Fatigue survival and damage modes of lithium disilicate and resin nanoceramic crowns

Abstract Polymer-based composite materials have been proposed as an alternative for single unit restorations, due to their resilient and shock absorbing behavior, in contrast to the brittleness of ceramic materials that could result in failure by fracture. Objective: To evaluate the fatigue strength and damage modes of monolithic posterior resin nanoceramic and lithium disilicate glass ceramic crowns. Methodology: Twenty-six resin nanoceramic (RNC) and lithium disilicate glass ceramic (LD) 2 mm monolithic crowns (n=13) were cemented on composite resin replicas of a prepared tooth and subjected to cyclic load with lithium disilicate indenters for 2 million cycles. Specimens and indenters were inspected every 500,000 cycles and suspended when presenting fractures or debonding. Surviving specimens were embedded in epoxy resin, polished and subsurface damage was analyzed. Specimens presenting fractures or severe subsurface damage were considered as failures. Survival data was subjected to Fisher's exact test; damage modes were subjected to Mann-Whitney test (p<0.05). Results: There were no debonding, cohesive or catastrophic failures. Considering subsurface damage, 53.8% of RNC and 46.2% of LD crowns survived the fatigue test, presenting no statistical difference. Chief damage modes were radial cracks for RNC and inner cone cracks for LD, presenting no statistical difference. Conclusions: The results suggest that if debonding issues can be resolved, resin nanoceramic figures can be an alternative to posterior crowns. Although distinct, damage modes revealed potential to cause bulk fracture in both glass ceramic and resin nanoceramic crowns.


Introduction
Full crowns have been widely used to restore extensively damaged teeth. The classic crown consists of a bilayer restoration: a strong and stiff ceramic core veneered with aesthetic porcelain. The structural reliability of this combination of materials is primarily controlled by the properties of the core, 1 which provides stress-shielding of the veneer layer as well as of the underlying soft dentin support. 2 However, the main complications reported for bilayer restorations are chipping of the weak ceramic veneer. 3 Lithium disilicate (LD) glass ceramics present high flexural and fatigue strength, and fracture toughness [4][5][6][7] when compared to other glass ceramics. These important mechanical properties associated to excellent optical properties 8  For the posterior area, monolithic crowns have been proposed, since lithium disilicate optical properties exempt veneering ceramics in most cases. 9 In this approach, marginal and internal fit, occlusal and proximal contacts may be checked in a single visit, once core and veneer are merged into a monolithic restoration. Additionally, by eliminating the veneering ceramics, these crowns seem to exhibit higher fatigue strength, 10  nanoceramic fillers, specifically 20 nm silica particles, 4 to 11 nm zirconia particles and silica-zirconia nanoclusters, all embedded into a highly cross-linked polymeric matrix. Industrial manufacturing and additional curing of composites reduce the porosity and the amount of flaws, which seems to result in higher fatigue and flexural resistance in comparison to direct composites with conventional layering and curing processes. 12 This material presented high fatigue strength when compared to glass-ceramics, and apparently meets the mechanical requirements for high stress-bearing areas. 6,13,14 Despite these promising results, debonding cases were reported for composite crowns cemented on zirconia abutments 15,16 and the manufacturer opted to change the indications, limiting the material to partial crowns and veneers.
Although bond strength studies do not report problems on adhesion to RNC compared to ceramics, 17   Subsurface damage analysis revealed that inner cone cracks were the dominant crack system mechanism for LD crowns, occurring in 9 crowns.
In 5 of them, the inner cone crack reached the cementation surface, which would eventually result in crown fracture ( Figure 5). Two crowns presented radial Figure 2-B, C and D depict side views from a LD crowns polished through the entire damage area, as shown in A (occlusal view, 0.8x). In B it is possible to identify a crack that seems to originate from the cementation surface (filled arrow). In C, the crack is propagating further (filled arrow). Finally in D, considering the angle relative to the occlusal surface and the presence of another similar crack (outlined arrow), we concluded it was an inner cone crack extending to the cementation surface (B/C/D Magnification 2x)   In general, lithium disilicate indenters loading RNC crowns presented no damage. Indenters loading LD crowns presented discrete wear facets and removal of the glaze (Figures 11a and 11b). Minor cracks were detected after 500,000 cycles but were followed up during the entire test. One indenter presented a large crack and one presented cohesive fracture; both cracks started close to the fixture base and were not related to contact damage (Figures 11c and 11d).

Discussion
The null hypothesis that there would be no In the present study, most resin nanoceramic crowns presented contrasting outcomes: while radial cracks occurred in 5 from 13 crowns, other 5 crowns did not show any type of detectable damage.
Radial cracks penetrated both the cement layer and supporting composite (Figure 8), or propagated through the entire crown thickness (Figure 7), which would probably lead to bulk fracture. Surprisingly, two of these radial cracks occurred far from the indentation area (Figures 7 and 9), which suggests some lateral movement of the indenter.
With regards to flexural strength, lithium disilicate glass ceramics is capable of withstanding higher stress before failure when compared to resin composite 5   The materials tested presented different damage modes: resin nanoceramic seems to be more susceptible to flexure-induced radial cracks, while lithium disilicate crowns presented radial and inner cone cracks. Although distinct, both damage modes showed potential to cause failure by bulk fracture in monolithic LD and RNC crowns.