency of sugar intake. In the above study, it was observed that some individuals with a PFRI score of 5 had consumed several bananas during the 24-hour period of free plaque accumulation. Many other factors are also related to plaque formation rate. For example, antimicrobial proteins of human whole saliva may influence plaque formation rate (Box 1). The PFRI has recently been applied in studies on different populations and age groups. From more than 1,000 17 to 19 year olds in the city of Karlstad, Sweden, 30% with the highest gingival index score were selected to participate in a 4-month doubleblind mouthrinse study. At baseline, most of the subjects had PFRI scores of 3 (more than 40%) or 4 (about 25%) (Axelsson et al, 1994). Subjects with the highest gingival index scores also had the highest PFRI scores. In addition, sites with gingival inflammation had significantly higher plaque formation rates than healthy gingival sites (Rahmberg et al, 1995). Brazil has the highest caries prevalence in the world. In Sao Paulo, a 3-year cariespreventive study based on self-diagnosis and self-care was carried out in 12- to 15year-old schoolchildren; the PFRI was used as a tool for self-diagnosis and establishment of needs-related oral hygiene habits. At baseline, almost 100% of the 12-year-old schoolchildren had a PFRI score of 5. The mean percentage of surfaces with reaccumulated plaque was more than 70%, probably because of the extremely high caries prevalence, a high gingival index, and the presence of erupting permanent teeth. At reexamination of the subjects 3 years later, the PFRI had dropped significantly: most of the 15 year olds had scores of 3 or 4. The main contributing factors were an improvement in oral hygiene habits and gingival health and the fact that all teeth were now fully erupted (Albander et al, 1995; Axelsson et al, 1994; Buischi et al, 1994). In Duisburg, Germany, the PFRI was evaluated in different age groups of children; preschool children, children with mixed dentitions and erupting permanent teeth, and children with fully erupted teeth. Children with erupting teeth had the highest PFRI scores (Fig 14). However, the German children generally had higher PFRI scores than did Swedish children of comparable age with very low caries prevalence, excellent gingival conditions, and good oral hygiene habits (Fig 15) (Axelsson, 1991; Cunea and Axelsson, 1997). According to the World Health Organization's Data Bank

(1993), caries prevalence is high among 12-year-old German children. Pattern of plaque reaccumulation As discussed earlier, plaque formation rate is influenced by such factors as (1) the anatomy and surface morphology of the teeth; (2) the stage of eruption and functional status of the teeth; (3) the wettability and surface tension of the tooth surfaces (both intact and restored surfaces); and (4) gingival health and volume of gingival exudate. The pattern of plaque reaccumulation will also be influenced by these factors, but may differ somewhat between, on the one hand, tooth surfaces exposed to chewing forces, abrasion from foods, and friction from the dorsum of the tongue, the lips, and the cheeks, and, on the other hand, less accessible areas, such as approximal sites, along the gingival margin, and in irregularities such as occlusal fissures. These less accessible areas are often designated "stagnation areas" for plaque. In a 6-week study by Lang et al (1973), plaque reaccumulation was registered in four groups of dental students who carried out oral hygiene procedures (mechanical toothcleaning by self-care) with different frequencies: twice daily or every second, third, or fourth day. Figure 16 shows the pattern of reaccumulated plaque according to the Silness and Loe (1964) Plaque Index (scores 0 to 3) on the distal, mesial, facial, and lingual surfaces of the maxillary and mandibular teeth. After only 12 hours of free plaque reaccumulation, there was visible plaque on some of the approximal surfaces of the molars and the lingual surfaces of the mandibular molars (score 2). After 48 hours, almost 100% of these surfaces and most of the remaining approximal surfaces had scores of 2 or 3. The pattern of visible plaque after 2 and 3 days seems to be similar, except on the facial surfaces. According to Listgarten (1976), freely accumulated plaque is about five times thicker after 3 days than after 2 days (see Fig 6). This explains why gingivitis developed in the group of students cleaning only every third or fourth day but not in those who cleaned at least every second day. It also explains the striking difference in response to rinsing with 10% sucrose solution shown by Imfeld (1978): a dramatic fall in pH on approximal surfaces covered by 3-day-old plaque compared to the pH on lingual surfaces covered by immature (12-hour) plaque (see Fig 10). Figure 17 presents the percentage of freely reaccumulated (de novo) plaque, 24 hours after PMTC, in 667 14-year-old children in the city of Karlstad (Axelsson, 1989, 1991). Plaque reaccumulation was greatest on the mesiolingual and distolingual mandibular surfaces (33%), particularly on the molars, followed by the mesiobuccal and distobuccal surfaces of both maxillary and mandibular teeth, particularly on the molars. There was almost no plaque reaccumulation (3%) on the palatal surfaces of the maxillary teeth, mainly because of friction from the rough dorsum of the tongue. Figures 18 and 19 illustrate the percentage of de novo plaque on maxillary and mandibular tooth surfaces, respectively, 24 hours after PMTC, in young German subjects (Cunea and Axelsson, 1997). The highest percentages are found in 6 to 14 year olds with many erupting teeth on distobuccal and mesiobuccal surfaces of molars, and on distolingual and mesiolingual surfaces of mandibular molars. Carvalho et al (1989) studied the pattern and amount of de novo plaque, 48 hours after

PMTC, on the occlusal surfaces of partly and fully erupted first molars. Figure 20 illustrates the heavy plaque reaccumulation, particularly in the distal and central fossae, in the eruptin maxillary and mandibular molars, in contrast to the reaccumulation in the fully erupted molars, which are subjected to normal chewing friction. Abrasion from normal mastication significantly limits plaque formation; this explains why almost 100% of occlusal caries in molars begins in the distal and central fossae during the eruption period of 14 to 18 months. It is important to differentiate between plaque indices and plaque reaccumulation rate (PFRI). For successful primary and secondary prevention of dental caries and periodontal diseases, an understanding of plaque formation rates and patterns is essential. Mechanical removal of dental plaque according to the nonspecific plaque hypothesis is a rational method for prevention and control of periodontal diseases as well as dental caries, because it is directed toward the cause (etiology) of these diseases. However, for cost-effectiveness, the program should be related to the pattern of plaque reaccumulation, PFRI, and predicted risk. (For reviews on plaque formation rate and the role of needs-related plaque control see Axelsson 1994, 1998.) Fig 8 The anterior teeth of a 12-year-old boy with gingivitis at the following sites: 13 mesiobuccal (mb); 12 distobuccal (db); 21 db; 22 mb, db; 23 mb; 43 mb; 42 db; 32 db; and 33 mb. Enamel caries is found at 13 mb, 43 mb, 42 b, 32 db, 33 mb, and 34 mb. There is a cavity on 22 d. Fig 9 A disclosing agent was used to visualize plaque.Observe the close relationship between the localization of gingivitis, carious lesions (Fig 8), and dental plaque.

Fig 10 Limited drop in pH in 12hour-old lingual plaque compared to critical drop in pH beneath 3-day-old (3d) interdental plaque after rinsing with 10% sucrose solution. (Courtesy T. Imfeld.)

Fig 11 pH drop in molars with approximal plaque at four different sites after rinsing with sucrose solution. (From Firestone et al, 1987.) Fig 12 Frequency distribution of PFRI scores (1 to 5) among 14-yearold schoolchildren. (From Axelsson, 1989, 1991.)

Fig 13 Relationship of caries prevalence and PFRI scores. (From Axelsson, 1989, 1991.)

Fig 14 Plaque accumulation, by age. (From Cunea and Axelsson, 1997.)

Fig 15 PFRI scores in German and Swedish schoolchildren. (From Cunea and Axelsson, 1997.) Fig 16 Pattern of reaccumulated plaque according to the Silness and Loe Plaque Index. (Modified from Lang et al, 1973.) Fig 17 Percentage of freely accumulated plaque, 24 hours after PMTC, in 667 14-year-old children in the city of Karlstad. (From Axelsson, 1989, 1991.) Fig 18 Percentage of freely accumulated plaque on maxillary tooth surfaces 24 hours after PMTC. (From Cunea and Axelsson, 1997.) Fig 19 Percentage of freely accumulated plaque on mandibular tooth surfaces 24 hours after PMTC. (From Cunea and Axelsson, 1997.) Fig 20 Pattern of freely accumulated plaque on the occlusal surfaces of partly and fully erupted first molars 48 hours after PMTC. (Modified from Carvalho et al, 1989.) Role of the Oral Environment Introduction In certain aspects, the oral cavity may be regarded as a single microbial ecosystem. A major regulatory factor is the flow rate of saliva, which decreases to almost 0.0 mL/min during sleep, is approximately 0.4 mL/min at rest, and increases to 2.0 mL/min after stimulation. Although saliva is not a good medium for supporting the growth of many bacteria, 1.0 mL of whole saliva may contain more than 200 million microorganisms, representing more than 300 different species. Most originate from local environments in the oral cavity, but a minority belong to the so-called normal microflora of saliva and obtain nutrients from salivary proteins. All the surfaces of the oral cavity are colonized by microorganisms. The facultatively anaerobic streptococci constitute an essential part of the microflora that constantly colonize the mucous membranes and the teeth. Microorganisms are regularly swallowed with saliva and the amount within the oral cavity fluctuates, simply because the microbial deposits building up on mucous membranes and, in particular, on tooth surfaces grow and multiply, thus providing a reservoir for the oral environment. Fluctuations also occur during sleeping and waking hours, and also as a result of such activities as eating and drinking and oral hygiene procedures. Because the composition of the microflora in mixed saliva is mainly a result of the microorganisms that colonize oral surfaces, the salivary microflora to some extent reflects the gross composition of the microbial deposits on the various oral surfaces.

The oral cavity presents two types of surface for colonization by bacteria, the soft tissues and the hard tooth surfaces, modified to some extent by a coating of saliva, or, in the case of the hard surfaces, by a pellicle formed by adsorption of salivary components. A distinct and important difference between the two types of surface is that the soft tissue surfaces are lost when the cells are shed; thus, readherence of bacteria is essential for survival. In contrast to the hard surfaces which will support heavy deposits of bacteria in dental plaque, the soft tissue surfaces do not support formation of complex layers of bacteria (biofilms). It is also likely that microbes attached to desquamated epithelial cells spread, via saliva, to different tooth surfaces and typically colonize sheltered regions: interproximal spaces, gingival margins, and occlusal fissures (Saxton, 1975). Colonization of microenvironments The oral cavity consists of several major and minor compartments, each constituting a separate microenvironment not easily affected by major events in the oral cavity. Examples of major compartments are the tongue, the oral mucosa, and the tonsils. The different approximal tooth surfaces, occlusal fissures, and gingival sulci are regarded as minor compartments. A specific area that supports a bacterial flora is termed a habitat. The flora of a habitat develops through a series of stages, collectively called colonization. Colonization is a complex process, because it involves not only interaction between bacteria and their environment but also interactions among different bacteria. The first important prerequisite for colonization is access. The organisms must be able to enter the habitat and consequently they must be able to be transmitted from one habitat to another. For example, mothers can serve as reservoirs for oral bacteria, which they transmit to their children. Within a single host, bacterial reservoirs can aid survival of the organism. In the human mouth, not only the oral mucosa but also the tongue and tonsils may serve as as reservoirs for bacteria, which, under favorable conditions, may colonize the teeth as well as the periodontal pockets. It is well known that the dorsum of the tongue is the main reservoir for Streptococcus salivarius, which is a very potent, cariogenic (acidogenic) bacteria. In one study, however, higher numbers of S mutans were repeatedly found on the dorsum of the tongue after five thorough scrapings with a tongue scraper than prior to scraping, indicating this to be an important reservoir (Axelsson et al, 1987). Lindquist et al (1989) found a significant correlation between the prevalence of S mutans in saliva and its prevalence on the dorsum of the tongue. These data support the inclusion of the dorsum of the tongue in oral hygiene procedures, at least in patients highly infected by periopathogens and/or cariogenic bacteria, such as S mutans. Although there are general definitions of habitats, studies of the oral microflora should always include careful definition of the habitats being examined. It is important to recognize that the physical dimensions of a habitat do not fall within specific limits: The whole oral cavity, an occlusal tooth surface, or even a defined area on the occlusal surface may be considered a habitat. In oral microbiology, changes in the flora of a habitat such as the saliva may indicate, for example, patients at risk of developing caries, while changes in tooth surface microenvironments can

identify a surface at risk of disease. Effect of plaque ecology Owing to differences in local environmental conditions, the microflora of mucosal surfaces differs in composition from that of dental plaque. Similarly, the plaque microflora varies in composition at distinct anatomic sites on the toothfor example, in fissures, on approximal surfaces, and in the gingival crevice. The resident microflora of a site acts as part of the host defenses by preventing colonization by exogenous (and often pathogenic) microorganisms. The early colonizers of the tooth surface include members of the genera Streptococcus, Actinomyces, Haemophilus, Neisseria, and Veillonella (Liljemark et al, 1986; Nyvad and Kilian, 1987). These bacteria adhere to the acquired enamel pellicle by specific and nonspecific molecular interactions between adhesions on the cell and receptors on the surface (Busscher et al, 1992; Gibbons, 1989). Once established, the microflora at a site remains relatively stable over time, despite regular minor disturbances in the oral environment (Marsh, 1989). This stability (termed microbial homeostasis) stems not from any metabolic indifference among the components of the microflora but rather from a dynamic balance of microbial interactions, including both synergism and antagonism (Sanders and Sanders, 1984). It has been proposed that the ability to maintain homeostasis within a microbial community increases with its species diversity (Alexander, 1971). In dental plaque, diversity is enhanced by the development of food chains between bacterial species and their use of complementary metabolic strategies for the catabolism of endogenous nutrients, such as glycoproteins and proteins. Individual species possess different but overlapping patterns of enzyme activity, so that certain mixed cultures of oral bacteria can synergistically degrade complex host molecules (van der Hoeven and Camp, 1991). Antagonism is also a major mechanism in maintaining microbial homeostasis in plaque (James and Tagg, 1988; Marsh, 1989). Unless removed by diligent oral hygiene, plaque accumulates preferentially at stagnant or retentive sites, such as the posterior approximal surfaces, the fissures of erupting molars, and along the gingival margin. Igarashi et al (1989), studying the effect of rinsing with sucrose solution on 4-day-old plaque, showed that the fall in pH in plaque was significantly greater on the approximal surfaces of the molars than in the fissures. The ecological plaque hypothesis, introduced by Marsh (1991), proposes that a change in a key environmental factor (or factors) will trigger a shift in the balance of the resident plaque microflora, and this might predispose a site to disease. The occurrence of potentially pathogenic species as minor members of the resident plaque microflora would be consistent with this proposal. Under the conditions that prevail in health, these organisms would be only weakly competitive and might also be suppressed by intermicrobial antagonisms, thus constituting only a small percentage of the plaque microflora, without clinical effect. Microbial specificity in disease would result from the fact that only certain species are competitive under the new (changed) environmental conditions.

It is a basic tenet of microbial ecology that a major change in an ecosystem produces a corresponding disturbance in the stability of the resident microbial community (Alexander, 1971; Brock, 1966; Fletcher et al, 1987). An increasing mass of plaque impedes penetration by saliva to protect the enamel. Microbial homeostasis can break down, and major shifts in the composition of the microflora can occur. For example, frequent consumption of fermentable dietary carbohydrates is associated with an increased risk of dental caries (Loesche, 1986). Such diets lead to a rise in the proportions of mutans streptococci (MS) and lactobacilli, with a concomitant fall in levels of other streptococci, especially members of the Streptococcus oralis group, which include S sanguis, S oralis, and S mitis (Dennis et al, 1975; de Stoppelaar et al, 1970; Minah et al, 1985; Staat et al, 1975). The metabolism of plaque also changes from a heterofermentative pattern to one in which sugars are converted primarily to lactic acid. Studies by Bradshaw et al (1989a) have shown that low pH, rather than the availability of carbohydrates per se, is the factor driving the selection of potentially cariogenic species. This selection is at the expense of acid-sensitive species, some of which are associated more with oral health. The experiment was repeated to determine if there were a "critical pH" at which this breakdown in homeostasis would occur. Plaque microorganisms were pulsed with glucose in three replicate experiments in which the pH was allowed to fall only to fixed values of pH 5.5, 5.0, or 4.5. The microbial community was disrupted irreversibly only when the pH fell regularly below 5.0 (Bradshaw et al, 1989b). The predominant species in these experiments always became Streptococcus mutans, Lactobacillus casei, and Veillonella dispar (Bradshaw et al, 1989b). These three species have been associated with nursing caries (Milnes and Bowden, 1985) and progressing caries (Boyar et al, 1989) in humans. Pure culture studies have also shown that the growth of these three species is less sensitive to low pH than is that of other oral bacteria (Bradshaw et al, 1989a; Harper and Loesche, 1986). Furthermore, mouthrinsing with acidic buffers (pH 3.9) was found to increase the proportions of mutans streptococci in human fissure plaque (Svanberg, 1980). Studies by van Houte et al (1991) have shown that streptococci other than mutans streptococci will increase the acidogenic potential of plaque at low pH. Collectively, these findings show that the selection of cariogenic species following regular sugar consumption is likely to be a consequence of their aciduric physiology, which enables them to compete successfully at low pH. On the other hand, in subjects with a conventional low-sugar diet, the composition of plaque microflora would be stable, only small amounts of acid would be produced at main meals, and the processes of demineralization and remineralization would be in equilibrium. If the frequency of fermentable dietary carbohydrate intake were to increase, however, there would be longer periods of low plaque pH (Loesche, 1986a). Such conditions would favor the proliferation of mutans streptococci and lactobacilli at the expense of less acid-tolerant species, tipping the equilibrium toward demineralization (Fig 21). Factors reducing the flow of saliva (eg, xerostomia) would lead to similar shifts in the microflora. Greater numbers of mutans streptococci and lactobacilli would lead to

even faster rates of acid production from sugars, enhancing demineralization still further, while the elevated levels of lactic acid in plaque would also select for Veillonella spp. Acid-sensitive species, such as members of the S oralis group (eg, S sanguis, S oralis, and S mitis), would decline in proportion, thereby accounting for the widely reported inverse relationship between S sanguis and mutans streptococci seen in plaque. Other bacteria could also produce significant amounts of acid under similar conditions, at slower rates (van Houte, 1993), but nevertheless providing an explanation for demineralization in the absence of mutans streptococci. Fig 21 Ecological plaque hypothesis and prevention of caries. (Modified from Marsh, 1994.) Strategies for prevention and control of caries based on plaque ecology hypothesis According to the plaque ecology hypothesis, low pH (less than 5) will promote overgrowth of aciduric microorganisms, such as the cariogenic mutans streptococci and lactobacilli, at the expense of less acid-tolerant plaque microorganisms, such as S oralis, which are associated with healthy tooth surfaces. Therefore the treatment strategy would be to increase plaque pH and thereby promote reestablishment of the harmless normal microflora of the tooth surfaces (see Fig 21). Increased pH can successfully be achieved by a combination of the following preventive measures: 1. Frequent mechanical removal of the plaque from all tooth surfaces: "Clean teeth never decay," and frequent removal of plaque (once or twice a day) limits the thickness of reaccumulated plaque, ensuring that saliva has accessibility for diluting and buffering the acid that is formed. 2. Reduction of sugar clearance time by reducing the intake of sticky sugar-containing products. 3. Use of sugarless chewing gums containing fluoride and chlorhexidine as a "dessert" for 15 to 20 minutes directly after every meal (including between-meal snacks). Use of this kind of gum has several beneficial effects: 1. Salivary flow is increased and the acid formed during the meal is diluted and buffered. 2. Fluoride will reduce acid formation by the acidogenic microorganisms at low pH. 3. Chlorhexidine has not only a nonspecific antiplaque effect but also a specific effect on mutans streptococci and acid formation by acidogenic microorganisms. 4. Fluoride ions and minerals from the increased salivary flow will accelerate remineralization directly after the acid attack during the meal. This recommendation is very important for caries-susceptible patients, particularly

those with xerostomia (for reviews on plaque ecology related to caries etiology, see Bowden, 1997; Bowden and Edwardsson, 1994; Marsh, 1993.) Role of Specific Cariogenic Microflora Introduction Microorganisms implicated in the etiology of dental caries must be acidogenic as well as aciduric. To initiate carious lesions in enamel, the microorganisms must also be able to colonize the tooth surface and survive in competition with less harmful species, forming biofilmsthe so-called dental plaque. As early as 1960, Fitzgerald and Keyes showed that certain microorganisms isolated from human dental plaque, when inoculated in germ-free rodents on a high-sucrose diet, resulted in the spread of rampant caries. Therefore, dental caries should be regarded as an infectious, transmissible disease. There is abundant support for the so-called specific plaque hypothesis, introduced by Loesche (1982, 1986), which proposes that some specific species of the plaque flora be regarded as major pathogens in the etiology of dental caries. Included in the major pathogens are those bacteria associated with caries in humans and also able to induce carious lesions in experimental animals. The most important are the mutans streptococci: there are seven species, of which two, S mutans and S sobrinus, are closely associated with caries in humans. The remainder are found in animals or, if present in humans, are not highly cariogenic. The relationship of S sobrinus to caries in humans is not as well understood as that between S mutans and caries, and only recent studies identify the species separately. The second genus closely associated with caries is Lactobacillus, commonly isolated from carious dentin (Edwardsson, 1974), thought to be its main habitat in the mouth. Compared to the extensive research into identification of the mutans streptococci, much less attention has been paid to speciating lactobacilli isolated from carious lesions. Also associated with the etiology of dental caries, but considered to be less cariogenic than S mutans, S sobrinus, and Lactobacillus, are Actinomyces odontologica, Actinomyces naeslundii, and some other species of MS. To clarify the role of S mutans in the etiology of dental caries, many cross-sectional and longitudinal human studies have been conducted over the past two to three decades, particularly by the Krasse and Bratthall research groups in Sweden. Cariogenicity of mutans streptococci Mutans streptococci are acidogenic as well as aciduric and can adhere to tooth surfaces (Gibbons et al, 1986). Mutans streptococci can produce extracellular and intracellular polysaccharides from sucrose. Intracellular polysaccharides in particular can be degraded during periods of low nutrient supply, indicating that these polysaccharides increase the virulence of some MS species (S mutans and S sobrinus). Because the microbial ecology of the mouth is highly complex, strains of the same species could vary considerably in virulence (Bowden and Edwardsson, 1994). In other words, MS fulfill all the requirements of caries-inducing bacteria.

Colonization of the teeth by mutans streptococci is highly localized; some tooth surfaces are colonized but not others. The amount of mutans streptococci in saliva is related to the number of colonized surfaces (Lindquist et al, 1989). This is the basis for saliva tests for MS. A high count in saliva (more than 1 million colony-forming units [CFUs] per 1 mL of saliva) indicates that most teeth are colonized by these bacteria, ie, that many tooth surfaces are subject to increased caries risk. However, a salivary MS count does not provide information about the origin of the bacteria, ie, the specific tooth surfaces which are colonized. The most common types of mutans streptococci, Streptococcus mutans (serotypes c, e, and f) and Streptococcus sobrinus (serotypes d and g), are present worldwide. Their prevalence differs among populations. Test values also differ, depending on the method of detection (Axelsson et al, 1987b; Bratthall et al, 1986; Beighton et al, 1989; Buischi et al, 1989; for reviews, see Bratthall, 1991; Carlsson, 1988). About 10% to 30% of a population may have little or no MS, 0 to 100,000 CFUs/mL of saliva. The percentage of individuals with very high levels of MS (> 1 million CFUs/mL of saliva) in a population may vary considerably, depending on age, caries prevalence, dietary habits, and so on. Evidence Several cross-sectional studies in human populations with relatively high caries prevalence have shown a correlation between very high salivary MS levels and very high caries prevalence (Axelsson et al, 1999b; Buischi et al, 1989; Klock and Krasse, 1977; Salonen et al, 1990; Zickert et al, 1982). This is exemplified in Fig 22, from a study of 12-year-old Brazilian children (Buischi et al, 1989). However, in populations with relatively low caries prevalence and high standards of oral hygiene, the threshold value of more than 1 million CFUs of MS/mL of saliva no longer seems to apply, as exemplified in Fig 23, from a study of 13- to 14-year-old schoolchildren in Karlstad, Sweden (Kristoffersson et al, 1986). In this population, the critical difference was between MS-negative and MS-positive subjects. As also shown in Fig 23, it was impossible to find any correlation between intake of sticky sugar products (estimated point scale) and caries prevalence, highlighting the multifactorial nature of dental caries: The lower the prevalence and incidence of caries in a population, the more difficult it is to demonstrate a significant correlation for one single etiologic or modifying factor. Transmission. Because MS require a hard, nondesquamating surface for colonization (Berkowitz et al, 1975; Carlsson et al, 1975; Catalanotto et al, 1975; Stiles et al, 1976), infants do not harbor MS until some time after tooth eruption: The major source of the infection is thought to be maternal. Evidence for this comes from several studies showing that isolates of MS harbored by mothers and their children exhibit similar or identical bacteriocin profiles (Berkowitz and Jordan, 1975; Berkowitz and Jones, 1985; Davey and Rogers, 1984) and identical plasmid or chromosomal DNA patterns (Caufield et al, 1985, 1986, 1988; Caufield and Walker, 1989; Hagan et al, 1989; Kulkarni et al, 1989). Several studies have suggested that the extent of MS colonization and, to some

degree, subsequent carious activity experienced by a child may be correlated with the mother's salivary level of MS: Mothers with high levels of MS tend to have children with high levels and vice versa (Caufield et al, 1988; Kohler et al, 1984; Kohler and Bratthall, 1978; van Houte et al, 1981). While correlations between caries or MS levels in mothers and those in their children may be explained in part by common genetic or environmental factors, others have suggested that a child's degree of colonization or disease may be dictated by the mother's levels of MS at the time of transmission. In a landmark study, Kohler and coworkers (1983, 1984) selected mothers with initially high levels of MS in saliva and determined the effects of various preventive and treatment regimens aimed at reducing MS below a predetermined threshold level. Children of these mothers were monitored for initial acquisition of MS and, subsequently, for carious activity over a 3-year period. A statistically significant difference was observed between control and experimental groups in terms of when a child acquired MS, the levels of MS harbored by the mother and child, and the child's caries outcome. Figure 24, from the longitudinal study by Kohler et al (1988), illustrates that the earlier the colonization by MS, the higher the caries prevalence at 4 years of age. In a recent study by Caufield et al (1993), oral bacterial levels of 46 mother-child pairs were monitored from the birth of the child to 5 years of age to study the acquisition of MS by the children. In 38 children, initial acquisition occurred at the median age of 26 months, during a discrete period that was designated as the "window of infectivity." In the remaining eight children (17%), MS was undetectable throughout the study (median age 56 months). No significant differences were found in salivary levels of MS or lactobacilli of mothers of children with and without MS. Comparisons between a caries-active cohort colonized by MS (9 of 38) and children without detectable MS revealed similar histories in terms of antibiotic usage, gestational age, and birth weight. Interestingly, half the children who were MS negative between the ages of 1 and 2 years were minded by caregivers other than the mother, while all the children who were caries active during this age interval were cared for by their mothers; the difference was statistically significant. This study by Caufield et al (1993) was the first to present evidence that MS is acquired during a defined period in the ontogeny of a child. Support for the notion of a discrete window of infectivity comes from other sources, including animal models. By studying mother-child and father-child pairs, Alaluusua et al (1991) found a strong correlation between teenagers and mothers with high numbers of decayed, missing, or filled surfaces and high salivary MS levels, but no such correlation in father-child pairs. The above studies in humans confirm the earlier animal studies by Fitzgerald and Keyes (1960) that dental caries is an infectious disease, transmissible by MS. Several experimental and clinical studies have also confirmed that MS can be isolated from dental plaque covering active carious lesions in enamel (Axelsson et al, 1987; Kristoffersson et al, 1985) and at the root (van Houte et al, 1990) as well as secondary carious lesions (Gonzales et al, 1995) and the margins of restorations (Wallman and Krasse, 1992; for reviews, see Bowden and Edwardsson, 1994; Loesche, 1986a). Caries incidence. Many longitudinal human clinical studies have shown correlations between high salivary MS counts and high caries incidence. In preschool-aged

children (primary dentition), correlations between salivary MS counts and caries incidence have been shown by Alaluusua et al, 1990; Kohler et al, 1988; Roeters et al, 1995; Thibodeau and O'Sullivan, 1996; and Twetman et al, 1996. In the permanent dentition, Klock and Krasse (1978, 1979) showed that 9 to 12 year olds with high salivary MS levels developed significantly more new carious surfaces than did children with low levels of MS during a 2-year period. However, when a test group of children with more than 1 million CFUs of MS/mL of saliva received high-quality plaque control by frequent professional mechanical toothcleaning, they developed fewer new carious surfaces than did the control groups with high and low salivary MS levels (0.9 new carious surfaces versus 2.2 and 4.3 new carious surfaces, respectively). In Molndal, Sweden, Zickert et al (1982b) also attained a significant correlation between the prevalence of MS in saliva and the incidence of new carious lesions. During a 3-year period, children with high levels of salivary MS (> 106 CFUs/mL) developed about three times as many new carious lesions as did control groups with lower levels of MS. Subjects in test groups on a treatment program including chlorhexidine developed significantly fewer cavities. In US adolescents, Kingman et al (1988a) also showed that subjects with high salivary MS levels developed more new carious surfaces than did subjects with lower MS levels. In particular, controlled intraindividual longitudinal studies monitoring the microflora at the tooth surface level have clarified the cariogenic potential of MS (Axelsson et al, 1987b; Kristoffersson et al, 1985; MacPherson et al, 1990). An advantage of such studies is that several other external and internal modifying factors such as diet, fluoride, and saliva are equal for test and control sites. These studies have clearly shown that, in the same mouth, a tooth surface colonized by MS is at greater risk for caries than is a similar surface without MS. During a 30-month period, S mutans on all the approximal surfaces was studied in a selected group of 13 year olds with more than 1 million CFUs of MS/mL of saliva. From a population of 720 13 year olds, subjects with more than 1 million CFUs/mL were selected (n = 187). Every 6 months, S mutans was sampled from saliva, the dorsum of the tongue, and every approximal tooth surface. Interproximal samples were obtained with a sterile wooden toothpick, as described by Kristoffersson and Bratthall (1982) (Fig 25). The contaminated sides of the toothpick were then pressed directly against selective (mitis-salivarius-bacitracin) agar plates (Fig 26). After incubation, the number of colonies formed (CFUs) was evaluated for every approximal surface. In 17 subjects who consistently had a minimum of one surface highly colonized with MS and a minimum of one MS-negative or lightly colonized surface, about 60% of the highly colonized surfaces developed carious lesions (Fig 27, left), but only 3% of the MS negative or sparsely colonized surfaces did (Fig 27, right) (Axelsson et al, 1987b). A prior study showed the surfaces most heavily colonized with MS to be the approximal surfaces of the molars and the second premolars (Fig 28) (Kristoffersson et al, 1984). In fact, a previously mentioned study of more than 600 14 year olds (Axelsson, 1989, 1991) showed that the same surfaces also had the highest PFRI

scores. These observations explain why, in toothbrushing populations, these approximal surfaces have the highest prevalence of decayed, missing, or filled surfaces (Fig 29). For optimal caries prevention in such populations, plaque control and topical application of fluorides should target these key-risk surfaces. Methods of sampling As mentioned earlier, the correlation between salivary MS counts and the number of MS-colonized tooth surfaces is relatively good (Lindquist et al, 1989), and simple salivary sampling methods are a more convenient and realistic means of assessing the severity of MS infection than sampling from individual tooth surfaces. Laboratory methods. Saliva is collected, mixed with a proper transport medium, and forwarded to a microbiologic laboratory. After incubation using a selective medium, mutans colonies are counted and the results are expressed as the number of colonyforming units per milliliter of saliva. Several selective media are available, and their properties are not identical: This fact must be taken into consideration in assessing the results. A common type of selective medium for plating mutans streptococci is the mitis-salivarius-bacitracin agar (Gold et al, 1973). With the exception of the rare serotype a, all types of mutans streptococci grow on this medium. Bacitracin is the main selective ingredient. Because the plates have a shelf life of only about 1 week, they are not convenient for chairside tests. For screening surveys using agar plates, the following simplified method has been described, eliminating the need for transportation, dilution, and plating of saliva (Kohler and Bratthall, 1979; Newbrun et al, 1984) (Fig 30): Wooden spatulas are contaminated with saliva and immediately pressed against selective agar plates. After incubation, the number of colonies on a predetermined area of the plate is calculated. Chairside method. The so-called Strip Mutans test (Fig 31), described by Jensen and Bratthall (1989), is based on the ability of mutans streptococci to grow on hard surfaces and the use of a selective broth (high sucrose concentration in combination with bacitracin). Because the bacitracin can be added to the broth just before use, the shelf life of the test can be prolonged considerably compared to that of agar plates. The test set is used as follows: A bacitracin disc is taken from the vial with forceps or a needle. The cap is reclosed tightly. The bacitracin disc is put in the culture broth vial and allowed to stand for at least 15 minutes (Fig 32). The vial is shaken gently after 15 minutes. When more than one Strip Mutans test is to be run, the bacitracin discs can be added to the vials beforehand. However, only one Strip Mutans test can be performed in each vial, and vials to which bacitracin has been added must be used on the same day. The patient is given a Dentocult paraffin pellet and instructed to chew it for 1 minute. The stimulated saliva should be swallowed or spat out (Fig 33). A test strip is removed from the Strip Mutans container, so that only the square end is touched. About two-thirds of the strip is placed in the patient's mouth and rotated on the surface of the tongue about 10 times. The strip is removed from the mouth, pulled

between the patient's closed lips to remove any excess saliva (Fig 34). The strip is placed in the culture vial containing the well-mixed bacitracin-broth solution. The cap is reclosed tightly (Fig 35). The patient data is written on the label, and the label is attached to the vial. The strip is incubated for 2 days at 35°C to 37°C (95°F to 99°F). The strip is removed from the culture vial and allowed to air dry. The treated side, which is marked with a line, can be examined immediately or later (Fig 36). After it is dried, the test strip can be stored for future reference in a plastic bag, plastic wrap, autoclave plastic, or other similar material. The bacitracin and the higher sugar content inhibit the growth of practically all microorganisms, other than S mutans, that grow in mitis-salivarius medium. In proportion to their actual amount in saliva, S mutans in the specimen will adhere to the treated side of the strip, and grow as small, dark or light blue colonies, 1 mm in diameter, or considerably less, when growth is very dense. The amount of S mutans per milliliter of saliva is estimated by comparing the colony density on the strip with the standard charts included in the instructions. If the number of S mutans is very high, the treated side of the test strip will turn blue, and separate colonies will be indistinguishable. A test strip without S mutans may have a blue shade as a result of precipitation of the color indicator present in the culture medium. A magnifying glass or a microscope should be used to verify questionable cases. The Strip Mutans method should not be used within 12 hours of an antibacterial mouthwash (eg, chlorhexidine) or 2 weeks of a course of antibiotics to avoid falsenegative results. Fig 22 A study of 12-yearold Brazilian children shows a correlation between very high salivary MS levels and very high caries prevalence . (From Buischi et al, 1989.)

