Accuracy of digitally fabricated drilling guide to form screw-access channels in cement-retained implant prostheses: A randomized clinical trial

Statement of problem: Accurate intraoral preparation of screw channels in cement-retained implant prostheses could be a treatment challenge for tilted or badly positioned implants. Purpose: The purpose of this randomized clinical trial was to evaluate screw channel accuracy for retrieving cement-retained implant prostheses by using a digitally fabricated drilling guide. Material and methods: Twenty-four participants requiring removal of cement-retained implant prostheses were recruited and consecutively assigned to 1 of 2 groups. In the first group, drilling guides were fabricated by using an intraoral scan, computer-aided design (CAD), and 3D printing to facilitate screw channel preparation (guide group). In the second group, screw channel preparation was based on panoramic radiographs without a drilling guide (freehand group). Screw channel accuracy was evaluated with a 3D analysis and subsequently compared with the ideal shape of the screw channel for linear deviation at coronal and apical levels, angular deviation, surface loss deviation, and volume loss deviation. The ideal shape of the screw-access channel was based on the Ø2.2-mm cylinder image marked along the long axis of the implant. The Mann-Whitney U and Kruskal-Wallis tests were used for statistical analyses (α=.05). Results: The drilling guide group showed higher dimensional accuracy than the freehand group for angular deviation (9.45 degrees versus 15.69 degrees; P=.014) and surface loss deviation (7.40 mm versus 9.53 mm; P=.001), especially when the implant was tilted more than 30 degrees. The deviation values of the screw channels were not significantly different according to implant tilt in either group (P>.05). Conclusions: Digitally fabricated drilling guides for intraoral preparation of screw channels in cement-retained implant prostheses can improve accuracy and provide a less-destructive screw channel.