| Sign In to gain access to subscriptions and/or personal tools. |
Does Caries in Primary Teeth Predict Enamel Defects in Permanent Teeth? A Longitudinal Study
1 Faculty of Dentistry, University of Otago, PO Box 647, Dunedin, New Zealand; Correspondence: * corresponding author, jonathan.broadbent{at}stonebow.otago.ac.nz
The notion that caries in primary teeth causes developmental defects of enamel in permanent teeth has been recently revived. The research objective was to test this hypothesis through analysis of data from the Dunedin Multidisciplinary Health and Development Study, a longstanding prospective cohort study. The maxillary incisors of 663 children were assessed for existing restorations and dental caries at age five and for developmental defects of enamel at age nine. Where a primary tooth had been carious, the permanent successor was more likely to have a demarcated opacity after adjustment for gender, family socio-economic status, years of exposure to water fluoridation, trauma to primary teeth, and early loss of primary teeth (unadjusted OR = 2.3, 95% CI 1.3, 4.1; adjusted OR = 2.2, 95% CI 1.1, 4.3). These findings support a time-ordered association between dental caries in primary maxillary incisors and demarcated opacities in their permanent successors.
Key Words: caries • enamel defect • longitudinal study
Early last century, Turner suggested that a primary tooth abscess may cause local sepsis around the developing permanent successor, thus interfering with normal development and resulting in an enamel defect (Turner, 1906). Turner later suggested that the buccal surfaces of the incisors are more likely to be affected than the lingual, due to the close approximation of the roots of the primary incisors to the former (Turner, 1912). Studies from Japan and the US in the 1960s found relationships between abscesses of primary molars and enamel defects of the succeeding premolars (Niswander and Sujaku, 1962; McCormick and Filostrat, 1967). A 1992 UK study of enamel defects found a greater prevalence of enamel defects among Asian children than in others. The authors noted that ‘previous local studies’ had found more primary caries in Asian children, and an etiological link between caries and enamel defects was suggested; however, no effort was made to confirm this (Elley and Charlton, 1993). In a second UK study, children of about age 12 with no enamel defects in permanent teeth were found to have a mean DMFS of 2.7, while those having only demarcated opacities had 3.4, those with only diffuse defects had 1.6, and those with a combination had 2.4. The authors speculated that, since children with caries by age 12 are likely to have had caries in their primary teeth, caries in the primary dentition may cause local conditions resulting in demarcated opacities in permanent teeth; the lower mean DMFS among those with diffuse defects was believed to be associated with fluoride exposure (Ellwood and O’Mullane, 1994). A study in China recently drew attention back to this possible association between caries in primary teeth and enamel defects in permanent successors. Some 452 children living in the non-fluoridated region of Conghua (Southern China) were examined for caries in the primary teeth between the ages of three and six. A total of 388 of these children had at least two examinations during this period, and, of these, 250 participated in a follow-up study of enamel defects at about 12 years of age. All erupted surfaces of the teeth of succession were examined for enamel defects at this age. Where the primary precursor of a permanent tooth had been carious, there was greater prevalence of demarcated opacities, and hypoplasia was greater, than where there had been no caries [7.5 vs. 3.8% and 1.9 vs. 0.4%, respectively (Lo et al., 2003)]. Furthermore, experimental studies have shown an association of artificially induced inflammation of the periapical tissues of primary teeth with defects of formation of permanent teeth (including hypoplasia) in monkeys (Kaplan et al., 1967; Winter and Kramer, 1972; Valderhaug, 1974) and dogs (Binns and Escobar, 1967), and these have been supported by human autopsy reports (Morningstar, 1937; Bauer, 1946). It can be difficult to establish the etiology of enamel defects (Suckling et al., 1987). Demarcated opacities and hypoplastic defects of enamel tend to be isolated and of a sporadic distribution; thus, a local cause would be most plausible. In contrast, diffuse opacities more frequently affect multiple teeth which have undergone enamel secretion and maturation during the same period; thus, a systemic cause (such as consumption of fluoridated water) is more likely. The close approximation of the buccal surface of a developing permanent incisor with the root end of its precursor means that insult to a developing permanent tooth (by caries-associated periapical inflammation of the primary tooth) may interfere with the normal process of enamel matrix deposition or mineralization, thus resulting in a demarcated opacity or hypoplastic defect. A further possibility is that early extraction of a primary tooth may cause trauma to the developing permanent tooth, thus damaging the enamel matrix and resulting in a defect of enamel. Since carious teeth are more likely to be extracted than non-carious primary teeth, this could result in a higher likelihood of enamel defects in the permanent successors of such teeth. Alternatively, a carious primary tooth might be more likely to be a target for topical fluoride application by a dentist; this high concentration of locally applied fluoride might result in a fluorotic defect of enamel. The aim of this study was to test the hypothesis that dental caries in primary teeth is a risk factor for enamel defects in their permanent successors.