Fig 23 A study of 13- to 14-yearold schoolch ildren in Karlstad with good oral hygiene and low caries prevalen ce shows no correlati on between caries prevalen ce and different dietary scores and different levels of salivary S. mutan. The cutoff is S mutans negative or positive. (From Kristoffe rsson et al, 1986.)

Fig 24 Dental caries in relation to colonization of MS. (From Kohler et al, 1988.)

Fig 25 Interproximal samples of S mutans are obtained with a sterile wooden toothpick. Fig 26 Slides containing S mutans are pressed on agar plates.

Fig 27 Development of carious lesions on surfaces highly colonized (left) and sparsely colonized (right) with MS. (From Axelsson et al, 1987b.) Fig 28 The approximal surfaces of molars and second premolars have been shown to be the most highly colonized with MS according to approximal MS scores 03. (From Kristoffersson et al, 1984.)

Fig 29 Caries prevalence in 12 year olds in the county of Vormland, Sweden, 1964-1994. (From Axelsson, 1998.)

Fig 30 Wooden spatula contaminated with saliva and pressed on an agar substrate (left). Typical S. mutans colonies on the surface of the agar substrate (right). Fig 31 Strip Mutans test. (Courtesy D. Brathall.)

Fig 32 Placement of a bacitricin disc in the culture broth vial in which Strip Mutans test is being performed. (Courtesy D. Brathall.)

Fig 33 Dentocult paraffin being chewed to stimulate saliva. Chewing friction loosens S mutans from the tooth surface. (Courtesy D. Brathall.)

Fig 34 The test strip is removed from the patient's mouth and placed in the vial containing bacitracin solution. (Courtesy D. Brathall.) Fig 35 The test strip is removed from the patient's mouth and placed in the vial containing bacitracin solution. (Courtesy D. Brathall.) Fig 36 Examination of test strip. (Courtesy D. Brathall.)

Cariogenicity of lactobacilli According to the specific plaque hypothesis, some strains of lactobacilli are considered to be major caries pathogens along with S mutans and S sobrinus. Lactobacilli are acidogenic and even more aciduric than MS. Mutans streptococci are strongly correlated to the etiology of initial enamel and root surface lesions, because they can adhere to and colonize the tooth surfaces. Lactobacilli are more dependent on retentive sites for heavy colonization: Mutans streptococci are regarded as the pioneers, followed by lactobacilli in the succession toward more cariogenic plaque. This has been shown in a study on the development of so-called nursing caries (Milnes and Bowden, 1985) and by Mac Pherson et al (1990) in another study on plaque flora associated with early enamel demineralization. Lactobacilli are most often found in the deepest part of the lesion (dentin), an environment with prolonged periods of very low pH. Lactobacilli are highly influenced by the dietary carbohydrate content and intake frequency, in addition to reflecting an acidogenic environment by their very presence, because they are so aciduric. They also indicate the presence of substrate for other bacteria, such as mutans streptococci. Persistently high levels of lactobacilli after elimination of retention sites such as cavitated carious lesions indicate a diet rich in carbohydrates.

Evidence Lactobacillus counts have been used to predict the incidence of new carious lesions. Crossner (1981) studied a group of children, who had been given dental treatment at baseline so that no open lesions were present at the bacterial sampling. Two subgroups in this material are of special interest: those with very low or very high lactobacillus counts. Very few individuals in the low lactobacillus group developed new carious lesions over a 64-week period. In the high lactobacillus group, many, but not all, developed new lesions. Some studies have shown that caries incidence is significantly increased when MS and lactobacilli occur in the same individual: During a 3-year longitudinal study, Alaluusua et al (1990) showed that teenagers with high salivary values for both MS and lactobacilli developed several times more new carious surfaces than did those with lower counts of either. The percentages of children who developed 0, 1 to 3, and more than 3 new carious surfaces were correlated to the total scores of MS and lactobacilli (Fig 37). Similar findings were reported by Stecksen-Blicks et al (1985) (Fig 38). In a 2-year study by Crossner et al (1989), samples from saliva, the tongue, and 276 interdental spaces were obtained from 23 7 year olds to (1) relate the presence of lactobacilli at various oral sites to the occurrence of lactobacilli in saliva, and (2) relate the presence of mutans streptococci and various types of lactobacilli interdentally to the development of proximal carious lesions. The results showed that the number of interdental samples containing lactobacilli increased as the number of salivary lactobacilli increased. Furthermore, lactobacilli were never found interdentally without the presence of mutans streptococci, and lactobacilli proved to be the more suitable microorganism for prediction of proximal carious lesions. Neither the number nor the differentiation into different species of interdental lactobacilli seemed to be of importance, merely their presence or absence. The presence of lactobacilli probably reflects a caries-inducing environment (etiologic microflora plus fermentable carbohydrates), thus explaining their high predictive ability despite their rather limited etiologic importance in the initiation of caries. Lactobacilli are also implicated in root surface caries. Fure and Zickert (1990a) studied a group of 208 randomly selected 55, 65, and 75 year olds. To estimate the number of root caries lesions, an index, DFS%(R), was used, indicating the number of decayed and filled root surfaces as a percentage of the exposed root surfaces. The mean DFS%(R) was 13 for the 100 subjects with low lactobacillus counts (less than104), and 23 for the 52 subjects with high counts (more than 105). In crosssectional and longitudinal studies, it has been demonstrated that lactobacillus counts are among the factors with a significant correlation to the development of root caries lesions (Ravald and Birkhed, 1991, 1994; van Houte et al, 1990). Ellen et al (1985) also showed that root surfaces harboring both MS and lactobacilli are most likely to develop root caries. A number of studies have tried to clarify the prevalence of lactobacilli in various populations, eg, two Swedish studies, one comprising 646 9 to 12 year olds (Klock and Krasse, 1977) and the other a group of 101 13 to 14 year olds (Zickert et al, 1982b). About 50% showed low salivary counts (less than 10,000 CFUs/mL of saliva)

and 10% to 20% had high counts (100,000 to 1 million CFUs/mL of saliva). Generally the MS counts were 10 times greater than the lactobacillus counts. Most children with low mutans streptococci levels also had low numbers of lactobacilli, but there were situations where one type of bacterium was high and the other low. In a dentate Swedish population of 80 and 85 year olds, Kohler and Persson (1991) found that 95% of the subjects had detectable salivary lactobacilli counts and 35% very high levels (more than 100,000 CFUs/mL of saliva). Almost 90% were MS positive, and 30% had more than 1 million CFUs/mL of saliva. Methods of sampling Laboratory method. The standard method for determining the number of lactobacilli includes the use of a selective medium, Rogosa selective Lactobacillus (SL) agar. Saliva obtained by chewing a piece of paraffin is shaken with glass beads to break up aggregates of bacteria. The saliva is then mixed with a buffer solution, and 1 mL of the dilutions 10-2 and 10-3 is mixed with 10 mL of melted SL agar. Another 10 mL is then poured into the Petri dish, and the plates are incubated at 37°C for 2 days. The plates are then ready for a colony count. Chairside method. A chairside dip-slide method (Dentocult LB) is also available for simplified evaluation of salivary lactobacilli counts (Fig 39). After aerobic incubation for 4 days at 37°C, the number of lactobacilli is estimated by comparing the slides with a chart supplied by the manufacturer. An advantage of this method is that the results can be shown directly to the patient. Figure 40 shows examples of low (less than 10,000 CFUs/mL of saliva) and high salivary lactobacillus counts. The high levels indicate a cariogenic enviroment (low pH) in the oral cavity. The number of lactobacilli in saliva seems to be fairly stable during the daytime. Significantly higher levels are often obtained if saliva is collected in the early morning, before breakfast and toothbrushing, than if samples are obtained during the rest of the day, especially for subjects with high numbers of lactobacilli. Fig 37 Distribution of caries increment in children with different combined levels of mutans streptococci and lactobacilli scores. Score systems based on Dentocult SM for mutans streptococci and Dentocult LB for lactobacilli, both systems ranging from score 0 (low levels of bacteria) to score 3. (From Alaluusua et al, 1990. Reprinted with permission.) Fig 38 Lactobacilli and S mutans among 13-year-old children with different net caries increment. (From Steckse′ n-Blicks, 1985. Reprinted with permission.)

Fig 39 Dip-slide used for simple evaluation of salivary lactobacilli counts.

Fig 40 Examples of low (left) and high (right) salivary lactobaccilli counts using the Dentocult method.

Cariogenicity of other bacteria There are overwhelming data from experimental and clinical studies in humans showing that S mutans and S sobrinus and lactobacilli are strongly correlated to caries etiology. However, the use of selective substrates in most of these studies may have introduced some bias. For example, Sansone et al (1993) found that plaque samples with and without MS and lactobacilli were equally acidogenic when cultured at low pH and in the presence of excess glucose. Borgstrom et al (1997) evaluated the pHlowering effect of plaque from carious lesions in enamel as a virulent variable and found a relatively weak correlation among lactobacilli, MS, and dental caries. Root caries is considered to develop at higher pH than enamel caries. Normally, demineralization precedes the breakdown of the organic part (about 40% by volume) of the root surface. Schupbach et al (1995), using a new sampling technique and unselective, anaerobic culturing methods, evaluated the composition of the plaque microbiota covering sound root surfaces, actively carious root surfaces, or arrested lesions. On all surfaces Actinomyces spp predominated, and streptococci and lactobacilli formed a minor part (less than 1%) of the microbiota. With respect to the detected proportions of anaerobes, microaerophiles, Actinomyces naeslundii, Prevotella buccae, and Selenomonas dianae, significant differences were observed among the three categories of root surfaces. The total numbers of CFUs were significantly higher on both caries-free and caries-active surfaces than on arrested lesions. In general, the results supported a polymicrobial etiology for caries initiation on root surfaces, during which A naeslundii, Capnocytophaga spp, and Prevotella spp make specific contributions to the processes of cementum and dentin breakdown. In other words, the roles of bacteria-producing proteolytic enzymes and of acidogenic bacteria other than MS and lactobacilli in the development of root surface caries warrant further investigation (for reviews on specific microflora related to the etiology of dental caries, see Bowden, 1997; Bowden and Edwardsson, 1994; Bratthall and Ericsson, 1994; Emilsson and Krasse, 1985; Loesche, 1986). Prediction of Caries Risk Principles of risk prediction Some basic principles have to be followed for successful and cost-effective caries prediction, caries prevention, and caries control: 1. The higher the risk of developing caries for most of the population, the more significant the effects of one single preventive measure and the stronger the correlations between one single etiologic or modifying risk factor and the risk for caries development.

2. In populations in which only a minority of the people will develop new carious lesions, it is necessary to use accurate risk predictive measures to select at-risk individuals and introduce needs-related combinations of caries-preventive measures, in other words, a "high-risk strategy." In most child populations today, caries incidence is skewed. The majority of the children in most age groups develop no or very few new carious surfaces, while a small minority, 5% to 10%, develop several new carious surfaces each year. Therefore, accurate caries risk predictors are useful. However, the dilemma is that no method will guarantee that 100% of the selected high-risk individuals are "true" highrisk individuals. The percentage of "true" high-risk individuals among the selected group of high-risk individuals is termed sensitivity of the risk predictive method. Similarly, methods are used to select nonrisk or low-risk individuals. The percentage of "true" nonrisk individuals is termed specificity of the predictive method used. These principles may be exemplified in Fig 41 (a to c) showing the outline of a typical study with the aim of evaluating the predictive power of a risk marker of dental caries. In the beginning, the baseline caries status and the level of the selected risk marker are assessed. Caries recordings at the end of the follow-up period make it possible to assess the true caries incidence during the period. Prediction studies deal with two dichotomies: (1) individuals for whom it is believed that the risk is high or low, and (2) the ones for whom true high or low caries incidence is observed. Thus group a in Fig 41 (a) consists of correctly classified individuals, true-positives, for whom it was believed that the risk was high and whose actual caries incidence was high. Correspondingly, group d represents correctly classified true-negatives. For individuals falling into groups b and c, misclassification has occurred. For the false positives, in group "b," a high risk was assumed, but the true caries incidence was low. Correspondingly the false-negatives in group c were believed to have a low risk, but their actual caries incidence was high. This design is only usable for one predictor at a time. In practice, several predictors are often regarded in prediction studies. In the case of multiple predictors, each of them can be considered separately, which leads to predictor-specific numbers of trueor false-positives and true- or false-negatives. Alternatively, the information of many predictors can be condensed into a single variable on the level of which the prediction of high or low risk is based. The techniques for such condensing range from simple summaries to sophisticated regression-based multivariable models. Even in the case of one predictor, the risk markers are seldom natural dichotomies. To generate the four groups of interesttrue- and false-positives, and true- and falsenegativesit is necessary to artificially dichotomize both multi- ple-level predictors and the outcome, true caries incidence, which, in the data collection, is usually regarded as the number of new carious tooth surfaces, not as a dichotomy. The dichotomization can be done in different ways. Each threshold level for believed high risk and for observed high true caries incidence leads to a different distribution of the study subjects into the four groups of interest (true- and false-positives, and true- and false-negatives). Thus, when the results of a prediction study are evaluated, it is of utmost importance to consider the threshold levels that have been used. To

estimate the accuracy of the classification of the four groups in Fig 41 (a), the quantities a, b, c, and d are organized in the form of a 2 x 2 contingency table (Fig 41 (b)). Six different measures of accuracy and their estimators are given in Fig 41 (c). As mentioned earlier, sensitivity is the proportion of those who were believed to have a high risk among the individuals whose actual caries incidence during the follow-up was high. Specificity is the proportion of those who were believed to have a low risk among the patients whose actual caries incidence during the follow-up was low. False-positive rate and false-negative rate carry exactly the same information as sensitivity and specificity but, in contrast, reveal proportions of misclassified subjects. False-positive rate is the proportion of those who were believed to have a high risk among the subjects whose actual caries incidence during the follow-up was low. False-negative rate is the proportion of subjects who were believed to have a low risk among those individuals whose actual caries incidence was high. Positive predictive value is the proportion of those whose actual caries incidence was high among the subjects who were believed to have a high risk. Negative predictive value is the proportion of subjects whose actual caries incidence was low among the patients for whom a low risk was predicted. All six of these measures should always be examined pairwise. For instance, sensitivity has no meaning if the specificity is not known. This is important from a cost-effectiveness point of view during screening of populations of children with relatively low caries prevalence and incidence before needs-related caries preventive programs are designed. However, for the individual patient the consequences of being a false nonrisk or low-risk individual are very different from those of being selected as a false-positive high-risk individual. Fig 41 (a and b) A study of the evaluation of the predictive power of a risk marker of dental caries. (c) Selected measures for evaluating the accuracy of predictions. N = study cohort; BHR = believed to have a high risk; BLR = believed to have a low risk; TCI = true caries incidence; Group a = believed to have a high caries risk, actual high caries incidence (true-positive); Group b = believed to have high caries risk; actual low caries incidence (falsepositives); Group c = believed to have low caries risk; actual high caries incidence (false-negatives); Group d = believed to have low caries risk; actual low caries incidence (true-negatives). (Modified from Hausen, 1997. Reprinted with permission.) Accuracy of risk assessments in practice A perfect risk marker would have a sensitivity of 100% and a specificity of 100%, implying no errors in risk assessment. Consequently, the false-positive and falsenegative rates would be 0%, and positive and negative predictive values would be 100%. Having perfect accuracy means that the predicted high-risk group would consist of only true high-risk individuals and that only true low-risk individuals would be included in the predicted low-risk group. Unfortunately, no such marker is available for the assessment of caries risk. A certain proportion of errors have to be

accepted. However, there are no generally accepted rules of what the acceptable level of error might be. It has been suggested that, in a risk model, the sum of sensitivity and specificity be at least 160% before a caries risk marker can be considered a legitimate candidate for targeting individualized prevention (Kingman, 1990). This is in agreement with an alternative suggestion that a sensitivity and specificity of 80% would be acceptable for practical use in the community. Although neither of these suggestions takes into account the fact that errors related to poor sensitivity do have consequences that are different from those related to poor specificity, both proposals can be used as a starting point for evaluating the performance of proposed markers for high caries risk. What would a combined sensitivity and specificity of 160% mean in practice? If both the sensitivity and specificity were 80%, every fifth individual with a true high risk would remain undetected in a risk assessment and thus fail to receive the intensified protection against caries that he or she needs. Correspondingly, every fifth individual with a true low risk would erroneously be included in the high-risk group and receive preventive measures to no or little purpose. Thus, even the proposed minimum acceptable level of accuracy would result in an uninvitingly high rate of misclassifications. If the proportion of caries-risk individuals in a population is close to half or more, this clearly implies that the occurrence of caries is not low enough to justify the effort and expense of identifying key-risk individuals. In such a situation, the preventive efforts should be targeted to the whole population. The proportion of the target population that can be given individual protection against further caries development naturally varies from one setting to another. In most cases, risk groups of a size exceeding 30% seem to be unworkable. In a thorough review by Hausen et al (1994), an effort was made to compare the predictive power of risk markers in a situation where the aim was to select the 30% of the target population with the highest risk of developing new lesions. For none of the markers aimed at assessing the risk for coronal caries did the predictive power reach the proposed combined sensitivity and specificity of 160% (Kingman, 1990). This level was surpassed in one study only, where a combination of several predictors had been used for assessing the risk of root caries (Scheinin et al, 1992). The difficulty of predicting caries is not unexpected. The multifactorial etiological and modifying factors of dental caries make it likely that even the most sophisticated risk models will be of limited value in predicting future caries development very accurately. Furthermore, even a perfect marker is capable of predicting a person 's future caries experience only if the conditions on which the prediction is based remain stable. In most industrialized countries, where virtually all the prediction studies have been conducted, the populations are exposed to a variety of professional prevention and treatment regimens as well as self-care, which, if applied selectively, most probably reduce the observed power of such studies. The living conditions and oral health behaviors may change over time, thus modifying a person's caries risk in either direction. In addition, the rational and ethical consequence of risk prediction in clinical practice is to introduce needs-related measures for caries prevention and caries control. The optimal outcome therefore should be no new carious lesions. For

these reasons, it is not likely that, even in the future, caries risk can be assessed accurately by using one single risk marker. Past caries experience (caries prevalencethe number of decayed, missing, or filled teeth and surfacesand caries incidencethe number of new carious teeth and surfaces in a year) has so far been the most powerful single predictor for future caries incidence, at least in children and young adults. That is because carious lesions represent the sum result of all the etiologic and modifying risk factors to which the individual has been exposed. For example, in a recent 3-year longitudinal study, Bjarnason and Kohler (1997) achieved 89% sensitivity value in a group of Swedish adolescents by comparing the prevalence of non-cavitated enamel caries and DFS at the baseline as predictors. Together, the sensitivity and specificity values reached 160% or more. High salivary MS and lactobacilli scores resulted in 71% sensitivity and 75% specificity, respectively (cutoff level for high caries risk was 5 or more new carious surfaces in 3 years). However, only baseline values of incipient enamel caries were significantly correlated to the caries incidence. The use of past caries experience as an indicator of future incidence has justly been criticized by the argument that the aim should be to determine the high-risk individuals before there are any signs of past caries experience. In other words, efforts should focus on primary prevention instead of secondary prevention. In particular this is important in infants and children with erupting permanent teeth. Wendt et al (1994) found that the caries incidence in infants and toddlers aged 1 to 3 years was strongly correlated to the plaque scores and oral hygiene regimens even at 1 year of age. Selection of caries-risk patients Inability of a sole salivary MS test to predict caries risk As already mentioned in this chapter, numerous cross-sectional as well as longitudinal studies have shown significant correlations between salivary MS levels and caries prevalence and caries incidence (for review, see Bratthall, 1991; Bratthall and Ericsson, 1994; Beighton et al, 1989). At the surface level, even more significant correlations between MS colonization and caries incidence have been found (Axelsson et al, 1987b; Kristoffersson et al, 1985). Most of the early salivary MS studies were carried out in child populations with relatively high caries prevalence (Sweden in the 1970s), and at that time more than 1 million CFUs of MS/mL of saliva was shown to be a good predictor of caries risk (Klock and Krasse, 1977; Zickert et al, 1982). However, since then, caries prevalence in Sweden and many other industrialized countries has decreased significantly. The correlation between one single etiologic factor, such as salivary MS levels, and caries prevalence and caries incidence tends to be weaker in such populations, because dental caries is a multifactorial disease. In a more recent 2-year longitudinal Swedish study in children (5 to 7 years and 12 to 14 years), Sullivan et al (1989) found that the correlation between caries incidence and both salivary MS and lactobacilli was weak at the individual level, particularly

after correction for confounding factors, such as oral hygiene status. In another study, Sullivan et al (1996) found that MS and lactobacilli, whether in saliva or in plaque, was not a powerful enough tool for caries prediction in a group of 14 to to 15 year olds. Kingman et al (1988a) also found that the predictive values for salivary MS and lactobacilli on caries incidence in 10- to 15-year-old US schoolchildren was low (31% and 39%, respectively). The moderate-to-low predictive value of salivary MS may partly be explained by differences in virulence, not only among species of MS but also among individual clones of S mutans and S sobrinus (Bowden, 1997). Even in relation to root caries, the important role of MS and lactobacilli has recently been questioned (Schupbach et al, 1995; for recent reviews on the importance of the specific microflora for prediction of caries risk, see Bowden, 1997; Bowden and Edwardsson, 1994; Bratthall and Ericsson, 1994; Hausen et al, 1994; Hausen, 1997. More recent cross-sectional studies in Swedish schoolchildren have repeatedly found that the cutoff for correlation between salivary MS counts and caries prevalence is MS negative or MS positive rather than > 1 million CFUs of MS/mL of saliva (see Fig 23) (Kristoffersson et al, 1986). However, the dilemma is that only 10% to 30% of the individuals are MS negative in most populations (higher in young children and lower in elderly). The question is how to select 5% to 25% high- and very high-caries risk individuals from among the 70% to 90% MS-positive subjects. Rationale for combining salivary MS tests and PFRI for prediction of caries risk Like the inflammation induced in gingival soft tissues adjacent to dental plaque, carious lesions that develop on the individual enamel surface beneath bacterial plaque should be regarded as the net result of an extraordinarily complex interplay between harmless and harmful bacteria, antagonistic and synergistic bacterial species, their metabolic products, and their interaction with the many other external (fermentable carbohydrates etc) and internal (saliva and other host factors) modifying factors, which are discussed in more detail in chapters 2 and 3. In other words, enamel carious lesions develop only on the specific tooth surfaces where thick plaque with a high percentage of acidogenic and aciduric bacteria remains too longits "acid slag products" demineralize the underlying tooth surface. As discussed earlier in the chapter, the quantity of plaque that forms on clean tooth surfaces during a given time represents the net result of interactions among etiologic factors, many internal and external risk indicators and risk factors, and protective factors. This observation was the rationale for the development of the Plaque Formation Rate Index by Axelsson (1984, 1989, 1991). The index, based on the amount of plaque freely accumulated (de novo) 24 hours after PMTC, is described in more detail earlier in the chapter. An earlier 30-month longitudinal study showed a very strong correlation between development of approximal carious lesions and the level of MS colonization (Axelsson et al, 1987). Other studies have shown that salivary MS counts are correlated to the number of tooth surfaces that are colonized by MS (Lindquist et al, 1989; for review, see Bowden, 1997; Bowden and Edwardsson, 1994; Bratthall and Ericsson, 1994). Therefore, it seems reasonable that MS-positive individuals with high and very high PFRI scores (4 and 5, respectively) should be more caries susceptible than MS-negative individuals or MS-positive individuals with very low or

low PFRI scores (1 and 2, respectively). That is because the total number of the most cariogenic bacteria (MS) should be significantly higher on tooth surfaces in subjects with a PFRI score of 4 or 5 than in subjects with a PFRI score of 1 or 2, if the percentage of MS in their plaque is the same. Prediction study. In 1984, a large-scale, combined cross-sectional and longitudinal study was initiated with the following objectives: 1. To determine the distribution of the PFRI in a large number of schoolchildren and the distribution characteristics of plaque formation on the individual tooth surfaces in the dentition. 2. To determine whether there is a correlation among the salivary S mutans level, a Cariostat test, and the PFRI score, separately or in combination, and the prevalence of smooth-surface caries. 3. To determine whether a combination of the SM level and the PFRI score is more closely related to caries prevalence than are the variables individually. 4. To determine whether there is any association between the SM level and the PFRI score. 5. To determine the influence of individual factors on the PFRI score (these data were not available at the time of writing and are not included in this chapter). 6. To determine whether caries development can be predicted by a combination of salivary S mutans levels and the PFRI. All 716 14 year olds in Karlstad, Sweden, were recruited to participate in the study. Each was given two dental appointments, precisely 24 hours apart. The first appointment comprised the following procedures (for details on methods and materials, see Axelsson, 1989, 1991): 1. Measurement of salivary secretion rate. 2. Salivary S mutans test using the spatula method described by Kohler and Bratthall (1979). 3. Cariostat test (based on a sample of approximal plaque). The acidogenic capacity of the sample was estimated from the colorimetric indicator in the test tube showing different pH values. 4. Gingival index. 5. Plaque index based on disclosed plaque (O'Leary et al, 1972). 6. Professional mechanical tooth cleaning. The subject was instructed to refrain from all oral hygiene until the appointment scheduled for the following day. 7. Examination of all smooth surfaces for caries, with the following diagnosis: sound

surface, enamel caries, dentin caries, or filled surface. This was recorded as the subject's mean decayed or filled surface (DFS) score. On day 2, 24 hours later, the appointment began with plaque disclosure. The presence of adherent plaque mesiobuccally, buccally, distobuccally, mesiolingually, lingually, and distolingually was noted for each tooth. The percentage of tooth surfaces with plaque was calculated according to the following formula: Total number of surfaces with plaque x 100 Number of teeth x 6 Each subject's PFRI was then scored according to the scale described earlier in the chapter. At the same visit, a thorough 24-hour dietary history was recorded for subsequent evaluation of the influence of dietary factors. Many other indicators and factors possibly related to the PFRI were also evaluated, including gingival inflammation, salivary levels of glucosyl transferase, agglutinin levels in resting saliva, and oral hygiene, dietary, and fluoride habits. Of the 716 children aged 14 years, 667 participated in the PFRI study. Figure 12 presents the frequency distribution of PFRI scores in the population. Figure 17 presents the plaque distribution on various surfaces. Six hundred fifty-four children, 333 boys and 321 girls, formed the population for further analysis; for each of these, complete examinations were available. The other 62 children among the 716 originally selected for the study were excluded from the statistical analysis mainly because of incomplete examinations, antibiotic treatment, orthodontic bands, refusal to participate, or illness. Results of prediction study. The examination for caries showed that 70% of the children had no dentin caries or restorations on smooth surfaces. Of the total number of lesions on these surfaces, enamel caries constituted more than 80%. Figure 42 shows the mean number of approximal carious lesions per individual in the extreme groups in relation to the PFRI score, salivary S mutans level, and Cariostat test. The group with a PFRI score of 5 had, on average, twice as many carious or restored surfaces as did the group with score 1. The difference between S mutans counts of fewer than 100,000 CFUs/mL of saliva and more than 1 million CFUs/mL of saliva was much less marked and of the same order as the differences between Cariostat blue (low acidity) and greenish yellow (high acidity). An analysis of caries prevalence (mean DFS) related to different PFRI scores indicated a thres- hold for caries risk between PFRI scores 2 and 3 (see fig 13), and this was subsequently confirmed in the longitudinal part of the study, over 5 years (Axelsson 1989, 1991). For S mutans, this critical level was between 0 and 100,000 CFUs/mL. The Cariostat test offered no additional diagnostic advantages over S mutans counts. Table 1 presents the mean values for caries prevalence per individual in relation to different scores of salivary SM and PFRI. The mean values were low in the

approximately 20% of subjects with 0 SM, irrespective of the PFRI score. On the other hand, individuals among the 80% of the SM-positive subjects with a PFRI score ≥ 3 showed markedly higher caries prevalence than did others. The level of salivary S mutans appeared to lack significance in this context. These results were are also confirmed by the lack of correlations between salivary S mutans levels and PFRI scores (Table 2). In the cross-sectional part of the study, a score of 0.0 DFS was regarded as a truenegative value for low caries risk and > 10.0 DFS was chosen as a true-positive cutoff value for high caries risk because the mean value for the whole group was only 3.5 DFS. The results in Table 2 indicate that the SM level and PFRI score were independent, ie, there was no association between the variables. The contingency coefficient was low: 0.16 (upper limit = 0.87). From the data in Table 3, it can be calculated that a combination of the PFRI and S mutans gave values of 92.1, 60.9, and 67.3%, for sensitivity, specificity, and diagnostic power, respectively; For these extremely low- or high-risk groups, the values obtained were better than were those for sensitivity, specificity, and diagnostic power for any of the variables alone (PFRI, Cariostat, SM > 0, and SM > 106 CFUs/mL). Five years later, the children were reexamined. Table 4 shows the mean caries incidence on the approximal surfaces in the predicted nonrisk (SM-negative) group and the predicted risk group (SM-positive subjects with PFRI scores 3 to 5). The risk group developed five times more new approximal carious lesions in dentin per individual per 5 years than did the no-risk group, in spite of ongoing preventive programs. The question remains, without answer, how big the difference would have been without any preventive program. The 14-year-old age group was selected as particularly appropriate for this investigation. At this age, smooth-surface cariesparticularly approximal lesionswould have developed within the previous 2 years; that is, the caries prevalence on these surfaces would largely correspond to the caries incidence over this period, and the subjects would still have a suboptimal number of intact approximal surfaces at risk. The occlusal surfaces of the molar teeth were deliberately excluded from the investigation because indications for restoration of these surfaces vary widely. Recommendations derived from prediction study. For caries prevention, S mutanspositive subjects with a PFRI score > 3 should be encouraged to clean their teeth more frequently than other 14 year olds. These subjects should probably clean their teeth twice as often as others, that is, morning and evening with an efficient use of fluoride toothpaste. For patients at extreme risk, cleaning immediately before meals and the use of fluoride chewing gum as a "dessert" after every meal may be recommended. In lingual plaque up to 12 hours old, critically low pH values do not occur after rinsing with a 10% sucrose solution. However, in approximal 3-day-old plaque, there is a potentially dangerous drop in pH. Rinsing with a sucrose solution does not cause a critical drop in pH approximally if these surfaces have been cleaned just prior to rinsing (Imfeld, 1978). Wright et al (1979) demonstrated, in a split-mouth study, that

approximal cleaning once a day gave a reduction of slightly more than 50% in approximal caries. Artificial cleaning of the palatal surfaces is, for caries control purposes, unnecessary. The extremely low plaque formation is probably attributable to the constant friction of the rough surface of the dorsum of the tongue on these surfaces. Needs-related tests of patients' oral hygiene have clearly shown that special efforts should be concentrated on the linguoapproximal surfaces of the mandibular molars and premolars and the buccoapproximal surfaces of the corresponding maxillary teeth. Experience has also shown that the recall visit on day 2 is the ideal occasion for successful introduction of such individual needs-related oral hygiene practices. Traditionally, a salivary S mutans count of > 1 million CFUs/mL has been regarded as a critical value in assessing caries risk. The results of the previously described investigation do not support this concept. Rather, the critical limit for salivary S mutans is 0 CFUs/mL. Similar findings were reported in another study of Karlstad schoolchildren of the same age (Kristoffersson et al, 1986). However, when the same material was analyzed using approximal tooth surfaces as the unit, a very clear association emerged between the different levels of S mutans colonization and caries risk (Axelsson et al, 1987b). Table 1 could serve as a guideline for selecting nonrisk and risk individuals. A salivary S mutans test screens out SM-negative subjects (about 25%) as not being at risk. Of the remaining 75% or so (SM-positive subjects), those with a PFRI > score 3 are selected as risk patients (approximately 20%). From these subjects, an extremely high-risk group may be further selected: those with a PFRI score of 4 or 5 and an SM score of 2 or 3 (around 5%). Such a guideline is illustrated in Fig 43. In general, if the aim of screening is to direct intensive preventive treatment toward high-risk subjects, a screening procedure offering high sensitivity and predictive value is preferable both for individual patients and for community dental health planning. A false-negative diagnosis would deny a subject at risk the benefit of additional preventive measures. The occurrence of many false-positive diagnoses would make unnecessary demands on community dental health resources. In this study, Cariostat tests and high salivary S mutans counts were less reliable as predictors than were salivary S mutans-positive status and high PFRI (scores 3 to 5). Evaluation of the influence of several individual risk factors on the PFRI is in progress. It is anticipated that the three or four main factors will be identified. Identification of which of these factors dominates in patients with a PFRI score > 3, should make it possible to design an individual preventive program in which, where feasible, preventive measures are specifically directed toward minimizing the influence of these factors. Individuals with a PFRI score of 1 or 2 were stable over 5 years, while scores in individuals with a PFRI score of 3 to 5 sometimes changed over time. This observation indicates that plaque formation rates in individuals with a PFRI score of 4 or 5 can be reduced: Such individuals should be thoroughly evaluated to identify the factors contributing to their rapid plaque formation. Needs-related preventive measures could then be introduced.

