The mandibular arch form at the levels of both the application point of the orthodontic bracket and the basal bone in adults and children with Class I malocclusion and Class II Division 1 malocclusion was investigated.
One hundred thirteen pretreatment mandibular casts were scanned to generate a 3-dimensional computer model of each cast. The casts were divided into Class I and Class II Division 1 malocclusion groups, and were further divided into adults (age, ≥25 years) and children (age, ≤18 years). Two reference points, FA and WALA, were assigned for each tooth. The FA and WALA arch forms were compared, and the distances between corresponding points and intercanine and intermolar widths were analyzed.
The mandibular intercanine FA point widths were significantly greater in the Class II Division 1 malocclusion group than in the Class I malocclusion group ( P <0.05) and were also significantly greater in the Class I adults than in the Class I children ( P <0.05). Both the canine FA and WALA point distances and the molar FA and WALA point distances were moderately to highly correlated (R 2 >0.55) and highly significant ( P <0.001) for all groups. The FA and WALA curves for all groups had individual differences, especially in the premolar and molar areas.
The Class II Division 1 mandible is essentially the same as the Class I mandible with respect to basal bone and dental arch dimensions. WALA points can be used to predict individual dental arch forms in adults and children. Dental and basal arch forms were not significantly different between adolescents and adults.
The size and shape of the dental arches have considerable implications for orthodontic diagnosis and treatment planning. An important goal of orthodontic treatment is to establish a dental arch form that is in harmony with the underlying, supporting basal bone. Studies to date that investigated the relationship between the dental arch form and the basal bone arch form used different methods for defining these arch forms, with conflicting results.
In 1925, Lundstrom coined the term “apical base” to refer to the junction of the alveolar and basal bones of the maxilla and the mandible near the apices of the teeth. He believed that the movement of teeth with mechanical force to achieve normal occlusion is not necessarily accompanied by growth of the apical base. He further postulated that this disharmony between the position of the teeth and the supporting bone led to an occlusion that cannot be maintained. Seventy-five years later, Andrews and Andrews defined what they called the WALA ridge as the band of keratinized soft tissue directly adjacent to the mucogingival junction. The WALA ridge served as a clinically observable structure representing the apical base described by Lundstrom. Andrews and Andrews assigned points to the midfacial axes of the teeth (FA points) to define the dental arch form and then defined the basal arch form by assigning corresponding points along the WALA ridge that were directly beneath the FA points. Using this method, they could investigate the arch form of the basal bone and the arch form characterized by the sites of the orthodontic brackets.
Studies of differences in the transverse dimensions of the mandibular arch form in subjects with different classes of occlusion commonly assess intercanine and intermolar widths. However, many of these studies have obtained conflicting results regarding the transverse dimensions of the mandibular arch. Staley et al compared arch widths of subjects with normal occlusion with those with Class II Division 1 malocclusion using canine and molar cusp tips as landmarks. They reported no differences in mandibular intercanine widths between the 2 groups, although male subjects with normal occlusion had significantly larger mandibular intermolar widths than did male subjects with Class II Division 1 malocclusion. Sayin and Turkkahraman compared the mandibular dental casts of Class I subjects with those of Class II Division 1 malocclusion subjects. Cusp tips were used as landmarks for the canines, premolars, and molars. They found no significant difference in interpremolar and intermolar widths. However, they found mandibular intercanine widths to be significantly larger in the Class II Division 1 group than in the Class I ideal occlusion group. Uysal et al compared arch widths of subjects with normal occlusion, Class II Division 1 malocclusion, and Class II Division 2 malocclusion. They also reported significantly greater mandibular intercanine widths in the Class II Division 1 group compared with the normal occlusion group. Frohlich compared mandibular intercanine and intermolar widths of children with Class II malocclusion with children with normal occlusion and found no significant difference in either intercanine or intermolar width between the 2 groups. Al-Khateeb and Alhaija compared tooth sizes, arch widths, and arch lengths of malocclusion patients with Class I, Class II Division 1, Class II Division 2, and Class III with all patients 13 to 15 years of age. Arch widths were assessed by using buccal and lingual cusp tips as well as the central fossae as landmarks for the molars and premolars and the cusp tips for the canines. They reported no significant differences in mandibular intercanine widths or arch lengths between Class I and Class II Division 1 malocclusions. However, they reported that the first premolar interarch width in the Class II Division 1 malocclusion group was significantly smaller than in the Class I malocclusion group.
