Influence of the dental arch and number of cutting-off and rescanning mesh holes on the accuracy of implant scans in partially edentulous situations.

Objectives: To evaluate the influence of the dental arch and cutting-off and rescanning procedures on the accuracy of complete-arch implant scans in partially edentulous arches.

Material And Methods: A maxillary and a mandibular partially edentulous typodont with implant abutment analogs placed in the right and left first molar and right central incisor sites were digitized to create reference models by using an industrial optical scanner (7 Series Desktop Scanner; Dentalwings). Two experimental groups were scanned using an intraoral scanner (IOS) (TRIOS 4; 3Shape A/S): the Maxillary group (Mx) and the Mandibular group (Mb). Four subgroups were generated depending on the number of rescanned mesh holes: No holes (Mx-G0, Mb-G0), 1 hole (Mx-G1, Mb-G1), 2 holes (Mx-G2, Mb-G2) and 3 holes (Mx-G3, Mb-G3). A 3-dimensional metrology software (Geomagic Control X; 3D Systems) was used to measure the difference between the reference and the experimental scans computing the root mean square (RMS) error calculation. Two-way ANOVA and a post-hoc Tukey test were used to analyze the trueness data (α=0.05). Levene test was used to evaluate the prevision (α=0.05).

Results: The Mx group obtained a trueness mean value of 54 ± 17 µm and a mean precision value of 54 ± 17 µm, while the Mb group presented a trueness mean value of 67 ± 23 µm and a mean precision value of 66 ± 22 µm. The Mx group demonstrated significantly better trueness than the Mb group (P<.001). The G0 and G1 subgroups had the highest trueness values among the subgroups tested. No significant difference was observed between G0 and G1, G1 and G2, and G2 and G3 subgroups in trueness and precision. However, the G0 had significantly better trueness and precision values compared to G2 and G3 subgroups. In addition, the G1 had significantly better trueness values than the G3 subgroup. However, the Levene test revealed no difference in the precision mean values among the subgroups tested.

Conclusions: Implant scanning trueness was affected by the dental arch and the number of rescanned mesh holes using the IOS tested. A higher number of rescanned mesh holes decreased the scanning trueness. The stitching algorithm of the IOS software tested after the mesh hole scan demonstrated a significant error, especially when multiples mesh holes are involved in the same arch.

Clinical Significance: Given that cutting-off and rescanning techniques can reduce trueness, clinicians should consider whether these techniques are necessary in complete digital workflows. This is particularly important when fabricating multiple single implant-supported restorations in the same arch.