A secondary analysis of data from the Dunedin Multidisciplinary Health and Development Study was conducted. The Dunedin Study is a prospective study of a birth cohort of 1037 children born in Queen Mary Hospital, Dunedin (New Zealand), between 1 April 1972 and 31 March 1973 (Silva and Stanton, 1996). Collection of health and development data was undertaken once every two years from children at ages three through 15, and from youth at ages 18, 21, and 26. The current study includes those who were dentally examined at ages five and nine. Ethical approval for each assessment phase was obtained from the Otago Ethics Committee. Study members and their parents gave informed consent before participating. Most of the age-five dental examinations were conducted within four to six weeks of the Study members’ fifth birthdays. Diagnosis and recording of dental caries were made according to the World Health Organization criteria (1977). The age-nine examinations for enamel defects were conducted within three months of each child’s ninth birthday. A tooth was considered present when more than half the clinical crown had erupted; scoring of enamel defects was according to the FDI Index of Developmental Defects of Dental Enamel. Because children are in the mixed-dentition stage at age nine, the permanent canines and premolars have not normally erupted. For this reason, such teeth were excluded from the analysis; also excluded were first molar teeth (since these are accessional teeth) and lower incisors (since their incidence of caries and prevalence of enamel defects are relatively low). The analysis of enamel defects was limited to the buccal surfaces of the maxillary incisors, since these teeth are usually erupted by the age of nine, and the buccal surfaces are the most important esthetically. It was not possible to determine whether missing primary teeth were missing due to caries, trauma, or early exfoliation. Detailed information was available on trauma to the primary teeth. Where a tooth had a history of trauma and was missing at age five, a dummy variable was created designated ‘Missing at five, trauma’. Where a tooth had a history of trauma but was not missing at age five, a dummy variable was created labeled ‘Trauma, not missing at five’. Where a tooth did not have a history of trauma but was missing by age five (e.g., due to extraction or early exfoliation), a dummy variable was created designated ‘Missing at five (not trauma)’. The socio-economic status (SES) of the Study members’ families was measured on the basis of the parents’ self-reported occupational status. Occupation was scaled into one of six categories (6 = unskilled laborer, 1 = professional) on the basis of the educational levels and income associated with that occupation in data from the New Zealand census (Irving and Elley, 1977; Elley and Irving, 1985). For the purposes of the multivariate analysis, dummy variables were created. Scores ‘5’ and ‘6’ (Low SES) formed the reference category; ‘3’ and ‘4’ were the ‘Medium SES’ category, and scores ‘1’ and ‘2’ comprised the ‘High SES’ category.
Univariate statistics were computed, and bivariate analyses were undertaken. The level of significance was set at P < 0.05; we used the
Some 922 Study members participated in dental examinations at age five (Evans et al., 1980), while 696 children were assessed for enamel defects at age nine (Suckling et al., 1985). Some 663 children were dentally examined at both ages five and nine, and the analyses are confined to these individuals. Males were more likely to be included than females (66.9% vs. 60.8%; P < 0.05), while no differences existed by SES. At age five, 2617 primary maxillary incisors were present; by age nine, 2395 permanent maxillary incisors were present. A total of 124 primary maxillary incisors (5.2%) were carious or restored (Table 1  ). A total of 616 permanent maxillary incisors (25.7%) had enamel defects (Table 2). Diffuse opacities were the most commonly observed defect and were most prevalent in the central incisors. Demarcated opacities were the next most common and were most prevalent in the central incisors. Hypoplastic defects were the least prevalent and showed no difference by tooth type.
Permanent teeth were more likely to have a demarcated defect or any defect if there had been experience of caries in the primary precursor by the age of five (Table 3 ). Before adjustment for potential confounding factors, a permanent tooth with a carious predecessor had 2.3 times the odds of having a demarcated opacity compared with a permanent tooth with a sound predecessor.
