INTRODUCTION

The positional relationship between the maxilla and the mandible greatly influences the occlusion, function, and esthetics of the craniofacial region. Growth modification of the mandible is considered a standard of care for adolescents with skeletal discrepancies and active rotation of the mandible through vertical control has become a routine part of conventional orthodontic protocol for adults.^1,2 Orthodontists sometimes confront unplanned or unexpected changes of the condyle, during or after treatment, which may cause negative effects on the overall outcome as well as affect stability.^3,4 Particularly, posterior displacement of the condyle may negatively affect the profile or occlusion, and may be associated with symptoms of temporomandibular joint disorder (TMD).^4,5

Posterior displacement or secondary displacement due to pathologic changes of the condyle is difficult to predict, differentially diagnose, or confirm with conventional radiographs.⁶ With advancements in digital imaging including cone-beam computed tomography (CBCT) and three-dimensional (3D) superimposition methods, positional or serial changes of the mandible as well as the condyle can be visualized more efficiently.⁷ For example, the use of rapid palatal expander in children, maxillary first premolar extraction for skeletal Class II malocclusion, and orthognathic surgery for skeletal Class III malocclusion reportedly induce approximately 0.5–0.7 mm displacement of the condyle on average.^8–11 Minor limited changes in temporomandibular joint (TMJ) anatomy, position, or joint space after orthodontic treatment do not cause major concerns and are not considered as the cause or cure for TMD related symptoms.^12–14 Despite these findings, orthodontists do take caution to maintain the TMJ position given that female adolescents and young adults with Class II malocclusion, who have a higher prevalence of TMD associated symptoms,^13–15 also account for a large proportion of orthodontic patients.¹⁶

On the other hand, patients with condylar displacement exceeding 1–2 mm after orthodontic treatment or orthognathic surgery are also frequently noted in the orthodontic literature. Clinically, patients with condylar displacement exceeding 1 mm after treatment are recommended to undergo continuous observation due to poor prognosis, the possibility of occlusal instability, or the development of TMD.^10,17 However, information on the overall incidence of unexpected or excessive condylar displacement that may follow comprehensive orthodontic treatment is still limited. Therefore, this study aimed to evaluate the incidence of unexpected condylar displacement exceeding the clinical threshold of 1 mm in adult orthodontic patients and to identify displacement patterns as well as the clinical risk factors associated with these changes.

MATERIALS AND METHODS

Participants

This study was approved by the Institutional Review Board of Gangnam Severance Hospital, Yonsei University. From the adult cohort (≥18 years) treated with fixed orthodontic appliances between 2011 and 2020 in the Department of Orthodontics Gangnam Severance Hospital, consecutive patients with pre-(T1) and post-treatment (T2) CBCTs were screened (N = 540). Patients who underwent orthognathic surgery (n = 232) and those with low-quality CBCT images (n = 17) were excluded, resulting in 291 participants (Figure 1). Sex, age, skeletal relationship, treatment duration, and extraction pattern were evaluated using the orthodontic records. The presence of TMD symptoms at T1 was based on patient self-reported questionnaires (Supplement 1). Age groups were classified as under 30 (18–29), 30s (30–39), 40s (40–49), and over 50s (≥50 years). Orthodontic extraction of one or more permanent teeth (excluding the third molar) or presence of edentulous regions due to missing teeth were classified as extraction treatment; otherwise, treatment was classified as nonextraction. Treatment modalities were further subclassified using the treatment plan as well as the treatment outcome in the antero-posterior (A-P) and vertical relationships compared to before treatment: anterior retraction, premolar extraction followed by space closure with more than 2 mm of incisor retraction; alignment, less then 2 mm of A-P or vertical movement of the incisors; upper or lower, distalization of the upper or/and lower arches with temporary anchorage devices (TADs) to correct A-P relationship; and upper or lower molar intrusion, intrusion of the molars with TADs. The demographics of the participants are shown in Table 1.

