Surgical-orthodontic treatment for class II asymmetry: outcome and influencing factors

The study aimed to evaluate the treatment outcome of bimaxillary surgery for class II asymmetry and find the influencing factors for residual asymmetry. Cone-beam computed tomographic images of 30 adults who had bimaxillary surgery were acquired, and midline and contour landmarks of soft tissue and teeth were identified to assess treatment changes and outcome of facial asymmetry. The postoperative positional asymmetry of each osteotomy segment was also measured. After surgery, the facial midline asymmetry of the mandible, chin, and lower incisors improved significantly (all p < 0.01). However, the residual chin deviation remained as high as 2.64 ± 1.80 mm, and the influencing factors were residual shift asymmetry of the mandible (p < 0.001), chin (p < 0.001), and ramus (p = 0.001). The facial contour asymmetry was not significantly improved after surgery, and the influencing factors were the initial contour asymmetry (p < 0.001), and the residual ramus roll (p < 0.001) or yaw (p < 0.01) asymmetry. The results showed that bimaxillary surgery significantly improved midline but not contour symmetry. The postoperative midline and contour asymmetry was mainly affected by the residual shift and rotational jaw asymmetry respectively.


Influencing factors for residual asymmetry.
Stepwise multiple linear regression analysis demonstrated the residual midline deviation of the nose and upper lip was associated with the initial deviation and the residual maxillary shift asymmetry (all p ≤ 0.001). The residual midline deviation of the lower lip was associated with the residual mandibular shift asymmetry (p = 0.001). The residual midline deviation of the mandible was affected by the residual shift and yaw asymmetry of the mandible (both p < 0.001). The residual midline deviation of the chin was affected by the residual shift asymmetry of the mandible (p < 0.001), chin (p < 0.001), and ramus (p = 0.001). The residual lip cant was affected by the residual mandibular roll asymmetry (p < 0.001) ( Table 2).
The residual upper contour asymmetry was associated with its initial asymmetry and the residual ramus roll asymmetry (both p < 0.001). The residual middle and lower contour asymmetry was associated with their initial asymmetry (both p < 0.001) and the residual ramus roll (both p < 0.001) and yaw asymmetry (both p < 0.01) ( Table 2).
The residual upper dental midline deviation was affected by the residual mandibular yaw asymmetry (p < 0.001) and maxillary shift asymmetry (p < 0.01). The residual lower dental midline deviation was affected by the residual yaw (p = 0.001) and shift (p < 0.01) asymmetry of the mandible ( Table 2).

