This study quantitatively assessed the amount of microleakage on Class V cavities prepared by Er:YAG laser and high-speed handpiece, varying the surface treatment and restoring with a resin-modified glass ionomer cement. Fifty cavities were prepared using either an Er:YAG laser device or a carbide bur at high speed. The surface treatment was performed as follows: Er:YAG laser irradiation (G1); 40% polyacrylic acid (G2); laser + acid (G3); finishing with low speed + laser + acid (G4); conventional bur preparation + acid (G5-control). The samples were restored with Fuji II LC, thermocycled, isolated and immersed in a 50% AgNO3 solution. The restorations were serially sectioned and the extent of dye penetration was measured in milimeters using specific computer software. Data were analyzed by two-way ANOVA and Tukey test. The lowest degree of microleakage was observed for G5, which was statistically similar (p>0.05) to G4 but different (p<0.05) from all the other experiental groups. Lesser microleakage was observed at the occlusal margins than at the cervical margins (p<0.05). It may be concluded that the use of Er:YAG laser for cavity preparation and surface treatment negatively affected the marginal sealing of resin-modified glass ionomer restorations.