Abstract
Statement of problem. Glass-ceramics are often selected for use in dental restorations based upon advertised flexural strengths obtained from standardized tests on prefabricated specimens (bars and disks); these may not accurately reflect their performance in dental applications. Purpose. The purpose of this in vitro study was to determine and compare 4-point flexural strength, biaxial flexural strength, hardness, and crown failure loads for 3 commercially available glass-ceramics. Material and methods. Specimens were pressed and prepared from 3 brands of glass-ceramics: Celtra (CEL), IPS e.max (EMA), and Lisi (LIS). Rectangular bars, circular disks, and fully contoured crowns were created (n=15 specimens per glass-ceramic, 45 specimens per geometry, totaling 135 specimens). Disks were tested for biaxial flexural strength by using a piston-on-3-ball (POB) test, while bars were tested for 4-point flexural strength by using 4-point bending (4PB) and Vickers hardness (VH) tests. Crown failure loads were assessed in "crunch-the-crown " (CTC) tests. The results were analyzed by using general linear modeling, the Pearson correlation coefficient, and Weibull analysis. Results. The general linear modeling revealed significant differences (P <.05) in the failure load for crown specimens (EMA > LIS > CEL), the 4-point flexural strength (EMA > LIS > CEL), and the biaxial flexural strength (EMA > LIS=CEL). The disk specimens had higher flexural strengths than the bar specimens for CEL and EMA materials. LIS had a higher Weibull modulus than EMA and CEL for bar and crown specimens. CEL had a higher Weibull modulus than LIS and EMA for disk specimens. There was no correlation among the VH (R-2=0.86 and P=.24), biaxial flexural strength (R-2=0.84 and P=.26), and crown failure load. However, there was a high correlation between the failure load (crown specimens) and 4-point flexural strength (bar specimens) (R2=0.99 and P=.03). Conclusions. The 4-point flexural strength correlated significantly with crown failure load.