Abstract
Objectives
The aim of this retrospective study was to characterize MRI findings of synovial chondromatosis in the temporomandibular joint (TMJ) by correlation with their pathological findings.
Methods
22 patients with synovial chondromatosis in unilateral TMJ were referred for plain MRI prior to surgical management and pathological examinations. Parasagittal and coronal proton density-weighted imaging and T2 weighted imaging were performed for each case.
Results
MRI demonstrated multiple chondroid nodules and joint effusion in all patients (100%) and amorphous iso-intensity signal tissues within expanded joint space and capsule in 19 patients (86.4%). On T2 weighted imaging, signs of low signal nodules within amorphous iso-intensity signal tissues were used to determine the presence of attached cartilaginous nodules in pathology, resulting in 100% sensitivity, 60% specificity and 90.9% accuracy. Signs of low and intermediate signal nodules within joint fluids were used to detect loose cartilaginous nodules and resulted in 80% sensitivity, 42.9% specificity and 68.2% accuracy.
Conclusions
MRI of synovial chondromatosis in TMJ was characterized by multiple chondroid nodules, joint effusion and amorphous iso-intensity signal tissues within the expanded space and capsule. The attached cartilaginous nodules in pathology were better recognized than the loose ones on MRI. Plain MRI was useful for clinical diagnosis of the disorder.
Keywords: synovial chondromatosis, tempromandibular joint, magnetic resonance imaging
Introduction
Synovial chondromatosis is defined by the World Health Organization1 as a benign nodular cartilaginous proliferation arising from the joint synovium, bursae or tendon sheaths. The condition of the primary growth characteristics and locally destructive behaviour suggest a neoplasm.1 Although the temporomandibular joint (TMJ) is infrequently affected by a tumour or tumour-like lesion, synovial chondromatosis is the most common neoplastic lesion of the joint.2 Evaluations of TMJ synovial chondromatosis by using CT and MRI have been frequently reported in the literature.2-14 However, an integrated depiction of MRI features on this condition based on the correlation with pathological findings was not seen in previous reports owing to few patients and variations of MRI sequences and planes.
Pathologically, synovial chondromatosis is characterized by the multiple cartilaginous nodules or bodies formed in the synovial membrane. The multiple cartilaginous bodies or nodules may be either covered (attached) or uncovered (detached) by the synovial lining cells and fibrous tissues.1 A correct differentiation between both cartilage bodies or nodules is helpful in reducing recurrence rate after a synovectomy, especially during the early phase of the disease.1
The purpose of the current study was to characterize MRI findings of TMJ synovial chondromatosis and determine the capabilities of common MRI sequences in the assessment of covered and uncovered cartilage bodies in the disorders.
Materials and methods
From 2003 to 2009, 22 patients with unilateral synovial chondromatosis underwent surgical therapy after MRI in the Ninth People's Hospital, School of Medicine, Shanghai Jiao Tong University, Shanghai, China. All patients were histopathologically confirmed and retrospectively reviewed. The patients included 15 females and 7 males, ranging in age from 17 years to 64 years, with a mean age of 45.3 years (Table 1).
Table 1. Sex, age, location and MRI findings of temporomandibular joint synovial chondromatosis in 22 cases.
| No. of case | Sex/age (years) | Left/right | Superior space | Inferior space | Joint fluids | Low and intermediate signal nodules | Amorphous iso-intensity signal tissues on T2 weighted images | Expanded joint space and capsule | Condyle erosion | Glenoid fossa erosion |
| 1 | F/42 | L | + | − | + | + | + | + | − | − |
| 2 | F/45 | L | + | − | + | + | + | + | − | − |
| 3 | F/48 | L | + | − | + | + | + | + | − | − |
| 4 | M/42 | R | + | − | + | + | + | + | − | − |
| 5 | M/27 | L | + | − | + | + | + | + | − | + |
| 6 | F/42 | L | + | − | + | + | − | − | − | − |
| 7 | F/53 | L | + | − | + | + | + | + | − | − |
| 8 | F/17 | R | + | − | + | + | − | − | − | − |
| 9 | M/38 | R | + | − | + | + | + | + | − | + |
| 10 | F/40 | R | + | − | + | + | + | + | − | − |
| 11 | F/45 | L | + | − | + | + | + | + | − | − |
| 12 | M/63 | L | + | − | + | + | + | + | − | − |
| 13 | F/61 | R | + | − | + | + | − | − | − | − |
| 14 | M/41 | R | + | + | + | + | + | + | + | − |
| 15 | M/64 | L | + | − | + | + | + | + | − | + |
| 16 | F/46 | R | + | − | + | + | + | + | − | − |
| 17 | F/51 | R | + | − | + | + | + | + | − | − |
| 18 | F/48 | L | + | − | + | + | + | + | − | + |
| 19 | M/57 | R | + | − | + | + | + | + | − | + |
| 20 | F/44 | R | − | + | + | + | + | + | + | − |
| 21 | F/45 | R | + | − | + | + | + | + | − | − |
| 22 | F/38 | R | + | − | + | + | + | + | − | − |
F, female; L, left; M, male; No., number; R, right; +, positive; −, negative.
