This study evaluated in vitro commercial desensitizing toothpastes with respect to the prevention of erosion and explored the effect of their agents alone or in combination with fluoride. Bovine enamel blocks were randomly allocated to five groups of 20 and exposed to: Sensodyne ProNamel (1,425 ppm F as NaF, 5% KNO(3)), Sensodyne Original (no fluoride, 10% SrCl(2)), Colgate Sensitive (1,450 ppm F as sodium monofluorophosphate, 5% K citrate), Crest (fluoride-only toothpaste, 1,100 ppm F as NaF) and water (negative control). A second experiment was conducted with experimental dentifrices containing fluoride (NaF, 1,100 ppm F), 10% SrCl(2), 5% KNO(3 )or 5% K citrate alone or the latter three combined with F. The samples were submitted to four cycles, alternating demineralization (cola, 10 min) and remineralization (artificial saliva, 1 h). Before and between cyclic de- and remineralization, blocks were treated with slurries of the respective toothpastes or water (1 min). Erosive tissue loss was analyzed by profilometry. Data were analyzed by Kruskal-Wallis and Dunn's tests (p < 0.05). The mean erosion depth (+/- SE, mum) was significantly less for Colgate Sensitive (0.04 +/- 0.00), Sensodyne Original (0.06 +/- 0.01) and Crest (0.07 +/- 0.01) than for Sensodyne ProNamel (2.36 +/- 0.25) or water (2.92 +/- 0.24), which did not significantly differ from each other. Both F and the desensitizing agents alone reduced erosion, but no additive effect was found. In addition, the combination of F and KNO(3) did not reduce erosion. These in vitro results suggest that the presence of fluoride or desensitizing substances in toothpastes, alone or in combination, can reduce erosion of enamel, but this is not valid for all the formulations.