Discussion
Patient satisfaction for correction of sagittal deformity and malocclusion via OGS requires improvement of facial asymmetry. Improvement of midline asymmetry of soft tissue and incisors shown in the frontal view, including midline deviation and lip cant, are of primary importance for patients to assess an asymmetry outcome positively 6,7,19 . Nevertheless, the frontal contour asymmetry, which is indeed altered by OGS and is of great clinical relevance 20 , has long been overlooked. Thus, the facial landmarks chosen in the present study covered midline and contour regions, intending to provide a measuring technique which is easily applicable and practical for clinicians to assess facial symmetry during and after operation.
This study also evaluated the underlying jaw characteristics contributing to the residual asymmetry after surgery. After bimaxillary surgery, the extent of residual facial asymmetry was highest in the lower contour region, and decreased in the order of middle and upper contour, chin, mandible, lower and upper incisors, lower lip, lip commissures, upper lip, and nose. The trend was almost the same as preoperative asymmetry (Table 1). For the contour region, the asymmetry was improved via surgery but not well enough to reach significance, which was correlated with the initial asymmetry and the roll or yaw asymmetry of the ramus (Table 2).
Although the mandibular body is the underlying skeletal support corresponding to the soft tissue envelope of the lower contour region, there was no significant correlation in between. Residual lower contour asymmetry was directly affected by the ramus, rather than the underlying mandibular body. One possible explanation is the modified Hunsuck technique, which extends the anterior cut of the osteotomy to the first molar. Another explanation is the limitation of the roll rotation of the proximal segment during surgery 14 . This speculation is supported by the significant correlation between the postoperative ramus roll asymmetry and the postoperative mandibular yaw asymmetry (r = −0.61, p < 0.001).
The threshold of clinical acceptance for midline asymmetry has been reported to be approximately 2 mm, including the upper dental midline 21,22 , lower dental midline 21 , and chin midline 2,21,23 . Thus, the results in Table 1 showing the treatment outcome of soft tissue midline asymmetry, with the exception of the chin, was favorable when the mean value was ≤2 mm. Although the soft tissue chin midline deviation in skeletal class II showed the greatest improvement (2.97 mm, 52.9%, p < 0.001), noticeable asymmetry (2.64 ± 1.80 mm) was still observed after surgery. This finding of residual chin deviation is consistent with previous OGS studies on different types  Table 1. Facial asymmetry a before and after bimaxillary surgery. a Absolute values were used to present the extent of facial asymmetry. of malocclusion [7][8][9]12,16 , suggesting the difficulty in the recognition of facial midline intra-operatively or relapse post-operatively. The present study provided further evidence that the residual chin asymmetry is correlated with the residual shift asymmetry of the mandible, chin, and ramus ( Table 2). The lip cant was insignificantly improved after bimaxillary OGS (0.64 mm, 37.0%, p = 0.013). We found that the mandibular roll asymmetry, rather than the maxillary roll asymmetry, affected the lip cant postoperatively ( Table 2). In addition, postoperative mandibular roll asymmetry was not necessarily correlated with postoperative maxillary roll asymmetry (r = 0.35, p = 0.055). In the study by Suzuki-Okumara et al., the preoperative measurements showed the same correlation, in which the preoperative lip cant was correlated with the preoperative mandibular roll asymmetry rather than the maxillary roll asymmetry 12 . Interestingly, Suzuki-Okumara et al., also reported that the change of lip cant was correlated with the change of the maxillary roll asymmetry, rather than the change of the mandibular roll asymmetry 12 . Although the present study did not measure the change in roll asymmetry of the maxilla or mandible, many of the OGS studies on lip cant consistently found a significant correlation between the change of lip cant and the change of the maxillary occlusal cant 6,10,19 . The study by Kim et al. demonstrated the average amount of lip cant correction was approximately 50% of the maxillary occlusal cant correction 19 . Therefore, in addition to correction of maxillary occlusal cant, correction of mandibular roll asymmetry might also play a role in further restoring lip symmetry.
Residual maxillary and mandibular shift asymmetry were found to be the most important factors influencing the postoperative upper and lower dental midline deviation, respectively. Residual mandibular yaw asymmetry also affected the upper and lower dental midline deviation postoperatively. Residual maxillary yaw asymmetry was found to have no significant influence on residual upper dental midline asymmetry, although mandibular yaw asymmetry was significantly correlated with maxillary yaw asymmetry (r = 0.63, p < 0.001). Song et al. analyzed pre-treatment variables and found that maxillary yaw asymmetry was the primary contributing factor for upper dental midline deviation in patients with upper dental midline deviation greater than 2 mm; however, www.nature.com/scientificreports www.nature.com/scientificreports/ analysis of mandibular variables was not shown 22 . Ryu et al. also analyzed pre-treatment variables and found that upper dental midline deviation was not significantly correlated with mandibular roll or yaw asymmetry, and lower dental midline deviation was significantly correlated with mandibular yaw asymmetry 24 . The possible reasons for the discordant findings about influencing factors for upper dental midline deviation include heterogeneity of samples (class III asymmetry vs. class II asymmetry), different variables for positional jaw asymmetry (line vs. plane), and the intervention of surgical-orthodontic treatment (no vs. yes).
The midline deviation of the nose became more severe after surgery (Table 1). This could be explained by the shift asymmetry of the maxilla after surgery ( Table 2). Sacrifice of the nasal and maxillary symmetry has also been shown to achieve favorable mandibular and overall facial symmetry 25 . Despite the deterioration, the mean residual midline deviation of the nose was less than 1 mm, which is usually clinically acceptable 2,21,22 .
This study has some limitations. First, the degree of facial asymmetry in patients with class II deformities is usually modestly remarkable and consequently the sample size of this study was small. Studies with larger sample size are needed to draw more robust conclusions. Second, the occlusal plane cant, which might play an important role on mandibular shift and chin deviation, was not measured. Finally, no size influence of soft tissue or jaws was analyzed. Future studies are needed to explore the impact of occlusal cant, soft and hard tissue volume on the treatment outcome of facial asymmetry.

conclusions
The findings of the present study showed bimaxillary OGS for patients with class II asymmetry significantly improved the midline asymmetry of the chin, mandible, and lower incisors. However, noticeable chin deviation was still observed after surgery, which was affected by the residual shift asymmetry of the mandible, chin, and ramus. The contour asymmetry was not significantly improved after surgery, which was affected by the initial severity of contour asymmetry, and the residual roll or yaw asymmetry of the ramus. Maxillary shift asymmetry was the primary factor influencing the postoperative midline deviation of the nose, upper lip, and upper incisors. Mandibular shift asymmetry was the primary factor influencing the postoperative midline deviation of the lower lip, mandible, chin, and lower incisors.

