SUMMARYPurpose: This study compared the microleakage of an experimental low-shrinkage resin composite (Hermes), a nanofilled resin composite material (Filtek Supreme) and a hybrid resin composite (Tetric Ceram) using a dye penetration method. Methods and Materials: Class I cavities prepared in 60 human molars were randomly divided into 3 groups according to the restorative material used. The preparations were restored using a bulk or an incremental technique. Half of the specimens from each group were subjected to 200,000 cycles of loading at 50 N, while the other half were stored in distilled water for 24 hours at 37°C. All specimens were immersed in 1% methylene blue (pH = 7.0) for 24 hours and sectioned into 3 slabs. The margins were evaluated for microleakage using an ordinal scoring system (0-4) under a stereomicroscope at 40x magnification. Data were subjected to the non-parametric Kruskal-Wallis analysis of variance and Mann-Whitney test (p<0.05). Data were expressed as median leakage scores and mean ranks.Results: All of the restorative systems had microleakage, regardless of the insertion technique and mechanical load cycling. Incremental placement significantly reduced microleakage as compared to the bulk technique, regardless of the restorative system used. Load cycling significantly affected incrementally placed restorations, except for the Hermes system.

The present study evaluated the long-term microtensile bond strength of collagen-depleted dentin followed by cyclic loading. Bovine flat, midcoronal dentin received one of the following surface treatments: acid-etch or acid-etched + 5% NaOCl for 2 min. The teeth were restored with Single Bond, Scotchbond Multi-purpose, One-Step Plus, or All-Bond 2 adhesive systems. Half of the specimens were randomly assigned to receive 200,000 cycles (50 N force). Teeth were sectioned into 1 x 1 mm thick slices and stored for 24 h, 1 month, 3 months, or 6 months. Specimens were subjected to tensile testing after elapsed storage time. Samples were analyzed by three-way ANOVA and Fisher's PLSD (p < 0.05). The results showed that the deproteinized groups without cyclic loading presented lower bond strength when compared with the control group, but the difference was only statistically significant for Single Bond and All-Bond 2 (p < 0.05). All adhesives presented a decrease in bond strength over time, regardless of the dentin treatment. When cyclic loaded, the decrease in bond strength for the deproteinized group was even greater when compared with the control group (p < 0.05). The results suggest that the hybrid layer is important as a stress-buffering layer when loading is applied, and thus the presence of collagen is essential.

No abstract