RESUMO
PURPOSE: The aim of this study was to evaluate the effectiveness of fabrication methods (computer-aided design/computer-aided manufacture [CAD/CAM], copy-milling, and conventional casting) in the fit accuracy of three-unit, screw-retained fixed dental prostheses. MATERIALS AND METHODS: Sixteen three-unit implant-supported screw-retained frameworks were fabricated to fit an in vitro model. Eight frameworks were fabricated using the CAD/CAM system, four in zirconia and four in cobalt-chromium. Four zirconia frameworks were fabricated using the copy-milled system, and four were cast in cobalt-chromium using conventional casting with premachined abutments. The vertical and horizontal misfit at the implant-framework interface was measured using scanning electron microscopy at ×250. The results for vertical misfit were analyzed using Kruskal-Wallis and Mann-Whitney tests. The horizontal misfits were categorized as underextended, equally extended, or overextended. Statistical analysis established differences between groups according to the chi-square test (α = .05). RESULTS: The mean vertical misfit was 5.9 ± 3.6 µm for CAD/CAM-fabricated zirconia, 1.2 ± 2.2 µm for CAD/CAM-fabricated cobalt-chromium frameworks, 7.6 ± 9.2 µm for copy-milling-fabricated zirconia frameworks, and 11.8 (9.8) µm for conventionally fabricated frameworks. The Mann-Whitney test revealed significant differences between all but the zirconia-fabricated frameworks. A significant association was observed between the horizontal misfits and the fabrication method. The percentage of horizontal misfits that were underextended and overextended was higher in milled zirconia (83.3%), CAD/CAM cobaltchromium (66.7%), cast cobalt-chromium (58.3%), and CAD/CAM zirconia (33.3%) frameworks. CONCLUSION: CAD/CAM-fabricated frameworks exhibit better vertical misfit and low variability compared with copy-milled and conventionally fabricated frameworks. The percentage of interfaces equally extended was higher when CAD/CAM and zirconia were used.
