Mandibular asymmetry types and differences in dental compensations of Class III patients analyzed with cone-beam computed tomography

Because of the increasing interest and demand in patients to correct facial asymmetry, orthodontic and surgical modalities are constantly being developed to improve treatment results. The accurate diagnosis in skeletal discrepancies and dental compensations of the patients can provide a guide for sufficient dental decompensation relative to the planes of each jaw, and this can help in achieving orthognathic surgery with accurate symmetric positioning of the jaws relative to cranial reference planes. Tooth position may play a crucial role as a guide in determining jaw position during surgery; thus, the accomplishment of proper tooth positioning by sufficient decompensation is one of the most important aspects leading to a successful outcome.

Regarding dental compensations in facial asymmetry, previous studies have highlighted the buccoversion of the maxillary molars and linguoversion of the mandibular molars on the deviated side (Dv), extrusion of the maxillary molar on the nondeviated side (NDv), and incisor tipping.^1–6  In addition, these dental changes can be accentuated or lessened depending on the positional variations of bones, particularly the mandible.^3,4  The mandible can be shifted to one side or rotated around horizontal and/or coronal planes, also known as roll or yaw rotation, and this can lead to different amounts and compositions of vertical, rotational, and transverse dental compensations of the maxillary and mandibular dentitions. Previous three-dimensional (3D) computed tomography studies have attempted to categorize the types of asymmetry.^7–9  However, detailed categorization methods were not provided or their classifications were based on only horizontal aspects of the mandible. With these methods, specific features of dental compensation of each type could not be characterized fully. Therefore, the categorization based on 3D rotation generally used in orthodontics, such as rolling or yawing, can be more effective in representing characteristics of mandibular asymmetries. A detailed understanding of dental compensations according to well-categorized asymmetry types may play a crucial role in the successful treatment of facial asymmetry.

Therefore, to clarify dental compensations based on different asymmetry types, this study investigated the distances (vertical, transverse, and anteroposterior) and angles of the incisors and first molars in skeletal Class III patients with roll-, yaw-, and translation-dominant type mandibular asymmetries. The null hypothesis was that there would be no significant difference in dental compensation among the three mandibular asymmetry types.

This retrospective study was approved by the institutional review board of Kyungpook National University Dental Hospital (KNUDH 2022-08-01-00).

The necessary sample size was calculated based on a previous cone-beam computed tomography (CBCT) study³  on skeletal and dental variables in patients with facial asymmetry using G*power (version 3.1.9.7; Heinrich Heine University of Düsseldorf, Düsseldorf, Germany). With a two-tailed significance level of .05, an effect size of 0.75, and a test power of 0.80, the estimated sample size was calculated to be at least 29 patients in each group; thus, to increase the power of this research, 30 patients were assigned to each group.

The inclusion criteria were patients with skeletal Class III relationship (Point A-nasion-point B angle [ANB] < 0°), no congenitally missing teeth except for the third molars, no dental prosthesis, no spacing, and tooth size–arch length discrepancy <3 mm. The exclusion criteria included patients with a history of prior orthodontic treatment, orthognathic surgery, or craniofacial syndrome or trauma.

This study comprised a total of 90 patients (63 men, 27 women; mean age, 22.00 ± 3.31 years; range, 18–37.9 years) with facial asymmetry (>4 mm menton [Me] deviations relative to the midsagittal plane [MSP]) who were diagnosed in the Department of Orthodontics at Kyungpook National University Dental Hospital, Daegu, Korea, between January 2010 and December 2021.

The sample was divided into three groups based on the extent of mandibular rolling (the angle between the mandibular horizontal plane [MHP] and Frankfort horizontal plane [FHP]) and yawing (the angle between the mandibular midsagittal plane [MnMSP] and MSP). The roll-dominant group included patients with moderate-to-high mandibular rolling (≥5° between the MHP and FHP) and low mandibular yawing (<3° between the MnMSP and MSP); the yaw-dominant group included patients with moderate-to-high mandibular yawing (≥5° between the MnMSP and MSP) and low mandibular rolling (<3° between the MHP and FHP); and, finally, the translation-dominant group included patients with low mandibular rolling and yawing (<3° between the MHP and FHP and <3° between the MnMSP and MSP) and >4 mm mental foramen (MF)–mid deviation relative to the MSP (Figure 1).

The CBCT scans (120 kVp, 15 mA, 19-cm field of view, 0.377-mm voxel size, 9.6-second scan time) were acquired using a computed tomography scanner (CB MercuRay, Hitachi, Osaka, Japan). All measurements were assessed using Invivo 6 Anatomy imaging software (Anatomage, San Jose, Calif).

