Torsion and linear accuracy in intraoral scans obtained with different scanning principles

Few investigations have examined the production of single restorations using intraoral scanners (IOS). Data on full-arch scans are rare, and data regarding torsion within the entire arch are very sparsely reported. Therefore, the aim of this study was to examine the deviations of torsion and linear distances in full-arch scans of three IOS based on different scanning principles.A cobalt-chrome-molybdenum alloy master model (CCMM) with four hemispheres was fabricated by laser sintering. The CCMM was digitized using a laboratory scanner (ATOS-Core/GOM) and scanned with three IOS (Omnicam/Sirona(OC); True Definition/3M(TD); TriosII/Cara-Version/Kulzer(TR)). All scan data were exported in a standard STL-file format and were analyzed with GOM Inspect software (V7.5/GOM). Torsion between the right and left side of the arch and linear accuracy (trueness and precision) were evaluated. After normality was confirmed, all data were subjected to parametric statistical analyses.The torsion ranged from 0.07±0.03°(OC) to 0.29±0.14°(TD). Pairwise comparisons showed significant differences between the OC and TD scanners and between the TR and TD scanners. The linear distances ranged from 6±5μm(OC) to 298±317μm(TD). Significant differences were observed among all investigated IOS (p=0.05).Although the highest torsion was observed for the TD scanner, it is still not clear whether the differences between the IOS are related to the scanning principle or to the scanning algorithm. Due to the high clinical relevance of full-arch restorations, future studies should consider torsion. Regarding linear accuracy, no general difference related to the scanning principles of the IOSs was observed.