Effect of brushing and thermocycling on the shade and surface roughness of CAD-CAM ceramic restorations

Statement of problem: The effects of toothbrushing (B) and thermocycling (TC) on the surface texture of different materials with various fabrication processes have been investigated. However, studies of computer-aided design and computer-aided manufacturing (CAD-CAM) ceramic restorations are limited. Purpose: The purpose of this in vitro study was to evaluate the effect of B and TC on the color stability and surface roughness of extrinsically characterized and glazed CAD-CAM ceramic restorations. Material and methods: Lithium disilicate CAD ceramic (n=90) and zirconia ceramic (n=90) were studied. All specimens were crystallized/sintered, characterized, and glazed following the manufacturer's recommendation. The specimens were divided into 9 different groups: B, TC, and a combination of B plus TC (B+TC). Brushing was performed at 50 000, 100 000, and 150 000 cycles, simulating an oral environment of 5, 10, and 15 years. Thermocycling was performed at 6000, 12 000, and 18 000 cycles, simulating an oral environment of 5, 10, and 15 years. Brushing plus TC was performed with the combination of the 50 000 cycles of B, then 6000 cycles of TC, and 10 000 cycles of B, then 12 000 cycles of TC, and 15 000 cycles of B, then 18 000 cycles of TC. The color and surface roughness of each specimen were measured before and after all interventions with simulated cycles. Color differences (ΔE) and surface roughness (ΔR_a) data were analyzed using 2-way ANOVA, followed by the least significant difference test (α=.05). The correlation between ΔE and ΔR_a was statistically analyzed using the Pearson correlation analysis. Results: Within the lithium disilicate CAD groups, intervention did not result in any significant differences in color change (P>.05). Within the zirconia groups, a 15-year clinical simulation revealed significantly higher ΔE values than a simulated 5-year exposure (P=.017). Increased simulated cycles showed significantly higher R_a values for all groups. Within the zirconia groups, B revealed significantly smoother surfaces than TC (P<.001) and B+TC interventions (P<.001). For the zirconia, simulating B+TC for15 years revealed significantly higher R_a values than the groups of B+TC for 5 years (P<.001) and B+TC for 10 years (P=.003). No correlation (lithium disilicate CAD, r=.079; P=.462; zirconia, r=.001; P=.989) was found between the color change and surface roughness. Conclusions: For both lithium disilicate CAD and zirconia, color changes were below the selected clinical perceptible threshold (ΔE=2.6) after all intervention and simulated cycles. All mean surface roughness measurements were below 0.2 μm. Generally, the surface of both lithium disilicate CAD and zirconia became rougher. No correlation was found between color difference and surface roughness for either material.