The effect of stiffness on the localization of tensile stress at the surface of bonded posterior restorations.

OBJECTIVES The aim of this study is to assess the distribution of tensile stress under the conditions representative of tooth-food contact in a restored posterior tooth. The ideal stiffness for a bonded restoration will be predicted. METHODS A two-dimensional plane-strain finite element representation of a restored first maxillary molar is created. The restoration with Class I/II bucco-lingual geometry is assigned a range of Young's moduli (E = 10-80 GPa), representative of the range of materials available. In addition, a hypothetical multi-phase model, in which stiffness is gradually increased from the intercuspal concavity to the adjacent enamel, is also created. A food particle is modeled and pushed close to the interface onto the site of (1) intact enamel and (2) restored surface. RESULTS For all single-phase models tested, the magnitude of tensile stress at the surface is far greater than that present elsewhere in the tooth. However, the exact position and magnitude of the maximum tensile stress varied according to the modulus assigned to the restoration and the position of load. The results of the model imply that interfacial problems are likely in low-modulus restorations (E = 10-20 GPa), whereas stress-related problems could occur in the intercuspal concavity of high-modulus restorations (E = 40-80 GPa). Of all the single-phase models, tensile stresses are lowest when the Young's modulus assigned to the restoration is 30 GPa. These tensile stresses are reduced in the multi-phase model. SIGNIFICANCE Given the limitations of the model, the results indicate that the most suitable modulus for a single-phase posterior bonded restoration is around 30 GPa. Such a modulus is approached in compact-filled composites. In addition, the highly desirable dissipation of load in the multi-phase model should warrant further investigation. It is suggested that the use of an incremental filling technique with region-specific proportions of hard-filler could be one way forward.