Polishing the bonding surface, before or after crystallization, does not alter the fatigue behavior of bonded CAD-CAM lithium disilicate

The aim of this study was to assess if the finishing/polishing of the bonding surface of lithium disilicate ceramic, prior to or after crystallization, would affect the fatigue behavior of a bonded restorations. For this, lithium disilicate ceramic (IPS e.max CAD) discs (n = 15) were milled and randomly divided into 3 groups: CAD-CAM group which remained untouched; PRE group which received a finishing/polishing protocol (OptraFine sys-tem) prior to its crystallization; and POST group, which received the treatment after its crystallization. After surface treatments, ceramic and glass-fiber reinforced epoxy resin discs were paired and bonded using a resin cement (Multilink N). A cyclical fatigue test was conducted (frequency 20 Hz, initial load 200 N for 5000 cycles, step-size of 100 N for 10,000 cycles/step) until failure occurrence. Surface roughness and topography were analyzed. An initial descriptive analysis of surface roughness, FFL and CFF was performed to obtain the mean, standard deviation and confidence interval values (SPSS v. 21, SPSS Inc.) for statistical analysis. Roughness data was using one-way ANOVA with Tukey's post hoc test (alpha = 0.05), while the fatigue data was submitted to survival analysis with Kaplan-Meier test (alpha = 0.05) and Weibull modulus (Weibull++, Reliasoft). Neither the finishing/polishing procedure of the bonding surface, nor the moment (prior to or after crystallization), affected the fatigue behavior of bonded milled lithium disilicate. There were also no differences for mechanical reliability among conditions. Despite this, finishing/polishing reduced surface roughness and led to smoother topography. Finishing/polishing the bonding surface of milled lithium disilicate, before or after crystallization, does not alter the fatigue behavior of the bonded restorative set, although there is some influence on roughness and topography.