This retrospective study evaluated the effect of surgical computer-assisted navigation in temporomandibular joint (TMJ) ankylosis gap arthroplasty. Eighteen patients (25 sides) with bony ankylosis who underwent surgical treatment under computer-assisted navigation (navigation group) from May 2011 to April 2013 were assessed, along with 19 such patients (25 sides) treated without computer-assisted navigation (non-navigation group) from March 2009 to April 2011. The navigation group patients underwent surgery with the preservation of ≥3 mm bone thickness in the skull base and anterior wall of the external auditory canal. Postoperatively, computed tomography (CT) was used to measure the residual bone thickness in the skull base and anterior wall of the external auditory canal. Maximum mouth opening (MMO) changes were evaluated at >1 year of follow-up. Postoperative CT measurements showed that the lowest skull base thickness in the navigation group was significantly lower than that in the non-navigation group (3.86 ± 1.95 mm vs. 6.01 ± 3.07 mm, P = 0.009). The lowest thicknesses of the anterior wall of the external auditory canal were similar in the two groups. Postoperative follow-up showed similar average MMO in the two groups. Therefore, with the navigation system, TMJ ankylosis gap arthroplasty can achieve more extensive removal of the ankylosed bone, at least towards the skull base, under the premise of ensuring a safety distance of 3 mm.
Temporomandibular joint ankylosis (TMJ ankylosis) is a mandibular movement disorder that results from intra-articular fibrous or bony adhesion caused by lesions in the TMJ. It usually leads to secondary facial malformation, occlusal disorders, and sleep apnoea syndrome, seriously affecting physical function, body development, and the appearance of patients. The surgical treatment of TMJ ankylosis by joint gap arthroplasty was proposed by Humphry et al. in 1878. In 1990, Kaban et al. proposed a treatment programme for TMJ ankylosis, which was further improved in 2009.
Regardless of the type of surgery selected, the first step in the surgical treatment of TMJ ankylosis is an extended resection of the ankylosed bone. Kaban et al. suggested that a bone gap of 1.5–2.5 cm should be preserved after completion of the osteotomy, because a too close bone section would result in new adhesions and consequently the recurrence of the ankylosis. Others have advocated that a bone gap of 0.5–1.0 cm should be maintained, because a distance exceeding 1 cm would shorten the mandibular ramus height, causing a postoperative open bite and mandibular deviation.
Although the ideal width of the osteotomy remains controversial, full removal of the bone adhesion is widely accepted by researchers to be a necessary condition for releasing the joint and reducing the recurrence of ankylosis. However, the removal of a sufficient amount of ankylosed bone is very difficult and highly risky. The anatomy of the TMJ region is very complex, with many important adjacent structures. In addition, bone adhesion makes the anatomical structure more ambiguous. The need to protect the surrounding structures also limits an adequate resection of the ankylosed bone, which may increase the risk of postoperative recurrence. Thus, how to achieve maximum removal of the ankylosed bone while ensuring safety constitutes a problem in the surgical treatment of the TMJ. Traditional treatment methods depend greatly on surgeon experience.
Computer-assisted navigation technologies are now used widely in the treatment of oral and maxillofacial diseases. Assisted by a computer navigation system, one can achieve accurate real-time intraoperative positioning, and accurately transfer the preoperative design to the actual operation. This is undoubtedly highly suitable for intraoperative real-time positioning in TMJ ankylosis surgery, as it allows the relationships with the surrounding tissues to be seen. The application of surgical navigation systems in TMJ ankylosis surgery has been reported widely and is advocated for its usefulness in improving the safety and accuracy of the operation. However, relevant large sample size studies are scarce. No data have been reported from studies assessing whether improvements are obtained postoperatively following the use of navigation surgery, in comparison with the traditional surgery.
This retrospective study evaluated the effect of computer-assisted navigation in TMJ ankylosis gap arthroplasty.
Materials and methods
A total of 18 patients with TMJ ankylosis, who underwent computer-assisted navigation surgery in the oral and maxillofacial trauma centre of the study institution from May 2011 to April 2013, were selected as the navigation group (group A). A further 19 such patients with TMJ ankylosis, who underwent surgery without the use of computer-assisted navigation from March 2009 to April 2011, were reviewed as the non-navigation group (group B). The two groups included 13 cases of bilateral TMJ ankylosis and 24 of unilateral TMJ ankylosis, giving a total of 50 affected sides ( Fig. 1 ).
Inclusion criteria were the following: (1) sustained limitation of mouth opening lasting ≥8 weeks; (2) mouth opening ≤25 mm; (3) imaging examination showing damage to the TMJ structure, with osseous adhesion and Sawhney classification type II–IV ankylosis (type I: fibrous ankylosis; type II: osseous adhesion formed in the joint anterolaterally; type III: a wide range of osseous adhesion in the joint, but not exceeding the scope of the joint; type IV: ankylosed bone adhesion exceeding the range of the joint, with formation of extensive adhesion in the skull base, roots of the zygomatic arch, and ramus) ; (4) operations performed by the same surgeon; (5) patient re-examined by computed tomography (CT) within 2 weeks of the operation. This study was approved by the ethics committee of the study institution; all patients provided informed consent.
