Dentomaxillofacial Radiology (2002) 31, 19 23 2002 Nature Publishing Group. All rights reserved 0250 832X/02 $25.00 www.nature.

com/dmfr

Imaging of temporomandibular joint ankylosis. A new radiographic classication

IE El-Hakim*,1 and SA Metwalli1
1

Department of Oral and Maxillofacial Surgery, School of Dentistry, Ain Shams University, Cairo, Egypt

Objectives: To compare the pre-operative clinical and radiographic ndings of temporomandibular joint (TMJ) ankylosis with those found at operation and propose new classication. Methods: Thirty-three patients were treated for TMJ ankylosis. There were 18 unilateral and 12 bilateral cases. All patients underwent pre-operative clinical and radiographic examination, (consisting of a panoramic radiograph, axial and coronal CT. The surgical ndings were compared with the imaging features. Results: Post-contrast coronal CT was the best imaging modality for planning surgery as it displayed the anatomical relationship between the ankylosed segment and the surrounding vital structures, particularly where the sphenoid and temporal bones were involved. Conclusions: Surgical planning should be based on coronal and axial CT. A new classication of TMJ ankylosis based on the CT ndings is proposed. Dentomaxillofacial Radiology (2002) 31, 1923. DOI: 10.1038/sj/dmfr/4600660 Keywords: tomography, X-ray computed; temporomandibular joint; temporomandibular joint disorders; ankylosis Introduction Bony ankylosis of the TMJ is a disabling disease that is not conned to the rst two decades, but can occur later in life.1 Dierent causes have been attributed to the condition, such as condylar fracture with involvement of the articular surface, advanced arthritis and trauma from obstetric forceps.2,3 Ankylosis can be classied into true (intra-articular) and false (extraarticular).4 True ankylosis has been classied as type I, II, III and IV. Type I occurs where the condyle is medially angulated and associated with a deformed articular fossa together with a mild-to-moderate amount of new bone formation. Type II is found where there is no recognizable condyle or fossa but instead a large mass of new bone extending from the ramus to the base of the skull. Type III ankylosis usually results from a medially displaced fracture dislocation with bone bridging the mandibular ramus to the zygomatic arch, while type IV is found when the joint architecture is replaced completely by bone with fusion of the condyle, sigmoid notch and coronoid process to the zygomatic arch and glenoid fossa.5 Radiography is an essential diagnostic tool for TMJ ankylosis. Current methods include panoramic radiography, and CT.6 Recently, the value of threedimensional CT (3D-CT) prior to surgery has been advocated.7 9 The purpose of this study was to correlate the preoperative clinical and radiographic ndings of joint ankylosis with those found at surgery. A new classication of TMJ ankylosis based on the CT ndings is proposed. Methodology A total of 33 patients (19 males and 14 females) suering from TMJ ankylosis (42 joints) were included in this study. Thirty-seven joints were operated for the rst time, three for the second time and two for the fourth time after recurrence. The patients ranged in age from 5 30 years with a mean age of 15 years. The cause of ankylosis and the time between the initial incident and the patient seeking treatment was recorded.

*Correspondence to: Ibrahim El-Hakim, 6 El-Gendy Street, Hadayek Helwan 11433, Cairo, Egypt. E-mail: imelhakim@hotmail.com Received 22 January 2001; revised 9 July 2001; accepted 10 September 2001

