Impact of different complete coverage onlay preparation designs and the intraoral scanner on the accuracy of digital scans

Abstract

Statement of problem: The trueness and precision of intraoral scanners (IOSs) and the effect of intracoronal restorations have been reported. However, studies addressing the accuracy of IOSs in reproducing different complete coverage onlay preparation designs are lacking. Purpose: The purpose of this in vitro study was to evaluate the influence of complete coverage onlay preparation design and intraoral scanning devices on the accuracy of digital scans in terms of trueness and precision. Material and methods: Three preparation designs on the mandibular first molar were considered: a traditional preparation design with isthmus reduction (IST), a traditional preparation design without isthmus reduction (wIST), and simplified nonretentive preparation (nRET). Digital scans of epoxy resin mandibular arch reference models of the preparations (containing second premolar, first molar, and second molar) were obtained by using 3 IOSs (iTero Element 2 [ELE], Trios 3 [TRI], and Primescan [PRI]) (n=10). Trueness (μm) and precision (μm) were analyzed by superimposing the digital scan on the digital reference models obtained with a high-accuracy industrial scanner (ATOS Core 80) in a tridimensional metrology software program. Accuracy was quantified by the absolute deviation (μm). Local and overall mean positive and negative deviations for trueness were also obtained. Data were analyzed by using the Kruskal-Wallis and Dunn tests with a statistical software program (α=.05). Results: The nonretentive preparation groups obtained higher trueness (3.8 μm) and precision (2.7 μm) than the IST and wIST groups (trueness=7.5 to 6.3 μm, precision=5.5 to 4.6 μm). Trueness values were lower with ELE×IST (16 μm), followed by ELE×wIST (13 μm), and PRI×IST (7.8 μm). In general, no difference was found between PRI and TRI scanners (6.3 to 5.9 μm), with lower performance for ELE (13 μm). Positive deviations were higher on the proximal box of the IST and wIST preparation and on the occlusal box of the IST group. Negative deviation was higher on the ELE×IST occlusal box. Conclusions: Different intraoral scanners and preparation designs influenced the accuracy of digital scans. A more complex preparation such as IST and wIST showed higher deviation. The iTero Element 2 scanner exhibited higher deviation for both trueness and precision.