Do cone‐beam computed tomography low‐dose protocols affect the evaluation of the temporomandibular joint?

Background There is no established acquisition protocol based on scientific evidence for the acquisition of cone‐beam computed tomography (CBCT) exams to evaluate the temporomandibular joint (TMJ). Objectives To evaluate the influence of acquisition protocols and jaw positioning on the diagnostic accuracy of TMJ condylar morphological alterations (CMA), dimension, position and excursion. Methods Thirty‐six TMJs on 18 dry skulls were imaged using a CBCT unit (OP300 Maxio, Instrumentarium, Tuusula, Finland) at two exposure settings (4.5 and 6.3 mA), three voxel resolutions (0.085, 0.125 and 0.280 mm), three jaw positions (concentric, anteriorised and posteriorised) and three jaw excursions (normoexcursion, hyperexcursion and hypoexcursion). The macroscopic anatomy examination and high‐resolution CBCT images were used as ground truth for CMA. Twenty‐five TMJs had at least one CMA with 11 healthy TMJs serving as controls. Three experienced oral and maxillofacial radiologists evaluated the parasagittal images for the presence of CMA, position and excursion and measured dimensions. The area under the ROC curve, sensitivity and specificity were calculated. Weighted Kappa (α = 0.05) was used to determine intra‐ and interexaminer reliability and comparisons between dependent variables analysed by Analysis of Variance at an a prior level of significance of 0.05. Results The agreement of the evaluation of the position and excursion with the reference standard was high, independent of the protocol (range, 0.75–0.91). Various combinations of acquisition protocols and jaw position did not influence the CMA evaluation. Erosion was overdiagnosed in protocols with larger voxel sizes and the detection of osteophytes greater in images with smaller voxel sizes. The anteroposterior dimension was greater in the open jaw position (p