| Sign In to gain access to subscriptions and/or personal tools. |
Is Fluoride Concentration in Dentin and Enamel a Good Indicator of Dental Fluorosis?
1 Faculty of Dentistry, University of Toronto; Correspondence: * corresponding author, grynpas{at}mshri.on.ca
Despite some studies correlating dental fluorosis (DF) and fluoride (F) concentration in dental enamel, no information is available about DF and dentin F concentration. Our objective was to determine the correlation between teeth F concentration and DF severity in unerupted human 3rd molars, and the correlation between dentin and enamel F concentrations in the same tooth. Ninety-nine 3rd molars were studied—53 from Fortaleza, Brazil (F water, 0.7 ppm), 22 from Toronto (1.0 ppm), and 24 from Montreal (0.2 ppm). DF severity was evaluated according to the Thylstrup-Fejerskov Index, while F concentration was analyzed by Instrumental Neutron Activation Analysis. DF severity varied between TF0 and TF4, while F concentration ranged between 39 and 550 ppm in enamel and 101 and 860 ppm in dentin. Our results showed correlation between dentin F concentration and DF (rS = 0.316, p = 0.001), but no correlation between enamel F concentration and DF (rS = 0.154, p = 0.133). No correlation was observed between dentin and enamel F concentrations in the same tooth (rS = 0.064, p = 0.536).
Key Words: fluoride • dentin • enamel • dental fluorosis
Dental fluorosis (DF) is a tooth malformation believed to be caused by chronic ingestion of high levels of fluoride (F) during tooth development (Murray et al., 1991a; Den Besten, 1994). However, in addition to the major risk factors previously identified (related to ingestion of F) (Mascarenhas, 2000), other factors, such as genetic susceptibility, have also been shown to be important (Everett et al, 2002) in DF prevalence. DF prevalence has increased throughout the world (Angmar-Mansson et al., 1994), ranging between 7.7% and 80.9% in areas with fluoridated water and between 2.9% and 42% in areas without fluoridated water (Clark, 1994; Mascarenhas, 2000; Pendrys, 2000; Everett et al., 2002). Few studies have analyzed the F concentration and distribution in mineralized tissues (Richards et al., 1977, 1989, 1992; Olsen and Johansen, 1978). It has been shown, for example, that F concentration is higher in cementum followed, respectively, by alveolar bone, dentin, and enamel (Kato et al., 1990; Ishiguro et al., 1994). Differences between F concentrations in the enamel surfaces of various tooth types in the same individual have also been found (Aasenden et al., 1973), and correlation between DF and F concentrations in enamel has been reported (Retief et al., 1979; Richards et al., 1989, 1992). However, a correlation between F concentration and DF severity has not been extensively investigated. A study done by Brudevold et al.(1978), looking at the correlation between DF severity and F concentration in rat incisors, found a consistent trend between F concentration and DF scores. However, each score had a wide range and overlapping F concentrations. Thus, teeth with normal appearance (no fluorosis) and teeth with a high fluorosis score might have the same F concentration, whereas teeth with diverging F concentrations were placed in the same DF group (WHO, 1984). Olsen and Johansen (1978) analyzed human teeth and showed that the F concentrations in enamel were essentially independent of surface appearance. On the other hand, a study by Richards et al.(1992), looking at unerupted human teeth, claimed that a correlation between F content and the degree of DF existed and was significant. However, the study had low power (n = 14), and, when only 3rd molars were re-analyzed (n = 9), the results showed a trend, not a statistical correlation, between increasing TF scores and F concentration in the enamel. Interestingly, there are no available data correlating DF and dentin F concentration. This information is important because of the potential use of dentin as a biomarker for total F body burden. Biomarkers are defined as indicators signaling events in biological systems or samples (Committee on Biological Markers of the National Research Council, 1987). Once validated, a biomarker can be used to assess the exposure risk of populations and individuals to various substances (Sampaio, 2000). Dentin, especially coronal, may be the best marker for the estimation of chronic F intake and the most suitable indicator of total F body burden. Dentin contains only the F that has been incorporated through systemic ingestion. It does not normally undergo resorption, continues to accumulate fluoride throughout life, is more easily obtained than bone, and is also protected from F exposure in the oral cavity and surrounding bone by the covering enamel and cementum (Ten Cate, 1994; WHO Expert Committee on Oral Health Status and Fluoride Use, 1994). The purpose of our study was to determine the relationship between F concentration in teeth (dentin and enamel) and the DF severity in unerupted 3rd molars from areas with different F concentrations in the drinking water. The correlation between dentin and enamel F concentration in the same tooth was also determined.
