|Magnetic resonance imaging of the lateral pterygoid muscle in temporomandibular disorders|
|Prev||Chapter 2. Review of literature||Next|
Most studies agree that TMD cannot be diagnosed only on the basis of findings by clinical examinations. Imaging diagnosis is very important (Wilkes1989a,b, Westesson et al. 1989, Gibbs et al. 1998). There are a number of imaging techniques, ranging from plain-film to MRI, that have been used in diagnosis of TMD. Arthroscopy is applied as a treatment for TMD rather than a diagnosis technique (Barkin & Weinberg 2000, Holmlund et al. 2001). Nuclear medicine imaging is rarely used for clinical diagnosis of TMD (Matin 1983). These two methods are therefore not included in this review.
Plain films of TMJ are made with a stationary x-ray source and film. In order to avoid superimposition of adjacent anatomic bony structures making visualization of all parts of TMJ, different projections of transcranial films have been applied, which include oblique transcranial view, transmaxillary view, submental-vertex view and transpharyngeal view (DelBalso 1990, Kopp & Rockler 1979, Kononen & Kilpinen 1990, Williamson et al. 1999). Transcranial plain films have been used in the past to diagnose TMD by evaluating the position of the mandibular condyle in the glenoid fossa and condylar osseous changes (Weinberg 1979, van Sickels et al. 1983, Kononen & Kilpinen 1990). However, some studies have reported that the relationship between the condyle and the glenoid fossa could not be accurately evaluated in transcranial film by comparing with tomography of TMJ (Pullinger et al.1985, Knoernschild et al. 1991). Concordance in the degree of condylar displacement was found in only 60% of the cases when comparing transcranial film and tomography (Pullinger & Hollender 1985). Currently, transcranial plain film is suggested to be a method for detecting condylar fractures rather than a method for diagnosing TMD (Lindqvist et al. 1988, Brooks et al. 1997, Thoren et al. 2001).
Panoramic radiography used to be considered a good imaging method for evaluating TMJ since information about the teeth and other parts of the jaws were also shown on the image (Kononen & Kilpinen 1990, Howard 1990). However, the relationship between the condyle and glenoid fossa cannot be evaluated in the panoramic film because the fossa cannot be seen with superimposition of the base of the skull and zygomatic arch. The morphology of the condyle becomes wider than the anatomic structure of the condyle (DelBalso 1990). The agreement between panoramic radiographs and lateral tomograms on osseous changes of the condyle is 60% to 70% (Habets et al. 1989, Ludlow et al. 1995). Panoramic radiography has also been used in evaluating condyle fractures (Silvennoinen et al. 1998).
Tomography of TMJ is generated through the synchronous movement of the x-ray tube and film cassette through an imaginary fulcrum located in the center of the desired imaging plane. Linear tomography and complex tomography are involved (DelBalso 1990).
Osseous changes in TMJ. Several studies have attested that tomography is a good method for depicting the osseous changes with arthrosis in TMJ (Berrett 1983, Rosenberg & Graczyk 1986). In studies of TMJ specimens obtained at autopsy, tomography has been shown to represent the anatomic structures better than transcranial radiography (Lindvall et al. 1976, Eckerdal & Lundberg 1979). Observer performance studies in the assessment of tomographic images have shown a 60% to 80% interobserver agreement and a similar intra-observer agreement (Kopp & Rockler 1978, Petersson & Rohlin 1988, Rohlin & Peterson 1989, Cholitgul et al. 1990).
Condyle position. For evaluation of condyle position in glenoid fossa of TMJ, tomography has been reported to be more reliable than plain film and panoramic radiography in a study comparing the three methods (Ludlow et al. 1995). Tomography has been used for evaluating the condyle position and joint space in many reports (Knoernschild et al. 1991, Kuboki et al. 1999, Ozawa et al. 1999). However, numerous studies have shown that condyle position appears to be highly variable in both asymptomatic (Pullinger et al. 1985, Blaschke & Blaschke 1981) and symptomatic subjects (Brand et al. 1989, Pullinger et al. 1986). On the other hand, the relationship between the condyle position and disc displacement is uncertain. The condyle position is not reliable in estimating the disc displacement of TMJ and related symptoms (Katzberg et al. 1983, Bonilla-Aragon et al. 1999, Kurita et al. 2001). Clinically, condyle position is still an important aspect in orthognathic surgery (Williamson et al. 1999, Arat et al. 2001) and orthodontic studies (Major et al. 1997, Zhou et al. 1999). The major disadvantage of tomography is the lack of visualization of the soft tissue of TMJ, a problem shared with plain film radiography.