The 18 patients in group A (navigation group) included seven bilateral cases and 11 unilateral cases, for a total of 25 affected sides. This group comprised 10 males and eight females, with an average age of 28.6 years (range 9–62 years). By aetiology, there were two cases of inflammation, 10 of traffic accident injury, and six of falls, with a duration of between 12 and 320 months (average 85.3 months). Preoperative CT scans indicated six sides with Sawhney type II ankylosis, 10 with Sawhney type III, and nine with Sawhney type IV. Maximum preoperative mouth opening was 5.39 ± 5.59 mm (range 0–16 mm). Group A patients underwent computer-assisted navigation surgery.
The 19 patients in group B (non-navigation group) included six bilateral cases and 13 unilateral cases, amounting to a total of 25 affected sides. This group comprised 10 males and nine females, with an average age of 21.2 years (range 6–61 years). By aetiology, there were four cases of inflammation, 10 of traffic accident injury, and five of falls, with a duration of between 4 and 204 months (average 61.2 months). Preoperative CT scans revealed five sides with Sawhney type II ankylosis, 12 with Sawhney type III, and eight with Sawhney type IV. Maximum preoperative mouth opening was 7.89 ± 6.02 mm (range 1–20 mm). Group B patients all underwent traditional surgical treatment, rather than computer-assisted navigation surgery.
There was no significant difference between the two groups in age ( P = 0.146) or preoperative maximum mouth opening ( P = 0.198) ( Table 1 ).
|Navigation group||Non-navigation group||P -value|
|Age, years||28.61 ± 15.91||21.21 ± 14.21||0.146|
|Preoperative maximum mouth opening (mm)||5.39 ± 5.59||7.89 ± 6.02||0.198|
|Thickness of the skull base (mm)||3.86 ± 1.95||6.01 ± 3.07||0.009 a|
|Thickness of the anterior wall of the external auditory canal (mm)||3.43 ± 2.40||4.26 ± 2.78||0.377|
|Maximum mouth opening (mm)||36.39 ± 7.36||32.47 ± 13.16||0.976|
|Ankylosis recurrence rate||5.6% (1/18)||21.1% (4/19)|
All patients were examined by spiral CT before and after the operation (helix with 0.625-mm slice thickness; Bright Speed 16, GE Healthcare, Buckinghamshire, UK). Preoperative CT data from the CT database were transferred to disc in DICOM format (Digital Imaging and Communications in Medicine). The CT data were then processed and transferred into iPlan CMF software (Brainlab AG, Feldkirchen, Germany), for preoperative surgical planning and postoperative evaluation. The VectorVision navigation system (Brainlab) was used for surgical navigation.
Preoperative patient CT data were imported into the computer navigation system (VectorVision infrared navigation workstation) from a CD-ROM or U disc. After the induction of general anaesthesia, a reference frame with three light-reflecting balls was rigidly fixed to the patient’s skull to identify his/her position. Registration was then completed through facial surface scanning using a Z-touch wireless laser pointer. The software automatically verified registration accuracy for all patients; the registration error was less than 1.5 mm in all cases.
A pre-auricular incision was applied to expose the TMJ region. Then, a fissure drill was used to perform the osteotomy, cutting and removing the ankylosed bone in big pieces. To render the procedure safe, the navigation system was used to indicate the distance between the actual position of the operation and the medial border of the ankylosed bone.
After completion of the osteotomy and TMJ release, a round bur was used to remove the ankylosed bone at the upper bone surface and to carefully shape the glenoid fossa. During this step, the distance from the top of the glenoid fossa to the skull base of the medial cranial fossa and that from the posterior margin of the fossa to the anterior border of the bony external auditory canal was monitored via the navigation system ( Figs. 2 and 3 ). The ‘tooltip’ function in the navigation software was used to display the safety distance, with an additional instruction for a 3-mm distance along the probe direction in the real-time position of the navigation probe ( Figs. 2B and 3B ). After shaping the glenoid fossa, at least 3 mm of bone thickness was maintained at the skull base and anterior wall of the bony external auditory canal. Finally, a 10–20-mm bone gap was eventually formed from the top of the glenoid fossa to the residual condyle or mandibular ramus. After gap arthroplasty of one or two sides, depending on the patient’s situation, the passive maximum mouth opening distance was checked. If this did not reach 35 mm, arthroplasty of the other side or resection of the coronoid process was done until more than 35 mm of passive maximum mouth opening was obtained.