Effect of voxel size on detection of fenestration, dehiscence and furcation defects using cone-beam computed tomography.

Abstract

This study aimed to assess the effect of voxel size on detection of fenestration, dehiscence, and furcation defects using cone-beam computed tomography (CBCT). This in vitro, experimental study evaluated 4 sheep skulls with both the maxilla and mandible accompanied by the surrounding soft tissue. Fenestration (n = 30), dehiscence (n = 65), and furcation defects (n = 46; 18 grade I, 25 grade II, and 3 grade III) were randomly created by round and needle burs in both jaws, and 40 areas served as control sites. CBCT scans were obtained with 0.300 and 0.150 mm3 voxel sizes and 8 × 11cm2 field of view (FOV), and were randomly observed by four observers (two oral and maxillofacial radiologists and two periodontists). The kappa values, sensitivity and specificity were calculated for each voxel size and compared using paired t test. By an increase in image resolution, diagnostic sensitivity increased while specificity decreased. The kappa values for fenestration (0.602-0.623), and grade III furcation defects (0.903-1.00) were optimal (> 0.6), and almost similar for both voxel sizes. The kappa values for dehiscence, and grades I and II furcation defects were unfavorable (< 0.6) and almost similar for both voxel sizes, except for grade I furcation defects, which had a significant difference in kappa values between the two voxel sizes (0.014 and 0.34). Smaller voxel size had higher sensitivity and lower specificity for detection of all defects except for grade I furcation defects, for which the smaller voxel size had higher sensitivity and higher specificity.