Introduction

Based on several examined cases, Angle [3] estimated that Class II malocclusions would have a prevalence of approximately 26% in the population. Since then, epidemiological studies of different populations and ethnic groups have been performed, observing a prevalence of 7–28% [4, 5, 9, 21] in permanent dentition. When only orthodontic patients were evaluated, the prevalence of Class II malocclusions can approach or exceed 50% [6, 31].

The literature shows a number of studies assessing skeletal Class II malocclusions in the sagittal and vertical dimensions [1, 18]. However, little attention has been devoted to changes in the transverse direction, other than those patients with maxillary constriction. This becomes important as it is not unusual for Class II malocclusions to appear only on one side of the arch (i.e., Class II subdivision) [3, 4]. In many of these cases, occlusal disharmony is associated with skeletal asymmetries [19, 23].

It is known, however, that even pleasant faces have a subclinical level of asymmetry [20]. In this context, the term relative symmetry or fluctuant asymmetry may be used as small random variations; it is also widely used as a measure of instability in the development of plants and other animals [22]. However, in moderate or severe asymmetries orthodontists often face the challenge of obtaining final bilateral harmony, either with orthodontic, orthopedic, or surgical approaches [29].

Thus, the identification of morphological components involved in the asymmetric expression is important in mapping out a treatment plan. Currently three-dimensional images allow a comprehensive evaluation of these patients, observing anatomic structures with real size and without anatomical superimpositions [7, 13, 14, 27]. Thus, the aim of this study was to evaluate by cone-beam computed tomography (CBCT) the craniofacial components related with different categories of mandibular asymmetry in adults with skeletal Class II. The null hypothesis is that there are no differences between the deviated and contralateral sides, as well as that there are no differences between different intensities of mandibular asymmetry.

Materials and methods

The study protocol (reference number: 1.591.220) was approved by the Ethics Committee at Universidade do Sul de Santa Catarina (UNISUL, SC, Brazil). All procedures were in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with the Helsinki Declaration of 1964, as updated in 2013.

This study was nested in a previous epidemiological investigation of mandibular asymmetries [28]. CBCT images of 120 orthodontic and orthognathic surgery patients were eligible, and the power calculation for the statistic tests applied demonstrated that this sample size would suffice for this study (β < 0.2 using α = 0.05).

All scans were obtained using the same type of device (iCAT®, Imaging Sciences International, Hatfield, PA, USA), adjusted to operate under the following specifications: extended field of view (16 × 22 cm or 17 × 23 cm) 120 KvP, 3-8 mA, and 0.4 mm voxel size. Patients were asked to occlude at maximum intercuspation and relax their lips.

The images of the CBCT were exported in DICOM format (Digital Imaging and Communication in Medicine), using the iCAT Vision® software. The DICOM files were imported into the SimPlant Ortho Pro® 2.0 software (Dental Materialise, Leuven, Belgium). Landmarks were located using the multiplanar reconstruction view, with a measurement scale of 0.01 mm and 0.01°.

The following inclusion criteria were adopted: CBCT images obtained with clinical justification or in case conventional radiographic techniques failed to meet clinical needs, in accordance with the guidelines of the SedentexCT project and the American Academy of Oral and Maxillofacial Radiology [2, 25]; all permanent teeth erupted (with the exception of third molars); subjects ages 19 through 60 years old and with Class II skeletal jaw relationship (ANB > 4.5°, as proposed by Tweed) [30]. The exclusion criteria were the following: a former history of orthodontic treatment, fractures or surgeries in the facial area, degenerative diseases in the temporomandibular joint, craniofacial syndromes and anomalies.

The outcome was categorized into three groups according to the intensity of mandibular asymmetry. Symmetry was defined through the analysis of gnathion deviation, since lateral displacement of the chin shows a higher influence on the perception of facial asymmetry [16, 26]. Regardless of the side of deviation, patients with a gnathion displacement of up to 2 mm in relation to the midsagittal plane were considered to have relative symmetry [15]. Patients with a displacement of more than 2 mm and up to 4 mm were considered to have moderate asymmetry. Lastly, those with a gnathion deviation greater than 4 mm were considered to have severe asymmetry [12, 17]. These cut-off points were previously suggested in the literature [12, 15, 17]. Each category of asymmetry contained 40 individuals, totaling 120 evaluated patients. The landmarks and reference planes used in this study are described in Table 1.

Tab. 1 Landmarks and reference planes used in the study
Full size table

Several measurements were analyzed (mandibular and maxillary components), divided into groups for transverse, sagittal, and vertical evaluation. These measurements were made three-dimensionally and are described in Table 2 and illustrated in Fig. 1.