Patients undergoing extraction of 3rd molars at hospital and private clinics in Toronto (Canada), Montreal (Canada), and Fortaleza (Brazil) were asked to donate their extracted teeth and to answer a brief questionnaire about their place of residence and water consumption. Ethical approval for this research was granted by the University of Toronto, University of Montreal, and Universidade Federal da Paraiba ethical committees. Information sheets and consent forms were used by all patients involved in this project. For this study, we collected a total of 169 unerupted 3rd molars, coming from Toronto (water artificially fluoridated at 1 ppm for over 35 yrs), Fortaleza (water artificially fluoridated at 0.7 ppm for over 12 yrs), and Montreal (water not artificially fluoridated; natural levels around 0.2 ppm). Teeth from Brazil were sent to Canada (where analysis was carried out) in wet gauze immersed in thymol and were immediately frozen upon arrival. Teeth originating in Canada were frozen immediately after collection. Storage of teeth by freezing has been shown to have no significant effect on shear-bond strength and to be an appropriate method of tooth storage (Titley et al., 1998).
Before being prepared, teeth were defrosted for approximately 12 hrs at room temperature (21°C). Teeth were then washed in running water, quickly dried, and analyzed by one of the authors (AV) for DF severity according to the Thylstrup-Fejerskov Index (TFI) (Thylstrup and Fejerskov, 1978) (Fig. 1B
After the DF severity assessment, teeth were transferred to molds and embedded in epoxy resin (Epoxycure resin, Buehler, Markham, Canada) that was allowed to polymerize at room temperature (21°C) overnight. The blocks were sectioned by means of a low-speed saw (Isomet, Buehler Ltd., Lake Bluff, IL, USA) in a buccal-lingual direction following the coronal-apical axis. Two sections (2 mm apart) were made on the central part of each block, dividing the tooth into thirds (mesial, central, and distal). The tooth coronal central part was further sectioned to provide buccal and lingual samples of enamel and dentin (Fig. 1A  ). Microdissection was used to separate the enamel and dentin from each other. The dentin samples from the buccal and lingual sides of each tooth were collectively analyzed for F concentration by instrumental neutron activation analysis (INAA). The same procedure was used for the enamel samples. In INAA, each sample is bombarded with thermal neutrons that produce short-lived radioisotopes from the elements in the sample. These radioisotopes decay with specific half-lives, emitting gamma rays of discrete and characteristic energies. The relative amounts of gamma rays detected are proportional to the concentrations of the elements in the sample (Mernagh et al., 1977). INAA gives the F concentration average of the sample and is capable of measuring element concentrations to the 10 ppm level (0.01%). Samples weighed approximately 30 mg for dentin and 70 mg for enamel. We used the Spearman correlation test to evaluate the correlation between DF severity and F concentration in dentin and enamel and to evaluate the correlation between dentin and enamel F concentrations in the same tooth. We used the Mann-Whitney U test to evaluate the differences in F concentrations in tooth structure between samples with and without DF. We performed the Kruskal-Wallis test to evaluate the differences in F concentrations in tooth structure in the 5 TF scores. Statistical analysis was done with the use of statistical analysis software (SPSS for Windows, SPSS Inc., Chicago, IL, USA).
From the 169 teeth collected, we analyzed only unerupted human 3rd molars with complete (n = 56) or almost complete (n = 43) roots and that did not exhibit any obvious damage as a result of extraction (n = 99). From the 99 3rd molars studied, 53 teeth were from Fortaleza, Brazil, 22 from Toronto, and 24 from Montreal. The mean age of the patients from whom the teeth were collected was 20 yrs (range, 13–38 yrs). Fifty-four percent of the teeth collected were maxillary 3rd molars, while 46% were mandibular 3rd molars. About 89% of the teeth came from patients who customarily cook with tap water, and about 40% of the teeth came from patients who customarily drink tap water. The F concentration varied between 39 ppm and 550 ppm in the enamel samples, and between 101 ppm and 860 ppm in the dentin samples. The DF severity varied between 0 and 4. The number of samples per group, the mean values of F concentration, the standard deviation, and the range of F concentrations for each DF group (presented by TFI score) for enamel and dentin are shown in Table 1  . We can see that the F concentration ranges for the different DF groups overlap (Fig. 2). However, the average F concentrations in dentin and enamel increase with DF severity. We can also see higher values of F concentration in dentin than in enamel at each DF level. In addition, the F concentration variance in enamel and dentin tends to increase with increasing severity of DF. The Mann-Whitney test showed no difference in F concentration between samples with and those without DF in enamel (p = 0.134) and a significant difference between those two groups in dentin (p = 0.020). The Kruskal-Wallis test showed no significant difference in F concentrations among the 5 different fluorotic groups in enamel (p = 0.502) and a trend in the dentin (p = 0.056).
A positive correlation was found between dentin F concentration and DF (rS = 0.316 and p = 0.001) (Fig. 2a ). However, this correlation explains only about 10% of the variance between the two variables. No correlation was found between enamel F concentration and DF (rS = 0.154 and p = 0.133) (Fig. 2b). Interestingly, even though there is a general increase of F concentration with an increase of DF in dentin (significant) and enamel (not significant), no correlation was found between dentin and enamel F concentrations in the same tooth (rS = 0.064 and p = 0.536) (Fig. 2c).
This is the first study to present data on dentin fluoride (F) concentrations in unerupted human teeth with different levels of DF, and to compare F concentrations in dentin and enamel in the same tooth. Although the results of this study are based on a relatively limited number of severe cases of DF, the samples covered the whole range of DF severity possible in unerupted teeth (TF0 to TF4) (Fejerskov et al., 1988; Richards et al., 1992). The use of a single type of tooth in this study avoided the problems associated with the variations of F concentration among different types of teeth in the same individual (Aasenden et al., 1973). Unerupted 3rd molars are not exposed to the oral environmental, are easily collected, and have a potential for being a good reservoir of F deposition.
