Vertical marginal discrepancy of ceramic copings with different ceramic materials finish lines and luting agents: An in vitro evaluation

Statement of problem. Prosthetic restorations that fit poorly may affect periodontal health and occlusion. Studies that have evaluated the accuracy of fit of ceramic restorations before and after cementation assessed primarily intracoronal restorations.Purpose. This in vitro study evaluated the effect of different finish lines, ceramic manufacturing techniques, and luting agents on the vertical discrepancy of ceramic copings.Material and methods. Two stainless steel molars were prepared for complete crowns with 2 different finish lines (heavy chamfer and rounded shoulder); each molar was duplicated to fabricate 90 copings. A total of 180 copings generated 18 groups (n=10 for each finish line-coping material-luting agent combination). Luting agents tested included zinc phosphate, resin-modified glass ionomer (Fuji Plus), and resin composite cements (Panavia F). A metal frame was developed on which to screw the stainless steel model and a ceramic coping; the distance (mum) between 2 predetermined points was measured before and after cementation by a profile projector under a torquing force. A 4-way ANOVA with repeated measurements was performed to assess the influence of each factor in the vertical marginal discrepancy: 3 between-coping factors (finish line-coping material-luting agent) and 1 within-coping factor (before and after cementation) (alpha=.05).Results. Procera copings presented the lowest mean values (P<.05) of vertical marginal discrepancy before and after cementation (25/44 mum) when compared to Empress 2 (68/110 mum) and InCeram Alumina copings (57/117 mum), regardless of any combinations among all finish lines and luting agents tested.Conclusion. Considering each factor separately, the ceramic manufacturing technique appeared to be the most important factor tested for the definitive vertical discrepancy of all-ceramic copings, with lower mean values for Procera copings.