The influence of voxel size and artifact reduction on the detection of vertical root fracture in endodontically treated teeth
Serdar Uysal, Gökçen Akçiçek, Eda Didem Yalçın, Behram Tuncel, Sema Dural
Abstract
OBJECTIVE: The aim of this study was to evaluate the effect of voxel size and artefact reduction (AR) on the identification of vertical root fractures (VRFs) in endodontically treated teeth. METHODS: A total of a hundred sound, extracted human mandibular single-rooted premolars were decoronated, after which root canal preparation was performed, canals were filled with gutta percha by single cone technique. Randomly selected fifty specimens were fractured, repositioned and glued together. The teeth were examined with cone beam computed tomography (CBCT) in five different voxel sizes (0.125, 0.200, 0.250, 0.300, and 0.400 voxels). Two scans were performed for each tooth, one with AR and one without AR. Two radiologists evaluated the CBCT scans. RESULTS: All voxel dimensions were successful in detecting VRFs in CBCT scans. But as the voxel size increased, the percentage of detecting VRFs decreased. High accuracy, sensitivity, specificity and predictive values were found for VRF detection on CBCT scans. Accuracy and sensitivity values decreased (from 100 to 82) while voxel dimensions increased (from 0.125 to 0.400). High-resolution images (0.125, 0.200, and 0.250 voxels) caused an increase in sensitivity for detection of VRFs. AR did not affect the accuracy, sensitivity, specificity and predictive values for VRF detection on CBCT scans. CONCLUSIONS: High-resolution CBCT images resulted in an increase in sensitivity and specificity for detection of VRFs compared with lower-resolution CBCT images. The use of AR did not further improve its diagnostic potential.