The Effect of Dentin Bonding and Material Thickness on the Flexural Properties of a Lithium-Disilicate Glass-Ceramic

Thanks to adhesive techniques and strengthened glass ceramics, ultrathin bonded occlusal veneers have been recently introduced. However, since a universally accepted thickness limit for ultrathin ceramics has yet to be established, their resistance to fracture needs to be better investigated. The purpose of this in vitro study was to evaluate the effect of dentin bonding on the flexural properties (ie, fracture load and flexural strength) of a lithium-disilicate (LD) glass ceramic when used in thicknesses equal to or less than the manufacturer's recommendations for occlusal restorations.

A total of 96 dentin slices (2.0 mm thick and 15 mm long) were obtained by sectioning bovine teeth along their long axes. LD slices of different thicknesses (1.5 mm/1.3 mm/1.0 mm/0.8 mm/0.6 mm) and 15 mm in length were cut from CAD/CAM LD blocks (IPS e.max CAD-C16). In each of 5 experimental groups, 16 dentin slices were adhesively luted to 16 LD slices (n = 16) of the same thickness, in order to create 16 bi-layered dentin-LD bonded assemblies. In the control group, the 16 remaining dentin slices were conventionally cemented to 1.5-mm-thick LD slices (n = 16) using a resin-modified glass-ionomer cement (FujiCEM 2). All dentin-LD assemblies were cut perpendicularly to their joint interface, in order to obtain 1-mm-wide, 15-mm-long bi-layered prismatic beams, having the following final thicknesses: for the 5 experimental groups, 2 mm (dentin layer) + 1.5 mm/ 1.3 mm/1.0 mm/0.8 mm/0.6 mm (LD layer); for the control group, 2 mm (dentin layer) + 1.5 mm (LD layer). All prismatic beams were subjected to a three-point bending test (14-mm span, load applied on the LD side). Fracture loads (N) and flexural strengths (MPa) were recorded. Data were analyzed using one-way ANOVA on ranks tests (α = 0.05). The correlations between the recorded flexural strengths and the dentin:LD thickness ratio and between the flexural strength and the luting strategy were also investigated. The failure modes were observed and classified.

No statistically significant differences were recorded between the conventionally luted control group (LD thickness 1.5 mm; fracture load 35.26 N; flexural strength 60.44 MPa) and the thinnest adhesively luted experimental group (LD thickness 0.6 mm; fracture load 28.97 N; flexural strength 90.01 MPa) in terms of fracture load and flexural strength. A fracture involving both the dentin and the LD of the bi-layered prismatic beam, but without any debonding between the LD and the dentin substrates of the broken specimen, was the most common failure mode observed on the adhesively luted samples.

Compared to conventional cementation, when LD is bonded to dentin, the flexural properties of the whole system are improved, and the two different substrates seem to behave like a single unit. Once adhesively luted, 0.6-mm-thick LD has the same fracture load and flexural strength as that of the conventionally luted 1.5-mm-thick LD.