Accuracy, reproducibility, and dimensional stability of additively manufactured surgical templates

Statement of problem. Additively manufactured surgical templates are commonly used for computer-guided implant placement. However, their accuracy, reproducibility, and dimensional stability have not been thoroughly investigated with the different 3D printers and materials used for their fabrication. Purpose. The purpose of this in vitro study was to evaluate the accuracy, reproducibility, and dimensional stability of additively manufactured surgical templates fabricated by using different 3D printers. Material and methods. Thirty surgical templates were designed and additively manufactured from 3 different 3D printers as follows: group SLA (n=10) was fabricated by using a desktop stereolithography (SLA) 3D printer and photopolymerizing resin; group PolyJet (n=10) was fabricated by using a PolyJet 3D printer and photopolymerizing resins; and group DMP (n=10) was fabricated by using a direct metal printing (DMP) system and Co-Cr metal alloy. All surgical templates were scanned by using a laser scanner within 36 hours of production and digitalized again 1 month later. All scanned files were compared with the corresponding designed files in a surface matching software program. The mean deviation root mean square (RMS, measured in mm, representing accuracy), percentage of measurement data points within 1 standard deviation of mean RMS (in %, representing reproducibility), and dimensional changes were determined and compared. Results. At the postproduction stage, group PolyJet was most accurate with the lowest RMS value of 0.10 +/- 0.02 mm and highest reproducibility with 93.07 +/- 1.54% of measurement data points within 1 standard deviation of mean RMS. After 1-month storage, group PolyJet((1month)) remained the most accurate with the lowest RMS value of 0.14 +/- 0.03 mm and the highest reproducibility value of 92.46 +/- 1.50%. For dimensional stability, group SLA versus group SLA((1month)) comparison showed a significant decrease in accuracy (RMS values of 0.20 +/- 0.08 mm versus 0.25 +/- 0.08 mm, P<.001) and reproducibility (88.16 +/- 3.66% versus 86.10 +/- 4.16%, P=.012). Group PolyJet versus group PolyJet((1month) )comparison only showed significant changes in accuracy (RMS values of 0.10 +/- 0.02 mm versus 0.14 +/- 0.03 mm, P=.011). Group DMP versus group DMP(1month) comparison showed no significant changes in accuracy (RMS values of 0.19 +/- 0.03 mm versus 0.20 +/- 0.04 mm, P=.981) or reproducibility (89.77 +/- 1.61% versus 89.74 +/- 2.24%, P=1.000). Conclusion. Printed resin surgical templates produced by using the PolyJet 3D printer showed higher accuracy and reproducibility than those produced by using the desktop SLA 3D printer and printed Co-Cr surgical templates at both the postproduction stage and after 1-month storage. The level of accuracy and reproducibility in printed Co-Cr surgical templates was not affected by 1-month storage.