Effect of Implant Analog Design and Additively Manufactured Casts’ Printing Layer Thickness on the Linear and Angular Accuracy of Analogs for Direct Digital Workflow: An In Vitro Study

Purpose: To evaluate how implant analog design and printing layer thickness can affect the linear and angular accuracy of implant analogs in additively manufactured casts compared to conventional implant analogs in stone casts. Materials and Methods: A reference cobalt chromium mandibular model with a single implant was digitized using an industrial optical scanner and scan bodies that were compatible with a pressure/friction fit (S) or a screw retained (N) implant analog for direct digital workflow. These scans were used to fabricate casts with 50-µm (S-50 and N-50) (n = 10) and 100- µm (S-100 and N-100) layer thickness (n = 10). A total of 10 stone casts were made after single-step closed-tray polyvinyl siloxane impressions of the model (CNV) were made. All casts were digitized with the same metal scan body and scanner used to digitize the master model. These scans were then superimposed over the scan of the master model to measure the linear (x, y, and z- axes) and angular (XY and YZ planes) deviations (Geomagic Control X). The precision of measured deviations was defined with the average deviation values. Generalized linear model analysis was used to compare the deviations within implant analogs for direct digital workflow, while a one-way analysis of variance (ANOVA) and Dunnett’s test were used to compare these analogs and conventional analogs (α = .05). Results: The analog design affected the linear deviations (y-axis), while the interaction between the analog design and the layer thickness affected the angular deviations (XY plane, P ≤ .030) of the analog design. S analogs had lower linear and angular deviations than N analogs, and S-50 led to lower angular deviations than N-50 (P ≤ .030). CNV led to higher linear accuracy (y-axis) than N-50, N-100, and S-100 and led to lower angular deviations than all test groups (XY plane) (P ≤ .025). Conclusions: The analogs in S-50 casts had positional trueness similar to or higher than those in other test groups, and their accuracy was mostly similar to those in CNV casts. Implant analogs for direct digital workflow deviated more toward lingual and gingival, and conventional analogs deviated more toward buccal, occlusal, and distal. All analogs had a tendency to tilt toward the lingual and distal directions.