Aim: This in vitro study evaluated gap width formation and marginal microleakage in Class II composite restorations light-cured with three different light-curing units. Methods: Standardized cavities in the proximal surfaces of 36 human third molars were made with margins located below the cementoenamel junction. Cavities were restored with Filtek P60 (3M ESPE), inserted with a photocondenser tip and light-cured with three different methods: GI-Optilux401 (halogen); GII-ColtoluxLED (LED) and

more »

... I-UltraLumeLED5 (LED). After finishing the restorations, teeth were subjected to a thermal cycling regimen of 500 cycles (5 o C ± 2 o C and 55 o C ± 2 o C), totalizing 500 cycles. Thereafter, the teeth were sectioned in a buccolingual direction and in the center of the restorations. Half of the specimens (18) were used to evaluate marginal microleakage, by measuring of dye penetration in cross-sectioned specimens, and the other half was used to analyzed the gap formation width by SEM observations (1000X). Data were submitted to Kruskal-Wallis (α=0.05). Results: The mean values of gap width (µm) were: GI 3.28±3.34; GII 1.48±1.89 and GIII 3.11±3.45, and microleakage was not affected by the light-curing units. Conclusions: There were no differences between the light-curing methods in gap formation and marginal microleakage.