With the recent interest in esthetics at an early age, prediction of mandibular incisor crowding is of significant importance. Since dental arch development is related to craniofacial growth, we conducted a cohort study to find a regression model for mandibular incisor crowding based on various craniofacial parameters in children.
A total of 250 children, all in the early mixed dentition, were selected randomly by cluster sampling from primary schools. Craniofacial parameters were measured by a caliper bow, and intercanine widths were measured on dental casts. After a 12-month follow-up period, mandibular incisor crowding and intercanine width were assessed on each subject’s dental cast. Discriminant and multiple regression analyses were performed separately for boys and girls.
Of 250 children, 148 returned for the 1-year follow-up and met the inclusion criteria. Regression analyses of patients with normal occlusion showed a statistically significant correlation between anterior dental crowding and facial height and bigonial width in both sexes. A significant inverse correlation was found between initial intercanine width and incisor crowding in girls. Furthermore, using the aforementioned parameters, the occurrence of mandibular incisor crowding could be predicted with an accuracy of 92.6%.
We found that the occurrence and severity of mandibular incisor crowding in the early mixed dentition can be predicted accurately based on certain craniofacial parameters.
Craniofacial parameters can predict occurrence and severity of mandibular incisor crowding.
Intercanine width is greater in children with mandibular incisor crowding.
Crowding of the permanent teeth, especially in the anterior part of the mandible, is one of the most prevalent forms of malocclusion among children. For esthetic reasons, these patients comprise a significant portion of visits to a dental office. Mandibular anterior crowding is the result of a discrepancy between the sum of mesiodistal widths of 4 permanent incisors widths and the available space in the alveolar process. Several factors are related to the development of mandibular incisor crowding in the mixed dentition: arch dimensions, increased intercanine width, and mandibular growth pattern. The ability to predict the development of mandibular incisor crowding, especially in the early mixed dentition, has significant value to clinicians for decisions regarding the beginning of preventive therapy in the form of space management and preservation of the leeway space. In line with the reported correlations between facial and cranial parameters with dental arch changes in the literature, various authors have evaluated the relationship between dental crowding and craniofacial measurements. However, the results of these studies have yielded dissimilar and conflicting results. Since the current literature on this topic consists of cross-sectional studies evaluating a limited number of variants, the need for a well-designed cohort study to find a predictable regression model based on craniofacial parameters seems apparent. To the best of our knowledge, this study is the first cohort study investigating the relationship between incisor crowding and anthropometric parameters. Our aim was to determine whether mandibular incisor crowding can be predicted in the early mixed dentition stage using craniofacial measurements; this could provide a valuable tool for treatment planning.
Material and methods
The individuals enrolled in the study were selected based on a cluster sampling method. To this end, various districts in Shiraz, Iran, were selected as clusters, and the primary schools in each cluster were each given a number. Five schools in each cluster were selected randomly, yielding a total of 1700 white children aged 7 to 8 years. From this population, 250 subjects were selected randomly to take part in the study.
The inclusion criteria for the study were children with a normal molar relationship defined as a flush terminal plane (1 mm deviation of the mandibular molars either mesial or distal from this position was deemed acceptable), an orthognathic growth pattern, erupted mandibular permanent central incisors, exfoliated or mobile mandibular deciduous lateral incisors, bilateral deciduous canines, no evidence of interproximal caries or cuspal wear in the mandibular deciduous canines, and erupted mandibular permanent first molars. The skeletal growth pattern was analyzed based on the positions of subnasale and pogonion relative to the true vertical line from the deepest point on the nasal bridge. Subjects with a history of previous orthodontic intervention and detected oral habits were excluded as were those who began orthodontic therapy during this study. Of the 250 children, 148 returned at the 1-year follow-up. There were many reasons for this drop in attendance. Some children changed schools, and others did not respond to the second visit call. Furthermore, some children were excluded due to loss or extraction of deciduous teeth or starting orthodontic treatment. After explaining the study procedure, written informed consent was obtained from the parents of all participants.
