| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
J Am Dent Assoc, Vol 138, No 1, 39-46.
© 2007 American Dental Association |  |
COVER STORY |
 |
ABSTRACT |
|---|
 TOP ABSTRACT SUBJECTS, METHODS AND MATERIALSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Methods. The authors conducted cross-sectional analyses of dietary data collected from the Iowa Fluoride Study using three-day diaries for subjects at ages 1, 2, 3, 4 and 5 years and for subjects aged 1 through 5 years according to dental caries experience at 4.5 to 6.9 years of age. They categorized foods and beverages as containing NME or intrinsic/milk sugars.
Results. Subjects’ total, NME, food NME and intrinsic/milk sugars intakes at ages studied did not differ between subjects with and without caries experience. Beverage NME sugars intakes at age 3 years predicted caries (P < .05) in logistic regression models adjusted for age at dental examination and for fluoride intake.
Conclusions. Dental caries is a complex, multifactorial disease process dependent on the presence of oral bacteria, a fermentable carbohydrate substrate and host enamel. A simple NME-intrinsic/milk sugars categorization appears insufficient to capture the complex dietary component of the caries process.
Clinical Implications. Cariogenicity is more likely a function of the food and/or beverage vehicle delivering the sugar and the nature of exposure—that is, frequency and length of eating events—than of the sugar’s categorization.
Key Words: Sugars; caries; diet
Abbreviations: DMFT: Decayed, missing, filled teeth • IFS: Iowa Fluoride Study • NMES: Nonmilk extrinsic sugars.
Historically, the role of sugars in the dental caries process has been supported by numerous epidemiologic and animal studies.1–5 However, relationships among sugars and dental caries in contemporary society appear weak, as reviewed by Zero6 and Burt and Pai.7 Woodward and Walker2 compared available sugars intakes data and decayed-missing-filled-teeth (DMFT) scores from July 1979 to June 1991 for both developed and developing countries. They reported that on a population level, sugars intakes were associated with DMFT in developing countries, but not in developed countries. The 2003 report Diet, Nutrition and the Prevention of Chronic Diseases by the World Health Organization and the Food and Agriculture Organization of the United Nations8 concluded that there is convincing evidence associating the quantity of free sugars with dental caries, noted that the frequency of sugars intake could be just as important as the amount of sugar, and stated that "no conclusion regarding the relative importance of these two variables [could] be drawn."
Increases in the quantities and changes in the types of sugars consumed could affect the risk of dental caries. Per capita energy intakes from sugars have increased in developing countries during the past 30 years.8 In the United States, per capita consumption of high-fructose corn syrup increased 10.6-fold between the 1970s and 2000, sucrose consumption decreased 33 percent between the 1960s and 2000 and total sugars consumption increased 33 percent.9
In 1989, the British Department of Health established a Committee on Medical Aspects of Food Policy to report on relationships among dietary sugars and both oral and systemic health.10,11 This committee categorized dietary sugars for oral health purposes as "nonmilk extrinsic" (NME) or "intrinsic" on the basis of the sugars’ location in terms of cellular structure and the degree to which they are processed. Nonmilk extrinsic sugars (NMES) are located outside of cells, being either released (as in juice) or added (as in confections) during processing. Intrinsic sugars occur naturally and typically reside within the cellular structure. Although lactose in milk resides extracellularly, it is considered less cariogenic than glucose, fructose, sucrose and maltose when in extrinsic form; therefore, milk sugars are not included in the NMES category. The rationale for this categorization is that foods with NMES are thought to be more cariogenic than foods with intrinsic or milk sugars.10–11 Investigations in Great Britain using this categorization system, including reports of relationships with dental caries, have been few11–13; to our knowledge, this system has not been used in the United States.
We hypothesized that children with dental caries have higher intakes of NMES, particularly from beverages, than do children without caries. In this article, we describe young children’s intakes of NME and intrinsic sugars, and we relate those intakes to dental caries experience.
Data collection.
Dental caries experience.
Dental examinations were conducted in the General Clinical Research Center at The University of Iowa or in one of several community locations by trained and calibrated examiners (including J.J.W.) using standardized, portable equipment.14,15 The examination was visual, but examiners used dental explorers to confirm questionable findings.15 Transillumination (using the DenLite system [Welch Allyn Medical Products, Skaneateles Falls, N.Y.]) augmented the visual and tactile examination.