High-caries rate prediction study. This study was followed up by a 3-year longitudinal study in Polish children (Axelsson et al, 2000a). Selected 12-year-old schoolchildren in Warsaw were randomly assigned to a test or a control group. At the baseline examination, caries prevalence (DFSs), salivary SM counts (Strip-SM), the PFRI, the Plaque Index (O'Leary, 1967), etc, were recorded. Figures 44, 45, and 46 show the frequency distribution of DFSs, PFRI scores, and Strip-SM scores, respectively, among all the children at the baseline examination. Based on the baseline examination, the subjects were assigned to low-risk, risk, and high-risk groups, according to the following criteria: 1. Low-risk groups: Streptococcus mutans-negative individuals with a PFRI score of 1 or 2 (test: n = 47; control: n = 43). 2. Risk groups: Streptococcus mutans-positive individuals with a PFRI score of 3 (test: n = 30; control: n = 32). 3. High-risk groups: Streptococcus mutans-positive individuals with a PFRI score of 4 or 5 (test: n = 14; control: n = 13). During the following 3 years, the children in the test group participated in a preventive program based on professional mechanical toothcleaning at needs-related intervals but with very simplified methods. For ethical reasons, the children in the control group were maintained in the regular, simple, school-based preventive program, based on oral hygiene instructions and topical fluoride administration. After 3 years, the children in the low-risk test group had developed significantly fewer new DFSs per individual per 3 years than had the children in the risk test group, who also had developed significantly fewer new DFSs than both high-risk groups. However, all the test groups had developed fewer new DFSs than had the control groups (Fig 47). This study showed that future caries development can be predicted, even in populations with very high caries incidence, by a combination of salivary S mutans counts and the PFRI (Axelsson et al, 2000a). Fig 42 Correlation of PFRI, S mutans, plaque pH, and caries prevalence interproximally. (From Axelsson, 1989.) Fig 43 Four-point scale for prediction of caries risk based on S mutans and PFRI. (From Axelsson, 1991.)

Fig 44 Caries prevalence in 12-yearold Polish schoolchildr en: Frequency distribution of DFSs. (From Axelsson et al, 2000a.)

Fig 45 Plaque formation rate index in Polish schoolchildr en. (From Axelsson et al, 2000a.)

Fig 46 Salivary strip S mutans in Polish schoolchildren. (From Axelsson et al, 2000a.) Fig 47 Effect of a needs-related preventive program for dental caries in Polish schoolchildren: Results after 3 years. (From Axelsson et al, 2000a.) Conclusions Etiology of caries The clinical carious lesion that develops on the tooth surface beneath undisturbed

bacterial plaque represents the net result of an extraordinarily complex interplay among harmless and harmful bacteria, antagonistic and synergistic bacterial species, their metabolic products, and their interaction with the many salivary and other host factors. In other words, dental caries not only is a multifactorial disease but also has a complicated etiology. It is more difficult to demonstrate a correlation between one single species of cariogenic bacteria and future caries development in populations with low caries prevalence than it is in populations with high caries prevalence. Three different theories for the etiology of dental caries have been proposed: the nonspecific plaque hypothesis, the ecological plaque hypothesis, and the specific plaque hypothesis. However, the true etiology is none of these, but a complex combination of all three processes. The criteria for cariogenic bacteria is that they must be acidogenic; that is, organic acids are formed as waste products of the fermentation of carbohydrates. In addition, the bacteria must be aciduric to survive in the resultant acidic environment (low pH) in the plaque and the carious lesion. Even bacteria-producing enzymes, which destroy the organic components of root cementum and dentin, may be involved in the development of root caries and dentin caries. The basic principle of the nonspecific plaque hypothesis is that thick plaque on the tooth surface, if left undisturbed for long periods, allows the total amount of acid produced within this plaque to initiate the development of a carious lesion. Accordingly, very high plaque formers (PFRI scores 4 and 5) would be expected to develop more carious lesions than low or very low plaque formers (PFRI scores 1 and 2), if the standards of oral hygiene and the composition of the microflora were the same in the two groups. In addition, carious lesions tend to develop on the particular tooth surfaces on which most plaque reaccumulates between toothcleaning procedures (mesiolingual and distolingual surfaces of the mandibular molars and mesiobuccal and distobuccal surfaces of the maxillary molars), and, in a toothbrushing population, where the toothbrush has limited access (the approximal surfaces of the molars and premolars). This is confirmed in studies on the pattern of caries prevalence in different populations. Not only the frequency but also the main target of needs-related oral hygiene procedures should be based on the score and the pattern of the PFRI. Because the quantity of plaque that forms on clean tooth surfaces during a given time represents the net result of interactions among etiologic factors, many internal and external risk factors, and protective factors, future research should be directed toward methods of identifying the major factors causing rapid plaque formation in the individual patient. If possible, these factors should be reduced or eliminated. The ecological plaque hypothesis is based on the principle that a change in a key environmental factor (or factors) will trigger a shift in the balance of the resident plaque microflora and this might predispose a site to disease. For example, the thicker the plaque, the less accessibility there is for the saliva to dilute and buffer the organic acids formed by fermentation of carbohydrates by acidogenic plaque bacteria. As a consequence, the pH will continuously decrease the more fermentable carbohydrates (for example, sucrose) are supplemented and the longer the plaque remains undisturbed. The lowered pH (< 5) in the plaque will promote a shift of the

composition of the plaque bacteria toward an increased number and assortment of acidogenic and aciduric species such as the cariogenic mutans streptococci and lactobacilli. According to this hypothesis, the strategy for caries prevention should be to maintain a high pH on all tooth surfaces (microenvironments) by frequent removal of plaque, thereby limiting the thickness of undisturbed plaque; and by reduction of the "sugar clearance time," through a diet that stimulates saliva, and use of supplementary fluoride chewing gum as a dessert directly after every meal, particularly in patients with reduced salivary flow. Future research should focus on efficient methods for achieving homeostasis of dental plaque, maintaining high pH (> 6). There is abundant support for the so-called specific plaque hypothesis, which proposes that some specific species of the plaque flora should be regarded as major pathogens in the etiology of dental caries. The most significant of these bacteria are some of the mutans streptococci. This group includes seven species, although two, S mutans and S sobrinus, are most closely associated with dental caries in humans. In longitudinal studies, particularly at surface level, a very strong correlation has been shown between S mutans and development of caries lesions on smooth surfaces. However, in populations with low caries prevalence, a correlation between various levels of salivary S mutans counts and caries incidence seems to be less significant: The threshold would appear to be S mutans-negative or S mutans-positive status. The second genus closely associated with caries etiology is Lactobacillus, commonly isolated from the dentin in both coronal and root caries lesions, its main habitat in the mouth. Compared to Streptococcus, Lactobacillus has been less extensively studied. Actinomyces odontologica, Actinomyces naeslundii, and other species of the MS group are also associated with the etiology of dental caries but are considered to be less cariogenic than S mutans, S sobrinius, and Lactobacillus. Future research involving DNA probes and so-called genetic-fingerprinting techniques will result in tools for evaluation of (1) how S mutans is transmitted among individuals; (2) how stable the oral population is; (3) how many types of S mutans an individual carries; (4) if particular clonal lines of S mutans are more virulent (cariogenic) than others; and (5) if individuals with higher levels of carious activity carry particular types of S mutans. Prediction and prevention of caries The younger the population and the lower the caries prevalence in the population, the higher the percentage of caries-free subjects. In these populations, it is necessary to focus on "high-risk strategy" and primary prevention, rather than secondary prevention. For practicing primary prevention according to the high-risk strategy, the etiologic factors used for caries prediction must be as sensitive as possible, that is, optimizing the percentage of true high-risk individuals for cost effectiveness. Because dental caries is a multifactorial disease with a complicated etiology, it is necessary to combine as many etiologic factors as possible to predict caries risk in children with

low caries prevalence, which is the situation among most children in the world. In this approach high and very high plaque formers (PFRI scores 4 and 5, respectively) with a high percentage of cariogenic bacteria such as S mutans would be expected to develop significantly more new carious surfaces than would those with a very low or low plaque formation rate (PFRI scores 1 and 2, respectively) and little or no S mutans in the plaque. There is a correlation between salivary S mutans counts and the number of tooth surfaces colonized with S mutans. Therefore, the combination of salivary S mutans counts and Plaque Formation Rate Index (PFRI scores 1 to 5) is recommended for caries risk prediction, according to the following scale: 1. No caries risk: Streptococcus mutans-negative individual 2. Low caries risk: Streptococcus mutans-positive individual with a PFRI score of 1 or 2 3. Caries risk: Streptococcus mutans-positive individual with a PFRI score of 3 4. High caries risk: Individual with high S mutans counts and a PFRI score of 4 or 5

Chapter 2. External Modifying Factors Involved in Dental Caries Introduction Awareness of the multifactorial nature of dental caries is of fundamental importance. Figure 48 illustrates the interdependence of most of the determinate variables associated with dental caries. Besides etiologic, preventive, and control factors, many other factors may modify the prevalence, onset, and progression of dental caries. Such factors may be divided into external (environmental) and internal (endogenous) factors (to be discussed in chapter 3). Factors that have proved, in cross-sectional studies, to be significantly associated with increased prevalence of a specific disease are termed risk indicators (RIs). Factors that have proved, in well-controlled prospective studies, to increase significantly the risk for onset or progression of a specific disease are termed risk factors (RFs) and prognostic risk factors (PRFs), respectively. The RF and PRF are often expressed as the odds ratio for the onset or progression of a specific disease. Among external modifying RIs, RFs, and PRFs for dental caries are fermentable carbohydrates, poor socioeconomic status, systemic disease, medication that impairs salivary function, and irregular dental care attendance habits. Fig 48 Relationship between the etiologic factor in dental caries (plaque) and determinants (inner yellow circle) and cofounders (outer blue circle) in dental caries.(Modified from Fejerskov and Manji, 1990.)

Role of Dietary Factors Role of fermentable carbohydrates (sugar and starch) A diet rich in fermentable carbohydrates (frequent sugar intake) is indisputably a very powerful external RF and PRF for dental caries in populations with poor oral hygiene habits and an associated lack of regular topical fluoride exposure from toothpaste. However, in populations with good oral hygiene and daily use of fluoride toothpaste, sugar is a very weak RF and PRF, because clean teeth never decay, and fluoride is a unique preventive factor. The biochemical role of fermentable carbohydrates such as sucrose in the development of an enamel caries lesion on a plaque-covered tooth surface is illustrated in Fig 2 (see chapter 1). Figure 49 depicts the most important variables and the interactions determining an eventual acid attack on enamel after dietary intake. The final (eventually cumulative) effect of single dietary intakes is dependent on their frequency. During and following consumption, depending on the quality of salivary gland function, a certain amount of saliva is stimulated by particular characteristics of the food, such as taste, acid content, and surface texture, and by the intensity of mastication. Together with the volume of saliva secreted, other substrate qualities, such as solubility, stickiness, and an individual food- and host-dependent intraoral distribution, will determine the specific oral clearance of the item in question. Thus, after each intake, distinct amounts of dietary fermentable carbohydrates, acids, and neutralizing agents will be present and capable of influencing the pH of the surface of the tongue, of plaque, and of saliva for a given time. The resolution of the interactions among these three factors, which is greatly influenced by the thickness and diffusion characteristics of the dental plaque on the specific tooth surface, determines the severity (fall in pH) and duration of the acid attack on the tooth surface. The "true" plaque acid production, potentially dangerous to the tooth, should therefore be measured at the enamel surface beneath the undisturbed plaque. Methods for measuring the pH of plaque will be discussed later in this chapter. Categories of fermentable carbohydrates Box 2 shows the fermentable carbohydrates, ranked in order of complexity. All can be fermented to acids by the plaque bacteria. In addition, the sugars may influence the quantity and quality, and thus the cariogenicity, of microbial plaque on the teeth. For several reasons, sucrose is regarded as the arch-criminal in dental caries. Sucrose refined from sugar cane or beet is the most common dietary sugar and is largely responsible for the above-described effects. Apart from familiar sweet products, such as candy, cakes, desserts, jam, dried fruits, and soft drinks, a surprisingly large variety of other everyday foods contains added sucrose: most breakfast cereals, many milk products, some meat and fish products, salad dressings, ketchup, etc. Sucrose also occurs naturally in fruit. The dietary sugars all diffuse rapidly into the plaque and are fermented to lactic and

other acids or can be stored as intracellular polysaccharides by the bacteria, prolonging the fall in pH and promoting a suitable environment for other aciduric and acidogenic bacteria (see chapter 1). Sucrose, however, is unique because it is the substrate for production of extracellular polysaccharides (fructan and glucan) and insoluble matrix polysaccharides (mutan). Thus, sucrose favors colonization by oral microorganisms and increases the stickiness of the plaque, allowing it to adhere in larger quantities to the teeth. Figure 50 (a and b) shows free plaque accumulation on the same tooth during a week on a "sugar-free" diet and a week of high sugar intake, respectively. The sugar-free diet is associated with a thick pellicle and a homogenous, relatively thin plaque, while frequent sugar intake promotes the development of a thick, sticky, sugar plaque with a high percentage of extracellular polysaccharides. Because of this effect on the quality of plaque, sucrose is considered to be somewhat more cariogenic than other sugars. Regardless of possible minor differences between the caries-inducing potential of sucrose and that of other sugars, for practical purposes all dietary monosaccharides and disaccharides are regarded as powerful risk factors: All are rapidly fermented on plaque-covered tooth surfaces. Glucose, fructose, maltose, and sucrose give identical curves for falls in the pH of plaque; for lactose, the fall in pH is somewhat smaller (Neff, 1967). Sucrose constitutes the bulk of dietary sugar. Lactose is present in milk, and maltose is derived mainly from hydrolysis of starch. Glucose and fructose occur naturally in fruit and honey and are also formed by acid hydrolysis of sucrose during the manufacture and storage of soft drinks, marmalade, and other acidic products. Some foods (Swedish baby food) are produced with invert sugar, which is hydrolyzed sucrose. In industrial food processing, the use of glucose is increasing, produced by hydrolysis of starch from cereals or potatoes and declared in the contents as dextrose, corn syrup, or glucose syrup. Therefore, a decrease in national sucrose consumption may not necessarily reflect a drop in sugar consumption. Starch, the major storage polysaccharide of plants, is the major dietary carbohydrate. In many countries, cereals, such as wheat, rice, maize, oats, and rye, provide about 70% of the calories; in the United States and Western Europe, the corresponding figure is 25%. Other important sources of starch are root vegetables (potato, sweet potato, cassava, yams, taro) and pulses (beans, lentils, and peas). Starch is a polysaccharide of glucose. The starch granules in plants are only slowly attacked by salivary amylase because the starch is in an insoluble form and protected by cellulose membranes. Heating at temperatures used in cooking and baking, however, causes partial degradation to a soluble form, which can be further broken down by salivary and bacterial amylases to maltose, maltotriose, dextrins, and small amounts of glucose. Although polysaccharide molecules are too large to diffuse into the plaque, low-molecular weight carbohydrates (released in saliva or at the plaque surface) become available for bacterial fermentation. Consumption of raw starch has little effect on the pH of plaque. The fall in pH following consumption of soluble (cooked) starch and starch-containing foods, such

as bread or crackers, while not as pronounced as for sugars, may easily reach pH 5.5 to 6.0, levels which may be critical for initiation of root caries. A combination of soluble starch and sucrose would be expected to be a more powerful caries risk factor than sucrose alone, because the increased retention of the food on the tooth surfaces would prolong sugar clearance time. Evidence from epidemiologic studies Numerous worldwide epidemiologic studies during the 20th century have shown that caries prevalence is low in developing countries or populations living on a local, carbohydrate-rich diet, based on starch instead of sucrose. Figure 51 shows sugar consumption in 1977 in a number of countries worldwide. Consumption is extremely low in China, and caries prevalence among 12 year olds is very low. On the other hand, sugar consumption in Japan is only about half that of other industrialized countries, but caries prevalence is moderate to high. In contrast, for the last 30 to 40 years, sugar consumption in Sweden has remained persistently high, at about 120 g/per day (Fig 52). At the same time, caries prevalence has decreased from very high to low. Since the early 1950s, it has been "common knowledge" in Sweden that caries is "caused" by frequent intake of sweets. Despite this, over the last 30 years, indirect sugar consumption in the form of sticky sweets, cakes, and so on has increased from about 30% to more than 60% of total sugar consumption (see Fig 52). The dramatic reduction in caries prevalence is therefore attributable not to a reduction in dietary sugar but to a marked improvement in oral hygiene habits, an associated widespread, regular use of fluoride toothpaste, and needs-related professional preventive measures. However, comparison of international data discloses an association between sugar consumption and caries development. Using information on sugar supplies in various countries, obtained from food balance sheet data prepared by the FAO, and data on caries prevalence from the World Health Organization for 6 year olds in 23 nations and 12 year olds in 47 nations, Sreebny (1982) demonstrated a significant positive correlation between the quantity of sugar available per capita in a country and caries prevalence in 12 year olds, but not in 6 year olds. In both age groups, the availability of less than 50 g sugar per person per day in a country was always associated with decayed, missing, or filled teeth scores of less than 3. However, this type of epidemiologic comparison is flawed: Sugar availability cannot directly be extrapolated to consumption specifically by 6 or 12 year olds. Both caries prevalence and sugar consumption vary among different age groups within each country. In wartime, the availability of sugar is usually restricted. In Japan, annual sugar consumption fell from 15 kg per person prior to World War II to 0.2 kg in 1946. Many attempts have been made to relate the level of sugar consumption before, during, and after World War II to caries prevalence in the children: In Norway, Finland, and Denmark there was a clear relationship between sugar consumption and caries development in permanent first molars in children. One of the most thorough literature surveys was made by Sognnaes (1948), who reviewed 27 wartime studies from 11 European countries, involving 750,000 children. Reductions in caries prevalence and severity were observed in all studies. Because of

the high prevalence of caries in Europeans, reductions in severity were usually greater than reductions in prevalence. Sognnaes observed that, in many of the studies, there appeared to be a delay of about 3 years between the reduction (or increase) in sugar consumption and a reduction (or increase) in caries severity. Evidence from cross-sectional studies Numerous cross-sectional observational studies in children have used dietary interview and questionnaire methods to study the relationship between caries prevalence and consumption of sugar and sweets. The results are somewhat conflicting (Rugg-Gunn, 1989): A significant, but not very strong, correlation between caries and the total quantity of sugar consumed has been found in some studies but not in others. A closer relationship has been demonstrated between caries and the quantities of sweets and confectionery consumed, probably because these products are consumed in ways that enhance cariogenicitybetween meals and over long periodswhereas consumption of even large quantities of sugar at meals seems to do little harm. There may be several reasons for these findings. Most of the studies have shown a weaker correlation between sugar intake and caries than might, in theory, be expected. A general methodologic weakness is that dietary data obtained by questionnaires, 24hour recall, or diet history interviews cover a very limited period, from only 1 day to some months, while caries data express total caries experience over the years (Birkhed, 1990). Furthermore, some people, such as the obese, are known to underreport their intake of sugar. Interstudy comparison is difficult, because the studies have been carried out with a large range of variables: with different age groups of subjects, at different times, in different countries, and in specific populations. For example, dietary information has been collected in a variety of ways: Some reports subdivided confectionery into types of sweets, only some of which were significantly related to caries experience. The term sugary food was seldom defined, which frequently made interpretation of the correlation between caries and frequency of sugar intake difficult. Some studies were limited to only one aspect of sugar consumption, such as bedtime eating habits. In most of the studies, children were not selected for inclusion on the basis of their level of caries experience, but some studies compared only the eating habits of children at the two extremes of the range of caries experience. Absolute figures for caries experience were not reported: In many studies only correlation coefficients were reported. In some studies, although significant correlations were found, the absolute differences in caries prevalence were small. In other cases large differences in sugar consumption habits were observed but insufficient data were presented to allow their inclusion. Several studies have investigated the effect of infant-feeding practices on caries, particularly "rampant caries" (or labial incisor caries) in the very young. Five British studies (Goose, 1967; Goose and Gittus, 1968; James et al, 1957; Winter et al, 1966; Winter et al, 1971) have all shown a strong relationship between labial incisor caries and sugared infant pacifiers, especially nursing bottles. One study which did not show such a relationship was reported by Richardson et al (1981a) in South Africa. The

worldwide use of comforters and their effect on oral health has been reviewed by Winter (1980). Two studies by Granath et al (1976, 1978) are of particular interest: Not only was the level of consumption of sugary foods compared with caries severity, but also two other important confounding factors, fluoride supplementation and oral hygiene practices, were taken into account. The first study, involving 6 year olds, was small (179 children) and the higher level of caries found in the children who consumed larger amounts of sugary foods between meals was not statistically significant. However, the second study, involving 4 year olds, was larger (515 children) and differences between the dietary groups were highly significant. When the effects of oral hygiene and fluoride were kept constant, the children with low between-meal sugar intake had 86% fewer buccal and lingual carious lesions and 68% fewer approximal carious lesions than did children with high between-meal sugar intake. Hausen et al (1981), in a study involving more than 2,000 Finnish children, aged 7 to 16 years, reported that water fluoride level, toothbrushing frequency, and sugar exposure were all important determinants of caries prevalence, but least important was sugar exposure. Similarly, in another study in Finland, involving 543 children in three age groups (5, 9, and 13 years), Kleemola-Kujala and Rasanen (1982) found a stronger relationship between poor oral hygiene and caries than between high sugar consumption and caries, although both relationships were important. However, the combination of poor oral hygiene and poor dietary habits seemed to be synergistic. Very similar results were reported among 159 12- to 16-year-old French Canadians (Lachapelle-Harvey and Sevigny, 1985). Holund et al (1985) reported more frequent consumption of sugary drinks in cariesactive than caries-inactive 14-year-old Danes. Continuing the work begun by Granath in the 1970s, Schroeder and Granath (1983) found that poor dietary habits and poor oral hygiene were both good predictors of caries in 3-year-old Swedish children. A few years later, Schroeder and Edwardsson (1987) reported that the predictive potential of diet and oral hygiene can be enhanced by the addition of salivary Lactobacillus and Streptococcus mutans counts. In another group of Swedish 13-year-old schoolchildren, positive salivary S mutans values were found to be a significant but weak risk indicator for caries, but evaluation of the intake of sticky sugar products according to an estimated point scale disclosed no correlation with caries prevalence (Kristoffersson et al, 1986). Stecksen-Blicks et al (1985) conducted a large survey of the relationship between dietary and toothbrushing habits and caries prevalence in children of three age groups (4, 8, and 13 years) living in two northern communities and one southern community in Sweden. Children from the south had considerably more carious lesions in both primary and permanent teeth. This was attributed to differences in toothbrushing frequency and the age at which dental care started. The lack of observed differences in diet between north and south indicated that diet was not an important factor. A large cross-sectional study in the US specifically investigated the relationship between consumption of soft drinks and caries prevalence (Ismail et al, 1984). Analyses of data from 3,194 Americans, aged 9 to 29 years, revealed significant

positive associations between frequency of between-meal consumption of soft drinks and high decayed, missing, or filled teeth scores. These associations remained even after the researchers allowed for the reported concurrent consumption of other sugary foods and other confounding variables. In some studies, caries experience has been correlated with the subjects' dietary habits some years previously. Persson et al (1985) reported a positive correlation between the consumption of sucrose-rich foods at 12 months of age and the presence of caries at 3 years of age in 275 Swedish children. Both factors were linked to the educational status of the mother. The importance of social factors as determinants of eating habits and caries experience of young children has been highlighted in a number of studies; for example, Blinkhorn (1982) reported caries and sugar consumption in Edinburgh, Scotland, to be much higher in children from socially deprived backgrounds. The role of socioeconomic factors will be discussed later in this chapter. In Hertford, England, Silver (1987) collected data on infant feeding and caries status in children aged 3 years and compared the data to dietary habits and caries status when the subjects were aged 8 to 10 years. "Poor infant feeding" (including the use of sugared foods and drinks) was positively correlated with the subjects' caries experience at 3 years and at 8 to 10 years. Children who in infancy had been bottlefed with sweet drinks were more likely to be consuming sugar-containing snacks at the age of 8 to 10 years, supporting the concept that a taste for sweet food, acquired in infancy, persists in later childhood. The importance of establishing good oral health habits as early as possible and postponing bad habits for as long as possible has recently been highlighted in a 2-year prospective study by Wendt (1995). Almost 700 infants were examined at the age of 1 year and reexamined after 1 and 2 years. At the baseline examination, the amount of plaque, gingival conditions, caries prevalence (decayed or filled surfaces), and salivary S mutans levels were recorded. At the annual examinations, the accompanying parent was interviewed about the child's oral hygiene and dietary habits. The percentage of caries-free children decreased from 99.5% to 71.7% at the age of 3 years. Among children (n = 61) of immigrant parents, only 35% were cariesfree (Wendt et al, 1992). Children who were regularly bottle-fed with sweet drinks at night or breast-fed for more than 12 months (mostly at night, when salivary function is at resting level) developed significantly more new carious lesions than did children with more disciplined dietary habits. Bottle-feeding with sweet drinks was common among children of immigrants (Wendt and Birkhed, 1995). Because this was a prospective study, it confirmed that regular bottle-feeding with sweet drinks, and prolonged breast-feeding at night, should be regarded as risk factors for caries development in infants and toddlers. Children who were caries free at 3 years of age had had their teeth brushed more regularly and frequently: At 1 and 2 years of age, these children already had less visible plaque than did children with caries. Immigrant children had had their teeth brushed less frequently, used fluoride toothpaste less frequently, and, at 1 year of age, already had a higher prevalence of visible plaque than did nonimmigrant children (Wendt et al, 1994).

If dietary risk behavior was already apparent at 1 year of age, the chance of remaining caries free until 3 years of age was highest if good oral hygiene habits were established by the age of 2 years. Caries-related behavioral patterns established during infancy, such as oral hygiene and dietary habits, persisted throughout early childhood (Wendt et al, 1996). In this context, it is of interest to note that in the county of Varmland, Sweden, largescale preventive programs at maternal and child welfare centers emphasize early establishment of good oral hygiene and dietary habits: As a result, from 1973 to 1993, the percentage of caries-free 3 year olds increased from 35% to 97%. Although a few studies (eg, those investigating sugar intake in infant feeding) have attempted to assess lifelong habits of sugar consumption, nearly all cross-sectional studies have attempted to relate current caries prevalence to current consumption of sugar or sweets or, at the most, consumption over the previous 3 to 7 days. As discussed earlier, this approach may be valid in young children, whose teeth have erupted and developed caries within the preceding few years and whose sugar consumption habits may have remained relatively constant since the time of tooth eruption; for older groups, its validity is questionable. In a child of 12 years, caries experience is typically confined mainly to the permanent first molars, which erupted 6 years previously and may have developed caries quite early. It cannot be assumed that there has been no change in sugar consumption habits between the ages of 6 and 12 years. It is therefore not valid to relate the dietary habits at one point in time (eg, at 12 years) to caries experience over a very much longer period (eg, 6 to 12 years). However, most cross-sectional studies have attempted to do just this. A typical example is the study by Mansbridge (1960), reporting that caries prevalence was 13% greater in 12 to 14 year olds who admitted consuming more than 8 oz (227 g) of sweets per week than it was in those claiming to consume less. The difference, although statistically significant, was modest. First molar caries prevalence was similar in the two groups, but the difference was pronounced for premolar and second molar caries. The first molars had erupted about 6 to 8 years before the sweet-eating habits were assessed, compared to fewer than 4 years for the premolars and second molars. The many cross-sectional studies conducted several decades ago showed that, at the time, frequent intake of sugar-containing products was often a risk indicator for caries in very young individuals with relatively high caries prevalence. However, recent studies of populations older than 12 years, with good oral hygiene, including regular daily use of fluoride toothpaste, generally show very weak or no correlation between intake of sugar-containing products and caries prevalence. However, a combination of poor oral hygiene and a high frequency of sugar intake seems to have a synergistic cariogenic effect. Evidence from human longitudinal, interventional, and experimental studies There are many reasons why there are so few planned interventional human studies of diet and dental cariesfor example, the problem of persuading groups of people to maintain rigid dietary regimens for long periods of time. Although most of such

studies involved providing daily sugar supplements to subjectsa practice that would be considered unethical todaythese studies made an important contribution to dental knowledge. However, 25 to 50 years ago, at the time of these studies, the standard of oral hygiene was very poor and fluoride toothpaste was unavailable: In most industrialized countries, both caries incidence and caries prevalence were high. Vipeholm study. The Vipeholm study is indisputably unique in the annals of caries research. All previous and most subsequent human studies of the diet-caries relationship have been either epidemiologic or cross-sectional surveys, based on dietary recall. Because such studies are noninterventional, the investigator has no control over either the amount or the frequency of sugar ingestion. Hence, the importance of the Vipeholm findings is matchless. The study was conducted in Sweden, over a 5-year period (1946 to 1951). The aim was to clarify the relationship between sugar intake and caries incidence (Gustavsson et al, 1954). The subjects comprised 436 institutionalized, mentally handicapped, or retarded adults. At the baseline examination in 1946, their mean age was 32 years. Because of their poor oral hygiene (fewer than 20% brushed their teeth regularly), they had abundant amounts of plaque, an important prerequisite for caries development. The subjects were therefore not representative of the general population. The stated objective was to study the relationship between sugar consumption and caries activity by varying the cariogenic substrate (present or absent), the amount of sugar (less than, equal to, or double the normal intake), the form of sugar (nonsticky or sticky), and the frequency of sugar intake (only at meals or at and between meals). The chronology of the study falls broadly into three categories. During the preparatory and vitamin period (1945 to 1947), all subjects received a diet relatively low in sugar (about half the normal intake) and no additional sugar at meals. The baseline caries incidence was low, about 0.34 new carious surfaces per patient per year. During the next 2 years (1947 to 1949), carbohydrate study I, most groups consumed about twice the normal amount of sugar, but only at meals. During the final 2 years (1949 to 1951), carbohydrate study II, most groups ate normal amounts of sugar, some only at meals and others both at and between meals. Seven distinct study groups were established: 1. Control group: continued on a low-sugar diet, only at meals 2. Sucrose group: received a high-sugar diet, mostly in drinks with meals 3. Bread group: received sugar intake either half that or equal to normal, but only in sweetened bread at meals 4. Caramel group: given 22 sticky candies, either in two portions at meals (carbohydrate study I) or in four portions between meals (carbohydrate study II) 5. Eight-toffee group: given eight toffees in two portions at meals (carbohydrate study I) or in four portions between meals (carbohydrate study II)

6. Twenty-four-toffee group: allowed to eat 24 toffees, at their pleasure throughout the day, with about twice the normal total intake of sugar 7. Chocolate group: given milk chocolate in four portions between meals (carbohydrate study II) Sugar consumption at meals in a nonsticky form, over a wide range of total daily intake, from 30 to 300 g (carbohydrate study I), had very little influence on the baseline caries rate of 0.3 to 0.5 new carious surfaces per year. The addition of sugar to the diet resulted in an increased caries incidence, but the increase varied depending on the manner of consumption (carbohydrate study II). Sugar consumed in sweet drinks with meals or in bread eaten at meals had little effect. The group receiving chocolate four times daily between meals showed a moderate increase in caries. However, a dramatic increase occurred in groups receiving 22 caramels, eight toffees, or 24 toffees between and after meals. Thus, caries risk was greatest if the sugar was consumed between meals, in a form that was retained in the mouth for a long time and provided high concentrations of sugar (Fig 53). However, there were wide individual variations. In fact, approximately 20% of the patients did not develop any caries, even after consuming 24 toffees daily. In the Vipeholm study the total sugar consumption by the subjects was about twice that of the normal Swedish diet, and their plaque accumulation was far heavier than normally is found today. Other caries-modifying variables were also different 50 years ago. The results, therefore, should not be extrapolated directly to modern societies. The main conclusions from the Vipeholm study were: 1. Consumption of sugar, even in large quantities, is associated with only a small increase in caries incidence, provided that ingestion is limited to mealtimes, at most four times a day. 2. In subjects with poor oral hygiene, consumption of sugar both between meals and at meals is associated with a marked increase in caries incidence. 3. Under uniform experimental conditions, the increase in caries incidence varies widely from person to person. 4. Caries activity subsides once sugar-rich foods are withdrawn from the diet. 5. In subjects with poor oral hygiene, carious lesions occur despite the avoidance of sugar. Turku sugar studies. By 1970, there was considerable evidence of variation in the rate of acid production from different sugars by plaque microorganisms: For example, the sweet polyalcohols (sorbitol, xylitol, and mannitol) produced virtually no acid. Animal experiments had also demonstrated that sugars differed in their cariogenicity. To test whether the same difference applied to humans, a clinical study was conducted in Turku, Finland, from 1972 to 1974 (Scheinin and Makinen, 1975). The objective was to study the effect on dental caries incidence of almost total substitution of sucrose, with either fructose or xylitol, in a normal diet.

