Purpose: To evaluate the effect of different finishing protocols on the wear behavior of lithium disilicate glass-ceramics (LD). Materials and Methods: Specimens were produced from LD prefabricated CAD/CAM blocks and divided into three groups according to the surface treatment (n = 8): control, polishing, and glaze. Ceramic specimens were subjected to wear testing using a dual-axis chewing simulator. A 49-N load was applied in the axial direction combined with a lateral movement (1-mm path) using an LD spherical piston for a total of 10⁶ cycles. Qualitative analysis of the wear surface was performed using an optical microscope. Quantitative analysis of surface roughness and volume loss was performed using a confocal microscope and a 3D-image editing software, respectively. Surface roughness and volume loss data were analyzed using Friedman’s nonparametric statistical test for repeated measures and the Student-Newman-Keuls test (α = .050). Results: There were statistical differences for surface roughness and volume loss of LD specimens in the different experimental conditions (P < .001). The control and polishing groups showed similar surface roughness and volume loss values for all testing times. The glaze group had greater wear volume after 10³, 10⁴, and 10⁵ cycles. After 10⁶ cycles, surface roughness and volume loss were similar in all groups. For the piston, surface roughness was similar over time and among groups. Conclusions: A distinct wear behavior was found for glazed glass-ceramic specimens compared to control and polished specimens. At the end of the simulation, the surface roughness and volume loss was similar for the groups.