Accuracy of IOS on vertical and horizontal deep subgingival preparations: A three-dimensional analysis

To evaluate the accuracy of an intraoral scanner (IOS) on both horizontal and vertical tooth preparations at different depth levels below the gingival margin and assess if the IOS detects the area beyond the finish line of these preparation geometries. Four abutments of an upper first molar were designed using CAD software, with vertical and horizontal preparation finish lines at 1 and 2 mm from the gingival margin. These abutments were printed in resin and placed each on a reference typodont. Ten scans were made with an IOS for each preparation geometry to obtain four experimental groups: for horizontal preparation "H1" at 1 mm from the gingival margin and "H2" at 2 mm, while for vertical preparation "V1" at 1 mm and "V2" at 2 mm. The scans were imported into a dedicated software, then trueness and precision were evaluated in µm. In addition to descriptive statistics (95%-CI), the Games-Howell was run to analyze differences among groups (α=.05). Mean values for trueness were H1=29.9(28.4-31.3), H2=22.2(21.3-23.1), V1=25.4(22.4-28.5), V2=26.4(25.2-27.5) with statistically significant differences for H1/H2 (P<.001), H1/V1 (P=.047), H2/V2 (P<.001). Mean values for precision were H1=14(12.3-15.7), H2=14.5(11.8-17.2), V1=23.3(15-31.5), V2=35.6(29.8-41.4), with significant differences for H2/V2 (P<.001). Only for vertical preparations, it was possible to record the area beyond the finish line. Although vertical preparations allow for registration beyond the finish line with IOS, they are less accurate than horizontal ones when the preparation margin is 2-mm from the gingival margin.