Because the study required the full cooperation of the subjects, including undergoing a wide range of biochemical and microbiologic tests, the study was restricted mainly to adults, most of whom were associated with the Turku dental or medical schools. Of the original 125 subjects, 115 remained after 2 years. There were three groups of subjects: sucrose (S), fructose (F), and xylitol (X). Because full cooperation was essential, the subjects were invited to choose which group they wished to join. All clinical caries examinations were conducted blindly by the observer throughout the study. Two standardized bitewing radiographs were taken of each side of the mouth. Precavitational and cavitational lesions were recorded, for both primary and secondary caries. The organization of the dietary regimens for the subjects in the three groups was very complex, requiring virtually all foods that normally contain sucrose to be manufactured with fructose or xylitol instead of sucrose. The cumulative development of caries, diagnosed both clinically and radiographically, is presented in Fig 54 (a). These results include both precavitational and cavitational lesions. The 24-month caries incidences in the S, F, and X groups were 7.2, 3.8, and 0.0, respectively. These refer to primary caries only. Inclusion of secondary caries (Fig 54 (b)) gives a considerable increase in the magnitude of caries incidence (10.5, 6.1, and 0.9 decayed, missing, or filled surfaces in the S, F, and X groups, respectively at 24 months), indicating the importance of secondary caries in adults. In 1987, to quantify changes in the size of approximal carious lesions, Rekola reexamined the radiographs from the Turku study using a planimetric method. At baseline, the mean size of the lesions was similar for the X and S groups, but at the end of the 2-year study the mean size was significantly smaller in the X group (P < 0.01): Lesion size increased almost linearly by 0.12 mm2/year in the S group but remained virtually unchanged in the X group. Analyses of the caries data over the 2-year period showed that substitution of xylitol for sucrose in a normal Finnish (high-sucrose) diet resulted in a markedly lower caries incidence for both initial and manifest lesions. Although subjects in the S group developed more initial lesions than did those in the F group, more lesions in the F group progressed to cavitation. The X diet was clearly less cariogenic than either the S or F diet, but it cannot be concluded that the F diet was less cariogenic than the S diet. Comprehensive biochemical and microbiologic tests were carried out parallel to the caries assessments. Although a very slight fall in plaque weight was observed in the S and F groups, a much greater decrease was recorded in the X group (P < 0.005). Substitution of dietary sucrose with xylitol did not affect the proportion of major bacterial groups in dental plaque but did reduce the number of most organisms, especially the acidogenic and aciduric flora, including S mutans. Plaque from X group subjects showed a reduced rate of sucrose hydrolysis. No adaptation by plaque organisms to produce acid from xylitol was observed. Experimental caries study. In experimental human studies (Von der Fehr et al, 1970), development of buccogingival enamel caries was evaluated by the use of a dissection microscope. Over a period of 23 days, dental students rinsing nine times daily with 10 mL of a 50% sucrose solution developed a higher Caries Index and more early lesions

than did the control group. Both groups abstained from oral hygiene. After 30 days of oral hygiene and daily fluoride rinses, the Caries Index returned to preexperimental levels (Fig 55). This experiment demonstrated the rapid cariogenic effect of sucrose in combination with dental plaque. The fact that sugar in solution proved highly cariogenic suggests that the critical factor is the duration and frequency of sugar exposure rather than the physical form of the sugar-containing food. However, during the 23 days without oral hygiene, the controls also developed enamel caries. Subsequently, the sucrose rinsing experiment was repeated for 3 weeks. This time, the subjects employed chemical plaque control by rinsing twice a day with 0.2% chlorhexidine solution but used no fluoride; no caries developed (Loe et al, 1972). These two short-term human experimental studies showed that: 1. Sugar is not an etiologic factor for caries development, but it is a modifying risk factor. 2. Dental plaque is an etiologic factor for caries development. 3. Despite frequent sugar intake, clean teeth do not develop caries, even in the absence of fluoride. These early longitudinal, interventional, and experimental studies in Scandinavian adults clearly showed sugar to be an external modifying risk factor for caries development. However, 25 to 50 years ago, both caries incidence and caries prevalence in Scandinavia were very high. Moreover, in the designs of both the Vipeholm study and the experimental caries study, the frequency of sugar intake (eight to 24 times per day) was extreme, and the major modifying preventive factors (plaque control and fluoride administration) were absent. In other words, the etiologic factor (thick, undisturbed dental plaque) was continuously in situ on most tooth surfaces and there was no intermittent supply of fluoride to modify the fall in plaque pH. For ethical reasons, under the Helsinki Declaration, such interventional human studies would no longer be permitted. Therefore, the relative role of sugar as an external modifying risk factor for caries development under present conditions in Scandinavia is unknown. Because of the excellent standard of plaque control and associated use of topical fluorides, particularly in toothpaste, both caries incidence and caries prevalence in children are very low, despite an increase in the consumption of sticky, sugar-containing products over the past 30 years. Observational studies. Although interventional human longitudinal studies with frequent sugar administration are no longer permitted, longitudinal observational human studies are still allowed, and a few have been conducted during the past two decades. The studies by Wendt et al (1992), Wendt (1995), Wendt and Birkhed (1996), and Wendt et al (1996), described earlier, documented the relationship between oral hygiene and dietary habits and caries development from the ages of 1 to 3 years.

In a region of Egypt where the water fluoride concentration was higher than 1 mg/L, Axelsson and El Tabakk (2000b) followed caries incidence in relation to dietary habits in a 2-year study of 685 12 year olds with very poor oral hygiene habits (fewer than 10% brushed their teeth daily). The diet was evaluated according to a cariogenicity point scale. The results showed that a diet rich in sugar was a risk factor for caries development, albeit a weak one. The caries incidence per individual over 2 years, related to cariogenicity scores 1 to 8, 9 to 13, and 14 to 17, was 0.8, 1.0, and 1.9 new carious surfaces, respectively. Two other important large-scale observational longitudinal studies were conducted in schoolchildren in Northumberland, England, by Rugg-Gunn et al (1984) and in Michigan by Burt et al (1988). Some data from the two studies are compared in Table 5. To avoid confounding effects, both investigations were conducted in communities with low concentrations of fluoride in water. In the English study, from 1979 to 1981, the subjects were initially aged 11.5 years; in the American study, from 1982 to 1985, the subjects were initially aged 11 to 15 years. Dietary analyses differed: Rugg-Gunn et al used 3-day diet diaries on five separate occasions, each followed by an interview, using models to assess portion size. Burt et al used 24-hour recall interviews, conducted with the aid of food models, on three or four occasions. To assess the importance of frequency of eating, the timing and grouping of food intakes were noted in both studies. Caries was scored by clinical and partial radiographic examination (Rugg-Gunn et al) or clinically only (Burt et al). In both studies, pit and fissure caries and approximal caries were scored separately. In the Rugg-Gunn et al study, caries incidence was related to a wide range of dietary variables and by examining groups of children with extremes (high versus low) of sugar intake and caries incidence (0 versus more than 7 new decayed, missing, or filled surfaces). For total daily intake of sugars and total caries incidence, the coefficient of correlation was low, but increased when the incidence of fissure caries was considered alone. The overall incidence for smooth-surface caries may have been too low to give statistically significant results. In a later (1987) analysis of the data, Rugg-Gunn et al examined the possible interaction between starch and sugar in the development of caries. The subjects were divided into a high-sucrose/low-starch group, and a low-sucrose/high-starch group. The former developed more new carious lesions than did the latter, but only the difference for fissure caries approached significance. No significant correlations were found between starch intake and any measure of caries incidence. The highsugar/low-starch group ate more frequently than did the low-sucrose/high-starch group (7.8 versus 5.7 times a day). Burt et al (1988) did not report correlation coefficients between caries incidence and dietary variables but divided the subjects into groups corresponding to contrasting caries incidences or dietary practices. Table 6 shows the selected dietary variables for their 0-increment group compared to those from children with 2 or more new carious approximal surfaces during the 3-year study. Only the comparisons of energy derived from carbohydrate (or sugar) in snacks and the percentage of total energy from sugars in snacks reached conventional levels of significance. In contrast to the conclusions of the English study, no difference was observed in the energy intakes from total sugars or from meals and snacks with one or more high-sugar foods. These differences were

unaffected when baseline age was taken into account. Social factors had a highly significant relationship with caries incidence but did not confound any of the relationships with dietary factors. When the subjects were grouped on the basis of high and low sugar intake, those consuming high levels of total sugars developed more new carious lesions, but the difference was not pronounced and approached significance only for approximal surfaces. There was no change in significance when the subjects were grouped according to frequency of eating or high intake of sugary snacks. However, significantly more new approximal carious lesions developed in those eating large amounts of sugars between meals, and here the differences in total caries incidence approached significance. In contrast to the findings of Rugg-Gunn et al (1984), none of the analyses disclosed any differences with respect to fissure caries. Differences in methods of collecting and analyzing the data preclude any direct comparison of these two well-conducted longitudinal trials. Nevertheless, in general, both studies confirmed that, while there is a relationship between caries development and aspects of dietary sugar consumption, it does not explain intersubject variance. Both studies also cast considerable doubt on the importance of the frequency of eating as a caries predictor. Differences between the findings have already been indicated. The English study found that the fissures were more sensitive than the approximal surfaces to intake of sugars and other dietary influences, although the number of fissures attacked was greater and would be expected to show significance more readily. In the American study, although more than 80% of the new carious lesions developed in pits and fissures, caries at these sites was apparently unaffected by sugar intake or eating frequency. Furthermore, the American study found a relationship between caries and the amount of sugar in snacks but not between caries and total sugar intake or intakes with one or more high-sugar items. The English data revealed high degrees of correlation between caries and total sugar intake and a significant relationship between caries and highsugar items. In both reports, the findings were interpreted in the light of the relatively low caries incidence observed (> 1 new carious surface per individual per year). Rugg-Gunn et al (1987) attributed the unimpressive correlation coefficients between caries and many of the sugar-related variables to problems of data collection (especially dietary data), intrasubject variation, and the low caries incidence and suggested, in hindsight, that a longer study and intergroup comparisons would have been preferable. These recommendations were adopted by the American workers, without an apparent increase in the sensitivity of the trial. In this context, it is of interest that comprehensive epidemiologic data collected annually from the total population in the county of Varmland, Sweden, disclosed that in the corresponding age groups, caries incidence decreased from 1.2 new DS in 1979 to 0.1 new DS in 1997 (Axelsson, 1998), even though there was no concurrent reduction in consumption of sugar-containing products. Both animal experiments and human intra- oral biochemical tests (plaque pH, which will be discussed further) strongly suggest that the most critical predictor of dietary

cariogenicity is frequency of intake and that foods containing sugars generally have the greatest potential to raise the acidogenicity of plaque and thus the cariogenic challenge. The two aforementioned clinical studies, in which the methods were optimized, indicate that these two factors (especially frequency) are not of overriding importance, and this brings into question the likely efficacy of current dietary advice to patients. For example, in the Rugg-Gunn study (1987), the low-sugar/high-starch group (the subjects presumably complying with recommendations to reduce sucrose and increase dietary starch intake) achieved only a 31.7% reduction in cariesa modest result compared to the groups whose dietary practices would certainly not be approved by dental health educators. The most significant predictors of caries risk identified in the study by Burt et al (1988) were social factors: parental education and income. The role of educational level as an external modifying factor will be discussed later in this chapter. Fig 2 Development of noncavitated enamel caries. (Modified from Fejerskov and Clarkson, 1996.) Fig 49 The variables and interactions that determine an eventual acid attack on enamel after eating. (Modified from Imfeld, 1983.)

Fig 50 Free plaque accumulation on the same tooth during a week of a sugar-free diet (left) and a week of frequent, high sugar consumption (right). (From Egelberg, 1965. Reprinted with permission.) Fig 51 Sugar consumption in selected countries, 1977.

Fig 52 Sugar consumption in Sweden, 1960-1990.

Fig 53 Results of the Vipeholm study, which clarified the relationship between sugar intake and caries incidence. - DMFT = decayed, missing, and filled teeth - Vit = vitamin period - CHI1 = carbohydrate study I, part 1 - CHI2 = carbohydrate study I, part 2 - CHII1 = carbohydrate study II, part 1 - CHII2 = carbohydrate study II, part 2 (From Gustafsson et al, 1954.) Fig 54 (a) Cumulative development of primary caries, including precavitational and cavitational lesions, diagnosed clinically and radiographically. (b) Caries incidence including primary and secondary caries. (From Scheinin and Mokinen, 1975.)

Fig 55 Experimental study showing the rapid cariogenic effect of sucrose and dental plaque. (From Von der Fehr et al, 1970. Reprinted with permission.)

Influence of hydrogen ion concentration (pH) of plaque It is generally accepted that enamel caries is the result of a disturbance in the equilibrium between enamel hydroxyapatite and the calcium and phosphate ion concentrations of the dental plaque covering the enamel surface. At neutral pH, plaque seems to be supersaturated with these ions. A fall in pH, however, caused by intraplaque bacterial fermentation of carbohydrates, leads to a shift in the equilibrium of concentrations and to dissolution of enamel. The "critical pH" for enamel dissolution ranges from 4.5 to 5.5, depending on such conditions as the presence of fluoride in the plaque and enamel crystal fluids. Because dental caries is a multifactorial disease, many factors influence the pH of plaque: 1. The amount, thickness, age, site, and composition of the plaque 2. The amount, concentration, composition, clearance time, and permeability into the plaque of fermentable carbohydrates in retentive microenvironments of the dentition, and in saliva and gingival exudate 3. The amount and quality of saliva, as well as its access to and ability to permeate the plaque 4. The concentration of fluoride, calcium, and phosphate ions in the plaque If the acidogenic theory of caries etiology is accepted, measurement of plaque pH before, during, and after a food is eaten should be a guide to its cariogenic potential. As a basis for counseling patients on the potential cariogenicity of their diet, the acidogenicicty of various foods, drinks, and meal patterns can be compared under standardized conditions. Although acidogenicity is measured, not cariogenicity, there should be a strong correlation between the two, modified only by the possible presence of protective factors, such as fluoride, which may protect the enamel against dissolution, even at low pH. Measurement of pH Three main methods have been used for measuring plaque pH. The original method, still in use, is the scraping, or harvesting, method developed by Fosdick et al (1941) and subsequently used in Sweden (Frostell, 1969), the United States (Edgar et al, 1975), and the United Kingdom (Rugg-Gunn et al, 1975, 1978). Small samples of plaque are obtained from representative tooth surfaces and pooled. The pH is measured in the laboratory with a pH meter. The touch-on/microtouch method was originally developed by Stephan (1940, 1943,

1944) in his renowned "Stephan curve" experiments. Microelectrode metal probes or glass probes are inserted in plaque in situ. The method was commonly adopted and later improved by the introduction of new thin palladium oxide microelectrodes, providing increased accessibility through the entire thickness of the plaque with less disturbance. A disadvantage is that the plaque is penetrated from the outer surface, and the "true" pH between the tooth surface and the deepest part of the plaque may be altered, for example, by saliva. The telemetric indwelling electrode method, developed by Graf and Muhlemann (1966), is the most technologically advanced and expensive, but also the most accurate method for measuring the true pH beneath undisturbed plaque. A glass electrode tip is built into either the crown of an extracted tooth or a denture tooth in a partial prosthesis, in such a way that the tip is positioned, for example, in the approximal space. Plaque is allowed to accumulate on the tip of the electrode. Wires or radiotransmitters can be used to relay readings from the mouth (Figs 56a, 56b, and 56c). Figure 57 is a detail of plaque, freely accumulated over 7 days, on the tip of an indwelling electrode inserted in the approximal surface of an extracted natural tooth crown fixed in a partial denture. This telemetric indwelling electrode method allows continuous readings of pH at the undisturbed glass-tooth surface-plaque interface, even in the least accessible areas interproximally, where metal or glass touch-on electrodes cannot be applied. The new microelectrodes for the touch-on method have partly overcome this problem, but will disturb the microflora each time they are inserted at the site of measurement, with a possible effect on plaque permeability. On the other hand, microelectrodes allow studies on large representative samples of individuals at any site in the mouth and can be used under field conditions. Recent comparative studies of the three different methods (for review, see Nyvad and Fejerskov, 1996) for measuring plaque pH have indicated that the microtouch and telemetric methods give more pronounced pH responses than does the sampling method and are therefore more appropriate for differentiation of the acidogenic potential of different foods. However, irrespective of the method used, the original observations by Stephan (1944) have been confirmed: When microbial deposits are exposed to a fermentable carbohydrate, such as sucrose, for a short period of time (1 to 2 minutes), pH falls rapidly within the ensuing minutes. The pH then gradually rises, although not as rapidly, and the baseline level is resumed within 30 to 60 minutes. The severity and duration of the fall in pH will depend somewhat on the developmental stage and age of the plaque covering the tooth surface. However, when the telemetric and Stephan methods are used concurrently on the same plaque-covered surface, the telemetric pH curve is more individualized and sensitive than the standard Stephan curve for different food items tested in sequence. The telemetric method is most frequently used on posterior approximal surfaces, which, in toothbrushing populations are the most caries susceptible. As a reference, a lingual "plaque-free" surface is used. A 10% sucrose solution is usually used as a positive control, after the subject has chewed a piece of paraffin wax for a few minutes. In plaque more than 3 days old, a 10% sucrose solution results in optimal pH

fall. Most of the telemetric studies have been conducted by Imfeld (1977, 1983). Relationship of plaque location to pH Stephan curves from approximal plaque show significant intraoral regional differences: Mandibular plaque has a less pronounced pH response than does maxillary plaque (Fig 58). Even within the maxilla, there are local differences in the Stephan response, attributable partly to variations in accessibility to saliva. The lowest pH values are recorded for anterior sites. The gradual resumption of baseline pH values probably results from diffusion of acids out of the plaque and the neutralizing effect of buffers within the plaque and in the salivary film covering the plaque surface (Fejerskov et al, 1992). Accessibility of saliva is influenced by tooth morphology and location and variations in the flow of saliva from the different salivary glands. Access to pits and fissures and approximal surfaces is poor, favoring plaque acidity (Kleinberg and Jenkins, 1964); other tooth surfaces in the vicinity of these sites will be more accessible to saliva and as a result, the plaque will be much less acidic (see Fig 11). Some teeth, such as the mandibular incisors, are located in regions of the mouth where saliva is abundant. The plaque on the maxillary incisors is less alkaline than the corresponding mandibular plaque and favors the development of caries, whereas there is a greater tendency to calculus formation on the mandibular incisors. The volume of saliva secreted by the major salivary glands varies considerably: The greatest flow is from the submandibular and sublingual glands, which have duct orifices in the floor of the mouth just lingual to the mandibular incisors. There is an important difference between the intraoral distribution of sugar ingested in solution and that of sugar in solid foods. Sugar in solution flows over the same tooth surfaces as the saliva and will rapidly be cleared from the oral cavity, except under certain conditions, eg, if ingested in high concentrations or if salivary secretion is seriously impaired. The sugar in solid foods that have to be chewed will enter pits and fissures and be retained in approximal embrasures, the stagnation sites in the dentition. By using the wire telemetric method, Igarashi et al (1989) showed that, after a 1minute rinse with 10% sucrose solution (Fig 59), the pH was much lower in 4-day-old approximal plaque than in the corresponding fissure plaque. Relationship of plaque age and composition to pH The telemetric method has been used to evaluate the influence of plaque age on pH, following a 2-minute rinse with 10% sucrose solution. Figure 60a shows the pH fall in 2-, 3-, 5-, and 6-day-old interdental plaque in a 14-year-old boy. Irrespective of the subject's age, and in experiments in the same test subject over a 2-year period, it seems that a critical fall in pH (to below 5) occurs only in 3-day-old plaque. Figure 60b shows the fall in pH in 3-day-old plaque in a 52-year-old woman, a 7-year-old girl, and a 7-year-old boy after they rinsed with sucrose (Imfeld, 1978, 1983). In a toothbrushing population, such mature plaque would be found, if at all, only on

the approximal surfaces of the molars and premolars. This explains why, in such a population, these surfaces are the most susceptible to caries. Plaque composition also may influence the pH of plaque. The fall in pH after a sucrose rinse would be expected to be more severe in cariogenic plaque with a high percentage of acidogenic bacteria than in noncariogenic plaque. This is illustrated in Fig 61. In a group of 14 year olds, the fall in plaque pH (Stephan curves) after a sucrose rinse was measured in intact occlusal surfaces, inactive occlusal carious lesions, and active occlusal carious lesions (Fejerskov et al, 1992). The age of the plaque, however, may also have varied from subject to subject. Relationship of different carbohydrates and sugar concentrations to pH and clearance time Neff (1967) used the Stephan method for evaluation of plaque pH changes associated with different fermentable carbohydrates. Figure 62 (a) shows the drop in pH for lactose, glucose, maltose, fructose, and sucrose. Figure 62 (b) shows the effect of raw starch, cooked starch, maltose, and sucrose. These experiments indicated that raw starch can be regarded as noncariogenic. However, under certain conditions, lactose and cooked starch may cause a drop in pH to critical values for initiation of root caries. In cariogenic plaque, glucose, maltose, fructose, and sucrose all seem to have the potential to cause a fall in pH to the critical value for the development of enamel caries. In a series of telemetric experiments, Imfeld (1978, 1983) measured the pH beneath 4day-old interdental plaque after 2-minute rinses with 0.025%, 1.25%, 2.5%, 5%, and 10% sucrose solutions. At the beginning and end of each session, as well as between treatments, the acidified plaque was neutralized by salivary flow, which was stimulated by the chewing of neutral paraffin. If very low plaque pH values were attained by glycolysis, rinsing with a 3% carbamide solution greatly improved the neutralization of plaque acids, through intraplaque ammonia formation. Although carbamide is cleared from plaque very rapidly, a further paraffin chewing phase was introduced to ensure its removal, and this always resulted in physiologic plaque pH values. Figure 63 shows the telemetrically recorded pH of 4-day-old interdental plaque in one subject during and after rinsing with increasing concentrations of sucrose solutions. The test solutions were always spit out after the subject rinsed. Sucrose is rapidly fermented in plaque. Regardless of the concentration of sucrose, intraplaque pH drops immediately on sucrose intake and throughout the entire 2minute rinsing period. The very small quantity of sucrose remaining after the 15-mL 0.025% sucrose solution has been spit out is sufficient to depress plaque pH below 5.7. Up to a certain limit (15 mL, 10%), the amount of fermentable substrate is negatively correlated with the lowest pH value reached. Higher sucrose concentrations do not further depress pH. This is the rationale for using a 10% sucrose solution as a positive control in most telemetric experiments. Figure 64 shows the falls in pH occurring on a plaque-free lingual surface and on 4day-old interdental plaque after rinsing with 0.1%, 0.5%, 1%, and 5% sucrose

solutions (Imfeld, 1978, 1983). Even weak sucrose solutions (2.5% to 5%) yield suboptimal drops in pH (to the level of pH 4.2 to 4.5), well within the critical pH range for enamel caries. For reference, Table 7 shows the concentrations of glucose, fructose, and sucrose in some common Swedish food products. It is clearly unrealistic to try to exclude sugar or reduce dietary concentrations to levels low enough to eliminate the risk of inducing critical plaque pH values in mature, cariogenic plaque. A more realistic approach to caries prevention and control would involve: 1. Removal of dental plaque from all tooth surfaces once or twice a day, with concurrent use of fluoride toothpaste. 2. Restriction of the total number of food intakes, including snacks, to four to six per day and exclusion of sticky sugar-containing products: This will reduce the total daily sugar clearance time. In addition to the chemical composition of foods, physical and organoleptic properties (particle size, solubility, adhesiveness, texture, and taste) are important for cariogenicity, because they influence eating patterns and intraoral retention of foods. The oral carbohydrate concentration and the length of time carbohydrates remain in the mouth during and after eating are important characteristics. Foods are eliminated during and after mastication by the flushing action of saliva and by the activities of the masticatory muscles, tongue, lips, and cheeks. Clearance times may be prolonged by retentive factors in the dentition (carious lesions, poor restorations, fixed partial dentures, and removable partial dentures), by low secretion rates, or by high viscosity of saliva. According to the telemetric method, initial oral carbohydrate concentrations and clearance times show large individual variations (Imfeld, 1983) (Figs 65, 66, 67, 68, and 69), and slow clearance increases caries risk. Different foods also vary greatly in initial oral carbohydrate concentration and clearance times. The carbohydrates in fruits with a high acid content, such as apples and oranges (see Figs 66 and 67), vegetables, and various drinks are eliminated within 5 minutes. Sweets, such as sugar-containing chewing gum, caramels, toffees, chocolates, and lozenges, generally result in high oral sucrose concentrations and clearance times ranging from 40 minutes for chewing gum to 15 to 20 minutes for other sweets. On the other hand, concluding each meal with sugar-free fluoride chewing gum is an excellent caries-preventive measure, particularly for high-risk xerostomic patients. Clearance times for bread and crackers may be reduced because the rough texture requires vigorous chewing, which stimulates a high salivary flow. The high secretion rate of saliva induced by vigorous chewing not only has a mechanical rinsing effect but also increases the buffering capacity of saliva, which accelerates neutralization of plaque acids. Increasing the contact time between dental plaque and sucrose leads to a continuously declining interdental plaque pH, thereby increasing its cariogenicity (see, for example, the effect of bananas in Fig 65). The effect of other products, such as some dried fruits

and cakes, is probably even greater. In marked contrast is the effect of cheese (see Fig 68). Bananas, which are pasty and contain 15% sugar (sucrose, fructose, glucose), have the potential to reduce the pH of 4-day-old interdental plaque to almost the same level as rinsing with 10% sucrose solution over a prolonged period (1.5 hours); apples (10% sugar) and oranges seem to be noncariogenic but tend to be erosive because of their acid content (see Figs 65, 66, and 67). This tendency may be counterbalanced by the increased salivary flow stimulated by the acids in fruit. Relationship of eating patterns to pH Studies of the effect of meal patterns on intraplaque acid formation have shown that the fall in plaque pH after consumption of sugary foods may be considerably modified by the consumption of less fermentable foods before, concurrently, or afterward. Imfeld (1983), using the telemetric method, demonstrated the pronounced influence of the last course of a meal on the duration of the postprandial fall in plaque pH. Eating 30 g of Camembert cheese after lunch (see Fig 68) raised the pH of 5-day-old interdental plaque, which had fallen during the meal, but eating chocolate cream as a dessert prolonged and exacerbated the low pH of the interdental plaque (see Fig 69). The observation of the effect of cheese is in agreement with other studies (Schachtele et al, 1982). Animal studies have shown that cheese reduces caries incidence in rats (Edgar et al, 1981). Eating cheese not only stimulates the flow of saliva but also releases calcium and phosphate, which enhance the buffer capacity and remineralization potential. Relationship of sugar substitutes to pH For thousands of years, humans have craved sweet food. Infants rapidly become accustomed to a sweet taste, and this is sometimes acquired prenatally. In frequently consumed snack foods such, as sweets and drinks, less fermentable and noncariogenic sweeteners are increasingly being used as substitutes for potentially cariogenic sugars (monosaccharides and disaccharides). These sugar substitutes are often classified as caloric or noncaloric sweeteners. Among the caloric sugar substitutes are the sugar alcohols (sorbitol, xylitol, and mannitol) and hydrogenated glucose (Lycasin). Examples of common noncaloric sugar substitutes are saccharin, cyclamate, and aspartame. Table 8 from Rugg-Gunn (1989) shows the sweetness of different sugars and sugar substitutes relative to sucrose. Most of the sugar substitutes have been tested by the telemetric method (Imfeld, 1983; Imfeld and Muhlemann, 1978). Figure 70 shows the telemetric pH of 5-day-old interdental plaque in one subject during and after rinsing with 10% aqueous solutions of Lycasin 80/55, xylitol, sorbitol, sorbose, and sucrose. The sugar substitutes Lycasin 80/55, xylitol, and sorbitol, and the sugar sorbose have been declared safe for teeth according to the criteria applied by the Swiss Office of Health. Sucrose, used as a positive control, and administered in the same way as the sugar substitutes and sorbose, resulted in a prolonged fall in pH to below 4.5. Among

others, the longitudinal clinical Turku study (Scheinin et al, 1975) described earlier, as well as the following chewing gum study, have shown that xylitol is noncariogenic. All noncaloric sweeteners are also noncariogenic: They cannot be fermented at all by the acidogenic plaque bacteria. Fig 11 pH drop in molars with approximal plaque at four different sites after rinsing with sucrose solution. (From Firestone et al, 1987.) Fig 56a Volunteer sitting in the Faraday cage during a test to measure the response of interdental plaque pH to the consumption of a European breakfast. (Courtesy T. Imfeld.) Fig 56b Oral wire telemetric prosthesis inserted with cable connection during a recording session. (Courtesy T. Imfeld.)

Fig 56c Tip of an interdental pH electrode covered with 6-day-old plaque. (Courtesy T. Imfeld.)

Fig 57 A close up of plaque, freely accumulated over 7 days, on the tip of an electrode. (Courtesy T. Imfeld.)

Fig 58 Stephan response curves obtained after a sucrose rinse from approximal spaces in the maxilla and the mandible. In the 7 year olds, pH was measured between deciduous molars, and in the 14 year olds, between premolars. Note that mandibular sites exhibit a less pronounced pH

drop. Bars indicate standard errors. (From Fejerskov et al, 1992. Reprinted with permission.) Fig 59 Results of the wire telemetric method show the low level of pH in approximal plaque compared to fissure plaque. (From Igarashi et al, 1989. Reprinted with permission.) Fig 60a Telemetrically recorded pH of 2-, 3-, 5-, and 6-day-old interdental plaque in a 14-year-old boy during and 15 minutes after a sucrose rinse (15mL, 10%). PC = 3 minutes paraffin chewing. Fig 60b Comparsion of telemetrically recorded pH values of 3-dayold interdental plaque in a 52 year old and two 1 year olds during and 15 minutes after a 2-minute sucrose rinse (15 mL, 10%). (From Imfeld. Reprinted with permission.) Fig 61 Stephan response curves obtained from sound occlusal surfaces, inactive occlusal caries lesions, and deep, active occlusal carious cavities following a sucrose rinse in a group of 14 year olds. (From Fejerskov et al, 1992. Reprinted with permission.) Fig 62 The Stephan method evaluates plaque pH changes associated with particular carbohydrates. (From Neff, 1967. Reprinted with permission.) Fig 63 Telemetrically recorded pH of 4-day-old interdental plaque after rinsing with increasing concentrations of sucrose solution. U = 2 min 3% urea rinse. PC = 3 min paraffin chewing. (Courtesy T. Imfeld.) Fig 64 Fall in pH on a plaque-free lingual surface (lg) and on 4-dayold interdental plaque (id) after rinsing with different concentrations of sucrose solutions. PC = 3 min paraffin chewing. (Courtesy T. Imfeld.) Fig 65 The telemetric method shows that initial oral carbohydrate concentrations and clearance times exhibit large individual variations, and that slow clearance increases caries risk. PC = 3 min paraffin chewing; D = days, age of plaque. (Courtesy T. Imfeld.) Fig 66 The telemetric method shows that initial oral carbohydrate concentrations and clearance times exhibit large individual variations, and that slow clearance increases caries risk. PC = 3 min paraffin chewing; D = days, age of plaque. (Courtesy T. Imfeld.) Fig 67 The telemetric method shows that initial oral carbohydrate concentrations and clearance times exhibit large individual variations, and that slow clearance increases caries risk. PC = 3 min paraffin chewing; D = days, age of plaque. (Courtesy T. Imfeld.) Fig 68 The telemetric method shows that initial oral carbohydrate concentrations and clearance times exhibit large individual variations, and that slow clearance increases caries risk. PC = 3 min paraffin chewing; D = days, age of plaque. (Courtesy T. Imfeld.)

Fig 69 The telemetric method shows that initial oral carbohydrate concentrations and clearance times exhibit large individual variations, and that slow clearance increases caries risk. PC = 3 min paraffin chewing; D = days, age of plaque. (Courtesy T. Imfeld.) Fig 70 The telemetrically recorded pH of 5-day-old interdental plaque during and after rinsing with water solution of some sugar substitutes (Lycasin, xylitol, sorbitol, sorbose) and 10% sucrose. U = 2 min 3% urea rinse. PC = paraffin chewing. (Courtesy T. Imfeld.) Evaluation of dietary factors The human longitudinal studies described earlier showed that, in individuals with little or no plaque control and no use of fluoride, frequent intake of sugar-containing products is a significant risk factor or prognostic risk factor for dental caries. In addition, in vivo plaque pH measurements have shown that the drop in pH and sugar clearance time in undisturbed plaque (more than 2 days old) is related to the sugar concentration and consistency of the food item being evaluated (see Figs 63, 64, 65, 66, 67, 68, and 69). Frequent intake of sticky sugar-containing products results in prolonged sugar clearance time, which further prolongs the drop in pH on all tooth surfaces covered with undisturbed cariogenic plaque. Because a prolonged drop in plaque pH eventually results in demineralization of the enamel and development of a carious lesion, the sugar clearance time, based on an evaluation of dietary habits and eating patterns, would seem to be essential information to obtain for all caries-active individuals. Data obtained from evaluation of dietary habits not only provides background material for caries risk assessment but also aids dietary counseling in needs-related caries control and the encouragement of good dietary habits in general health promotion. Dietary assessment in dental practice is aimed at estimating the cariogenic challenge caused by carbohydrates and assessing the general nutritive value of a diet. This means that information on eating patterns and intake of fermentable carbohydrates, as well as energy and other nutrients, should be collected and evaluated. The goal is to establish the absolute magnitude of these variables with the least degree of measurement error. These objectives form the basis for selecting a method. Of several available methods, the following are suitable for dental practice: dietary history, 24hour recall, dietary record, and food frequency questionnaires. Dietary history All methods can be used in dental practice, but, in its original version, the dietary history method takes 1 to 2 hours. This method is considered accurate when validated with nitrogen excretion in urine, but is generally too time consuming for dental practice. However, modified forms may be combined with one of the other interview methods. 24-hour recall This method is widely used. A trained member of the dental team interviews the

patient about the intake of food and beverages during the latest 24-hour period. Consistency in the technique and the skill of the interviewer are important factors influencing communication and patient cooperation and thereby the result. Food models or life-sized illustrations are recommended by most researchers as an aid in estimating quantities. The portion size can also be given in household measures, such as glasses, cups, tablespoons, ounces, and pounds. To reduce bias, the 24-hour recall is done without prior notice to the patient. It should then be repeated for at least 4 days to establish the eating pattern and intake of energyproviding nutrients. For nutrients with large day-to-day variations, the number of days is increased. For example, the time required to estimate the true intake of vitamin A is reported to be approximately 40 days. The days should be selected to represent ordinary days and include weekdays as well as a weekend. Boxes 3 and 4 present examples of 24-hour recalls of a highly cariogenic diet and a noncariogenic diet, respectively. Dietary record In dietary records, also called food diaries, the patient records the type and quantity of all food and drink consumed over a prescribed period, usually 3 to 7 days. Estimates of portion sizes and selection of days are the same as for the 24-hour recall. The patient is given the following detailed instructions: 1. To make the evaluation as accurate as possible, ordinary dietary habits should be kept. Record carefully and precisely. For example: a. How many slices and what kind of bread is used for sandwicheswhat kind of spread is used and what filling? b. What is drunk with or between meals? c. Is jam or sugar used with milk, buttermilk, or yogurt? d. How many lumps of sugar in tea or coffee? e. Are vegetables raw or boiled? 2. Include all snacks: soft drinks, sweet rolls, fruit puree, milk with a sandwich, fruit or sweets, chewing gum, and throat lozenges. This prospective strategy may increase measurement error because of incomplete registration or deliberate or inadvertent changes in diet. Both the dietary record and the 24-hour recall method are reported to underestimate intake slightly compared to the dietary history method and excretion of urinary nitrogen. Food-frequency questionnaires A food-frequency questionnaire contains a list of food items, usually 50 to 150 items, selected to illustrate the whole diet or a specific nutrient, eg, sucrose. The patient marks his or her consumption on a scale, ranging from never to several times per day.