Definitions of landmarks and reference planes are described in Table 1 and Figure 2. The FHP and MSP were used as cranial reference planes in this study. For the mandible, the MHP and MnMSP were used to measure the angles of mandibular rolling or yawing and dental variables.

The skeletal variables were measured to compare the skeletal characteristics among the groups (Table 2, Figure 3). To investigate dental compensation, the distances (vertical, transverse, and anteroposterior) and angles of the central incisors and first molars of the jaws were measured relative to the reference planes (Table 2, Figures 4 and 5).

A single investigator (Dr Kim) performed all measurements, and 15 randomly selected patients were remeasured by the same investigator after 4 weeks. The intraclass correlation coefficient was 0.987 (mean; range, 0.976–0.998), and the Dahlberg errors were 0.49 mm (mean; range, 0.32–0.58) and 0.78° (mean; range, 0.76–0.81). Because of the confirmation of the normality of data using the Kolmogorov–Smirnov test, one-way analysis of variance with a Tukey post hoc test was used to compare each variable of the three groups. A linear-by-linear association was used to compare the sex distribution of the sample among the groups. A paired t-test was used for the comparison of the variables at the Dv and NDv. All statistical analyses were performed using SPSS statistical software (version 22; IBM, Chicago, Ill), and the significance level was set at P < .05.

The samples showed no significant difference in sex distribution, age, or cephalometric measurements among the three groups (Table 3).

Regarding 3D skeletal measurements (Table 4), the amount of Me deviation was significantly greater in the roll- and yaw-dominant groups (roll, 9.40 ± 2.94 mm; yaw, 8.65 ± 3.25 mm) compared with the translation-dominant group (6.24 ± 1.70 mm; P < .01). The roll-dominant group showed significantly greater values of the bilateral difference (ΔNDv−Dv) in ramus height (8.45 ± 3.73 mm) and angle between the MHP to FHP (6.45 ± 1.50°) than the other groups (P < .001). The yaw-dominant group exhibited significantly greater values of the ΔNDv−Dv in body length (5.40 ± 2.77 mm; P < .001) and angle between the MnMSP to MSP (7.63 ± 1.95°; P < .001) than the other groups. For the comparisons of bilateral measurements, all samples showed significant differences between the sides, excluding in the distance between the condylion and MSP.

Regarding the dental linear measurements (Table 5 and Figure 6), the roll-dominant groups showed significantly higher ΔNDv−Dv in the vertical distance of the maxillary (UM, 2.42 ± 1.24 mm) and mandibular first molars (LM, 2.23 ± 1.28 mm) than the yaw- and translation-dominant groups (P < .001). In a comparison of these measurements between the sides, the roll-dominant groups showed a significant difference in both jaws (P < .001); however, the other groups presented significance for the UM only (P < .01), not for the LM. Regarding the transverse dental distance relative to the MSP, the deviations of the midpoint of both maxillary (UI, 2.19 ± 1.51 mm) and mandibular incisor edges (LI, 6.62 ± 2.68 mm) were greater in the yaw-dominant groups than those of the other groups. The transverse distances of the UM and LM relative to the MSP were significantly longer at the Dv than those at the NDv in all samples (P < .001); however, the ΔNDv−Dv in the distance was not significantly different among the groups. Regarding the distance from the MnMSP, the yaw-dominant group showed the greatest amount of LI deviation toward the NDv (−2.11 ± 1.39 mm) among the three groups (P < .001). In addition, the LM distance at the Dv was significantly shorter than that at the NDv in all patients (P < .01); however, no significant difference in the ΔNDv−Dv of the LM distance was observed among the groups. For the anteroposterior distance relative to the coronal or mandibular coronal plane, no significant difference was found among the groups and between the sides, except for the LM distance from the mandibular coronal plane of the yaw-dominant group (P < .001). This indicated that the yaw-dominant group exhibited mesial and distal crown tipping of the mandibular molars at the Dv and NDv, respectively; however, not for the other groups.