Cranial and facial dimensions (height and width) of the sample population were measured on soft tissue landmarks by a caliper bow (ICS-Spreading Caliper-SPCG01P, Industrial and Commercial Services, Telangana, India). The measurements and the landmarks are summarized in Table I . The measurements were performed 3 times, and an average was reported for each parameter. Some subjects (10% of the total sample population) were reassessed 2 days later to analyze the intraexaminer reliability using the intraclass correlation coefficient (ICC); this was determined to be 0.87. The mean error of the 2 measurements was 0.23 ± 0.21 cm.
|Facial height||From nasion ∗ to menton † measured on the soft tissue|
|Facial width||Distance between the most prominent points of zygomatic bones in soft tissue from the frontal view|
|Cranial height||Summit of glabella ‡ to furthest occipital point|
|Cranial width||Widest measurement of the cranium at right angles to median plane|
|Bigonial width||Measurement from gonion § to gonion on soft tissue|
Mandibular arch impressions of the children were taken at the first visit by mixing 52 g of alginate powder (Tropicalgin; Zhermack, Badio Polesine, Italy) and 40 mL of water for 15 seconds. The paste was immediately placed on a tray while the patient rinsed with warm water. Once the tray was seated on the dental arch, it was kept in place under finger pressure for 90 seconds. The impressions were rinsed with cold water and disinfected using a glutaraldehide solution for 10 minutes. After a final rinse, the impressions were stored in a damp and cool environment for 1 hour and subsequently poured using type IV dental stone (GC Corporation, Tokyo, Japan) where 100 g of powder was hand-mixed with 30 mL of water. The impressions were subjected to vibration for 20 seconds and allowed to set for an hour at room temperature.
Dental casts were used to obtain the mandibular intercanine width. The intercanine width was measured directly on the casts using a digital caliper (Shoka Gulf, Spain) in millimeters with an accuracy of 0.01 mm from the canine cusp tip of 1 side to the other. A secondary alginate impression was obtained at the 12-month follow up and used to acquire dental measurements including mandibular intercanine widths, incisor widths, and available space. The measurements were performed 3 times by a board-certified orthodontist (A.S.), and an average was used. Ten percent of the casts were remeasured, and the intraexaminer reliability was calculated as 0.88 using the ICC method. The mean error of the 2 measurements was 0.38 ± 0.27 mm.
Mandibular incisor crowding was measured on dental casts with a digital caliper. Incisor widths were measured as described by Hunter and Priest. The tips of the digital caliper were introduced from the labial side of the teeth and held incisally parallel to the occlusal plane, and the distances between the anatomic contact points of the teeth were measured. The available incisor space was measured between the mesial surfaces of the deciduous canines by allocating the dental arch into 2 straight-line segments (from the mesial contact point of the mandibular canine to the mesial contact point of the central incisor). To calculate the severity of crowding, the sum of incisor widths was subtracted from the available incisor space. For the prevalence of incisor crowding. the children were classified into 2 groups based on the amount of incisor crowding. Crowding greater than 2 mm was regarded as the threshold for clinical significance because of reports of spontaneous improvement of irregularities less than this amount. All measurements were performed 3 times by a board-certified orthodontist, and an average was used. Ten percent of the casts were remeasured, and the intraexaminer reliability was calculated as 0.91 using the ICC method. The mean error of the 2 measurements was 0.12 ± 0.23 mm.
The Student t test was used to compare the values for the craniofacial parameters and the measurements from the dental casts between boys and girls. Both groups had a normal distribution. To compare the prevalence of mandibular incisor crowding between boys and girls, the chi-square test was applied. The increase in intercanine width was evaluated using the paired t test. To analyze the relationship between the initial intercanine width and the subsequent increase in intercanine width, the Pearson coefficient was used. A stepwise discriminant analysis was performed to render a quick tool for the prediction of patients who will have crowding (>2 mm) later. Stepwise multiple regression to analyze the potency of the craniofacial parameters in predicting the severity of mandibular incisor crowding was performed. For all statistical tests, P values less than 0.05 were considered significant. The tests were performed using the Statistical Package for the Social Sciences (version 18; SPSS, Chicago, Ill).
