Diagnostic accuracy of panoramic radiography and MRI for detecting signs of TMJ degenerative joint disease

Abstract

The objective of this study was to determine the diagnostic accuracy of panoramic radiography and magnetic resonance imaging (MRI) for detection of signs of temporomandibular joint (TMJ) degenerative joint disease (DJD). Panoramic radiography and bilateral MRI and computed tomography (CT) of the TMJs were performed for 705 subjects. Three calibrated board-certified radiologists who were blinded to the clinical findings interpreted all images. The diagnoses of DJD established using the panoramic radiographs and MRIs were compared to the reference standard diagnoses derived from the CTs. DJD was defined as the presence of at least 1 of the following 4 signs: a subcortical cyst, surface erosion, osteophyte formation, or generalized sclerosis. The target values for sensitivity and specificity were 70% or greater and 95% or greater, respectively. Compared to the reference standard CTs, the panoramic radiographs had the following sensitivity and specificity values: subcortical cysts, 14% and 100%, respectively; erosion, 20% and 100%, respectively; osteophytes, 12% and 100%, respectively; and sclerosis, 33% and 100%, respectively. The MRIs achieved the following sensitivity and specificity values: subcortical cysts, 32% and 100% respectively; erosion, 35% and 99% respectively; osteophytes, 71% and 98%, respectively; and sclerosis, 50% and 100%, respectively. The radiologists’ interexaminer reliability was slight (κ = 0.16) when using panoramic radiographs, moderate (κ = 0.47) when using MRIs, and substantial when using CTs (κ = 0.71) for diagnosis of signs of DJD. Panoramic radiographs and MRIs had below-target sensitivity but above-target specificity in detecting all CT-depicted signs of DJD with the exception of osteophytes, for which MRIs demonstrated adequate diagnostic accuracy. Use of CT for diagnosis of TMJ DJD is recommended to avoid the false-negative findings that can occur if panoramic radiographs and MRIs are used.

Degenerative joint disease (DJD) is the most common pathologic condition affecting the temporomandibular joint (TMJ).¹ According to Tanaka et al, “The degenerative changes in the TMJ are believed to result from dysfunctional remodeling, due to a decreased host-adaptive capacity of the articulating surfaces and/or functional overloading of the joint that exceeds the normal adaptive capacity.”¹ DJD is characterized by a progressive loss of articular cartilage with increased loading of the subchondral bone. This results in signs and symptoms of focal degeneration and osteophyte formation.² Diagnosis of TMJ DJD is important, as progression of the condition is associated with loss of joint function and occlusal disharmony, including anterior open bite.³

A valid diagnosis of TMJ DJD is based on radiographic findings, because clinical assessment is associated with poor diagnostic accuracy.⁴ Established imaging analysis criteria use the presence of subcortical cysts, surface erosion, osteophytes, and/or generalized sclerosis to confirm a diagnosis of TMJ DJD (Fig 1).⁵ Currently, computed tomography (CT) is considered the reference (gold) standard method for diagnosis of DJD.

Fig 1.

Sagittal (1) and coronal (2) computed tomographic (CT) views of a normal condyle and the 4 radiographic signs of temporomandibular joint degenerative joint disease. A1. Normal condyle; A2. Normal condyle. B1. Surface erosion (arrow). B2. Surface erosion (arrow). C1. Subcortical cyst (arrow). C2. Subcortical cyst (arrow). D1. Osteophyte formation (arrow). D2. Sclerosis of the condylar head. E1. Generalized sclerosis of the condylar head, neck of the condyle, and articular fossa. Osteophyte formation is also present. E2. Generalized sclerosis of the condylar head, neck of the condyle, and articular fossa.

Panoramic radiography is widely used in dentistry, and magnetic resonance imaging (MRI) is commonly used to assess the TMJ. Panoramic radiography is readily available and cost-effective for the general dentist, and it requires a relatively low radiation dose. MRI has the advantage of creating no exposure to ionizing radiation. In addition, MRI provides greater soft tissue detail and is well suited to identify features such as bone marrow changes, joint effusion, and disc position, shape, and integrity.

It has been suggested that both panoramic radiography and MRI are accurate screening instruments to detect gross osseous changes associated with TMJ DJD.⁶ However, compared to CTs, panoramic radiographs used for detection of TMJ DJD were found to have a sensitivity of 26% and specificity of 99%, suggesting that panoramic radiographs fail to reveal approximately three-fourths of DJD detected on CTs.⁵ MRI is primarily used to evaluate the soft tissue structures in the TMJ, since its capacity to assess osseous structures is limited. Previous research has shown that MRIs have reduced sensitivity (59.4%) but excellent specificity (98.0%) for detection of TMJ DJD.⁵ The aim of the present study was to evaluate the diagnostic accuracy of panoramic radiography and MRI in comparison to CT for detecting specific signs associated with TMJ DJD. The signs were subcortical cysts, surface erosion, osteophyte formation, and generalized sclerosis.

