Ceramic inlays and partial ceramic crowns: influence of remaining cusp wall thickness on the marginal integrity and enamel crack formation in vitro.

No information is currently available about what the critical cavity wall thickness is and its influence upon 1) the marginal integrity of ceramic inlays (CI) and partial ceramic crowns (PCC) and 2) the crack formation of dental tissues. This in vitro study of CI and PCC tested the effects of different remaining cusp wall thicknesses on marginal integrity and enamel crack formation. CI (n = 25) and PCC (n = 26) preparations were performed in extracted human molars. Functional cusps of CI and PCC were adjusted to a 2.5 mm thickness; for PCC, the functional cusps were reduced to a thickness of 2.0 mm. Non-functional cusps were adjusted to wall thicknesses of 1) 1.0 mm and 2) 2.0 mm. Ceramic restorations (Vita Mark II, Cerec3 System) were fabricated and adhesively luted to the cavities with Excite/Variolink II. The specimens were exposed to thermocycling and central mechanical loading (TCML: 5000 x 5 degrees C-55 degrees C; 30 seconds/cycle; 500000 x 72.5N, 1.6Hz). Marginal integrity was assessed by evaluating a) dye penetration (fuchsin) on multiple sections after TCML and by using b) quantitative margin analysis in the scanning electron microscope (SEM) before and after TCML. Ceramic- and tooth-luting agent interfaces (LA) were evaluated separately. Enamel cracks were documented under a reflective light microscope. The data were statistically analyzed with the Mann Whitney U-test (alpha = 0.05) and the Error Rates Method (ERM). Crack formation was analyzed with the Chi-Square-test (alpha = 0.05) and ERM. In general, the remaining cusp wall thickness, interface, cavity design and TCML had no statistically significant influence on marginal integrity for both CI and PCC (ERM). Single pairwise comparisons showed that the CI and PCC of Group 2 had a tendency towards less microleakage along the dentin/LA interface than Group 1. Cavity design and location had no statistically significant influence on crack formation, but the specimens with 1.0 mm of remaining wall thickness had statistically significantly more crack formation after TCML than the group with 2.0 mm of remaining cusp wall thickness for CI. The remaining cusp wall thickness of non-functional cusps of adhesively bonded restorations (especially for CI) should have a thickness of at least 2.0 mm to avoid cracks and marginal deficiency at the dentin/LA interface.