The authors evaluated the local accuracy of intraoral scanning (IOS) systems for single-tooth preparation impressions with an in vitro setup.
The authors digitized a mandibular complete-arch model with 2 full-contour crowns and 2 multisurface inlay preparations with a highly accurate reference scanner. Teeth were made from zirconia-reinforced glass ceramic material to simulate toothlike optical behavior. Impressions were obtained either conventionally (PRESIDENT, Coltène) or digitally using the IOS systems TRIOS 3 and TRIOS 3 using insane scan speed mode (3Shape), Medit i500, Version 1.2.1 (Medit), iTero Element 2, Version 1.7 (Align Technology), CS 3600, Version 3.1.0 (Carestream Dental), CEREC Omnicam, Version 4.6.1, CEREC Omnicam, Version 5.0.0, and Primescan (Dentsply Sirona). Impressions were repeated 10 times per test group. Conventional (CO) impressions were poured with type IV gypsum and digitized with a laboratory scanner. The authors evaluated trueness and precision for preparation margin (MA) and preparation surface (SU) using 3-dimensional superimposition and 3-dimensional difference analysis method using (95% - 5%) / 2 percentile values. Statistical analysis was performed using Kruskal-Wallis test. Results were presented as median (interquartile range) values in micrometers.
The authors found statistically significant differences for MA and SU among different test groups for both trueness and precision (P < .05). Median (interquartile range) trueness values ranged from 11.8 (2.0) μm (CO) up to 40.5 (10.9) μm (CEREC Omnicam, Version 5.0.0) for SU parameter and from 17.7 (2.6) μm (CO) up to 55.9 (15.5) μm (CEREC Omnicam, Version 5.0.0) for MA parameter.
IOS systems differ in terms of local accuracy. Preparation MA had higher deviations compared with preparation SU for all test groups.
Trueness and precision values for both MA and SU of single-unit preparations are equal or close to CO impression for several IOS systems.