The aim of this study was to evaluate microleakage of composite restorations submitted to marginal treatment. Class V preparations with walls located in enamel were performed at buccal and lingual surfaces of eighteen recently extracted non-carious human premolars. Cavities were restored with composite resin (Z100 - 3M) and adhesive system (Scotchbond Multi Purpose Plus). Samples were stored in distilled water for 48h and polished with Sof-Lex discs (3M). Teeth were randomly divided in six groups: G1 - Control; G2 - marginal treatment with surface sealant (Fortify- Bisco); G3 - Nd:YAG 25Hz, 80mJ, 2W; G4 - Nd:YAG 20Hz, 100mJ, 2W; G5 - Nd: YAG 30Hz 60mJ, 1.8W; G6 - Nd: YAG 30Hz, 40mJ, 1.2W. Contact fiberoptic (300 mu m) pulsed (1.064 mu m) Nd:YAG laser was used for 30sec, under air cooling Teeth were impermeabilized, immersed in a dye (Rhodamine B) for 4h at 37 degrees C, and sectioned. Specimens were evaluated under light microscopy and evaluated with scores. Results were analyzed with Kruskal-Wallis test (p=0.05) and showed that: there were significant differences between marginal treatments; there were no significant differences between groups 1, 2, 4, and 3, 5, 6; lower values of microleakage were at groups 3, 5, 6. NdYAG laser showed marginal sealing ability and decreased microleakage of composite resins restorations.