In this study, we quantitatively assessed 3-dimensional condylar displacement during counterclockwise maxillomandibular advancement surgery (CMMA) with or without articular disc repositioning, focusing on surgical stability in the follow-up period.
The 79 patients treated with CMMA had cone-beam computed tomography scans taken before surgery, immediately after surgery, and, on average, 15 months postsurgery. We divided the 142 condyles into 3 groups: group 1 (n = 105), condyles of patients diagnosed with symptomatic presurgical temporomandibular joint articular disc displacement who had articular disc repositioning concomitantly with CMMA; group 2 (n = 23), condyles of patients with clinical verification of presurgical articular disc displacement who had only CMMA; and group 3 (n = 14), condyles of patients with healthy temporomandibular joints who had CMMA. Presurgical and postsurgical 3-dimensional models were superimposed using voxel-based registration on the cranial base. Three-dimensional cephalometrics and shape correspondence were applied to assess surgical and postsurgical displacement changes.
Immediately after surgery, the condyles moved mostly backward and medially and experienced lateral yaw, medial roll, and upward pitch in the 3 groups. Condyles in group 1 showed downward displacement, whereas the condyles moved upward in groups 2 and 3 ( P ≤0.001). Although condylar displacement changes occurred in the 3 groups, the overall surgical procedure appeared to be fairly stable, particularly for groups 1 and 3. Group 2 had the greatest amount of relapse ( P ≤0.05).
CMMA has been shown to be a stable procedure for patients with healthy temporomandibular joints and for those who had simultaneous articular disc repositioning surgery.
Counterclockwise maxillomandibular advancement surgery has shown stable results when TMJ pathology is appropriately managed.
Point-to-point correspondence across all measured surfaces is used.
A new 3D cephalometric method is presented that showed high reproducibility.
Counterclockwise maxillomandibular advancement surgery (CMMA) has often been used to treat hyperdivergent skeletal Class II patients. This surgical technique was developed as an effective means to achieve optimal functional and esthetic outcomes in patients with high occlusal plane facial deformities.
However, skeletal relapse after that orthognathic surgery has been a major issue because of problems related to the stretching of suprahyoid, pterygoid, and masseter muscles, as well as adverse effects on the temporomandibular joints (TMJs). Clinical concerns have been raised regarding the influence of suboptimal intraoperative positioning of the proximal segments: ie, condylar torque, which may be associated with progressive condylar resorption and subsequent postoperative relapse.
CMMA has been described as a stable procedure for patients with healthy TMJs. However, controversial opinions surround the appropriate treatment plan for those with preexisting TMJ disorders who need such orthognathic surgery for correcting jaw deformities and malocclusions.
Some authors have suggested that orthognathic surgery alone may reduce or eliminate TMJ dysfunction and symptoms, whereas others have reported damaging effects to the condyles from such surgery when there is internal derangement of the TMJs. For instance, after mandibular advancement, it may happen from muscular activity, which causes the discs to remain displaced as the condyles assume a superoposterior position in the fossae by an increase in mechanical loading.
Some studies have shown that concomitant surgical correction of dentofacial deformities and TMJ disorders by repositioning and stabilizing the articular disc using the Mitek anchor technique (Mitek Products, Westwood, Mass) provides great treatment outcomes for most patients concerning functional, esthetic, and psychological aspects. Contrariwise, specific condylar displacement changes during articular disc repositioning surgery might be investigated as potential factors inducing condylar remodeling in the long-term follow-up, because of the condylar loading alteration. The current literature is still not clear about the best treatment option for preventing degenerative condylar changes after bimaxillary surgical advancement.
Cone-beam computed tomography (CBCT) has been used for assessing condylar changes and surgical relapse. However, most previous studies have measured longitudinal changes by using 2-dimensional tools, which are susceptible to errors in determining corresponding landmark positions when bone remodeling occurs. Accurate quantitative 3-dimensional (3D) image techniques are now available, giving clinicians a new imaging modality to evaluate postoperative skeletal relapse as well as positional and dimensional condylar changes.