Tab. 2 Measurements performed to evaluate bilateral differences of mandibular and maxillary components
Full size table
Fig. 1
figure 1

Illustration of the measurements used in the study: a Gn-MSP, Go-MSP, J-MSP, Cap-MSP, UDM, and LDM; b ANB angle, Go-Coronal, Cap-Coronal, GoGn, CoGo, Go-Camper, and J-Camper. For the bilateral points, the measurements were made both in the contralateral (C) and in the deviated (D) sides

Abb. 1 Darstellung der in der Studie verwendeten Messungen: a Gn-MSP, Go-MSP, J-MSP, Cap-MSP, UDM und LDM; b ANB-Winkel, Go-Coronal, Cap-Coronal, GoGn, CoGo, Go-Camper und J-Camper. Bei bilateralen Punkten wurde sowohl auf der kontralateralen (C) als auch auf der abweichenden (D) Seite gemessen

Full size image

The methodology used to determine the midsagittal plane in this study was previously validated by the work of Damstra et al. [8]. The deviation of gnathion from the midsagittal plane was calculated in absolute values, regardless of the side of deviation. For other measurements made in midpoints, a positive value was given when the displacement of the point coincided with the side of the gnathion deviation (deviated side); a negative value was given when the displacement occurred on the opposite side (contralateral side). To determine the asymmetry between the measurements made in bilateral cephalometric landmarks, the difference (/dif) of the contralateral side minus the side of mandibular deviation was analyzed.

To estimate the error of the method, 20% of the sample was evaluated at two different times by a single examiner, with an interval of 2 weeks between evaluations. The intraclass coefficient of correlation (ICC) was used, and a value of >0.75 was obtained for all evaluated measurements (with the lowest ICC being 0.77 and the highest 0.99), thus, demonstrating good reliability of the method.

Statistical analyses were conducted using the SPSS® 20.0 software (IBM, Chicago, IL, USA). The Shapiro–Wilk test was applied, demonstrating a normal distribution of the values obtained in bilateral measurements and the abnormal distribution of the values obtained in midpoint measurements. The values obtained on the contralateral side and on the deviated side were compared using the Student’s t test for paired samples. In order to verify possible differences in the measurements between different categories of mandibular asymmetry, the analysis of variance (ANOVA) was conducted (complemented by the Tukey test) when the data were considered normal. The Kruskal–Wallis test, on the other hand, was conducted when the normality criterion was not met (complemented by the Mann–Whitney test with Bonferroni correction). To determine the correlation of the gnathion deviation with other variables evaluated, the Spearman correlation coefficient was applied. A 5% significance level was considered.

Results

The mean of gnathion displacement in relation to the MSP was, in absolute terms, 3.21 mm [standard deviation (SD) 2.75], varying from 0.01–21.49 mm. Other characteristics of the sample and the analyzed groups can be visualized in Table 3.

Tab. 3 General characteristics of the analyzed sample
Full size table

The null hypothesis was rejected. As to the measurements obtained on the contralateral side and on the deviated side for the bilateral cephalometric measurements, Table 4 presents the means and standard deviations considering the different categories of mandibular asymmetry for these patients. Using the paired samples t test, several differences between the measurements of the contralateral and deviated sides were observed.

Tab. 4 Comparison of the values obtained (in mm) for the contralateral side and the deviated side in bilateral measurements, in each category of mandibular asymmetry
Full size table

A comparison between patients with relative symmetry, moderate asymmetry and severe asymmetry is shown in Table 5. It was possible to verify that the values of Gn-MSP, Go-MSP/dif, UDM, LDM, CoGo/dif, Go-Camper/dif, and J-Camper/dif differed significantly between the groups. For Gn-MSP and Go-MSP/dif, there was a statistically significant difference between all groups. For UDM, LDM, CoGo/dif, Go-Camper/dif, and J-Camper/dif, there was a statistically significant difference between the patients with severe mandibular symmetry and the other ones. Other analyzed variables showed no statistical differences between the groups. It is worthwhile to point out that when the value of the bilateral difference between the measurements (/dif) is negative, this means that the deviated side presented a higher average dimension than that of the contralateral side.

Tab. 5 Comparison of the groups regarding the differences between the measurements (in mm) obtained for contralateral and deviated sides, as well as of measurements from midpoints
Full size table

The correlation test (Table 6) demonstrated that there was no variable significantly correlated to the gnathion deviation for patients with moderate asymmetry. For patients with severe asymmetry, the variables Go-MSP/dif, LDM, CoGo/dif, and J-Camper/dif were found to be significantly correlated to gnathion deviation.

Tab. 6 Correlation between the lateral deviation of the gnathion point and the analyzed measurements for both moderate and severe mandibular asymmetry
Full size table

Discussion

The results from this study highlight that asymmetrical Class II patients have several disharmonic features that affect the expression of their asymmetries. This could be verified by comparing the values obtained on the contralateral side and on the deviated side of those patients. It is important to mention that it is fundamental to discriminate between different categories of mandibular asymmetry, since these differences can distinguish between a compensatory orthodontic/orthopedic treatment and an orthognatic surgery approach [16, 29].

There are certain limitations in the present study. Functional shifts that could lead to a postural component of the mandibular asymmetries were not analyzed. Since the main goal was to address skeletal mandibular asymmetries, mainly skeletal landmarks were used. However, dental components that could affect craniofacial asymmetries were not assessed. Vertical or transverse deviations of the landmarks A, N, and B can pretend changes in the sagittal dimension of a three-dimensional angle [10], and this may have resulted in errors for establishing the ANB angle, specially in the severe asymmetry group. For establishing the groups, mandibular asymmetry was determined by the lateral displacement of the gnathion point. Nevertheless, it is known that mandibular asymmetries can also be characterized in the vertical plane. In addition, since this study presents a cross-sectional design, it becomes impossible to state the order of occurrence of events in time, and therefore it cannot be defined which abnormal structure was the preliminary cause of the mandibular asymmetry. On the other hand, this study offers advancement with a large sample with controlled data, addressing skeletal alterations in three distinct categories of mandibular asymmetry.

