Accuracy of 3D-printed models and positional trueness of removable dies fabricated by two milling centers

To evaluate the accuracy of 3D-printed models and positional trueness of the removable dies.

A typodont incisor, canine, premolar, and molar were prepared for single crowns and digitized. Based on this scan, two sets of models (n = 10 each) were 3D-printed in two milling centers, resulting in a total of 80 dies. The two model types varied in die configuration (conical and root-formed), resin type, and post-processing techniques. Fabrication accuracy was evaluated after model delivery. The positional trueness was measured at baseline, after 10 and 50 repositioning cycles, by evaluating metric and angular die displacements. Deviations were compared between the groups using non-parametric Mann-Whitney-U tests at a significance level of 0.05.

Fabrication accuracy was similar for the two model sets. Trueness/precision were 72 ± 15 µm/76 ± 19 µm for center 1. For center 2, the corresponding values were 60 ± 5 µm/79 ± 4 µm. Significant differences in positional trueness were observed between the two model types after both 10 and 50 repositioning cycles. After 10 cycles, mean die displacements were 78 ± 76 µm for center 1 and 526 ± 470 µm for center 2 (p < 0.001). These values showed minimal variation after 50 cycles, with mean die displacements of 98 ± 80 µm for center 1 and 470 ± 386 µm for center 2 (p < 0.001). Angular deviations were minimal for both groups, even after 50 cycles, with deviations below 1°.

While the accuracy of the 3D-printed models was adequate, the positional trueness of the dies requires improvement, as some currently available models demonstrate unacceptable discrepancies.

The quality of 3D-printed models with removable dies for CAD/CAM crowns can vary substantially between milling centers. While some models can merely serve as restoration carriers, others are equipped with accurate dies that can be removed and repositioned reliably.