This study evaluated the influence of surface treatment on the shear bond strength of a composite resin (CR), previously submitted to the application of a temporary cement (TC), to an adhesive luting cement. Eight-four CR cylinders (5 mm diameter and 3 mm high) were fabricated and embedded in acrylic resin. The sets were divided into 6 groups (G1 to G6) (n=12). Groups 2 to 6 received a coat of TC. After 24 h, TC was removed and the CR surfaces received the following treatments: G2: ethanol; G3: rotary brush and pumice; G4: air-abrasion; G5: air-abrasion and adhesive system; G6: air-abrasion, acid etching and adhesive system. G1 (control) did not receive TC or any surface treatment. The sets were adapted to a matrix and received an increment of an adhesive luting cement. The specimens were subjected to the shear bond strength test. ANOVA and Tukey's tests showed that G3 (8.53 MPa) and G4 (8.63 MPa) differed significantly (p=0.001) from G1 (13.34 MPa). The highest mean shear bond strength values were found in G5 (14.78 MPa) and G6 (15.86 MPa). Air-abrasion of CR surface associated with an adhesive system provided an effective bond of the CR to the adhesive luting cement, regardless the pre-treatment with the phosphoric acid.