An example of this is a questionnaire aiming to measure intake of food items (Fig 71). The patients mark with a cross the most appropriate square. Figure 72 shows another questionnaire with special reference to the frequency of sugar-containing products. The frequency questionnaire also can be used to estimate nutrient intake. There is a strong correlation between consumption frequency and intake of energy and nutrients. The frequency questionnaire method is uncomplicated and inexpensive and may be useful as a screening instrument or for obtaining dietary data at a group level. Analysis of dietary data When the advantages and disadvantages of these methods are assessed, use of the repeated 24-hour recall and the food record method for 4 to 7 days seems to be the most appropriate for dental practice. The 24-hour recall method is preferable for adolescents, for the elderly, and when communication is poor. The length of the study period is decided according to demands for precision of micronutrients, such as vitamins and minerals. For caries, a 4-day record usually meets the requirements. After completion of data collection and a check on the plausibility of the reported consumption, the intake is evaluated. Evaluation of the cariogenic potential includes an estimation of factors such as the number of intakes containing fermentable carbohydrates, the consumption of snacks and sugar-containing drinks at night, and the retentiveness of the cariogenic products. In children and adolescents with an uncomplicated pattern of caries, simply scoring sucrose intake is often adequate. Several inexpensive computer software programs for evaluation of energy and nutrients in the diet are available in Scandinavia, and computer-based analysis of dietary registrations is common. This is a convenient way to evaluate the nutritive value of the intake. The results of changes to the diet are readily demonstrated, which is of great educational value for patients. Another way to estimate nutritive value of the diet is to score the number of intakes representing six specific food groups (Fig 73). Samples of charts that can be used in such a food group-based evaluation are shown in Tables 9 and 10. When the cariogenic potential and nutritive value are assessed, other properties of the food, known to modify the carious process, for example, food that requires chewing, should also be considered. The ensuing stimulation of salivary secretion and distribution reduces the duration of a drop in plaque pH. Fig 71 Dietary questionnaire for adults. (From Holm et al, 1983.)

Fig 72 Dietary habits questionnaire. (Courtesy D. Birkhed.)

Fig 73 The six main food groups arranged in a food circle with recommended daily intake for the average adult. (From the Danish Goverment Home Economics Council, Copenhagen, 1989.) Dietary recommendations for general health promotion General recommendations General guidelines for energy and nutrient intake are given in the Nordic recommendations from 1989, and in recommendations specific for each Scandinavian country. They give age- and sex-specific recommendations for daily energy and nutrient intake as well as minimal daily required amounts for healthy individuals older than 3 years. It is recommended that energy intake be at a level that does not cause obesity and that there be five or six daily intakes of food at even intervals throughout the day. Recent studies indicate that a more frequent eating schedule would offer physiologic benefits, for example, a decrease in total serum cholesterol concentration. However, other aspects of such recommendations must be considered before they are generally adopted. The daily recommended energy intake originates mainly from carbohydrates (55% to 60% of the total energy). Fat should provide a minimum of 20% and a maximum of 30% of the energy. A fat intake below 20% to 25% of total energy may lead to deficiency of essential fatty acids. Protein provides the remaining 10% to 15% of daily energy. Specific recommendations are also given for fiber, salt, alcohol, and micronutrients. To fulfill these recommendations, the average diet in all Scandinavian countries would need modifications described in Box 5. These recommendations are useful guidelines for medical as well as dental practice and can also be applied to reduce caries risk. The dietary recommendations for diabetics are also in general agreement with these guidelines. Individual recommendations After assessment of the dietary information, the advised plan for the individual is formulated. A useful tool may be "sugar clocks," demonstrating the high caries risk associated with frequent eating (Fig 74). In some patients, carious activity may be attributable to a single habit, eg, frequent consumption of sugar-containing lozenges or snacking or drinking soft drinks at night. Such habits are readily identified and usually easily rectified. In other patients, eating habits may be more complex, comprising snacks only and no main meals. In such cases, a change in basic behavior is required.

This process is complicated by the fact that humans dislike change. Therefore, enforced dietary changes will not succeed unless the benefit accrues rapidly and is of demonstrable advantage. This can be seen in some weight-reducing programs or, for example, when uremic patients adopt a protein-reduced diet. Otherwise, a successful change in dietary behavior requires a program of repeated, small steps. This applies to the introduction of new food items and habits in small children as well as in adults. It is also important that the advice be compatible with possible disease conditions or medication in the individual patient and that the proposed changes be acceptable to the patient. Of further importance is that dietary counseling take into account the patient's social situation. The basis for designing advice sheets on proper energy and nutrient intake is beyond the scope of this book. For further information the reader is referred to textbooks on nutrition. The objective of dietary evaluations and recommendations related to dental caries should be to reduce the total sugar clearance time per day. However, because root caries can develop at a pH as high as 6, the intake of sticky, starch-containing products must also be regarded as a powerful modifying risk factor in elderly people with exposed root surfaces and impaired salivary function. High salivary levels of lactobacilli indicate a high sugar intake and low intraoral pH. The Lactobacillus test is therefore a valuable objective supplement to the dietary questionnaire. For caries prevention and control, compliance with the following dietary recommendations is essential (Box 6). Fig 74 Sugar clocks. (left) Frequent eating results in many periods of acid formation in dental plaque (rod areas). (right) Eating occurs five times a day, resulting in long periods (green area) with no acid formation. (Modified from Johansson and Birkhed, 1994. Reprinted with permission.) Influence of other risk factors on diet-related caries Certain conditions may predispose people to risk for diet-related dental caries. Systemic diseases and regular medication may affect caries risk. The disease or medication per se might increase caries risk, but sometimes the increased risk is related to treatment. The increased need for energy and nutrients during a disease episode is often not met, and the patient may be undernourished. Intake of medicines containing sucrose must be noted, eg, fiber supplements for constipation, cough mixtures, and antibiotics. Further, the intake of soft drinks and sweets is found to be high in hospitalized patients. In some diseases, dietary treatment relieves disease symptoms. Thus, a reduced-fat diet eases diarrhea associated with Crohn's disease or irradiation of the abdominal tract. A low-protein diet defers the need for dialysis in patients with uremia. To compensate for the reduced fat or protein intake, carbohydrate intake is increased, and this increases caries risk. Monosaccharides and disaccharides are used generously; otherwise, the meals would be too large. Dental caries in patients with psychiatric disorders may be complex to explain.

Carbohydrates favor the uptake of tryptophan to the brain, and serotonin production is enhanced. Thus, carbohydrates can have a sedative effect, and frequent eating may induce relaxation. Caries resistance may be lowered by concurrent medication with psychiatric drugs which often impair salivary secretion, as will be discussed in chapter 3. Abuse of recreational drugs, such as hashish, may be associated with a craving for sweets. These patients frequently have high caries activity, typically with smoothsurface lesions. A few decades ago, pregnancy was regarded as a cause of tooth loss resulting from dental caries. Although this is no longer the case, pregnancy may be associated with increased caries risk in some women. During the first trimester, problems with oral hygiene may result from nausea. Pregnancy is often associated with cravings for sweets and more frequent eating. Hormonal changes will also reduce the amount and quality of saliva during the final months of pregnancy. Studies have shown an association between obesity and caries prevalence. However, the association with diet has not been clear. Several studies have shown that the obese underreport total energy, fat, and sucrose intake, but overreport vitamin C and fiber. It could, therefore, be assumed that the sucrose intake in obese individuals with a caries problem is higher than is disclosed by the patient during the dietary registration. Occupations in which frequent food sampling is possible, or even a necessary aspect of work, are associated with an increased risk for dental caries. Examples of such occupations are workers in the confectionery industry and restaurant personnel. Bakery workers were also once considered to be at higher risk for caries (for reviews on dietary factors related to dental caries, see Imfeld, 1983; Rugg-Gunn, 1989, in Murray, 1989; Edgar and Higham, 1991, Geddes, 1991, Bowen, 1994, Geddes, 1994, Imfeld, 1994a,b, Marsh, 1994, Johansson and Birkhed, 1994, Nyvad and Fejerskov, 1994, Carlsson and Hamilton, 1994, Rugg-Gunn, 1994). Role of Socioeconomic and Behavioral Factors Introduction At group and population level, socioeconomic factors, particularly educational levels, are emerging as the most important external factors related to dental caries today. History has clearly shown a relationship between social characteristics and dental disease patterns and, in particular, how social changes have influenced those patterns. Wartime, urbanization, and industrialization, to mention a few, have affected caries prevalence. Most often, we think of social class when we talk about social factors. There are various classifications of social class, usually based on the income of the head of the household and the length and type of education. The links between social class and dental caries have been demonstrated in many studies (Antoft et al, 1988; Beal, 1989; Holm et al, 1975; Koch and Martinsson, 1970; Milen, 1987; Schwarz, 1985; Zadik, 1978). Throughout the 20th century, in temperate and industrialized countries, caries prevalence in primary teeth has been found to

increase with decreasing socioeconomic status. In contrast, during the first half of the century, caries experience in permanent teeth was more prevalent in the highest social class, but the situation is now reversed (Milen and Tala, 1986). In most tropical and developing countries, on the other hand, caries prevalence has been reported to increase with increasing socioeconomic status (for review, see Enwonwu, 1981). Social factors are closely linked to behavioral factors, and a great number of behaviors, particularly health behaviors, are characteristic and distinctive for each social class. Other indicators have also been used, for example, which newspaper the household reads, whether the family has a car, the number of households with no bath, the absence of an inside toilet, shared toilet facilities, residents per room, and similar factors. With such information, geographic areas can be ranked separately for each variable. A combination of variables gives a clear indication of the most and least advantaged areas: these appear at opposite ends of the scale. Such a system is valid for extremes, such as high-risk groups, but less reliable for the middle ranges. Palmer and Pitter (1988) used such a classification and clearly demonstrated wide variations in caries status and treatment levels in 8-year-old English children from different social backgrounds. Socially disadvantaged children had a much higher level of dental caries than did their more socially advantaged contemporaries. The potential number of social and behavioral indicators of deprived or disadvantaged groups or individuals is enormous; such indicators must be chosen with care, with special reference not only to relevance to a given society but also to the changing nature of these indicators with time. The Korner Report (Department of Health and Social Security [DHSS], 1982) recently questioned the validity of social class as a health-related variable and set up an inquiry to study alternatives. Sarll et al (1984) have studied the advantages and limitations of a composite indicator, A Classification Of Residential Neighborhoods (ACORN), a system based on census statistics, in terms of its use in planning dental services. In the industrial area in the north of England they found that the socioeconomic ACORN analyses effectively identified differences in caries prevalence. Data collection was simple, and a high proportion of subjects could be classified. In addition, the ACORN classification, relying on postal address, avoided the need for questions about occupation and economic circumstances, which may be particularly difficult in studies of children. The national survey of children's dental health in the United Kingdom (Todd and Dodd, 1985) disclosed regional inequalities in dental health: Children living in England had the least dental caries, and those in Northern Ireland had the most. The findings showed that, at the national level, where a child lives is a more important factor than social class in determining caries experience. Influence of socioeconomic status Social class The relationship between parents' social status and children's dental health has been demonstrated in numerous studies. Many studies in Western industrialized countries have also shown a relationship between on the one hand, the parents' dental health

status, dental knowledge, and dental care habits and on the other, the prevalence and incidence of dental caries in their children (Martinsson, 1973; Martinsson and Petersson, 1972). For example, Martinsson and Petersson (1972) found a much higher percentage of edentulous parents among children with high caries experience than children with low caries experience. Asher et al (1986) reported a significant correlation between parents' dietary carbohydrate intake and the oral health of the dependent child. Beal (1989) has detailed risk factors contributing to higher caries prevalence in children of low socioeconomic background. These include infant-feeding practices conducive to nursing bottle caries, lesser parental involvement in hygiene practices, and a much lesser parental knowledge of, and involvement in, topical and supplementary fluoride regimens. In the US caries prevalence in schoolchildren in relation to the educational level of the children's mothers was evaluated. Because the area had fluoridated drinking water, caries prevalence in the children was generally low; nevertheless, there was still significantly less caries in children with well-educated mothers than in those whose mothers had less education. In the 3-year longitudinal study in almost 500 US schoolchildren, discussed earlier, Burt et al (1988) failed to find any correlation between frequency of intake of sugary products and children who developed 0 or more than 2 approximal carious lesions. Social factors (parents' income and educational level) had a highly significant relationship with caries incidence, but these factors did not confound any of the relationships with dietary factors. In a longitudinal study, Grytten et al (1988c) examined the influence of various social and behavioral variables on, and the predictability of, caries experience in early childhood. Data were collected when the children were 6, 18, and 36 months old, through parental questionnaires and, at 36 months of age, clinical examination. Caries experience at 36 months showed a statistically significant association with the child's sugar consumption as well as with the mother's dental health, dental care attendance pattern, and level of education. However, when a multivariate model was constructed of predictors that bivariately had shown a statistically significant association with caries experience, only the number of missing teeth in the mother was significantly associated with caries experience, and the explained variance of the dependent variable was low. Primosch (1982) investigated the effect of family structure on dental caries experience of children, in an attempt to identify those at greatest risk. Multiple linear regression analysis showed that none of the selected variables in family structure was sensitive enough to predict children at greatest risk. Maternal age at marriage and family size, however, seemed to show the most promise for predictive value. Comparison of the family structure of children with high and low caries experience disclosed the following: 1. Children of parents who married young (mother younger than 20 years and father younger than 22 years) had significantly greater caries prevalence. 2. Children born to mothers younger than 23 years and fathers younger than 28 years were also more susceptible to caries.

3. Children with birth ranks or family size at either extreme (one child or more than three children) were significantly more susceptible to caries. 4. Age-span differences between siblings had little effect on the caries experience of the subject. The predictive power of a number of sociologic and behavioral variables was investigated by Poulsen (1988) in a study of the public child dental service in Denmark. A multivariate logistic regression analysis, expressing caries risk by the odds ratio, showed high risk of caries (Table 11) in the following cases: 1. Learning disability in the child 2. A high level of pocket money spent on sweets 3. Little support from family 4. Little or no discussion about dental health 5. Negative attitudes toward dental health 6. Negative parental attitudes toward a healthy diet 7. Low educational level 8. Economic pressures in the family The sensitivity was 66%, the specificity 80%, and the predictive power 71%. However, when sociologic variables as well as epidemiologic variables were included in the analysis, sensitivity increased to 95%, specificity to 91%, and the predictive power to 91%. This study clearly shows the value of integrating family health support and living conditions in caries-predictive models. The validity and practical application of these promising findings warrant testing on another pediatric population. Ethnicity Several studies have shown highly significant caries differences between racial groups (eg, Clerehugh and Lennon, 1986). In the English city of Coventry, Paul and Bradnock (1986) found the dental health of Asian children to be considerably poorer overall than that of indigenous children. In Sweden, Widstrom and Nilsson (1986) found that the proportion of each immigrant group who visited a dentist was significantly smaller than the corresponding proportion of Swedes, and extractions, endodontic procedures, and dentures were more common in all the immigrant groups. In Britain, the Dental Strategy Review Group (DHSS, 1981) recommended that the community dental service look to the requirements of "special needs groups." Gelbier and Taylor (1985) stated that young Asian children, and possibly children of other ethnic minorities, are dentally disadvantaged through language, primary socialization,

and the lack of appreciation of minority cultures and needs among the ethnic majority. There is extensive evidence of dietary differences between Asians and other groups within the community, not only resulting from different cultural backgrounds but also associated with social deprivation and communication problems. In a study of 5-year-old Asian schoolchildren in an area of Britain with multiple deprivation (Bedi, 1989), three distinct dental high-risk groups were identified: (1) children of Muslim, English-speaking mothers; (2) children of Muslim, non-Englishspeaking mothers; and (3) children of non-Muslim, non-English-speaking mothers. In West Birmingham, where Asian children were shown to have a rate of decayed, missing, or filled teeth nearly twice as high as that of white children, special programs, tailored to meet the needs of special groups, have been recommended (Bradnock et al, 1988). Apart from language and cultural problems and an often low standard of education among immigrants, emigration disrupts traditional eating habits and leads to exposure to new foods. Studies consistently show that breakfast and snacks, the meals with the least symbolic importance, are the first to change. Therefore, immigrants with poor standards of oral hygiene and an associated irregular use of fluoride toothpaste are at high risk of developing caries when they come to Western countries, and this can partly be attributed to dietary changes. The role of the parents' immigrant background on caries development in infants and toddlers was recently highlighted in the longitudinal studies by Wendt et al (1994) and Wendt and Birkhed (1995), mentioned earlier. The aim of the initial studies was to describe oral hygiene factors in infants and toddlers living in Sweden, with special reference to caries prevalence at 2 and 3 years of age and to immigrant status. The study was designed as a prospective, longitudinal study starting with 671 children, aged 1 year. At 3 years, all the children were offered a further examination. A total of 298 children, randomly selected from the original group, were also examined at 2 years. The accompanying parent was interviewed about the child's oral health habits. Compared to the children with caries at age 3 years, the caries-free children had had their teeth brushed more frequently at 1 and 2 years of age, had used fluoride toothpaste more often at 2 years of age, and had a lower prevalence of visible plaque at 1 and 2 years of age. Immigrant children had had their teeth brushed less frequently, had used fluoride toothpaste less often, and had a higher prevalence of visible plaque at 1 year of age than did nonimmigrant children. Seventy-eight percent of 3-year-old nonimmigrant children were caries free, compared to only 50% of the children of immigrant parents. The authors concluded that early establishment of good oral hygiene habits and regular use of fluoride toothpaste seem to be important for achieving good oral health in preschool children. These goals are achieved less commonly in children of parents with an immigrant background (Wendt et al, 1994). The purpose of the second study was to describe dietary habits in infants and toddlers living in Sweden, with special reference to caries prevalence at 2 and 3 years of age and to immigrant status. The study was designed as a prospective, longitudinal study starting with children aged 1 year. At 3 years, all children were offered a further examination. The accompanying parent was interviewed about the child's dietary habits.

Children with caries at 2 and 3 years of age and immigrant children, at the age of 1 year, had consumed caries-risk products, had been fed at night, and had been bottlefed with sweet drinks more often than caries-free 2 and 3 year olds and nonimmigrant children. Although a great variation in dietary habits in infants and toddlers was recorded, the use of sugar-containing products is widespread in Sweden even in early childhood (Wendt and Birkhed, 1996). In contrast to many immigrant parents, however, almost all nonimmigrant parents of today's infants and toddlers are educated at least to matriculation level and have had access to regular preventive programs since birth. With respect to ethnic minorities, the main problems are therefore not the prediction and identification of high risk but the lack of programs tailored to meet their special needs. Few studies of this kind have been conducted in developing countries; these would be of great interest, because the particular parental characteristics associated with children's caries experience are bound to differ in different cultures. Influence of social and behavioral variables As discussed earlier in this chapter, the development of dental caries is a complex interaction of etiologic factors and many modifying risk and protective factors. Social factors influence behavior directly related to dental caries, such as oral hygiene, dietary habits, and dental care habits. Besides the influence of social and sociobehavioral factors on, and interaction with, sugar intake, one behavior in particular influences the caries-promoting effect of sugar intake, namely oral hygiene. It is generally accepted that caries occurs only after plaque has accumulated on susceptible tooth surfaces in individuals who eat sugar frequently. One reason for the difficulty in proving the direct relationship between oral cleanliness and dental caries in point prevalence surveys, as well as in longitudinal retrospective or even prospective studies, is the complex interaction of a number of factors. Several studies have shown an interaction between sugar intake and oral cleanliness. Kleemola-Kujala and Rasanen (1982) found, in a study of 543 Finnish children, a significant relationship between the amount of plaque and dental caries at all levels of sugar consumption. With increasing total sugar consumption, the risk of caries increased significantly only when oral hygiene was also poor. Further analysis showed that the effect estimates for the two factors in combination were always greater than the sums of the separate effects, indicating a synergistic interaction between the two caries determinants. Granath et al (1976) also found an interaction between oral hygiene and dietary habits, but the significance was low in individuals with low caries prevalence. Rajala et al (1980) found, in a study of male adults, that caries experience was consistently higher for sporadic toothbrushers. Their findings indicated that the positive association between reported daily toothbrushing and low caries experience may be more pronounced in groups with higher overall risk status, for example, in the strata where education and income are low, frequency of dental visits is irregular, use of sucrose is high, and fluoride exposure is low. However, in a well-controlled longitudinal 3-year study in 12-year-old Brazilians, it

was recently shown that oral hygiene habits could be improved, and caries incidence thereby reduced by more than 50%, even though test and control subjects all lived in an area with fluoridated water and were supplied with fluoride toothpaste once a month (Axelsson et al, 1994). In most of the aforementioned, the most commonly investigated social factor related to dental caries is social class or socioeconomic status, and the most commonly studied behavioral factors are oral hygiene or dietary habits. This is not surprising, because diet and oral hygiene are the factors most obviously and directly related to caries development. Socioeconomic status also has been recognized for years as one of the main factors influencing equality, or rather inequality, in both general and dental health. Because few studies have addressed the question, little is known of the caries-predictive value of other social and behavioral factors. However, there is some indirect evidence of a relationship with other social and behavioral factors. In a major questionnaire survey in Scotland, toothbrushing was studied in relation to a number of other health-related behaviors in 4,935 11, 13, and 15 year olds (Schou et al, 1990). Toothbrushing was shown to be significantly related to the subjects' health perception, smoking habits, alcohol consumption, breakfast habits, bedtime, sweet consumption, fruit consumption, and video watching. The minority of children who reported low toothbrushing frequency also reported unfavorable behavior in all the other areas. Awareness that many health-related behaviors may interact with each other and with other environmental factors to determine individual health outcome has led to the socalled lifestyle approach in health promotion. Dental health factors are seldom included in analyses of the influence of lifestyle factors on health, and conversely, a person's lifestyle is seldom taken into account in studies of determinants and predictors of dental health. In recent analytic epidemiologic studies in adults we found higher caries prevalence in 35- and 50-year-old smokers than nonsmokers (Axelsson et al, 1998). In another randomized study in almost 600 50 to 55 year olds, we found that subjects who seldom or never exercised had significantly greater tooth loss than those who exercised regularly (Axelsson and Paulander, 1994). The role of social class and educational level of the parents in the dental status of young children has already been discussed. In a randomized analytic epidemiologic study in 35, 50, 65, and 75 year olds in the county of Varmland, Sweden, one of the factors evaluated was the relationship between dental status and educational level. The following clinical data were collected: the percentage of edentulous subjects, the number of remaining teeth, masticatory function according to the modified Eichner Index, prevalence of removable and fixed prostheses, probing attachment level, furcation involvement, Community Periodontal Index of Treatment Needs, caries prevalence (decayed, missing, or filled surfaces and root caries), prevalence of endodontics and apical periodontitis, oral mucosal lesions, and Plaque Index (O'Leary et al, 1972). The subjects also filled in a questionnaire about educational level, other socioeconomic conditions, diseases, use of drugs, body mass index, dental care, and oral hygiene and dietary habits. For evaluation of educational level, the subjects were randomized into elementary school level (low) and more than elementary school level (high) (Axelsson et al, 1990).

Figure 75, from the data collected in 1988, shows the percentage of subjects with low and high educational levels in the four age groups. Among 35 year olds only, 22% had a low level of education, in contrast to 69% and 72% among the 65 and 75 year olds, respectively. However, today almost all 35 year olds are educated to matriculation or tertiary level. For the last 20 years, Sweden has had compulsory education to the end of secondary school (at least 12 to 14 years of education). It is also estimated that, of the current 50 year olds, only 25% have low educational levels, because of the continuous improvement in every cohort of age groups, and the availability of adult education programs. Figure 76 illustrates the dental care habits among all the subjects related to educational level: Irregular dental attendance is much more common among subjects with low educational levels (82.5%) than among those with higher education (17.5%). Figure 77, from the 1988 data, shows the percentage of edentulous subjects in the four age groups in relation to educational level: Among the well-educated subjects, except the 75 year olds, edentulousness was extremely rare. However, the percentage of edentulousness has declined dramatically, even among those with lower educational standards. This can be attributed mainly to changes in indications for extraction of teeth and the introduction in 1973 of a national dental insurance scheme that covers all residents of Sweden. Figure 78 shows the mean numbers of teeth (excluding third molars) in persons with low and high educational levels. Figure 79 shows the percentage of sound and decayed, missing, or filled surfaces in 50 year olds in relation to elementary school (low), secondary school (middle), or tertiary (high) educational level. Subjects with higher levels of education have a greater percentage of intact surfaces and a lower percentage of missing surfaces than do those with less education. However, the percentage of carious surfaces is almost negligible, indicating that the available resources for provision of dental care are adequate, at least with respect to treatment of caries. The results of this large-scale, analytic, cross-sectional study show that low educational level is a very significant risk indicator for tooth loss, dental caries, and periodontal diseases, not necessarily because highly educated people are more intelligent or wealthier. (In Sweden, there are very limited differences in net income, after tax, between occupational categories such as poorly educated laborers and welleducated teachers.) The difference in dental health status is attributable to the fact that highly educated people know how to learn from written information, to seek information about health promotion, and to apply theoretical information, for example, to self-care. Fig 75 Age and level of education. - Epidemiologic study for the evaluation of the relationship between dental status and level of education. (From Axelsson et al, 1990.) Fig 76 Level of education and regularity of dental care visits. Epidemiologic study for the evaluation of the relationship between dental status and level of education. (From Axelsson et al, 1990.)

Fig 77 Percentage of edentulous patients in relation to level of education. - Epidemiologic study for the evaluation of the relationship between dental status and level of education. (From Axelsson et al, 1990.) Fig 78 Mean number of teeth in relation to level of education. Epidemiologic study for the evaluation of the relationship between dental status and level of education. (From Axelsson et al, 1990.) Fig 79 Percentage of intact and DMFSs in 50 year olds in relation to level of education. DSs = decayed surfaces; - Epidemiologic study for the evaluation of the relationship between dental status and level of education. (From Axelsson et al, 1990.) Conclusions Introduction The most important external modifying factors related to dental caries are frequent intake of fermentable carbohydrates and socioeconomic factors. Dietary factors The fermentable carbohydrates may be ranked in order of complexity, as monosaccharides (glucose and fructose), disaccharides (sucrose, maltose, and lactose), polysaccharides (glucan, fructan, and mutan) and starch. If there is undisturbed cariogenic plaque on an accessible tooth surface, intake of any of the fermentable carbohydrates will result in a drop in pH in the plaque and on the underlying tooth surface, where some demineralization may occur (see Fig 2). The most precipitous fall in pH is induced by sucrose, closely followed by glucose, and fructose, while the effect of raw starch is negligible. Sucrose, glucose, and fructose are therefore considered to be highly cariogenic. "Sugar" (sucrose) is used universally as a sweetener and an inexpensive source of energy. Excluding China and some other developing countries, the average annual consumption is about 50 kg per individual. In Sweden, for example, daily consumption has remained persistently high (about 120 g per individual) for 40 years, although the proportion of indirect consumption, in the form of drinks and sticky sweets, has doubled, increasing from about 30% to more than 60%. Nevertheless, during the same period, a dramatic decrease in caries has been achieved in Sweden. Experimental studies have shown that, in germ-free animals, frequent intake of sugar does not result in caries (Orland et al, 1954). However, if cariogenic human bacteria (mutans streptococci) are inoculated into the mouth of one animal in a group being fed on fermentable carbohydrates, rampant caries develops in the whole group (Fitzgerald and Keyes, 1960). In other words, dental caries is an infectious, transmissible, but multifactorial disease. Frequent sugar intake is not an etiologic factor, but an external (environmental) modifying risk factor for development of caries on tooth surfaces covered with cariogenic plaque.

Conflicting results are reported from the numerous cross-sectional human clinical studies investigating the correlation between sugar consumption and caries prevalence. Most of the early studies, conducted in populations with high caries prevalence, showed that high intake of sugar-containing products was a significant risk indicator for dental caries. In more recent studies, in populations where caries prevalence is low, because of high standards of oral hygiene and regular use of fluoride toothpaste (for example, in Scandinavia), little or no such correlation has been found, because "clean teeth never decay," and caries prevalence (experience) expresses the cumulative caries incidence (increment), since eruption of the tooth. A few human longitudinal interventional or observational studies have been designed to evaluate possible correlations between intake of sugar-containing products and caries incidence. Experimental interventional human studies have been carried out in the absence of plaque control and fluoride (Gustavsson et al, 1954; Scheinin and Makinen, 1975; von der Vehr et al, 1970). These early Scandinavian interventional studies in adults demonstrated the following: 1. In the absence of plaque control and fluoride, frequent intake of sugar-containing products is a significant risk factor and prognostic risk factor for dental caries. 2. If sugar is substituted with nonfermentable sweeteners, a significant reduction in caries may be achieved. Recent longitudinal observations in children, however, have shown little or no correlation between the intake of sugar-containing products and caries incidence. Because there is a strong correlation between the in vivo fall in the pH of the plaque and demineralization of the underlying tooth surface, the effect on plaque pH of dietary products containing different fermentable carbohydrates has been investigated extensively. The cariogenic outcome of falls in plaque pH is influenced, however, by the concentrations of fluoride, calcium, and phosphate ions in the plaque fluids and by the microbial composition of the plaque. In vivo plaque pH measurements have shown the following: 1. Plaque pH after rinsing with a sucrose solution is related to plaque age and site. The lowest values are recorded in the maxillary teeth and on the most central part of the approximal surfaces of molars. The pH drops below 5 in interdental plaque more than 3 days old but not in less mature plaque. In a toothbrushing population, interdental plaque more than 3 days old, if present at all, should be located only between molars and premolars. 2. Of the fermentable carbohydrates, the lowest plaque pH is induced by sucrose, closely followed by glucose, fructose, and maltose. The fall in pH associated with lactose and cooked starch (to pH 5.5 to 6.0) is not as severe but is critical for initiation of root caries. 3. Plaque pH is related to the sugar concentration. Rinsing with even a weak sucrose solution (2.5% to 5.0%) results in a suboptimal pH drop (below 5) in interdental plaque that is more than 3 days old. The optimal pH drop occurs with 10% sucrose

solution; concentrations greater than 10% do not further depress plaque pH. Many dietary products, such as mustard, ketchup, salad dressing, soft drinks, and ice cream, contain 8% to 13% sucrose. While it is therefore unrealistic to exclude all products containing more than 2% sucrose, the daily number of intakes should be restricted. 4. Plaque pH is correlated not only to the sugar concentration of the product but also to the consistency (texture) and the pattern of consumption. For example, eating cheese directly after sugar-containing products will rapidly raise the plaque pH, in contrast to pasty bananas or sugary desserts. A habit that could be recommended for caries prevention is a combination of the Southern European custom of finishing a meal with a cheese platter followed by the new Scandinavian recommendation of using sugarless fluoride chewing gum after meals. 5. Neither caloric sugar substitutes (sorbitol, xylitol, lycasine, and sorbose) nor noncaloric sugar substitutes (saccharin, cyclamate, aspartame, etc) induce critical falls in plaque pH, even to levels critical for root caries development. While these are now widely used as sweeteners in products frequently consumed between meals, it is unrealistic, nutritionally and economically, to recommend sugar substitutes in food consumed mainly at mealtimes. Clinical cross-sectional studies, longitudinal interventional and observational studies in humans, and animal experiments, as well as in vivo plaque pH measurements indicate a synergistic cariogenic effect of dental plaque and fermentable carbohydrates (particularly sucrose) on plaque-covered tooth surfaces. Evaluation of dietary habits is important, particularly in caries-susceptible individuals. Because caries is a multifactorial disease, dietary data complement clinical and case history data used to compile the patient's riskprofile (see chapter 4). The most common methods for evaluation of dietary habits in relation to dental caries are the dietary history and the 24-hour recall. Emphasis is on the frequency of intake of sticky, sugar-containing products, which prolong sugar clearance time. Dietary recommendations for caries control, while emphasizing noncariogenic or lowcariogenic food habits, should also meet nutritional requirements and recommendations for general health: fortunately, a healthy diet is not cariogenic. The diet recommended for diabetics is in general agreement with such recommendations: a high intake of fresh vegetables and fruits, carbohydrate intake from starch instead of sucrose, and a low intake of fat. For caries prevention and control, there are five major dietary recommendations: 1. Breakfast should be a balanced composition of dairy products, grains, and fruits. 2. The total daily number of intakes, including snacks, should be limited to about four. 3. Sticky sugar-containing products, which prolong sugar clearance time, should be eliminated. Sugarless sweets and soft drinks are available as substitutes. 4. Each meal should include fiber-rich products, which stimulate chewing and salivary flow. Cheese is recommended at the end of the meal.

