LAUSR

OBJECTIVES To compare the efficacy of digital periapical (PA) radiography with cone-beam computed tomography (CBCT) by altering the gray-scale range for the detection of bone-implant interface gaps. METHODS Titanium implants were placed in 52 blocks of bovine rib such that 42 had no gaps and served as the control group. The implants were subsequently placed in the same blocks with 0.25 mm increase in the osteotomy size to serve as the test group. The remaining 10 blocks were used for random arrangement of the test and control blocks within a fabricated wax arch. CBCT with change in the gray-scale range and digital PA were obtained and evaluated by two observers regarding the presence/absence of gaps. The two imaging modalities were compared by the receiver operating characteristic (ROC) curve, Kappa and McNemar tests (α = 0.05). RESULTS PA and CBCT showed moderate agreement for gap detection (k = 0.60). There were no significant differences in the area under the ROC curve (AUC) between CBCT and PA (P = 0.45). The frequency of correct diagnoses on PA radiographs was significantly higher in the test group (P = 0.016), while not significant on the CBCT images (P = 0.344). PA showed higher sensitivity (100%) compared to CBCT (83.33%). However, the specificity of CBCT (92.86%) was greater than that of PA (83.33%). CONCLUSIONS Certain gray-scale ranges in CBCT enhance the assessment of bone-implant interface which brings the accuracy of CBCT closer to digital PA radiography as the modality of choice.