In vitro study on digital splint effect to the accuracy of digital dental implant impression

Background

Digital implant impressions (DII) with intraoral scanners (IOS) are a relatively novel, but continuously improving technique. Since IOS devices can only capture part of the object at a time, images have to be stitched together to form a 3D object and therefore it is the source of possible errors of the scan. Digital splinting at edentulous areas can possibly improve the accuracy of DII.

Aim/Hypothesis

The aim of this in vitro study was to compare the trueness and precision of three different IOS scanning partially and fully edentulous models with 2 or 4 implants with attached scan bodies and digital splints.

Material and Methods

Two types of maxilla models were printed with Asiga Max 3D printer. The first model was missing both premolars and molars on the right side, so Straumann BL dental implants were inserted instead first premolar (straight) and second molar (tilted 20° mesially). Four implants were inserted in the second edentulous model symmetrically at second incisors (straight) and first molar areas (tilted 20° mesially). Scan bodies were attached to the implants and models were scanned with Nikon Altera 10.7.6. coordinate measurement machine (CMM) to form a reference scan. DII was taken with a Primescan (version 5.0.1), CS 3600 (version 3.1.0), Trios3 (version 1.18.2.10) IOS ten times each (n = 10) without digital splint. After that, tablets of hardened Fuji Plus cement was glued in edentulous areas to form digital splint and all models were scanned with three different IOS. Scanning data were exported in standard tessellation language format for analysis.

Results

Trueness of distance and angle in Carestream partially edentulous models was 185 μm in the group with splint and 280 μm without one and 0.22° in the group with splint and 0.29° in the group without respectively. Precision of distance and angle measurements in the splint groups were 87 μm and 0.13°, in the groups without −202 μm and 0.25°. In fully edentulous models trueness of distance varied 53–106 μm in the groups with splint and 67–8 μm in the groups without. Trueness of Primescan in partially edentulous models with splints was 21 μm and 0.16° in distance and angular measurements. Without splints −27 μm and 0.21°. For fully edentulous models trueness and precision of distance and angle was better n groups with splint than without. Trueness of distance and angle of Trios3 in partially edentulous splinted models was 15 μm and 0.3°+53 μm and 0.11° in unsplinted models respectively.

Conclusion and Clinical Implications

Primescan showed the best results of trueness and precision of distance and angle measurements. Since digital splints improve the accuracy of DII, the impact of their forms and materials should be more researched.