Influence of implant angulation and clinical implant scan body height on the accuracy of complete arch intraoral digital scans

Statement of problem The accuracy of digital implant scans can be affected by the implant angulation, implant depth, or interimplant distance. However, studies analyzing intraoral scanning accuracy with different implant angulations and different scan body heights are scarce. Purpose The purpose of this in vitro study was to determine the influence of the implant angulation and clinical implant scan body height on the accuracy of complete arch scans. Material and methods Two definitive implant casts with 6 implant analogs (Zimmer Biomet) were obtained: 1 cast had all the implant analogs parallel (GP group), and 1 cast had the implant analogs with divergence of up to 30 degrees (GD group). A coordinate measurement machine (Global Evo 09.15.08) was used to measure the positions of the implant analogs. Each group was divided into 3 subgroups depending on the clinical implant scan body height: 10, 6, and 3 mm. An implant scan body (Elos Accurate Scan Body Brånemark system) was positioned on each implant analog. A total of 10 scans of each subgroup were recorded by using an intraoral scanner (TRIOS 3). Each STL file obtained was imported into a reverse engineering software program (Geomagic), and linear and angular Euclidean measurements were obtained. The Euclidean calculations between the implant analog positions of the definitive implant casts were used as a reference to calculate the discrepancies among the corresponding subgroups. The Kolmogorov-Smirnov test revealed that the lineal measurements were not normally distributed, so the Kruskal-Wallis and pairwise comparison Dunn tests were used (α=.05). The Kolmogorov-Smirnov test revealed that the angular measurements were normally distributed. Therefore, the 2-way ANOVA and pairwise comparison Tukey tests were used (α=.05). Results The Kruskal-Wallis test revealed significant differences in the linear Euclidean medians between the GP and GD groups with different clinical implant scan body heights (H(5)=23.18, P<.001). Significant differences in the linear Euclidean medians were computed between the GP-6 and GD-10 subgroups (P=.009), GD-3 and GD-6 subgroups (P=.029), and GD-3 and GD-10 subgroups (P=.001). Two-way ANOVA revealed that the implant angulation (F(1, 3.3437)=28.93, P<.001) and clinical implant scan body height (F(2, 0.4358)=3.77, P=.029) were significant predictors of discrepancies in the angular measurement. Conclusions Implant angulation and clinical scan body height influenced scanning accuracy. The lowest clinical implant scan body height tested had the lowest accuracy in both parallel and angled implants, but statistically significant differences were found only in the angled group.