For this study, we defined caries experience as the presence of at least one cavitated (d2–3) or filled surface. We initially categorized lesions as noncavitated (d1) or cavitated (d2–3).15 Specifically, d1 lesions manifested as distinct, chalky white enamel with no clinically visible or irreversible loss of enamel structure or break in the enamel surface. In contrast, d2–3 lesions manifested as demonstrable loss of enamel structure. The criteria did not differentiate between cavitated enamel (d2) and dentinal (d3) lesions.15
Sugars intake.
We mailed parents three-day food and beverage diaries when their children were 6 weeks of age; 3, 6, 9 and 12 months of age; every four months through 3 years of age; and every six months thereafter until the child was 9 years of age, though we used only the data through the age of 5 years. (The data span the years 2001 through 2004.) In these analyses, we used diaries completed when children were 1 (n = 631), 2 (n = 526), 3 (n = 447), 4 (n = 415) and 5 (n = 411) years of age (as described in Warren and colleagues14,15). If the subject did not return a specific three-day food and beverage diary (for example, the 24-month diary), then we substituted the previous diary (for example, the 20-month diary). If this diary also was missing, then we substituted the subsequent diary (for example, the 28-month diary) for the yearly diary. If neither was available, we omitted the subject for that year. Inclusion in area-under-the-curve analyses (that is, a weighted average of the one-through five-year intakes) required a minimum of four diaries, including the one- and five-year diaries (n = 400). We calculated the area under the curve by using the trapezoidal rule (as described in Marshall and colleagues16 and Marshall and colleagues17).
We asked parents to record the types (including brand names) and quantities of all foods and beverages consumed by their children for one weekend day and two weekdays. Parents received written instructions on procedures, including how to measure and record quantities of foods using household measures (such as ounces or cups) or package units (such as "small serving fast-food fries," "can of pop"). Research assistants who were registered dietitians reviewed diaries for completeness. They also coded and verified all entries from the three-day food and beverage diaries (as described in Marshall and colleagues16 and Marshall and colleagues17). We calculated weighted averages based on weekend or weekday consumption to reflect average consumption across a week (as described in Eichenberger-Gilmore and colleagues18). We used these data to create a food and beverage intake table.
We created a nutrient table from nutrient data obtained from the U.S. Department of Agriculture’s Agriculture Research Service (USDA Nutrient Database for Standard Reference, Release 1228), the Minnesota Food and Nutrient Database (Nutrition Cooordinating Center NDS-R, Version 4.01; University of Minnesota, Minneapolis) and data from food and beverage manufacturers (as described in Marshall and colleagues16 and Marshall and colleagues17). We used a relational database software package (Microsoft Access, Version SR-1 [Microsoft, Redmond, Wash.]) to link the food and beverage intake table and the nutrient table to calculate individual sugar and energy intakes.
We categorized foods and beverages as containing NMES if the foods were highly processed (for instance, 100 percent juice) or contained added sugars (for instance, fruit pie, ice cream, jam, breakfast cereals, juice drinks, yogurts sweetened with sugars). We subsequently categorized NMES as being contained in food or beverages. Foods and beverages were categorized as containing intrinsic/milk sugars if the foods were minimally processed (such as fresh, frozen or dried fruit) or of dairy origin (such as milk) and did not contain added sugars. For clarity, we refer to sugars as "NMES" or "intrinsic/milk" throughout this article.
Fluoride intake.
We used IFS questionnaires completed for children aged from 6 weeks through 5 years to estimate fluoride intakes from water consumed as a beverage and water added during preparation of beverages and selected foods (such as pasta, soup, hot cereal), other beverages, dietary fluoride supplements and fluoride dentifrices.19–21 We analyzed the fluoride content of the nonmunicipal water in homes and child-care locations, of filtered municipal water and of beverages available for purchase by the subjects in our study.22–25 These values were used to calculate subject-specific water, beverage and select food (that prepared using home water) fluoride concentrations. We defined fluoride intakes as the sum of fluoride from water, other beverages, select foods, fluoride supplements and dentifrices. We estimated cumulative fluoride intakes (in milligrams per day) from fluoride intakes in children aged from 6 weeks through 5 years using the trapezoidal method (that is, summation of trapezoidal areas under the line connecting sequential time points).
Statistical analyses.
We conducted analyses using SAS (Version 9, SAS Institute, Cary, N.C.). We categorized subject characteristics, which are presented as percentages. We used the Cochran-Mantel-Haenszel test to identify demographic differences among subject groups. Daily sugars intakes are presented as medians (25th, 75th percentiles). To compare sugars intakes between subjects with and without caries, we used the Mann-Whitney U test. Logistic regression models helped us predict caries experience from sugars intakes while adjusting for age at dental examination and fluoride intake. We used the sign test to identify significant changes in sequential yearly trends in intakes of sugars. We considered a P value of less than .05 to be statistically significant.