5. Certain caries-susceptible individuals, particularly subjects with reduced salivary flow, should use sugarless fluoride chewing gum for 20 minutes after every meal. In future, refinement of the intraoral wire telemetric and the microtouch methods for in vivo plaque pH measurements is expected. This will allow a more systematic classification of the cariogenicity of food, for example, a scoring system from 1 to 5. Several years ago, a similar system for sweets, assessed by the intraoral wire telemetric method, was introduced in Switzerland. For ethical reasons, human interventional longitudinal clinical studies are no longer allowed. Therefore, in vivo plaque pH measurement is the only available method for evaluation of cariogenicity of dietary products in humans. Further improvements may also be expected in sugar substitutes, noncaloric as well as caloric, with respect to taste and side effects. Use of such sweeteners will become more widespread in snack foods such as sweets, confectionery, and soft drinks. A concerted effort should be made to prevent infants from acquiring a taste for sweet foods. Animal experiments have shown that, by frequent intake of sucrose during pregnancy, a sweet taste can be acquired prenatally. The studies by Wendt and Birkhed (1996) clearly showed that bottle-feeding of sweet drinks, particularly at night, resulted in significantly increased caries development from the age of 1 to 3 years. Finally, how do international experts perceive the past and present relationship between dietary sugar and dental caries? Bratthall et al (1996), in a recent questionnaire, sought the opinions of 55 international experts on dental caries and preventive dentistry as to the main reasons for the caries decline in many Western countries during the last three decades, specifically in 20 to 25 year olds. The respondents were asked to rank reduced sugar consumption, reduced sugar frequency, fluoride toothpaste, school fluoride programs, reduced amount of plaque, and fissure sealants. The respondents ranked these in the following order: 1. Fluoride toothpaste 2. Reduced amount of plaque 3. School fluoride programs 4. Reduced sugar frequency 5. Fissure sealants 6. Reduced sugar consumption However, it should be noted that fewer than 10% of people worldwide use fluoride toothpaste, and the Western industrialized countries represent 30% to 40% of the world's population. In addition, fewer than 1% of school-aged children worldwide have access to school-based fluoride programs.

Socioeconomic and behavioral factors Early establishment of good oral hygiene and dietary habits and regular use of fluoride toothpaste are of utmost importance. Several studies in infants and toddlers have clearly shown that such habits, as well as dental status, are strongly correlated to the parents' social class (particularly educational level), dental status, regularity of dental care (particularly preventive programs), and ethnic background (immigrants). Organized oral health education programs at maternal and child welfare centers are therefore important strategies for reducing such inequalities. In particular, especially disadvantaged parents, such as some immigrant groups, should be identified and offered special oral health promotion programs tailored to their ethnic background, language, culture, dietary customs, oral hygiene habits, and educational level. It has also been shown that the socioeconomic and educational level of the parents is much more significantly related to caries incidence in children than, for example, the frequency of intake of sugar-containing products. On the other hand, health-related behavior that influences dental caries development, eg, dietary, oral hygiene and dental care habits, and the use of fluorides, is strongly correlated to parental socioeconomic class and particularly to educational level. Prediction of caries risk in early childhood and in schoolchildren might therefore be improved by combining data on behavioral and social factors with clinical examination, rather than analysis of behavioral or parental social variables only. Social class, oral hygiene and dietary habits, and the use of fluorides are the variables conventionally related to caries prevalence. However, many other social and behavioral variables may also influence oral health status. The role of parental educational level on children's dental health status has already been discussed. Even more important is the role of educational level on oral health status in the adult population. Generally, the trend in the industrialized countries is toward an acceleration in the percentage of well-educated adults, particularly among 20 to 50 year olds. There is increasing exposure to information and education about self-care, self-diagnosis, and so on from departments of health, oral health personnel, the media, and others. Such conditions favor a positive outcome for oral health promotion and a consequent improvement in oral health status in all age groups. Other behavioral factors that have also been shown to correlate with oral health status are socalled lifestyle behaviors, such as smoking habits, regular or irregular exercise, and a vegetarian diet. Conflicting results have been reported from studies of caries in mentally and physically handicapped people: Although prevalence is often no greater and sometimes lower than in normal children or adults, more of the caries present in handicapped people remains untreated, and more teeth are extracted. For mentally retarded children, the most important determinant of caries risk is the poor standard of oral hygiene. A mental or physical handicap does not in itself seem to be a predictor of high risk, but handicapped people need special care, and this is not always as readily available as is routine care for the nonhandicapped population. Multivariate predictive methods are superior to single analysis of any social and behavioral variables. Models that include not only sociologic and behavioral but also

clinical variables are superior to those based only on sociologic or epidemiologic variables. However, despite the relatively high sensitivity and specificity of models, few studies have analyzed their practical application. The decision to initiate high-risk programs is not merely an academic question: the impact, politically and philosophically, is of far greater consequence. If a philosophy of equality in resource allocation prevails, equality in health may never be achieved. The high-risk strategy will probably require unequal allocation of resources to achieve equality in health. It is hoped that application of current knowledge and the results of ongoing research about prediction of risk groups and risk individuals will help to advance equality in health. It is, however, more difficult to predict caries risk at that individual level than to identify groups in the population at high caries risk. Social and behavioral markers, although not perfect, are the best available markers for identification of groups but less satisfactory at the individual level. Based on current knowledge of dental disease patterns, public dental health strategies should specifically target those in need, rather than the whole population, irrespective of need.

Chapter 3. Internal Modifying Factors Involved in Dental Caries Introduction As discussed in chapter 2, many factors modify the prevalence, onset, and progression of dental caries. The major internal (endogenous) modifying risk indicators, risk factors, and prognostic risk factors related to dental caries are reduced salivary secretion rate (SSR), poor salivary quality, impaired host factors, chronic diseases, unfavorable macroanatomy and microanatomy of the teeth, and the stage of eruption, all of which favor plaque retention, poor quality and maturation of enamel, and exposed root cementum or dentin. Impaired salivary function, particularly an inadequate SSR, is of utmost importance. Role of Saliva Introduction The secretion rate and quality of saliva are important not only in caries development but also for remineralization. Function of saliva Saliva serves as a first line of both nonspecific and specific defense in the oral cavity against infectious diseases, erosion, attrition, and traumatic lesions of the oral mucosa. Saliva is vital to the integrity of the mineralized tissues (teeth) as well as the soft tissues; to the selection, ingestion, and preparation of food for digestion; and to the ability to communicate. Maintenance of the integrity of the oral tissues is primarily a function of the unstimulated (resting) basal secretions; the functions related to digestion are served by salivary flow stimulated by the intake of food.

Saliva has manifold functions in protecting the integrity of the oral cavity from food residue, debris, and bacteria: 1. Saliva has some buffering effect against strong acids and bases. 2. Saliva provides the ions needed to remineralize the teeth. 3. Saliva has antibacterial, antifungal, and antiviral capacities. Components of saliva also facilitate the motor functions of chewing, swallowing, and speaking, as well as sensory and chemosensory functions in the oral cavity. These functions are summarized in Table 12. Secretion of saliva The normal daily volume produced by the salivary glands is about 0.5 to 1.0 L, of which only about 2% to 10% is produced during sleep. About 80%, stimulated by chewing, is produced during meals; that is, the mechanisms of salivary production are capable of rapid response to physiologic demand. About 90% of the total volume is produced by three major pairs of symmetrically located salivary glands: glandulae parotidea, glandulae sublingualis, and glandulae submandibularis (Fig 80). In humans, salivary glands are classified, according to the nature of their secretion, as serous, mucous, or mixed. Serous glands, for example, the parotids, produce a thin, watery secretion rich in enzymes. Mucous glands, for example, the minor glands of the soft palate, produce a viscid secretion. In mixed salivary glands, such as the submandibular and sublingual glands, the secretory product varies, depending on the proportion of mucous to serous cells within the gland. The submandibular glands are mainly serous, and the sublingual glands are mainly mucous. Salivary glands can also be classified as simple or compound, according to their duct system. The glands comprise mainly ducts and acini (Fig 81). The duct system of the submandibular and parotid glands is well developed and branched, containing intercalated, striated, and excretory ducts. In sublingual glands, the intercalated and striated ducts are sparsely distributed. The minor salivary glands are classified as simple branched tubular glands. Of the major salivary glands (see Fig 80), the parotids are the largest, weighing 20 to 30 g each. The parotid duct (Stensen's duct) is about 5 cm long and opens into the oral cavity opposite the buccal surface of the maxillary second molar (the parotid papilla). The submandibular glands are smaller than the parotids and are surrounded by a welldefined capsule. The main duct (Wharton's duct) is about 5 cm long and opens at the summit of the sublingual papilla just lateral to the frenulum of the tongue. The sublingual gland is composed of several smaller glands; the main duct (Bartholin's duct) opens close to the duct of the submandibular gland. The minor salivary glands, numbering between 200 and 400, produce about 10% of the total volume of saliva. They occur throughout the oral mucosa, with the exception of the gingivae and anterior part of the hard palate. They are named, according to their

location, as labial, buccal, palatine, lingual, glossopalatine, and minor sublingual glands. Salivary secretion from major and minor glands is controlled by both parasympathetic and sympathetic stimuli. Depending on the nature of the stimulus, this also affects the composition of saliva. In general, parasympathetic stimuli increase the output of water and electrolytes, whereas sympathetic stimuli enhance protein synthesis and secretion. Clinically, this difference may have relevance, because both the volume of fluid and the concentration and nature of salivary proteins are important for protection against microbial diseases, such as dental caries: In the clearance process, the waterelectrolyte fraction is important, and the actual antimicrobial activity is determined by the protein fraction. Saliva is secreted in response to neurotransmitter stimuli. For most of the day, neurotransmitter release is low and salivary flow is basal, or unstimulated. During food ingestion, in response to gustatory and masticatory stimuli (via mechanical stimulation of the nerves in the periodontal ligaments), there is a pronounced increase in neurotransmitter release, and secretion is stimulated. Resting secretion is considered to be mainly protective, while the larger volume of stimulated saliva is needed to facilitate ingestion (formation and swallowing of a food bolus) and communication. The bulk of the stimulated saliva is secreted by the parotid gland, which is estimated to contribute about 10% of unstimulated and more than 50% of stimulated whole saliva. Salivary secretion rate (SSR) "Normal" values and thresholds. Of the many studies of SSRs in presumably healthy individuals in different countries, the most remarkable finding is the enormous variability; the SSR ranges between 0.08 and 1.83 mL/min, a 23-fold range, for resting whole saliva and between 0.2 and 5.7 mL/min, an almost 30-fold range, for stimulated saliva. Throughout these vast ranges, individuals are generally free of subjective complaints and objective signs of salivary gland dysfunction. These studies show that normal oral function can be maintained with wide individual ranges in saliva production. Because of this heterogeneity, it is difficult to assess the status of a patient's salivary gland function from a single measurement of SSR. In the absence of complaints or signs, it is difficult to determine the presence of a salivary gland disorder. Furthermore, it is evident that caution should be exercised in comparing a single SSR with a population standard. Changes in a patient's SSR over time are probably a more reliable indicator of oral health. If clinicians routinely assessed saliva production in all their patients, they would be able to establish a patient`s normal SSR and be alert to any decline. This would allow early intervention to prevent or limit the deleterious consequences of salivary gland dysfunction. "Whole saliva," the fluid present in the mouth, comprises not only pure secretions from the major and minor salivary glands but also gingival exudate, microorganisms and their products, epithelial cells, food debris, and, to some extent, nasal exudate. Whole saliva is of clinical relevance for susceptibility to caries and carious activity.

However, there is no linear association between SSR and carious activity, but rather a "threshold effect." Although, for clinical purposes, there is consensus on the thresholds presented in Table 13, this is clearly an oversimplification, particularly at the individual level. The so-called normal values for unstimulated and stimulated SSR exhibit a large biologic variation, which should be considered in relation to sex, body weight, and age. For example, 3- to 4-year-old children, with limited experience of different tastes, seem to have an extremely high SSR per kilogram of body weight, about five times as high as that of 10 year olds. On the other hand, in healthy adults, there is only a limited decline in stimulated SSR with age. Studies by Heintze et al (1983) established the gender-related ranges of unstimulated whole SSR (Fig 82) and stimulated SSR (Fig 83) in adults, with peaks at about 0.3 to 0.4 mL/min and 1.5 mL/min, respectively. However, the secretion rates for both unstimulated and stimulated saliva were significantly lower for females than for males. In most studies, the reported mean value for stimulated SSR is about 1.5 mL/min in females and 2.0 mL/min in males; the difference is attributed mainly to the greater body weight of males. When the secretion rate is evaluated, body weight (reflecting glandular size) should be taken into account. Although Heintze et al (1983) found a significant correlation between unstimulated (resting) and stimulated SSRs, the individual variations were so large that one type of SSR could not readily be predicted from the other. A study by Percival et al (1994) compared the SSR of unstimulated whole saliva and stimulated saliva from the parotid gland in relation to age and gender in "healthy" adults (without medication). The mean values were lower in females than in males. However, while the unstimulated whole SSR (Fig 84) was significantly lower in the older age groups (80 or older) than in the younger groups (20 to 39 years), there was no correspondingly significant difference for the stimulated parotid SSR (Fig 85). Randomized samples of adults will, however, include both healthy and unhealthy individuals. Particularly among the elderly, there is widespread regular use of pharmaceuticals that have systemic depressive effects on SSR as well as the quality of the saliva. Loss of teeth is also strictly age related; because of total or partial edentulousness, chewing stimulation is reduced in a relatively high percentage of the elderly (about 20% to 50% of 65 to 90 year olds). In a randomized sample of about 1,000 50, 65, and 75 year olds, one of many clinical and anamnestic variables evaluated was stimulated whole SSR (Axelsson et al, 1990). Figure 86 shows the percentage of individuals with 0.0 to 0.7 mL/min, 0.8 to 1.4 mL/min, and more than 1.5 mL/min in the three age groups. Link to caries risk. The relationship between stimulated SSR and the development of carious lesions has been studied extensively. Although caries risk is extreme in the absence of saliva or in the presence of very low secretion rates, there does not seem to be a strictly linear correlation. An inverse relationship between stimulated SSR and caries incidence, for both enamel and root caries, is found in most studies, and statistical significance has also been demonstrated in some cross-sectional investigations. Stimulated SSR values of less than 0.7 mL/min are regarded as a

threshold for considerably increased risk of further caries development. Therefore, it is interesting that in randomized samples of 50, 65, and 75 year olds, such low values were recorded in as many as 15%, 20%, and 25%, respectively, about one subject in five over the age of 50 years (see Fig 86) (Axelsson et al, 1990). However, SSR cannot be assessed qualitatively. As the most important clinical variable of saliva affecting susceptibility to dental caries, simple quantitative assessment of stimulated whole saliva should be a routine clinical procedure in the adult population. The same saliva sample can also be used to measure salivary buffering capacity and the levels of salivary mutans streptococci and lactobacilli. In clinical practice, measurement of saliva (sialometry) is particularly indicated: 1. As part of the initial examination of a new patient to be treated for dental caries. 2. During evaluation of preventive and restorative treatment of dental caries, to assess how the overall treatment has affected oral health. 3. In elderly patients who take regular medication, and/or have exposed root surfaces. 4. As part of the investigative procedures for suspected hyposalivation associated with, for example, regular use of medicines with systemic depressive effects on SSR, Sjogren's syndrome and other diseases associated with reduced SSR, or irradiation to the head and neck region. The data gathered by Axelsson et al (1990) were used to analyze the relationship between SSR and dental health. Figure 87 shows the mean values of stimulated whole SSR in dentate and edentulous individuals. Figure 88 shows the mean stimulated whole SSRs related to sex, regular medication, and edentulous versus dentate. Figure 89 shows the mean number of teeth (third molars excluded) in 50, 65, and 75 year olds with low and high stimulated SSRs (0.0 to 0.7 and more than 1.5 mL/min, respectively). The results indicated that the stimulated SSR values may influence the number of teeth lost. Figure 90, from the same study (Axelsson et al, 1990), shows the percentage of intact, decayed, filled, and missing surfaces among 50-, 65-, and 75year-old dentate individuals and all individuals with low and high stimulated SSRs. In a more recent cross-sectional study of a randomized sample of more than 600 50 to 55 year olds, among many clinical and anamnestic variables caries prevalence was related to stimulated whole SSR and regular use of medicines with known systemic effects on SSR (Axelsson and Paulander, 1994). Twenty-nine percent of the subjects were taking medication regularly, and 22% used medicines that impair SSR. Figure 91 compares the frequency distribution of intact, decayed, missing, and filled surfaces in subjects with a stimulated SSR of less than 0.7 mL/min, versus subjects with an SSR of greater than 1.5 mL/min, and in subjects using drugs that impair salivary function versus subjects not taking any medication. These data show conclusively that the SSR is an important factor in caries severity and should be considered when caries risk is assessed. Very low stimulated SSR (hyposialosis) (less than 0.7 mL/min, and particularly less than 0.4 mL/min) results in a high risk of caries. Clinically, it is therefore important to determine whether SSR is

normal or impaired. Symptoms of salivary gland hypofunction resulting in hyposalivation Apart from an increased susceptibility to caries, other oral and systemic disturbances may also be associated with hyposalivation (Box 7). Hyposalivation, or reduced SSR, is not synonymous with xerostomia, which is a symptom reflecting the end result of the process of inflow of pure saliva, evaporation, adsorption to the oral mucosa, and outflow of saliva. Of the saliva that enters the mouth, as much as 0.20 to 0.25 mL/min may evaporate, causing a sensation of dryness, especially in mouth breathers. Smokers may also experience dryness. A study on dental health status related to smoking habits found that, although smokers had higher mean values for stimulated SSR than did nonsmokers, significantly more smokers reported symptoms of dry mouth (Axelsson et al, 1998). Experiments by Dawes (1987) showed that subjects experienced a feeling of dry mouth when their normal SSR was temporarily reduced by 50% (eg, a reduction in stimulated SSR from 2 to 1 mL/min or in unstimulated SSR from 0.4 to 0.2 mL/min). Even healthy individuals with "normal" salivary flow rates can experience symptoms of dry mouth: Studies have reported that 20% of 30 year olds and 50% of 55 year olds are so discomforted by dry mouth that they resort to salivary stimulation or rinsing. The sensation of dryness is usually attributable to hypofunction of the minor salivary glands, which produce a mucin-rich, high-viscosity secretion, rather than to hypofunction of the major glands. Because there is little, if any, relationship between subjective complaints of xerostomia and actual quantitative salivary flow, it is important to measure the SSR in each individual patient. The data in Table 13 show that proper individual diagnosis is almost impossible, although on a population level these numbers are relatively reliable. To assess susceptibility to caries or carious activity, the SSR should be monitored regularly in individual patients and not assessed as "normal" or "abnormal" on the basis of just one measurement. A very low SSR, particularly for unstimulated saliva, results in clinical changes in the oral cavity (Box 8). The most conspicuous feature of salivary gland hypofunction is dryness of the lining oral tissues. The oral mucosa may appear thin and pale, lose its glossy sheen, and feel dry: a tongue blade or mirror drawn across the surface may adhere. Such dry, thin mucosa can also be diagnosed by optical measurements with infrared light (Fig 92) or by mechanical friction measurements (Fig 93). Other clinical changes are increases in dental caries; oral infections, especially candidiasis; fissuring and lobulation of the dorsum of the tongue and occasionally the lips; angular cheilosis (Figs 94 and 95); and occasional swelling of the salivary glands. Milking of the salivary glands may not yield any saliva. New carious lesions are common, and develop rapidlywithin weeks or months rather than yearsand often at atypical sites: the mandibular anterior teeth, at the cervical margins of recent restorations (Fig 96), and on the incisal edges. Candidiasis may appear as smooth red patches or as a diffuse area of intense redness (the erythematous or atrophic form); as white-to-ecru, removable plaques (the

pseudomembranous form or thrush); or as white plaques that cannot be removed by scraping (the hyperplastic form). These lesions often appear on the dorsum of the tongue and the palate. The presence of Candida on the mucosal surfaces and in the saliva can readily be determined by a simple dip-slide test. Patients with xerostomia may also have a wide variety of nonoral clinical signs (see also Box 7). Ocular changes include xerophthalmia, keratoconjunctivitis, decreased lacrimation, and the accumulation of viscous secretions in the conjunctival sac. Involvement of the exocrine glands may lead to pharyngitis and laryngitis, persistent hoarseness, a dry cough, and difficulty with speech. Nasal dryness may induce scab formation, epistaxis, and loss of olfactory acuity. A decrease in the production of saliva, as well as in secretions from the gastrointestinal tract, may lead to reflux esophagitis, heartburn, and constipation. Causes of hyposalivation and xerostomia The salivary glands derive their fluid from the circulating blood. This fluid, with its electrolytes and small organic molecules, is modified by the glands and, together with the macromolecules synthesized by the gland cells, secreted into the oral cavity (see Figs 80 and 81). Secretion occurs in response to neural stimulation. Disturbances of the blood supply to the gland, of its secretory apparatus, or of the stimuli that elicit secretion may lead to a decrease in the production of saliva. As mentioned earlier, a person experiences oral dryness when the volume of saliva decreases to about half the normal flow rate; in xerostomia, the most extreme form of dry mouth, the decrease is significantly greater. For the resting flow of saliva to fall to such a level, more than one gland must be affected. The loss of activity of a single gland, observed in patients with salivary gland tumors and of sialoliths, does not result in oral dryness. Thus, xerostomia is the result of multiglandular salivary hypofunction, frequently as a result of the intake of xerogenic drugs, therapeutic irradiation, or certain systemic conditions. Age and decreased mastication may also contribute to the feeling of oral dryness. The most common causes of salivary gland hypofuction and xerostomia are listed in Box 9. Medicines. The most common cause of hyposalivation and xerostomia is the use of xerogenic drugs. It is estimated that more than 400 drugs, some commonly used, can cause oral dryness and induce salivary gland hypofunction. These include anticholinergics, anorectics, antihistamines, antidepressants, antipsychotics, antihypertensives, diuretics, and antiparkinsonian drugs (Box 10). Many physicians are still unaware of these side effects, and patients are not informed about the increased risk of caries. The dentist should therefore always ask for detailed information about the patient's medication. When the medication has a known saliva-inhibiting effect, the physician should be consulted: Although management of systemic disease has first priority, and medication should not be changed for dental reasons alone, the physician may be able to suggest an alternative drug or modify the dose. Irradiation. Patients irradiated for the treatment of oral, head, and neck carcinomas frequently experience severe hyposalivation (and even the absence of salivation),

xerostomia, mucositis, and dysgeusia. The effects of the radiation are dose, time, and gland dependent. As far as possible, radiation oncologists shield the glands from the full dose of radiation, but, when bilateral exposure of the salivary glands is unavoidable, xerostomia may be permanent. Patients often experience oral dryness at an early stage of treatment, and this is exacerbated as the therapy proceeds. In one study, a 50% reduction in the resting flow rate of parotid saliva was recorded 24 hours after the administration of only 2.25 Gy; after 6 weeks of treatment (60.00 Gy/fraction), the flow rate had declined by more than 75% (Sreebny et al, 1992). In most patients, the disturbance of salivary gland function and the associated xerostomia are irreversible. A more than 95% reduction in salivary secretion has been found to persist 3 years after treatment. The mechanisms underlying the acute effects of radiation on salivary function are not known; the early effects may result from damage to the blood supply or interference with transmission of nerve impulses. The later effects are due to destruction of the secretory apparatus and its subsequent replacement by fibrous connective tissue and to specific vascular damage (endarteritis). The secretory cells, the blood supply and the nerves may all be affected by ionizing radiation. Serous cells are more sensitive to radiation than are mucous cells: the parotid gland, which contributes the bulk of the serous component of saliva, is therefore most vulnerable to damage, while the minor salivary glands may still function normally. Thus there is not only a pronounced reduction in salivary flow rate but also a change in salivary composition. The saliva becomes a viscous, white, yellow, or brownish fluid with a reduced pH, reduced buffering capacity, and altered electrolyte and protein content (see Fig 94). Among the irradiation-induced changes in the mouth is also a pronounced increase in the numbers of acidogenic, cariogenic microorganisms at the expense of noncariogenic bacteria. Clinically, the most significant changes are increases in mutans streptococci, lactobacilli, and Candida species. The quantitative and qualitative salivary changes predispose the irradiated patient to a variety of oral problems, such as extreme xerostomia and all its symptoms, listed in Box 7, and rapid and extensive dental caries (see Fig 96), unless intensive caries-preventive measures are stringently followed. In addition to having rapid onset and progression, radiation caries characteristically occurs at sites normally relatively resistant to dental caries (lingual and incisal surfaces). The areas just below the contact points, usually the sites most susceptible to caries, are often the last to be affected. Because therapeutic doses of radiation cause no direct damage to the tooth structure, the associated enormous increase in the cariogenic challenge is attributable to hyposalivation-related alterations in microbial, chemical, immunologic, and dietary variables. The salivary glands are usually located within the treatment portals for head and neck cancer, and, at present, there is no proper clinically acceptable radioprotection. Treatment of the resulting severe hyposalivation is therefore partly palliative and partly directed toward caries prevention: mechanical plaque control, use of antimicrobial and slow-release fluoride agents, stimulation of residual salivary gland tissue by masticatory and gustatory stimuli, such as fluoride chewing gums, and symptomatic relief of oral dryness with artificial saliva that contains fluoride. These measures are not limited to irradiated patients but are appropriate for all patients with severe hyposalivation and xerostomia.

Systemic diseases. Systemic diseases, and the drugs used in their management, frequently cause a marked reduction in salivary secretion. Xerostomia and salivary gland hypofunction are intimately associated with a number of systemic diseases and conditions, some of which cause progressive destruction, usually irreversible, of the gland parenchyma. Others may have vascular or neural effects that are transient and reversible. Included among the diseases are rheumatoid conditions (sometimes referred to as collagen-vascular, connective tissue, or autoimmune disorders); hyposecretory states; certain common diseases (eg, hypertension and diabetes mellitus); cystic fibrosis; certain neurologic conditions; depression and dehydration, anorexia nervosa, and hormonal changes (see Box 9). The classic example of the rheumatoid conditions is Sjogren's syndrome. The primary form is characterized by salivary and lacrimal gland involvement, usually presenting as dry mouth and dry eyes. The secondary form involves at least one of these organs and, in addition, is an associated collagen disorder, most commonly rheumatoid arthritis. Systemic lupus erythematosus, scleroderma, dermatomyositis, and primary biliary cirrhosis may also be associated with secondary Sjogren's syndrome. In the early stages, there may be little change in the SSR; as the disease progresses, there is a corresponding decrease in SSR, resulting from the gradual destruction of the gland parenchyma by a lymphoreticular cell infiltrate. There is massive irreversible acinar cell degeneration and atrophy. Changes resulting from Sjogren's syndrome are not restricted to the mouth and eyes; there may be extraglandular manifestations in the gastrointestinal, renal, genitourinary, and pulmonary systems. The condition is associated with increased risk of pseudolymphoma and malignant lymphoreticular disease. The diagnosis of Sjogren's syndrome is usually made several years after onset, and, in many patients, the dentition is severely damaged before the definitive diagnosis is made. Meticulous assessment of salivary secretion rate by the dentist is an aid to early identification of these patients. The presence of Sjogren's syndrome is often confirmed by biopsy of the minor salivary glands of the lower lip. Sjogren's syndrome is believed to occur in about 1% of the population. It affects eight times as many women as men, most above the age of 45 years. Although it is the cause of profound distress in older women, it is a condition that has been largely overlooked, and surprisingly little information is available in the medical and dental literature. There is increasing evidence that xerostomia is associated with a number of common disorders and diseases such as hypertension and diabetes mellitus. The evidence for the link between diabetes mellitus and xerostomia is of two types. First, there is a greater incidence of diabetes mellitus in xerostomic subjects than in nonxerostomic controls. Second, diabetic subjects with no other diseases and taking no drugs other than insulin have a far greater prevalence of xerostomia than do matched nondiabetic controls (Sreebny et al, 1992). However, insulin-dependent (type 1) diabetes mellitus, as such, does not damage salivary glands to such an extent that hyposalivation can be regarded as a common complication. Reduced salivary secretion is typical only during periods of diabetic instability or during the onset of the disease: During these periods,

increased glucose levels in salivary secretions are common, heightening caries risk. This is one reason why such diabetic patients are at least as caries susceptible as nondiabetic patients. Other conditions associated with reduced SSR. The changes in SSR associated with chronic depression are generally persistent. When oral dryness cannot be attributed to organic change, the patient should be advised to consult a psychologist or psychiatrist to explore possible psychogenic factors. Although psychic states can induce oral dryness, the underlying mechanisms are not well understood. Depression is frequently treated with tricyclic antidepressants, which tend to aggravate the severity of oral dryness. In patients suffering from severe or prolonged malnutrition or anorexia nervosa, deterioration in SSR and the quality of the saliva may lead to oral symptoms, for example, increased susceptibility to dental caries, erosion, and dry mouth. Although short-term fasting also reduces the SSR significantly, it does not lead to true hyposalivation, and the flow rate is restored to normal values soon after the fasting period ends. Hormonal changes may also affect the SSR and the composition of human saliva. The most profound changes are postmenopausal, many studies confirming that the salivary secretion rate is lower in postmenopausal women than in younger women. However, there is a widely varying individual response: Some postmenopausal women experience no detectable change in SSR, while others experience distressing oral dryness (with a number of symptoms, such as "burning mouth," sore tongue, difficulties with speech and swallowing, and fungal infections). Estrogen supplementation therapy has little, if any, effect on SSR after menopause. Age as such does not have a clinically obvious effect on SSR: In healthy individuals, the stimulated whole SSR does not decline. Age-related decreases in secretions from both minor and submandibular glands have been observed, similar decreases in SSR from the parotid glands have not been found. [au: Reference?]Comparative studies of stimulated minor gland secretion in older and younger adults have reported an agerelated decline of more than 50% in secretion, a functional decrease consistent with morphologic studies showing a 40% to 50% reduction in acinar volume with age (Percival et al, 1994). These physiologic changes may explain why many old people experience discomfort from dry mouth, even though the secretion rate for paraffin-stimulated whole saliva is normal. Alterations in submandibular and sublingual gland functions have the greatest impact on the sensation of oral dryness. These changes will also increase the risk for development of denture stomatitis and reduced denture retention in edentulous elderly. Both human and animal studies (Dawes, 1987) have shown salivary gland atrophy to be associated with decreased masticatory activity. In humans, this has been reported in subjects on liquid diets and in patients whose jaws were wired together after orthognathic surgery. The extent to which varying degrees of decreased masticatory activity contribute to salivary gland hypofunction and xerostomia in humans is not known. On the other hand, as will be discussed later in this chapter, studies have

shown that introduction of agents to stimulate chewing may increase the SSR in patients with hyposalivation (Axelsson et al, 1997a). Evaluation of hyposalivation Based on the principle that "status is determined by clinical examination but explained by the case history," the following points should be considered for proper diagnosis of hyposalivation: 1. Stimulated salivary secretion rate 2. Resting SSR 3. Anamnestic data: possible side effects of medication; systemic diseases known to cause salivary gland hypofunction; difficulty in swallowing dry food; difficulty in speaking; soreness of the oral mucosa; frequent episodes of sore throat; difficulty in tolerating removable dentures 4. Tenderness of salivary glands to palpation or swelling of the glands 5. Inflammatory changes in the oral mucosa or tongue 6. Indicative test: Does the examination mirror stick to the cheek mucosa? 7. Atypical pattern of caries (lesions on smooth surfaces or on the tips of incisors or cusps) If most of these above features are present and the resting SSR is low, the diagnosis almost certainly includes hyposalivation, and the patient should be considered at high risk of developing caries. Apart from measurement of SSR, the other anamnestic and clinical variables have already been discussed in detail. The importance of repeated measurements of SSR in the individual patient, and particularly in caries-susceptible individuals, cannot be overemphasized. There are noninvasive, painless techniques for sampling not only whole saliva but also saliva from the individual major and minor salivary glands. Whole saliva is easily obtained and in most cases a good indicator of whole-mouth dryness. Disease in a major salivary gland can often be diagnosed from secretion collected directly from the gland. The purpose and method of the collection procedures should be explained to the patient beforehand. Saliva should be collected about 1 1/2 to 2 hours after eating or after an overnight fast. Patients should be instructed to do nothing that might stimulate the SSR prior to the collection, including mastication of food, chewing gum, or candy; smoking; toothbrushing; mouthwashing; or drinking. The test should be conducted in quiet surroundings. Detailed instructions for standardized measurements of SSR are shown in Box 11. To obtain a mean SSR, sampling should be repeated at least once, at about the same time of day. If the patient's baseline has been established earlier, the values obtained

can be used as a comparative indicator of the patient's present salivary status. If the baseline is not known, as is usually the case, the SSR is compared with the relevant standards for the population (see Table 13). As with any test, the results should be interpreted in light of the patient's history, the presence of any signs of disease, and the results of other tests. Whole saliva may be collected and measured by a variety of volumetric and gravimetric techniques: draining (drooling), spitting, suction, and swab. The volumetric methods to be described, especially a combination of the drooling and spitting techniques, are easily carried out in the dental or medical office. Either of two measuring devices may be used: a Sialometer or any finely calibrated measuring cylinder. The Sialometer is a specially constructed, reusable device that allows collection of both resting and stimulated saliva in a single vessel. Alternatively, the following equipment can be obtained from chemical supply houses: two funnels and two measuring cylinders, each with a volume of about 12 mL, calibrated to no less than 0.1 mL. Measurement of stimulated whole SSR. In the clinic, the usual procedure is to measure SSR during masticatory stimulation (ie, while the patient chews a piece of paraffin). To achieve reliable, standardized results, the patient should be given detailed instructions (see Box 11). The patient is instructed to chew the 1-g piece of paraffin for 1 minute to soften it and then to swallow or spit out all saliva. The patient then chews the softened bolus of paraffin for a fixed time (5 minutes), spitting the saliva into the graduated cylinder. Foam can be avoided or reduced by using an ice-chilled beaker or by the addition of a drop of octanol. The secretion rate is calculated in milliliters per minute. As an alternative to mechanical stimulation by chewing, saliva may be stimulated chemically by a 2% solution of citric acid (made up at a local pharmacy), applied to the laterodorsal surface of the tongue at 30-second intervals, for 2 minutes. The patient then spits the saliva into the receiving vessel. The procedure is repeated twice more, for a total collection time of 6 minutes. As before, the SSR is expressed as milliliters per minute. Measurement of unstimulated (resting) SSR. It is impossible to sample true "resting" saliva, because the SSR is always influenced by some kind of stimulus during consciousness. Nevertheless, a sample collected by passive drooling, without any deliberate physical or chemical stimulation, is more reliable than stimulated saliva as an indicator of reduced SSR and hyposalivation. When resting secretion is collected, the patient is instructed to sit in a relaxed position, with the elbows resting on the knees and the head lowered between the arms, the so-called coachman's position. This position is also good for the collection of stimulated saliva. Even slight movements of the tongue, cheeks, jaws, or lips should be avoided. The lips are only slightly apart, and the patient allows the saliva to drool passively over the lower lip into the measuring cylinder, avoiding actively spitting. For healthy adults, the resting SSR should exceed 0.1 mL/min. In patients suspected to have hyposalivation, the sampling period should be 15 minutes, to avoid bias

caused by fluctuations in the SSR. For clarity, the results should be expressed in milliliters per minute and in milliliters per 15 minutes. Measurement of SSR from the major salivary glands. Parotid saliva is usually obtained in a modified, two-chambered Carlson-Crittenden collector). The inner chamber is placed over the orifice of Stensen's duct; the outer chamber is attached, via thin tubing, to a rubber bulb that, when compressed, creates a slight negative pressure and permits the device to adhere to the surrounding mucosa. This device makes it possible to collect pure parotid saliva in a noninvasive manner. A simple method that makes it easy for the dentist to collect submandibular and sublingual saliva has also been reported. The region of Wharton's ducts is isolated with gauze, and the orifices of Stensen's ducts are covered. The saliva, resting or stimulated, that has collected during a known time is aspirated with a plastic micropipette. The flow rate is expressed as milliliters per minute per pair of submandibular or sublingual glands. Measurement of SSR from minor salivary glands. Saliva may be obtained from the minor salivary glands of the lower lip or the palate. The minor glands are dried and isolated with gauze or cotton rolls. For resting saliva, the fluid that is present at the orifice of one or more of these glands after 2 minutes is adsorbed onto filter strips (Perio-Paper). The volume of fluid on each strip is read electronically in a special device (Periotron). For stimulated minor gland saliva, the tongue is swabbed with 2% citric acid solution, as described earlier. The results are expressed as microliters per minute. Because the number of glands and the area sampled vary, the SSR is semiquantitative. Fig 80 Major salivary glands. (Modified from Tenovuo and Lagerlof, 1994. Reprinted with permission.)

