Is there an association between dental caries, fluorosis, and molar-incisor hypomineralization?

Abstract Objective: This cross-sectional study aimed to determine the prevalence of dental caries, dental fluorosis, and molar-incisor hypomineralization, and their associations in a group of Brazilian schoolchildren. Methodology: Adolescents (n=411) were evaluated by two calibrated examiners for dental caries (DC), dental fluorosis (DF), and molar-incisor hypomineralization (MIH) using the CAST (Caries Assessment Spectrum and Treatment) instrument, Thylstrup and Fejerskov (TF) index, and MIH Severity Scoring System (MIH-SSS), respectively. Descriptive statistics, chi-square tests, and logistic regression were used for statistical analysis. Results: The sample comprised 42.75% boys and 57.25% girls. The prevalence of DC in permanent dentition was 94.75%, of which 29% were represented by dentin lesions. For DF, a prevalence of 40.75% was observed, with 69.32% mild, 12.88% moderate, and 17.79% severe. A positive association between the source of water and fluorosis was detected (p=0.01). The prevalence of MIH was 18%. Thirty adolescents (41.7%) presented with severe MIH. No association was found between DF or MIH and dentin DC or between MIH and DF at the individual level. However, a significant negative relationship was detected between DF and dentin carious lesions ( p <0.005) and DF and MIH ( p <0.00001) at the tooth level, whereas a positive association was observed between MIH and dentin carious lesions ( p <0.00001). A positive association was also observed between the severity of both conditions ( p <0.00001). Mild DF was the most prevalent problem observed. Cases of teeth with mild MIH were the most predominant in MIH-affected teeth. Conclusions: No association was observed among the dentin carious lesions, MIH, and DF at the participant level. However, a positive association between MIH and dentin carious lesions was found at the tooth level, whereas MIH, DF, and DF and dentin carious lesions showed a negative relationship.

because the longer the teeth are in the oral cavity, the more susceptible they are to post-eruptive breakdown. 11,12 However, information about the oral health condition of individuals affected by MIH when they grow older is sparse, since prevalence studies are typically conducted in children from 6 to 10 years of age. 13 Additionally, identification of DF may be more straightforward in adolescents than in children. This is because they have more permanent teeth, especially homologous teeth, thus contributing to a more reliable diagnosis of fluorosis. 14 To our knowledge, only two studies have investigated the correlation between DF, DC, and MIH.
They reported an association between MIH and DC and between MIH and DF. 11,15 However, the diagnostic indexes used in both studies to classify all conditions are not the most discriminative. Additionally, the 7 to 9 years age group in one of the studies was not ideal for DF diagnosis. Therefore, this study aimed to investigate the prevalence, severity, and possible association between MIH, DF, and DC in a population of adolescents (11 to 14 years old), since there is a lack of information in this Brazilian population.  The inclusion criteria were healthy children aged between 11 and 14 years. Children 1) with dental hypersensitivity and enamel defects other than MIH and DF, 2) under orthodontic treatment, 3) with special needs that would impair proper examination, and 4) whose parents did not sign the informed consent were excluded from the analysis. Paulo, Brazil) were used. Data were recorded in a specific form by two trained note-takers. For DC and MIH, every tooth surface was recorded, with the most severe score registered in case of two or more defects on the same surface, and tabulated using CAST and MIH-SSS, respectively. For DF, the tooth was used as a unit to record the presence and severity using the TF index.

Methodology Study design and ethical aspects
A sociodemographic questionnaire with questions on the use of fluoridated toothpaste, brushing times per day, swallowing of toothpaste during infancy, and the source of drinking water was answered by the parents after clinical examination by phone interview.
The examined population presented a late mixed dentition, and data from primary and permanent dentitions were presented separately (336 and 10,455 deciduous and permanent teeth, respectively).
Although the study population presented with mixed dentition, only the permanent teeth were considered for the evaluation of the association between the three conditions on the tooth level. For the individuallevel analysis, it was observed that all children with J Appl Oral Sci. 2021;29:e20200890 4/10 DC in the primary dentition also presented with DC in the permanent dentition. In total, 10,455 teeth (only permanent dentition) were used for DC and DF analysis. Meanwhile, only the first permanent molars and permanent incisors (central and lateral) were considered for MIH (4,780 teeth).
Bivariate analysis using Chi-square test (χ²) with a 5% significance level was used to evaluate the associations among the three conditions. For the tooth level analysis, a logistic regression model was used to obtain the odds ratio using the independent variables with p<0.20. In the bivariate analysis, a model according to stepwise forward selection was considered. The software used was Stata / SE 15.1 (StataCorp, College Station, TX 77845, USA).