The aim of this study was to quantitatively assess 3D condylar displacement changes during CMMA with or without articular disc repositioning, focusing on surgical stability in the follow-up period.
Material and methods
This retrospective study sample was composed of CBCT scans and clinical records from patients who had CMMA by the same surgeon (L.M.W.). Inclusion criteria were (1) osteotomies performed and stabilized with rigid internal fixation; (2) female patients at least 15 years old and male patients at least 17 years old; (3) patients with no TMJ abnormalities and with TMJ disc displacement assessed in clinical examinations and on magnetic resonance imaging interpreted by 2 experienced and calibrated doctors (L.M.W. and J.R.G.); and (4) CBCT scans acquired at 3 time points: before surgery (T1), immediately after surgery (T2), and at least 6 months postsurgery (T3). The exclusion criteria were patients with (1) craniofacial syndromes, (2) systemic degenerative conditions, (3) severe facial asymmetry, (4) previous TMJ surgery, and (5) previous arthroscopy, arthrocentesis, or viscosupplementation.
Records from 226 subjects consecutively treated from October 2008 to January 2011 were evaluated. One hundred nine patients were excluded for having undergone total prostheses of the TMJ. Thirty-eight patients were excluded for not having CBCT scans at all 3 time points (12 had TMJ articular disc repositioning surgery using the Mitek anchor technique (Mitek Products, Westwood, Mass), and the other 26 had no TMJ intervention). Therefore, 79 patients matched the inclusion criteria for this study.
A total of 158 condyles were analyzed; 16 condyles were excluded due to previous arthroplasty. The final sample included 142 condyles divided into 3 groups: group 1 (n = 105), condyles of patients diagnosed with symptomatic presurgical TMJ articular disc displacement who underwent articular disc repositioning concomitantly with CMMA. Many condyles in this group had osteoarthritis, showing severe flattening of the condylar surface, subchondral cysts, erosions, and osteophytes, causing considerable deformation of the condylar structure. In group 2 (n = 23), the condyles were of patients with clinical verification of presurgical bilateral TMJ articular disc displacement, mostly without osteoarthritic signs or symptoms, who underwent only CMMA. In group 3 (n = 14), the condyles were of patients with healthy TMJs who had CMMA. All patients signed an informed consent form for hospital admission, surgical procedures, and release of information for research purposes. This study was approved by the institutional review board of the University of Michigan and complied with the Helsinki Declaration.
If indicated, articular disc repositioning surgery was performed using the Mitek anchor technique. This is an open-joint procedure performed simultaneously with the orthognathic surgery. Only salvageable discs were indicated for this surgery. A modified endaural incision was used to access the TMJ. The superior joint space was entered by incising the capsular ligaments, and the inferior joint space was entered with an incision just above the lateral pole of the condyle. The hyperplasic bilaminar tissue was wedge resected. The disc was mobilized and passively positioned over the condyle, with the lateral pterygoid muscle attachment preserved. The Mitek anchor with two 0 Ethibond sutures (Ethicon, Somerville, NJ) attached was inserted in the posterolateral surface of the condylar head, approximately 8 mm below the condylar top. The Ethibond sutures were attached to the posterior aspect of the posterior band of the disc for stabilization. The joint was then irrigated and the incision closed.
After the TMJ surgery, the orthognathic surgery was performed. Counterclockwise rotation and advancement of the maxillomandibular complex was routinely performed on these patients that included bilateral mandibular ramus osteotomies and multiple maxillary (LeFort 1) osteotomies. Bilateral mandibular ramus sagittal split osteotomies were performed; the mandible was placed into its final position with an intermediate splint and intermaxillary fixation, and internal rigid fixation using bone plates and screws. Maxillary osteotomies were then performed, internasal procedures were completed if indicated, a palatal splint was inserted, intermaxillary fixation was placed, and rigid fixation was applied using 4 bone plates fixated with 2.0-mm diameter screws.
