Survival of Dicor Glass-Ceramic Dental Restorations Over 20 Years: Part IV. The Effects of Combinations of Variables

Purpose: Previous studies have shown the relationship of individual clinical variables to the survival of Dicor (Corning Glass Works) restorations. The purpose of the present investigation was to examine the effect of combinations of these variables on the intraoral survival of Dicor restorations. Materials and Methods: Dicor glass-ceramic restorations (n = 1,444) were placed in 417 adult patients. Failure was defined as a restoration that required remake because of material fracture. The survival of restorations with different combinations of variables that were each individually associated with survival was described using Kaplan-Meier survivor functions. The statistical significance of differences in survival between different combinations of specific predictor variables was examined using the proportional hazards model. Results: Kaplan-Meier survival analysis indicated that significantly worse survival rates were found for restorations that included combinations of molar teeth, a dentin core, and a glass-ionomer luting agent; molar teeth, a dentin core, and a resin luting agent; and single-rooted teeth, a dentin core, and a glass-ionomer luting agent than for any other combinations tested. The Cox proportional hazards model described a hazard ratio of 3.37 (95% confidence interval [CI]: 2.23 to 5.08) for molar teeth (versus single-rooted teeth), 2.65 (95% CI: 1.44 to 4.87) for dentin core (versus gold core), 2.35 (95% CI: 1.58 to 3.51) for men (versus women), and 1.72 (95% CI: 1.13 to 2.60) for glass-ionomer luting agent (versus resin) after adjusting for the other variables in the model. Conclusion: Factors beyond individual restoration design impact the survival of Dicor glass-ceramic. These include sex, tooth position, and restorations luted to gold core foundation bases.