Evaluation and comparison of the accuracy of three intraoral scanners for replicating a complete denture

Statement of problem. Technological advances in digital acquisition tools have increased the scope of intraoral scanners (IOSs), including scanning a removable complete denture (RCD) to replicate it. However, studies assessing the accuracy of IOSs for replicating a maxillary or mandibular RCD are lacking. Purpose. The purpose of this in vitro study was to evaluate the accuracy (trueness and precision) of 3 IOSs while replicating a maxillary and mandibular RCD. Material and methods. One maxillary and 1 mandibular RCD were scanned with a desktop scanner (D2000) to obtain the reference model. Two operators scanned each RCD 5 times with 3 different IOSs (TRIOS 4, Primescan, and IS3800), following a predefined acquisition protocol. The 60 study models obtained were compared with the reference model using the Geomagic software program. For each comparison, the mean and standard deviation of discrepancy were calculated. Distances were measured on both the reference and the study model, and differences were calculated to assess whether sagittal or transverse deformations were present. The tolerance percentage of the volume of the digital model compared with the volume of the reference model was determined (difference tolerance was set at 0.1 mm). A univariate analysis of variance followed by a post hoc analysis using the Student-Newman-Keuls (alpha=.05) test was performed to determine the truest and the most precise IOS. Results. The TRIOS 4 and Primescan IOSs had comparable trueness, with mean dimensional variations of 47 +/- 27 mu m and 57 +/- 8 mu m respectively compared with the reference model. The IS3800 had a lower trueness (98 +/- 35 mu m). Primescan was significantly more precise with a mean standard deviation of 64 +/- 15 mu m (P<.05). The TRIOS 4 (141 +/- 48 <mu>m) and IS3800 (129 +/- 24 mu m) had comparable precision. Primescan showed the least sagittal and transverse deformation. Conclusions. This study determined that an RCD can be replicated using an IOS, although all IOSs did not have equal accuracy. An in vivo study needs to assess whether this procedure is clinically acceptable.