The protocol for image acquisition was carried out with the patients sitting upright, keeping the Frankfort horizontal plane (trago-infraorbital) parallel to the ground. The mandible was positioned in centric relationship with the lips relaxed, and the patients were instructed not to swallow. CBCT images were obtained in the same machine (i-CAT CBCT, 120 kV, 5 mA; Imaging Sciences International, Hatfield, Pa) using a 17 × 23-cm extended field of view protocol, during a 17.8-second scan, with a 0.3-mm isotropic voxel size. Records were taken 1 day (range, 1-2 days) before the surgery (T1), 5 days (range, 3-9 days) after surgery (T2), and in the longest follow-up (T3), on average, 15.4 months after surgery (range, 6-52 months).
The CBCT images were reformatted to 0.5-mm isotropic voxel size for the segmentation of the anatomic structures of interest. Three-dimensional models of the cranial base, maxilla, and mandible were constructed by outlining the cortical threshold using a semiautomatic procedure (ITK-SNAP software; www.itksnap.org ).
The ITK-SNAP software was also used for cropping the cranial base model. This model indicated the registration program (CMF registration, 3DSlicer), the specific place where we wanted the different time-point models to be superimposed. The cranial base was used as the reference for registration because it remains stable over time and does not change with surgical treatment. By using an automated voxel-wise rigid registration method that allowed 6 degrees of freedom, the program compared and matched different time point images considering the intensities of the voxel grey scales at the cranial base.
Three-dimensional cephalometric analysis was used to determine the facial skeletal pattern before surgery and to assess the surgical changes (T1-T2) and postsurgical stability (T2-T3) (Q3DC, 3DSlicer). First, landmarks were positioned in specific places in the cranium as described in Table I . Then the software automatically calculated the SNA and SNB angles to express the anteroposterior positions of the maxilla and mandible, respectively, relative to the cranial base. The SN.GoMe angle was also calculated to show the mandibular plane inclination ( Fig 1 ).
|Nasion||N||Anterior point on the frontonasal suture in the midsagittal plane|
|Sella||S||Midpoint at the posterior wall of sella turcica, obtained by projection of the geometric center of sella passing through nasion|
|Subspinale||A||Deepest point on the anterior contour of the maxillary alveolar process in the midsagittal plane|
|Supramentale||B||Deepest point on the anterior contour of the mandibular alveolar process in the midsagittal plane|
|Menton||Me||Lowest point on the lower border of the mandibular symphysis in the midsagittal plane|
|Gonion||Go||Midpoint at the angle of the mandible, obtained by the mean distance between the right and left sides|
For analyzing specific mandibular condylar displacement changes, superimposed models were simultaneously cropped (Easy Clip, 3DSlicer). All left condyles were mirrored in the sagittal plane to form right condyles. Then condylar models were compared by subtraction to compute the surgical (T1-T2) and postsurgical (T2-T3) changes by using the shape correspondence analysis (SPHARM-PDM, 3DSlicer).
A mesh with 4002 correspondent points was generated by the shape correspondence analysis via spherical mapping and parameterization of each volume. 3DSlicer tool was then used to calculate the 3D point-wise linear distances between each time-point model (model to model distance, 3DSlicer).
Semitransparent overlays and vector maps were used to visually compare condylar displacement changes. The magnitudes of the computed 4002 differences were displayed on the condyle surface, and vector images pointed out the direction of the change.
Shape correspondence made it possible to mark the interest regions in 1 condyle alone (at T1) and propagate such regions for the other surgical time points (T2 and T3), obtaining x, y, and z coordinates for each point (Pick ‘n Paint module, 3DSlicer) ( Fig 2 ). Then the Q3DC module in the 3D Slicer software allowed measuring both translational and rotational displacements ( Fig 3 ). Positive or negative signs indicated displacement directions ( Table II ).