Of the 250 children, 148—83 boys (56%) and 65 girls (44%),—with a mean gge of 7.4 ± 0.48 years participated in the follow-up. This reduction of the sample was because some participants changed schools and were out of reach or refused to comply, and others were excluded because they began orthodontic treatment or had tooth extractions. Mandibular incisor crowding (>2 mm) was found in 36% of the sample population (boys, 28; girls, 25). Applying the chi-square test, we found no significant difference in the prevalence of crowding between boys and girls ( P >0.05). The amounts of mean crowding of the mandibular incisors were 2.1 mm for girls and 2.7 mm for boys, with no statistically significant difference between them ( P >0.05). Intercanine width measurements were significantly greater in boys at the initial measurement ( Table II ). There was also a significant inverse correlation between intercanine width at the first visit and the amount of increase in this parameter during the 1-year period. The Pearson coefficients demonstrating this correlation were –0.73 for girls and –0.65 for boys. Stronger correlations were present between the 2 parameters in subjects with crowding ( Table III ).
|Boys||Girls||P value ∗||Crowding||No crowding||P value|
|Facial neight)cm)||10.31 ± 0.73||9.92 ± 0.58||<0.001||10.55 ± 0.64||9.91 ± 0.61||<0.001|
|Facial width (cm)||9.48 ± 0.92||8.63 ± 0.52||<0.001||8.75 ± 0.78||9.3 ± 0.87||<0.001|
|Height of head (cm)||16.25 ± 0.63||16.16 ± 0.53||0.28||16.3 ± 0.54||16.17 ± 0.61||0.2|
|Width of head (cm)||12.37 ± 0.67||11.51 ± 0.64||<0.001||11.83 ± 0.79||12.08 ± 0.78||0.07|
|Bigonial width (cm)||6.65 ± 0.65||6.12 ± 0.39||<0.001||5.89 ± 0.33||6.7 ± 0.52||<0.001|
|Tooth size (mm)||20.60 ± 0.92||20.53 ± 0.72||0.59||20.69 ± 0.74||20.5 ± 0.88||0.18|
|Intercanine width (mm)||28.15 ± 1.02||26.42 ± 0.97||<0.001||28.03 ± 1.52||27.69 ± 1.69||0.15|
|Crowding||No crowding||Crowding||No crowding|
The craniofacial parameters selected for this study had significantly greater values in boys ( Table II ). A stepwise discriminant analysis in girls provided a model in which bigonial width, facial height, and initial intercanine width could predict those who would develop mandibluar incisor crowding (>2 mm). The same model in boys included only bigonial width and facial height. The results of the discriminant analysis are shown in Table IV . As is apparent, excluding initial intercanine width, all variables remaining in the models for both sexes were significantly different between the crowding and noncrowding groups. Subsequently, a formula was derived for predicting whether a subject would develop crowding in the future ( Table IV ). The letter D stands for the threshold of the discriminant analysis where one can predict whether crowding will occur if the value obtained from the formula is more negative than the threshold. This threshold number has no units and only has a numeric value for a yes or no answer regarding crowding. The threshold values for the prediction of crowding were −0.396 for girls and −0.412 for boys. We then used the model to see whether it could accurately predict those in our sample who developed crowding. The result for girls was sensitivity of 88%, specificity of 95.3%, and total accuracy of 92.6%. The same analysis in boys resulted in sensitivity of 96.4%, specificity of 90.9%, and total accuracy of 92.8%.
|Sex||Parameter||Crowding||P value||Wilk’s λ||DCF|
|Female ∗||Bigonial width (cm)||5.75 ± 0.22||6.34 ± 0.3||<0.001||0.702||3.401|
|Facial height (cm)||10.3 ± 0.61||9.7 ± 0.44||<0.001||0.474||−1.547|
|Initial intercanine width (mm)||26.6 ± 1.43||26.75 ± 1.1||0.628||0.346||0.305|
|Male †||Bigonial width (cm)||5.99 ± 0.35||6.99 ± 0.48||<0.001||0.804||2.065|
|Facial height (cm)||10.75 ± 0.6||10.07 ± 0.68||<0.001||0.468||−0.722|