Obesity and Dental Caries: Systematic Review Obesidade E Cárie Dentária: Revisão Sistemática

OBJECTIVE: Identifying, through a systematic literature review, evidence of a possible association between obesity and dental caries. METHODS: A search of articles published between 2005 and January 2012 was performed in the Medline/PubMed, LILACS and Web of Science databases. The quality of scientifi c evidence of the selected articles was assessed by the items proposed for observational studies in the Downs & Black instrument. RESULTS: Initially, 537 references were found; after checking the titles and abstracts by two independent researchers, twenty-eight articles were selected for complete reading. Ten of them that assessed the primary and/or permanent dentition observed a positive association between obesity and dental caries and one study found an inverse association. According to the Downs & Black classifi cation, thirteen articles with good scientifi c evidence were found. CONCLUSIONS: The present review did not fi nd suffi cient evidence regarding the association between obesity and dental caries, and it did not clarify the possible role of diet and other possible effect modifi ers on this association. 800 Obesity and dental caries Silva AER et al Overweight and obesity, defined as excess body fat compared to lean body mass, 19 are a growing public-health problem in the world. 15 Decreasing physical activity, increasing sedentary lifestyles and dietary changes are factors strongly associated with the development of overweight and obesity. 12 Studies have observed increases in being overweight in childhood and adolescence since the beginning of 2000, 21 resulting in the increased risk of cardiovascular diseases, respiratory disorders and other chronic diseases during adulthood. 32 Diet plays an important role in the increased prevalence of obesity 8 due to the higher consumption of foods rich in fat and carbohydrates. 34 Overweight or obese children and adolescents reported higher consumption of sugary drinks and foods such as " fast food " compared with those who reported normal weight. 34 Besides being directly associated with obesity, eating habits, especially regarding the intake of sucrose, have a well-established causal relationship with tooth decay, along with socio-demographic aspects, such as low socioeconomic status. 27 A systematic review and meta-analysis 7 published in 2010 identifi ed a positive association between obesity and periodontal disease. However, the relationship between obesity and dental caries is unclear. Dental caries remains the most prevalent problem in oral health and it is associated with pain 4 and a decrease in quality of life.

Overweight and obesity, defined as excess body fat compared to lean body mass, 19 are a growing public-health problem in the world. 15Decreasing physical activity, increasing sedentary lifestyles and dietary changes are factors strongly associated with the development of overweight and obesity. 12Studies have observed increases in being overweight in childhood and adolescence since the beginning of 2000, 21 resulting in the increased risk of cardiovascular diseases, respiratory disorders and other chronic diseases during adulthood. 32et plays an important role in the increased prevalence of obesity 8 due to the higher consumption of foods rich in fat and carbohydrates. 34Overweight or obese children and adolescents reported higher consumption of sugary drinks and foods such as "fast food" compared with those who reported normal weight. 34Besides being directly associated with obesity, eating habits, especially regarding the intake of sucrose, have a well-established causal relationship with tooth decay, along with socio-demographic aspects, such as low socioeconomic status. 27systematic review and meta-analysis 7 published in 2010 identifi ed a positive association between obesity and periodontal disease.However, the relationship between obesity and dental caries is unclear.Dental caries remains the most prevalent problem in oral health and it is associated with pain 4 and a decrease in quality of life. 16

INTRODUCTION
Because caries is associated with poor dietary habits and inappropriate diets promote obesity, a biologically plausible association between obesity and dental caries has been inferred in the literature. 14However, the presence of a common factor such as a high-sugar diet, caused by the ingestion of food such as soft drinks, cake, and ice cream, seems to increase the probability of both diseases and seems to be the most acceptable theory to explain the relationship between obesity and dental caries. 21The biggest challenge in understanding this relationship is to measure the possible confounding variables (diet and income) and effect modifi ers (age, oral hygiene habits and use of fl uoridated water) in a standardized manner. 32systematic review 21 on the subject was indexed by Bireme, Medline, ISI (Thomson Scientifi c) and Cochrane Library in 2006, covering the age groups of childhood, adolescence, and adulthood.The exclusion criteria identifi ed 33 articles for reading.At the end of the review, three cross-sectional studies, all performed with children, were identifi ed as having good methodological quality.They presented confl icting results regarding the presence of a positive association between obesity and dental caries.Another systematic review and meta-analysis 17 including studies published before 2010 indicated the presence of a positive association between obesity and dental caries in permanent dentition without clarifying the direction of this association.