TMJ ankylosis IE El-Hakim and SA Metwalli

20

Panoramic views were obtained in all cases. Pre and post-contrast axial and coronal CT was performed for 30 patients to dene clearly the relation of the major blood vessels to the ankylotic segment in the area of the proposed surgery. Transcranial TMJ views were taken in three cases. 3D-CT was employed in 10 cases. Both authors examined the radiographs separately and the ndings were then compared with those at surgery. Condylectomy was performed through a standard pre-auricular incision. The subsequent procedure varied according to the operative ndings. In class I and II ankylosis, the brous adhesions were excised and the condylar head rounded and smoothed until free movement was achieved.10 In cases of class III and IV ankylosis where there was close relationship to a vital structure, the resection was started using a surgical bur and then completed with chisels, especially when there was elongation of the lateral pterygoid plate.11 In recurrent cases gap arthroplasty was performed through the same pre-auricular approach as described by Sawhney.5 Results Table 1 summarizes the clinical and radiographic ndings. The average duration of ankylosis was 2.5 years and ranged from 9 months to 18 years. Trauma was a predisposing factor in 32 cases (97% of cases) while condylar hyperplasia secondary to brous dysplasia was found in the remaining case (Figure 1). Unilateral cases of ankylosis showed clinical evidence of a hypoplastic mandibular body and the ramus on the aected side. This was evident radiographically when the onset of the lesion was in early childhood. 3D-CT conrmed the prominent mandibular angle and accentuation of the antegonial notch together with reduced vertical height of the ramus. In longstanding cases, elongation of the coronoid process was found which was not apparent on axial or coronal CT (Figure 2). The elongated coronoid process was conrmed at surgery. When coronoidectomy was performed in such cases, an improvement in mouth opening was achieved. In bilateral cases, there was severe mandibular micrognathia giving the patients the characteristic bird-face appearance. Generally, panoramic views showed that there was joint deformity, with complete loss of the joint space and abnormal bone formation in and around the joint, but did not reveal the nature and the extent of the pathology, in particular the medial and lateral extension of the ankylosed bony mass, and its relation to surrounding vital structures. These details were clearly apparent in coronal CT which illustrated whether the ankylotic mass was brous or bony. Post-contrast CT demonstrated the relation to surrounding vital structures, especially the maxillary artery (Figure 3). These ndings assisted in better

Table 1 Summary of the clinical and radiographic findings in 33 patients with TMJ ankylosis Clinical details 19 males and 14 females Age: 5 30 years (mean 15 years) Forty-two ankylosed joints: 37 joints operated for the first time, three for the second time and two for the fourth time after recurrence Cause: trauma in all cases except one was secondary to condylar hyperplasia Average ankylotic period: 2.5 years, range 9 months to 18 years Dental malocclusion in varying degree Multiple carious lesions Multiple impacted teeth in older individuals Reduced vertical height of the ramus Complete obliteration of the joint space Block of bone bridging the ramus and zygomatic arch The ankylosis appeared to extend from the lateral aspect of the zygomatic arch medially as far as the foramen spinosum and carotid canal in some cases Elongated lateral pterygoid plate that was fused to the mandible The exact relation of the maxillary artery to the ankylotic segment The sigmoid notch and the coronoid process displaced upward behind the zygomatic arch Reduced vertical height of the ramus Accentuated antegonial notch Mandibular recession in longstanding cases CT findings very similar to operative findings The use of plain radiographs led to a change in the plan during surgery

Plain radiography (panoramic and transcranial TMJ views in three cases) Axial CT scan

Coronal CT scan

3D-CT reconstruction

Surgical findings

Figure 1 Coronal CT showing bony exostoses in the glenoid fossa superiorly as well as medial on the condylar head, resulting in bony ankylosis

surgical planning and therefore reduced operating time and surgical complications. Axial CT was of great value in illustrating the relation of the vital structures at the base of the skull

Dentomaxillofacial Radiology

TMJ ankylosis IE El-Hakim and SA Metwalli

21

Figure 2 3D CT of an ankylosed joint with an elongated coronoid process (c) that projects clearly under the zygomatic arch (z)

Figure 3 Post-contrast coronal CT scan of an ankylosed joint showing its relationship to the maxillary artery (M)

Figure 4 Axial CT scans of two dierent patients illustrating the relationship of the ankylosed bone mass to the surrounding vital structures at the base of the skull. (a) C=Carotid canal; (b) Arrow=Foramen spinosum. Arrowhead=Foramen ovale

to the ankylotic mass (Figure 4). In some cases the lateral pterygoid plate appeared elongated and fused to the ankylotic bony mass (Figure 5). It was not possible to restore mouth opening after condylectomy without releasing the fusion between the lateral pterygoid plate and the mandible. In cases of recurrence, coronal CT revealed that the cause could be attributed to the inadequate resection of the ankylosed segment, with an insucient gap between the two bone ends.