Fig 81 Simplified diagram of a secretory unit in a major salivary gland. The ducts usually branch leading to several acini. (Modified from Tenovuo and Lagerlof, 1994. Reprinted with permission.) Fig 82 Mean unstimulated whole salivary secretion rates in males and females. (From Heintze et al, 1983. Reprinted with permission.) Fig 83 Mean stimulated salivary secretion rates in males and females. (From Heintze et al, 1983. Reprinted with permission.) Fig 84 Mean unstimulated whole secretion rates in healthy (unmedicated) males and females, by age group. *Statistically significant difference between the 80+ females and the other groups. (From Percival et al, 1994.)

Fig 85 Mean stimulated parotid salivary secretion rates in healthy (unmedicated) males and females, by age group. (From Percival et al, 1994.) Fig 86 Frequency distribution of stimulated whole salivary secretion rates, by age. (From Axelsson et al, 1990.) Fig 87 Mean stimulate d whole salivary secretion rates in dentate and edentulou s individua ls, by age. (From Axelsson et al, 1990.)

Fig 88 Mean stimulated whole salivary secretion rates, by sex; by use of drugsregular medication (Med) or no regular medication (No med); and by status of the dentitionedentulous (Edent) or dentate (Dent). (From Axelsson et al, 1990.) Fig 89 Mean numbers of teeth (excluding third molars) in individuals with low (0.0 to 0.7 mL/min) and high (>1.5 mL/min) stimulated salivary secretion rates, by age. (From Axelsson et al, 1990.) Fig 90 Frequency distribution of intact surfaces, decayed surfaces (DSs), filled surfaces (FSs), and missing surfaces (MSs) in individuals with low (0.0 to 0.7 mL/min) and high (>1.5 mL/min) stimulated salivary secretion rates, by age. (From Axelsson et al, 1990.)

Fig 91 Frequency distribution of missing, filled, decayed, and intact surfaces in individuals with low (1.5 mL/min) stimulated salivary secretion rates; and in individuals using drugs that impair salivary function (Med) and those not taking any medication (No med). (From Axelsson and Paulander, 1994.) Fig 92 Optical measurement of the mucosa with infrared light. (Courtesy of T. Axell.)

Fig 93 Mechanical friction measurement of dry, thin oral mucasa in a patient with hyposalivation. (Courtesy of T. Axell.)

Fig 94 Typical dorsum of the tongue in a patient with xerostomia. (Courtesy of T. Axell.)

Fig 95 Angular cheilosis and typical dorsum of the tongue in a patient with xerostomia. (Courtesy of T. Axell.) Fig 96 Typical pattern of carious lesions and restorations in mandibular teeth in an older person with xerostomia. (Courtesy of T. Axell.) Composition of saliva Although composed mainly of water, saliva is a complex secretion. As discussed earlier, so-called whole saliva consists primarily of the secretions from the major and minor salivary glands. Whole saliva also contains a number of constituents of nonsalivary origin: crevicular fluid, serum, and blood cells; bacteria and bacterial products; desquamated epithelial cells and cellular components; viruses and fungi; food debris; fluoride; and some bronchial secretions. It is estimated that 1 mL of whole saliva contains about 700 million viable bacteria, 0.5 million leukocytes (more than 90% polymorphonuclear neutrophil leukocyte-cells), thousands of desquamated epithelial cells, 2 mg of proteins, 800 mg of lipids, 100 mg of immunoglobulins, and some inorganic electrolytes, such as calcium, phosphates, bicarbonate, sodium, chloride, and fluoride. Even pure secretions collected directly from the orifices of the main excretory ducts of the parotid, submandibular, or sublingual glands contain the saliva synthesized by the secretory cells, along with certain substances supplied by the circulation. The composition of whole saliva is influenced by a number of physiologic factors. Important among these are the source, the method of collection, and the degree of stimulation. As described earlier, the major salivary glands are composed of different acinar cells, programmed to synthesize quite different secretions. The parotid glands

have serous acinar cells and produce a proteinaceous, watery secretion. The secretion from the sublingual glands is mucous and hence more viscous. The submandibular glands have both serous and mucous acinar cells and produce a saliva with lower protein content and higher viscosity than do the parotid glands. The minor salivary glands are purely mucous glands and produce a saliva that is particularly viscous and rich in secretory immunoglobulin A (IgA). In response to stimulation, there may be a manifold increase in salivary output, with significant changes in consistency and in the concentration of many of its constituents. About 99% of the saliva is water. The remaining 1% consists mainly of large organic molecules, (eg, proteins, glycoproteins, and lipids); small organic molecules (eg, glucose and urea); and electrolytes (chiefly sodium, calcium, chloride, and phosphates). Most of the organic molecules are produced by the acinar cells; some are synthesized in the ducts, and some are transported into the saliva from the blood. A list of salivary constituents, subclassified as proteins, small organic molecules, and electrolytes, is presented in alphabetical order in Box 12. The major proteins of the salivary glands are produced by the acinar cells and exist as families. Each family has a number of distinct but closely related members (genetic polymorphism). They include the proline-rich proteins (with at least 13 discrete members); the histatins (histidine-rich proteins with five related components); the cystatins (cystine-containing proteins); the tyrosine-rich proteins (statherin and others); mucins of high and low molecular weight; glycosylated and nonglycosylated amylases; and several salivary peroxidases. Other salivary proteins exist in a single form, some produced by acinar and some by ductal cells. Among the acinar proteins are epidermal growth factor, secretory component, and lactoferrin. Lysozyme is known to be produced by duct cells, but, for many other constituents, the site of origin is unknown. Included among the compounds that are transported from the blood into the salivary secretion are the major electrolytes; albumin, IgA, immunoglobulin G (IgG), and immunoglobulin M (IgM); and vitamins, drugs, hormones, and water. There is a good correlation between plasma and salivary levels of a number of hormones and medications. This correlation forms the basis for proposals to use saliva collection as a noninvasive means of monitoring hormones and both therapeutic and illicit drugs. Salivary sampling is currently being tested as a screening method for the presence of antibodies to human immunodeficiency virus 1. However, such methods are complicated by the fact that glandular inflammation results in a marked increase in the number and concentration of serum elements in saliva. Role of saliva as a modifying factor in dental caries Indisputably, an adequate secretion rate and saliva of good quality are essential for oral health. Saliva is well known to have specific protective effects against dental caries. The most direct evidence of this is the rampant caries that can occur following the loss of salivary function as a result of irradiation for head and neck tumors. Within a few weeks, tooth surfaces not normally susceptible to caries may be affected, leading to complete coronal destruction. The principal properties of saliva that protect the teeth against caries are:

1. Dilution and clearance of dietary sugars 2. Neutralization and buffering of the acids in plaque 3. Supply of ions for remineralization 4. Both endogenous and exogenous antiplaque and antimicrobial factors The major functions of different salivary components have been presented in Table 12. Dilution and clearance of food components and clearance of microorganisms Of the many functions of saliva, the most important is the clearance of oral microorganisms and food components from the mouth to the gut. Therefore, an adequate volume to flush noxious (and also commensal) microorganisms out of the oral cavity is a prerequisite for a healthy balance between host defense and endogenous and exogenous microbial attack in the mouth. This balance can be disturbed by either extensive growth of bacteriaas a consequence, for example, of poor oral hygiene, excessive dietary intake of fermentable carbohydrates, or some systemic diseasesor reduced SSR (hyposalivation). In the most highly caries-susceptible individuals, a combination of these factors is common. Although caries research has concentrated on salivary clearance of sucrose and fluoride, the principles that apply to sucrose clearance are valid for any substance introduced to the oral cavity. Besides sugars and fluoride, other substances are of relevance to the clinician: chemical plaque control agents (chlorhexidine, etc), chloride in relation to corrosion of amalgam, and citric acid and other acidic products that might be implicated in tooth erosion. The study of sugar clearance was pioneered by Swenander-Lanke (1957), who found that, following the consumption of solid carbohydrate foods, the concentration of sugar in saliva fell exponentially, at first rapidly and then more slowly. Sreebny et al (1985) noted that sugar solutions were cleared in a two-stage pattern and that the rapid clearance rates over the first 6 minutes and the slower ones thereafter were proportional to the shifts in the SSR at those times. In 1983, Dawes developed a computer model for sugar clearance, based on the following postulate: that the important factors in clearance were (1) the volume of saliva just before and after swallowing and (2) the unstimulated SSR. The computer predictions based on this postulate were confirmed in studies using an "artificial mouth" system and in human experiments. The computer predicted that the clearance was rapid when both salivary volumes were low and the unstimulated SSR was high. Figure 97 shows a physiologic model of the oral cavity that includes the most important properties for understanding the clearance process. The events after intake of sucrose may be described as follows: In the oral cavity, there is a minimum volume of saliva after swallowing, the residual volume. Spread out as a thin film, this volume has been estimated to be, on average, 0.8 mL, but the interindividual variation is large. Dissolution of a small amount of sucrose in this small volume of saliva will

give rise to a very high sucrose concentration. For example, dissolving one tenth (0.3 g) of a sugar lump in the residual volume will result in sucrose concentrations much higher than would be found in an ordinary sucrose-containing beverage. The taste of sucrose, together with optional flavoring agents, stimulates the salivary glands to respond in a few seconds with an increase in the flow rate. The volume of saliva will increase until a maximum value is reached. This maximum value is about 1.1 mL (ie, a normal swallow can be estimated to be 0.3 mL). The swallowing reflex is stimulated, and some of the sucrose is eliminated. The remaining sucrose is then progressively diluted by the saliva entering the mouth until the maximum volume is reached, triggering another swallow, and so on. After some time, the concentrations of sucrose and the optional flavoring agent reach such low levels that the stimulation of the glands decreases to an unstimulated state, which results in a slower clearance process, dependent on the unstimulated SSR. The time it takes to reach a given detectable low level has been used as a measure of clearance rate. Several variables are important for the clearance rate, the most important being the SSR and the volumes of saliva in the mouth before and after swallowing. A high SSR will result in rapid clearance, compared to the slow clearance obtained at low SSR. From the great differences in clearance rates, it is clear that caries risk increases enormously with a low SSR. The clearance rate is an individual property that is constant over time. However, if changes in health status cause a decrease in the SSR, a drastic change in clearance rate will ensue. The clearance rate also differs considerably at different sites, because of the complicated rheology of the oral cavity. The film overlying the mucous membrane and the teeth moves at varying rates, from 0.8 to 8.0 mm/min. In sites where the salivary film may be expected to move rapidly, for example, in proximity to the ductal orifices, the clearance rate is considerably greater than in sites where the saliva is stagnant (eg, the buccal areas of the maxillary anterior teeth and mandibular molars), which may explain in part the pattern of caries on different teeth and tooth surfaces. Sucrose in saliva and in the salivary film diffuses readily into dental plaque. A few minutes after sugar intake, the plaque will be overloaded with sucrose, with a greater sugar concentration than is present in the saliva. Provided that the plaque is not too thick to impede accessibility of the saliva, the flow of sucrose will be reversed. Therefore, there is a correlation between pH changes in the plaque and the salivary clearance of sucrose. In contrast to rapid clearance, slow clearance resulting from limited salivary accessibility will cause steep Stephan curves (see chapter 2). After sucrose rinsing, the pH fall in approximal plaque on molars is much more severe in the center of the surface than it is lingually, because the central area is inaccessible to saliva for dilution and buffering of the plaque acids. Neutralization and buffering of acids Although while the effect of saliva in facilitating sugar clearance can partly explain why saliva reduces formation of plaque acids and therefore caries, the neutralizing and buffering actions of saliva are more dramatic. These are due predominantly to salivary bicarbonate, originating mainly from the parotid gland. In unstimulated saliva, the bicarbonate level is low; at the greater secretion rates of stimulated saliva, the concentration is higher, the pH rises, and the buffering power of saliva increases

dramatically. There are also other less important buffering systems in saliva, such as macromolecular proteins. Ingestion of sugar causes a drop in plaque pH. When saliva is experimentally prevented from entering the mouth (by cannulating the excretory ducts and discharging the saliva extraorally), the fall in plaque pH after ingestion of sugar is greater and more prolonged than when salivary access is normal. If, after ingestion of sugar, flow is stimulated by chewing of paraffin or cheese, the plaque exhibits an immediate and dramatic rise in pH and a fall in lactic acid concentration, accompanied by a change in its amino acid spectrum. Similar effects are seen with sugar-free chewing gum and even with sucrose-sweetened gum, provided that this is chewed for longer than the time it takes for the sugar to be dissolved. Although the plaque of caries-resistant patients and the plaque of caries-susceptible patients respond similarly to a sugar challenge, the levels at which these responses occur are quite different. In the plaque from a caries-resistant person, the presugar pH is higher and the fall in pH after the sugar challenge is smaller. Studies have also shown that the capacity to buffer plaque acids is greater in caries-resistant patients than it is in caries-susceptible patients. The buffering effects of saliva are mostly measured in vitro by laboratory methods or chairside methods. In the laboratory, 1.0 mL of saliva is mixed with 3.0 mL of hydrochloric acid (0.0033 M for resting saliva; 0.005 M for stimulated saliva). A stream of air is then passed through the mixture for 20 minutes and the pH (the "final pH"), is measured. If the air stream step, which removes carbon dioxide, is excluded, about the same results are obtained for saliva with low buffering effect, final pH 5 or lower. Chairside tests are available, allowing the clinician to evaluate the salivary buffering effect directly after sampling and to discuss the results with the patient. In the Dentobuff Strip system (Fig 98a to 98c), one drop of stimulated saliva is placed on a test strip containing an acid and a pH indicator. After the reaction between saliva and acid, the color of the test pad is compared to a chart, and the final pH value is obtained. This test is highly simplified and will discriminate among low, medium, and high buffering capacities. The method is particularly useful for identifying individuals with risk values, that is, low buffering capacity (final pH of 4 or less). As with secretion rates, there is a normal range of buffer capacity, with no apparent relation to caries risk. However, below a threshold value (final pH less than 4), the carious process seems to be facilitated. Figure 99 shows the frequency distribution of the buffering effects in males and females, taken from the previously described salivary study in adults by Heinze et al (1983); more females had low values (pH less than 4.0) for both resting and stimulated saliva. Notably, other studies have shown a dramatic reduction in salivary buffering effects during the last months of pregnancy, which may explain in part why caries incidence seems to increase during pregnancy. On a population basis, there is a positive correlation between SSR and buffering effect, but there are many individual exceptions. A low SSR combined with a low or moderate buffering effect clearly indicates poor salivary resistance to microbial

attack: Clearance of microorganisms is slow, and the residual saliva, ranging in various individuals from 0.5 to 1.0 mL, is spread as a thin film on the oral surfaces. Fermentable carbohydrates dissolved in this small volume of saliva would be neutralized only slowly, because of the low buffering effect. The interpretation of salivary buffering tests in isolation is questionable. In most investigations, there is little or no correlation with variables measuring different aspects of dental caries. One important explanation is that the decisive events in a carious attack take place in the plaque and below the enamel surface. In these loci, the buffering mechanisms are very different from those found in saliva. It is unlikely that salivary buffering substances could significantly influence pH changes in the depth of the plaque, particularly in areas of limited accessibility, for example, the approximal surfaces of the molars. The buffering capacity of the plaque may have greater relevance, but test methods are as yet unavailable. On more accessible mandibular lingual surfaces covered with only a thin plaque, the salivary buffering effect may play a more significant role as a modifying factor in lesion development. The human mouth is quite frequently exposed to agents that have a pH different from that of saliva (6.5 to 7.5) and are potentially damaging to the teeth (erosion) or to the mucosa. Under these conditions, the role of the buffering agents in saliva is to restore the pH to the normal range as quickly as possible. Demineralization and remineralization of tooth surfaces The physicochemical integrity of dental enamel in the oral environment is entirely dependent on the composition and chemical behavior of the surrounding fluids: saliva and plaque fluids. The main factors governing the stability of enamel apatite are pH and the free active concentrations of calcium, phosphate, and fluoride in solution, all of which can be derived from the saliva (see Box 12). The development of a clinical carious lesion involves a complicated interplay between a number of factors in the oral environment and the dental hard tissues. The carious process is initiated by bacterial fermentation of carbohydrates, leading to the formation of a variety of organic acids and a fall in pH. Initially, H+ will be taken up by buffers in plaque and saliva; when the pH continues to fall (H+ increases), however, the fluid medium will be depleted of OH- and PO34-, which react with H+ to form H2O and HPO24-. On total depletion, the pH can fall below the critical value of 5.5, where the aqueous phase becomes undersaturated with respect to hydroxyapatite (HA). Therefore, whenever surface enamel is covered by a microbial deposit, the ongoing metabolic processes within this biomass result in fluctuations in pH and occasional steep falls in pH, which may result in dissolution of the mineralized surface. The role of the saliva in this process is highly dependent on accessibility, which is closely related to the thickness of the plaque (for review see Pearce, 1991; Tenovuo, 1997). Caries versus erosion. As discussed earlier, dissolution of enamel can result in the development of either a carious lesion or an erosive lesion. Caries is defined as the result of chemical dissolution of the dental hard tissues, caused by bacterial degradation products, that is, acids produced by bacterial metabolism of lowmolecular weight sugars in the diet. Erosion is defined as chemical dissolution of

tooth substance caused by any other acid-containing agent. Mixed lesions may well exist, particularly when the dentin has been exposed by erosion, causing hypersensitivity, which may lead to inadequate plaque control and, subsequently, to caries. This condition occurs frequently on exposed root surfaces. The appearance of the two lesions differs. The carious lesion is characterized by a subsurface demineralized lesion body, covered by a rather well-mineralized surface layer. In erosion, the surface has been etched away layer by layer, and there is no subsurface demineralization. Under normal conditions, in the absence of thick undisturbed plaque and/or very high frequency of acidic dietary products, teeth do not dissolve in saliva, because it is supersaturated with calcium, phosphate, and hydroxyl ions, which constitute the mineral salts of the tooth. The degree of supersaturation is even greater in plaque, especially in its extracellular fluid phase, which is in direct contact with the tooth surface. In addition, in individuals who have a regular daily source of fluoride (eg, fluoride toothpaste), both the saliva and the plaque fluid should contain an abundance of fluoride ions. In the dynamic equilibrium of the carious process, the supersaturation of saliva provides a barrier for demineralization and a driving force for remineralization. This equilibrium is greatly affected by fluoride, which reduces demineralization and enhances remineralization. Salivary saturation is overcome only when the plaque pH falls so far that the hydroxyl and phosphate ion concentrations are reduced below a critical value (through conversion of PO43- to HPO42- and H2PO4-). In principle, dental enamel can be dissolved under two different chemical conditions. When the surrounding aqueous phase is undersaturated with respect to hydroxyapatite and supersaturated with respect to fluorapatite (FA), HA dissolves and FA is formed. The resulting lesion is a carious lesion in which the dissolving HA originates from subsurface enamel and FA is formed in the surface enamel layers. The higher the supersaturation with respect to FA, the more fluoride is taken up in the enamel surface, the better mineralized the surface enamel layer becomes, and the less demineralized is the subsurface body of the lesion. On the other hand, if there is undersaturation with respect to both HA and FA, both apatites dissolve concurrently, and layer after layer is removed. This will result in an erosive lesion. Fresh acidic fruit, fruit juices, acidic carbonated soft drinks, and some champagnes are all unsaturated with respect to both apatites and are able to cause erosive demineralization of the teeth. These mechanisms for enamel dissolution are illustrated in Figure 100. Role of calcium. In these processes, the most important inorganic ions are calcium, phosphate, and fluoride. Calcium is a bivalent ion excreted, together with zymogen proteins, into the lumen of the acinus. Therefore, the concentration of calcium found in the saliva is dependent on the SSR. Going from an unstimulated state to a "somewhat stimulated" state, the calcium concentration decreases somewhat. However, the excretion pattern is complicated by the different calcium concentrations found in the secretions from different glands: the concentration in the submandibular or sublingual fluid is about twice as high as that in the parotid saliva. As the proportion of parotid secretion in the total volume of saliva increases with stimulation, the resulting flow pattern in whole saliva is a linear increase related to the calcium concentration.

Depending on the pH, calcium is distributed in saliva in ionized and bound forms. The free, ionized calcium is especially important in the carious process, because it participates in establishing the equilibrium between the calcium phosphates of the dental hard tissue and its surrounding liquid. At pH values close to normal, the ionized calcium constitutes approximately 50% of the total calcium concentration, but it increases if salivary pH is lowered. At pH values below 4, most of the salivary calcium is in ionized form. The nonionized calcium is distributed on a diversity of ligands with association constants in a large range; that is, the calcium is more or less firmly bound to inorganic ions such as inorganic phosphate, bicarbonate, and fluoride (10% to 20% of the total calcium concentration, depending on pH and SSR), to small organic ions (less than 10%) such as citrate, and to many macromolecules (10% to 30%). Some salivary macromolecules have been attributed a special role in oral calcium homeostasis. The influence of SSR on the distribution of calcium on free and bound fractions is complex. As already pointed out, the pH of saliva is strongly dependent on the SSR, as are the concentrations of most of the calcium-complexing substances. At low SSR, the bicarbonate concentration is very low, with a correspondingly low concentration of the calcium bicarbonate complex. The tooth is usually separated from the saliva by an intermediate layer of integuments, in the form of a pellicle or a plaque. The total calcium concentration in these compartments is slightly higher, sometimes much higher, than in the saliva, because of a high concentration of binding sites for calcium and because of the presence of precipitated calcium salts. There is a strong correlation between both total and ionized calcium in saliva and dental plaque, showing a flow of calcium over the plaque-saliva interface following existing diffusion gradients in ionized calcium. This gradient will be large after sugar intake, liberating bound calcium; as the plaque pH slowly increases, the concentrations of ionized calcium in saliva, pellicle, and plaque will slowly reach an equilibrium. Role of inorganic phosphate. The inorganic orthophosphate in saliva consists of phosphoric acid (H3PO4) and the primary (H2PO4-), secondary (HPO42-), and tertiary (PO43-) inorganic phosphate ions. The concentrations of these ions are dependent on the pH of the saliva. The sum of the ions and the molecule constitutes the total phosphate concentration. The lower the pH, the less the concentration of the tertiary ion, implying that the ion product of hydroxyapatite decreases considerably with decreasing pH. This phenomenon is the main cause of demineralization of the teeth. As with calcium, it is evident that the content of inorganic phosphate in saliva is a prerequisite for the stability of the tooth mineral in the oral environment. The concentration of total inorganic phosphate decreases with increasing SSR. As is the case for calcium, the different glands differ in phosphate excretion: the phosphate concentration in the submandibular glands is only about one third that in parotid saliva, but is about six times higher than that in the minor mucous glands, the glands nearest the tooth surfaces. Therefore, it may be assumed that the inorganic phosphate concentration shows a large variation in the microenvironment.

About 10% to 25% of the inorganic phosphate, depending on pH, is complexed to inorganic ions such as calcium or is bound to proteins. A small part, less than 10%, is in the dimer form, pyrophosphate (H4P2O7), which is a potent inhibitor of the precipitation of calcium phosphate and influences the formation of calculus. This is the rationale for the inclusion of pyrophosphate in toothpastes intended to inhibit calculus formation. However, the prevalence and incidence of caries in individuals who exhibit rapid formation of salivary calculus tend to be less than average. In a recent 3-year, longitudinal, double-blind study of fluoride toothpaste in more than 4,000 11 to 12 year olds, those with salivary-derived calculus developed significantly fewer new carious surfaces than did the others (Stephan et al, 1994). Among adult patients maintained for several years in a needs-related preventive program and with very high standards of oral hygiene, a subselection of participants with very rapid formation of salivary-derived calculus had more intact tooth surfaces than did patients with little or no calculus formation (Axelsson et al, 1995, unpublished). The inorganic phosphate of saliva has several important biologic functions, the most important from a caries aspect being its contribution to solubility products with respect to calcium phosphates and thus its role in the maintenance of the tooth structure. Its minor role in salivary buffering has already been discussed. Role of fluoride. Fluoride in the fluids surrounding the enamel crystals has been shown to have the potential to reduce the rate of demineralization. When present in the liquid phase of remineralization, fluoride will be incorporated into the enamel crystals and the enamel will become more resistant to demineralization (Fig 103). Fluoride has also been shown to reduce acid production in dental plaque. Therefore, in caries-preventive programs, the aim of fluoride administration should be to ensure that fluoride levels in the oral fluids are adequate to prevent and inhibit caries. The fluid bathing a plaque-covered tooth surface consists of saliva, plaque fluid, and the fluid surrounding the enamel crystals and is sometimes influenced by the crevicular fluid. These fluids constitute a continuous system, and ions will diffuse according to their concentration gradients. Fluoride introduced into the oral cavity will be distributed in saliva and thus influence the fluoride concentration in plaque fluid and enamel crystal fluid. Fluoride is present in saliva in concentrations that depend on fluoride in the environment, especially in drinking water. Other important sources are fluoride toothpastes and other fluoride products used for caries prevention and control. In areas with low concentrations of fluoride in the drinking water (below 10 umol [0.2 ppm]), the basal concentration of fluoride in whole saliva is usually less than 1 uM. The concentration may be much higher in areas with higher water fluoride concentrations. After an intake of fluoride, the levels of fluoride in the blood increase, reaching a peak after 30 minutes to 1 hour. The fluoride enters the saliva by simple diffusion over the membranes of the acinar cells. The concentration of fluoride in the duct saliva will therefore follow the plasma values, at a 30% to 40% lower level. This results in an increase of fluoride concentration in whole saliva, although only 0.1% to 0.2% of the ingested fluoride is excreted via the salivary glands.

In some foods and beverages, the fluoride is mainly in ionized form, which readily dissolves in the saliva; in others, fluoride may be firmly bound, making it difficult to predict the resulting fluoride concentration after exposure. This variable should be taken into account in the formulation of caries-preventive topical agents: For example, fluoride tablets and fluoride chewing gum have very different solubility rates. In addition, some slow-release fluoride agents, such as fluoride varnish, may contain up to 2% to 5% fluoride; glass-ionomer cements may be intermittently reloaded with fluoride, and as a consequence, release fluctuating amounts of fluoride. Because the oral cavity contains only a small volume of saliva in a thin film, even if only very small amounts of fluoride dissolve in the residual saliva, the resulting concentration may be very high. For example, if a fluoride tablet of 0.25 mg is dissolved in 1 mL of saliva, the resulting fluoride concentration is about 13 mol (about 200 ppm), more than 10,000 times higher than the basal fluoride concentration. Even higher fluoride concentrations could be expected in loci close to the fluoride source. For example, if a fluoride tablet is placed on one side of the oral cavity, very large differences in salivary fluoride concentration are found between the exposed and unexposed sides of the mouth (Sjogren et al, 1993). Therefore, slowly dissolving fluoride tablets and fluoride chewing gum should be moved around the mouth continuously to distribute fluoride to as many microenvironments as possible, and slow-release fluoride agents should be applied to key-risk teeth and key-risk surfaces. The high initial fluoride concentration in the salivary film after fluoride exposure will establish a concentration gradient between the dental integuments and the plaque. Fluoride will diffuse from saliva into the pellicle and the plaque, rapidly elevating the concentration of fluoride in the plaque fluid. Mineral calcium fluoride (CaF2) may form in saliva, in the pellicle, and in plaque fluid (see Fig 101) (Ogaard et al, 1983a, b). The limiting factor for the formation of CaF2 is the calcium content of the oral fluids. Therefore, the use of fluoride chewing gum after every meal as a combined salivastimulating and fluoride agent, resulting in increased calcium release from the saliva, fluoride release, and increased buffering effect, offers a rational, self-administered measure for caries control during or just after the fall in pH. Calcium fluoride releases fluoride slowly (Fig 102). Fluoride diffusing into microorganisms also prevents participation of the enzyme enolase in the glycolytic pathway by binding magnesium, essential for optimal function of the enzyme. However, this will not occur on plaquefree tooth surfaces. After the initial exposure to fluoride, the salivary concentration of fluoride decreases rapidly, by the same mechanisms involved in sugar clearance. The most important factor for the fluoride clearance rate is, as for sugar, the salivary secretion rate, which is dependent on the degree of stimulation. Fortunately, fluoride clearance is significantly slower in patients with hyposalivation than in individuals with normal SSR. Clearance varies markedly at different sites in the oral cavity, is generally more rapid from lingual than from buccal sites, and is most rapid beneath the tongue. However, there are some important differences between salivary fluoride clearance and salivary sugar clearance. The saliva contains a certain basal level of fluoride, which results in a gradual, theoretically asymptomatic decrease of fluoride to the

basal level. This slow adaptation is often prolonged for several reasons. First, swallowed fluoride will partly reenter the saliva, increasing the amount of fluoride, but the effect on the fluoride concentration is probably minor. Second, after a few minutes, the fluoride concentration in pellicle and plaque fluids is higher than it is in the saliva, causing the concentration gradient to reverse direction. Some fluoride will therefore diffuse back from the pellicle into the saliva. Third, after the fluoride concentration in the pellicle has fallen to a level that makes the fluid undersaturated with respect to calcium fluoride, this salt may start to dissolve slowly, increasing the ionized fluoride concentration. This last factor is complicated by its dependency on the pH of the pellicle, because at pH values in the normal range, calcium fluoride dissolution is inhibited by adsorbed phosphate ions. When pH approaches 5, this coating of the calcium fluoride particles vanishes (see Fig 102). When pH rises again, phosphate and protein-coated CaF2 is reformed in the pellicle (Fig 103). Antimicrobial and other protective properties The saliva contains many different proteins and some other small organic proteins that together protect the oral cavity (the soft tissues as well as the teeth) from frictional wear, dryness, erosion, pathogenic bacteria, and so on (see Box 12). Lubrication and other protective properties. Almost all salivary proteins are glycoproteins; that is, they contain variable amounts of carbohydrates linked to the protein core. Glycoproteins are often classified according to their cellular origin and subclassified on the basis of their biochemical properties. A characteristic feature is that many occur in multiple forms, constituting families; these families, may, however, exhibit remarkable functional differences. Mucous glycoproteins, the mucins, are of acinar cell origin, have a high molecular weight, and contain more than 40% carbohydrate. The mucins are produced by the minor salivary glands in the palate and provide a nonfrictional, lubricant layer that protects the soft tissues from wear and tear and facilitates swallowing of food. Because the mucins have a strongly negative charge, other negatively charged molecules, such as those contained in the cell walls of many oral bacteria, are repelled from the mucin-coated oral mucosa. Among other properties, the mucins also bind water and thereby protect the oral mucosa from drying out. Serous glycoproteins have a much lower molecular weight than mucins and contain less than 50% carbohydrate: Many belong to a group called proline-rich glycoproteins (PRPs), of which several are phosphorylated. These proteins are secreted from the parotid and submandibular glands. The collective name glycoprotein refers to all carbohydrate-linked proteins, making this group very heterogenous and large. Most salivary proteins, such as secretory IgA, lactoferrin, peroxidases, and agglutinins, belong to this group. Because human saliva is supersaturated with respect to most calcium phosphate salts, some proteins are necessary to inhibit their spontaneous precipitation in the salivary glands and their secretions. Such proteins include statherin and PRPs. The resulting stable but supersaturated state of the saliva with respect to calcium phosphate salts constitutes a

