Purpose: To investigate the clinical performance of a silorane-based composite resin used for repairing dimethacrylate- based composite restorations. Materials and Methods: One operator repaired defective dimethacrylate-based resin restorations that were randomly assigned to one of two treatment groups: control (n = 50), repaired with Adper SE Plus and Filtek P60 Posterior Restorative (3M ESPE); or test (n = 50), repaired with P90 System Adhesive Self-Etch Primer and Bond and Filtek P90 Low Shrink Posterior Restorative (3M ESPE). After 1 week, restorations were finished and polished. Two calibrated examiners (weighted Kappa ≥ 0.78) evaluated the repaired restorations, blindly and independently, at baseline, after 6 months, 1 and 2 years. The parameters examined were marginal adaptation, anatomic form, surface roughness, marginal discoloration, postoperative sensitivity, and secondary caries. The restorations were classified as Alfa, Bravo, or Charlie, according to modified US Public Health Service criteria. Variation in the levels of clinical parameters over time was evaluated by Friedman's ANOVA (α = 0.05). The Mann-Whitney test assessed the differences between the materials for all clinical criteria at baseline, 6-month, 1- and 2-year recalls (α = 0.05). The Wilcoxon test compared each composite resin for all clinical criteria at the same recalls (α = 0.05). Results: After two years, 79 repaired restorations were re-examined. No statistically significant differences were found between the materials at baseline or at the 2-year recall (p > 0.05). Comparing baseline and 2-year recall, there was a statistically significant difference for marginal discoloration (p = 0.029) in silorane-based composite restorations. Conclusion: After two years, the clinical performance of the silorane-based composite was similar to that of the dimethacrylate-based composite when used to make repairs. Keywords: low-shrinkage silorane-based composite, dimethacrylate-based composite, resin-based restoration, repair