Background: With the introduction of digital systems in the orthodontic field, it is not still clear if such systems can accurately substitute analogical systems in the diagnosis and orthodontic treatment plane. Objective: The study compared the Arch Length Discrepancy (TALD) and Bolton ratio obtained from plaster dental casts (gold standard) and digital models and tested the null hypothesis that TALD and Bolton ratio measurements in digital models are affected by the degree of crowding. Methods: The sample included 40 dental models divided into five sub-groups, according to the American Board of Orthodontists (ABO) score of crowding. Plaster casts were scanned by a 3D laser scanner to obtain digital models. In digital and plaster models, the mesiodistal width of each tooth and arch lengths (maxillary and mandibular) were measured to calculate anterior and total Bolton ratios and TALD. Three operators performed measurements on plaster and digital models using a digital caliper and OrthoAnalyzerTM 3D software (3Shape A/S, Copenhagen, Denmark). Results: No statistically significant differences were obtained for intra and inter-examiner reliability (p > 0.05). When comparing plaster and digital findings, statistically significant differences were obtained for all measurements except for mandibular arch length and anterior Bolton ratio. In general, there was an overestimation of tooth size and arch length discrepancy in digital models. Moreover, the mean difference of measurements between both the methods increased in more crowded models. Conclusions: Crowding affects, the accuracy of Bolton ratio, and the arch length measurements. since the overestimation is relatively small, it should not be considered of clinical significance.
