ABSTRACT
OBJECTIVES: The aim of this study was to evaluate the efficacy of different high-resolution cone beam computed tomography protocols in the diagnosis of chemically simulated external root resorption (ERR). STUDY DESIGN: ERR was simulated in 3 sizes by using an acid decalcification protocol in the cervical, middle, and apical thirds of 30 single-rooted premolars. Four scans of each tooth were acquired with different voxel sizes: 0.080, 0.085, 0.120, and 0.133 mm. The images were analyzed by 2 blinded evaluators for the presence of ERR. The evaluation of diagnostic performance was based on receiver operating characteristic (ROC) curve analysis. RESULTS: No statistically significant differences were found in the detection of ERR with regard to lesion sizes or the different root thirds. Areas under the ROC curve were 0.901, 0.892, 0.887, and 0.767 for voxel sizes of 0.080, 0.085, 0.120, and 0.133 mm, respectively. Voxel sizes of 0.080 and 0.085 mm presented significantly larger areas under the ROC curve compared with the voxel size of 0.133 mm. There were no statistically significant differences among the protocols with the use of voxel sizes of 0.080, 0.085, and 0.120 mm. CONCLUSIONS: Cone beam computed tomography examinations with higher-resolution protocols demonstrated better performance in the diagnosis of ERR. Further investigations using simulations that more closely resemble the actual ERR process are warranted.
