This study assessed the use of different voxel resolutions in a cone beam computed tomography (CBCT) unit for the detection of artificially induced periodontal bone defects created using burr, burr and chemicals, and only chemicals.Study Design.
Seven dry skulls were used in this study. In total, 65 dehiscence defects, 43 fenestration defects, and 76 control sites with no periodontal defects were examined. Images were obtained from a CBCT unit (3 D Accuitomo; J Morita Mfg. Corp., Kyoto, Japan), using 3 field-of-view (FOV) sizes (4 × 4 cm; 6 × 6 cm; and 8 × 8 cm) and 4 scan modes (standard, high-definition, high-fidelity, and low-dose).Results.
Overall, κ coefficients for interobserver agreement on burr-induced periodontal defects ranged from 0.336 to 0.795, with the lowest κ value (indicating a fair degree of agreement) obtained for images acquired in standard mode with a voxel size of 0.160 mm3. κ coefficients for the detection of periodontal defects were highest (indicating moderate to high degrees of interobserver agreement) for smaller voxel sizes and high-resolution images. Statistical comparison among groups (burr, burr + chemicals, and chemicals only) was performed using 1-way analysis of variance with post hoc tests.Conclusions.
The CBCT scan mode may affect the diagnosis of periodontal defects. The technique used to create periodontal defects also affected diagnosis. For this kind of experiment, burr-induced or burr + chemical–induced defects should be used, rather than those induced solely using a chemical technique.