Materials and methods

In this multicenter, cross-sectional validation project, participants were consecutively recruited from August 2003 to September 2006 at the University of Minnesota, University of Washington, and University at Buffalo. The institutional review boards of these 3 universities approved the study. Informed consent was obtained for all participants. Guidelines of the Health Insurance Portability and Accountability Act (HIPAA) were followed.⁷

Study population

A total of 1410 joints in 705 subjects were assessed with panoramic radiography, CT, and MRI. The demographics and clinical characteristics of the participants as well as details on study settings and location, recruitment methods, informed consent process, and participant reimbursement have been reported previously.⁴

Image acquisition and interpretation

Panoramic radiography, MRI, and multidetector CT (MDCT) were used in this study. At the University of Minnesota, the Orthophos digital panoramic X-ray machine (Dentsply Sirona), Vision 1.5T and Avanto 1.5T MRI scanners (Siemens), and Sensation 16 MDCT (Siemens) were used. At the University of Washington, the Orthophos panoramic X-ray machine, Signa 1.5T MRI scanner (GE Healthcare Life Sciences), and LightSpeed VCT (GE Healthcare Life Sciences) were used. At the University at Buffalo, the Orthophos 3 panoramic X-ray machine (Dentsply Sirona), Symphony 1.5T MRI scanner (Siemens), and Aquilion CT (Toshiba) were used.⁵ Panoramic radiographs were obtained with proper subject positioning as recommended by the manufacturer. No modifications of the protocols were used at the 3 sites. The criteria used for participant preparation and image acquisition for both CTs and MRIs have been previously reported.⁵

Three board-certified radiologists interpreted the images. The examiners were blinded to the subjects’ clinical histories and clinical diagnoses. Previously reported, specific image analysis criteria were used.⁵

The kappa statistic (κ) was used to measure the diagnostic agreement among the radiologists, taking into account chance agreement. A value of κ = 1 indicates perfect agreement, whereas a value of κ = 0 indicates agreement that is equivalent to chance.⁸ For osseous tissue diagnosis of DJD based on panoramic radiographs, the interexaminer agreement of the radiologists was slight (κ = 0.16).⁹ The reliability of the radiologists in diagnosing hard tissue status was moderate (κ = 0.47) when they were using MRIs and substantial when using CT images (κ= 0.71).⁹ The rates of positive agreement for diagnosing DJD were 19% for panoramic radiographs, 59% for MRIs, and 84% for CTs. The rates of negative agreement were 88% or greater.⁵

Scoring criteria

Evaluation of the osseous components of the TMJs was recorded on a scoring form. Four criteria were used in diagnosing DJD (Table 1). Erosion was evaluated in the condylar head and fossa/eminence, while subcortical cysts, osteophyte formation, and generalized sclerosis were evaluated only in the condylar head. Each scoring factor had a yes/no option.

Table 1.

Diagnostic criteria for TMJ degenerative joint disease.^a

Diagnosis	Definition	Diagram
No DJD	Normal relative size of the condylar head; no subcortical sclerosis or articular surface flattening; and no deformation due to subcortical cyst, surface erosion, osteophyte formation, or generalized sclerosis
DJD	Deformation due to subcortical cyst, surface erosion, osteophyte formation, or generalized sclerosis
	Subcortical cyst: cavity (arrow) below the articular surface that deviates from normal marrow pattern
	Surface erosion: loss of continuity (arrow) of the articular cortex or cortical margin
	Osteophyte formation (arrow): marginal hypertrophy with sclerotic borders and exophytic, angular formation of osseous tissue arising from the surface
	Generalized sclerosis: no clear trabecular orientation and no delineation between the cortical layer and the trabecular bone that extends throughout the condylar head

^aAdapted from Ahmad M, Hollender L, Anderson Q, et al. Research Diagnostic Criteria for Temporomandibular Disorders (RDC/TMD): development of image analysis criteria and examiner reliability for image analysis.