The value of panoramic radiography, plain TMJ views and CT on the patients' management was also evaluated retrospectively. The CT ndings supported the surgical planning and ndings (n=30). The surgical plan was changed during the procedure when only plain radiographs had been used to evaluate the patient (n=3). A case which appeared to have unilateral bony ankylosis had to be operated bilaterally as there was brous ankylosis on the other joint that prevented adequate mouth opening which
Dentomaxillofacial Radiology

TMJ ankylosis IE El-Hakim and SA Metwalli

22

graphy underestimated the extent of the bony ankylosis that is found at the time of operation. We support the views expressed by Helms et al. that CT is a valuable diagnostic tool for evaluation of TMJ ankylosis.14 The currently accepted classication of TMJ ankylosis described in the Introduction to the paper4,5,15 does not correlate the ankylotic bony mass to the surrounding vital structures. The need for better understanding of this relationship is the basis for our suggested new classication which is based on postcontrast axial and coronal CT. We propose that ankylosed joints can be grouped according to the relation of the ankylosed mass to the surrounding vital structures, especially at the base of the skull as follows: . Class I: includes unilateral and bilateral brous ankylosis. The condyle and glenoid fossa retain their original shape, and the maxillary artery is in normal anatomical relation to the ankylosed mass. Class II: there is unilateral or bilateral bony fusion between the condyle and the temporal bone. The maxillary artery lies in normal anatomical relation to the ankylosed mass. Class III: the distance between the maxillary artery and the medial pole of the mandibular condyle is less on the ankylosed than in the normal side or the maxillary artery runs within the ankylotic bony mass. This is best seen on coronal CT. Class IV: the ankylosed mass appeared fused to the base of the skull and there is extensive bone formation, especially from the medial aspect of the condyle to the extent that the ankylosed bony mass is in close relationship to the vital structures at the base of the skull such as the pterygoid plates, the carotid and jugular foramina and foramen spinosum and no joint anatomy can be dened from the radiograph. This is best visualized on axial CT.

Figure 5 Axial (a) and coronal (b) CT scans of an ankylosed joint showing an elongated lateral pterygoid plate (P) which appears fused to the bony exostosis

was not visible on the plain lms. The length of the coronoid process and the relation of vital structures to the ankylotic mass was not clear and resulted in injury to the maxillary artery that necessitated blood transfusion. Therefore the surgical procedure took longer in these patients compared with those who had a CT scan. Discussion The clinical and radiographic ndings of this study are in agreement with those of Sawhney.5 long-standing, early-onset ankylosis in childhood results in marked facial asymmetry, whereas the bony changes are minimal when the problem occurs during adolescence or the patient has early treatment. The eective treatment of TMJ ankylosis requires detailed preoperative evaluation of the type and extent of the deformity.8 The information is important for precise surgical treatment planning as adequate resection is necessary to reduce the incidence of recurrence.12 Plain and panoramic radiography did not provide additional information to that already gathered clinically. Sanders et al.13 reported that conventional radioDentomaxillofacial Radiology

This new classication gives the surgeon the opportunity for careful surgical planning and achieves better surgical results with minimum operative complications. We agree with de Bont7 who reported that CT has great potential for imaging intra- and extracapsular hard tissue abnormalities of the joint and facilitated the surgical procedure, whereas we disagree with Posnick and Goldstein16 who considered that it added little diagnostic information to that already gathered from clinical examination and plain lms. Metwalli6 measured the distance between the internal carotid artery, the internal jugular vein, the maxillary artery and the medial pole of the mandibular condyle and found that this distance decreased on the ankylosed side compared with the normal. Concern about the possible risk of damage to any of these structures, compromises the exposure necessary for adequate resection of the ankylotic segment and is often the cause of the subsequent re-ankylosis.17 Our ndings support this view; we recommend that coronal and axial

TMJ ankylosis IE El-Hakim and SA Metwalli

CT are essential in identifying the relationship of these vital structures to the ankylosed bony mass. Our results strongly suggest that TMJ ankylosis aects not only the condylar head, glenoid fossa and

mandibular growth but also aects structures of the skull base such as sphenoid and temporal bone. A separate study is in progress to verify these ndings in detail.

