| Sign In to gain access to subscriptions and/or personal tools. |
A Longitudinal Study of Tooth Erosion in AdolescentsDepartment of Preventive and Restorative Dentistry, Radboud University Nijmegen Medical Centre, PO Box 9101, NL-6500 HB, Nijmegen, Netherlands; Correspondence: * corresponding author, H.ElAidi{at}dent.umcn.nl
Incidence studies on tooth erosion among adolescents are scarce. This longitudinal study aimed at estimating the prevalence, incidence, progression, and distribution of erosion in young adolescents over a 1.5-year period. Erosion at baseline was present in 32.2% of the 622 children (mean age, 11.9 ± 0.9 yrs) and increased to 42.8% over 1.5 yrs. The prevalence of deep enamel or dentin erosion increased from 1.8% to 13.3%. It was hypothesized that more boys will suffer from erosion than girls. At baseline, no difference was found (p = 0.139), whereas 1.5 yrs later, more boys showed erosion than girls (p = 0.026). Of the erosion-free children, 24.2% developed erosion. Progression of erosion in terms of lesion depth and/or in number of lesions was 61.0% and was found more frequently in boys than in girls (p = 0.034). The distribution of erosion showed a predominance of occlusal surfaces of molars, especially the mesiobuccal cusp top, and palatal surfaces of upper anterior teeth.
Key Words: tooth erosion • adolescents • prevalence • incidence • distribution
Tooth erosion is defined as the loss of dental hard tissue due to the chemical influence of acids without bacterial involvement (Imfeld, 1996). Erosive tooth loss cannot be regarded as pathological per se (Ganss and Lussi, 2006). Erosion becomes pathological when it leads to pain, endodontic problems, or loss of function or esthetics. The distinction of acceptable and pathological tooth erosion at a given age is based upon the prediction of whether the tooth will survive the rate of wear (Smith and Knight, 1984a). A review of epidemiological studies carried out among children in the age range 9–17 yrs showed a prevalence of tooth erosion between 11 and 100% (Jaeggi and Lussi, 2006). This wide prevalence range is due to different standards of examination and populations. Furthermore, these studies often addressed tooth wear in general and not erosion specifically, which makes direct comparisons of the results difficult. In the Netherlands, only two prevalence studies of tooth erosion have been conducted. The first study examined smooth-bordered tooth wear in two different age groups (van Rijkom et al., 2002). Three percent of the 10- to 13-year-olds and 30% of the 15- to 16-year-olds had visible smooth-surface wear. This suggests that tooth erosion increases with age. It should be noted, however, that cross-sectional studies are not very suitable for investigating the progression or incidence of tooth erosion. The second study identified 24% of the 12-year-old children as having erosive wear (Truin et al., 2005). Internationally, two longitudinal studies have been performed to assess the incidence of tooth erosion among adolescents. In a cohort of 12-year-old children, an incidence of 12.3% over 2 yrs was found (Dugmore and Rock, 2003). The longitudinal observation of orthodontic study models of adolescents revealed an incidence of 18% over 5 yrs (Ganss et al., 2001). The aim of the present study was to estimate the prevalence, incidence, progression, and distribution of tooth erosion over a 1.5-year period in a sample of children aged 10 to 12 yrs. It was hypothesized that significantly more boys would suffer from tooth erosion than girls (Al-Dlaigan et al., 2001; van Rijkom et al., 2002; Dugmore and Rock, 2003; Truin et al., 2005), as would children with low socio-economic status compared with children from other social backgrounds (Al-Dlaigan et al., 2001; Harding et al., 2003). To address these issues, we performed an interim analysis after 1.5 yrs of a three-year longitudinal study on the etiology of tooth erosion.
Sample The study sample consisted of regular attendees, aged 10 to 12 yrs, of a pediatric dental clinic in Oss, a medium-sized town with 57,300 inhabitants (January, 2005) in the central part of the Netherlands. Approximately 35% of the 10- to 12-year-old children in Oss visit this dental clinic. On 1 April 2004, the dental clinic’s patient records included 1012 children in that age range. Before conducting the main etiology study, we carried out a power calculation. For a multivariate analysis of the relationship between 12 presumed risk factors and the presence of tooth erosion, we required a total of 120 incident cases within the study period of 3 yrs. Assuming an incidence of 7% per year (van Rijkom et al., 2002), a non-response of 20% at baseline, a drop-out of 5% yearly, and 120 incident cases after 3 yrs, we needed to invite 834 children to include 666 persons in the study. Since the non-response appeared higher than expected (31%), 950 children were invited to participate in the study. The study protocol was approved by Radboud University’s research committee (CEOM No:2003-207). Informed consent was obtained from the parents or legal representatives of all participating children. Of the invited children, 656 participated in the study. The enrollment of participants started in April, 2004, and took 1 yr. After 1.5 yrs, all children, minus 34, were reexamined. The main reason for this dropout was that the individuals concerned were no longer registered as patients in the dental clinic. The socio-economic status (SES; low, medium, and high) was based on income, level of education, and home ownership. SES was assessed based on the children’s six-digit residential postal codes. For six children, the SES could not be assessed.