Results
In total, 400 adolescents, 171 boys and 229 girls with a mean age of 12 ±1 years, were examined (234 from school 1 and 166 from school 2), and 291 parents responded to the sociodemographic questionnaire (response rate of 72.75%).
Concerning the sociodemographic information, approximately 56% of the mothers worked outside the home and, most of them had not completed elementary/middle school (35%), 11% had not completed high school, and 26% had completed high school. Regarding the father's educational level, most of them also had not completed elementary/ middle school (34%), 9% had not completed high school, and only 17% had completed high school.
Most families lived in houses (56%) that were rented (32%) and had a monthly income of up to one (48%) or between 1 and 2 (38%) Brazilian minimum wages (approximately 200 US dollars). Most of the families had only one family member that works (64%). The mean number of people living in the same household was around 5±2.
For dental caries, the majority of teeth, whether primary or permanent, were classified as sound (CAST 0). However, when the maximum CAST per participant was estimated, for the permanent dentition, 94.75% presented DC (CAST 3 to 7). Considering the enamel (CAST 3) or dentin carious lesions (CAST 4, 5, 6, and 7), approximately 66% and 29% of the sample had at least one of these types of lesions, respectively. For DF, 10,455 teeth (400 adolescents) were evaluated. Table 1 shows the distribution of the adolescents according to the TF index and for severity.
A prevalence of 40.75% (163 adolescents) was observed, with mild cases (TF 1 and 2) being the  Table   2 shows the distribution of adolescents according to the variables related to fluoride sources. A significant difference was observed for the water source. A greater number of adolescents without fluorosis ingested well water than adolescents with fluorosis (χ², p=0.01).
When the three conditions (DC, MIH, and DF) were evaluated per adolescent, dentin carious lesions, MIH, and DF coexisted in about 2.5% of the adolescents.  10 had moderate, and 52 had severe MIH. Table 5 shows that a significant positive relationship between the MIH and dentin carious lesions was detected at the tooth level was detected (χ², p<0.00001).
Furthermore, there was a significant correlation between dentin carious lesions and MIH severity. Of the 227 teeth classified with MIH, 33 presented dentin carious lesions, 28 of which were classified as having severe MIH (χ², p<0.00001).
Finally, when considering the relationship between MIH and DF, we observed a significant negative association between these two variables (χ²,