Regarding tooth inclinations, tipping of the maxillary incisor was significantly greater in the roll-dominant group (3.22 ± 2.82°) than that of the translation-dominant group (1.07 ± 2.06°; P = .01). The maxillary first molar inclination at the Dv was significantly greater than that at the NDv in all groups (P < .001); however, there was no significant difference in the ΔNDv−Dv of the molar inclination among the groups. Based on the MnMSP and MHP, the yaw-dominant group yielded the greatest tipping of the mandibular incisor (−4.13 ± 3.30°) compared with the other groups (P < .001); the inclination of the mandibular first molar demonstrated a significantly lower value at the Dv than at the NDv (P < .05), indicating molar lingual tipping at the Dv. In addition, the roll-dominant group showed the lowest value of ΔNDv−Dv in the mandibular molar inclination (2.28 ± 4.84°) among all groups (P < .001; yaw, 9.05 ± 5.94°; translation, 6.83 ± 4.86°).

This study investigated the amount of mandibular rolling and yawing to distinguish among the asymmetry types. The body length and ramus height or inclination used in previous studies^4,7  may not fully reflect the mandibular rolling or yawing because the position of the glenoid fossa may affect the mandibular asymmetry.^10–13  For this reason, in this study, the angles between the mandibular and cranial reference planes were used to accurately classify the mandibular asymmetry types. In addition, when establishing these mandibular planes, the mental foramen was used as the reference landmark in this study because of its reliability and stability.^3,14–18 

The roll-dominant group exhibited vertical dental compensation with the extruded posterior teeth at the NDv in both jaws (Figure 6A). Considering the angular dental compensation in the mandible in contrast to the maxilla, the tipping of the teeth toward the NDv was lesser in the roll-dominant group than in the other groups. Thus, it can be inferred that, by the mandibular rolling itself, the mandibular teeth tipped spontaneously toward the NDv; thus, the compensatory tipping of the teeth might be less. Accordingly, the transverse occlusal cant and maxillary tooth tipping should be a focus (Figure 7A). Particularly, in the maxilla, the canted-down molar should be intruded and tipped buccally using elastomeric thread from the buccal microimplant, and/or the canted-up molar should be extruded and tipped palatally using an auxiliary extrusion spring supported by the buccal microimplant or buccal intermaxillary elastics to the lingual button of the ipsilateral mandibular molar.¹⁹  Notably, to achieve successful improvement in patients with roll-type asymmetry, the vertical dental discrepancies between the sides should be completely removed. If not, this may result in a remaining mandibular roll and unsatisfactory surgical outcomes.

The yaw rotation generally represents a large horizontal deviation in the anterior part of the mandible. Hence, compensatory tipping of the incisors might be more predominant in the yaw-dominant group than in the other groups (Figure 6B). The maxillary molars on the Dv showed buccal tipping, whereas the mandibular molars showed lingual tipping. Anteroposteriorly, the mandibular molars tipped mesially at the Dv and distally at the NDv. For dental decompensation in patients with yaw-dominant mandibular asymmetry, buccolingual tipping of the posterior teeth should be corrected including anterior tooth angulation and anteroposterior tipping of the mandibular posterior teeth; however, vertical tooth movement is rarely required (Figure 7B). Hence, the use of Class II intermaxillary elastics on the NDv may be effective in performing appropriate tooth movement anteroposteriorly.

The translation-dominant group had the lowest Me deviation among the three types. Regarding dental compensation, differences were observed in the transverse distance and angle between the sides; however, there were no differences in the vertical distance in the mandible and sagittal difference in both jaws (Figure 6C). Therefore, transverse dental correction is required, including lateral movement and tipping correction. Notably, their extent may be greater in the maxilla than in the mandible (Figure 7C). This dental decompensation should be sufficient to surgically move the posterior side of the mandible toward the NDv.

Collectively, the differential tooth movement based on the mandibular asymmetry types can provide strong assurance of accomplishing better facial symmetry in patients. In addition, after proper dental decompensation and precise jaw surgery based on the reference planes, the supplementary osteotomies for compensatory bone modeling, such as the mandibular inferior border and gonion, can be taken into account to maximize improvement of the face.^20,21 

Although this study provided valuable findings on dental compensation in patients with different asymmetry types, it did not include atypical asymmetry. Hence, dental compensation of the atypical asymmetry type should be evaluated in future research.

• The null hypothesis of this study was rejected, and the asymmetry types showed different and unique features of dental compensations.

• The roll-dominant group showed transverse occlusal canting of both jaws and buccal tipping to the Dv in the maxilla; the yaw-dominant group presented buccolingual tipping of the molars in both jaws and a large midline discrepancy in the mandible; the translation-dominant group demonstrated greater transverse tipping in the maxilla than in the mandible at the Dv.

• A complete understanding of the differences in dental compensation according to asymmetry types may enable clinicians to achieve satisfactory treatment outcomes in patients with facial asymmetry.