Clinical Investigation The criteria used for the clinical assessment of tooth erosion were a modification (van Rijkom et al., 2002) of the diagnostic criteria developed by Lussi (1996). The tooth erosion index codes used in this study were as follows: 0 (no visible erosion), 1 (initial enamel erosion: surface with silky-shining, "melted" appearance, the bottom of the lesion is white), 2 (deep enamel erosion: more pronounced signs than score 1, dentin is shining through, light yellow), 3 (dentin erosion, the lesion bottom is yellow). For a more detailed assessment of tooth erosion, the occlusal surfaces of permanent molars and the palatal surfaces of permanent upper anterior teeth were subdivided into several areas. Cusp tops and slopes were scored separately. Palatal surfaces of upper anterior teeth were examined, distinguishing between the cervical one-thirds of the surfaces and the remaining parts. Two examiners conducted the baseline and interim examinations. Both examiners received training and were calibrated by the examiner who had previously applied the index in a study in The Hague (GJT). Independently, the examiners blindly conducted duplicate measurements in 11% of the participants.
Statistical Analysis
Prevalence, Incidence, and Progression The mean age of the children at baseline was 11.9 yrs (SD = 0.9). The group consisted of 51% boys and 49% girls. Erosion at baseline was present in 32.2% of the children (95%CI: [28.5; 35.8]) (Table 1  ). This prevalence increased over 1.5 yrs to 42.8% (95%CI: [38.9; 46.7]). The proportion of children with deep enamel erosion increased from 1.8% (95% CI: [0.7; 2.8]) to 10.8% (95% CI: [8.3; 13.2]). None of the children at baseline had erosion exposing the dentin; at the 1.5-year evaluation, 2.6% of the children (95% CI: [1.4; 3.8]) showed dentin exposure. At baseline, the prevalence of tooth erosion did not differ significantly between boys and girls (p = 0.139, OR = 1.3, 95%CI: [0.9; 1.8]), or between the low- and medium-SES groups (p = 0.112, OR = 1.3, 95%CI: [0.9; 2.2]), or between the low- and high-SES group (p = 0.190, OR = 1.4, 95%CI: [0.9; 1.6]). At the 1.5-year evaluation, significantly more boys than girls showed tooth erosion (p = 0.026, OR = 1.4, 95%CI: [1.0; 2.0]), as did children with a low SES compared with children with a medium (p = 0.021, OR = 1.5, 95%CI: [1.1; 2.2]) or high SES (p = 0.018, OR = 1.7, 95%CI: [1.1; 2.7]). For children with erosion at baseline, the mean number of affected teeth was 1.73, involving 1.74 tooth surfaces. Over 1.5 years, the mean number of affected teeth increased to 3.63, involving 4.01 tooth surfaces.
We could not calculate the incidence simply by subtracting prevalence at baseline from prevalence at the 1.5-year evaluation, because 36 children (18.0%) who showed signs of erosion at baseline did not do so at the 1.5-year evaluation. One hundred and two children (24.2%) free of erosion at baseline developed erosion over the subsequent 1.5 yrs (Table 2 ). The majority of them (88.2%) developed only initial enamel erosion. The incidence of tooth erosion did not differ significantly (p = 0.120, OR = 1.4, 95%CI: [0.9; 2.3]) between boys and girls (27.9% and 20.6%, respectively), or between the low- and medium-SES groups (p = 0.150, OR = 1.4, 95%CI: [0.9; 2.4]), or between the low- and high-SES groups (p = 0.127, OR = 1.6, 95%CI: [0.9; 3.1]).
The progression of tooth erosion was analyzed for those children who showed erosion at baseline (n = 200). The percentage of children showing any form of progression (increase in lesion depth and/or number of lesions) was 61.0% and was significantly (p = 0.034, OR = 1.8, 95%CI: [1.0; 3.3]) higher in boys (67.6%) than in girls (52.8%). The progression did not differ significantly between the low- and medium-SES groups (p = 0.925, OR = 1.0, 95%CI: [0.5; 1.8]), or between the low- and high-SES groups (p = 0.812, OR = 0.9, 95%CI: [0.4; 2.0]).
Distribution
In lower molars, cusp tops (especially the mesiobuccal cusp top) were more frequently and more severely affected by tooth erosion than were cusp slopes (Fig. 2 ).
