INTRODUCTION

Facial asymmetry results in occlusal and lip cant. Facial asymmetry, however, might be masked by severe facial skeletal imbalance, dental malalignment, soft tissue compensation, or a tilting head posture.1 Haraguchi et al.2 reported differences between the degrees of actual skeletal asymmetry and soft tissue asymmetry perceived in Class III patients.

It has been suggested that even people considered to be facially symmetric in clinical examinations have, upon radiographic examination, at least some facial asymmetry.1–6 Ferrario et al.3,4 and Haraguchi et al.2 suggested that soft tissue asymmetry is less severe than skeletal asymmetry. For these reasons, the proper analysis of facial asymmetry should include examinations of both hard and soft tissue. Postero-anterior (PA) two-dimensional (2D) cephalograms combined with frontal facial photographs can be cost-effective tools for determining the presence and extent of facial asymmetry.7 However, midfacial areas, because of their geometric position, are difficult to evaluate by those means. Among the several three-dimensional (3D) analysis strategies,8–11 cone-beam computed tomography (CBCT) offers simultaneous access to the facial soft tissue and the underlying skeletal structure. Also, voxel-based superimposition, working from a stable reference structure, can provide orthodontists and oral surgeons with information on treatment-related facial changes.12–14

Bimaxillary orthognathic surgery is a common method of treating patients with facial asymmetry.15–17 For correction of the maxillary occlusal plane, Le Fort I osteotomy proceeds by intrusion of the long side or differential intrusion of the short and long sides. Most patients with a canted maxillary occlusal plane are corrected with a minor decrease in the original vertical length of the facial dimension. Therefore, differences in the extent of vertical reduction between the short and long sides might influence soft tissue asymmetry in the midfacial area.

The purpose of the present study was to compare the midfacial soft tissue changes in patients who had undergone differential leveling Le Fort I osteotomies on the short and long sides for facial asymmetry. The null hypothesis was that there were no significant differences in the midfacial soft tissue changes between the greater- and lesser-reduction sides.

MATERIALS AND METHODS

The study included 25 skeletal Class III patients with facial asymmetry. All of them had undergone leveling Le Fort I osteotomy and mandibular setback sagittal split ramus osteotomy (SSRO) between January 2008 and December 2010 at the Department of Oral and Maxillofacial Surgery, Pusan National University Hospital. The average age of the patients was 22.3 ± 2.8 years (range, 18–27 years). They included 13 women (21.3 ± 2.7 years) and 12 men (23.3 ± 1.8 years). For inclusion in the subject group, patients had to have more than 3 mm of occlusal cant at the maxillary first molar bilaterally. Bimaxillary surgery was performed by the same surgeon. None of the subjects showed any presence of syndromes. All of them had previously undergone surgical orthodontic treatment at the Department of Orthodontics, Pusan National University Hospital. The canted maxillary occlusal plane was corrected by impaction of the long side of the maxilla in 11 patients and by a combination of differential impaction of the long and short sides in 14 patients. This study was reviewed and approved by the Ethics Committee at Pusan National University Hospital.

The patients were subjected to CBCT (DCT pro, Vatech Co, Seoul, Korea) 1 month before (T0) and 7.9 ± 1.1 months (T1) after surgery with centric relation and reposed lip. The maxillofacial regions were scanned for 24 seconds using CBCT with a field of view of 20 × 19 cm, a tube voltage of 90 kVp, a tube current of 4.0 mA, and a voxel size of 0.3 mm. The CBCT raw data were reformatted to 3D images using 3D imaging software (Ondemand 3D, Cybermed Co, Seoul, Korea).

To evaluate the midfacial hard and soft tissue changes after leveling Le Fort I osteotomy (reduction on the longer side only or differential reduction on both sides) and mandibular setback SSRO, two CBCT data images were superimposed on the anterior cranial base according to maximization of mutual information theory.13

Linear measurements were obtained under consistent scanning conditions. The measurement points and reference planes are defined in Table 1. To assess the horizontal and vertical skeletal changes after surgery, the distances on the sagittal plane from both the Frankfort horizontal (FH) plane and the Na perpendicular plane to the A-point on the maxilla and the Me on the mandible were evaluated before (T0) and after surgery (T1; FH plane to A-point, Na-perpendicular plane to A-point; FH plane to Me, Na-perpendicular plane to Me).

Table 1 Reference Points and Planes

View Fullscreen

Table 2 Means and Standard Deviations of Soft Tissue Measurement Points as Determined From Preoperative (T0) and Postoperative (T1) Data

View Fullscreen

For comparison with maxillary occlusal canting, the maxillary height difference was defined as the short- and long-side difference in the distance between the FH plane and the furcation of the maxillary first molar (Figure 1). The extent of the mandibular setback on the short and long sides was defined as the preoperative and postoperative difference in the distance between the Na-perpendicular plane and the furcation of the mandibular first molar (Figure 1).

