Digital versus conventional implant impressions for partially edentulous arches: An evaluation of accuracy

Statement of problem. To the authors' knowledge, while accuracy outcomes of the TRIOS scanner have been compared with conventional impressions, no available data are available regarding the accuracy of digital scans with the Omnicam and True Definition scanners versus conventional impressions for partially edentulous arches. Purpose. The purpose of this in vitro study was to compare the accuracy of digital implant scans using 2 different intraoral scanners (IOSs) with that of conventional impressions for partially edentulous arches. Material and methods. Two partially edentulous mandibular casts with 2 implant analogs with a 30-degree angulation from 2 different implant systems (Replace Select RP; Nobel Biocare and Tissue level RN; Straumann) were used as controls. Sixty digital models were made from these 2 definitive casts in 6 different groups (n=10). Splinted implant-level impression procedures followed by digitization were used to produce the first 2 groups. The next 2 groups were produced by digital scanning with Omnicam. The last 2 groups were produced by digital scanning with the True Definition scanner. Accuracy was evaluated by superimposing the digital files of each test group onto the digital file of the controls with inspection software. Results. The difference in 3-dimensional (3D) deviations (median interquartile range) among the 3 impression groups for Nobel Biocare was statistically significant among all groups (P<.001), except for the Omnicam (20 +/- 4 mu m) and True Definition (15 +/- 6 mu m) groups; the median +/- interquartile range for the conventional group was 39 +/- 18 mu m. The difference in 3D deviations among the 3 impression groups for Straumann was statistically significant among all groups (P=.003), except for the conventional impression (22 +/- 5 mu m) and True Definition (17 +/- 5 mu m) groups; the median interquartile range for the Omnicam group was 26 +/- 15 mu m. The difference in 3D deviations between the 2 implant systems was significant for the Omnicam (P=.011) and conventional (P<.001) impression techniques but not for the True Definition technique (P=.247). Conclusions. Within the limitations of this study, both the impression technique and the implant system affected accuracy. The True Definition technique had the fewest 3D deviations compared with the other 2 techniques; however, the accuracy of all impression techniques was within clinically acceptable levels, and not all differences were statistically significant.