RESUMO
PURPOSE: Previous studies considering retention of cast metal restorations to implant abutments incorporated some degree of frictional fit due to internal surface nodules and roughness of the restoration. In comparison, CAD/CAM restorations have minimal surface irregularities, possibly impacting retention. There is insufficient knowledge of retentive force of CAD/CAM restorations to titanium abutments, and therefore the topic warrants further investigation. This in vitro study investigated the retention of all-ceramic CAD/CAM restorations to three different prefabricated implant abutments using five different cements. MATERIALS AND METHODS: A total of 150 Astra Tech dental implant abutments were used, with each group of 50 being subdivided into five groups of 10. An optical impression of each size of abutment was made with the CEREC 3D intraoral camera. A full-coverage restoration was designed and milled with an enlarged, conical-shaped occlusal surface, which served to secure the restoration into a brass jig used with a universal testing machine. Five different cements were used with three different-sized abutments. Following cementation, the implant/abutment/restoration assemblies were stored for 24 hours at 37°C in 100% humidity. A pull-out test using a universal testing machine, set at a 0.5 mm/min crosshead speed, was used to evaluate retention of the individual restorations. The load required to remove each all-ceramic restoration was recorded. Retention values were analyzed using ANOVA and Fisher's PLSD multiple comparisons test at the 0.05 level of significance. RESULTS: Peak loads for two provisional cements and a resin-modified glass ionomer cement ranged from 56 N to 127 N. Peak loads for two resin cements ranged from 184 N to 318 N. Two-way ANOVA showed significant effects upon retentive forces for both the cement and abutment design. Post hoc Fisher's PLSD multiple comparisons test found significant differences in retention for 7 of the 10 pairings of cements at a 0.05 level of significance. In addition, Fisher's PLSD multiple comparisons test found significant differences between Astra Tech Direct Abutments 4 and Astra Tech Direct Abutments 5 as well as Astra Tech Direct Abutments 4 and Astra Tech Direct Abutments 6 at a 0.05 level of significance. No significant difference was found between Astra Tech Direct Abutments 5 and Astra Tech Direct Abutments 6. CONCLUSIONS: Of the five cements tested, the most retrievable CAD/CAM restorations were luted with Temp Bond NE and Improv Temporary Cement. Resin-modified glass ionomer retentive forces were closer to those of the "temporary cements" than those of the permanent adhesive-resin cements. The abutment surface area became less important when using adhesive-resin cements. Retention of CAD/CAM all-ceramic restorations to prefabricated abutments has not been reported in the literature. This in vitro study demonstrated clinically significant variation among the selected cements used to retain all-ceramic CAD/CAM restorations to implant abutments. In addition, abutment size influenced the retention of all-ceramic CAD/CAM restorations.