Aim To investigate pulp chamber penetration of bleaching agents in teeth following restorative procedures. Methodology Bovine lateral incisors were sectioned 3 mm apical to the cemento-enamel junction and the coronal pulpal tissue was removed. Teeth were divided into six groups (n = 10): G1, G2 and G3 were not submitted to any restorative procedure, while G4, G5 and G6 were submitted to Class V preparations and restored with composite resin. Acetate buffer was placed in the pulp chamber and treatment agents were applied for 60 min at 37 degreesC as follows: G1 and G4, immersion into distilled water; G2 and G5, 10% carbamide peroxide (CP) exposure; G3 and G6, 35% CP bleaching. The buffer solution was removed and transferred to a glass tube where leuco crystal violet and horseradish peroxidase were added, producing a blue solution. The optical density of the blue solution was determined spectrophotometrically at 596 nm. A standard curve made with known amounts of hydrogen peroxide was used to convert the optical density values of the coloured samples into microgram equivalents of hydrogen peroxide. Data were submitted to <smallcapitals>anova</smallcapitals> and Tukey's test (5%). Results Amounts of hydrogen peroxide found in the pulp chamber of G2 and G5 specimens (0.1833 +/- 0.2003 mug) were significantly lower (P = 0.001) when compared to G3 and G6 specimens (0.4604 +/- 0.3981 mug). Restored teeth held significantly higher (P = 0.001) hydrogen peroxide concentrations in the pulp chamber than intact teeth. Conclusion Higher concentrations of the bleaching agent produced higher levels of hydrogen peroxide in the pulp chamber, especially in restored teeth.