Oral Surg Oral Med Oral Pathol Oral Radiol Endod. 2009;107(6):844–860, with permission from Elsevier.⁵

Statistical analysis

The sensitivity and specificity of the panoramic radiographs and MRIs were estimated, using CT diagnoses as the reference standard, for each of the 4 signs of DJD: subcortical cysts, surface erosion, osteophyte formation, and generalized sclerosis. Sensitivity was the ability of the test images (panoramic radiographs and MRIs) to correctly identify TMJs with signs of DJD when the CT was positive for those signs (true-positive rate). Specificity was the ability of the test images (panoramic radiographs and MRIs) to correctly identify TMJs without signs “of DJD when the CT was negative for those signs (true-negative rate). Sensitivity and specificity values vary between 0 and 1 (ie, 100.0%). The 95% confidence interval for each of these measures was calculated. The target values for sensitivity and specificity were 70% or greater and 95% or greater, respectively.¹⁰

Results

Panoramic radiographs, MRIs, and CTs from 1410 joints in 705 subjects were evaluated for surface erosion in the condylar head and eminence. Some data were missing for the other individual variables (subcortical cysts, osteophyte formation, and generalized sclerosis), so a range of 1396–1408 joints were evaluated for those categories. The subjects included 579 females (82.1%) and 126 males (17.9%). The demographic details of the study subjects have been reported previously.⁴

Panoramic radiographs showed poor sensitivity but excellent specificity for detection of the 4 signs of DJD compared to the reference standard, the CT (Table 2). The MRIs were superior to panoramic radiographs in detecting osseous changes associated with DJD. The MRIs had poor sensitivity for subcortical cysts (32.1%) and erosion (35.9%), marginal sensitivity (50.0%) for generalized sclerosis, and excellent (70.7%) sensitivity for osteophyte formation.⁵ The specificity was excellent for all 4 signs of DJD when MRIs were viewed (Table 3). Figure 2 illustrates the findings for DJD on the panoramic radiograph, MRI, and CT views of a single subject.

Table 2.

Diagnostic accuracy of panoramic radiography compared to CT for DJD.

Signs of DJD	Sensitivity (%)	95% CI	Specificity (%)	95% CI
Subcortical cysts (n = 56)	14.3	5.5–32.4	99.7	98.8–99.9
Surface erosion (n = 256)	19.5	13.6–27.3	99.7	98.6–99.9
Osteophyte formation (n = 182)	12.1	6.8–20.5	99.8	98.8–100.0
Generalized sclerosis (n = 24)	33.3	13.1–62.4	100.0	NA

Abbreviations: CT, computed tomography; DJD, degenerative joint disease; NA, not applicable.

Table 3.

Diagnostic accuracy of MRI compared to CT for DJD.

Signs of DJD	Sensitivity (%)	95% CI	Specificity (%)	95% CI
Subcortical cysts (n = 56)	32.1	17.6–51.1	99.9	99.0–100.0
Surface erosion (n = 256)	35.9	28.1–44.6	99.0	97.7–99.5
Osteophyte formation (n = 184)	70.7	60.6–79.0	97.9	96.4–98.8
Generalized sclerosis (n = 24)	50.0	24.4–75.6	99.7	98.9–99.9

Abbreviations: CT, computed tomography; DJD, degenerative joint disease; MRI, magnetic resonance imaging.

Fig 2.

Osseous changes in a study subject. A. Panoramic radiograph showing poor visualization of the right and left condylar heads. B. Sagittal CT view of the left condyle, showing osteophyte formation (white arrow), a subcortical cyst (open arrow), and surface erosion (white arrowhead). Sclerosis of the fossa (black arrow) and eminence (black arrowhead) is also visible. C. Axially corrected coronal CT of the left condyle showing surface erosion (white arrowhead) and a subcortical cyst (open arrow). Sclerosis of the fossa (black arrow) is also present. D. Sagittal proton density-weighted magnetic resonance image (MRI), taken in the closed-mouth position, showing osteophyte formation (white arrow) and a subcortical cyst (open arrow). E. Sagittal proton density-weighted MRI, taken in the open-mouth position, showing osteophyte formation (arrow) of the condylar head.

Discussion

The results of this study indicated that panoramic radiographs and MRIs of the TMJ have excellent specificity but inadequate sensitivity for the detection of subcortical cysts, surface erosion, and generalized sclerosis. For osteophytes, panoramic radiographs also have poor sensitivity, but MRIs have excellent sensitivity. The implication is that positive findings of signs of DJD on a panoramic radiograph or MRI are definitive, while a negative finding on a panoramic radiograph or MRI does not rule out the possibility that the patient has a sign of TMJ DJD that was not detected with these imaging modalities.

