Enhancement of the facial profile has become a significant priority in orthognathic surgery due to increasing social concerns about esthetics. Currently, a symmetric and tapered facial profile in the frontal view is preferred, and a straight facial profile is preferred over a convex or concave profile in the lateral view.1 However, Asians tend to have a large percentage of prognathic and squared facial profiles.2^,3

Orthodontic treatment must be accompanied by orthognathic surgery to improve severe facial asymmetry and mandibular prognathism. In practice, two main methods, internal vertical ramus osteotomy (IVRO) and sagittal split ramus osteotomy (SSRO) are used for mandibular orthognathic surgery.4 Unlike IVRO, SSRO can be used for various mandibular deformities, including prognathism, retrognathism, and asymmetry.

There have been many studies on the stability of SSRO since its introduction.5^,6 Factors influencing the stability of SSRO include presurgical orthodontic treatment, internal fixation method, quality of postsurgical occlusion, and location of the proximal segment.6 The site of the proximal segment is crucial not only for stability but also for esthetics. The shape of the mandibular ramus, especially the gonial area, may influence patient esthetics and, thus, patient satisfaction.

Previous studies7^-9 using two-dimensional (2D) images have shown changes in mandibular width after mandibular setback surgery using SSRO. However, a 2D image can be distorted depending on the magnification and orientation of the image. To overcome these limitations, there have been increasing attempts to assess the changes after mandibular setback surgery using three-dimensional (3D) cone-beam computed tomography (CBCT). Most studies have focused on evaluating condylar stability and not the mandibular width. A study10 assessing the change in mandibular width after SSRO in patients without mandibular asymmetry has been published. However, for patients with asymmetry of the mandible, when the mandibular midline shifts towards the facial midline, the distal segment of the non-deviated side tends to move toward the inside and the deviated side toward the outside.11 This may increase the mandibular width due to flaring or a gap between segments. Therefore, assessing the mandibular width in patients with mandibular asymmetry is more clinically meaningful.

In this study, our objective was to evaluate the 3D changes in mandibular width after SSRO in patients with mandibular asymmetric prognathism using CBCT.

This study was reviewed by the Institutional Review Board of Pusan National University Dental Hospital (PNUDH -2019037).

Subjects

Seventy patients who underwent SSRO for mandibular setback surgery between 2013 and 2018 at the Pusan National University Dental Hospital were included in this retrospective study. Patients were divided into two groups according to the differences in their right and left setback amounts. The asymmetric group (ASG; n = 35) included patients with a setback difference of 2 mm or greater. The symmetric group (SYG; n = 35) included patients with a setback difference of less than 2 mm (Table 1).9^,12 All patients met the following criteria: 1) mandibular prognathism, 2) pre-operative orthodontic treatment, 3) no mandibular angle osteotomy, 4) no history of temporomandibular joint disorder, 5) no maxillofacial syndromes, and 6) no maxillary asymmetry. The anatomic landmarks used to determine maxillary asymmetry were based on a previous study.13 For 2-jaw surgery, only Le Fort I maxillary impaction or advancement surgery was included. The sample size of 35 subjects in each group was calculated based on the following: β error = 0.20, a error = 0.05, and effect size = 0.5.14

Table 1 . Descriptive statistics.

Variable	Symmetrical group (n = 35)	Asymmetrical group (n = 35)	p-value
Sex (Male/Female)	18/17	19/16	0.811†
Age (yr)	22.7 ± 2.4	23.1 ± 2.7	0.887*
1-jaw (SSRO only)/2-jaw	17/18	15/20	0.631†
Chin deviation before surgery (mm)	1.10 ± 0.74	4.58 ± 2.10	< 0.001*
Chin deviation after surgery (mm)	0.01 ± 0.02	0.03 ± 0.02	0.571*
Amount of setback (mm)
Both sides	8.52 ± 2.73	8.10 ± 4.28	0.623*
Deviated side		5.80 ± 3.68
Non-deviated side		10.39 ± 3.58

Data are presented as mean ± standard deviation or data only..

SSRO, sagittal split ramus osteotomy..

*Independent t-test..

^†Chi-square test..