Discussion
Our study evaluated the presence and severity of DC, DF, and MIH and their association in a group of socially vulnerable adolescents aged 11 to 14 years. We found an association among the three conditions; however, we decided to also report the J Appl Oral Sci. 2021;29:e20200890 7/10 findings of each condition separately. The population's socioeconomic background can partially explain the high prevalence of caries reported in a community with access to fluoridated water. When the variables "parental level of education" and "family income" were analyzed together, we learned that the adolescents investigated came from poor families. Besides, previous investigations in the same region have reported that the affordability of dental care in that locality is limited, 19 which also contributed to the high level of participants with untreated carious lesions.
Regarding DC detection, the CAST instrument used in our study is considered a suitable system for caries assessment in epidemiological surveys. 20 This takes only one minute longer to be performed than the DMF/dmf index. 21 Moreover, CAST provides a more detailed evaluation, since it includes the recordings of the enamel lesions. By using this strategy, we could identify a worrying scenario for permanent dentition: less than 4% of the adolescents were caries-free and approximately 66% presented at least one carious enamel lesion and 29% had dentinal caries. This emphasizes the importance of preventive dentistry and restorative treatment of dentin lesions in this population. Furthermore, the only treatment Regarding DF, the fact that we included adolescents aged from 11 to 14 years was considered an advantage. It allowed for the evaluation of the anterior teeth that were fully erupted, contrary to other studies in which only the younger age groups were assessed, thus impeding an accurate diagnosis. The prevalence of DF was 40.75%, and the most prevalent cases were the mild cases, which are not considered a public health concern. 16 Both outcomes are consistent with previous studies, since the prevalence of DF varies from 16.7% to 65%, 22,23 and a predominance of mild cases is frequently reported. 16,23 However, 12.50% of the adolescents presented either moderate or severe DF, which can affect their quality of life. 16,24 DF severity is influenced by factors such as the amount of fluoride ingested, age and time of exposure, body weight, and some systemic conditions. 14,24 In our investigation, special attention was given to the sources of fluoride, as Paranoá is supplied with fluoridated water (0.6 to 0.8 mg/L) and most children reported using fluoridated toothpaste. According to our findings, the only variable associated with DF was the source of the water consumed. Adolescents that consumed water from artesian wells or bottled water in their childhood presented significantly less fluorosis than those who consumed public drinking water. This finding contradicts a systematic review that concluded that individuals exposed to artesian well water are at high risk of developing DF; however, it is important to emphasize that the comparison made in the review was between regions with non-fluoridated water and locations that used groundwater. 24 As mentioned, Paranoá's piped water contains fluoride, but the amount of fluoride in groundwater in the region is not known and was not measured, which limited the interpretation of our findings. Moreover, memory bias might have influenced the responses given by parents.
Regarding MIH, the prevalence found was 18%, similar to that of other studies in Brazil 25 and slightly higher than that reported globally (14.2%). 13 Differences in the examination methods, diagnostic criteria, and age of the participants can help to explain these variations. In our investigation, the decision to include adolescents (11-14 years old) is justified by the fact that there is a lack of information about the oral health status of participants in that age group. This is relevant, especially in a socially vulnerable population with a lack of access to dental care. One of the main concerns related to MIH is the post-eruptive breakdown over time, increasing the chances of tooth loss. 12 Our results found that 41.7% of the adolescents diagnosed with MIH already presented this condition at a severe stage, which was considered high. At the tooth level, 22.9% of the MIH-affected teeth were diagnosed as severe. This outcome was influenced by the patient's age, since the longer the affected enamel is exposed to masticatory forces, the greater the odds of damage to the tooth structure. 7,26 To our knowledge, only one study evaluated this specific age range, in which 10.75% of MIH-affected teeth were classified as severe. 27 However, an increased severity has been J Appl Oral Sci. 2021;29:e20200890 8/10 reported in children older than 10 years. 25,26 Since the definition of severe cases varied considerably among the studies, the results are difficult to compare. Both Costa-Silva, et al. 25 (2010) and Bhaskar and Hegde 26 (2014) included post-eruptive breakdown restricted to the enamel as severe cases, which is different from the study by de Lima, et al. 27 (2015) and the present investigation. However, if we compare our findings to a prevalence study conducted in the same region using the same diagnostic criteria (MIH-SSS), but assessing children aged 8 years, 10 a disturbing finding was observed: a dramatic increase in severe cases from 13% described by Raposo, et al. 10 (2019) to approximately 42% reported in our study. This finding reinforces the importance of early diagnosis of MIH followed by monitoring of the affected teeth. It is important to emphasize that the diagnosis of MIH does not condemn the tooth to develop DC. The assessment of an older group of participants than what is usually recommended 6,13 helped in the observation of the severity of MIH and its consequences. Previous studies showed a relationship between dental caries and MIH, 5,29 which corroborates our findings that showed a positive significant association between the two conditions at the tooth level. According to Negre-Barber, et al. 30 (2018) in areas with a high prevalence of caries, MIH can remain difficult to diagnose, since rapid caries progression eliminates any trace of this condition. Besides, the more severely the tooth is affected, the greater the chances of requiring restorative treatment and re-interventions throughout life, which can eventually lead to endodontic treatment or tooth loss prematurely. 31  Considering that a fluorotic enamel is more porous, the fluoride from dentifrice used by the adolescents might have been able to diffuse throughout the enamel, protecting from demineralization. 32 Limitations of our study include the memory bias that might have affected the parents' answers regarding the use of fluoride and the fact that no information about the amount of fluoride in the well water was available. Moreover, an eating and dietary habits survey was not performed, and, since sugar consumption directly influences the development of DC, the higher number of DCs found in this population could also be explained, at least in part, by their dietary patterns. 3 However, our study also presents important strengths, such as the inclusion of more than 300 participants for MIH, as recommended by Elfrink, et al. 33 (2015) and the application of detailed diagnostic criteria for all conditions investigated.

The high number of untreated DCs emphasizes
Is there an association between dental caries, fluorosis, and molar-incisor hypomineralization? J Appl Oral Sci. 2021;29:e20200890 9/10 the studied population's lack of access to dental care. Cases of teeth with mild MIH were the most predominant in MIH-affected teeth. DF was the most prevalent problem detected, whose cases were mostly mild. Further studies are necessary to estimate the prevalence and severities of these dental conditions and their associations, especially in areas where access to dental care is limited.

Conclusion
No association between DC, MIH, and DF was observed at the participant level; however, at the tooth level, MIH was positively associated with dentin carious lesions, whereas MIH, DF, and DF and dentin carious lesions showed a negative association.

Author disclosure statement
The authors declare no conflicts of interest.