The maxillary right first molar of the typodont model (PRO2001-UL-SP-FEM-32; Nissin Dental Product, Kyoto, Japan) was used for the fabrication of the reference model. Tooth preparation was performed under the following conditions: reduction of 1.5 mm in the occlusal direction, reduction of 1.2 mm in the axial direction, and the finish line design in the form of chamfer. The tooth preparation model was scanned using a laboratory scanner (E1; 3Shape, Copenhagen, Denmark). The scanned model was formed in standard tessellation language (STL). The formed STL file was saved separately for the abutment and adjacent teeth. A 3D printer (Meg-printer 2; Megagen, Daegu, Republic of Korea) was used to fabricate the adjacent teeth, except the abutment. The abutment was fabricated as a reference model using the milling equipment (EZIS HM; DDS, Seoul, Republic of Korea). For abutment teeth, lithium disilicate ceramic (IPS e.max CAD; Ivoclar Vivadent AG, Schaan, Liechtenstein) with transparency similar to natural teeth was used. The milled abutment and printed adjacent teeth were post-processed according to the manufacturer’s instructions. A diamond bur (Cylindrical medium diamond; KG Sorensen, São Paulo, Brazil) was used to prepare the surface polish of the abutment. Five STL files were acquired for the abutment of the reference model using a contact scanner (DS10; Renishaw plc, Gloucestershire, UK). These files were aligned and merged using 3D reverse engineering software (Geomagic Design X; 3D Systems, Rock Hill, USA) to obtain 1 high-resolution CRM. The gingiva of the adjacent teeth and abutment was fabricated using the tissues from the neck region of a pig to reproduce the oral environment similar to the actual clinical situation (Fig. 1A). Because edible pig tissues were used, approval of the Institutional Review Board was not required. The location of the subgingival finish line should not exceed the depth of 1-mm to confirm the accuracy of IOSs according to the subgingival finish line depth.22,23,29,30 Therefore, the finish line of the abutment was arranged according to 5 depths (0-mm, 0.25-mm, 0.5-mm, 0.75-mm, and 1-mm) (Fig. 1B). The subgingival finish line depth was confirmed using 3D mesh viewer software (3Shape 3D viewer, 3Shape, Copenhagen, Denmark) and a periodontal probe (CP 15 UNC; HU-Friedy, CHI, USA). For gingival retraction, gingival displacement cords (Z-Twist Weave Retraction Cords; GINGI-PAK, Camarillo, CA, USA) were inserted into the gingival sulcus under the margin of the abutment (Fig. 1C). The gingival retraction was manipulated using a cord packer (Packing1; atria, Seoul, Republic of Korea). To calculate the mean and standard deviation (SD) of the subgingival finish line depth at random locations, vernier calipers (500-151-30; Mitutoyo, Takatsu-ku, Japan) were used. The subgingival finish line depth were 0.2521 ± 0.0039 mm at 0.25-mm, 0.5008 ± 0.0005 mm at 0.5-mm, 0.7507 ± 0.0008 mm at 0.75-mm, and 1.0006 ± 0.0005 mm at 1-mm.
CTMs were obtained by scanning the reference model with two types of IOSs: i500 (MEDIT, Seoul, Republic of Korea) and CS3600 (Carestream Dental, Atlanta, USA) (Fig. 1D). To confirm the effect of the subgingival finish line depth, IOSs were used to scan at 0-mm, 0.25-mm, 0.5-mm, 0.75-mm, and 1-mm. After scanning with 2 IOSs without gingival retraction, the gingival retraction was performed immediately and scanned. 400 CTMs were obtained by scanning 20 times before gingival retraction and 20 times after gingival retraction for each IOS depending on the depth. To prevent the gingiva replaced with pig tissue from drying and shrinking, distilled water was wetted in the gingiva region before scanning, and moisture on the abutment was removed so as not to affect the scanning environment. All areas except the area to be evaluated were removed using CAD software (Meshmixer; Autodesk, San Rafael, USA) so that the acquired CTM was not disturbed during the superimposition of data. All scanning was performed at an ambient temperature of 23 ± 2°C and a humidity of 50 ± 5% in consideration of the oral environment. All scanning procedures were evaluated by a trained investigator.
Scan data of CRM and CTM were 3D analyzed using 3D inspection software (Geomagic Control X; 3D Systems, Rock Hill, USA). For detailed analysis, the occlusal region, axial region, and marginal region were evaluated separately. 3D analysis was performed on the whole region (all regions above the finish line of abutment), occlusal region (the region down to 1 mm below the occlusal plane), axial region (between the occlusal region and the marginal region), and marginal region (the region up to 1 mm above the finish line) of the abutment. CRM and CTM were aligned using best-fit alignment, and the correspondence between CRM and CTM was evaluated using the 3D comparison function. The root mean square (RMS) was calculated based on all cloud points of CRM using the following formula:
$$RMS=\frac{1}{\sqrt{n}}\bullet \sqrt{{\sum }_{i=1}^{n}{\left({X}_{1,i}-{X}_{2,i}\right)}^{2}}$$
,
where \({X}_{1,i}\) indicates a measurement point at \(i\)-th in CRM and \({X}_{2,i}\) indicates a measurement point at \(i\)-th in CTM. n is the number of all points evaluated. Therefore, the RMS value is the absolute average distance of all cloud points and means the degree of agreement between CRM and CTM.
CAD software (Meshmixer) was used to evaluate the difference in surface area of the scanned surface area according to the subgingival finish line depth. After loading the saved CRM and CTM data into CAD software (Meshmixer), all areas except the upper area were deleted based on the finish line. The surface area of CRM and CTM was measured using the stability function. The difference in surface area was evaluated by calculating the difference between the measured CRM surface area and CTM surface area data. All measured surface area data were recorded, analyzed, and evaluated.
All data were analyzed using SPSS statistical software (release 23.0; IBM Corp, Chicago, IL, USA) (α = .05). One-way analysis of variance (ANOVA) and the Tukey HSD test were used to compare the accuracy of IOSs according to the subgingival finish line depth and according to the evaluated region. The paired t test was used to compare the accuracy of IOSs with and without gingival retraction, and an independent t test was used to compare the accuracy according to the IOSs. Two- and three-way ANOVA were used to verify the interaction effects.