Do Ultra-Low Multidetector Computed Tomography Doses and Iterative Reconstruction Techniques Affect Subjective Classification of Bone Type at Dental Implant Sites?

Abstract

To investigate whether ultra-low-dose multidetector computed tomography (MDCT) combined with the reconstruction techniques filtered backprojection (FBP), adaptive statistical iterative reconstruction (ASIR), or model-based iterative reconstruction (MBIR) alters the subjective classification of revised Lekholm and Zarb (LZ) bone types at prospective dental implant sites. Three cadavers underwent a reference MDCT examination using a standard dose volume (CT dose index [CTDIvol]: 29.4 mGy) and reconstructed with FBP in addition to five test protocols (LD1-LD5) using ultra-low doses (CTDIvol: 4.19, 2.64, 0.99, 0.53, and 0.29 mGy, respectively) and reconstructed with FBP, ASIR 50, ASIR 100, and MBIR. Transverse cross-sectional images of the jawbones were obtained, and three examiners subjectively classified the bone type in each image using the revised LZ classification. The bone type classifications obtained using the reference examination were compared with those obtained from the test protocols for each examiner, and kappa statistic was used to analyze the level of agreement between the reference and test protocols. The clinical significance of the differences was analyzed with Wilcoxon signed rank test. Examiners 1 and 2 found moderate to strong agreement between the reference and test protocols, while Examiner 3 found strong to almost perfect agreement (P < .001). The Wilcoxon signed rank test did not demonstrate a clinical significance of the differences between the reference and test protocols for any of the three examiners. MDCT dose reductions of up to 99% did not significantly alter the subjective classification of bone at dental implant sites.