In vitro diagnostic accuracy of low-dose CBCT for evaluation of peri-implant bone lesions

Objectives To investigate the in vitro diagnostic accuracy of low-dose cone-beam computed tomography (LD-CBCT) for the detection, classification, and measurement of peri-implant bone lesions. Material and methods Titanium dental implants with all-ceramic single crowns (n = 24) were inserted into bovine bone. At twelve implants, four types of peri-implant bone lesions were created. Radiographic imaging was performed using three techniques: LD-CBCT, high-definition CBCT (HD-CBCT), and intraoral radiography (IR). The datasets were presented twice to four observers in a random order. Diagnostic accuracy was measured by calculating sensitivity and specificity, and analyzed using the McNemar's test at a significance level of 0.05. Absolute agreement between the defect sizes was assessed by means of intraclass correlation coefficients (ICC). Results For all three techniques, diagnostic accuracy and reliability for the detection of defects were almost perfect. The following order was found for classification of the different defect types (sensitivity/specificity): HD-CBCT (0.96/0.99) > LD-CBCT (0.93/0.98) > IR (0.71/0.95). No significant difference was found between the two CBCT techniques. Their performance was superior to that of IR. With regard to absolute agreement of defect size, LD-CBCT agreement with HD-CBCT (ICC, 95% confidence interval) was slight for defect depth (0.342, 0.181-0.625) and substantial for defect width (0.911, 0.775-0.963). Conclusion Intraoral radiography is useful for detecting peri-implant bone lesions. LD-CBCT provides additional information regarding the geometry of defects. The even higher diagnostic accuracy of the HD-CBCT protocol tested needs to be carefully weighed against its radiation dose, which is 14 times higher than that of LD-CBCT.