OBJECTIVE: This study aimed to evaluate the effectiveness of two ion-releasing, resin-based composites (RBCs), ACTIVA BioACTIVE-Restorative (BIO1) and ACTIVA Presto (BIO2), in preventing demineralization around restorations. METHODS: Class I and II cavities were prepared on bovine (n = 4) and human teeth (n = 8) and restored with BIO1, BIO2, a conventional composite (RBC, Filtek Supreme XTE), and a resin-modified glass ionomer cement (RMGIC, Ionolux). Following restorations (class I, n= 16/material
class II, n= 8/material), the specimens were exposed to Streptococcus mutans biofilm in a continuous-flow bioreactor over two weeks. Micro-computed tomography (micro-CT) assessed demineralization depths at restoration margins, and supernatant pH changes were measured after a 24-h acidic challenge. Statistical analyses included one and two-way ANOVA and Tukey's test (p <
0.05). RESULTS: On enamel surfaces, RMGIC showed no demineralization, followed by BIO1 (≈50 μm), BIO2 (≈125 μm), and RBC (≈150 μm). No difference between human or bovine enamel was observed. In dentin, RMGIC showed the least demineralization (≈190 μm), followed by BIO1 (≈230 μm), BIO2 (≈280 μm), and RBC (≈400 μm). pH buffering was highest in RMGIC (+ 0.24 pH), while BIO1, BIO2, and RBC showed similar buffering capacities (∼ +0.1 pH). Gaps were found at several interfaces for BIO1 and RMGIC. CONCLUSIONS: Ion-releasing RBCs varied in effectiveness for reducing demineralization of surrounding tissues, with limited pH buffering capacity. RMGIC exhibited better performance. Gaps between cavity walls and BIO1/RMGIC raised concerns about long-term adhesion. CLINICAL SIGNIFICANCE: ACTIVA BioACTIVE Restorative (BIO1) and ACTIVA Presto (BIO2) resin-based composites promoted slight inhibition of demineralization in an in-vitro biofilm model, suggesting that further modifications in their chemical composition are necessary.