Surgery

Every patient was treated with modified bilateral SSRO surgery15 by a single oral maxillofacial surgeon with more than 16 years of experience in orthognathic surgery. The medial pterygoid muscle was detached to minimize muscle extension. The mandible was located using a pre-manufactured surgical stent. Early contact between segments was removed using a surgical round bur. Internal fixation was performed by using mini-plates with monocortical screws. Postsurgical orthodontic treatment began one month after surgery.

CBCT data

CBCT (Pax-Zenith3D; Vatech Co., Seoul, Korea) images taken immediately before surgery (T1), 3 days after surgery (T2), and 6 months after surgery (T3) were used to evaluate changes in mandibular width after SSRO. The following CBCT settings were used: tube voltage of 90 kVp; scanning time of 24 seconds; tube current of 4 mA; voxel size of 0.3 mm; and field of view of 20 × 19 cm. The Ondemand3D (Cybermed, Seoul, Korea) software was used to convert the CBCT data into a 3D model. To re-orient the transformed 3D models, the horizontal reference plane (HRP, passing through both orbitale and right porion) and vertical reference plane (VRP, perpendicular to HRP and connecting the nasion and basion) were used. Based on the anterior cranial base, CBCT images at T2 and T3 were superimposed on those at T1. The landmarks are based on previous studies (Figures 1 and 2, Table 2).9 The measurements of the hard tissue were as follows: (1) inter-gonial width (IGW), a straight line connecting the left and right gonion points (R_go and L_go); (2) inter-ramal width (IRW), a straight line connecting the landmarks (R_ra and L_ra) of the ramus located on the mastoid process line; and (3) unilateral gonial width (UGW), the distance from VRP to the unilateral gonion point (R_go or L_go). The soft tissue was measured using (4) the soft tissue width (STW), a straight line connecting the soft tissue landmarks (R_lip and L_lip) located on the lip commissure line. The menton was used as the landmark for measuring the amount of mandibular setback. The dental midline of the mandible and the menton were well aligned in all subjects through preoperative orthodontic treatments. The amount of mandibular setback was evaluated by the following procedure.16

Table 2 . Landmarks and description.

Landmarks	Description
Horizontal reference plane (HRP)	Left orbitale - right orbitale - right porion
Vertical reference plane (VRP)	Perpendicular to the HRP plane and passes through nasion and basion
R_go	Most lateral point of right gonion area
L_go	Most lateral point of left gonion area
Gonial plane	Parallel to the HRP and passes through the midpoint of R_go and L_go
Coronal plane	Perpendicular to the HRP and passes through the midpoint of the R_go and L_go
Ramus plane	Parallel to the HRP plane and passes the midpoint of the most inferior point of the mastoid process on both sides
R_ra	Intersection point of the ramus plane and the lateral border of right ramus
L_ra	Intersection point of the ramus plane and the lateral border of left ramus
Lip plane	Parallel to the HRP plane and passes the midpoint of the lip commissure on both sides
R_lip	Right intersection point of the coronal plane and the lip plane
L_lip	Left intersection point of the coronal plane and the lip plane
IGW	Distance from R_go to L_go
IRW	Distance from R_ra to L_ra
UGW	Distance from VRP to L_go or R_go
STW	Distance from R_lip to L_lip

IGW, intergonial width; IRW, interramal width; UGW, unilateral gonial width; STW, soft tissue width..

Figure 1. Hard tissue landmarks.
See Table 2 for definitions of each landmark.

Figure 2. Soft tissue landmarks.
See Table 2 for definitions of each landmark.

1. Using the facebow transfer, the cast is mounted to an articulator.

2. Horizontal markings, parallel to the baseplate, are created. Vertical markings are then placed at the facial midline and first molar cusp. The distance from the mandibular dental midline to the menton on the CBCT is measured. Then, the horizontal position of the menton is drawn on the mandibular cast using this measured distance.

3. The planned movement, which rotates the menton to match the facial midline, is carried out, and the segments are reassembled in the postoperative position.

4. The setback amount is measured using vertical reference lines at the first molar cusp.

The average chin deviation of SYG and ASG, which was the distance from the VRP to the menton on the CBCT, was 1.1 and 4.58 mm, respectively. All patients' chin deviations were successfully corrected through orthognathic surgery (Table 1).

