RESUMO
OBJECTIVES: To evaluate the influence of CBCT enhancement filters on the diagnosis of vertical root fractures (VRFs) in teeth with and without metal posts. METHODS: The crowns of 40 uniradicular human teeth were removed and all roots were prepared. 20 teeth were randomly selected, and VRFs were induced using a universal testing machine. The i-CAT (Imaging Sciences International, Hatfield, PA) CBCT was used to scan teeth with and without intracanal metal posts using the following parameters: 0.2 voxel size, 8 × 8-cm scan size and acquisition time of 26.9 s. Images were evaluated by three observers with and without the use of the following filters: S9, smooth, smooth 3 × 3, sharpen, sharpen-mild and sharpen 3 × 3. RESULTS: Intra- and interobserver agreement ranged from poor to moderate. Images with and without CBCT filters did not show significant differences regarding the area under the receiver operating characteristic curve, as well as sensitivity (p > 0.05). As for accuracy, the sharpen-mild filter was superior to the sharpen (p = 0.03), but these filters did not differ from all others. For specificity, S9, smooth and original images were superior to sharpen (p < 0.01). Results for teeth without posts differed from those for teeth with metal posts in all cases (p < 0.05). CONCLUSIONS: The use of enhancement filters in CBCT images has no influence on the diagnosis of VRFs in teeth with metal posts, and their use is not justified.
