Reliability and accuracy of cephalometric analysis in cone-beam computed tomography. A review

Aim:The aim of this study is to evaluate the reproducibility and reliability of cephalometric landmarks with threedimensional (3D) cone-beam computed tomography (CBCT) in comparison with conventional twodimensional (2D) cephalometric radiography. Methods:The research was made on Pubmed using these keywords: cefalometric AND landmarks AND cbct AND 3D AND reliability. The inclusion criterias used were: - methodological reliability and reproducibility studies - studies using CBCT - articles with a sample size at least of 12 cases - studies in human or dry skulls - studies including both skeletal and dental landmarks The exclusion criterias used were: - studies using only conventional radiographies or 2D images generated from CBCT - studies in animals - case reports and author opinion articles - studies with a number of observers inferior to 3 Results: Electronic database searches provided 21 articles; excluding repetition the number resulting was 18, of which a fnal total of 8 titles satisfed the selection criteria and were included in this review. Gribel et al. (Angle Orthod. 2011) in this article investigated the accuracy of craniometric measurements made on lateral cephalograms versus CBCT images. The conclusion of their study is that CBCT craniometric measurements are accurate and can be used for craniofacial analysis. In addition, lateral cephalograms have some limitations due to distorted images. De Oliveira et al. (Oral Surg Oral Med Oral Pathol Oral Radiol Endod. 2009) evaluated reliability in 3D landmark identifcation using CBCT between intra-observer and inter-observer. They found that following a protocol for operator training and calibration offers consistent and reproducible data. Lisboa Cde O. et al. (J Appl Oral Sci. 2015) in this systematic review found an higher reliability of the dental landmarks and the median sagittal line, while the landmarks of the curve surfaces such as condyle, porion and the orbitale presented a lower level of faithfulness. The author found important variations according to the parameters of image acquisition, software, types of visualization and in the anatomic references marked. In their research, Neiva MB. et al. (Angle Orthod. 2015) investigated the reliability of landmarks identifcation using two different visualization techniques: 3D reconstruction and multiplanar views. The author found a better reproducibility of cephalometric landmarks using CBCT in multiplanar views than in 3D image reconstructions. Furthermore, the points with lower reliability were the condyle, ramus point and sella turcica, while the most reliable are the ones found on the midsagittal plane. Conclusion: 2D conventional images may have some issues with distortion and magnifcation that can lead to cephalometric measurements inaccuracy. Cephalometric landmarks and measurements on 3D CBCT are reliable and can possibly be used as a quantitative orthodontic diagnostic tool. The midsagittal plane and the dental landmarks demonstrated the highest reliability. Landmarks with the lowest reliability included those marked on the condyle and other anatomic structures with prominent curvatures.