Comparison of the accuracy of implants placed with CAD-CAM surgical templates manufactured with various 3D printers: An in vitro study

The fit of a 3D printed surgical template will directly influence the accuracy of guided implant surgery. Various 3D printing technologies are currently available with different levels of resolution and printing accuracy; however, how the different systems affect accuracy is unclear. The purpose of this in vitro study was to assess the effect of using various 3D printers for the fabrication of implant surgical templates and its effect on the definitive implant position compared with the planned implant position. A cone beam computed tomography scan from a partially edentulous patient and an extraoral digital scan of a dental cast obtained from the same patient were used. The digital imaging and communications in medicine and standard tessellation language (STL) files were imported to an implant planning software program and merged, and an implant was digitally positioned in the mandibular right first molar region. A surgical template was designed and exported as an STL file. Ten surgical templates were printed for each of the following groups: stereolithography (SLA) printing, PolyJet, and MultiJet. The region where the implant was planned was cut away from the cast onto which the surgical templates were seated, allowing a passive positioning of the implant through the template, which was held in place with polyvinyl siloxane material. A scan body was inserted in the implant, and the cast was scanned with a laboratory scanner. The STL files obtained from the definitive implant position were imported into an implant planning software program and registered with the planned implant position, allowing for a comparison between the planned and actual implant position. Mean deviations were measured for angle deviation, entry point offset, and apex offset. Data normality was tested by using the Shapiro-Wilk test. The Kruskal-Wallis test was used to determine whether the outcomes of angle deviation, apex offset, and entry offset were statistically different between groups (α=.05). The median and interquartile range for the angle deviation (degrees) were 1.30 (0.62) for SLA; 1.15 (1.23) for Polyjet; and 1.10 (0.65) for Multijet. No statistically significant differences were found in the angular deviation among groups (χ2(2)=3.08, P=.21). The median and interquartile range for the entry offset and apex offset (mm) were 0.19 (0.16) and 0.36 (0.16) for SLA, respectively; 0.20 (0.13) and 0.34 (0.26) for Polyjet, respectively; and 0.23 (0.10) and 0.32 (0.08) for Multije