TMJ MRI was performed on a 1.5 T Signa TwinSpeed system (GE Healthcare, Milwaukee, WI) by using the dual TMJ array coil. In the closed mouth position, both oblique saggital and coronal slices were obtained with proton density-weighted imaging (PDWI) fast spin-echo sequence. The parameters of PDWI were as follows: time of repetition (TR)/time of echo (TE), 1360–2000/20–26 ms; 3 acquisitions; field of view (FOV), 10 × 10 cm; bandwidth, 31.2 Hz; matrix, 320 × 192; slice thickness, 2 mm (oblique sagittal) and 1.5 mm (oblique coronal); spacing, 1 mm (oblique sagittal) and 0.5 mm (oblique coronal). In the open mouth position, oblique saggital images were obtained with T2 weighted imaging (T2WI) fast spin-echo sequence. The parameters of T2WI were as follows: TR/TE, 3760–4300/82–90 ms; 3 acquisitions; FOV, 10 × 10 cm; bandwidth, 31.2 Hz; matrix, 320 × 192; slice thickness, 2 mm; spacing, 1 mm.
MR images of 22 patients were independently reviewed by 3 radiologists who were blinded to the pathological reports and the final interpretations of the MRI findings of the synovial chondromatosis in a TMJ were based on a consensus of at least 2 of the radiologists. The described parameters on MRI for each subject comprised (1) internal structural change of synovial chondromatosis, (2) contour change of joint space and capsule and (3) change of glenoid fossa of temporal bone and mandibular condyle. Contralateral TMJ of each subject was used as an internal comparison. The observed items on surgical procedures and pathological findings for each subject included (1) cartilage nodules attached to the synovium and fibrous tissues (synovial metaplasia) and (2) cartilage nodules detached to the synovium and fibrous tissues (loose cartilage bodies). The sensitivities, specificities and accuracy of MRI in determining both cartilage bodies were calculated.
Results
The surgical procedures and pathology confirmed that of 22 patients, 20 (90.9%) lesions originated from the superior space, 1 (4.5%) lesion from the inferior space and 1 (4.5%) from both spaces of the TMJ. There were no extracapsular soft-tissue involvements which were proven by pathology and MRI in all subjects. The sex, age, location and abnormal MRI findings for each case are listed in Table 1.
The abnormal MRI findings of TMJ synovial chondromatosis included: (1) joint effusion (Figures 1–4) in all patients; (2) multiple nodules within the joint space (Figures 1–4) in all patients; (3) amorphous iso-intensive signal tissues to joint fluids and calcified nodules on T2WI within the expanded joint space and capsule (Figures 2–4) in 19 patients (86.4%); (4) erosion of the glenoid fossa of temporal bone (Figure 2) in 5 patients (22.7%); and (5) erosion of the mandiblar condyle (Figure 4) in 2 patients (9.1%).
Figure 1.
Synovial chondromatosis in the right temporomandibular joint (TMJ) (patient 8). Sagittal (a) proton density-weighted imaging in closed mouth position and (b) T2 weighted imaging (T2WI) in open mouth position show the lesion in unexpanded upper joint space of the TMJ. The multiple ovoid chondroid nodules (white arrow) are within high signals of joint effusion on T2WI. (c) Pathological figure with haematoxylin–eosin stain (original magnification ×12.5) shows the loose calcified nodules in the lesion
Figure 4.
Synovial chondromatosis in the right temporomandibular joint (TMJ) (patient 20). (a) Sagittal T2 weighted imaging (T2WI) shows the lesion located in expanded inferior space of the TMJ. The amorphous intermediate signal tissues (white arrowhead), low signal nodules (white arrow) and small amount of high signals effusion are seen on T2WI. (b) Coronal proton density-weighted imaging in closed mouth position shows the expanded inferior space and joint capsule of right TMJ, and erosion of mandibular condyle (white arrow). (c) Pathological figure with haematoxylin–eosin stain (original magnification×100) shows the multiple hyaline cartilage nodules (black arrow) on the surface of the mandibular condyle (white star), which attached to the synovial lining cells
Figure 2.
