Statement of problem: Additive manufacturing is commonly used for the fabrication of definitive casts with removable dies. However, how the trueness and fit of removable dies are affected by printing layer thickness is lacking.

Purpose: The purpose of this in vitro study was to investigate the trueness and fit of additively manufactured removable dies printed in different layer thicknesses.

Material and methods: A mandibular cast with a prepared right first molar tooth was digitized (CEREC Primescan), and its standard tessellation language (STL) file was imported into a software program (DentalCAD 3.0). A removable die (D-STL) and a hollow cast with (M-STL) or without the die (SM-STL) were designed. D-STL and SM-STL were imported into a nesting software program (Composer), and 45 removable dies in 3 layer thicknesses (100 μm, 50 μm, and 50 to 100 μm) (n=15) and 1 cast (100-μm) were additively manufactured. Each removable die (TD-STLs), the cast with each die (TM-STLs), and the cast without the die (TSM-STL) were digitized by using the same scanner. All STL files were imported into a software program (Medit Link v 2.4.4), and TD-STLs were superimposed over D-STL. The root mean square (RMS) method was used to analyze the trueness of the dies at 2 different areas (crown and root portion) and as a complete unit (overall). Overall RMS values of the cast with and without the die were also calculated after superimposing TM-STLs over M-STL. The fit of the dies in the cast was evaluated by using a triple-scan protocol to measure deviations at 5 different points (point M: most mesial point of the margin; point TM: tip of the mesial cusp; point O: deepest point of the occlusal fossa; point TD: tip of the distal cusp; point D: most distal point of the margin) on the crown portion. One-way ANOVA and Tukey honestly significant difference tests were used to evaluate data (α=.05).

Results: The RMS values of removable dies showed significant differences at each area (P≤.002). The 50- to 100-μm group had higher overall RMS values than the 100-μm group (P=.017). The 100-μm group had the highest RMS values for the crown portion (P≤.019), while the 50-μm group had the highest RMS values for the root portion (P<.001). The 50-μm group had the lowest RMS values for the crown portion when the die was in the cast (P<.001). Except for point TM (P=.228), significant differences were observed among the test groups at all points (P<.001). The 50-μm group had the lowest distance deviations at points M, TD, and D (P≤.005), while the 100-μm group had the highest distance deviations at points O and D (P≤.010).

Conclusions: Removable dies fabricated by using a 100-μm or 50- to 100-μm combined layer thickness had trueness that was either similar to or better than that of dies fabricated with a 50-μm layer thickness. When the die was on the cast, the 50-μm layer thickness resulted in the best crown portion trueness. However, because the deviation differences among groups were clinically small, the 100-μm layer thickness can be considered for the efficient fabrication of removable dies when the tested printer and resin are used.