Influence of splinting scan bodies or incorporating three-dimensionally printed scan aids on the trueness of complete arch digital scans

Statement of problem

Studies are sparse on how splinting scan bodies or incorporating 3-dimensionally (3D) printed scan aids influence the trueness of complete arch digital scans.

Purpose

The purpose of this in vitro study was to compare the trueness of multisite implant recordings obtained using 6 different methods on an edentulous mandible.

Material and methods

A definitive cast of an edentulous mandible with 4 multi-unit analogs placed at different angles and interanalog distances was extraorally scanned, and the resulting data were saved as a reference file. To obtain experimental files, 6 distinct methods were used: conventional impression with splinted open-tray impression copings (IC), intraoral scanning (IOS) without splinting scan bodies or using any scan aids (SB), IOS with pattern resin-splinted scan bodies (PR), IOS with composite resin-splinted scan bodies (CR), IOS with 3D printed custom scan bodies (CSB), and IOS with 3D printed auxiliary apparatus (AA). The experimental files were aligned to the reference file in a metrology software program. The 3D comparison algorithm was run to quantify the root mean square estimate error (RMS). Scan bodies in the files were converted to hollow virtual cylinders, and the Cartesian coordinates of the lines passing through the centers of these cylinders were recorded to analyze angular (AD) and linear distortion (LD). LD was further analyzed along the x (∆X), y (∆Y), and z axes (∆Z). One-way ANOVAs with the Tukey HSD test were used for statistical analysis (α=.05).

Results

AD at all sites, LD at all sites, and the RMS error showed significant differences (P<.05). The IC group showed the lowest AD values across all sites, followed by the AA, CSB, PR, CR, and SB groups. The SB group had the greatest LD values at all sites, while the IC group indicated the lowest LD values at all sites except the left anterior site. In terms of 3D distortions, the SB group had the largest RMS value, whereas the IC group showed the lowest RMS value. ∆X, ∆Y, and ∆Z values also showed significant differences at all sites (P<.001) except for the ∆Z values at the right anterior site (P=.194). The highest mean ∆X, ∆Y, and ∆Z values were recorded in the SB group except for the ∆Z measurement of the left posterior site.

Conclusions

The IC group outperformed the other groups. The AA group exhibited distortion comparable with that of the IC group. Splinting scan bodies or using scan aids enhanced the trueness.