|Condylar displacement||Orthogonal view planes||Negative values||Positive values|
|Anteroposterior ∗||Sagittal and axial||Anterior translation||Posterior translation|
|Vertical ∗||Sagittal and coronal||Upward translation||Downward translation|
|Lateral ∗||Coronal and axial||Lateral translation||Medial translation|
|Yaw||Axial||Posterior rotation of the medial pole and/or anterior rotation of the lateral pole (medial yaw)||Anterior rotation of the medial pole and/or posterior rotation of the lateral pole (lateral yaw)|
|Roll||Coronal||Medial rotation||Lateral rotation|
|Pitch||Sagittal||Counterclockwise rotation (upward pitch)||Clockwise rotation (downward pitch)|
The reliability of the 3D cephalometric analysis and condylar displacement changes were assessed by repeating landmark positioning and measurements on the CBCT images of 10 randomly selected subjects. Two examiners (L.R.G., M.R.G.) were carefully calibrated. For intraobserver reproducibility, each examiner performed landmark positioning and measurements at 2 times, with an interval of at least 1 week between the assessments. For interobserver reproducibility, landmark positioning and measurements by each examiner were compared. We used the intraclass correlation coefficient (ICC).
Kolmogorov-Smirnov and Shapiro-Wilk tests were used to check the normality of data distribution in each group. Descriptive statistics reported presurgical (T1), surgical (T1-T2), and postsurgical changes (T2-T3) in each of the 3 groups. For normally distributed data, the differences among the groups were tested by using 1-way analysis of variance followed by the Hochberg GT2 post hoc test, appropriate for unequal sample sizes. For nonparametric data, the Kruskal-Wallis test compared the overall significance of the differences among the 3 groups, whereas the Mann-Whitney test compared 2 groups at a time (version 16.0; SPSS, Chicago, Ill). A significance level of P ≤0.05 was applied.
Three-dimensional cephalometric analysis showed high intraobserver and interobserver reproducibilities for all diagnostic variables (ICC ≥0.9) ( Table III ). The method used to measure condylar displacement changes also showed high intraobserver and interobserver reproducibilities (ICC ≥0.8) ( Table IV ). Demographic characteristics of the sample are listed in Table V . The 3 groups had similar mean ages, follow-up periods, and craniofacial patterns ( P ≥0.05). The patients on average had high mandibular plane angles and bimaxillary retrusions.
|Examiner 1, ICC||0.97||0.95||0.98|
|Examiner 2, ICC||0.99||0.98||0.97|
|Examiner 1, ICC||0.96||0.99||0.97||0.99||0.99||0.91|
|Examiner 2, ICC||0.85||0.99||0.96||0.99||0.99||0.82|
|Age (y)||Follow-up (mo)||SNGoMe (°)||SNA (°)||SNB (°)|
|105 condyles from 57 patients (75 from female and 30 from male subjects)|
|23 condyles from 15 patients (14 from female and 9 from male subjects)|
|14 condyles from 7 patients (6 from female and 8 from male subjects)|
The amounts of counterclockwise rotation and maxillary advancement were similar in the 3 groups. However, patients in groups 1 and 2 experienced greater mandibular advancement compared with group 3 ( P ≤0.01) ( Tables VI and VII ).
|Surgical change ∗ (T1-T2)||Group 1||Group 2||Group 3|
∗ Positive values indicate counterclockwise rotation, and negative values indicate clockwise rotation for SNGoMe measurements; for SNA and SNB angles, negative values indicate that the maxilla or mandible moved anteriorly, and positive values indicate that it moved posteriorly.
|Surgical change (T1-T2)||Group 1-group 2||Group 1-group 3||Group 2-group 3|
|Mean difference||P value||Mean difference||P value||Mean difference||P value|
For condylar translational changes during surgery, it was observed that, on average, the condyles moved backward and medially in the 3 groups. Patients having disc repositioning surgery (group 1) showed, on average, downward condylar displacement, whereas the condyle moved upward in groups 2 and 3 ( P ≤0.001).
Regarding mean condylar rotational changes, lateral yaw, medial roll, and upward pitch were observed in the 3 groups. However, group 1 showed a greater upward pitch compared with group 2 ( P ≤0.05) and group 3 ( P ≤0.01). Medial roll was also significantly larger in group 1 relative to group 3 ( P ≤0.01) ( Tables VIII and IX ).
|Condylar displacement ∗ (T1-T2)||Group 1 (n = 105)||Group 2 (n = 23)||Group 3 (n = 14)|