DESCRITORES: Dental Caries, Epidemiology. Obesity. Observational Studies. Review.
The present study deals with a systematic review in order to use the new evidence to update the literature between 2005 and January 2012 on the subject.

METHODS
A systematic review on obesity and dental caries was performed in the electronic databases of the Medline/ PubMed, LILACS, and the Web of Science, between 2005 and January 2012.This review included studies published after 2010, which were not included in the systematic review and meta-analysis published earlier on the subject, new scientifi c evidence to try to clarify the relationship between obesity and dental caries have been added to this study.The fi rst search used the English terms "dental caries" and "obesity".These two terms are the same as those used in the systematic review performed on the same subject by Kantovitz & Pascon; Rontani & Gavião 21 (2006) with articles published between 1984 and 2004.The following keywords were used for the defi nition of caries: "DMFT index", "tooth decay"; for the explanatory variable the keywords "waist circumference", "skinfolds" and "body mass index" were used.These descriptors were found in the articles selected in the fi rst search conducted on the subject.For the combination of descriptors, the logical operators "AND" and "OR" were used.
The process of fi nding the scientifi c papers to write the review consisted of four steps.The fi rst was the identifi cation of titles in the databases PubMed, LILACS, and Web of Science.Exclusion criteria used in reading the titles consisted of: review articles, protocols on nutrition and food issues, studies on oral health and diet-specifi c items and obesity studies related to other oral diseases.
The second stage consisted of two researchers independently reading the eligible abstracts.The search also excluded those items that assessed participants aged six and under, articles published before 2004, articles evaluating specifi c populations such as schizophrenic patients, and master's dissertations or PhD theses.The few disagreements that occurred were discussed by the two researchers, who then reached a consensus.
The third stage consisted of reading the articles, applying the same exclusion criteria mentioned above.
In the fourth step, all the references of the selected articles were checked in order to fi nd any study not found in the databases previously consulted by the authors.
The quality of the scientifi c evidence in the selected articles was assessed using the criteria of Downs & Black 13 (1998).Of the 27 items proposed by the authors, 18 were selected.Those items evaluating intervention studies were excluded because none of the 28 articles selected for review had used this design.The items excluded from Downs & Black's classifi cation were: Are the interventions of interest clearly described?Does the study provide estimates of the random variability in the data for the main outcomes?Have all important adverse events that may be a consequence of the intervention been reported?Were the staff, places, and facilities where the patients were treated, representative of the treatment the majority of patients receive?Was an attempt made to blind study subjects to the intervention they have received?Was an attempt made to blind those measuring the main outcomes of the intervention?Was compliance with the intervention/s reliable?Were study subjects randomized to intervention groups?and Was the randomized intervention assignment concealed from both patients and health care staff until recruitment was complete and irrevocable?The articles with ten or more items scored by both evaluators in the review were found to have good scientifi c evidence.The criteria to consider an article as producing good evidence were given by the study authors.The choice of a cut-off point of ten or more items is due to the fact that the Article evaluated as having good scientifi c evidence had to have more than 50% of the items from the Downs & Black classifi cation positively evaluated by two researchers of the study.For cases in which there was disagreement about the scores of articles between the two researchers, i.e., one researcher considered the article to have good scientifi c evidence and the other does not, a new assessment was performed and a consensus reached without the need for a third appraiser's evaluation.