There is some evidence that the F content is greatest in teeth where the time from completion of enamel formation to tooth eruption is extended (Ten Cate, 1994). This could be the case for the 3rd molars. These teeth start mineralization relatively late and commonly stay unerupted for long periods. On average, a period of 6 yrs can be expected between completion of enamel formation and tooth eruption (Scully et al., 1998). This period is relatively similar to that found in pre-molars, the teeth known to be most severely affected by DF (Moller, 1965; Larsen et al., 1986). Table 2
A study from Richards et al (1992) that analyzed different types of unerupted teeth showed increasing enamel F concentration in teeth with increasing severity of DF. However, when only 3rd molars from this study were analyzed, the results confirm our finding of no correlation between enamel F concentration and DF (rS = 0.571 and p = 0.108) in the F range analyzed. The stronger correlation between the two variables in the study by Richards et al.(1992), compared with that found in our study, might be due to the fact that the whole TF scale was used in their study. The authors "predicted" the DF severity of unerupted teeth as if the tooth had erupted. No reliable means of applying this technique in our study was possible. We therefore elected not to attempt to use this "prediction" method in the present study. Our investigation showed, for the first time, a positive correlation between F concentration in dentin and DF severity. However, no correlation between enamel F concentration and DF severity was found. Interestingly, the coefficient of determination (r2), which expresses the proportion of variance in the dependent variable explained by the independent variable (Streiner and Norman, 2000), is very low for our results (dentin r2 = 0.1). This fact calls into question the generally accepted hypothesis that the main factor responsible for DF severity is the F concentration in tooth structure. We believe that other factors, such as individual genetic variation, that can influence DF susceptibility can also play an important role in DF severity. Support for this hypothesis has been provided in a recent study in different mouse strains (Everett et al., 2002). That study showed that different inbred strains of mice that have been controlled for genotype, age, gender, food, housing, and drinking water F level presented different susceptibilities to DF. This potential effect of genetic variation may explain the low correlation of dentin F concentration with DF severity and the lack of correlation between enamel F concentration and DF severity. In addition, genetic variation may also explain the general increase in F concentration variability with increasing DF severity found in our study. The influence of individual genetic susceptibility becomes more obvious in the more severe cases of DF, where the individuals are being challenged to a greater extent by the higher F levels. One of the characteristics of dentin which makes its study as a biomarker for F exposure interesting is the fact that it continues to form throughout life (narrowing the pulp chamber). By its continuous formation, dentin keeps accumulating fluoride throughout life, being then a good biomarker for total fluoride body burden. This peculiarity has been proven by a study that showed a correlation between age and fluoride concentration in dentin (Nakagaki et al., 1987). In our study, this correlation was not shown in any of the three localities studied (Montreal, rs = –0.204, p = 0.352; Toronto, rs = –0.053, p = 0.811; and Fortaleza, rs = –0.233, p = 0.081). We believe that this is due to the narrow age range of the subjects analyzed (mean = 20.55 yrs, SE = 0.43 yrs). However, it is important to note that dentin formation slows with age. Therefore, different ratios of dentin F exposure relative to total body burden may exist at different ages. In our study, significantly higher values of F concentration were found in dentin when compared with enamel (p < 0.001), and no correlation was found between enamel and dentin F concentrations in the same tooth. Apatite crystallites in dentin are considerably smaller and less crystalline than enamel. This increased surface area of the crystallites, together with the higher degree of tissue hydration (tubular structure) in dentin compared with enamel, increases the capacity for fluoride uptake by dentin (Fejerskov et al., 1996) and may explain the higher values of F in this tissue compared with enamel. The explanation for the non-correlation between F concentrations in enamel and dentin in the same tooth is unclear. It may derive from the different embryological origin of these tissues (Ten Cate, 1994). Furthermore, the fact that this enamel-dentin non-correlation is associated with a significant correlation between DF severity and dentin F concentration reinforces the biomarker potential of dentin for F exposure, which needs to be further investigated.
The authors express their thanks to those who helped in tooth collection: Mrs. Silva in Fortaleza, Brazil; Dr. Clokie, Dr. Caminiti, Dr. Baker, and oral and maxillofacial surgery residents in Toronto; and Dr. Schwartz, Dr. Gornitsky, and Dr. Beaudet-Roy in Montreal. We also thank Drs. Bull and Kopciuk for their help in the statistical analysis. This study was supported by a grant from the Canadian Institute of Health Research (CIHR) and by Harron and Connaught scholarships (AV). Received for publication December 18, 2002. Revision received September 24, 2003. Accepted for publication November 4, 2003.
Journal of Dental Research, Vol. 83, No. 1,
76-80 (2004) This article has been cited by other articles:
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
Sign In to gain access to subscriptions and/or personal tools.
TOP
ABSTRACT
INTRODUCTION
