The temporomandibular joint (TMJ) is a diarthrosis between the mandibular condyle and the glenoid fossa of the temporal bone. Its movement is both ginglymus and arthrodial, permitting a gliding forward-and-backward motion. Mandibulectomy that includes the condyle is sometimes necessary during treatment of patients with osteomyelitis, trauma, or tumor. During such surgery, the attachment of masticatory muscles and fibers around the mandibular ramus and condyle is cut off. Although vascularized tissue transfer can be used to rebuild the anatomy of the mandibular ramus and condyle, it cannot maintain the appropriate condyle–disc relationship during mandibular movements. Without TMJ reconstruction, postoperative neocondyle dislocation will affect the appearance and function of patients.

Many methods have been used for reconstruction of the masticatory muscle attachment to stabilize the neocondyle position^1,2. González-García et al.¹ used sutures between the masseter, pterygoid muscle, and fibular flap to stabilize the neocondyle in the glenoid fossa. Gravvanis et al.² found that free fibular flap transfers with direct seating of the fibula into the condylar fossa, followed by masseter muscle reinsertion, provided acceptable functional reconstruction of the mandibulectomy–condylectomy defect. However, these methods did not restore the appropriate condyle–disc relationship. Furthermore, these earlier studies did not use objective measurements of TMJ function. We have devised a TMJ capsule suspension technique to stabilize the neocondyle during TMJ reconstruction with a free fibular flap. In this study, we describe the technique and determine the efficacy and reliability of the technique using objective assessments of postoperative TMJ structure and function.

This prospective cohort study was performed in accordance with the tenets of the Declaration of Helsinki. Written informed consent was obtained from all patients, and approval was obtained from the Institutional Review Board (IRB) of the Peking University School and Hospital of Stomatology (approval No. PKUSSIRB-202055073).

1. Study population

For the study, we prospectively recruited patients who underwent mandibulectomy involving the condyle and mandibular reconstruction with a free fibular flap at Peking University School and Hospital of Stomatology between June 2020 and June 2021. All patients were treated with the TMJ capsule suspension technique. To form a control group, we selected patients who had undergone mandibulectomy with condylectomy and mandibular reconstruction with free fibular flap between September 2018 and September 2020; these patients had not received reconstruction of the masticatory muscle attachment and suspension of the neocondyle.

The inclusion criteria for participants were (1) mandible defect type H according to the HCL classification described by Jewer et al.³; (2) stable occlusion before and after the operation; (3) unilateral mandibular defect involving the condyle; and (4) fixation of fibular segments with miniplates. The exclusion criteria were (1) communication disorders or any difficulty in understanding commands to complete an action or (2) edentulous state or preserved teeth not sufficient to meet the requirement for para-occlusal attachment retention.

2. Surgical technique

All tumor resections and mandibular reconstructions were performed by the same surgeon (XP). The computer-aided design (CAD) and intraoperative navigation system that were used have been described in a previous study⁴. The TMJ capsule and disk were carefully preserved in all cases. The distal end of the reconstructed fibula was rounded and shaped to form a neocondyle to fit underneath the TMJ disc.

After vascular anastomoses, the TMJ capsule suspension was performed. The TMJ capsule and disc were identified. First, a circular line was marked 3 mm beneath the distal end of the fibula. Two holes were drilled on the circular line anteriorly and posteriorly. A needle with a tough nonabsorbable suture was passed through the anteromedial wall of the TMJ capsule and then through the anteromedial hole on the fibula. The puncture points on the TMJ capsule were all 5 mm from the border of the TMJ disc. Second, the needle was turned around and passed through the anterolateral hole on the fibula from the marrow cavity to the outside. The needle was then passed through the anterolateral wall of the TMJ capsule. The anterior suspension suture was completed using a horizontal mattress suture style. The posterior suspension suture was completed in a similar way. With this, the neocondyle was seated in the glenoid fossa. The position of the free fibular flap was identified by the intraoperative navigation system in relation to osseous reference points. Finally, the anterior and posterior suspension sutures were tightened and knotted.(Fig. 1) The suspension suture between the neocondyle and the joint capsule was completed to rebuild the TMJ inferior space.

Assessment of mandibular movements and surface electromyography of masseter

At least 6 months following the operation, the trajectory of mandibular movement and the surface electromyogram (EMG) signals of bilateral masseters were recorded using the zebris System (zebris Medical GmbH). As recommended in the literature^5,6, the following were recorded: (1) trajectory of mandibular movement from maximum mouth opening position to intercuspal position (three trials); (2) trajectory of mandibular movement from marginal position (left and right side) to intercuspal position (three trials); (3) trajectory of mandibular movement and the surface EMG signals of bilateral masseters during normal mastication (10 trials); (4) trajectory of mandibular movement and the surface EMG signals of bilateral masseters during unilateral (left and right side separately) chewing of sugar-free gum (10 trials); and (5) mean myoelectric values of bilateral masseters at rest state and at maximum bite state for 5 seconds.

