Is magnetic resonance imaging or cone beam computed tomography alone adequate for the radiological diagnosis of symptomatic temporomandibular joint osteoarthritis? A retrospective study.

This study was performed to compare the diagnostic accuracy of cone beam computed tomography (CBCT) alone and magnetic resonance imaging (MRI) alone in patients with clinical symptoms of temporomandibular joint osteoarthritis (TMJ-OA). Fifty-two patients (83 joints) with clinical signs of TMJ-OA were included in the study. Two examiners evaluated CBCT and MRI images. McNemar and kappa tests and Spearman's correlation analysis were applied. Radiological findings of TMJ-OA were detected in all 83 joints on CBCT or MRI . Seventy-four joints (89.2%) were positive for degenerative osseous changes on CBCT. MRI findings were positive in 50 joints (60.2%). Osseous changes were found in 22 joints, joint effusion in 30 joints, and disc perforation/degeneration in 11 joints on MRI. CBCT was more sensitive than MRI in detecting condylar erosion (P = 0.001), osteophyte (P = 0.001), and flattening (P = 0.002) and flattening of the articular eminence (P = 0.013) . Poor agreement (κ = -0.21) and weak correlations were found between CBCT and MRI. The study findings suggest that CBCT is superior to MRI in evaluating osseous changes of TMJ-OA, and that CBCT is more sensitive than MRI in detecting condylar erosion, condylar osteophyte, and flattening of the condyle and articular eminence.