RESULTS
Initially, 346 titles were identifi ed in PubMed, 21 in LILACS, and 170 in Web of Science.After reading the titles, 81 were selected for reading the abstracts: 66 from PubMed, six from LILACS, and nine from Web of Science.Studies involving participants < 6 years old were excluded after reading the articles.Twenty-eight articles published between 2005 and 2012 were identifi ed.All the references were checked and six articles were found that did not meet the review's inclusion criteria (Figure).Among the 28 articles, 25 were in English, two in Spanish and one was in Portuguese.Twenty-four studies had a cross-sectional design, three were prospective cohort and one was a case control (Tables 1 and 2); Table 1 lists all studies that did not perform adjusted statistical analysis, while those with adjusted statistical analysis are shown in Table 2.
The ages evaluated ranged from three to 20.Although one of the exclusion criteria was age < 6 years, few studies have evaluated various age groups including children < 6 years, and therefore, they were included in the review.For those studies that performed separate analyses by age groups, only groups with participants > 6 years were included.
The measure of obesity in the 28 studies was Body Mass Index (BMI), and BMI information was self-reported in two of these studies. 9,10The classifi cation of BMI showed no uniformity.Nine studies used the International Obesity Task Force (IOTF) for age and sex and seven assessed BMI through percentiles developed in the United States to determine the population's standard weight (Center for Disease Control and Prevention -CDC 2000).Three studies used BMI for age and one used for age and weight.Finally, eight studies were also checked which had other methods for obtaining BMI.
Seventeen studies assessed decay through teeth or decayed, missing, or fi lled surfaces -DMFT or DMFS, or, by means of DMFT or DMFS, in case of the primary dentition.The dental caries were assessed by interproximal radiographs in two studies, through the teeth or decayed and fi lled surfaces in four, by means of decayed teeth and interproximal radiographs in one study, by decayed surfaces and interporximal radiographs in another study, and through surfaces or decayed teeth in two studies (Tables 1 and 2).
The positive association between obesity and dental caries in permanent teeth was identifi ed in six studies, a negative association was found in one, and eight showed no association.For studies evaluating mixed dentition (deciduous and permanent), four of them found a positive association, one found positive correlation, and seven showed no association.There was no correlation between obesity and caries in a study in which permanent dentition was analyzed (Tables 1 and 2).
3,36,40,41 Eight studies included adjustments for age and sex, 6,19,23,27,29,33,36,41 two only for sex, 15,22 and fi ve for consumption of any type of food, such as soft drinks, 10,24 fruits or snacks between meals, 10 "fast food", 10 and milk, 9 as well as daily sugar consumption 19 and "snack" consumption more than three times a day. 29ong the 28 articles, 13 were considered to have good scientifi c evidence according to the criteria of Downs & Black.Eight of them with adjusted analysis and fi ve articles without adjusted analysis were considered to be of good scientifi c quality (Tables 1 and 2).
Of these 13 articles, six found a positive association, showing that individuals with higher BMI values had more caries, and there was no association between obesity and dental caries in the other seven studies (Tables 1 and 2).