Typically, screening tests for detection of nonmorbid targeted diseases have high sensitivity and low specificity. Therefore, panoramic radiography and MRI are not suited for screening for DJD. For TMD, it has been recommended that sensitivity be 70% or greater and specificity be 95% or greater.^10,11 This criterion for diagnostic accuracy is based on the objective of avoiding overdiagnosis of these disorders. Given this criterion, MRIs did provide acceptable diagnostic accuracy for detection of osteophytes.

Beyond its potential to affect jaw pain and function, TMJ DJD can cause malocclusions. It has been previously reported that TMJ DJD is associated with the development of skeletal anterior open bite, overjet greater than 6 mm, and retruded contact position–intercuspal position slides greater than 4 mm.¹² These malocclusions can compromise chewing ability and cause esthetic changes, thus impacting the quality of life of an individual.

Dental restorative procedures are dependent on a stable maxillomandibular position and occlusion, and TMJ DJD can change these relationships. Although patients can have TMJ DJD without occlusal or skeletal changes, there is a risk that dental interventions can alter this stability, resulting in occlusal changes to relationships that previously were compensated for by proprioceptive input from the patient’s teeth. Therefore, clinicians have an interest in detection of TMJ DJD, especially if extensive dental intervention is being recommended, since the disease has a vital influence on both treatment approach and prognosis. Diagnosis of TMJ DJD is also important from a medicolegal perspective, as the patient with TMJ DJD must be informed about the potential instability of the occlusion and the guarded prognosis for results of certain dental procedures.

The results of the present study are consistent with those of prior reports in the literature, which found low sensitivity and high specificity when clinicians are evaluating the signs of TMJ DJD.^13–15 An earlier study showed similar results when panoramic radiography was compared with tomography.¹³ Although the variables were slightly different, when Dahlström & Lindvall evaluated for the presence of condylar flattening and osteophyte formation, they found that panoramic radiographs had high specificity for the absence of osteophytes (0.90) and condylar flattening (0.85) while sensitivity was low (0.29 and 0.33, respectively).¹³ Another study compared panoramic examination, sagittal (lateral) scanography, and tomography for detection of TMJ condylar flattening, defects (defined as local areas of rarefaction), and osteophytes.¹⁴ Mean sensitivity values ranged from 0.10–0.50, while specificity values were high, ranging from 0.86–0.99. The researchers found no significant difference in diagnostic accuracy between the methods and favored the use of panoramic examination, which is simpler to undertake and results in less radiation burden than sagittal cross-sectional tomography.¹⁴ Another study evaluated the efficacy of panoramic radiography in diagnosing TMJ DJD and found it had a low diagnostic value.¹⁵

A systematic review evaluated the role of different imaging modalities in diagnosing TMJ erosion and osteophytes and concluded that only extensive erosion and large osteophytes in the TMJ can be identified with panoramic imaging.¹⁶ This conclusion is in agreement with current guidelines described in a position paper by the American Academy of Oral and Maxillofacial Radiology (AAOMR) for the use of panoramic radiography.¹⁷ The results in the present study do not support the position paper of the AAOMR. However, this position paper was issued in 1997, prior to the introduction of cone beam CT (CBCT). Currently, the AAOMR is developing new guidelines (Mansur Ahmad, BDS, PhD, personal oral communication, June 2016).

Some of the discrepancy between the AAOMR’s guidelines and the present results also may be explained by the imaging technique. The positioning of the patient is a critical factor in the quality of panoramic radiographs. If the patient’s head is inclined posteriorly, the condyle is flattened and may mimic the appearance of an osteophyte. If the patient’s head is inclined anteriorly, the condyle may look sclerotic.¹⁸ Fallon et al evaluated panoramic imaging of the TMJ using cadaveric skulls.¹⁹ They found that it was not possible to accurately determine condylar morphology because of the radiographic variations produced by differences in condylar angulation. A disadvantage of panoramic radiography is that the glenoid fossa and articular eminences are not well visualized because of the superimposition of the base of the skull and zygomatic arches. It would be beneficial to study the diagnostic accuracy of panoramic radiographs taken with the patient’s jaw positioned anteriorly to see if this improved visualization of the TMJ.

