Objectives: To evaluate the depth of wear gaps of new self-adhesive cements after toothbrush abrasion and ACTA wear test. Methods: Luting spaces (325 +/- 25 mu m width, 2 mm. depth) were produced in Empress 2 ceramic blocks with a diamond saw to obtain flat substrate segments for toothbrush abrasion (n = 24) and ACTA wear (n = 27). After etching and silanization, the slits were filled with 8 self-adhesive cements, 2 conventional resin cements and 1 flowable composite, stored for 2 weeks in distilled water at 37 degrees C and planished to the cement level. Toothbrush abrasion was carried out in a toothbrush simulator (Willytec, Germany) for 20,000 cycles (load 1N) using an abrasive slurry based on a commercial toothpaste (Elmex, Gaba, Germany, RDA = 77). The ACTA wear experiment was performed following the ACTA protocol in millet seed slurry for 400,000 cycles (Willytec). The gap replicas were measured for vertical wear loss under a confocal laser scanning microscope (CLSM). The data were analyzed using one-way ANOVA and a mod-LSD test at p < 0.05. Results: Toothbrush wear values were lower than the ones obtained for the ACTA wear test for all cements. In the toothbrush test Bifix SE, Clearfil SA, Smar-tCem 2, G-Cem and Maxcem Elite obtained the highest values together with Grandio Flow. Grandio Flow and AllCem showed to be the most resistant to the ACTA wear test, while SpeedCem the least resistant. No correlation was found between the two wear test experiments. Conclusion: Self-adhesive cements have good wear resistance to toothbrush abrasion but most of them wear more rapidly under higher loads in the ACTA test than conventional resin cements and flowable composites. (C) 2009 Elsevier Ltd. All rights reserved.