PURPOSE: To assess the influence of an additional shoulder preparation on the fracture strength of a cusp-replacing direct resin composite restoration in a premolar that previously had an amalgam MOD restoration followed by fracture of a cusp. MATERIALS AND METHODS: Two preparation designs were tested. In extracted sound premolars an MOD amalgam cavity was simulated followed by fracture of a cusp (Group A). Group B was the same as Group A but a shoulder was added along the cervical outline of the preparation. As a control group, cusp fracture was simulated but no MOD amalgam cavities were prepared. For each group, 14 preparations were made using a copy milling technique. All 42 preparations were restored using a direct resin composite restoration technique. Restorations were loaded until fracture. RESULTS: The fracture strength of the preparation with shoulder was not significantly different from the preparation without the shoulder. The fracture strength of the preparations in the reference group, which had no resistance form, was significantly less than in Groups A and B (ANOVA, P<0.05).
PURPOSE: To assess the influence of an additional shoulder preparation on the fracture strength of a cusp-replacing direct resin composite restoration in a premolar that previously had an amalgam MOD restoration followed by fracture of a cusp. MATERIALS AND METHODS: Two preparation designs were tested. In extracted sound premolars an MOD amalgam cavity was simulated followed by fracture of a cusp (Group A). Group B was the same as Group A but a shoulder was added along the cervical outline of the preparation. As a control group, cusp fracture was simulated but no MOD amalgam cavities were prepared. For each group, 14 preparations were made using a copy milling technique. All 42 preparations were restored using a direct resin composite restoration technique. Restorations were loaded until fracture. RESULTS: The fracture strength of the preparation with shoulder was not significantly different from the preparation without the shoulder. The fracture strength of the preparations in the reference group, which had no resistance form, was significantly less than in Groups A and B (ANOVA, P<0.05).