Subjects with caries experience had mothers and fathers who were younger, had lower levels of education and had lower incomes than did those without caries (data not shown, P < .05). Similarly, subjects in the IFS cohort who did not participate in dental examinations or provide regular dietary records had younger parents with lower levels of income and less education than did active participants (data not shown, P < .05).
Table 1
SUBJECTS, METHODS AND MATERIALS
TOP
ABSTRACT
SUBJECTS, METHODS AND MATERIALS
RESULTS
DISCUSSION
CONCLUSIONS
REFERENCES
Subjects.
Subjects initially were recruited from 1992 through 1995 for the Iowa Fluoride Study (IFS), a longitudinal investigation of fluoride intakes, dental fluorosis and dental caries.14–27 Subjects who participated in one dental examination between the ages of 4.5 and 6.9 years of age and whose parents and/or caregivers completed periodic IFS questionnaires and three-day food and beverage diaries are the focus of this report (n = 634). The institutional review board at The University of Iowa approved all components of the IFS; research assistants obtained written informed consent from mothers at recruitment and from either parent at the examination.
RESULTS
TOP
ABSTRACT
SUBJECTS, METHODS AND MATERIALS
RESULTS
DISCUSSION
CONCLUSIONS
REFERENCES
Demographic characteristics of subjects (n = 634) and their parents at enrollment have been reported in detail elsewhere.17–19 Subjects were 51.9 percent female (n = 329) and 42.7 percent firstborn (n = 271). At the time of recruitment (1992 through 1995), 27 percent of household incomes were less than $30,000, 39 percent were between $30,000 and $49,999, 31 percent were $50,000 or greater, and 3 percent were unknown. Seventeen percent of mothers were aged between 16 and 24 years, 32 percent between 25 and 29 years, 31 percent between 30 and 34 years and 20 percent older than 35 years. Seven percent of fathers were aged between 16 and 24 years, 27 percent between 25 and 29 years, 33 percent between 30 and 34 years, 28 percent older than 35 years and 5 percent unknown. Nineteen percent and 25 percent, respectively, of mothers and fathers had an educational level of high school equivalent or less; 35 percent and 28 percent, respectively, had attended some college; and 46 percent and 40 percent, respectively, had a college degree or higher education. Approximately seven percent of fathers’ educational status was unknown. 
shows median (25th, 75th percentiles) daily intakes of NME and intrinsic/milk sugars at ages 1, 2, 3, 4 and 5 years and for 1 through 5 years of age according to caries experience at about 5 years of age (at some point between 4.5 and 6.9 years of age). Subjects with caries had total, NME, food NME and intrinsic/milk sugars intakes similar to those of subjects without caries at all ages. Subjects with caries had higher beverage NME sugars intakes at year three than did subjects without caries. Although not statistically significant at other ages, this trend was observable at all time points, most notably at age 5 years and at ages 1 through 5 years.
|
). Subjects with caries had sucrose, glucose and lactose intakes similar to those of subjects without caries at all ages. Although not statistically significant, the results suggest that subjects with caries had higher fructose intakes at ages 3 and 5 years than did subjects without caries.
|
We calculated median (25th, 75th percentiles) percentage of total energy intakes from sugars (data not shown). The percentage of energy from total sugars ranged from 30 percent (26, 35) at age 1 year to 27 percent (24, 31) at age 5 years for subjects without caries and from 32 percent (27, 36) at age 1 year to 27 percent (23, 31) at age 5 years for subjects with caries. The percentage of energy from NMES ranged from 10 percent (6, 15) at age 1 year to 19 percent (14, 23) at age 4 years for subjects without caries and from 9 percent (6, 14) at age 1 year to 21 percent (14, 25) at age 4 years for subjects with caries. Subjects with caries had percentages of energies from total, NME, food NME and intrinsic/milk sugars intakes similar to those of subjects without caries at all ages; the percentage of energy from beverage NMES was higher in subjects with caries (20 percent [15, 25]) than in subjects without caries (19 percent [14, 23]) at age 3 years.
We investigated yearly changes in median daily intakes of sugars by individual type (Figure 1) and classified the types as NME or intrinsic/milk (Figure 2). Sucrose and total sugars intakes increased each year from ages 1 to 4 years for subjects with and without caries (all P < .05) and then remained steady. Total NMES intakes increased each year from ages 1 to 4 years for subjects with caries, and from ages 1 year to 2 years for subjects without caries (all P < .05). Beverage NMES intakes increased from age 1 year to age 2 years for subjects with and without caries (P < .05). Intakes of solid NMES increased each year from ages 1 to 3 years for subjects with caries, and from ages 1 to 4 years for subjects without caries (all P < .05). Intrinsic/milk sugars intakes declined between ages 1 and 2 years for both groups and then remained steady.