These inconsistent findings might be attributed to differences in inclusion criteria used for the study samples, such as age or severity of malocclusion. Additionally, a wide range of anatomic landmarks has been used for the measurements. Consequently, significantly different results have been obtained when applying these different measurement methods to the same dental casts.
Recently, Ronay et al compared the FA and WALA points derived from arches using 3-dimensional (3D) computer models of mandibular casts from 35 Class I malocclusion patients. They found a highly significant correlation of WALA and FA point widths in the canine and molar areas, showing that the practitioner could estimate the FA-derived arch form from the WALA-derived arch form for a patient. They also concluded that FA and WALA arch forms are highly individualized and therefore cannot be defined by 1 generalized formula. Ball et al, using methods identical to those of Ronay et al, investigated the arch forms derived from WALA and FA in 35 Class II Division 1 malocclusion patients and made similar conclusions.
It is likely that mandibular remodeling and growth of the jaws can change the relationship between the dental and basal arches. Further studies comparing the dental and basal arch forms in adolescents vs adults and other classes of malocclusion would help to determine whether the findings of Ronay et al and Ball et al apply to patients of different ages and different types of malocclusion. In this study, we compared the mandibular arch form at both the level of the clinically relevant application points of the orthodontic bracket (FA points) and the level of the underlying anatomic structure of the basal bone (WALA points) in adults and children with Class I and Class II Division 1 malocclusions.
Material and methods
Sixty-three pretreatment mandibular casts from patients with Class I malocclusion (mean age, 23.17 ± 12.02 years) and 58 pretreatment mandibular casts from patients classified as Class II Division 1 (mean age, 22.24 ± 11.97 years) were selected from a sample of 750 patients for this study. Portions of the data collected by Ronay et al and Ball et al were used for the Class I and Class II Division 1 samples, respectively. Both groups were further divided into adults (age, ≥25 years) and children (age, ≤18 years). The Class I group included 31 adults (mean age, 34.09 ± 7.32 years) and 32 children (mean age, 12.6 ± 1.56 years). The Class II Division 1 group included 28 adults (mean age, 33.31 ± 7.44 years) and 30 children (mean age, 11.17 ± 0.99 years). Previous studies found few difference between the sexes, and therefore the data were combined. The sample size for each group was calculated based on an alpha significance level of 0.05 and a beta of 0.1 to achieve 90% power. The ad-hoc power analysis showed that 25 patients in each group were needed. Subjects were selected to participate in the study by visual inspection of dental casts and review of treatment records. Inclusion criteria for the Class I malocclusion group required patients to be skeletal and dental Class I with an ANB angle of 0° to 4°, and have Class I molar and canine relationships. To be included in the Class II Division 1 malocclusion group, patients were required to be skeletal and dental Class II with an ANB angle of >4°, and have Class II molar and canine relationships. Additional inclusion criteria for both groups were fully developed permanent dentitions from first molar to first molar with no crossbites; no prosthetic crowns, large restorations, or gingival defects; and minimal crowding or spacing, with no tooth size-arch length discrepancy greater than 2 mm. Second molars were excluded from consideration because of the inability to ensure complete eruption in patients of all ages. Also, casts with unidentifiable mucogingival junctions were excluded from the sample. Approval for this study was granted by Harvard Medical School and Tufts University School of Dental Medicine.
Three-dimensional computer models of the dental casts were generated with a laser scanning unit. The setup consisted of a computer-assisted noncontact high-definition 3D scanning system including a laser-scanning unit (Dental Plaster Model Shape Scanning System, Surflacer model VMS-100f UNISN, Osaka, Japan), computer-aided-design software program (Dent-Merge, version 5.0, UNISN) and a dental cast analyzing software (Surfacer version 9.0, Imageware, Structural Dynamics Research Corporation, Milford, OH). This system was used to create and edit 3D digital models of the dental casts and identify anatomic landmarks for arch-form characterization. Ball et al described the accuracy and performance characteristics of this system in a previous study.
