ObjectivesTo compare the accuracy of an original and two newly designed CAD/CAM scan bodies used in digital impressions with one another as well as conventional implant impressions.Material and methodsA reference model containing four implants was fabricated. Digital impressions were taken using an intraoral scanner with different scan bodies: original scan bodies for Group I (DO), CAD/CAM scan bodies without extensional structure for Group II (DC), and CAD/CAM scan bodies with extensional structure for Group III (DCE). For Group IV, conventional splinted open‐tray impressions (CI) were taken. The reference model and conventional stone casts were digitalized with a laboratory reference scanner. The Standard Tessellation Language datasets were imported into an inspection software for trueness and precision assessment. Statistical analysis was performed with a Kruskal–Wallis test and Dunn–Bonferroni test. The level of significance was set at α = .05.ResultsThe median of trueness was 35.85, 38.50, 28.45, and 25.55 μm for Group I, II, III, and IV, respectively. CI was more accurate than DO (p = .015) and DC (p = .002). The median of precision was 48.40, 48.90, 27.30, and 19.00 for Group I, II, III, and IV, respectively. CI was more accurate than DO (p < .001), DC (p < .001), and DCE (p = .007). DCE was more accurate than DC (p < .001) and DO (p < .001).ConclusionsThe design of the extensional structure could significantly improve scanning accuracy. Conventional splinted open‐tray impressions were more accurate than digital impressions for full‐arch implant rehabilitation.

Background The accuracy of digital impressions for fully edentulous cases is currently insufficient for routinely clinical application. To overcome the challenge, a modified scan body was introduced, which demonstrated satisfactory accuracy in vitro. The aim of this study was to evaluate the accuracy of digital impressions using the modified scan bodies with extensional structure versus scan bodies without extensional structure in mandible with two implants in beagle dogs. Methods The unilateral mandibular second premolar to second molar were extracted in four beagle dogs. Twelve weeks later, two implants were placed. Five repeated digital impressions were performed with an intraoral scanner on each dog using each of the two different scan bodies: Group I—scan body without extensional structure (SB); Group II—scan body with extensional structure (SBE). The scans were exported to Standard Tessellation Language (STL) files to serve as test data. The dogs were sacrificed and the dissected mandibles were digitalized with a lab scanner to provide reference data. Linear and angular deviations were calculated in an inspection software for accuracy assessment. Statistical analysis was performed with two-way ANOVA. The level of significance was set at α = 0.05. Results For trueness assessment, the mean of absolute linear/angular deviations were 119.53 μm/0.75 degrees in Group I and 68.89 μm/0.36 degrees in Group II. SBE was more accurate than SB regarding both linear (p = 0.008) and angular (p = 0.049) deviations. For precision assessment, the mean of absolute linear/angular deviations were 63.01 μm/0.47 degrees in Group I and 38.38 μm/0.24 degrees in Group II. No significant difference was found. Conclusions The application of SBE significantly improved the trueness of digital impressions in mandible with two implants compared to SB. No significant difference was found in terms of precision.