protective and reparative environment of importance for the integrity of the teeth. Statherin is present in both submandibular and parotid salivas. Proline-rich proteins form a complex group of proteins with large numbers of genetic variants, some of which also have the ability to inhibit spontaneous precipitation of calcium phosphate salts. The molecular size of PRPs ranges from 106 to 150 amino acid residues. Like statherin, PRPs are remarkable for their high degree of compositional and charge asymmetry. Proline-rich proteins are readily adsorbed from saliva to hydroxyapatite surfaces and it is most likely that these adsorbed PRPs inhibit the crystal growth of calcium phosphate salts. Although present in whole saliva, PRPs are also susceptible to proteolytic degradation by oral microorganisms. α -Amylase is one of the most important salivary enzymes, accounting for as much as 40% to 50% of the total salivary gland-produced protein. Most (80%) is synthesized in the parotid glands and the remainder in the submandibular glands. The biologic role of salivary amylase is to split starch into maltose, maltotriose, and dextrins. Maltose can be further fermented by oral bacteria. Therefore, although amylase in saliva clears starch-containing food debris from the mouth, acids are formed in this process. In this way, starch may have some cariogenic potential. Salivary α -amylase is inactivated in the acidic parts of the gastrointestinal tract, and therefore its action is limited to the oral area. Antimicrobial properties. As described earlier, saliva plays a significant role in maintaining an appropriate balance within the ecosystem associated with tooth surfaces. This balance is of great significance in the control of dental caries, because saliva will enhance the ability of some bacteria to survive and will reduce the competitiveness of others. Saliva achieves this control over the oral flora through its components, which can be constantly present or activated by a specific host response. The major antimicrobial proteins are listed in Box 13. Many studies have shown that most of these proteins can inhibit the metabolism, adherence, or even the viability of cariogenic microorganisms in vitro (for review see Tenovuo, 1997). However, their role in vivo is largely unknown: It seems that they are important for the control of microbial overgrowth in the mouth, but their selectivity against pathogens has not been determined. A newly proposed biologic function for PRPs is the ability of adsorbed acidic PRPs to selectively mediate bacterial adhesion on tooth surfaces. Recently it was shown that the negative charge of these acidic PRPs binds electrostatically to calcium on the tooth surfaces, while the outer ends, consisting of proline and glutamine amino acids, attract and bind very strongly to the harmless and protective normal microflora of the teeth (Streptococcus oralis, Streptococcus sangius, and Streptococcus mitis). This may explain early scanning electron micrographs obtained by Lie (1978), showing how a gram-positive "pioneer colonizer" attaches to the pellicle-covered tooth surface, in contrast to a gram-negative bacterium and the pellicle (Figs 104 and 105). This primary colonization of the protective normal microflora occurs during the first 24 hours after cleaning. However, recent research has shown that the so-called secondary colonization by other, more pathogenic microorganisms (gram-positive as well as gram-negative) is strongly related to the binding between galactose amine

structures on the surfaces of the normal microflora as well as the secondary colonizers. The production and the individual structures of acidic PRPs and galactose amines are genetically related and may partly explain individual variations in plaque formation rates. This is a field of ongoing research (Stromberg, 1996). The lysozyme in whole saliva is derived from the major and minor salivary glands, gingival crevicular fluid, and salivary leukocytes (polymorphonuclear neutrophil leukocytes). Salivary lysozyme is present in newborn babies at levels equal to those of adults, suggesting a preeruptive antimicrobial function. The classic concept of the antimicrobial action of lysozyme is based on its muramidase activity, ie, the ability to hydrolyze the bond between N-acetylmuramic acid and N-acetylglucosamine in the peptidoglycan layer of the bacterial cell wall. Gram-negative bacteria are more resistant to lysozyme because of the protective function of the outer lipopolysaccharide layer. In addition to its muramidase activity, lysozyme is strongly cationic, and can activate bacterial "autolysins," which can destroy the cell wall components. Lactoferrin is an iron-binding glycoprotein secreted by the serous cells of the major and minor salivary glands. Polymorphonuclear leukocytes are also rich in lactoferrin and release it into gingival fluid and whole saliva. The biologic function of lactoferrin is attributed to its high affinity for iron and its consequent expropriation of this essential metal from pathogenic microorganisms. This bacteriostatic effect is lost if the lactoferrin molecule is saturated with iron, a factor that should be taken into account in areas where the drinking water is rich in iron. In its iron-free state (apolactoferrin), it has a bactericidal, irreversible effect against a variety of microorganisms, including mutans streptococci. Apolactoferrin can also agglutinate Streptococcus mutans cells. Salivary peroxidase is produced in the acinar cells of the parotid and submandibular glands but not in the minor salivary glands. Salivary peroxidase systems have two major biologic functions: (1) antimicrobial activity and (2) protection of host proteins and cells from hydrogen peroxide toxicity. Salivary agglutinins are glycoproteins that have the capacity to interact with unattached bacteria, resulting in clumping of bacteria into large aggregates that are more easily flushed away by saliva and swallowed: the term aggregation is therefore often used synonymously with agglutination. Listed in Box 13 are salivary proteins with agglutinating capacity. The most potent agglutinin is a high-molecular weight glycoprotein that has been isolated from human parotid saliva. Despite a concentration in parotid saliva of only 0.001%, it is very effective. Mucins are also able to agglutinate bacteria. In high-molecular weight glycoproteins, sugar residues and sialic acid are important for the interaction with bacteria. The secretory immunoglobulins, most notably secretory IgA, act by aggregating bacteria. They target specific bacterial molecules, such as adhesins, or enzymes, such as glucosyl transferase. Studies of the correlation between secretory IgA levels and caries prevalence have reported conflicting results (Riviere and Papaginnoulis, 1987). The saliva also contains IgG and IgM from serum and local production in the gingival tissues.

The conflicting results of recent longitudinal clinical studies of the relative predictive values of antimicrobial salivary components for caries incidence (Tenovuo et al, 1997) may be attributed to the fact that dental caries is a multifactorial disease. Fig 97 Physiologic model of the oral cavity, demonstrating the clearance of sucrose by saliva. C1= volume of saliva that has an upper level before swallowing (Vmax) and a lower level after swallowing (Vresid). C2= volume of integuments and plaque on teeth and mucosal membrane.(From Lagerlof and Oliveby, 1990. Reprinted with permission.) Figs 98a to 98c Chairside testing of the salivary pH with the Dentobuff Strip system. One drop of stimulated saliva is placed on a test strip containing an acid and a pH indicator. After the reaction between the saliva and acid is completed, the color of the test pad is compared to a chart, and the pH value is determined. Fig 99 Frequency distribution of the buffering effects of saliva (final pH) in males and females. (From Heinze et al, 1983. Reprinted with permission.) Fig 100 Mechanisms for enamel dissolution at different pH challenge and saturation of minerals. Fig 101 Tooth surface after topical fluoride treatment (pH 7.0). Fig 102 Tooth surface during cariogenic challenge (4.5 < pH > 5.5): Undersaturated with hydroxyapatite; supersaturated with fluorapatite. Fig 103 Tooth surface after cariogenic challenge (pH 7.0). Fig 104 Attachment of a gram-positive (G+) pioneer colonizer to the pellicle-covered tooth surface (white arrow) in contrast to the relationship between a gram-negative bacterium (G-) and the pellicle (black arrows). (From Lie, 1978. Reprinted with permission.) Fig 105 Close-up of the attachment (arrows) between a gram-positive bacterium (B) and the pellicle (P). (From Lie, 1978. Reprinted with permission.) Formation and functions of pellicle Saliva is seldom in direct contact with the tooth surface but is separated from it by the acquired pellicle, defined as an acellular layer of salivary proteins and other macromolecules, approximately 10 um thick, adsorbed onto the enamel surface. It forms a base for subsequent adhesion of microorganisms, which under certain conditions may develop into dental plaque. The pellicle layer, although thin, has an important role in protecting the enamel from abrasion and attrition, but it also serves as a diffusion barrier.

Figure 106 shows that the undisturbed pellicle is formed in different layers. There are many nonattaching bacteria close to the outer surface of the pellicle. Because of abrasion, for example from toothbrushing, the thickness will vary between 2 and 10 um, depending on the toothbrushing intervals. Saxton (1976) showed that complete removal of the pellicle requires about 5 minutes' cleaning with pumice in a rotating rubber cup. Figure 107 shows a groove made in the pellicle with a knife. Such grooves were earlier thought to be abrasive defects in the enamel surface resulting from the use of abrasive toothpaste. The pellicle shown in Figure 108 was removed by intensive cleaning for about 5 minutes. The pellicle-free enamel surface was partly covered with nail varnish, while the outer part was exposed to saliva in vivo for several hours. Figure 108 shows the thickness of the new pellicle compared to the naked enamel surface after removal of the nail varnish. Movement of molecules by forces other than diffusion is less frequent in the pellicle than in most other parts of the salivary film. The relatively undisturbed layer of liquid in the pellicle influences the solubility behavior of the enamel surface. Adsorption to the enamel of macromo- lecules, usually originating from the saliva, is selective; certain macromolecules show a higher affinity for the mineral surface than do others. In the normal oral pH range, the enamel surface has a negative net charge, because of the structure of hydroxyapatite, in which phosphate groups are arranged close to the surface. Count -erions (of opposite charge), for example, calcium, are attracted to the surface, forming a hydration layer of unevenly distributed charges. The exact composition of this layer will be determined by several factors (eg, pH, ionic strength, and the type of ions present in the saliva). Normally the hydration layer close to the enamel surface contains mainly calcium and phosphate ions in the proportion of 10:1, but other ions, such as sodium, potassium, fluoride, and chloride must also be present (see the formation of phosphate- and protein-coated CaF2 crystals in the pellicle; see Figs 101, 102, and 103). Because of the domination of calcium, the resulting net charge of the enamel surface with its hydration layer is positive, implying that the hydration layer will attract negatively charged macromolecules, as illustrated by Waerhaug more than 25 years ago (Figs 109 and 110). Negative charges on macromolecules are found in acidic side chains with end groups of phosphate or sulfate. These side chains have a high affinity to the tooth surface. Recent research has shown that the bulk of the pellicle consists of salivary micellelike structures of great importance for reducing diffusion through the pellicle and reducing friction between the teeth and other oral tissues. Not all the proteins contributing to the pellicle are well defined. However, salivary proteins, such as amylase, lysozyme, peroxidase, IgA, IgG, and glycosyltransferase, participate in formation of the pellicle matrix, together with mucins and breakdown products from macromolecules of both salivary and bacterial origin. Of special interest are the acidic PRPs, mentioned earlier, which bind via their amino-terminal segments to the tooth surface, leaving their carboxy-terminal regions directed toward the oral cavity, where they may interact with oral microorganisms.

During the first hour, pellicle formation is rapid, and then decreases. It seems likely that adsorption of the first layer of molecules onto a clean surface is instantaneous. The formation rate varies among individuals, probably as a result of differences in salivary composition, the frequency of oral hygiene, and diet composition. Fig 106 Undisturbed pellicle in cross section. It is formed in different layers. Many nonattaching bacteria (left) are close to the outer surface of the pellicle (right). (From Saxton, 1976. Reprinted with permission.)

Fig 107 Groove in the pellicle, down to the enamel surface. The groove was made with a knife. (From Saxton, 1976. Reprinted with permission.)

Fig 108 Thickness of new pellicle (right) compared with a naked enamel surface that had been protected by nail varnish (left). The groove shown in Fig 107 was removed by 5 minutes' intensive cleaning. The pellicle-free enamel surface was partly covered with nail varnish (left), while the other part (right) was exposed to saliva for several hours. The surface is shown after removal of the nail varnish. (From Saxton, 1976. Reprinted with permission.) Fig 109 Negatively charged salivary macromolecules. (Illustrated by J. Waerhaug. Courtesy Department of Periodontics, University of Oslo.) Fig 110 Negatively charged salivary macromolecules attached to the positively charged enamel surface. (Illustrated by J. Waerhaug. Courtesy Department of Periodontics, University of Oslo.)

Salivary stimulation and substitution in patients with hyposalivation and xerostomia Stimulation of saliva Recognition of the key role of saliva in maintaining normal oral function has stimulated research on its protective properties against caries and on the treatment of xerostomia and salivary hypofunction. Salivary clearance, buffering power, and degree of saturation with respect to tooth mineral are the major protective properties (for review, see Sreebny et al, 1992; Tenuvuo, 1997), their effect increasing with salivary stimulation: The saliva stimulated by consumption of fermentable carbohydrates reduces the fall in plaque pH that could lead to demineralization and increases the potential for remineralization. When saliva is stimulated after carbohydrate intake, acids produced in the plaque are neutralized, and experimental lesions in enamel are remineralized. The pH-raising effects are more easily explained by the buffering action of stimulated saliva than by clearance of carbohydrates. Remineralization is dependent on the presence of fluoride.

These findings suggest that the protective properties of saliva can be potentiated by appropriate salivary stimulation. In addition to established procedures, such as diligent oral hygiene and fluoride regimens, general recommendations for caries prevention might therefore include eating patterns that stimulate secretion of saliva. There are now a number of management options for protecting the oral cavity from the devastating effect of inadequate salivary function and for relieving the patient's discomfort. Treatment is determined by the degree of functional impairment. For the patient who has some remaining glandular function, stimulation of secretion is the optimal approach. Patients with negligible natural function can be offered symptomatic treatment to relieve oral dryness. For either patient category, selection of specific treatment measures is determined by a number of factors, including the patient's medical status. The practitioner must also be able to manage the complications of salivary hypofunction: increased incidence of caries, oral candidiasis, altered oral function, and pain. Stimulation of secretion, locally or systemically, has the great advantage of providing the benefits of natural saliva. Systemic. There has been increasing interest in systemic pharmacologic stimulation of salivary function. Three agents have been studied in some detail: bromhexine hydrochloride, anethole trithione, and pilocarpine hydrochloride. All three should be used only under specialist supervision and following medical examination. Bromhexine is a mucolytic agent used in the management of chronic bronchitis. Its use in managing dryness of the eyes associated with Sjogren's syndrome is controversial. No beneficial effects on salivary dysfunction have been demonstrated. Anethole trithione has been proposed as a treatment for salivary hypofunction caused by psychotropic drugs, radiation, and Sjogren's syndrome. Conflicting results have been reported on the efficiency of treatment. In one study, 74% of patients with Sjogren's syndrome had increases in the output of unstimulated whole saliva, whereas Swedish studies, on patients with more pronounced salivary dysfunction, have failed to show any improvement in salivary function. Patients with postradiation xerostomia showed no improvement following drug treatment, compared to controls. Further trials are necessary to delimit the ability of this drug to improve salivary function(for review see Pearce, 1991; Edgar et al, 1994). Pilocarpine hydrocholoride is a parasympathomimetic drug that functions primarily as a muscarinic-cholinergic agonist with mild β -adrenergic stimulatory properties. It has been used for more than 100 years as a potent stimulant of exocrine secretion. In the last decade, carefully controlled studies have shown that it can increase salivary output in normal volunteers and effectively relieve oral dryness in patients with salivary gland hypofunction. In a 6-month trial by Fox et al (1986) in patients with irradiation-induced salivary hypofunction and in patients with Sjogren's syndrome, 5 mg of pilocarpine, three times a day, was effective; side effects were well tolerated; and there were no significant alterations in heart rate, blood pressure, or electrocardiographic parameters. Greenspan and Daniels (1987) reported that pilocarpine treatment resulted in subjective and objective improvement in about 80% of patients with postradiation xerostomia. A synergistic effect on salivary stimulation from a combination of pilocarpine and anethole trithione was reported by Epstein and Schubert (1987).

Although pilocarpine appears to be the most effective systemic sialagogue currently available, it is of limited use in the management of salivary hypofunction. It is ineffective if insufficient functional tissue remains, as in advanced stages of Sjogren's syndrome or following head and neck radiotherapy. Possible interactions with other medications or potential adverse cardiovascular and pulmonary effects further limit patient eligibility. Further clinical studies are necessary to determine optimal doses, administration schedules, and systemic effects of pilocarpine. For the patient with an irreversible condition requiring long-term or lifelong management of dry mouth, a sustained-acting preparation would be ideal. Local. Local stimulation is feasible, because the salivary glands are highly responsive to stimulation from taste, masticatory activity, and the sensory nerves of the mucosa and periodontium. Because the salivary secretion rate usually increases during meals (a physiologic response), an important first step to potentiate salivary flow should be a diet of fiber-rich, well-flavored aromatic food, such as fruit. In addition, finishing a meal with matured cheese (for instance, cheddar) has been shown to increase SSR and decrease plaque pH significantly (Imfeld, 1983). In developed societies, the reduced masticatory activity required to chew highly processed modern foods may favor a measure of salivary hypofunction, because of disuse atrophy. Studies in both animals (Johnson and Sreebny, 1982) and humans (Axelsson et al, 1997a; Dodds et al, 1991; Jenkins and Edgar, 1989) have indicated that prolonged increases in levels of salivary stimulation after dietary changes result in greater salivary output. Adequate water intake is another prerequisite for normal salivation. For optimal prevention and control of dental caries, every meal containing easily fermented carbohydrates, particularly sucrose, should be followed immediately by supplementary local salivary stimulation to increase the sugar clearance, buffering effect, plaque pH, and the access of calcium and phosphate ions. As described earlier, demineralization is thereby decreased and remineralization is enhanced. Local saliva-stimulating agents that contain fluoride will further enhance the remineralization potential significantly and should therefore be recommended in preference to similar agents without fluoride. Specially formulated lozenges and chewing gum, both with and without fluoride, are available. Because frequent intake is recommended (four to six times per day, directly after meals and snacks), it is important that these agents not contain sugar and not be potentially erosive. The sweetening agents commonly used in these products are sorbitol, xylitol, and saccharin, separately or in combination. Fluoride lozenges containing 0.25, 0.50, 0.75, and 1.00 mg of fluoride are also commercially available. The 0.25-mg lozenges are recommended for children older than 5 years and 0.25 to 1.00 mg lozenges are recommended for selected young adults and adults, particularly in cases of hyposalivation, for combined salivary stimulation and fluoride delivery directly after meals. A recent study by Sjogren et al (1995) showed that in subjects with reduced SSR, fluoride lozenges and chewing gum not only improved the SSR, but also significantly prolonged fluoride clearance time, an important factor in prevention and control of caries in such patients, who are generally at high caries risk.

To date, the most promising system for local salivary stimulation is the recently introduced fluoride chewing gum (Fluorette and Fludent, used widely in Scandinavia). It is sugarless and sweetened with xylitol and sorbitol. Each piece contains 0.25 mg of fluoride. Chewing gum is unique because it is usually chewed for a prolonged periodaround 30 minutesbut its caloric value is almost negligible, both important characteristics in the context of salivary stimulation. Chewing gum has been shown to elicit a continued flow of saliva during prolonged mastication (Dawes and Macpherson, 1992), but the level of stimulus gradually declines: As the flavoring agents are released and swallowed, the gustatory stimulatory component is rapidly depleted, and the intensity of the masticatory stimulus abates, because of softening of the gum (Rosenhek et al, 1993). The effect of varying not only the duration of chewing, but also the interval elapsing between the caries challenge (drop in pH) and the subsequent gum chewing, has been studied. For maximum neutralization of the plaque pH, the gum must be chewed for at least 15 minutes, immediately following the caries challenge (Park et al, 1993). Recently, Sjogren et al (1997) compared the effect of chewing fluoride gum for 5, 10, 15, 20, 30, or 45 minutes on approximal plaque pH and salivary fluoride concentration on the chewing and nonchewing sides. The subjects had undisturbed approximal plaque, 3 days old, and rinsed for 1 minute with 10 mL of 10% sucrose solution. The resultant fluoride concentrations were two to three times higher on the chewing side than on the nonchewing side (Fig 111). The best recovery in approximal plaque pH was also noted on the chewing side, but the difference was not as pronounced as for the salivary fluoride concentration. Significantly higher values of plaque pH were found during prolonged chewing (Fig 112), while variations in chewing duration caused only minor variations in salivary fluoride concentration. This study showed that to attain optimal fluoride- and plaque pH-raising effects throughout the entire dentition, the gum should be chewed for at least 20 minutes, using both sides of the mouth. Studies in patients with low salivary flow rates (Abelson et al, 1990; Markovic et al, 1988) also showed that use of a sorbitol-sweetened gum raised the pH of the plaque on both enamel and root surfaces. Two principal mechanisms are implicated in the plaque pH-raising effects of chewing gum or foods such as cheese: clearance of carbohydrates and buffering of plaque acids. For chewing gum, the relative contributions of these two factors were studied by Dawson (1993). After a sucrose mouthrinse, the test subjects chewed sugar-free or sucrose-sweetened gum; the control subjects did not chew any gum. Over 45 minutes, SSRs, pH, sugar concentrations, and bicarbonate levels were measured. The results were compared with plaque pH data noted under similar conditions (Manning and Edgar, 1993). Both gums had pH-raising effects on plaque. The sugar-free gum accelerated clearance of the sucrose rinse, while the sugared gum released sugar, at potentially acidogenic concentrations, throughout the 45-minute period. The gums had a similar effect on salivary secretion rates, but the sugared gum resulted in lower levels for salivary pH and bicarbonate, indicating active participation of salivary bicarbonate in neutralizing and buffering acids produced in the mouth from the sugars released from the gum. Thus, the buffering effect was capable of overwhelming the acids formed

from sugars derived from sucrose-sweetened gum. If the bicarbonate level of saliva is thus paramount in eliciting the pH-raising action of chewing gum (and, by analogy, other pH-raising foods), then it is important to note that the relationship between salivary flow rate and bicarbonate concentration is not linear, but approaches a maximum at an intermediate flow rate. Thus, increasing salivary flow above this rate would not be expected to result in a comparable increase in pH-raising action on plaque, contrary to what would be obtained if the major plaque pH-controlling factor were salivary clearance of sugars or acids. This point merits further study with various levels of stimulation of saliva and observation of the effects on plaque pH. Sugar-free chewing gum containing urea (V6) is available in Europe and Scandinavia, and is claimed to exert plaque pH-raising effects superior to those of gum without added urea, presumably because of the increased synthesis of ammonia in plaque resulting from ureolysis. Recently, Imfeld et al (1995) used the telemetric technique to compare the effect of chewing xylitol or xylitol-carbamide (urea) gum on approximal plaque pH after a sucrose rinse. Figure 113 shows the extraordinary effect of the carbamide-containing gum compared to the xylitol gum and no chewing gum. Together, these studies offer convincing evidence that, following consumption of fermentable carbohydrates, the chewing of sugarless gum rapidly elevates plaque pH toward resting levels, where it persists for the duration of the experiment. In addition to causing an increase in the buffering power by stimulating saliva, the chewing increases bicarbonate levels leading to an increase in salivary pH and thus to the degree of supersaturation of stimulated saliva with respect to calcium phosphate solids, including hydroxyapatite. The increase in the degree of supersaturation with calcium and phosphate leads to the conclusion that stimulated saliva can influence the equilibrium between demineralization and remineralization in the development of caries, not only by reducing the duration of demineralization resulting from the pH changes in plaque but also by enhancing the potential for remineralization. This hypothesis was tested by Leach et al (1989). An intraoral device carrying a piece of partially demineralized human enamel, covered with gauze to encourage deposition of plaque, was attached to a mandibular first molar tooth in volunteers. The subjects used a sorbitol-sweetened chewing gum for 20 minutes after three meals and two sugary snacks each day. After 3 weeks, the enamel particle was replaced, and the subjects consumed the same meals and snacks but without using gum. The order (gum versus no gum) was reversed in half the subjects. The subjects continued to use their usual fluoride-containing dentifrice thoughout the study. Analysis of the mineral content of the carieslike lesions after intraoral exposure showed a significantly greater increase in the mineral content after gum chewing than was found without gum chewing, indicating a potentially beneficial remineralizing effect of the stimulated saliva. The increase in remineralization after the use of gum could have occurred either because of a reduction in the degree of demineralization via the effect of gum chewing on plaque pH or because of an increase in remineralizing potential. These results should not be interpreted as indicating that early white-spot lesions in enamel can necessarily be completely remineralized through the use of sugar-free gum. Rather, the data indicate the possibility of

favorable alteration of the equilibrium between demineralization and remineralization, preventing the development of an initial carious lesion. In such studies of salivary stimulation, the environment in which remineralization of the enamel lesion occurs is the fluid phase of plaque, which differs from saliva in pH and concentration of calcium and phosphate ions. While an increase in the supersaturation of saliva resulting from stimulation would be expected to have a direct effect on the supersaturation of plaque fluid, to date no such effect has been demonstrated. Sternberg et al (1992) found, however, that chewing of sorbitol and xylitol gum, besides reducing plaque accumulation and gingival inflammation, increased the concentration of acid-extractable calcium in plaque by more than one third. This would be expected to increase the remineralizing potential of the plaque, and it was suggested that the effect was due to complexing of calcium by both xylitol and sorbitol, leading to their retention in plaque. Equally, it may be that the elevation of plaque pH that would have followed gum chewing resulted in increased retention of calcium in plaque in the form of insoluble calcium phosphate deposits. Fluoride chewing gums should increase the reservoir of phosphate and protein-coated CaF2 crystals in the pellicle, as well as in any plaque that might remain. In the remineralization studies already described, a therapeutic fluoride environment was provided by the use of a fluoride dentifrice. The essential role of fluoride in the remineralizing potential of sugared gum was shown in preliminary data. Subjects chewed sucrose gum for 20 minutes after meals and snacks for successive 21-day periods, during which the fluoride content of the dentifrice was varied between 0 and 1,000 ppm. In the presence of fluoride, some remineralization was observed, although it was statistically insignificant; with the nonfluoridated dentifrice, significant demineralization occurred (Manning and Edgar 1993). In caries-susceptible patients with impaired salivary secretion, use of the new sugarless fluoride chewing gum directly after meals is therefore a promising adjunctive measure. Compared to subjects with normal salivary flow, patients with reduced salivary flow have a fluoride clearance time that is fortuitously prolonged, as shown in the aforementioned study by Sjogren et al (1993). The effect of a fluoride sugar-free chewing gum on stimulated salivary secretion rate, Plaque Index, Gingival Index, and Plaque Formation Rate Index, as well as on salivary mutans streptococci scores, was recently evaluated. The selected group of patients (n = 53) had less than 0.7 mL/min stimulated SSR (mean = 0.4 mL/min). The subjects were instructed to chew a stick of fluoride chewing gum (0.25 mg of fluoride) for 15 to 20 minutes after every meal (four to six times per day) for 6 months (Axelsson et al, 1997a). The results showed an increase in the mean stimulated salivary secretion rate from 0.4 to 0.6 mL/min. This finding is important because it shows that through regular salivary stimulation with chewing gum after every meal, a gradual increase in salivary secretion is possible. An average reduction of about 35% was achieved for Plaque Index, Gingival Index, and Plaque Formation Rate Index (Fig 114). The percentage of subjects with salivary mutans streptococci scores of 0 to 3 at baseline and after 6 months is shown in Fig 115. There was a pronounced shift from high scores to low; in particular a decline in score 2 and an increase in score 1 (Axelsson et al, 1997a). These results indicate that regular use of a fluoride chewing gum after every meal

would have a very significant caries-controlling effect in patients with hyposalivation, over and above the primary effect of the fluoride release from the chewing gum. In two recent studies on experimentally induced caries (enamel and root lesions), the remineralizating effect of fluoride chewing gum (0.1 mg of fluoride) used five times per day for 21 days was compared with an in situ slow-release fluoride device that released 0.5 mg of fluoride/day. A nonfluoride toothpaste was used for oral hygiene three times a day during the test period. The degree of remineralization of enamel lesions was 35.5% for the chewing gum and 34.0% for the slow-release device (Wang et al, 1993). On root lesions, De los Santos et al (1994) achieved similar results (36.0% and 35.8% for the gum and the device, respectively) for remineralization. However, the chewing gum resulted in higher stimulated SSR than did the fluoridereleasing device and the control (2.1, 1.8, and 1.7 mL/min, respectively) and higher mean salivary fluoride concentration during stimulation (3.0, 0.2, and less than 0.02 ppm of fluoride, respectively). The potential clinical effect of saliva stimulation per se has not been tested in a clinical trial, although it is possible to interpret results from certain clinical studies as effects of salivary stimulation. Thus, the caries-preventive effects of xylitolincluding apparent reversals of carious lesions (remineralization) shown in the Turku chewing gum trial, when sucrose- or xylitol-sweetened gum was chewed ad libitum over a 12-month period (Scheinin et al, 1975)could be due to the enhanced remineralizing potential, although inhibition of plaque acidogenicity and other effects of xylitol cannot be excluded. More conclusive indications of a beneficial effect of salivary stimulation are disclosed in studies by Moller and Poulsen (1973) in which chewing of sorbitol gum was associated with a small but significant reduction in caries incidence; by Isokangas et al (1989), in which significant long-term benefits of xylitol gum, used two or three times daily were shown; and by Kandelman and Gagnon (1990), in which a decrease of 65% in caries progression was found in children who chewed xylitol gum as part of a preventive program. It should be noted that the study designs did not stipulate routine postprandial chewing for 20 minutes, which would have maximized the influence on saliva. The only direct comparison of xylitol and sorbitol gums appears to be the recently published 2-year study by Makinen et al (1996), in which South American children chewed sorbitol or xylitol gum daily. Compared with controls not using any gum, the observed reduction in caries incidence was greater for xylitol than for sorbitol gum, while an increase in caries incidence occurred in the children using sucrose gum. The caries onset risks for xylitol and sorbitol pellet chewing gum were 35% and 44%, respectively of that in the non-gum group. The effects were greater with pellet gums than with stick gums and with increasing frequency of gum chewing. Both xylitol and sorbitol mixtures in pellet form were associated with a caries onset rate comparable with that of the xylitol stick gum. The largest reduction in caries risk was observed in the group receiving xylitol pellet gum. Thus, the protective action of saliva, essential for normal dental health, can be enhanced by stimulation through appropriate dietary manipulation and selection. Sugar-free chewing gum may be of particular value in stimulating salivation over a

prolonged period without increasing the energy content or acidogenicity of the diet. With the use of such gum after meals and snacks, for normal gum-chewing periods of 20 minutes or more, the effects of fluoride in favoring remineralization may be enhanced, and the benefits of salivary neutralization, buffering, and sugar clearance in opposing demineralization may be mobilized, as part of a program of prevention against caries. The use of gum not only as a salivary stimulant but also as a vehicle for preventive agents may further extend its potential applications. Recently, a new chewing gum containing chlorhexidine (10 mg per stick, available in Scandinavia) has shown significant antiplaque effects (Smith et al, 1996; Tellefsen et al, 1996) comparable to 0.2% chlorhexidine mouthrinse. It seems possible that in the near future a combined chlorhexidine-fluoride chewing gum will become available. Meanwhile, concurrent chewing of one stick each of the fluoride and the chlorhexidine chewing gum for 20 minutes directly after every meal could be recommended for high-caries-risk patients with reduced salivary secretion rate, high or very high Plaque Formation Rate Index (score 4 or 5) and high salivary mutans streptococci levels. In this way, chemical plaque control, fluoride, and salivary stimulation will act synergistically at the crucial time; during the acid attack. In addition, because the stimulatory effect is related to the volume of the stimulating agent, an even greater effect on salivary stimulation should be achieved if twice the amount of gum is chewed. Symptomatic therapy In the absence of natural salivation, it is essential to try to protect the oral hard and soft tissues by salivary substitution. Saliva substitutes, also called artificial salivas, are frequently recommended for patients complaining of dry mouth (xerostomia). Although many studies suggest that saliva substitutes are useful in the management of xerostomia, clinical experience has shown that these products are not well accepted by patients. Most patients do not continue to use the substitutes regularly, relying instead on water or other fluids to relieve their symptoms. One reason may be that most saliva substitutes are more viscous than natural saliva and may be uncomfortable for an individual with dry mucosal surfaces. Another reason may be that the need for frequent application to keep the mouth moist makes these substitutes inconvenient and expensive. Also, the artificial salivas fail to provide the broad spectrum of antimicrobial and other protective functions of natural saliva. There is a pressing need for more effective saliva substitutes and better delivery systems. Meanwhile, frequent sips of water or other fluids for the relief of oral dryness are often as effective as saliva substitutes. Patients should be advised to carry fluids with them at all times. (The water bottles used by cyclists or plastic glasses with snap-on lids are convenient.) Often, this simple suggestion will bring substantial relief at minimal cost, will improve mucosal hydration, and ease swallowing and speaking. Individuals could (and should) be cautioned to avoid not only fluids containing sugar but also those containing alcohol or caffeine, as these too may worsen the xerostomia or increase the risk of caries. A common complaint is dryness and cracking of the lips. If applied regularly, petroleum jelly-based compounds may be helpful. Patients may prefer lanolin-

containing creams, which help hydrate the tissues. Patients should be advised to use room humidifiers, especially at night, as an aid to relieving frequent symptoms of dryness of the throat and tongue. For institutionalized patients, demented, and other severely handicapped patients, a recently introduced aid is Saliswab (available in Europe), which acts as a combined salivary substitute and stimulating agent. In contrast to Lemon-Glycerin Swabs, it is not erosive. The practitioner must be prepared to manage the complications of salivary hypofunction: increased caries, oral candidiasis, altered oral function, and pain. Initially, patients with xerostomia do not have extensive restorative treatment needs because it takes some time for clinical caries to develop. Therefore, it is important to diagnose impaired salivary function and xerostomia as early as possible and introduce intensive needs-related preventive programs before caries has developed. In patients who have already developed several carious lesions, restorative treatment should be carried out in stages, beginning with excavation of caries and placement of provisional restorations using slow-release fluoride materials, such as glass-ionomer cements or resin-modified glass-ionomer cements, combined with an initially intensive, individually tailored preventive program. Once carious activity is under control, the next stage is definitive therapyrestorations in the form of complete crowns and fixed partial dentures. Most patients with severely impaired salivation and xerostomia should be regarded as lifelong high-caries-risk patients; they must continue on an intensive maintenance preventive program. Fig 111 Mean fluoride concentrations on the chewing and nonchewing sides after use of fluoride chewing gum, by duration of chewing. *=P 1 million to < 10,000 CFU/mL. 4. The lactobacilli count was reduced from >500,000 to