The laser-scanning unit had 4 main components: a slit-ray laser projector, 2 sets of video cameras, 2 x-y object tables and an r-table to measure the circumference of the object. The mandibular dental casts were scanned at 3 angles in the frontal and sagittal planes. Each dental cast required 60 to 80 minutes to be scanned. Scanned images, including about 90,000 sets of x, y, and z coordinates per cast, were captured by the computer software. The data from the coordinates of both the frontal and sagittal planes were then joined and manually corrected for scanning artifacts. Finally, a 3D model of the entire cast was generated by using the cast-analyzing software (Surfacer, Imageware, Structural Dynamics Research Corporation).
Using the Surfacer software, the anatomic reference points were subjectively identified by the primary author (D.G.). The reference points included the FA points and WALA points, defined below.
For all teeth except the molars, the FA point is defined as the most prominent part of the central lobe of the clinical crown or the midpoint of the facial axis of the clinical crown. The FA point for the first molars is the most prominent point of the clinical crown in line with the mesiobuccal groove that separates the 2 large facial cusps. The FA point is the point on the crown where an orthodontic bracket would be placed for a fully preadjusted appliance system.
The WALA point is defined as the point along the WALA ridge directly beneath the FA point of each tooth. The WALA ridge is a band of soft tissue immediately superior to the mandible’s mucogingival junction. It is located at or nearly at the same vertical level as the horizontal center of rotation of the teeth in an arch. The primary author was calibrated with Ball et al and Ronay et al to ensure that the points were selected in the same manner. To evaluate the reliability of landmark location, 10 mandibular dental casts were randomly selected from the sample. The FA and WALA points were determined twice by the primary author to evaluate intraoperator reliability, 2 weeks apart, on the right and left sides of the dental arch, totaling 120 teeth. Means and standard deviations of the 3D distance between 2 locations were calculated for reliability of landmark location. The same determination was made once by Ball et al and Ronay et al to evaluate interoperator reliability. The intraoperator and interoperator reliability values for the FA point were 0.27 mm (SD, 0.22 mm) and 0.49 mm (SD, 0.63 mm), respectively. The intraoperator and interoperator reliability values for the WALA point were 0.71 mm (SD, 0.42 mm) and 0.95 mm (SD, 0.75 mm), respectively.
Each set of points was digitized as x, y, and z coordinates with the Surfacer software. The sets of points were then exported into Excel software (version 2007, Microsoft, Redmond, Wash) with an ASCII format. A standard graph format was used to create graphs of both WALA and FA arch forms by connecting the individual FA and WALA point values by linear interpolation. Distances between the FA and WALA points for each tooth were then calculated. Mandibular intercanine and intermolar widths for the FA and WALA points were also calculated.
The data were analyzed with the Excel software. Statistical analyses of the mean distances and standard deviations between FA and WALA points and a comparison of the mandibular intercanine and intermolar widths for all groups were performed with independent samples t tests. Linear regression and correlation analyses were also used to assess the correlation between FA and WALA point distances at the canine and molar areas for the groups.
The mean distances and standard deviations between corresponding FA and WALA points for the Class I and Class II Division 1 malocclusion are shown in Figure 1 , and similar data for adults and children are shown in Figure 2 . The graphs illustrate that, for all groups, the FA points were located more lingually (positive values) in the premolar and molar areas in relation to the corresponding WALA points, and more buccally (negative values) in the area of the central and lateral incisors. The Class I malocclusion group had a negative mean distance at the canines, whereas the Class II Division 1 malocclusion group had a positive value. Similarly, the entire adult group had a negative mean distance at the canines, and the entire group of children had a positive value. An independent samples t test showed that no difference between the Class I and Class II Division 1 malocclusion groups, or between the adult and child groups, was statistically significant ( P <0.05).