Prosthesis accuracy of fit on 3D-printed casts versus stone casts: A comparative study in the anterior maxilla.

Abstract

To compare in vitro the accuracy of fit of a reference prosthesis seated on three-dimensional (3D) printed casts generated from digital implant scans vs stone casts made by conventional implant impressions. A partially edentulous maxillary master cast with two internal connection implants was generated, while a reference implant-supported prosthesis was fabricated. Conventional splinted open-tray impressions were taken to create stone casts (n= 10) (control group). A digital scan was taken of the master cast using a white light intraoral optical scanner (IOS) (TRIOS, 3Shape), and a Standard Tessellation Language (STL) file was obtained. Four 3D printers were used to print the casts (n= 10 from each 3D printer): Straumann® P30+, Varseo S, Form 3b+and M2 Carbon. Accuracy of fit of the reference prosthesis on all control and test casts was assessed using the screw resistance test and radiographic test. Additionally, all casts were digitized using the same IOS, and the STL files were superimposed to the master cast STL file (reference) to evaluate the 3D accuracy with inspection Geomagic Control software using the root-mean-square (RMS) error. The reference prosthesis presented with clinically acceptable fit on all casts. The highest median RMS error was found in the stone cast group (94.6μm) while the lowest median was in the M2 Carbon group (46.9μm). The Kruskal-Wallis test revealed a statistically significant difference between the groups (p< 0.001). For post hoc comparisons, Dunn's test with the Holm-Bonferroni correction resulted in a statistically significant difference in four tests, with M2 Carbon exhibiting lower RMS error than the stone cast (p < 0.001) and P30+ (p < 0.001) groups, Form 3b exhibiting lower RMS error than the stone cast (p < 0.001) group, and Varseo S exhibiting lower RMS error than the stone cast (p=0.006) group. Using the screw-resistance test and radiographic assessment, the reference prosthesis fit presented with clinically acceptable accuracy of fit on all casts. Printed casts from 3 different printers demonstrated statistically significant lower 3D deviations than stone casts generated using a conventional implant impression for the present partially edentulous scenario with two implants, but this did not affect prosthesis fit. Even though there were 3D deviations between the master cast and all control and test casts generated from conventional impressions and digital scans respectively, the reference prosthesis presented with accurate fit on all casts. This indicates that there is a threshold for clinically acceptable accuracy of fit and that 3D-printed casts may be used as definitive master casts to fabricate implant-supported fixed dental prostheses for the partially edentulous anterior maxilla.