The technique of TMJ arthrography was introduced in the 1940s (Nørgaard 1947), but it was not extensively used until the late 1970s (Westesson et al. 1980, Westesson 1993). Arthrography improved rapidly after the late 1970s (Wilkes 1978, Farrar & McCarty 1979, Katzberg et al. 1979). There are two technical methods for arthrography of TMJ. In single-contrast arthrography, radiopaque material is injected into either the lower or upper joint space, or into both compartments (Wilkes 1978a,b, Brand 1990). In double-contrast arthrography, a small amount of air is injected into the joint space after the injection of contrast materials (Westesson et al. 1980, Westesson 1982, Ma 1987). A comparative study reported that there was no statistically significant difference in the diagnostic accuracy between these two techniques (Westesson & Bronstein 1987).
Several studies have shown that arthrography is an accurate imaging method for evaluating anterior disc displacement. The accuracy for diagnosing the position of the disc ranged from 84% to 100% compared with the corresponding cryosectional morphology and surgical findings (Westesson & Rohlin 1984b, Westesson et al. 1986, Tanimoto et al. 1989, Schellhas et al. 1988). Perforation and adhesion of the disc can also be shown by this technique (Helms et al. 1980, Westesson et al. 1986, Schellhas et al. 1988, Ryan et al. 1990). These studies have given important evidence for diagnosis and identification of TMJ internal derangement (Wilkes 1989). Arthrography is based on plain film (Helms et al. 1980, Van Sickels et al. 1983) and tomography (Wilkes 1978a,b, Katzberg et al 1979). A recent study reported that using the arthrography technique might improve the accuracy of diagnosing perforations and adhesions of the disc in magnetic resonance imaging of TMJ (Toyama et al. 2000).
There are some advantages of this technique. Arthrography is a method that depends upon more technical training and experience in the observation of images, it also has minimally invasive (Westesson et al. 1993).
In the 1980s, computed tomography (CT) began to be applied in the diagnosis of TMJ ankylosis (Kaban & Bertolami 1981), condyle fracture (Avrahami et al. 1984, Sahm & Witt 1989), disc displacement (Helms et al. 1982, Thompson et al. 1984) and osseous changes (Larheim & Kolbenstvedt 1984, Katzberg et al. 1984, Cohen et al. 1985).
In an earlier report, the accuracy for disc displacement was high (81%) when comparing imaging observations of CT and surgical findings (Thompson et al. 1984). Some reports considered that CT might replace the technically difficult and invasive arthrography in the diagnosis of disc displacement in TMD (Helms et al. 1982, Avrahami et al. 1984, Cohen et al. 1985). However, the accuracy of the disc displacement was only 40%-67% in CT in studies of autopsy specimen materials (Westesson et al. 1987b, Tanimoto et al. 1990). The accuracy of osseous changes of TMJ in CT compared with cadaver material was 66%-87% (Tanimoto et al. 1990, Westesson et al. 1987a). Some reports pointed out that radiographic evidence of arthrosis may or may not be associated with clinical symptoms of pain dysfunction. Thus patients without osseous changes in TMJ may have pain, and those with clear signs of bony abnormalities may be pain-free (Westesson 1993, Brooks et al. 1992, Bertram et al. 2001). CT was not considered as a good method for the diagnosis of TMD in many later reports (Katzberg 1989, Kaplan & Helms 1989, Helms & Kaplan 1990, Westesson 1993, Brooks et al. 1997).
Magnetic resonance imaging (MRI) is unique in that there is no associated risk of ionizing x-ray. For MRI, the patient placed in a strong static magnetic field. The hydrogen nuclei, or protons, in the body align with the direction of the main magnetic field, a short radiofrequency (RF) pulse at the proper frequency and duration is then transmitted into the body. The protons absorb RF energy and flip over into a plane that is at an angle with the direction of the main magnetic field, the protons reemit some of the absorbed energy, which induces an electric current in a specially designed RF receiver coil. The induced current, so-called the magnetic resonance (MR) signal, is then transformed into an image by computerized mathematical methods (Jarenwattananon & Gentry 1990).
