Effects of cross-sectional area on resin-enamel tensile bond strength.

OBJECTIVE: It was hypothesized that there is an inverse relationship between resin-enamel bond strength and bonded cross-sectional area, and that there are regional differences in resin-enamel bond strength. METHODS: The facial and lingual surfaces of extracted human third molars were ground down 0.3 mm using 240 grit abrasive paper and were then bonded with either Clearfil Liner Bond 2 or Scotchbond Multi-Purpose Plus adhesive systems using the manufacturer's instructions. The bonded surfaces then received a resin composite build-up. After 24 h of storage in water, the bonded teeth were vertically serially sectioned into 1.0 mm thick slabs using a diamond saw, and the bonded surface area at the resin-enamel interface was varied from 0.5 to 3.0 mm2 using a diamond saw under microscopic observation. The trimmed region was varied from the occlusal third of the facial or lingual enamel to the middle third, to the cervical third. The trimmed specimens were then glued to a Bencor Multi-T device, placed in an Instron testing machine and stressed to failure at 1 mm/min. A three-factor ANOVA was used to compare bond strengths (buccal vs. lingual, occlusal vs. middle vs. cervical-third, vs. materials). Regression analysis was used to examine the relationship between bond strength and bonded cross-sectional area for each material on occlusal enamel. RESULTS: For both bonding systems, there was a highly significant (p < 0.001) inverse exponential relationship between tensile bond strength (y axis) and bonded cross-sectional area (x axis) with y intercepts of 51 and 59 MPa for Clearfill Liner Bond 2 and Multi-Purpose Plus, respectively. Using both materials, the highest bond strengths were measured in the occlusal third, which were significantly higher (p < 0.05) than those made to cervical enamel. SIGNIFICANCE: Like resin-dentin bonds, resin-enamel bonds exhibit an inverse relationship with cross-sectional area. This relationship becomes more apparent at bonded surface areas below 2 mm2 and is probably due to reductions in the number of interfacial stress-raisers as samples are made smaller.