View Figure

• Download as Powerpoint
• Download Figure

Table 1. Cohort Demographics

View Fullscreen

Data Acquisition and 3D Coordinate System

CBCT scans were performed by an experienced radiological technologist using PaX Zenith 3D (Vatech, Seoul, Korea). Participants were instructed to stand straight in the maximum intercuspal position with light contact between the lips and a relaxed face. The scan captured 632 slices, with exposure parameters of 105 kV and 5.4 mAs, acquired for 24 seconds, with a voxel size of 0.3 mm and a field of view of 24 × 19 cm, encompassing the face, jaws, and entire cranial base. T1 and T2 CBCT data were superimposed by voxel-by-voxel registration on stable structures of the anterior cranial base using ITK-SNAP software (www.itksnap.org) (Figure 2A).¹⁸ The T1 and registered T2 CBCTs were then reconstructed into 3D images using On Demand 3D software (Cybermed Co., Seoul, Korea) sharing the same orientation.

View Figure

• Download as Powerpoint
• Download Figure

For landmark detection and measurements, the horizontal reference plane was set parallel to the Frankfort Horizontal (FH) plane, constructed with the right porion (Po) and the left and right orbitale (Or) landmarks, passing through nasion (N). The midsagittal reference plane was set perpendicular to the FH plane and passed through N and basion, whereas the coronal reference plane was perpendicular to the horizontal and midsagittal reference planes and passed through N.^19,20 A Cartesian coordinate system was used with N as the origin of the coordinate system (0, 0, and 0) with x-y, x-z, and y-z planes to represent the axial, coronal, and horizontal planes, respectively, at T1 (Figure 2B).

Landmark positioning was performed on the 3D volume-rendered skeletal surface images with subsequent adjustment of multiplanar reconstructed cross-sectional image. Pogonion (Pog), B point (B) and sigmoid notch were positioned to represent the mandible. In general, automatic 3D reconstruction of the condyle is not fully reliable compared to the outer surface of the mandibular body and ramus.²¹ Therefore, the condylar neck point (CdN), the most posterior midpoint of the condylar neck at the same vertical level as the sigmoid notch, was used to represent the condylar position (Figure 2C and D).²¹ The Gonion posterior point (Go post), the most posterior midpoint of the mandibular angle area, ramus line, tangent line from the CdN to Go post, and ramus line angle, the angle between the ramus line and the horizontal reference plane, were used to measure the rotation of the mandible (Table 2).

Table 2. Landmark and Reference Plane Definitions

View Fullscreen

The 3D displacement of landmark coordinates at T2 were compared to T1 (Δ) on the x (transverse; +, lateral displacement), y (anterior-posterior (AP); +, anterior), and z (vertical; +, inferior) axes. The amount of displacement in CdN from T1 to T2 (ΔCdN) was evaluated on both sides per participants. Based on previous reports, a 1 mm threshold was applied to determine condylar displacement.^10,22,23 Those with ΔCdN >1 mm in any of the x, y, and z axes on either or both sides of the condyle were classified as unexpected condylar displacement (+) while, otherwise, they were classified as displacement (−) (Figure 2E–G). The incidence of unexpected condylar displacement was calculated as the percentage (%) of displacement (+) among 291 participants.

Morphologic changes of the condyle were compared using T1 and T2 mandibular 3D surface models of displacement (+) using ITK-SNAP software. Segmentation consisted of outlining the cortical boundaries of the condylar region using semi-automatic and manual discrimination procedures that allowed manual editing and slice-by-slice analysis in all three planes (sagittal, coronal, and axial).²⁴ T2 mandibular models were regionally superimposed based on CdN and its surrounding area on the condylar neck surface of the T1 models using the surface registration module of 3D slicer (Figure 2H). The distance between the T1 and T2 superimposed model was calculated to represent the differences using the model to model the distance module. The surface models were converted to a color-coded map using the shape population viewer module.^24,25 The American Board of Orthodontists Objective Grading System (ABO-OGS) was used to evaluate the overall orthodontic outcome.