Material and Methods
Patients. The retrospective study was conducted in accordance with the World Medical Association Declaration of Helsinki on medical research ethics. The approval of the study was granted by the Ethics Committee for Human Research at the Chang Gung Memorial Hospital in Taoyuan, Taiwan. The need for informed consent was waived by the Ethics Committee that approved the study due to the retrospective design of the study. Thirty Taiwanese adults (age ≥18 years) with class II deformity (A point-nasion-B point angle >4 degrees) and significant facial asymmetry (skeletal menton deviation >2 mm or lip cant >2 mm or significant  www.nature.com/scientificreports www.nature.com/scientificreports/ contour asymmetry) were selected based on the following criteria: (1) consecutive Le Fort I osteotomy and bilateral sagittal split osteotomy (BSSO) advancement surgery by the attending surgeons supervised by one senior surgeon with more than 40 years of experience at the Chang Gung Craniofacial Center during a 3-year period, (2) completion of postsurgical orthodontic treatment, (3) no progressive or chronic temporomandibular joint disorder, (4) no other craniofacial deformities or genetic syndromes, (5) no history of craniofacial surgery or trauma, and (6) available CBCT taken at two time points, before surgery and at least 12 months after surgery, on the day of orthodontic debonding. The informed consent for publication of identifying information/images in an online open-access publication was obtained from the patient whose images were displayed. Surgical technique. The BSSO was modified from Hunsuck 26 by extending the anterior cut of the osteotomy to the first molar 27,28 . The Le Fort I osteotomy was performed with a technique similar to that popularized by Bell 29 . No additional surgical intervention other than genioplasty was performed. Rigid fixation was performed with bone plates or screws. On average, the anterior maxilla (incisive foramen) moved backward (1.16 mm), upward (2.36 mm), and toward the opposite side (opposite to menton-deviated side, 0.23 mm). The posterior maxilla (greater palatine foramen) moved forward (0.49 mm and 0.61 mm, respectively for the deviated and opposite sides), upward (0.74 mm and 1.93 mm, respectively for the deviated and opposite sides), and toward the opposite side (0.11 mm and 0.25 mm, respectively for the deviated and opposite sides). The anterior mandible (genial tubercle) moved forward (3.87 mm) and toward the opposite side (2.13 mm). The posterior mandible (mental foramen) moved forward (2.86 mm and 3.04 mm, respectively for the deviated and opposite sides), downward (2.31 mm and 0.95 mm, respectively for the deviated and opposite sides), and toward the opposite side (2.23 mm and 1.28 mm, respectively for the deviated and opposite sides).
CBCT. CBCT of the head and neck was performed using an i-CAT 3D Dental Imaging System (Imaging Sciences International, Hatfield, PA, USA) with the following parameters: 120 kVp, 0.4 mm × 0.4 mm × 0.4 mm voxel size, 40 second scan time, and 16 cm × 16 cm field of view. The patient's head was positioned with the Frankfort horizontal plane parallel to the ground. Throughout the scan, patients were asked not to swallow.
Images were stored in the Digital Imaging and Communications in Medicine (DICOM) format and then transferred to a workstation (Avizo v7.0.0 software, FEI, Mérignac, France) where they were rendered into volumetric images, segmented and analyzed by one single investigator (CYF) blinded to the patients' treatment histories. Before analysis, six skeletal landmarks were selected for registration of the 3D images in a 3D coordinate system (x, y, z) given in millimeters with nasion as the zero point: nasion, bilateral porion, bilateral orbitale, and basion. The horizontal reference plane was parallel with the FH plane (the best-fit plane passing through bilateral porion and orbitale) and passing through nasion. The midsagittal plane was perpendicular to the horizontal reference plane and passing through nasion and basion. The coronal reference plane was perpendicular to the horizontal and midsagittal reference planes and passing through nasion. A positive value indicates the left, posterior and superior side of the face. After registration of the 3D images, landmarks 30-32 and planes used for measurement  Table 4. Landmarks and planes used for analysis of positional jaw asymmetry. a For the two OGS patients without genioplasty, the designation of LBs were started from the sagittal level of the mental foramen.  www.nature.com/scientificreports www.nature.com/scientificreports/ (Tables 3 and 4, and Figs. 1 and 2) were located on the 3D surface models by the same investigator. Multiplanar reconstruction views were also used to identify the landmarks when necessary.
Asymmetry outcome. To demonstrate the extent of asymmetry before and after surgery, absolute values of the following measurements for soft tissue and dental asymmetry were used. Real values were used for regression analysis.
(1) Midline asymmetry: (a) Midline deviation: the transverse distances between the midline landmarks of UIE and LIE and the MSP were measured as the midline deviation of the upper incisors and lower incisors (Table 3).
Influencing factors. The possible influencing factors for residual asymmetry included preoperative asymmetry and postoperative positional jaw asymmetry in terms of shift, roll or yaw asymmetry. To evaluate the positional jaw asymmetry, we first defined five planes (maxillary central plane, mandibular central plane, chin central plane, and bilateral ramal planes) for each osteotomy segment (Table 4). Then, the discrepancy of the five planes from the reference planes was calculated to quantify the shift, roll or yaw asymmetry of the maxilla, mandible, chin and ramus (  Statistical analysis. Statistical analyses were performed using the statistical software package SPSS version 19.0 for Windows (SPSS Inc, Chicago, USA). All descriptive statistics are presented as mean ± standard deviation (SD). Paired t test was used to compare the difference in facial asymmetry before and after surgery. To identify the influencing factors for residual asymmetry, stepwise multiple linear regression analysis was used with the postoperative soft tissue and dental asymmetry as the dependent variables and the preoperative soft tissue and dental asymmetry and the postoperative positional asymmetry of the osteotomy segments as the independent variables. To account for multiple comparison, p ≤ 0.01 was considered statistically significant.

Data availability
The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.