Background As there is no standard method for re-cementing debonded partial ceramic restorations, the aim of this study was to evaluate the use of a non-invasive thermal protocol for cleaning and retreatment, and to study its influence on shear bond strength. Material and Methods Twenty ceramic samples (IPS e.max CAD®) were bonded to composite cement cylinders and underwent a shear bond strength test (G1, n=20). A second group was created (G2, n=20), representing debonded restorations. To simulate debonding, the samples were artificially contaminated with composite cement. After debonding, these underwent a thermal protocol to remove remaining adhesive. After rebonding to the composite cement cylinders, samples underwent the shear bond strength test. Results Median bond strengths for G1 and G2 were 7.28±3.23; 7.06±3.41 MPa, respectively, without significant difference between the groups ( p =0.983). Conclusions Debonded lithium disilicate glass-ceramic restorations should undergo a laboratory cleaning and retreatment protocol before being returned to the clinic for rebonding. Key words: Ceramic, adhesive debonding, shear bond strength, porcelain laminated veneers.