RESUMO
PURPOSE: To investigate in vitro the retention and the resistance form, as well as the failure modes of maxillary premolars restored with cast metal crowns and different core materials. MATERIALS AND METHODS: Sixty human extracted maxillary premolars were selected according to their size and were embedded in PMMA resin blocks. After removing a part of their clinical crowns, the teeth were randomly divided into 3 groups of 20 teeth and were either left unrestored, or they were restored with amalgam or composite resin. All teeth were prepared for a cast metal complete coverage restoration. The restorations were cemented on the prepared teeth with a resin-modified glass ionomer luting agent (GC Fuji Plus). All specimens were subjected to static loading at 1 mm/min by a universal testing machine, until failure. Half the specimens of each group were subjected to tensile loading along the long axis of the teeth. The other half were subjected to compressive loading at a 30° angle. Failure loads and failure modes for each tooth were recorded. The statistical analysis included descriptive statistics, one-way ANOVA, and Tukey's HSD test. RESULTS: One-way ANOVA revealed statistically significant differences among the 3 tested groups (p < 0.05) for both tests. The group of teeth with no core material presented the highest failure loads for both the tensile and the compressive loading tests, with mean loads of 381.02 Ν and 741.21 Ν, respectively. Mean tensile and compressive failure loads for the amalgam group were 277.34 Ν and 584.75 Ν, while the composite resin group presented the lowest tensile and compressive failure values, which were 250.77 Ν and 465.78 Ν, respectively. The compression loading test resulted in the same failure mode for all specimens, which included unfavorable fracture of the teeth in combination with detachment of the cast metal complete coverage restorations. The tensile loading test resulted in different failure modes between the groups that used a core material and the group with no core material. CONCLUSIONS: Teeth that lost more than half of their coronal structure presented better retention and resistance form if no core material was used, provided that a minimum of 2-mm axial wall height was present at the missing part. Teeth that have lost more than half of their coronal structure and were restored with amalgam core presented better retention and resistance form than those restored with composite resin. Catastrophic fractures, extending to the root, were associated with compression forces but not with tensile forces.