Objectives: The aim of this study was to investigate the accuracy of CAD/CAM-fabricated bite splints in dependence of fabrication method (milling vs 3D printing), positioning (horizontal vs vertical), selection of material, and method of deviation measurement. Materials and methods: Bite splints were 3D-printed in either horizontal or vertical position (n = 10) using four different resins (Dental LT, Ortho Clear, Freeprint Splint, V-Splint). As control, ten bite splints were fabricated by CNC milling (ProArt CAD Splint). The splints were scanned and deviations between the CAD-file (trueness) and between each other within one group (precision) were measured by two different software applications and methods (cloud-to-cloud vs cloud-to-mesh). Data were analyzed using univariate analysis, Kolmogorov-Smirnov, Kruskal-Wallis, and Mann-Whitney U tests. Results: The highest impact on accuracy was exerted by the selection of the material (trueness: η_P² = 0.871, P < 0.001; precision: η_P² = 0.715, P < 0.001). Milled splints showed the highest trueness (P < 0.01) but not the highest precision at the same time. Horizontally positioned 3D-printed bite splints showed the least deviations in terms of trueness while vertical positioning resulted in the highest precision. The cloud-to-cloud method showed higher measured deviations than the other methods (P < 0.001–P = 0.002). Conclusion: Milled splints show higher trueness than 3D-printed ones, while the latter reveal higher reproducibility. The calculated deviations vary according to the measurement method used. Clinical relevance: In terms of accuracy, milled and 3D-printed bite splints seem to be of equal quality.

Aim: To introduce virtual reality (VR) into the endodontic curriculum for teaching root canal anatomy and to evaluate the effectiveness of this new method on third‐year undergraduate students.Methodology: Extracted human teeth were digitized using a CBCT scan and converted into STL files. The corresponding files were either 3D printed or imported into a VR software program. Subsequently, forty‐two third‐year undergraduate dental students in preclinical training were asked to respond to a questionnaire analysing their ability to detect all the anatomic features of the replica teeth and their comprehension of the underlying root canal anatomy. The investigation was based on three different methods: two‐dimensional radiography, CBCT scanning and VR simulation. Data were analysed using McNemar’s and binomial test, and the level of significance was set to 0.05 (P = 0.05).Results: Students reported that CBCT and VR allowed them to detect all anatomic features more than radiography (P < 0.001 – P = 0.049). Because it allowed improved comprehension of root canal anatomy, the VR simulation was considered better than CBCT scanning and radiography. Most of the students adapted well to the VR simulation.Conclusions: Dental students greatly appreciated the integration of VR simulation into the endodontic curriculum. From a didactic point of view, VR has considerable advantages over three‐dimensional reconstructions and two‐dimensional radiographs when teaching root canal anatomy.