View Figure

• Download as Powerpoint
• Download Figure

In the measurements of the soft to hard tissue response, 10 × 27 grids at 4.5 mm (vertical) and 5 mm (horizontal) intervals were employed.18 A total of 270 measurement points were placed on the grids. The soft tissue measurement procedure was as follows: an MRP image corresponding to each CBCT data set was manipulated according to the reference planes (FH and midsagittal reference [MSR] planes), after which 11 axial views in 4.5-mm intervals were employed to measure the distance from the Na perpendicular plane to the soft tissue measurement points. Points outside the face were not used. Finally, the means and standard deviations of the postsurgery differences were calculated (Figure 2).

View Figure

• Download as Powerpoint
• Download Figure

RESULTS

DISCUSSION

CBCT imaging can be used to show hard and soft tissue simultaneously, but its application has been limited to determinations of reliable measurement points on 3D images. Recently, many CBCT superimposition techniques have been introduced. In addition, Choi and Mah14 introduced a CBCT comparison method that incorporates the mutual information theory for accurate, rapid, and automatic comparison without segmentation. In the present study, voxel-by-voxel registration of the mutual information theory was applied for superimposition, with the anterior cranial base, appropriately, being used for best registration. The resultant image data were then analyzed.

Facial asymmetry in skeletal Class III patients is known to occur mainly owing to overgrowth, a more mesial inclination of the mandibular ramus, and a maxillary vertical excess on one side.19 A canted maxillary occlusal plane is the representative characteristic of facial asymmetry. Since the advent and popularization of orthognathic surgery, this can easily be corrected. However, soft tissue canting and soft tissue asymmetry often remain after surgery. Preoperative analysis of facial soft tissue and assessment of associated postoperative outcomes, therefore, are necessary if patients' esthetic postoperative satisfaction is to be met. In the present study, after leveling the maxillary occlusal plane by differential reduction on the short and long sides, the midfacial soft tissue response was determined from preoperative and postoperative data by the superimposition of CBCT volumes, and the difference in the soft tissue response between the greater- and lesser-reduction sides was compared. In the previous study already mentioned, the results showed that the vertical reduction effected by Le Fort I osteotomies provoked an overall hard to soft tissue response in the midfacial areas, with the soft tissue in the triangular area (which encompasses both the nasolabial grooves and the upper lip) moving in the anterior direction.18,20 Therefore, we could assume that the more significant vertical reduction on the long side would incur more redundant soft tissue. In the results, the overall area was changed. Even though the midfacial soft tissue areas below the FH plane to 13.5 mm showed different soft tissue response on the greater- compared with the lesser-reduction side, changes greater than the 2.5-mm average were common below the plane to 13.5 mm and lateral to the MSR plane to 30 mm for both the male and female groups (P < .05).

Jung et al.21 showed that only differential mandibular setback movement produced more soft tissue redundancy in the gonial angle area on the greater-setback side than on the lesser-setback side. However, in the present study, the extent of mandibular setback movement was not statistically different (P > .05; Table 3). Therefore, the differential mandibular setback might not affect any midfacial soft tissue difference. In the case of the leveling LeFort I osteotomies, there was more redundant midfacial soft tissue on the greater-reduction side than on the lesser-reduction side. This redundant soft tissue induced midfacial volumetric differences after bimaxillary surgery, especially in the buccal cheek area below the zygoma.

Table 3 Gender Difference in Skeletal Changes by Mann-Whitney U-Test

View Fullscreen

Table 4 Mean and Standard Deviation of Soft Tissue Changes After Le Fort I Osteotomies With Differential Reduction in Male Subjects^a

View Fullscreen

Table 5 Mean and Standard Deviation of Soft Tissue Changes After Le Fort I Osteotomies With Differential Reduction in Female Subjects^a

View Fullscreen

As solutions to the problem of volumetric differences in midfacial soft tissue between the short and long sides, the conventional face-lift procedure,22,23 the injection of botulinum toxin into the masseter muscle on the bulkier side,24 and fat grafting or fat reduction25 have been introduced. In orthognathic surgery, buccal fat reduction has been the method of choice. In the results, the distinctly changed areas have been distributed in the buccal cheek, where the buccal fat pad is one of several encapsulated fat masses. The buccal fat pad is located on either side of the face between the buccinators muscle and several superficial muscles including the masseter and zygomaticus major and minor.26 The anatomical position of the buccal fat pad is easily accessed in maxillary Le Fort I osteotomy. Its removal is one of the better choices among the procedures available for reduction of the cheek.

In this study, the subject group was divided between patients who had undergone unilateral maxillary impaction and those for whom bilateral differential maxillary impaction had been used. The potential influence of the group's lack of homogeneity on the results should be considered in follow-up or future studies. However, the present results suggest nonetheless that oral surgeons and orthodontists need to be concerned about midfacial soft tissue differences after leveling the maxillary occlusal plane at the preoperative stage and thus should take appropriate measures to improve it.

CONCLUSIONS

• The null hypothesis was rejected.

• Most midfacial soft tissue areas below the FH plane to 13 mm showed differences in the soft tissue response between the greater- and lesser-reduction sides (P < .05).

• The statistically significant >2.5-mm soft tissue difference between the greater- and lesser-reduction sides appeared commonly below the FH plane to 13 mm and lateral to the MSR plane to 30 mm in the male and female groups (P < .05).