RESULTS

Displacement of the Mandible following Comprehensive Orthodontic Treatment in Adults

The mean changes of B, Pog, and CdN between T1 and T2 were approximately 0.1–0.3 mm. Displacements of B (ΔB) and Pog (ΔPog) were greater than CdN (ΔCdN). Overall, A-P (Δy) or vertical changes (Δz) were greater than the transverse changes (Δx) (P < .05) (Table 3). The mean rotational change was 0.1 ± 1.12° (max/min, 4.65/−3.71).

Table 3. 3D Displacement of Mandibular Landmarks in Adult Orthodontic Population

View Fullscreen

Incidence and Pattern of Unexpected Condylar Displacement

ΔCdN exceeding 1 mm (displacement [+]) was noted in 18 participants among the total of 291. The condylar displacement pattern varied. All 18 participants showed AP (lΔyl >1 mm) displacement either in the anterior (9, 3.1%) or posterior direction (9, 3.1%) combined with vertical displacement (lΔzl >1 mm). The incidence rate of unexpected condylar displacement was 6.2% (18/291) (Table 4).

Table 4. Incidence of Unexpected Condylar Displacement in Adult Orthodontic Population

View Fullscreen

Among the 18 displacement (+) patients, five (1.7%) showed unilateral displacement, and 13 (4.5%) showed bilateral displacement. The majority underwent extraction for anterior retraction (N = 11) followed by upper (N = 3) or both upper and lower arch distalization (N = 2) with TADs, and simple alignment (N = 2). Vertically, two patients underwent upper molar intrusion with TADs (Table 5).

Table 5. Pattern of Condylar Displacement Among Displacement (+) Subjects

View Fullscreen

Clinical Factors Associated with Unexpected Condylar Displacement

Condylar displacement was more frequent in females (8.7%, 17/195) than in males (1.0%, 1/96) (P = .011) and in skeletal Class II (12.5%, 14/112) than in skeletal Class I (2.6%, 4/154) and Class III malocclusions (0.0%, 0/25) (P = .002). Age, treatment duration, extraction pattern, or the presence of a prior TMD history were not associated with the incidence of condylar displacement (Table 6). Females (vs males; OR: 9.07 [95% CI: 1.189–69.228]) and skeletal Class II (vs Classes I and III; OR: 4.57 [95% CI: 1.582–13.196]) had significantly higher odds of unexpected condylar displacement after orthodontic treatment (P < .05) (Table 7).

Table 6. Clinical Factors Associated With Unexpected Condylar Displacement

View Fullscreen

Table 7. Odds of Unexpected Condylar Displacement

View Fullscreen

Morphologic Evaluation of the Condyles and Clinical Outcome

Condylar displacement may be induced directly by positional changes in the condyle within the fossa and/or by morphologic changes such as condylar resorption. To determine the extent to which condylar resorption was a factor, T1 and T2 3D reconstructed mandibular surface models of displacement (+) were superimposed based on the CdN and its surrounding area of the condylar neck surface. Based on the color-coded maps, distinct morphological changes in the condyles >1 mm were not observed in the displacement (+) group (Figure 3, red outward, blue inward). The average ABO-OGS score of the displacement (+) group was 21.6 ± 11.7 point.

View Figure

• Download as Powerpoint
• Download Figure

DISCUSSION

The incidence of unexpected/excessive condylar displacement >1 mm shortly after conventional orthodontic treatment in adults was approximately 6%. Despite these concerns, condylar resorption or negative effects on final orthodontic outcomes were not evident in patients with unexpected condylar displacement.

Although short-term temporal changes or age-related changes in the condyle/fossa relationship among untreated/asymptomatic adults are limited,^11,26 in general, minor changes in the condyle position after various orthodontic modalities are considered clinically insignificant.^8–11 However, when >1–2 mm of condylar displacement is observed, close monitoring is recommended due to the possible association with TMD symptoms and poor prognosis.^10,17 Given that the main goal of this study was to monitor the incidence of unexpected excessive condylar displacement that may negatively affect treatment outcomes in a large-scale clinical setting, condylar displacement >1 mm was set as the cut-off margin within the orthodontic cohort.

