Accuracy of Full Arch Scans in Contemporary Intraoral Scanners

Objectives: In recent years digital impressions gained more and more importance. With regard to their accuracy the reproduction of a single tooth can be compared to conventional impressions. However, the few studies considering full arch scans show contradictory results. Therefore the objective of this study was to analyze the accuracy of full arch scans with three contemporary digital scanners in a laboratory set up.
Methods: A cast of an upper jaw with full dentition was manufactured from a chrome-cobalt-molybdenum-alloy using a Laser sintering processes. Afterwards four metal balls, 5mm in diameter, were fixed on the canines respectively the second molars. The cast was digitized with a high precision scanner (ATOS Core, GOM mbH). The three-dimensional position of the balls was assessed and saved as reference. Ten full arch scans were made with each of the following intraoral scanners: CEREC AC Omnicam (Sirona), True Definition (3M) and Cara (Heraeus-Kulzer). The resulting scans were compared to the reference using GOM Inspect V.7.5. The change of the inter-ball-distances and the tilt of the different planes spread out in between three balls at a time were calculated and checked for significance differences using ANOVA.
Results: The deviation of the distances (mean±std) ranged from 17±9µm (Cara) up to 471± 15µm (Omnicam) in the posterior region (sign. p<0,05) and 22µm ± 16µm (Cara) up to 420µm ± 20µm (Omnicam) in the anterior region (sign. p<0,05). The tilt ranged from 0,2± 0,12 (Cara) up to 0,3±0,19 (Omnicam). There are partly significant differences between the scanning systems.
Conclusions: Within the limits of this study it can be concluded that some scanning systems, especially Cara (Heraeus) are able to reproduce full arches with an accuracy comparable to data reported for conventional impressions.