The aim of this investigation was to assess the influence of several surface treatments on the shear bond strength (SBS) of a self-adhesive resin cement containing 10-methacryloxydecyl-dihydrogenphosphate monomer to densely sintered zirconia ceramic, before and after thermal cycles. Hundred densely sintered zirconia cylinders were divided into five groups (n = 20). Each of them received a different surface treatment: (1) control [No_T], with the zirconia surface unconditioned, (2) low pressure air abrasion [Sand_S], (50 μm, 1 bar), (3) standardized air abrasion [Sand_H], (50 μm, 2.8 bar), (4) standardized Rocatec™ Plus (silica-coated alumina oxide) air abrasion (2.8 bar) and silanization [Roc_H], (5) low pressure Rocatec™ Plus (silica-coated alumina oxide) air abrasion (1 bar) and silanization [Roc_S]. Five more surface-treated specimens were addressed to scanning electron microscope for qualitative observations. After specimen fabrication, subgroups of 10 bonded samples were stored in water either for 24 h (T1) or subjected to 5000 thermal cycles (T2); SBSs were determined with a universal testing machine at a crosshead speed of 1 mm/min. At T1, mean SBSs (MPa) obtained for the examined groups were: [Sand_H] 16.24 ± 2.95; [Sand_S] 16.01 ± 2.68; [Roc_H] 17.17 ± 1.64; [Roc_S] 15.92 ± 1.99. All surface treatments positively affected (p T2, SBS values ranged from 7.76 ± 2.37 (No_T) to 8.89 ± 1.74 (Sand_S), with no statistically significant difference between groups (p = 0.5293). Both air abrasion with alumina oxide and Rocatec™ universal bonding system, used with hard or low air pressure, produced comparable effects on cement–zirconia interface before and after thermal cycles. After artificial aging, minimal differences in bond strength values between sandblasted and control groups were not of statistical significance.