Purpose: To compare marginal microleakage in porcelain veneer restorations following dental finishing using two types of instruments to test the hypothesis that microleakage will be less when teeth are prepared with sonic oscillating burs than when prepared with high-speed rotating burs. Materials and Methods: Fifty-six extracted human maxillary central incisors were selected and divided randomly into two groups. Group 1 samples underwent dental finishing using high-speed rotating diamond burs, while group 2 used sonic oscillating diamond burs. Buccal chamfer preparation was carried out for both groups. Fortyeight of the samples (24 per group) were restored using IPS Empress ceramic veneers. 2% methylene blue was used to evaluate microleakage at the tooth/composite veneer interface. Teeth were sectioned lengthwise into three parts and microleakage was measured at two points - cervical and incisal - on each section. Before bonding, four teeth per group underwent SEM examination. Results: Evaluation of microleakage at the cervical dentin margin showed a value of 10.5% in group 1 and 6.6% in group 2, which was statistically significantly different (p 0.05). Incisal microleakage was 1.3% for group 1 and 1.2% for group 2, which was not significantly different. SEM revealed different patterns of surface texture in both areas according to the instrument used. Group 1 exhibited parallel horizontal abrasion grooves with a milled effect and thick smear layers; group 2 showed abrasive erosion, discontinuous perpendicular depressions, and thin smear layers. Conclusion: Tooth preparations finished with sonic burs produced significantly less microleakage in the cervical dentin area of bonded veneer restorations. No differences were found in the incisal enamel area. Keywords: porcelain veneer, microleakage, dentin adhesion, enamel adhesion, oscillating burs