Accuracy of different digital acquisition methods in complete arch implant-supported prostheses: An in vitro study

Digital methods such as intraoral scanners for recording the location of implants supporting complete arch prostheses have limitations. Photogrammetry devices should be able to digitize implant positions accurately, but standardized comparisons between different digital acquisition methods are lacking.

The purpose of this in vitro study was to compare the repeatability of different digital acquisition methods for complete arch prostheses supported by 6 and 4 implants.

A master cast was created with 6 and 4 dental implants with multiunit abutments to obtain the master digital casts. The evaluated devices were the industrial high-resolution 12-megapixel scanner (reference) Atos Compact Scan 12M (GOM), the laboratory scanners D2000 (3Shape A/S) and S900 Arti (Zirkonzahn), the photogrammetry devices iCam (iMetric4D) and PIC (PIC Dental), and the intraoral scanners TRIOS 3 (3Shape A/S) and iTero Element 5D (Align Technology). The resulting files were imported to a computer-aided design software program (exocad GmbH) to obtain the implant replicas as standard tessellation language (STL) files. These files were imported into a software program (Geomagic Control X) and superimposed per group through the best-fit algorithm to determine repeatability, defined as the closeness of agreement between each group's scanned results as root mean square (RMS) values. The normality of distribution was tested by the Shapiro-Wilk normality test, and the Kruskal-Wallis test with adjustment with the Bonferroni correction method was used accordingly (α=.05).

The repeatability means and 95% confidence intervals for the 4 implant scans were: 1.07 µm (0.86; 1.29) for GOM, 2.05 µm (1.89; 2.21) for D2000, 3.61 µm (3.23; 3.99) for S900, 7.01 µm (6.11; 7.91) for iCam, 5.18 µm (4.6; 5.76) for PIC, 20.52 µm (18.33; 22.72) for TRIOS 3, and 20.5 µm (17.37; 23.63) for iTero. Statistically significant differences were found between devices, except for iCam versus PIC, GOM versus S900, iCam versus D2000, PIC versus D2000, and TRIOS 3 versus iTero. The repeatability means and 95% confidence intervals for the 6 implant groups were: 1.36 µm (1.08; 1.65) for GOM, 3.17 µm (3.01; 3.33) for D2000, 2.15 µm (2.04; 2.25) for S900, 8.67 µm (8.06; 9.28) for iCam, 13.88 µm (12.62; 15.14) for PIC, 40.32 µm (36.29; 44.36) for TRIOS 3, and 38.86 µm (34.01; 43.71) for iTero. Statistically significant differences were detected between devices, except for S900 versus GOM, PIC versus iCam, and iTero versus TRIOS 3.

The results suggest that photogrammetry could be a suitable alternative for recording implant locations of complete arch prostheses supported by 4 or 6 implants, with better repeatability than intraoral scanners. Increasing the number of implants decreased the repeatability of every device tested except the laboratory scanners.