Removal of noncollagenous components affects dentin bonding.

The structural integrity of fibrillar type I collagen is critical for effective dentin bonding. Since most noncollagenous matrix components in dentin are closely associated with collagen, we hypothesized that they may also contribute to dentin bonding. To test this hypothesis, bovine dentin was acid-etched, treated with chondroitinase ABC (C-ABC), endo-beta-galactosidase (Endo-beta), or trypsin. Controls were prepared in the same manner but without the enzymes. All control and experimental specimens were then bonded with One-Step. Bond strength data were analyzed by one-way ANOVA and Fisher's PLSD test (p < 0.05). When dentin was treated with C-ABC or trypsin, bond strengths significantly decreased for the rewetted groups (p < 0.05). The treatment with Endo-beta showed no effects on bond strengths (p > 0.05). When the treated dentin surfaces were observed under SEM, the C-ABC and trypsin treated groups revealed significant loss of collagen fibril architecture. The results indicate that chondroitin sulfate glycosaminoglycans and trypsin-digestible noncollagenous proteins play roles in maintaining the open dimensions of the collagen fibril scaffold, which is essential for optimal dentin bonding.

Effect of thermal and mechanical load cycling on microtensile bond strength of a total-etch adhesive system.

To evaluate the effect of thermal and mechanical cycles on dentin bond strength to cervical margins of Class II restorations, 80 box-type Class II cavities were prepared on the surfaces of bovine incisors. The cavities were restored with Single Bond (3M-ESPE) and Z-250 composite (3M-ESPE) according to manufacturer's instructions. The incisors were divided into four groups: G1-Control, G2- Thermal cycling (2,000 cycles, 5 degrees C-55 degrees C), G3- Mechanical cycling (100,000 cycles; 50N) and G4- Thermal and mechanical cycling (2,000 cycles 5 degrees C-55 degrees C/100,000 cycles; 50N). The restorations were sectioned perpendicular to the cervical bonded interface into 0.7 +/- 0.2 mm-thick slabs. The slabs were further trimmed at the interface to 1.4 +/- 0.2 mm with a fine diamond bur to produce a cross-sectional surface area of 1 mm2. All specimens were then subjected to microtensile bond testing. Means and standard deviations were expressed in MPa. The bond strength data were analyzed by one-way ANOVA and Fisher's PLSD test (p<0.05). Fracture mode analysis was performed using SEM. Bond strengths were significantly lower when thermal and mechanical cycling were performed [G4-2.41 (8.57)] when compared to the other groups [G1-28.15 (14.03); G2-27.60 (10.14); G3-27.59 (8.67)]. No differences were observed among Groups 1, 2 and 3. Interfacial fracture of the control (G1) and thermocycling (G2) groups mainly occurred between the deepest portion of the adhesive resin and the top layer of the demineralized dentin (Interphase). Mixed failure was predominant and increased when mechanical cycling was applied (G3 and G4).