The location of osteophytes has also been cited as a possible cause for the limited diagnostic capabilities of panoramic radiography. Osteophytes usually form on the anterior surface of the condyle; in images taken in the coronal plane, osteophytes will be superimposed on the condylar head and hidden. Lesions in the central and medial locations are more accurately detected than those in lateral locations.²⁰

MRI of the TMJ is intended to visualize soft tissue components, such as disc position and effusions. Its capacity to identify osseous structures is limited.²¹ However, the results of the present study indicated that it is a much better tool for diagnosing DJD, especially osteophyte formation, than panoramic radiography. Early degenerative changes in the TMJ, such as erosions, are not well detected on MRIs. Osteophyte formation represents a late change created to spread the load over a greater surface area and appear radiographically as marginal hypertrophic bone formation.¹⁶

In the present study population, the prevalence rates for both erosion and osteophyte formation were close to 30%. In an earlier study, Schmitter et al used gadolinium-enhanced MRI to evaluate DJD in a sample of older individuals (73–75 years) who were virtually asymptomatic and found a prevalence of 70%.²² They concluded that DJD was detected at high rates because a contrast agent was used, which allowed the MRIs to display fine details.²² The researchers also suggested that signs of TMJ DJD are related to aging. The present study was carried out without the use of a contrast agent, and it is possible that minor alterations in form or structure may have been missed. If this is true, contrast-enhanced MRI has the potential to become a viable diagnostic tool for the diagnosis of DJD, since it may more accurately depict both hard and soft tissues.

With MRI, information about disc position, joint fluid, bone marrow changes, and bone structure at multiple levels of the joint can be obtained without exposing the patient to ionizing radiation. However, MRI is expensive compared to panoramic radiography and CT. MRI is also contraindicated for many patients, including those with pacemakers, intracranial vascular clips, and metal particles in vital structures. Other factors that may limit the use of MRI include the patient’s obesity, claustrophobia, or inability to limit motion during the examination.²³

The present study used MDCT scanners. Given that CT is the gold standard for the diagnosis of TMJ DJD, CBCT provides a viable alternative to conventional CT, since they have similar diagnostic accuracy.^24,25 CBCT requires a shorter scanning time and lower radiation doses than MDCT while producing images of high diagnostic quality. Thus, when a definitive diagnosis of TMJ DJD is needed, CBCT is an excellent imaging technique to use. Larheim et al examined the use of CBCT for evaluation of TMJ involvement in different conditions, including DJD.²⁶ They concluded that CBCT was comparable to MDCT for evaluation of cortical bone. However, they found that CBCT is more sensitive than MDCT to patient movements, making the diagnosis of small cortical abnormalities uncertain.²⁶

A strength of the present study is the large sample of participants whose demographic characteristics are comparable to those of the general population.¹¹ The 3 radiologists interpreting these images were calibrated and blinded to the clinical histories and clinical diagnoses of the study participants. Additionally, 4 different calibration and reliability studies for the radiologists were undertaken over the course of this study.⁵ Statistical analyses for this study used sensitivity and specificity estimates, which theoretically are independent of the prevalence of the target condition, in this case TMJ DJD.²⁷ A literature search did not reveal any large population-based studies on the prevalence of TMJ DJD.

A possible limitation of this study could be the exclusion of articular surface flattening and subcortical sclerosis as signs of TMJ DJD. These 2 signs were designated as “indeterminate” for TMJ DJD.⁵ Flattening can be present in normal joints and a variant of normal morphology. Both signs can be present due to aging. These 2 signs may also indicate remodeling of the TMJ and as such are not specific for the presence of TMJ DJD. Finally, they may be a precursor to development of TMJ DJD. To address these issues, longitudinal follow-up of study subjects would be needed.

The diagnostic criteria established for radiologic interpretation of TMJ DJD, including the designation of indeterminate for these 2 signs, had content validity, since the criteria were developed after a review of the literature and based on recommendations by the members of an external advisory panel appointed by the National Institute of Dental and Craniofacial Research for this study and suggestions from members of the TMD and radiology communities.⁵

Conclusion

The present study assessed the diagnostic accuracy of panoramic radiography and MRI of the TMJ for identification of subcortical cysts, surface erosion, osteophyte formation, and generalized sclerosis, which are signs of TMJ DJD. Images obtained with MDCT were used as the reference standard for diagnosis. The results indicated that panoramic radiography is not suitable for definitive diagnosis of these 4 signs of DJD because of inadequate sensitivity. MRI of the TMJ is much better than panoramic radiography for identifying the 4 signs of DJD, especially osteophyte formation. However, CT is needed for accurate diagnosis of DJD. In the future, CBCT may well replace MDCT for this purpose, because CBCT clearly depicts the osseous structures of the TMJ at a lower radiation dose.

Clinicians may need to consider CT, including CBCT, as the imaging method of choice when assessing patients for signs of TMJ DJD. An advantage of CBCT is that a panoramic view is also obtainable. Conventional panoramic radiographs are still useful if clinicians need to rule out odontogenic or nonodontogenic causes of orofacial pain.