To characterize whether the bonding performance and fatigue strength of resin-dentin interfaces created by a universal adhesive would be affected by different H3PO4-application times to more accurately assess long-term durability. Mid-coronal flat dentin surfaces with standardized smear-layers were produced on sound third molars, etched with 32% H3PO4 for 0, 3 and 15 s, bonded with a mild universal adhesive (3M-ESPE) and restored with a nanofilled composite. Bonded specimens (0.9 × 0.9 mm) were stored in deionized water for 24 h and sectioned into beams for microtensile testing (n = 10). Resin-dentin beams were tested under tension until failure (0.5 mm/min) after 24 h or 6 month storage in artificial saliva at 37 ̊C. Bar-shaped resin-dentin beams (0.9 × 0.9 × 12 mm) were tested under 4-point-flexure initially at quasi-static loads (n = 22) and then under cyclic loads (n > 50). The stress-life fatigue behavior was evaluated using the twin-bonded interface approach by the staircase method at 4 Hz. Fractured interfaces and the tension side of unfractured beams were evaluated under SEM, along with the micro-morphology of the etched dentin surfaces and hybrid layers. Data were analyzed by ANOVA and Tukey test and Wilcoxon Rank Sum Test (α = 0.05). Quasi-static loads were limited to discriminate the bonding performance of resin-dentin interfaces. Application modes significantly affected etching patterns, fatigue strength, endurance limits and hybrid layer morphology (p < 0.001). Reductions in fatigue strength of self-etched bonded interfaces raise concerns about the true ability of universal adhesives to properly bond to dentin.

To determine whether the composition of universal adhesives and the use of silane coupling agents could affect the fatigue strength of composite repair. Composite samples were aged in water at 37 °C for 90 days and bonded to fresh composite to produce twin-bonded bar-shaped composite specimens (2 × 2 × 12 mm). Five universal adhesives, a multistep composite repair system and a hydrophobic solvent-free resin associated to a separate silane coupling agent application were used for bonding. Composite samples were tested under 4-pointflexure initially at quasi-static loading (n = 12) followed by cyclic loading (n = 25). The stress-life fatigue behavior was evaluated following the staircase method at 4 Hz. The unfractured side of cyclic loaded beams were evaluated under SEM to determine crack initiation sites. Fatigue data was analyzed by ANOVA and Tukey test and Wilcoxon Rank Sum Test (α = 0.05). Bonding protocols were unable to restore the cohesive strength of the nanofilled composite (p < 0.05). Fatigue testing was more discriminative to reveal discrepancies in composite repair than conventional quasi-static loading. While the composition of universal adhesives affected composite repair potential, the highest endurance limits occurred for the separate silane coupling agent application. Crack propagation sites were mostly located on the aged composite surface. Although a trend for simplification invariably overruns current adhesive dentistry, composite repair using solely universal adhesives may result in inferior repair potential. The additonal use of silane coupling agents remains as an important procedure in composite repairs.

To determine whether smear layer management, via conservative etching protocols, and the hydrophobic-rich content of hybrid layers would affect the fatigue strength of resin-dentin interfaces. Bar-shaped dentin beams obtained from sound third molars were wet-polished for 30 s. Dentin was etched with 32 % ortho-phosphoric acid for 3 or 15 s, 10 % meta-phosphoric acid for 15 s or by a prime-and-rinse application using a mild universal adhesive (Scotchbond Universal, 3M ESPE). Self-etch application served as control. Coating was performed with a solvent-free bisGMA-based resin. Composite buildups were made with a nanofilled composite. Resin-dentin beams with twin-bonded interfaces were sectioned and stored in deionized water for 24 h at 37 ℃ before 4-point flexural quasi-static monotonic testing (n = 16). Stress-life fatigue behavior was evaluated under cyclic loading (n = 35) by the staircase method at 4 Hz. The tension side of cyclic-loaded unfractured beams were evaluated under SEM, along with the micro-morphology of etched dentin surfaces. Monotonic data was analyzed by two-way ANOVA followed by the Tukey Test and cyclic-loaded data by Kruskal-Wallis on Ranks (α = 0.05). Etching protocols and higher hydrophobic-rich content produced significantly higher fatigue life distributions (p < 0.05). Dentin demineralization was ranked as OPA 15 s > MPA 15 s > OPA 3 s > P + R > SE. Less aggressive etching and coating reduced crack formation at hybrid layers. Current oversimplification trends in resin-dentin bonding constitute a trade-off between hybridization quality and easier adhesive handling. Controlled dentin etching and increasing the hydrophobic-rich content of hybrid layers may be necessary to extend the longevity of mild universal adhesives.