An MR image is produced from signals coming from the hydrogen nuclei, or protons, in the body. The contrast of the image is provided by differences in signal intensity from protons in different tissues. Several parameters affect the signals intensity: (1) amount of the hydrogen nuclei in tissue; (2) the characteristics of the tissue, as determined by two different relaxation time constants (T1 and T2); (3) the bulk flow of protons in tissues. Varying the pulse sequence imaging variables, such as pulse repetition time (TR) or echo delay time (TE), allows discrimination of different tissues. Special selected parameters can generate images in which contrast differences are dependent on these factors. The most frequently used ones in TMJ images are T1-weighted image (short TR and TE), T2-weighted image (long TR and TE) and proton-density (PD) image (long TR and short TE) (Jarenwattananon & Gentry 1990). Typical values for T2 in tissue range from 0.02 to 0.30 seconds. In general, the more water a tissue contains, the longer the T2. Thus, areas of long T2 can be interpreted as areas of edema, effusion, or inflammation. Typical T1 values for tissue range from 0.2 to 3 seconds. Much of the power of MRI comes from the fact that various tissues have different values of T1 and T2 and contrast can be varied over a wide range by adjusting TE and TR (Kircos & Ortendahl 1990). For example, in T1-weighted and PD image, the fibrocartilaginous disc is in low signal intensity. The lateral pterygoid muscle (LPM) and bilaminar zone are of intermediate signal intensity. The position and morphology of the disc can be clearly observed (Katzberg et al. 1985). However, in a T2-weighted image, the disc, muscles and bilaminar zone are all in low signal intensity, effusion (if it exists) of the bilaminar zone or in joint compartments can be well observed with high-intensity signal (Schellhas & Wilkes 1989). Skeletal muscles, fibrous, blood and fat tissues all have their own characteristic signal intensities in MRI (Siegelman & Outwater 1999, Weinreb et al. 1985, Murphy et al. 1986).
MRI with surface coil was introduced applied to TMJ imaging in the 1980s (Harms et al. 1985, Katzberg et al. 1986, Schellhas et al. 1986). Several studies have compared MRI of TMJ with arthrography and CT (Schellhas et al. 1988, Liedberg et al. 1996, Trumpy et al. 1997). The MRI findings were also compared with anatomical and histological observations (Westesson et al. 1987a,b, Larheim et al. 1999). In studies on autopsy specimens, the accuracy of MRI in evaluating osseous changes in TMJ was 60% to100% (Westesson et al. 1987b, Hansson 1989, Tasaki & Westesson 1993) and the accuracy in evaluating disc displacement was 73% to 95% (Westesson et al. 1987b, Hansson 1989, Schwaighofer et al. 1990, Tasaki & Westesson 1993). All these studies showed that MRI was the best method of imaging both the hard and soft tissues of the TMJ (Raustia et al. 1995, Brooks et al. 1997, Gibbs et al. 1998).
Several studies have confirmed that disc displacement in MRI showed close associations with clicking, pain and other dysfunction symptoms of TMJ (Katzberg et al. 1996, Rammelsberg et al. 1997, Emshoff et al. 2001, Bertram et al. 2001). MRI was considered as a golden standard to evaluate the disc position (Gibbs et al. 1998). Whenever the clinical pain and dysfunction symptoms of TMJ were found to have no relationship with disc displacement in MRI, a false-positive or false-negative imaging diagnosis was suspected (Haley et al. 2001).
Although many studies agree that muscular pain is another major aspect of TMD (Solberg 1986, Greene & Laskin 1988, Rauhala et al. 1999, Molina et al. 2000, Greene 2001, De Laat 2001), the evidence of pathological changes of the masticatory muscles may have been ignored in imaging diagnosis. The results of some reports have shown that MRI is not only an accurate method to detect the disc position but also a potential technique to evaluate the pathological changes of the masticatory muscles in TMD (Katzberg et al. 1985, Schellhas 1989b, van Spronsen et al. 1989, Westesson 1993, Quemar et al. 1993). However, no reports concerning the relationship between the abnormalities of the masticatory muscles in MRI and clinical symptoms have been published.