When analyzing Table 4, it can be seen that for individuals with relative mandibular symmetry, the only statistical difference between the contralateral and deviated sides was in the transverse and vertical position of the jugale point in the maxilla. Each was less than 1 mm on average and is clinically nonsignificant. For individuals with asymmetries, bilateral differences could be found in the transverse positioning of the gonion and jugale, as well as the vertical positioning of the gonion and rami height. Moreover, for moderate asymmetry, differences between the contralateral and deviated sides were also found in mandibular body lengths and for severe asymmetry in the vertical position of the jugale points. All these variables should be carefully evaluated, especially for severe asymmetric cases that undergo orthognatic surgery, since this knowledge could determine an atypical osteotomy and/or whether a one- or two-jaw procedure would be indicated [24].

As previously stated by other authors [7, 13, 14, 27], our findings suggest that chin deviations in skeletal Class II individuals seem to be associated with bilateral differences in the position of the gonion and jugale. Bilateral alterations in the gonion position can influence differences in rami heights, mandibular corpus lengths, as well as frontal and lateral mandibular ramal inclination. Alterations in the jugale point positioning can influence maxillary cant and/or laterality.

When the categories of mandibular asymmetry were compared (Table 5), the significant differences, when found, were most often between the severe asymmetry and other groups. This was also stated by Masuoka et al. [16] when they compared patients exhibiting relative symmetry, moderate asymmetry, and severe asymmetry. Hence, the results of the present study indicate that severe mandibular asymmetry has a greater upper and lower dental midline deviation, as well as bilateral difference in the heights of mandibular rami and the vertical position of gonion and jugale points when compared to relative symmetry and moderate asymmetry. For the variables that evaluated lateral gnathion deviation and differences in lateral positioning of the gonion points, all groups were different from each other. For other analyzed variables, there were no differences between the groups.

Ramal inclinations have been shown to be related to severe mandibular asymmetries [7, 13], but most often it is not mentioned if those are related to alterations in the positioning of the condyles or the gonion points. The results presented in Table 5 demonstrated that, for severe mandibular asymmetries, the bilateral difference of the sagittal and lateral position of the condyles occurred to a lesser intensity than the bilateral difference on the sagittal and lateral position of the gonion points (Cap-MSP/dif = −0.51 ± 4.36; Go-MSP/dif = −4.34 ± 4.24; Cap-Coronal/dif = −0.22 ± 2.73; Go-Coronal/dif = 1.27 ± 4.34). Therefore, these findings suggest that frontal and lateral mandibular ramal inclination in severe asymmetries results more from the spatial displacement of the gonion points than that of the condyles.

Chen et al. [7] mentioned that for some patients the lateral deviation of the gonion point from the MSP could be greater than the gnathion displacement to the deviated side. Moreover, few patients exhibited a ramus asymmetry contralateral to the side of chin deviation.

This may also be seen in our findings, where for patients with severe asymmetry the bilateral difference of the positioning of the gonion point in the transverse plane was almost as great as chin deviation, and presented a high standard deviation (Table 5: Gn-MSP = 6.02 ± 2.88; Go-MSP/dif = −4.34 ± 4.24). These aspects need to be carefully evaluated, as the surgical approach, if instituted, should result in the best possible skeletal balance.

Correlation tests (Table 6) were only significant for severe asymmetric patients, and determined that chin deviation was significantly correlated with lower dental midline deviation and difference in the gonion lateral positions, difference in the mandibular rami heights, and difference in the jugale vertical displacements.

The results of this study therefore highlight that for asymmetrical Class II patients not only was the chin deviated, but other craniofacial features were disharmonious as well. The gonion area must be evaluated with caution in all three dimensions. Hajeer et al. [11] analyzed soft-tissue asymmetry in patients before and after orthognatic surgery and stated that in Class II patients surgical changes increased asymmetry. Therefore, for those patients all morphological aspects pointed to in the study are important considerations not only for the surgeon, but also for the orthodontist as the goal should be to create dental asymmetry during decompensation equal in magnitude to the skeletal asymmetry.

Often facial asymmetry is more noticeable in Class III patients, due to the projection of the chin; it is important therefore to not overlook asymmetries in Class II patients whose deformity may be hidden by the “drape” of the soft tissue. This misdiagnosis could lead to extended treatment time and compromised outcomes. Therefore, for severe asymmetric patients, the proper evaluation of characteristics related to mandibular lateral deviation is essential to provide patients with the best treatment possible.

Conclusions

For Class II patients with severe asymmetry, chin deviation was significantly correlated with lower dental midline deviation, differences in the gonion lateral positions, differences in the mandibular rami heights, and differences in the jugale vertical displacements.