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTSUBJECTS, METHODS AND MATERIALSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
NME and intrinsic/milk sugars intakes were similar at all ages for the subjects with and without caries, suggesting that dietary factors beyond the sugars source and extent of processing could be significant contributors to caries risk. Although we did not consider timing of consumption in our study, we previously showed that in the same study population, the cariogenicity of sugars- and starch-containing foods was higher when consumed more frequently (as with snacks) than when included with meals.19 Consumption of mixed foods containing fats, proteins and calcium at meals, perhaps through stimulation of saliva, could limit the cariogenicity of NMES.29–31
Our results support the observations of other investigators who concluded that factors other than NMES intakes were more important determinants of caries. Gibson and Williams12 analyzed data for children aged 1.5 to 4.5 years from the British National Diet and Nutrition Survey. They observed an association between the percentage of energy intake from NMES and dental caries only in children who brushed their teeth less than twice a day, suggesting that oral hygiene was more important than sugars consumption. Habibian and colleagues13 evaluated associations between dietary behaviors, oral hygiene and mutans streptococci in British children at 12 and 18 months of age. They found that early toothbrushing and the number of eating and drinking events were more closely associated with mutans streptococci than were sugars intakes.
However, in support of our hypothesis, after adjusting for age at dental examination and for fluoride intakes, we found that beverage NMES intakes at age 3 years predicted caries. Although this relationship was significant only at age 3 years, we observed similar trends at all ages. Previous reports by our group17 and Sohn and colleagues32 have shown that soda (that is, carbonated beverages) is more cariogenic than 100 percent juice;, we combined data regarding sugars from both of these beverages in our analyses, which potentially dilutes the effect of the greater cariogenicity of soda. After further investigation, we found that the frequency of soda consumption at snacks or meals was associated with caries to a greater extent than was the frequency of consumption of 100 percent juice.19 The cariogenicity of sugars found in soda could be enhanced by oligosaccharides present in the high-fructose corn syrup used to sweeten the beverage.33,34 Another explanation is that the cariogenicity of sugars found in 100 percent juice could be limited by the nonnutrient compounds (for example, phytochemicals) that that juice contains.
After 1 year of age, the percentage of energy from NMES in subjects in our study is higher than the 10 percent recommended by the World Health Organization.8 NMES intakes of subjects in our study are consistent with the added sugars intakes of preschool children in the United States (71 grams/day calculated for 18 percent total energy) reported by Kranz and colleagues,35 while our total sugars intakes are consistent with those (75–93 g/day) reported by Skinner and colleagues.36 The trends we observed in sugars intakes are consistent with changes in dietary patterns in early childhood. Decreases in lactose and intrinsic/milk sugars intakes between years one and two are consistent with a decreased consumption of infant formula and milk products as children make the transition to solid food. Similarly, the increased fructose intakes by the subjects in our study after 1 year of age are consistent with greater consumption of 100 percent juices in early childhood. Higher intakes of sucrose and food extrinsic sugars during the first three years of life are consistent with increased food acceptance and consumption of more confections.
Dietary methodology limitations of this study have been reported previously16,17 and are typical of self-reported data, including the possibility that reported intakes might not reflect actual intakes. Changes in dietary patterns resulting from preventive guidance provided by local health practitioners could limit our ability to identify sugars-caries relationships. Detection of caries by means of visual and tactile examinations and without the use of radiographs likely underestimated the caries experience. Defining caries as a cavitated or filled surface could overestimate caries experience if noncarious lesions were filled. Also, we defined caries as being simply present or absent; degrees of caries severity could have a different association with sugars categories. Subjects and their families are representative of a middle-to-upper socioeconomic class, and associations between sugars types and caries experience might be more pronounced in a more diverse socioeconomic group.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTSUBJECTS, METHODS AND MATERIALSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
FOOTNOTES |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTSUBJECTS, METHODS AND MATERIALSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  P. Stoodley, J. Wefel, A. Gieseke, D. deBeer, and C. von Ohle Biofilm Plaque and Hydrodynamic Effects on Mass Transfer, Fluoride Delivery and Caries J Am Dent Assoc, September 1, 2008; 139(9): 1182 - 1190. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. A. Morio, T. A. Marshall, F. Qian, and T. A. Morgan Comparing diet, oral hygiene and caries status of adult methamphetamine users and nonusers: A pilot study J Am Dent Assoc, February 1, 2008; 139(2): 171 - 176. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