Synovial chondromatosis in the right temporomandibular joint (TMJ) (patient 9). Sagittal (a) proton density-weighted imaging in closed mouth position and (b) T2 weighted imaging (T2WI) in open mouth position show the lesion in expanded upper joint space (white arrowheads) of the TMJ, accompanied erosion of glenoid fossa of temporal bone. There are low signal nodules within amorphous iso-intensity signal tissues (white arrow) and high signal fluids on T2WI. (c) Pathological figure with haematoxylin–eosin stain (original magnification ×400) shows the multiple hyaline cartilage nodules with local calcification (black arrow) and fibrous connective tissues
Figure 3.
Synovial chondromatosis in the right temporomandibular joint (TMJ) (patient 22). Sagittal (a) proton density-weighted imaging and (b) T2 weighted imaging (T2WI) in closed mouth position show the lesion located in expanded superior space (white arrowheads) of the TMJ. The low signal nodules within amorphous iso-intensity signal structures (white arrow) and high signal joint effusion are seen on T2WI. (c) Pathological figure with haematoxylin–eosin stain (original magnification ×400) shows the multiple hyaline cartilage nodules with lack of calcification and ossification covered by the hyaline fibrous tissues
The multiple chondroid nodules appeared in punctuate form in 20 (90.9%) patients and spherical ring-like forms in 11 (50%) patients. Amorphous iso-intensity signal tissues within the expanded joint space and capsule on T2WI (Figures 2–4) were mainly composed of uncalcified or unossified hyaline cartilage nodules in pathology, which were either covered or uncovered by the synovial tissues and fibrous tissues. 3 out of 22 patients (13.6%) were without the amorphous iso-intensive signal structures and expanded joint space and capsule (Figure 1). Skeleton erosion of the mandibular condyle was only seen in patients with lesions involved in the inferior space of the TMJ (Figure 4), while erosion of the glenoid fossa of the temporal bone was only visible in patients with lesions in the superior space of the TMJ (Figure 2).
On the basis of the relationship among low and iso-intensity signal nodules, joint effusion and iso-intensity signal tissues shown on T2WI, two MRI signs were noticed: low signal nodules within intermediate intensive signal tissues (19 patients, 86.4%, Figures 2–4) and low and iso-intensity signal nodules within high signal joint effusion (16 patients, 72.7%, Figure 1). If both MRI signs represented attached and loose cartilage nodules in pathology respectively, the correlations between MRI and pathological findings for each case are seen in Table 2. The diagnostic values of both MRI signs in determining the correlation with pathological findings are listed in Table 3.
Table 2. The match between pathological findings and MRI signs in temporomandibular joint synovial chondromatosis of 22 cases.
| No. of case | Pathological findings |
MRI findings |
Match between pathological findings and MRI findings | ||
| Chondroid nodules attached to synovium | Loose chondroid nodules | Sign 1 | Sign 2 | ||
| 1 | + | + | + | − | No |
| 2 | + | + | + | + | Yes |
| 3 | + | + | + | + | Yes |
| 4 | − | + | + | + | No |
| 5 | + | + | + | − | No |
| 6 | − | + | − | + | Yes |
| 7 | + | − | + | + | No |
| 8 | − | + | − | + | Yes |
| 9 | + | − | + | + | No |
| 10 | + | + | + | + | Yes |
| 11 | + | + | + | + | Yes |
| 12 | + | − | + | − | Yes |
| 13 | − | + | − | + | Yes |
| 14 | + | + | + | + | Yes |
| 15 | + | + | + | + | Yes |
| 16 | − | + | + | + | No |
| 17 | + | + | + | − | No |
| 18 | + | − | + | + | No |
| 19 | + | − | + | + | No |
| 20 | + | − | + | − | Yes |
| 21 | + | + | + | + | Yes |
| 22 | + | − | + | − | Yes |
No., number; +, positive; −, negative.
Sign 1: low signal nodules within amorphous iso-intensity signal tissues.
Sign 2: low and intermediate signal nodules within joint fluids.
Table 3. MRI value in judging attached and detached condroid nodules in temporomandibular joint synovial chondromatosis of 22 cases.
| MRI findings | True positive | True negative | False positive | False negative | Sensitivity | Specificity | Accuracy | Positive predictive value | Negative predictive value |
| Sign 1 | 17 | 3 | 2 | 0 | 100% | 60% | 90.9% | 89.5% | 100% |
| Sign 2 | 12 | 3 | 4 | 3 | 80% | 42.9% | 68.2% | 75% | 50% |
Sign 1: low signal nodules within amorphous iso-intensity signal tissues.