If the progression of tooth erosion is to be halted in an initial stage, early diagnosis is essential. However, early diagnosis of tooth erosion, especially for occlusal surfaces, introduces the danger of overestimating the tooth erosion problem (Ganss, 2006). In this study, we examined a group of young adolescents for the presence of tooth erosion on all accessible tooth surfaces and found a prevalence of 32.2% at baseline, which increased to 42.8% over 1.5 yrs. Regarding the numbers of affected teeth, the types of tooth surfaces, and the grades of the lesions, one could argue that these children do not suffer severely from erosion. However, when taking account of the young age of the children and the progression in prevalent cases, the risk for the development of more advanced erosion is likely. In the present study, a prevalence of 32.2% was found. This is higher than the previously reported prevalence of erosion in the Netherlands of 24% (Truin et al., 2005). In that study, only index teeth were examined. A two-year longitudinal study revealed a prevalence of 56.3% in 12-year-olds, which increased to 64.1% over 2 yrs (Dugmore and Rock, 2003). The presence of erosion found in the present study was substantially lower than the tooth wear reported previously (Milosevic et al., 1994). This is probably due to the fact that these authors applied the Tooth Wear Index (Smith and Knight, 1984a,b), which includes pure attrition. As hypothesized and in agreement with results from other studies (Al-Dlaigan et al., 2001; van Rijkom et al., 2002; Dugmore and Rock, 2003: Truin et al., 2005), a significantly higher prevalence of tooth erosion was found in boys than in girls. In the literature, the influence of socio-economic status is somewhat ambivalent. In line with reports from other studies (Al-Dlaigan et al., 2001; Harding et al., 2003), the present study demonstrated a significantly higher prevalence in children with a low socio-economic status compared with other children, while other research groups (Millward et al., 1994; van Rijkom et al., 2002; Bardsley et al., 2004) found more erosion in the high category. All indices for the clinical diagnosis of tooth erosion attempt visual assessment of the presence and severity of tooth erosion. In a recent study (Ganss et al., 2006), the accuracy of the visual diagnosis "dentin exposed" was found to be poor. So the validity of currently used indices is not fully established. The reproducibility between the examiners in this study was fair to good. However, until now no efforts have been made, either by us or by the working groups that previously used this index, to validate the diagnostic criteria and grading. Incidence is defined as the percentage of the sound population developing erosion during a given period. The incidence of tooth erosion in this study was 24.2%. When one is analyzing changes in erosion prevalence, regression due to measurement errors should also be considered. Regression in the present study was 18.0%; however, for 86.1% of the children, this involved initial enamel erosion on just one tooth surface. Nevertheless, to enable articles to be compared, we used a simpler estimate of the incidence, being the difference of subsequent prevalence. With this approach, the incidence in this study was 10.6%. This is in accordance with our hypothesis. Dugmore and Rock (2003) found an incidence of 12.3% over 2 yrs. Since clinicians may fail to see tooth erosion and the difficulty involved in halting its progression, it can be expected that, in prevalent cases, erosion will progress. The progression in the present study was 61.0% and was found to be significantly higher in boys than in girls. Dugmore and Rock (2003), over a two-year period, reported a lower progression of 47.7% in 12-year-old children. In this study, tooth erosion was predominantly found on palatal surfaces of upper anterior teeth and on occlusal surfaces of lower first molars. This is in agreement with the findings of Jaeggi and Lussi (2006). A novelty in the present study is the detailed assessment of occlusal surfaces of permanent molars and palatal surfaces of upper anterior teeth. In first molars, cusp tops, especially the mesiobuccal cusp tops, were more often eroded than were cusp slopes. A possible explanation is that the enamel layer around the cusp tops is thinner (Kono et al., 2002). The lesions on the cusp tops were cup-shaped. In most cases, contact with the antagonists during occlusion or dynamic occlusion was no longer present. Therefore, the progression of these lesions was due not to attrition, but to erosion and/or three-body wear. To conclude, in our young study population, severe tooth erosion was not found. However, the level of prevalence, especially of incidence and progression, raises concern. Therefore, the monitoring of adolescents with initial signs of tooth erosion, as well as a high standard of awareness on the part of dental professionals, is recommended.
The study was supported by grants from Radboud University Nijmegen, the Dutch Dairy Association, the Dutch Sugar Bureau, and the Dutch Soft Drinks Association. Received for publication July 23, 2007. Revision received February 21, 2008. Accepted for publication April 15, 2008.
Journal of Dental Research, Vol. 87, No. 8,
731-735 (2008)
|
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
|
 |
|
Sign In to gain access to subscriptions and/or personal tools.
TOP
ABSTRACT
INTRODUCTION
