Inglês (pdf)
Artigo em XML
Referências do artigo
Tradução automática
Permalink
INTRODUCTION
Dental development is a prolonged and highly complex biological process that occurs from birth to about 16 years old1,2. Dental age (DA) estimation uses the development and calcification of the tooth to estimate the chronological age. The method proposed by Demirjian et al (1976)3 is the method widely used to DA estimation in dental research4. Several factors are associated with delayed dental maturation-dental age, such as malnutrition5,6, non-syndromic tooth agenesis7,8 and hormonal deficiency9,10. Others factors are also associated with earlier dental maturation, such as obesity6,11 and genetic mutations12.
Several studies in different populations investigated if DA may be a biological predictor for skeletal malocclusion patterns development, however, the association between these two traits are still controversial13-24In fact, the skeletal complex, including maxilla and mandible, development share the same etiological factors with dental development12,25-28.
The knowledge whether different craniofacial growth patterns present different DA is of clinical importance in determining the timing of the orthodontic treatment intervention; for example, if specific craniofacial growth pattern shows a link to delaying or advancing dental development, the DA may be a tool to malocclusion predicting, once that DA can be estimate from three years old, and skeletal malocclusion can be developed until the end of growth peak3. In this way, DA may be a biological predictor for skeletal malocclusion estabilishment and aids the early diagnosis. The results between the current studies13-24 are divergent, which makes it impossible to reach a robust conclusion on the topic. Thus, this study aimed to investigate the association between skeletal malocclusion patterns, cephalometric measurements and DA in a group of Brazilian children.
METHODS
This cross-sectional multicentric study was delineated through the STrengthening the Reporting of OBservational studies in Epidemiology (STROBE) checklist.
A sample size calculation was performed through the software G*Power Version 3.1.9.7. The parameter to calculate was obtained from the study of Esenlik et al (2014)16. Considering the r = 0.20516, α = 0.05 and β = 0.80, the sample size calculation predicted a minimum of 145 individuals for this study.
The sample was recruited by convenience in orthodontic clinics of Faculdade de Odontologia de Ribeirão Preto, Univerisdade de São Paulo, and Universidade Federal Fluminense, Nova Friburgo. Healthy children and teenagers biologically unrelated, both of sexes, no history of orthodontic treatment or dental extraction, and aged ranged between 7 and 16 years old were recruited by a simple invitation. None remuneration or benefits was offered, and none impact occurs on the orthodontic treatment of patients who did not accept to participate in the research. Patients with dental agenesis of the seven permanent teeth (central and lateral incisors, canine, first and second premolars, and first and second molars) in both sides of the mandible were excluded. Patients with significant geometric distortion and numerous artifacts in their radiography were also excluded. The informed consent was obtained from both parents (or legal guardians) and children (informed assent) included in this study by handwritten signature. Children received an age-appropriate document and also indicated their assent with a handwritten signature. All patients were capable of reading and writing.
Dental age (DA) determination
DA was estimated according to Demirjian (1976)3. The development of the seven permanent teeth (central and lateral incisors, canine, first and second premolars, and first and second molars) of the mandibular left side was assessed using panoramic radiographs. If a missing tooth/tooth agenesis was founded in the mandibular left side, the right side was evaluated. If bilateral missing tooth/tooth agenesis was founded, the child was excluded. Third molar agenesis was not an exclusion criterion, as Demirjian’s method does not assess this tooth. Two observers were previously trained by an orthodontist and calibrated to standardize the evaluations. Ten percent of the sample was randomized selected to calibration. Weighted Cohen’s Kappa was performed twice, to test the intra and inter-observers’ reliability (concordance) of dental age estimation. The concordance tests (Kappa scores) ranged from 0.80 to 1.0029.
Skeletal malocclusion patterns determination
Digital cephalometric tracings were performed by two orthodontists previously calibrated. Steiner’s angles were measured to diagnoses skeletal malocclusion patterns. The relationship of maxilla to the cranial base was assessed by the angle formed by the landmark’s SNA (Sella, Nasion and subspinale point A), and the sample was classified as well-positioned maxilla (80°-84°), maxillary retrusion (<80°), or maxillary protrusion (>84°). The relationship of the mandible to the cranial base was assessed by the angle formed by the landmark’s SNB (Sella, Nasion and supramentale point B), and the sample was classified as well-positioned mandible (78°-82°), mandibular retrusion (<78°), or mandibular prognathism (>82°). The relationship of maxilla to the mandible was assessed by the angle formed by the landmark’s ANB (subspinale point A, Nasion and supramentale point B), and the sample was classified as class I (0°-4°), class II (>4°) or class III (<0°). Ten percent of the sample was randomized selected to calibration. The Intraclass Correlation Coefficient (ICC) was performed twice, to test the intra and inter-observers’ reliability of cephalometric tracings. The ICC scores ranged from 0.79 to 0.8729.
