Influence of the patient's clinical information on the diagnostic reproducibility and accuracy of MRI scans of temporomandibular joint pathologies.

PURPOSE: To evaluate the influence of the patient's clinical information on the accuracy as well as interexaminer and intraexaminer reproducibilities of temporomandibular joint (TMJ) magnetic resonance imaging (MRI) scans.

MATERIALS AND METHODS: Forty MRI scans from 20 TMJs corresponding to 7 TMJ pathologies (ie, degenerative alterations of the condyle, degenerative alterations of the mandibular fossa, alterations in the morphology of the TMJ disc, disc displacement with reduction, disc displacement without reduction, TMJ effusion, and TMJ hypermobility) were assessed by seven uncalibrated specialists in temporomandibular disorders (TMD) at baseline, 30 day-, and 60-day follow-ups for accuracy and reproducibility. No clinical information was provided before the 60-day follow up.

RESULTS: Examiners had a poor to regular accuracy (0.10 to 0.36), kappa index and 5% to 60% correct positive diagnosis) when compared with the radiologist's diagnoses (gold standard). The interexaminer reproducibility ranged from moderate to substantial (kappa = 0.32 to 0.71), and the intraexaminer reproducibility ranged from moderate to perfect (kappa = 0.38 to 1.00). Provision of clinical information improved neither the accuracy nor the reproducibility of the results (P < .05), with the exception of the intraexaminer reproducibility of one examiner.

CONCLUSIONS: Calibration is needed in assessing TMJ MRI scans, even when trained specialists are provided with clinical information from the patient.