Title Dental Measurements from CBCT-Generated Models Are Not as Accurate as Those from Intraoral Scanners
Clinical Question To implement CBCT in orthodontic practice, could dental measurements from CBCT models be just as accurate as the measurements from intraoral scanned models for diagnosis and treatment planning?
Clinical Bottom Line Dental measurements from CBCT-generated models are not as accurate as those from intraoral scanners; however, they are clinically acceptable to obtain diagnostic information when treatment planning. This is supported by controlled studies that evaluated measurements of tooth size and Bolton’s index for models that were generated by CBCT and intraoral scanner compared to measurements from plaster models. The first study showed that there were a few significant differences in the linear measurements made from CBCT-generated models whereas intraoral laser scanned models did not. The second study also showed that the CBCT models underestimated dental measurements whereas intraoral laser scanned models did not. Although there were significant differences in both studies for measurements made in CBCT models, they were still considered clinically acceptable to obtain diagnostic information for patients.
Best Evidence  
PubMed ID Author / Year Patient Group Study type
(level of evidence)
23273370Wiranto / 201222 volunteers (employees from Department of Orthodontics and Students from Department of Dentistry in The NetherlandsGold standard-controlled diagnostic study
Key resultsThe differences between the tooth-width measurements of DigiModels (digital models from CBCT scan) compared with the plaster models ranged from -0.04 to 0.16 mm whereas the differences from the Lava models (digital models from intraoral scan) to plaster models ranged from -0.24 to 0.07 mm. Although there were a few significant differences from DigiModels, they were not larger than 0.2 mm. For the Lava models, there were no significant differences compared to the plaster models. Furthermore, the anterior and overall Bolton ratios for the digital models were smaller than the plaster models and were statistically significant (P<0.05); however, the measurements never exceeded 1.5 mm and are considered clinically insignificant.
29410764San Jose / 2017100 patients from Orthodontic Department in Spain who were treated from January 2014 and April 2015Gold standard-controlled diagnostic study
Key resultsThe mean differences between direct measurements (tooth size, intercanine distance, intermolar distance, and arch length) of 3D models from intraoral laser scanner (ILS) are not statistically significant compared to the 2D digital methods (gold standard). The greatest difference was -0.77±2.47 mm. For the segmented CBCT models, there were statistically significant differences, “especially in the lower arch with the highest difference being 0.49±0.38 mm” (San Jose et al). There were no statistically significant differences in the indirect measurements (anterior and overall Bolton index) between the 3D models from ILS and CBCT compared to the 2D models; the greatest difference was 0.55±0.34% between 2D digital and CBCT.
Evidence Search ((cone-beam scan) AND (intraoral scanner)) AND (validity) ((cbct scan) AND (intraoral scan)) AND (plaster models)
Comments on
The Evidence
Wiranto et al. focused on the reproducibility and reliability of obtaining diagnostic measurements such as tooth-width and Bolton ratios from digitized models by CBCT scan and intraoral scanner compared to the gold standard, plaster models. 22 volunteers from the Department of Orthodontics and Dental School were in the experimental group. Each volunteer had two sets of alginate impressions taken – one set was used to make plaster models and the other set was sent to Orthoproof to create digital models from CBCT scans of the impressions. In addition, the 22 volunteers also had an intraoral scan performed. There were three examiners who measured the tooth-width and Bolton ratios independently and were not aware of the scanning methods for the digital models. There were no inclusion or exclusion criteria listed and the experimental group patients were not randomly selected. In San Jose et al.’s study, the reproducibility and reliability of indirect and direct measurements of 3D digital models from intraoral laser scanner and segmented CBCT were compared to the gold standard, which was a 2D image of a plaster model with a conventional scanner. 240 patients who received treatment at the Department of Orthodontics were screened for the inclusion and exclusion criteria, which ultimately led to 100 patients being selected for the study. The inclusion criteria indicated that the patient needed to have initial records of a CBCT, intraoral scan, and plaster models. There was a main observer who took all dental measurements in the digital models and 2D models and then randomly selected 40 patients to take a second set of direct measurements the week after for intraobserver errors. A second observer was trained to take direct measurements on the same 40 patients to determine interobserver errors.
Applicability The studies included patients who had a CBCT radiograph taken for other reasons or participants whose impressions were sent to a laboratory to get scanned by a CBCT; therefore, there was no additional radiation exposure for participants. Although CBCT offers both skeletal and dental information in comparison with intraoral scanners, which captures only the dentition, there is an increase in radiation exposure to patients and an additional cost and training for orthodontists to incorporate this type of imaging into their practices. However, for patients who warrant CBCT imaging such as orthognathic cases, impacted teeth or other dental reasons, CBCT is a great diagnostic and treatment planning tool as orthodontists can incorporate additional information about skeletal and jaw relationships that otherwise cannot be captured from intraoral scanners.
Specialty (Orthodontics)
Keywords CBCT, intraoral scanner, plaster models
ID# 3504
Date of submission 11/26/2022
E-mail parkj11@livemail.uthscsa.edu
Author Janet Park
Co-author(s) Asma Samara
Co-author(s) e-mail samara@uthscsa.edu
Faculty mentor Dr. Karakousoglou
Faculty mentor e-mail karakousoglo@uthscsa.edu
Basic Science Rationale
(Mechanisms that may account for and/or explain the clinical question, i.e. is the answer to the clinical question consistent with basic biological, physical and/or behavioral science principles, laws and research?)
None available
Comments and Evidence-Based Updates on the CAT
None available