Statistical analysis
Chronological age (CA) was considered as the decimal number obtained by subtracting a patient’s birthdate from the date of radiograph accessed. To determine the agreement between the DA estimation and the CA, Bland-Altman analysis was performed according to Esan and Schepartz (2018)30. Firstly, a linear regression was performed to evaluate proportional bias between CA and DA, and bias was not observed in the regression (p>0.05). After, a Bland-Altman plot was generated (figure 1).
Figure 1 : Bland-Altman Graph to evaluate proportional bias between Chronological Age and Dental Age
The difference (delta) between DA and CA was the dependent variable (DA-CA). The normality of the data was assessed by the Shapiro Wilk test. None variable has normal distribution (p<0.05), thus, the non-parametric tests Mann-Whitney and Spearman correlation test were performed. Also, Generalized Linear Regression adjusted by sex and age was performed29.
All analyses were made using IBM SPSS version 25.0 (IBM Corp. Armonk, USA), and values of p<0.05 indicated a statistical significance difference.
RESULTS
One hundred forty-five children were included, 61 (42.1%) boys and 84 (67.9%) girls. The characteristics of the studied sample are showed in table 1. When evaluated the CA-DA per sex, the mean for the boys was 0.73 (SD=0.88), and for the girls was 0.55 (SD=1.17). The difference between sexes was not statistically significant according to Mann-Whitney test (p=0.267) (table1).
Table 1 : Characteristics of sample recruited in southern of Brazil, during 2016 to 2018
| Chronological Age (CA) | Dental Age (DA) | Delta DA-CA | SNA | SNB | ANB | |
|---|---|---|---|---|---|---|
| Mean | 12.55 | 13.18 | 0.62 | 82.69 | 79.20 | 3.48 |
| Standard Deviation | 1.93 | 1.78 | 0.60 | 4.29 | 4.56 | 3.01 |
| Median | 12.49 | 13.00 | 0.60 | 82.70 | 79.10 | 3.20 |
| Minimum | 7.49 | 8.48 | -2.83 | 94.20 | 93.80 | -3.90 |
| Maximum | 16.08 | 16.00 | 4.75 | 72.00 | 66.50 | 11.50 |
Spearman correlations was performed between DA-CA and cephalometric measurements and are presented in the table 2. No statistical significant association was founded (p > 0.05) (table2).
Table 2 : Correlation between DA-CA and cephalometric measurements
| r2 (95% Confidence Interval) | p-value | |
|---|---|---|
| SNA | 0.006 (-0.16 to 0.17) | 0.937 |
| SNB | -0.019 (-0.18 to 0.15) | 0.815 |
| ANB | 0.089 (-0.08 to 0.25) | 0.289 |
Note: Spearman correlation test was performed.
The mean of DA-DA was compared between sexes and skeletal malocclusion patterns groups (table 3). No statistical difference between groups was observed (p>0.05) (table 3).
Table 3 : Comparison between DA-CA and groups (southern of Brazil, during 2016 to 2018)
| n (%) | Mean | Standard Deviation | Median | p-value | |
|---|---|---|---|---|---|
| Male | 61 (42.1) | 0.73 | 0.88 | 0.63 | 0.267 |
| Female | 84 (57.9) | 0.55 | 1.17 | 0.57 | |
| Well-Positioned Maxilla | 52 (35.9) | 0.66 | 0.81 | 0.60 | Reference |
| Maxillary Protusion | 44 (30.3) | 0.56 | 1.13 | 0.59 | 0.805 |
| Maxillary Retrusion | 49 (33.8) | 0.65 | 1.24 | 0.62 | 0.870 |
| Well-Positioned Mandible | 40 (27.6) | 0.79 | 1.10 | 0.64 | Reference |
| Mandibular Retrognathism | 61 (42.1) | 0.65 | 1.06 | 0.61 | 0.662 |
| Mandibular Prognathism | 44 (30.3) | 0.43 | 1.03 | 0.53 | 0.387 |
| Class I | 69 (48.9) | 0.61 | 0.85 | 0.60 | Reference |
| Class II | 55 (39.0) | 0.71 | 1.28 | 0.64 | 0.627 |
| Class III | 17 (12.1) | 0.62 | 0.84 | 0.52 | 0.965 |
The multivariate analysis adjusted by sex and chronological age evaluated the impact of cephalometric measurements on the DA, however, no significant association was observed (table 4).
DISCUSSION
The main findings of this study indicate that dental maturation, assessed by Demirjian’s method, is not associated with skeletal malocclusion patterns according to Steiner’s cephalometric measurements (ANB, SNA, and SNB angles). To the best of our knowledge, only Esenlik et al. (2014)16 investigated the association between DA and the same parameters. Just like us, the authors did not find any association. This result suggests that DA estimation may not be a reliable biological predictor for the development of skeletal malocclusion in Brazilian children and adolescents.
