RESUMO
STATEMENT OF PROBLEM: More data are needed on the influence of preparation design on the fracture strength, failure type, repairability, and polymerization-induced cracks of molar teeth restored with direct composite resin restorations. PURPOSE: This in vitro and finite element analysis study investigated the effect of different preparation designs on fracture strength, failure type, repairability, tooth deformation, and the formation of polymerization-induced cracks of compromised molars restored with direct composite resin restorations. MATERIAL AND METHODS: Human molars (n=64) were randomly assigned to 4 different preparation designs: undermined inlay (UI), extended inlay (EI), restricted overlay (RO), and extended overlay (EO). The teeth were restored using direct composite resin and subjected to artificial thermomechanical aging in a mastication simulator, followed by load-to-failure testing. Three-dimensional (3D) finite element analysis was conducted to assess tooth deformation. Polymerization-induced cracks were evaluated using optical microscopy and transillumination. The fracture strength data were analyzed using a Kruskal-Wallis test, while the failure mode, repairability, and polymerization cracks were analyzed using the Fisher exact test (α=.05). RESULTS: All specimens withstood thermomechanical aging, and no statistically significant difference in fracture strength was observed among the 4 preparation designs (P>.05). The finite element analysis showed differences in tooth deformation, but no correlation was observed with in vitro fracture resistance. The RO and EO groups presented significantly more destructive failures compared with the UI and EI groups (P<.01). The RO group had significantly fewer repairable failures than the UI and EI groups (P=.024). A correlation was found between higher frequencies of repairability and higher tooth deformation. A significant correlation between the increase in microfractures and preparation design was observed (P<.01), with the UI group exhibiting a higher increase in microfracture size compared with the EO group (P<.05). CONCLUSIONS: No influence of preparation design on the fracture strength of compromised molars restored with direct composite resin restorations was evident in this study, but the failure mode of cusp coverage restorations was more destructive and often less repairable. The finite element analysis showed more tooth deformation in inlay preparations, with lower stresses within the root, leading to more reparable fractures. Since cusp coverage direct composite resin restorations fractured in a more destructive manner, this study suggests that even a tooth with undermined cusps should be restored without cusp coverage.
