Background: Moisture control is very important in restorative procedures in dentistry. Use of hemostatic agents helps control moisture; however, they might result in changes on enamel and dentin surfaces, affecting composite resin bond quality. The aim of this in vitro study was to evaluate the marginal microleakage of two different composite resins with the use of three different hemostatic agents. Material and Methods: Standardized Class V cavities were prepared on the buccal and lingual surfaces of 48 premolars with cervical margins 1 mm apical to the cementoenamel junction (CEJ). The samples were randomly divided into 8 groups. In groups 1 to 4, an etch-and-rinse adhesive (Adper Single Bond) was applied as the bonding system, followed by exposure to different hemostatic agent: group 1: no hemostatic agent (control); group 2: ViscoStat; group 3: ViscoStat Clear; and group 4: trichloracetic acid, as hemostatic agents. The cavities were restored with Z-250 composite resin. In group 5 to 8 Silorane System Adhesive (Filtek P90 Adhesive) was applied as a bonding agent, followed by exposure to different hemostatic agents in a manner similar to that in groups 1to 4. The cavities were restored with Filtek P90, a low-shrinkage composite resin. The samples in each group were evaluated for dye penetration under a stereomicroscope at ×36 after 24 hours and a 500-round thermocycling procedure at enamel and dentin margins. Statistical analysis was carried out using Kruskal-Wallis and Mann-Whitney tests (α=0.05). Results: Z-250 composite resin exhibited significantly higher dentin microleakage scores compared to Filtek P90 ( P = 0.004). Trichloracetic acid increased dentin microleakage with Filtek P90 ( P =0.033). Conclusions: Under the limitations of this in vitro study, application of hemostatic agents did not affect microleakage of the two tested composite resins except for trichloracetic acid that increased marginal microleakage when used with Filtek P90.