ABSTRACT
STATEMENT OF PROBLEM: Complex prosthodontic treatments frequently require capturing an accurate impression of abutment teeth throughout the dental arch. Intraoral digital scanning has been reported to be accurate within single dental quadrants; however, the positional trueness of complete arch intraoral digital scans is not well understood. PURPOSE: The purpose of this in vitro study was to compare the positional trueness of a complete arch digital scan generated by digitally merging portions of a direct digital scan from an intraoral scanner (IOS) with a digitized polyvinyl siloxane (PVS) complete arch impression. MATERIAL AND METHODS: A 3D-printed reference cast was scanned with a desktop scanner, and reference standard tessellation language (STL) data sets were obtained. The reference cast was used to generate 4 nonmerged (NM) groups and 2 die-merged (DM) groups. In the NM groups (n=10), a direct digital scan of the reference cast was made with an IOS (NM-IOS), the PVS impression of the reference cast was digitized by using a cone beam computed tomography (CBCT) scanner (NM-PVS-CBCT) or a desktop scanner (NM-PVS-DESK), and the gypsum cast made from the PVS impression was digitized by using a desktop scanner (NM-STONE). In the DM groups (n=10), individual dies were cropped from the complete arch digital scan from the NM-IOS group and merged with the digital scans from the NM-PVS-DESK and NM-PVS-CBCT groups to generate the DM-PVS-DESK and DM-PVS-CBCT groups, respectively. Deviation was measured as the absolute value of the distance from a reference position on the reference cast. The Kruskal-Wallis and Dunn pairwise comparison tests were used to compare the difference in deviation between the groups (α=.05). RESULTS: The NM-STONE and NM-PVS-DESK groups demonstrated the highest positional trueness, with global deviations of 19.6 and 17.7 µm, respectively, with no statistically significant difference (P=1.00). However, both the NM-IOS and NM-PVS-CBCT groups differed significantly from the NM-STONE group (P<.001 and P=.003, respectively) and the NM-PVS-DESK group (P<.001 and P=.004, respectively). In the DM groups, the DM-PVS-CBCT group presented a higher deviation than the DM-PVS-DESK group. CONCLUSIONS: Complete arch digital scans of PVS impressions digitized with a desktop scanner exhibited less positional deviation than those digitized with a CBCT scanner or complete arch digital scans generated with an IOS. Complete arch digital scans generated with an IOS and CBCT scanner result in more deviations in the posterior regions, and their positional trueness may not be sufficient to construct an accurate digital scan. Generating a complete arch digital scan by digitally merging portions of a direct digital scan from an IOS with a digitized PVS complete arch impression is a suitable alternative to the contemporary workflow of digitizing a stone cast with a desktop scanner.