Purpose: To investigate regional shear bond strength to lateral walls of ceramic inlays in occlusal and occlusoproximal cavities using etch-and-rinse and self-adhesive resin cements and a glass-ionomer luting agent. Materials and Methods: IPS e.max Press ceramic inlays were made in 50 Class I and 50 Class II standardized cavities in intact extracted human molars and divided into 5 luting agent subgroups (n = 10): Variolink II (VL); Multilink Sprint (MLS); Multilink Automix (MLA); RelyX Unicem (RLX), and Ketac Cem Aplicap (KC). Inlays were pre-etched with IPS Ceramic etching gel for 60s. After 48 h, two disks of ca 1.0 mm thickness, one of superficial and the other of deep dentin, were push-out tested in a universal testing machine at a crosshead speed of 1.0 mm/min. The mode of failure was determined under a stereomicroscope at 20X. Data were analyzed with oneway ANOVA, and Scheffé's test was used for post-hoc comparisons (α = 0.05). Results: There were no significant differences in shear bond strength between Class I and Class II cavities for the dual-curing system in light-curing mode (VL=MLS=RLX), except that RLX demonstrated greater bond strength to deep dentin in Class II cavities. Bond strength values were significantly higher on deep than on superficial dentin. KC showed the worst result. Failures were mixed (adhesive/cohesive) for the resin luting cements and solely adhesive (cement/ceramic) for the glass-ionomer luting agent. Conclusion: Dual-curing etch-and-rinse or self-etching self-adhesive resin luting cements achieved greater bond strength when light curing was applied, with no differences between Class I and Class II cavities but higher values for deep vs superficial dentin. The weakest adhesion was obtained with glass-ionomer luting agent in both cavity types. Keywords: ceramic inlays, push-out, bond strength, etch-and-rinse, self-etching, resin cement