Our study included 145 children and teenagers, being sixty-nine (48.9%) skeletal Class I malocclusion, 55 (39%) skeletal Class II malocclusion, and 17 (12.1%%) skeletal Class III malocclusion. The mean difference between Chronological age and Dental age (CA-DA) for this study was 0.62 (0.73 for the boys and 0.55 for the girls). These findings indicate an overestimation of the Demirjian’s method on Brazilian population, highlighting the need for specific adjustments when applying this method to the Brazilian population.
Our study is relevant to clinical practice and contributes to improving the oral health care of children and teenagers, especially in determining the ideal timing for orthodontic interventions. We suggest, according to our study, that orthodontists should focuses on other indicators, such as genetic polymorphisms, to predict the skeletal malocclusion development.
Previous similar and divergent results
The overestimation between CA-DA showed by this study is according to a previous study31, which also explored the DA in children from Southeastern Brazil. Besides that, a systematic review with the global population showed that Demirjian method overestimate the age of children by more than six months32. Several factors can explain this overestimation: the method by Demirjian et al. (1976)3was created according to a French–Canadian data set. The inconsistency of this method when applied in other populations may be due to genetic, socioeconomic, and environmental variations between the nations32.
Several studies13,14 subdivided the results between DA and skeletal malocclusion per sexes. In our study, we observed that the difference between CA-DA and sexes was not statistically significant (p=0.267). Demirjian’s method contain a previous adjust to obtain the dental age differently for the boys and girls, which may explain the absent of statistical difference between sexes.
In this way, we chose not to perform a stratified univariate analysis to avoid that some subgroup be underrepresented. We consider the sex as a confounding variable in multivariate analysis.
Our aim was to investigate whether imbalances in dental maturation could serve as a biological indicator for skeletal malocclusion. Some studies have already investigated this topic13-24 using another skeletal malocclusion diagnosis parameters. We decided use the skeletal malocclusion diagnosis according to Steiner’s (ANB, SNA, and SNB angles) because it is the method most widely used worldwide. Esenlik et al. (2014)16was the only study that also investigated these same parameters. They also found no statistically significant association between dental age and skeletal malocclusion patterns. Some studies13-15,17-24 that founded statistical association between DA and skeletal malocclusion containing several bias that prevent a robust conclusion on the topic, such as absent of sample size calculation, exclusion criteria and skeletal malocclusion diagnosis poorly defined, and absent of multivariate analysis. In this way, the conclusion about this topic is limited.
Erudition of the authors and strengths of this study
Several studies by our research team have investigated factors that can predict the establishment of malocclusion6,12,33-35. Our research team have evaluating the impact of single nucleotide polymorphisms, a type of genetic variation, in dental and skeletal maturation12. We showed that polymorphisms associated previously with skeletal malocclusion were associated with skeletal maturation delayed. In this way, we hypothesized that dental maturation and skeletal malocclusion share of similar risk factors, and we decided to investigate this hypothesis here. Our study distinguishes of previous study in some aspects: we performed a sample size calculation, evaluated a population not yet explored, the researchers were adequately calibrated, and we performed a multivariate analysis. This strengths contribute to a result similar to the population reality.
Study’s limitations and suggestions for future research
Several factors were associated with dental development disruptions, malnutrition or obesity, hormonal deficiency, and genetic mutations5-11. Some local factors, such as early tooth extractions, absence of space to tooth permanent eruption, endodontic infections, or bruxism, can alter tooth maturation and do not necessarily change the development of malocclusions. Due to the difficulty in obtaining the patient’s oral health history, the information about local factors was not collected, and it can be considered a limitation of this study. Another limitation is that dental maturation is a long process that would be better investigate in cohort studies.
Bias selection was a limitation of our study, since only orthodontic patients were included in this study. Besides that, we could have made an specific adjustment to Demirjian analysis to our population to decrease the overestimation. We encourage that the future research focus in third molar development and skeletal malocclusion, to evaluate if the development of this tooth may be impacted by specific malocclusion. This research encourages the development of more accurate diagnostic tools tailored to local contexts. This ultimately enhances early diagnosis and personalized treatment planning, potentially reducing the risk of inappropriate interventions and improving overall orthodontic outcomes for young patients.
CONCLUSION
We concluded that:
Tooth maturation was not associated with skeletal malocclusion patterns according to Steiner analysis (ANB, SNB and SNA);
The dental age may not be a good biological indicator to predict the development of malocclusion;
Demirjian method overestimate the dental age of children and adolescents from southeastern Brazil.
Cohort studies should be developed to secure the certainty of this evidence.
Disclosures
Ethics approval and consent to participate: Informed consent/assent was obtained from all participants and/or their legal guardians where age-appropriate. This study was previously approved by the Research Ethics Committees of the Faculdade de Odontologia de Ribeirão Preto, Universidade de São Paulo (protocol 01451418.3.0000.5419) and the Universidade Federal Fluminense, Nova Friburgo (protocol 33791314.3.0000.5243).