After adjustment according to gender, SES, number of years of exposure to water fluoridation, primary tooth loss, and primary tooth trauma, where the primary precursor had been carious, the permanent tooth was 2.2 times more likely to have a demarcated defect. Where the primary precursor had been lost for a reason other than trauma (such as extraction due to caries/abscess or early exfoliation), the permanent tooth was associated with a 5.0 times increased odds of having a demarcated opacity. The odds of a permanent tooth having a diffuse opacity increased by 1.1 for every year the child lived in a fluoridated area. Enamel hypoplasia was more common among females and where a precursor tooth was missing at age 5 and the subject had a history of trauma. The effect of hypoplasia in primary teeth upon hypoplasia in permanent teeth did not reach statistical significance. Other defects (and combinations of defects) were more common among persons of high childhood SES or where a precursor tooth was missing as a result of trauma (Table 4 ).
This study reports an association between dental caries in primary incisors and the presence of demarcated opacities in their permanent successors, and supports the findings of Lo et al.(2003). Unlike the Chinese study, the association between primary caries and permanent hypoplasia did not reach statistical significance. In considering this difference, one must consider that the Chinese study included all of the primary and successional permanent teeth, whereas this study included only maxillary incisor teeth, which may be at lesser risk of either caries or hypoplasia. A further possibility is that hypoplasia may be related to severe and long-term untreated caries—in the Chinese study, only caries that was present and untreated at two (or more) assessments was considered for the analysis, whereas the current study included all caries (treated or untreated) which was present at one assessment. This study has found that if caries occurs in a primary tooth, the successor tooth is more than twice as likely to have a demarcated enamel defect. In the case of early tooth loss (for reasons other than trauma, e.g., extraction due to caries/abscess), the permanent successor tooth was five times more likely to have a demarcated defect. No significant relationship was observed between caries in the primary incisors and the presence of diffuse opacities in the permanent incisors. However, an association was found between diffuse defects of enamel and exposure to fluoridated water supplies. Analysis of these data supports the idea that diffuse enamel defects are related to systemic causes. Other significant findings were the high odds ratios relating tooth loss due to trauma to both hypoplasia and the ‘other/combination’ defects. One problem with secondary analyses is that there is no control over the data which are collected. Unfortunately, data on sources of fluoride exposure other than water fluoridation were not collected. Also, the reasons for missing teeth at age five were not recorded; hence, some teeth that were missing due to caries (i.e., extracted) were designated as non-carious. However, these teeth were accounted for by the use of the dummy variable ‘missing teeth (not related to trauma)’. The high odds ratio for this variable relating to demarcated opacities supports the study hypothesis, since such teeth are more likely to have had severe caries with periapical abscesses. This study supports the hypothesis that caries in a primary tooth will increase the odds of the succeeding permanent tooth having a demarcated defect, with the risk more than doubling. The precise mechanism for this process is unclear as yet, but it may be related to the presence of periapical infection. An interesting perspective on these findings is that perhaps fluoridation of municipal water supplies (at optimal levels) could indirectly lead to a decrease in the rate of demarcated defects.
Jonathan Broadbent is supported by Grant R01 DE-015260-01A1 from the National Institute of Dental and Craniofacial Research, National Institutes of Health, Bethesda, MD, USA 20892. We thank Dr. P.A. Silva, Dr. R.H. Brown, and all those involved in the administration of the Study. We are also indebted to Dr. R.W. Evans and Dr. G.W. Suckling for collecting and reporting on the data from Phases 5 and 9 of the Study. Dr. D. Welch is thanked for his suggestions regarding the multivariate analysis. The Dunedin Study would not be possible but for the ongoing participation of the Study members. The Health Research Council of New Zealand (previously the Medical Research Council), the New Zealand Department of Education, the New Zealand Department of Health, and the National Children’s Health Research Foundation provided funding for the assessments. Received for publication April 29, 2004. Revision received November 7, 2004. Accepted for publication December 31, 2004.
Journal of Dental Research, Vol. 84, No. 3,
260-264 (2005)
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
Sign In to gain access to subscriptions and/or personal tools.
TOP
ABSTRACT
INTRODUCTION
2 test to test for significance of observed associations. Multivariate analysis was conducted in Intercooled Stata 8.0 (Stata Corporation, College Station, TX 77840, USA, 2003) by generalized estimating equations, with independent working matrices and robust standard errors.