Sign 2: low and intermediate signal nodules within joint fluids.
Sensitivity = true positive / true positive + false negative.
Specificity = true negative / true negative + false positive.
Accuracy = true positive + true negative / total.
Positive predictive value = true positive / true positive + false positive.
Negative predictive value = true negative / true negative + false negative.
Discussion
Synovial chondromatosis was considered a rare condition of the TMJ in the past. Recently, reports on the disease have been increased owing to the development of new imaging modalities, including CT and MRI, which dramatically improved the capabilities in the diagnosis of bone and joint diseases.2-14 MRI appearances of TMJ synovial chondromatosis included joint effusion, multiple chondroid nodules or bodies, proliferative synovium, expanded joint capsule, anterior displacement of the mandibular condyle and intracranial involvement. However, an appropriate correlation between MRI and pathological findings on the condition was usually seen in case reports.5,7-10 In the current study, the authors not only characterize MRI findings of the disease based on the numerous cases but also compare the MRI findings with surgical and pathological outcomes.
TMJ synovial chondromatosis has frequently occurred in the upper compartment of the TMJ,2,11,13 which could lead to expansion of the joint space or capsule2,4-6, 9, 11 and intrajoint fluid collections.2,4-6, 11, 13 The data from the current study supported these findings. In addition, close relationships between expanded joint capsule and lesion contents, including joint effusion, metaplastic synovium containing attached cartilage nodules and loose cartilage nodules, were noticed. We found that the more effusion, metaplastic synovium and chondroid nodules accumulated in the joint space, the more expansion there was in the joint space and capsule.
On MRI, TMJ synovial chondromatosis is characterized by multiple chondroid nodules with low and iso-intensity signals within the joint space. The low signal intensity nodules might appear as both small round and punctuate forms, which correlated with calcified and ossified ones in pathology. Meng et al13 described some spherical cartilaginous bodies as a “ring-like” form and deemed that this sign is a MRI feature of TMJ synovial chondromatosis. The iso-intensity signal nodules were only shown on T2WI, which often appeared as a punctuate form under the background of high signal joint effusion in our series. Our study indicated that both punctuate and spherical ring-like chondroid nodules might appear in the same lesion. The spherical ring-like nodules only occurred in 50% of the patients in this study and were less frequently visible than punctuate nodules.
Amorphous intermediate signal structures to low signal cartilaginous nodules and high signal joint effusion within the expanded joint spaces or capsules on T2WI were found in most subjects in the current study. In comparison with pathological findings we found that the irregular intermediate signal tissues were mainly composed of uncalcified and unossified nodules, which were attached to and detached from synovial and fibrous tissues. Murphey et al15 considered that nodules with intermediate signal intensity characteristics reflected the high water content of the cartilaginous bodies.
On the basis of the relationship among joint effusion, amorphous iso-intensity signal tissues and low and iso-intensive signal nodules shown on T2WI, two MRI signs were noticed in the current study. One was low signal nodules within the amorphous intermediate signal tissues and the other was low and iso-intensity signal nodules within the high signal joint effusion. Both signs were not emphatically described in the previous literatures. The authors of this study assumed that both MRI signs represented the attached cartilaginous nodules and loose cartilaginous nodules in surgical pathology, respectively. The accuracy of both MRI signs in determining attached cartilaginous nodules and loose cartilaginous nodules in pathology were 90.9% and 68.2%, respectively. The different accuracy between both MRI signs indicated that the capability of MRI in detecting the attached cartilaginous nodules was superior to the loose ones in TMJ synovial chondromatosis. In the current study, we found that a correct match between MRI findings and pathological findings was established in 13 of 22 (59%) patients. The main reasons for mismatch may be that in plain MRI it is difficult to discriminate the amorphous intermediate signal tissues on T2WI as the tissues may be either uncalcified and unossified chondroid nodules or synovial and fibrous tissues.
TMJ synovial chondromatosis with erosion of the glenoid fossa and condyle shown on CT and MRI has been reported.3,12,14 In the current study, signs of destruction of the glenoid fossa were only visible in the patients involving the superior space of the TMJ, whereas signs of condyle erosion were only visible in the patients involving the inferior space of the TMJ. The phenomena were compatible with the descriptions from Ida et al.12 We assume that a corresponding relationship between the lesion site (superior or inferior space of the TMJ) and the location of TMJ bone erosion (glenoid fossa or condyle) may exist. The TMJ synovial chondromatosis with involvement of intracranial structures, such as dura and temporal lobe of the cerebrum, was even rarer owing to self-limiting growth of this disorder. However, this bony erosion may be an important indication that lesions will develop towards the intracranial structures.