DISCUSSION
Between 2005 and 2012, 28 eligible articles which analyzed the relationship between obesity and dental caries were found.After applying the exclusion criteria for the revision and those proposed by Downs & Black, 13 thirteen studies were shown to have good scientifi c evidence.However, they did not shed light on the link between obesity and dental caries.
Unlike the systematic review by Kantovitz & Pascon; Rontani & Gavião 21 (2006), this review excluded studies on individuals < 6 years old, as it has been observed that the use of cariogenic foods have a greater impact on the appearance of dental caries increasing the likelihood of association between obesity and caries, after the age of six. 39Moreover, in early childhood there is greater parental control over food intake and oral hygiene practices, which can mean lower prevalence of both diseases. 39The opposite is also true.Parents that do not perform preventive oral health behaviors and food control, especially regarding the sugar consumption, increase the chances of caries in the primary dentition, and later in the permanent dentition, concomitantly with the presence of obesity. 5udies that evaluated the association between dental caries and obesity indicate a variety of factors to be discussed in order to understand this relationship.The fi rst issue refers to methods for the diagnosis of both diseases.The main diagnostic method for caries is the visual examination of the teeth or surfaces, and this method was used in most studies.However, this diagnostic method can often lead to lower levels of detection of the disease.Some studies using this method considered only the decayed teeth, not those restored or extracted due to caries. 6Other studies 1,2 used interproximal radiographs, a method more sensitive to caries diagnosis.Using this method is likely to identify higher prevalence of the disease by detecting caries at earlier stages of cavitation.Using this type of examination for diagnosis in epidemiological surveys is not suitable because it is an expensive method and is primarily used by exposing the study population to examination when there is no clinical indication of disease.
There are also aspects that can interfere with measuring obesity.All studies included in this review measured obesity by BMI.There was no study measuring obesity using other diagnostic methods such as skinfold thickness, waist circumference, waist/hip ratio, or x-ray densitometry (DXA), all of which are more accurate methods for identifying levels of obesity.The probable explanation for this is the ease of implementation, objectivity, and comparability with a BMI reference standard, especially in population studies. 14The limitation of this index, which does not differentiate between lean body mass and fat mass and therefore often produces false positive results, must be considered   prior to the study.The non-use of DXA as a more appropriate method to measure obesity is probably due to the examination cost, which often makes it impossible to use in population studies.
A second point refers to the population's access to oral health services and the use of fl uorinated substances.These two factors are considered as potential effect modifi ers that may lead to a weak or negative association between obesity and caries. 19However, these factors were treated as potential confounders in studies in this review.In the present work, most studies showing an association were performed in European countries where oral health services are organized and available to the public, in addition to various forms of administering fl uorinated substances being made available.The lack of association between obesity and dental caries was identifi ed mainly in Latin American countries.Four of these studies were conducted in Brazil at the following locations: Bauru, 35 Southeastern Brazil, with 6.8 to 8.5% of adolescents studied presenting overweight and obesity; in cities belonging to the state of Paraíba, Northeastern Brazil, 28 with 30% participants aged between 12 and 15 obese; in Londrina, 38 20 are considered important factors for both dental caries and obesity.Dietary habits related to obesity and low incomes may provide a higher quantity and frequency of sucrose intake, which are important factors in caries etiology. 35Higher fat intake and decreased fi ber may be associated with increased prevalence of caries and overweight/obesity among individuals with lower socioeconomic conditions. 3However, other authors reported that excessive consumption of fatty foods has less infl uence on the development of caries than diets rich in sugar, which could be somewhat responsible for the lack of association between obesity and the presence of caries observed in some studies. 39is review presents some limitations.A small number of studies (fi ve) analyzed longitudinal data.When considering only those studies with good scientifi c evidence according to the criteria of Downs & Black 13 (1998), four used longitudinal data.The use of longitudinal data including information on various points of time is important because obesity in early life often leads individuals to change their habits to reduce body weight, resulting in a change in BMI scores.However, the scores for caries by means of DMF can remain high because this index considers caries throughout the individuals' lives. 39Moreover, children identifi ed by parents as obese may be guided to restrict their sugar intake; however, if there is still a high consumption of fats and if their caloric intake is greater than expenditure, they will become overweight. 36e second limitation is the lack of adjustment in several studies for variables such as diet and socioeconomics.These two factors have been cited as associated with both diseases, and they should be considered in studies that aim to explain the relationship between obesity and dental caries.However, of the 28 articles examined, fi ve had some kind of adjustment for diet, and four adjusted for socioeconomic issues.
A third limitation involves using the Downs & Black instrument 13 (1998) to evaluate the quality of scientifi c evidence from the studies.This instrument was originally developed to evaluate intervention studies, but it has been used for studies with other designs, excluding some of their items. 26,30,31Despite this limitation, it was considered useful for the present review because of its ideas about article quality.
It was not possible to perform a meta-analysis.First, due to the heterogeneity of the studies, particularly the methods used to obtain exposure -obesity of this review, and according to non-submission of primary data by most authors of the articles considered as good scientifi c evidence.A meta-analysis would help clarify the relationship between obesity and dental caries.
This review identifi ed six studies with good scientifi c evidence, which observed a positive association between obesity and dental caries in children and adolescents, in contrast to what was described in the review by Kantovitz & Pascon; Rontani & Gavião 21 (2006).They noted three studies with good scientifi c evidence, and of these one with positive association in children.Although this review found six studies with a positive association between obesity and dental caries, the causal mechanism remains unclear.It is unknown if the associations actually occurred because there is a direct relationship between obesity and dental caries or if diet, a factor common to both diseases, is responsible for this association.None of the studies was designed to consider all the possible effect modifi ers (access to health services, use of fl uorides, and oral health habits) and confusion factors (diet and income) from the literature.These results were similar to the recent systematic review and meta-analysis, 17 although the present study had included articles from the literature published between 2011 and 2012.Thus, the studies identifi ed in the literature provide clues about this relationship and indicate the need for new and more clearly delineated studies.

1 .
There is difference only among those with BMI < 25 and BMI  30 regarding the decayed and fi lled surfaces (RO (crude) = 2.8 (1.0, 7.7) 20 years old:There is difference only among those with BMI <18.5 to 24.9 and BMI  30 regarding the decayed and fi lled surfaces (OR (crude) = There was no association between obese and normal weight regarding the DMFT (p> 0.05).2.Inverse correlation was observed between BMI and DMFT (r = -0.06,p = 0.41) after adjustment for gender ethnicity.

Figure .
Figure.Search for scientifi c articles for the systematic review on obesity and dental caries, 2005-2012.

Table 1 .
Review of studies on obesity and caries without adjusted statistical analysis, 2005-2012.

Table 2 .
Review of studies on obesity and caries with adjusted statistical analysis, 2005-2012.Body Mass Index; DMFT: decayed, missing and fi lled teeth; DMFS: decayed, missing and fi lled surfaces most of the population's needs.The lack of association in studies conducted in Latin America probably addresses socioeconomic and dietary conditions, different factors in countries of these two continents.Dietary habits and income