The aim of this study was to evaluate the effect of different polishing protocols on surface roughness and morphology of lithium disilicate ceramics (IPS e.max CAD). Fifty ceramic bars (2.0x3.0x2.5 mm) were produced and randomly divided into five groups (n=10), according to surface treatments: C: Control; Cer: Ceramaster (60 s); CW: cloth wheel with large-grit diamond (30 s) + cloth wheel with fine-grit diamond paste (30 s); E: grey Exa-Cerapol (20 s) + pink Exa-Cerapol (20 s) + Cerapol Super (20 s); F: Felt wheel with large-grit diamond paste (30 s) + Felt wheel with fine-grit diamond paste (30 s). The samples were evaluated quantitatively for surface roughness (Ra) and qualitatively by SEM, at different moments: initial (prior to roughening), after roughening and after polishing procedures. Ra means were submitted to two-way ANOVA and Tukey test (p<0.05). After the different polishing protocols, no statistically significant differences were observed between the Cer and E groups (p=0.6347), E and CW groups (p=0.1366), and CW and FW groups (p=0.5433). Comparing different stages in the same group, it was possible to observe that CW (p<0.0001) and FW groups (p<0.0001) showed significant differences. SEM analysis revealed that both Cer and E groups exhibited a surface morphological pattern similar to that of control group. All polishing systems were effective in reducing the surface roughness of the studied ceramic type.