Statistical analyses

All measurements were recorded by a single researcher. Dahlberg's formula and intraclass correlation coefficient (ICC) were used to assess intra-examiner reliability. Twenty-one patients were randomly selected and the measurements were re-done 1 month later. The average Dahlberg error was 0.32 mm, and all ICC values were greater than 0.913. The Shapiro–Wilk and Levene's tests were used to determine whether the data were normal and homogeneous. Repeated measures analysis of variance was applied to statistically verify the differences in mandibular width at T1, T2, and T3 for each group. The correlation between the setback amount and the mandibular width was determined using Pearson correlation analysis. The SPSS program (version 19.0; IBM Corp., Armonk, NY, USA) was used for statistics.

The changes in IGW, IRW, and STW at T1, T2, and T3 are listed in Table 3. IGW, IRW, and STW values showed a statistically significant increase in both groups postoperatively (T2). However, after 6 months (T3), the values decreased significantly (Figure 3). In particular, no significant differences were noted in IRW and STW between T1 and T3. Moreover, a statistical difference between ASG and SYG was not found (p > 0.05).

Table 3 . Changes in hard and soft tissue width.

Variable	T1		T2		T3		T2-T1			T3-T2			T3-T1
	Mean ± SD		Mean ± SD		Mean ± SD		Mean ± SD	p-value		Mean ± SD	p-value		Mean ± SD	p-value
IGW (mm)
SYG	99.49 ± 6.19		102.40 ± 6.42		100.77 ± 6.29		2.91 ± 1.67	< 0.001*		–1.62 ± 1.30	< 0.001*		1.29 ± 1.69	< 0.001*
ASG	100.38 ± 6.40		102.85 ± 6.36		101.53 ± 6.09		2.47 ± 1.88	< 0.001*		–1.32 ± 1.08	< 0.001*		1.15 ± 1.85	0.002*
IRW (mm)
SYG	109.85 ± 4.90		112.49 ± 4.72		109.95 ± 4.95		2.64 ± 1.77	< 0.001*		–2.54 ± 1.38	< 0.001*		0.10 ± 1.58	0.674
ASG	110.38 ± 5.81		112.19 ± 6.10		110.64 ± 6.003		1.80 ± 1.85	< 0.001*		–1.54 ± 1.65	< 0.001*		0.26 ± 1.67	0.386
STW (mm)
SYG	131.57 ± 10.15		143.26 ± 11.89		131.33 ± 9.84		11.68 ± 7.60	< 0.001*		–11.92 ± 7.28	< 0.001*		–0.24 ± 4.53	0.713
ASG	133.00 ± 12.29		143.40 ± 14.08		132.70 ± 13.52		10.39 ± 4.87	< 0.001*		–10.70 ± 5.21	< 0.001*		–0.30 ± 3.17	0.592

SYG, symmetric group; ASG, asymmetric group; IGW, intergonial width; IRW, interramal width; STW, soft tissue width; SD, standard deviation; T1, immediately before surgery; T2, 3 days after surgery; T3, 6 months after surgery..

*p < 0.05..

Figure 3. Changes of mandibular width in the asymmetric group.

Table 4 shows UGW in the ASG. On the deviated side, UGW showed a significant increase at T2 and a decrease at T3. The deviated side had a more substantial change compared to the non-deviated side. The asymmetry of the deviated and non-deviated sides for UGW remained after surgery. However, these differences were not statistically significant (p > 0.05). The amount of setback had no statistically significant correlation with IGW, IRW, or STW values (Table 5).

Table 4 . Unilateral gonial width in deviated and non-deviated sides of asymmetric patients.

Variable	T1		T2		T3		T2-T1			T3-T2			T3-T1
	Mean ± SD		Mean ± SD		Mean ± SD		Mean ± SD	p-value		Mean ± SD	p-value		Mean ± SD	p-value
UGW (mm)
Deviated	51.05 ± 3.25		52.72 ± 3.03		51.78 ± 3.52		1.67 ± 2.01	0.003*		–0.96 ± 1.56	0.003*		0.73 ± 2.39	0.554
Non-deviated	49.30 ± 4.37		50.12 ± 4.38		49.76 ± 3.73		0.82 ± 2.10	0.105		–0.36 ± 1.69	0.370		0.46 ± 2.44	0.416

UGW, unilateral gonial width; SD, standard deviation; T1, immediately before surgery; T2, 3 days after surgery; T3, 6 months after surgery..

