Radiographic evaluation of implant-obscured bone. Comparison of digitally subtracted tomographic and periapical techniques.

This study evaluated the diagnostic accuracy of digitized and substracted periapical and tomographic film images in the assessment of facial and lingual bone loss at implant-obscured sites. An edentulous dried human mandible was prepared with implants in the incisor, premolar, and molar regions. Successive bur sizes from quarter round to no. 6 were used to create circumferential infrabony periodontal defects in the alveolar crest adjacent to the implants. Periapical radiographs were made with the use of custom stents. Linear tomograms imaged sites in a cross-sectional orientation. An equal number of radiographs with and without lesions were digitized and subtracted from baseline images. Radiographic anatomy on the mesial and distal aspects of each implant was electronically masked before the evaluation by a series of eight observers. Receiver operating characteristics analysis and analysis of variance were used to evaluate accuracy of interpretation of the presence or absence of crestal defects. Observer accuracy in detecting facial or lingual bone defects was significantly greater with subtracted tomographic images than with substracted periapical images (p < 0.001).