OBJECTIVES: The purpose of this in vitro study is to measure and compare the accuracy of the conventional impression, the intraoral scanner, and the photogrammetry techniques for full-arch implant-supported dental prostheses at the abutment level. METHODS: An edentulous maxillary master model containing 6 implant abutment replicas (RP analog for screw-retained abutment straight from NobelReplace® Multi-unit Abutment Plus Replica) was fabricated. A reference STL of the master model was obtained using a desktop scanner (inEos X5, Dentsply Sirona) with high trueness and precision and served as the control STL. Three impression techniques were performed: the intraoral scanning (IOS) group (TRIOS 3 Battery Cart, 3Shape A/S), the photogrammetry (PTG) group (ICam 4D Generation 3, Imetric), and the conventional (CNV) group. Ten impressions from each group were tested. Scan bodies in each STL file acquired from the different impression techniques were converted to implant abutment replicas using a digital library. Three tests were completed to compare the different registration techniques. A 3D deviation test between the experimental group and the reference was done on an inspection software (IScan4D Dental Version 9.1.104; Imetric) using a “best fit” algorithm to obtain the root mean square values, and on another inspection software (IScan3D Dental Version 9.1.104; Imetric) using spatial similarity transformation. The second test was meant to assess the angular deviations of the implant abutment replicas using a reverse engineering software (Geomagic Control X 2020.1; 3D Systems). The final test for cross-arch distances was done on an inspection software (IScan3D Dental Version 9.1.104; Imetric) which allows to determine the 3-dimensional coordinates for each implant by using the origin point and compare the cross-arch distance deviations as well as deviations at the x-, y-, and z- coordination. Trueness and precision were the two parameters used to define the accuracy of a system. The term "trueness" was used to see how close the measurements from the experimental files of each group were to those of the reference file. The term "precision" was defined as to see how close the measurements of each experimental files were to each other within the same group. The 3D discrepancies were then calculated and the trueness and precision of the three impression techniques were assessed and compared statistically (α = 0.05). RESULTS: The root mean square of 3D deviation values through the ICP “best-fit” method showed statistically significant differences between the PTG and CNV group (p < .0001), and the PTG and IOS group (p < .0001). The CNV and IOS group did not show statistically significant differences (p = 0.8626) through the “best-fit” method, but significant differences were observed via the “spatial similarity” method (p = 0.0041). Both methods however showed the best results in terms of trueness for the PTG group, followed by the CNV group and least the IOS group. In terms of precision, PTG showed the best results, followed by the IOS group and least the CNV group. The angular deviation test using the “best-fit” alignment method showed that the PTG and CNV had no statistically significant difference (p = 0.7955) and were equivalent. However, both showed a statistically significant difference to the IOS group (p < .0001), which had the highest angular deviation. Finally, in terms of cross-arch distances, the photogrammetry group showed optimal results followed by the IOS group and then the CNV group. In general, the shorter the inter-arch distance, the lower the deviation was. A larger deviation was observed on longer inter-arch distances. Considering the deviations on the 3 axes of each implant, the CNV technique had the highest deviation in the X-axis (longitudinal) and the IOS technique showed the highest deviation on the Y-axis (lateral) and Z-axis (vertical). The PTG technique experienced significantly less deviation on the X-axis, Y-axis, and the Z-axis. CONCLUSION: Within the scope of this study, the photogrammetry technique reported the best accuracy in terms of trueness and precision of implant positions for complete-arch implant rehabilitation. Conventional impressions showed better accuracy results than intraoral scanning in the 3D deviations test and global angular deviation test, however the latter exhibited better results in terms of accuracy in terms of cross-arch distances test. / 2023-08-11T00:00:00Z