*p < 0.05..

Table 5 . Correlation in association with the setback amount.

Variable	T2-T1			T3-T2			T3-T1
	r value	p-value		r value	p-value		r value	p-value
IGW	–0.197	0.080		–0.091	0.424		–0.112	0.325
IRW	–0.122	0.281		0.050	0.661		–0.086	0.447
STW	0.002	0.984		0.008	0.947		0.002	0.989

IGW, intergonial width; IRW, interramal width; STW, soft tissue width; T1, immediately before surgery; T2, 3 days after surgery; T3, 6 months after surgery; r, correlation coefficient..

This study showed that immediately after SSRO, both groups showed an increase in mandibular width. These results were consistent with those of previous 2D studies.7^,9 However, a statistical difference between ASG and SYG was not found. This may be due to the removal of the early contact between the segments with a surgical round bur during surgery. In addition, since the arch form of the mandible is wider toward the posterior and the width of the distal segment in the gonial area decreases when the distal segment goes setback, mandibular rotation is assumed to be less affected during rotational setback surgery.

The mandibular width decreased at T3 compared to that at T2. Although IGW was statistically different between T1 and T3, the difference was about 1 mm. In particular, no significant differences were identified between T1 and T3 for IRW and STW, which could be because of remodeling caused by the separation of the medial pterygoid muscle, which reduced muscle extension in the gonial region.17 Alternatively, the proximal segment may return to its original position with semi-rigid fixation and neuromuscular adaption. The distal segment can easily change postoperatively, whereas the proximal segment does not change as easily because of its adaptation to the preoperative environment.

The STW of the gonial region almost returned to its original width at T3 after a significant increase at T2. The increase in STW after surgery is mainly due to edema, whereas the subsequent decrease is due to soft tissue adaptation. However, since changes in the soft tissue are influenced by various parameters, such as body weight, facial expressions, and muscle mass, the results should be interpreted cautiously.

The deviated side of the asymmetric patient showed a considerable increase in UGW than the non-deviated side. When the distal segment is shifted toward the facial midline, the rotation causes the proximal segment to flare more on the deviated than on the non-deviated side. Although not statistically significant, the results were similar to previous 2D studies.12^,18 If the asymmetry of the deviated and non-deviated sides remains significant after surgery, additional surgery, such as mandibular angle osteotomy, is required.

No statistically significant correlation between the mandibular width and the amount of setback was found, which is also consistent with the previous results.7 The mandibular width does not change proportionally, even when the setback amount is larger.

The head's orientation must be kept constant to compare 3D data at different times accurately. To ensure identical head orientation, CBCT images taken at T2 and T3 were superimposed with those taken at T1 based on the anterior cranial base.19 Furthermore, to reduce measurement inaccuracies, this study selected landmarks such as the gonion point, mastoid process, and lip commissures that can be easily identified.

Despite the valuable findings of this study, there are some limitations. First, the results may lead to bias since a single oral maxillofacial surgeon from a single center treated patients. A multicenter study with a large sample size is required to overcome these limitations. Second, the arch form of the mandible was not fully considered. Third, mandibular asymmetry is not sufficiently categorized in detail. The number of canting-, translation-, and yawing-dominant types in the ASG was 0, 15, and 20, respectively.20 The change in mandibular width may vary depending on the type.

Regardless of the mandibular asymmetry, the mandibular width expanded immediately after mandibular setback surgery using SSRO. Six months after surgery, the mandibular width was similar to its original width.

A video presentation of this article is available at https://youtu.be/xg30oaGH0G0 or www.e-kjo.org.

None.

Conceptualization: SHK. Data curation: YIK. Formal analysis: YIK. Investigation: YIK. Methodology: SSK. Project administration: SHK. Resources: SSK. Software: SBP. Supervision: SHK. Validation: SSK. Visualization: SBP. Writing–original draft: SSK. Writing–review & editing: all authors.

No potential conflict of interest relevant to this article was reported.

None to declare.