This study evaluated the effect of ceramic surface treatments on bond strength of ceramic brackets to machine-able ceramics and ceramic veneering metal. Machined ceramic specimens (10x10x1.5 mm) were prepared from Empress® CAD (EP), and e.max® CAD (EM). Ceramic veneering metal specimens (PF) were fabricated from sintered d.Sign® porcelain (1.27 mm thickness) over d.Sign®10 metal (0.23 mm thickness). Each ceramic was divided into 3-groups and treated surface by Er-YAG laser (LE) or etching with 9.6% HF acid for 5 seconds (A5) or 15 seconds (A15). Resin adhesive (Transbond?-XT) was used for attaching ceramic brackets for each group (n=15) and cured with LED (Bluephase®) for 50 seconds. Specimens were immersed in distilled water for 24 hours before testing for shear bond at crosshead speed of 1.0 mm/min. The data were analyzed for the differences in bond strength with ANOVA and Tukey?s multiple comparisons (? = 0.05). De-bond surfaces were microscopically examined. Bond strength (MPa) were 12.65±1.14 for EMA5, 14.50±2.21 for EMA15, 13.97±1.17 for EMLE, 12.40±1.95 for PFA5, 15.85±3.13 for PFA15, 14.06±2.17 for PFLE, 12.12±1.54 for EPA5, 15.65±1.57 for EPA15, 12.89±1.17 for EPLE. Significant differences in bond strength among groups were found related to surface treatment (p<0.05), but not significant difference upon type of ceramics (p>0.05). A15 provided higher bond strength than LE and A5 (P<0.05). No damage of ceramic surface upon de-bonding was indicated except for A15 tends to exhibit ditching. LE showed more uniform treated surface for bonding and no surface destruction upon de-bond compared to others. Bond strength was affected by surface treatment. Both LE and A15 treated surface provided higher bond strength than A5. Considering possibly inducing defect on ceramic surface, LE seems to provide better favorable surface preparation than others. Treated ceramic surface with Er-YAG prior to bracket bonding is recommended.