The loss of resin cement adhesion to ceramic influences the fatigue behavior of bonded lithium disilicate restorations.

When partial and/or non-retentive preparation, such as those for occlusal veneers, is indicated, a proper and stable adhesion is essential. Therefore, the aim of this in vitro study was to evaluate the effect of loss of adhesion in different regions of the bonding interface on the fatigue behavior of simplified lithium disilicate restorations. For this, lithium disilicate (IPS e.max CAD) discs (1 mm thick and Ø = 10 mm) were fabricated, polished with #400-, #600-, #1200-grit silicon carbide (SiC) papers, and crystallized. As substrate, fiber-reinforced resin epoxy discs (2.5 mm thick and Ø = 10 mm) were fabricated and polished with #600-grit SiC paper. The ceramic bonding surface was treated with 5% hydrofluoric acid and a silane-containing primer (Monobond N), while the substrate was etched with 10% hydrofluoric acid followed by the application of the bonding system primers (Primer A + B). A lacquer (nail polish) was used to simulate the loss of adhesion in specific areas according to the study design to compose the testing groups: bonded (control; did not received nail polish application); - non-bonded (loss of adhesion in the whole specimen area); - margin (loss of adhesion in the ceramic margin); - center (loss of adhesion in the ceramic central area). The adhesive area of partially bonded groups was 50% of the adhesive surface. Then, the discs (n = 12) were bonded to the respective substrate using a resin cement (Multilink N), light-cured, water-stored for 90 days, and subjected to thermocycling (25,000 cycles, 5° to 55 °C) before testing. A cyclic fatigue test was run (20 Hz, initial load of 200 N for 5000 cycles, 50 N step size for 10,000 cycles each until specimen failure), and the fatigue failure load and number of cycles for failure were recorded. As complementary analysis, finite element analysis (FEA) and scanning electron microscopy analysis were performed. Kaplan-Meier log-rank (Mantel-Cox) was conducted for survival analysis. The results showed that as the loss of adhesion reaches the central area, the worse is the fatigue behavior and the higher is the stress peak concentration in the ceramic bonding surface. The bonded specimens presented better fatigue behavior and stress distribution compared to the others. In conclusion in a non-retentive preparation situation, proper adhesion is a must for the restoration fatigue behavior even after aging; while the loss of adhesion reaches central areas the mechanical functioning is compromised.