The study variables of mandibular movements included: (1) maximum opening; (2) lateral (left and right) movement distance; and (3) condyle path length. The asymmetry index of the condyle path length was used to compare the difference in condyle path length between the study group and the control group: asymmetry index of condyle path length=(condyle path length on normal side−condyle path length on affected side)/(condyle path length on normal side+condyle path length on affected side). This value represented the discrepancy in bilateral condyle movement trajectory. The study variables of the surface EMG were (1) mean electromyographic value of affected side masseter at rest; (2) mean electromyographic value of affected side masseter during maximal bite; and (3) mean electromyographic value of affected side masseter during chewing.

3. Acquisition and analysis of TMJ magnetic resonance imaging (MRI) data

All patients in the study group underwent MRI of the TMJ at the same imaging center. Oblique sagittal and coronal proton density–weighted images at mouth opening state and at mouth closing state were obtained at least 6 months following the operation to assess the postoperative neocondyle–disc relationship and the morphology of the TMJ disc. The following parameters were measured: (1) TMJ disc thickness at the levels of the anterior, intermediate, and posterior bands and (2) TMJ disc length.(Fig. 2)

4. Statistical analysis

R 4.0.0 (https://www.r-project.org/) was used for data analysis. Quantitative data were checked for normality of distribution and homogeneity of variance using the Shapiro–Wilk test and Bartlett’s test. The independent samples t-test was used if the measurement data were normally distributed; otherwise, a non-parametric test was used. Statistical significance was set at P≤0.05.

The study group (TMJ capsule suspension group) comprised 9 patients (8 male, 1 female; mean age, 40.00±22.59 years; age range, 12-70 years). The indication for mandibulectomy was tumor in all cases (benign in 3 and malignant in 6). The retrospectively assembled control group comprised 9 patients (6 male, 3 female; mean age, 33.67±17.99 years; age range, 8-57 years). The indications for mandibulectomy were ameloblastoma (n=7), ossifying fibroma (n=1), and odontogenic keratocyst (n=1). Intraoperatively, the TMJ disc and capsule were carefully preserved in all cases. All free fibular flaps survived without vascular crisis; there were no donor site or recipient site complications.

1. Mandibular movements and surface electromyography of masseter

The postoperative mean maximum mouth opening was 37.24±2.14 mm in the study group vs. 37.83±4.54 mm in the control group. No patient had limitations in mouth opening. Table 1 presents the data on postoperative maximum mouth opening, bilateral marginal movement distances, and asymmetry index of the condyle path length in the two groups. The asymmetry index of the condyle path length was significantly higher in the control group (P=0.02). In the study group, bilateral mouth opening trajectories were almost symmetric in 7 patients and deviated to the affected side in 2 patients. In the control group, the mouth opening trajectory was deviated to the affected side in all patients.

The mean electromyographic values of the masseter on the affected side in rest state, maximum bite state, and chewing state were 4.96±0.91 μV, 46.28±16.77 μV, and 14.57±6.32 μV, respectively, in the study group vs. 4.33±0.72 μV, 49.93±17.11 μV, and 15.29±6.72 μV, respectively, in the control group. The values in the three states were not significantly different between the two groups (P=0.13, P=0.65, P=0.82).(Fig. 3) On account of the dentition defect, the electromyographic values on the affected side were assessed in the gum chewing state.

2. MRI assessment

The mean time to follow-up TMJ MRI in the study group was 12.78±1.39 months. The mean thicknesses of the TMJ disc at the levels of the anterior, intermediate, and posterior bands were 2.27±0.21 mm, 1.64±0.11 mm, and 2.36±0.21 mm, respectively, on the affected side vs. 2.33±0.27 mm, 1.74±0.15 mm, and 2.43±0.24 mm on the normal side. The mean lengths of the TMJ disc on the affected side and normal side were 12.50±0.70 mm and 12.44±0.70 mm, respectively. The differences in these parameters between the two sides were not statistically significant (P=0.57, P=0.13, P=0.48, and P=0.87, respectively.(Fig. 4)

3. A representative case

A 15-year-old adolescent presented with osteofibroma. After obtaining informed consent, the patient was recruited in the TMJ capsule suspension group. With the help of CAD and the intraoperative navigation system, mandibulectomy involving the condyle was performed, followed by mandibular reconstruction with free fibular flap. The TMJ capsule suspension technique was used.(Fig. 5) The postoperative course was uneventful. At 12 months after the operation, TMJ MRI showed normal superior TMJ disc position both in mouth opening and mouth closing.(Fig. 6) The mouth opening trajectory showed no deviation. During mouth opening, the movement trajectory of the normal side condyle and the neocondyle were symmetric.(Fig. 7) The esthetic and functional outcomes were satisfactory.

The TMJ between the mandibular condyle and the temporal glenoid fossa is one of the most complex joints in the body⁷. Several techniques have been used to reconstruct the mandibular condyle after mandibulectomy with condylectomy; for example, autogenous costochondral graft, total alloplastic joint prosthesis, and vascularized free tissue transfer. Titanium prostheses have also been advocated, but problems such as plate fracture and wear of the head are likely over the long term.