In conclusion, MRI features of TMJ synovial chondromatosis mainly include the multiple cartilaginous nodules, joint effusion and amorphous iso-intensity signal tissues in the expanded space and capsule. The attached cartilage nodules in pathology may be better recognized on MRI than the loose cartilage nodules. MRI findings could be used to diagnose and classify approximately two thirds of the patients with TMJ synovial chondromatosis.
References
- 1.Miller MV, King A, Mertens F. Synovial chondromatosis. Fletcher CDM, Unni KK, Mertens F. (eds) World Health Organization classification of tumours. Pathology and genetics of tumours of soft tissue and bone. Lyon: IARC Press, 2002, pp 246 [Google Scholar]
- 2.Heffez LB. Imaging of internal derangements and synovial chondromatosis of the temporomandibular joint. Radiol Clin North Am 1993;31:149–162 [PubMed] [Google Scholar]
- 3.Nokes SR, King PS, Garcia R, Jr, Silbiger ML, Jones JD, 3rd, Castellano ND. Temporomandibular joint chondromatosis with intracranial extension: MR and CT contributions. AJR Am J Roentgenol 1987;148:1173–1174 [DOI] [PubMed] [Google Scholar]
- 4.Westesson PL, Yamamoto M, Sano T, Okano T. Temporomandibular joint. Som PM, Curtin HD. (eds) Head and neck imaging. St. Louis, Mo: Mosby, 2003, pp 1032–1033 [Google Scholar]
- 5.Wong WC, Cheng PW, Chan FL. MRI appearance of synovial chondromatosis in the temporomandibular joint. Clin Radiol 2001;56:773–774 [DOI] [PubMed] [Google Scholar]
- 6.Herzog S, Mafee M. Synovial chondromatosis of the TMJ: MR and CT findings. AJNR Am J Neuroradiol 1990;11:742–725 [PMC free article] [PubMed] [Google Scholar]
- 7.Dolan EA, Vogler JB, Angelillo JC. Synovial chondromatosis of the temporomandibular joint diagnosed by magnetic resonance imaging: report of a case. J Oral Maxillofac Surg 1989;47:411–413 [DOI] [PubMed] [Google Scholar]
- 8.Koyama J, Ito J, Hayashi T, Kobayashi F. Synovial chondromatosis in the temporomandibular joint complicated by displacement and calcification of the articular disk: report of two cases. AJNR Am J Neuroradiol 2001;22:1203–1206 [PMC free article] [PubMed] [Google Scholar]
- 9.Balliu E, Medina V, Vilanova J, Pelaez I, Puig J, Trull JM, et al. Synovial chondromatosis of the temporomandibular joint: CT and MRI findings. Dentomaxillofac Radiol 2007;36:55–58 [DOI] [PubMed] [Google Scholar]
- 10.Yildiz ST, Demir A, Kaya A. Synovial chondromatosis of the temporomandibular joint extending to temporalis, masticator, and parotid spaces. J Comput Assist Tomogr 2001;25:126–129 [DOI] [PubMed] [Google Scholar]
- 11.Kim HG, Park KH, Huh JK, Song YB, Choi HS. Magnetic resonance imaging characteristics of synovial chondromatosis of the temporomandibular joint. J Orofac Pain 2002;16:148–153 [PubMed] [Google Scholar]
- 12.Ida M, Yoshitake H, Okoch K, Tetsumura A, Ohbayashi N, Amagasa T, et al. An investigation of magnetic resonance imaging features in 14 patients with synovial chondromatosis of the temporomandibular joint. Dentomaxillofac Radiol 2008;37:213–219 [DOI] [PubMed] [Google Scholar]
- 13.Meng J, Guo C, Yi B, Zhao Y, Luo H, Ma X. Clinical and radiologic findings of synovial chondromatosis affecting the temporomandibular joint. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2010;109:441–448 [DOI] [PubMed] [Google Scholar]
- 14.Yu Q, Yang J, Wang P, Shi H, Luo J. CT features of synovial chondromatosis in the temporomandibular joint. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2004;97:524–528 [DOI] [PubMed] [Google Scholar]
- 15.Murphey MD, Vidal JA, Fanburg-Smith JC, Gajewski DA. Imaging of synovial chondromatosis with radiologic-pathologic correlation. Radiographics 2007;27:1465–1488 [DOI] [PubMed] [Google Scholar]