Comparison of the accuracy of virtual and direct bonding of orthodontic accessories.

INTRODUCTION: Conventional direct and indirect bonding techniques fail to obtain the ideal bracket position. OBJECTIVE: To compare the accuracy of virtual and conventional direct bonding of orthodontic accessories. METHODS: A single virtual configuration (dental mannequin with Class I malocclusion) served as basis for generating the reference model (treated virtually) and the intervention models (10 digital models and 10 solid models, obtained by means of prototyping). A total of 560 teeth were then equally distributed between a group of orthodontists (Group I, direct bonding; and Group II, virtual bonding), working in two different time intervals. The individual positions of the accessories were measured after three-dimensional superimposition with customized software. The Student's-t test for paired samples, and Chi-square tests were used for statistical analysis, both at the level of significance of 5%. RESULTS: In comparison of the errors in raw values, there were significant differences only in the vertical (p< 0.001) and horizontal dimensions (p< 0.001). Considering the groups of ranges by clinical limits of the deviations, these differences were significant in the three dimensions, vertical (p< 0.001), horizontal (p= 0.044) and angular (p= 0.044). CONCLUSION: Virtual bonding made it possible to obtain more precise/accurate positioning of the orthodontic accessories. The potential accuracy of this method brings new perspectives to refining the indirect bonding protocols.