Correlations between anatomic and MRI sections of human cadaver temporomandibular joints in the coronal and sagittal planes.

Cadaver material was used in this study to correlate sequential sagittal and coronal T1-weighted magnetic resonance images against anatomic detail. Magnetic resonance imaging (MRI) was found to accurately represent soft tissues in normal and deranged joints. In contrast to previous reports, MRI was found to accurately represent the discal-retrodiscal junction and did not appear to give false positive findings for disc displacement. Magnetic resonance imaging provided good images of bony outline, particularly in coronal views. Difficulties in interpretation arose when different adjoining tissues produced the same MR image; the central tendon of the lateral pterygoid muscle can appear as an extension of the disc, imaging as a distorted and displaced disc. In anatomic sections, a medial hernia sac in the lower joint space was seen as a constant indicator of the medial component of disc displacement; however, this was not evident in sagittal and coronal T1-weighted images. Fibrocartilaginous remodeling of the articular surface projecting into a discal perforation presented the same image as normal discal tissue. Because discs are often thinned over the lateral pole, it is difficult to determine whether discal tissue is present between the articular surfaces when MRI is at its present resolution. Subcortical bone spaces may be misinterpreted as areas of avascular necrosis and osteochondritis dissecans. It is recommended that an imaging sequence of the TMJ include a midcondyle image and lateral, central, and medial sagittal images; however, the lateral sagittal image is the most difficult to interpret.