Abstract
The purpose of this work was to study the compatibility of several commercially available dentin adhesives with composite resins. In this study, V-shaped cavity preparations were created on both buccal and lingual surfaces of 60 extracted human premolars($3mm{\times}3mm{\times}2mm$ deep). Preparations were located such that the occlusal margins were on the enamel and the gingival margins were on the cementum(1mm below the CEJ). These specimens were randomly divided into three equal groups. Three dentin adhesives(Scotchbond Multi-Purpose, All Bond 2, Prisma Universal Bond 3)were applied to the cavity in accordance with each manufacturer's instructions. The teeth in each group were restored with four composite resins(Silux plus, Z100, Bisfil, Prisma TPH) in three increments, with each increment light-cured for 40 seconds. All specimens were exposed to 500 cycles of thermal stress. Specimens then placed in a silver nitrate solution(50% by weight) and stored in darkness for 24 hours. They were then immersed for 6 hours in photographic developing solution under flourescent light. Specimens were sectioned buccolingually through the center of the restoration and observed under a binocular stereoscope. To compare the marginal leakage, ANOVA and Dunkan's multiple range tests were used in analysis. Selected samples were further studied using scanning electron microscopy(XL20, Philips, The Netherlands). The results were as follows. 1. In all groups, the enamel margin showed significantly less leakage than the cementum margin. 2. No statistically significant differences were found in microleakage when composite resins were used with their proprietary dentin adhesives. 3. In comparison between dentin adhesives, Prisma Universal Bond 3 showed significantly less leakage at the enamel margin and Scotch bond multi-purpose showed significantly less leakage at the cementum margin than the other groups. 4. In comparison between composite resins, Silux Plus showed significantly more leakage than other groups at the enamel margin and no statistically significant differences were found in microleakage at the cementum margin. 5. According to the backscattered scanning electron microscope, microleakage occurred via gaps at the dentin adhesives/dentin interface.