The intrinsic properties of the fibular flap have made it widely popular for use in mandibular reconstruction^8,9. The fibular flap is particularly well suited for condylar reconstruction because of its tubular shape and adaptability to the glenoid fossa. Contouring of the pole of the fibula will facilitate passive fit in the glenoid fossa. Leaving a single side of the mandible unsecured to the glenoid fossa—in essence allowing the mandible to “hang”—is a technique that has been used by plastic and reconstructive surgeons after condylectomy. With vascularized tissue transfer, it is possible to rebuild the anatomy of the mandibular ramus and condyle, but it cannot maintain an appropriate condyle–disc relationship during mandibular movements. Our previous study demonstrated that the unstable position of the neocondyle following reconstruction with free fibular flap could affect postoperative appearance and function¹⁰. Several methods for physical stabilization of the neocondyle have been proposed. After reviewing the literature, we identified three reconstructive approaches that attempted to guarantee correct positioning of the neocondyle in the fossa.

One approach is reconstruction of the masticatory muscle attachment. Many surgeons believed that suturing the masseter muscle to the angle of the neocondyle or the reconstruction plate could seat the fibula in the glenoid fossa and maintain its position^2,11,12. Thor et al.¹³ sutured the lateral pterygoid muscle to the condylar part of the fibular flap with polydioxanone to maintain the neocondyle position. Of the five patients treated by this method, two developed incorrect positioning of the neocondyle (location anterior to the eminence), while one developed temporo-fibular ankylosis with severe limitation in mandibular mobility and mouth opening. We believe that the position of the masticatory muscle reattachment and the postoperative changes in masticatory muscle force would make it difficult to maintain the position of the neocondyle.

The second approach is suturing of the neocondyle to the articular disc. In a retrospective study of 17 patients, Wax et al.¹⁴ reported satisfactory functional and aesthetic results with free fibular transfer with intention to preserve the disc. The articular disc was attached to the neocondyle with a permanent suture. The surrounding soft tissue of the pedicle and periosteum from the fibula were sutured, covering the end of the neocondyle to prevent ankylosis. Among the 17 patients treated by this method, 2 developed postoperative displacement of the fibular head out of the fossa, while 1 patient had excessive loosening of the fixation suture. We believe that, with this method, the tightness of the suture and the presence of soft tissue between the disc and the neocondyle would affect the postoperative position of the neocondyle.

The third method is suturing of the neocondyle to a peripheral bony structure such as the anterior lip of the fossa or the zygomatic arch. Wax et al.¹⁴ recommended a suspension suture to anchor the neocondyle to the anterior lip of the fossa if oncologic margins necessitated removal of the capsule. However, the tightness of sutures, which is determined by the experience of the surgeon, would affect the position of the neocondyle.

None of the above three approaches are fully effective in maintaining an appropriate condyle–disc relationship during mandibular movements. Our previous studies showed that, with preservation of the articular disc, remodeling of the neocondyle after free fibular flap reconstruction would not lead to temporo-fibular ankylosis^4,10. In the TMJ capsule suspension technique, the fibrous connective tissue surrounding the neocondyle is restored, and this helps stabilize the position of the neocondyle. In addition, this surgical technique ensures much flexibility for the neocondyle during mandibular movements.

In previous studies, postoperative mandibular movements and TMJ function after free fibular flap reconstruction were evaluated by clinical observation and simple measurements; there were no objective measurements. Wang et al.¹⁵ used the mandibular movement recording system to evaluate TMJ function after free fibular flap reconstruction and found that the mouth opening trajectory was skewed to the affected side in all patients. In our study, the mandibular movements assessment also demonstrated the reliability and validity of this new technique.

MRI is the ideal imaging modality for all types of TMJ disorders. The standard imaging protocol consists of oblique sagittal and oblique coronal images of the TMJ that are obtained perpendicular or parallel to the long axis of the mandibular condyle¹⁶. This allows visualization of the position and morphology of the disc as well as of the bony structures. In a study on fresh autopsy material, the diagnostic accuracy of oblique sagittal and oblique coronal sections for determining disc position and bone status were 95% and 93%, respectively¹⁷. However, MRI has rarely been used postoperatively to visualize the disc and neocondyle position. The current study is the first to report the use of MRI to evaluate TMJ function after reconstruction with free fibular flap; the results showed that patients regained a normal relationship between the articular disc and neocondyle both in mouth opening and mouth closing. During the mouth opening movement, the neocondyle could freely slide on the lower surface of the articular disc.

We present a novel neocondyle stabilization technique that has the potential to improve the postoperative TMJ structure and function. Objective assessments of TMJ function showed that the technique can achieve satisfactory reconstruction of the mandibular condyle.

S.B. and Y.Y. participated in data collection and writing the manuscript. W.B.Z. and Y.Q.M. participated in the study design and performed the statistical analysis. Y.W., C.M., D.C.W., and X.P. participated in the study design and coordination and helped to draft the manuscript. All authors read and approved the final manuscript.

Written informed consent was obtained from all patients, and approval was obtained from the Institutional Review Board (IRB) of the Peking University School and Hospital of Stomatology (approval No. PKUSSIRB-202055073).

Written informed consent was obtained from the patient’s legal guardian for publication of this article and accompanying images.

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