Acurácia de diferentes métodos para obtenção de modelos digitais de arco completo: um estudo in vivo

Abstract

Intraoral scanner systems (IOSs) are now routinely used and their accuracy has been primarily tested by laboratory studies. However, in vivo investigations are required to assess the accuracy of different digital equipment used for full-arch scans. This study aimed to investigate the accuracy of different IOSs used by experienced operators for full-arch scanning. Full-arch scans were taken from 15 subjects using two IOSs (Trios 3 and iTero 5D); after, PVS impressions and stone casts were obtained and digitized in a desktop scanner (inEos X5). Intraoral scans were performed at T0 and T1 periods for both IOSs; PVS impressions were taken at T0. The precision between the scanner systems was calculated by the mean deviation among the overlaps of the 4 scans from each group, at both periods (n=12). The trueness between the groups was assessed by superimposing the 8 intraoral scans of each participant on their reference models obtained from the desktop scanner (n=10). The models were analyzed in a proper 3D software to extract the data. ANOVA in split-plots in time was used for test trueness and in a completely randomized design for test precision. The mean precision for Trios 3 was 7.0 and 8.6 µm for maxillary and mandibular models, respectively. For iTero 5D, the mean precision was 9.0 and 8.8 µm for maxillary and mandibular models, respectively. No significant differences were observed for the precision of the IOS systems (P>0.05). The desktop scanner, showed significant differences for precision, presenting reduced variations (better precision) compared to the IOSs, with 1.0 and 0.9 µm precision for maxillary and mandibular models, respectively (P<0.05). The Trios 3 system presented better trueness for the lower models, in the maximum inferior parameter (P<0.05). The numeric values of trueness were improved from the T0 to T1 period, for the both IOS systems (P>0.05). Analyzing the trueness as a function of time, an improvement was observed in T1 for the maxillary models (maximum and minimum superior parameters), irrespective of the IOSs (P<0.05). The desktop scanner showed better precision for full-arch scanning comparing to the both IOS systems. Similar precision was achieved by experienced operators in full-arch scanning, regardless of the IOS system. The trueness of the maxillary models improved with the repeated scanning experience, irrespective of the IOS systems.