23

References
1. Aggarwal S, Mukhopadhyay S, Berry M, Bhargava S. Bony ankylosis of the temporomandibular joint: a computed tomography study. Oral Surg Oral Med Oral Pathol 1990; 69: 128 132. 2. Silver CM, Motamed M, Carlotti AE. Arthroplasty of the temporomandibular joint with use of vitalium condyle prosthesis. J Oral Surg 1977; 35: 909 913. 3. Obiechina AE, Arotiba JT, Fasola AO. Ankylosis of the temporomandibular joint as a complication of forceps delivery: report of a case. West Afr J Med 1999; 18: 144 146. 4. Kazanjiian VH. Ankylosis of the temporomandibular joint. Surg, Gynecol, Obstet 1938; 67: 333 348. 5. Sawhney CP. Bony ankylosis of the temporomandibular joint: Follow up of 70 patients treated with Arthroplasty and acrylic spacer interposition. Plast Reconst Surg 1986; 77: 29 38. 6. Metwalli S. Computerized tomography of the temporal bone in TMJ ankylosis. MSc Thesis, Cairo University, 1993; pp. 20 36. 7. de Bont LG, van der Kuijl B, Stegenga B, Vencken LM, Boering G. Computed tomography in dierential diagnosis of temporomandibular joint disorders. Int J Oral Maxillofac Surg 1993; 22: 200 209. 8. Kao SY, Chou J, Lo J, Yang J, Chou AP, Joe CJ, et al. Using a three-dimensional-computerized tomography as a diagnostic tool for temporomandibular joint ankylosis: a case report. Chung Hua I Hsueh Tsa Chih (Taipei) 1999; 62: 244 249. 9. Gorgu M, Erdogan B, Akoz T, Kosar U, Dag F. Threedimensional computed tomography in evaluation of ankylosis of the temporomandibular joint. Scand J Plast Reconst Surg 2000; 34: 117 120. 10. Munro IR, Chen YR, Park BY. Simultaneous total correction of temporomandibular ankylosis and facial asymmetry. Plast Reconst Surg 1986; 77: 517 528. 11. Quinn P. Color atlas of temporomandibular joint surgery. St Louis: CV Mosby 1998; pp 112. 12. Rotsko KS. Management of hypomobility and hypermobility disorders of the temporomandibular joint. In: Peterson LJ (ed). Principles of Oral and Maxillofacial Surgery. Philadelphia: JB Lippincott 1992, pp 1989 2014. 13. Sanders R, MacEwen CJ, McCulloch AS. The value of skull radiography in ophthalmology. Acta Radiology 1994; 35: 429 433. 14. Helms CA, Katzberg RW, Manzione JV. Computed tomography. In: Helms CA, Katzberg RW and Dolwick (eds). Internal Derangement of the Temporomandibular Joint. California: Radiology and Research and Education Foundation, 1983, pp 165. 15. Raw NC. Ankylosis of the temporomandibular joint. J Roy Coll Edinb 1982; 67: 96 99. 16. Posnick J, Goldstein JA. Surgical management of TMJ ankylosis in the pediatric population. Plastic and Reconst Surg 1993; 9: 791 798. 17. Behery MG, Helmy ES, Hakam MM. A new utilization of an old technique: Interocclusal block for TMJ ankylosis patients. Egyp Dent J 1992; 38: 251 256.

Dentomaxillofacial Radiology