Statement of problemThe effect of additional reference objects on the accuracy of different intraoral scanners for partially and completely edentulous patients has not been investigated sufficiently. PurposeThe purpose of this in vitro study was to evaluate the effect of an additional reference object in the form of additional artificial landmarks on the trueness and precision of different intraoral scanners in partially and completely edentulous areas. Material and methodsPartially and completely edentulous models with 2 and 4 implants (BLT, RC, Institut Straumann AG), respectively, were used in the study. For the digital scan, scan bodies (CARES Mono Scanbody) were attached, and reference data obtained by using industrial scanners. Ten digital scans of the same model were made with each intraoral scanner: PRIMESCAN, TRIOS 3, TRIOS 4, Carestream 3600, and Medit. Then, additional artificial landmarks were attached, and 10 more intraoral scans were made with each device. Computer-aided design files of the scan bodies were aligned to obtain 3-dimensional surfaces with reference and test scanners. Trueness and precision of distance, angulations, and vertical shift between scan bodies were estimated. The Mann-Whitney Wilcoxon or Student 2-sample t test was applied to estimate statistically significant differences between groups (α=.05). ResultsIn the partially edentulous model, distance trueness mean ±standard deviation values ranged from −46.7 ±15.4 μm (TRIOS 3) to 392.1 ±314.3 μm (Medit) in models without additional artificial landmarks. When additional artificial landmarks were applied, trueness of distance mean ±standard deviation values ranged between −35 ±13 μm (TRIOS 4) and 117.7 ±232.3 μm (CARESTREAM). Trueness mean ±standard deviation values of angulation varied from −0.0 ±0.5 degrees (CARESTREAM) to 0.2 ±0.0 degrees (PRIMESCAN) without additional artificial landmarks and from 0.0 ±0.2 degrees (TRIOS 3) to 0.4 ±0.5 degrees (CARESTREAM) with additional artificial landmarks. Vertical shift trueness measurements varied from −108 ±47.1 μm (TRIOS 4) to 107.2 ±103.5 μm (Medit) without additional artificial landmarks and from −15.0 ±45.0 μm (CARESTREAM) to −86.9 ±42.1 μm (TRIOS 4) with additional artificial landmarks. The additional artificial landmark technique improved the trueness of all measured parameters for the 5 tested intraoral scanners. No statistically significant differences were found among models with or without additional artificial landmarks, except for Medit in all parameters and PRIMESCAN in angle measurements (P<.05). The best precision for distance was found with TRIOS 3 and with PRIMESCAN for angulation and vertical shift. Larger deviations were observed in the completely edentulous situation. The effect of additional artificial landmarks was limited when the accuracy parameters of digital scans were considered. ConclusionsScans with and without additional artificial landmarks of partially edentulous conditions scanned by any of the intraoral scanners tested did not influence precision and trueness, except for Medit i500 in the distance and vertical shift parameters and CARESTREAM3600 in vertical shift. Precision and trueness of digital scans of completely edentulous areas were affected, except for Medit i500 for distance, PRIMESCAN and TRIOS 4 for angle, and all systems except TRIOS 4 for vertical shift precision.