Based on this study, the incidence of condylar displacement >1 mm shortly after comprehensive orthodontic treatment among adult patients was 6.2%, similar or slightly higher than other well-known complications or risks associated with orthodontic treatment, such as root resorption (1%–5%), gingival recession (1.3%–10%), dental caries (2.3%), and differential alveolar bone modeling (3.2%).^27–31 The incidence of unexpected condylar displacement was more frequent in the A-P direction than the transverse direction, similar to previous reports,^9,10,32 possibly due to the anatomical shape of the glenoid fossa and positional relationship of the TMJ disc.^7,33 Anterior displacement (with or without inferior displacement) of the condyle was observed about as often as was posterior displacement (with or without superior displacement).

Although the direct comparison of segmented reconstructed 3D images of the condyle before and after treatment allowed visualization of displacements in the condylar region, precise and efficient reconstruction of the condylar head requires a semi-automatic or manual segmentation process that limits its application in large-scale studies.²¹ Thus, after 3D superimposition of the stable anterior cranial base structures of T1 and T2, the amount of condylar displacement and its pattern in the adult orthodontic cohort were estimated by changes in the landmark CdN, a bilateral landmark located in a relatively stable structure of the condylar neck, along any of the three axes. To rule out the possibility of secondary displacement due to morphological changes in the condyle, segmented 3D models of the condyles in the displacement (+) group were monitored in detail by regional superimposition of the condylar neck. Color-coded distance maps of T1 and registered T2 3D surface models can illustrate condylar modeling including resorption (− value, in blue) and new bone formation (+value, in red) if any.^34,35 Although the range of remodeling can be accurately measured in a 0.1 mm scale or visualized in different colors in a 0.2 mm scale,³⁴ the current results indicated limited changes in the condylar shape and active remodeling after displacement.

Clinically, females are known to have a higher risk of TMD than males due to hormonal factors as well as differences in pain sensitivity.³⁶ Also, Class II patients have an increased risk of joint instability.^36,37 Similarly, the results also indicated that initial host conditions, such as females (vs males) with Class II malocclusion had a higher risk of unexpected condylar displacement than did specific treatment factors or having a prior TMD history. Although excessive condylar displacement was not associated with TMD or condylar resorption/remodeling, this should be interpreted with caution. TMD-associated symptoms were evaluated based on patient self-reported questionnaires collected before treatment and these were not monitored after treatment or during the retention period. Thus, long-term follow-up of clinical symptoms associated with unexpected condylar displacement is necessary in the future.

Due to the nature of this study being designed to represent the overall incidence of displacement in the adult orthodontic population, the demographics of the participants were not controlled or matched. Adult Asians who were treated with various comprehensive orthodontic treatment modalities including anterior retraction and total arch distalization with or without extraction were included. Adult dentofacial orthopedics using TADs for active molar intrusion or full-arch intrusion to induce autorotation of the mandible was also used when necessary. Based on this orthodontic cohort, extraction pattern per se was not associated with unexpected condylar displacement. However, comparison among different treatment modalities was not feasible due to the limited incidence of condylar displacement found. Further case control studies to compare condylar changes with different treatment modalities could be beneficial, especially in cases with mandibular rotation.

Although changes in CdN were used to estimate condylar changes, this measurement did not directly represent the regional displacement of the condylar head within the fossa. However, applying 3D cranial base superimposition and landmark measurements was a simple and effective method to screen for the overall incidence in a large-scale orthodontic population. Clinicians should consider that approximately 6% of adult patients in the daily orthodontic practice may experience unexpected excessive condylar displacement after orthodontic treatment, especially females with skeletal Class II malocclusion. Informed consent may be necessary in advance, with close monitoring of the mandibular position in high-risk patients. However, it is worth noting that the condylar displacement per se was not associated with poor treatment outcome or condylar resorption.

CONCLUSIONS

• Unexpected 3D condylar displacement was noted in approximately 6% of adult patients who underwent orthodontic treatment.

• However, the condylar displacement was not associated with condylar resorption and, overall, final orthodontic outcomes were acceptable.

SUPPLEMENTAL DATA

Supplement #1 available online.