Periapical radiographs vs cone beam CT imaging for the evaluation of peri-implant bone defects: an ex vivo study.

BACKGROUND: Data on the radiographic interpretation of peri-implantitis is still controversial. Thus, our study aimed to: a) investigate the detectability rate of ex-vivo induced peri-implant bone defects (PBDs) between observers using two different imaging methods; Cone Beam Computed Tomography (CBCT) and Periapical Radiographs (PAs), b) investigate the observers' agreement on their ability to detect PBDs according to their level of expertise and, c) determine the sensitivity and specificity of the imaging methods used to detect induced PBDs. MATERIAL AND METHODS: Two dried human mandibles were used in which ten dental implants were placed and eight PBDs were created simulating clinical conditions. Radiographic examination using PAs and two CBCT modes [CBCT/N (normal/0.3mm3), and CBCT/HR (HiRes/0.15mm3)] was performed at all experimental stages. All PBDs were recorded for their dimensions using a dental periodontal probe as they were used as a gold standard (GS). Finally, 145 images (49 PAs, 48 CBCT/N, and 48 CBCT/HR) were created and evaluated by nine independent observers. Three oral radiologists (OR), three implantologists (IS), and three general practitioners (GP). RESULTS: PBDs were detected at a higher rate by ORs compared to ISs, and GPs. However, the rate of their agreement, did not reach the nominal level of significance (z-test p-value> 0.05), and also between observers of the same expertise, and between the different imaging methods used: CBCT and PAs (z-test p-value> 0.05). In total, the sensitivity of the CBCTs and PAs method was 95% and 80.5%, respectively. While the specificity for all methods was lower, 57%, 62.2% and 50.4% for CBCT/N, CBCT/H and PAs methods, respectively. CONCLUSIONS: Although CBCT performs better than PAs in ex-vivo induced PBDs, further research is needed to evaluate if the present results can be extrapolated to other clinical scenarios and defect configurations.