In vitro evaluation of microleakage of indirect composite inlays cemented with four luting agents

STATEMENT OF PROBLEM: Microleakage around dental restorations is implicated in the occurrence of secondary carious lesions, adverse pulpal response, and reduced restoration longevity. PURPOSE: The aim of this in vitro study was to evaluate the microleakage of indirect resin composite inlays cemented with 4 luting agents. MATERIAL AND METHODS: Standardized Class V inlay preparations overlapping the cemento-enamel junction were prepared on the buccal and lingual surfaces of 40 extracted human mandibular third molars. Eighty postpolymerized, heat-treated resin composite inlays (Targis, 72 specimens, 8 controls) were processed in stone replicas and cemented into the preparations using 4 luting agents (n = 18 + 2 controls for each cement group): a resin composite used with a bonding agent (Variolink II/Excite), a resin composite used with a self-etching primer, but without bonding agent (Panavia F/ED Primer), a modified resin composite used with a bonding agent (Resinomer/One Step), and a resin-modified glass-ionomer cement (Fuji Plus). Thirty-six inlays (n = 9 + 1 control) were subjected to thermal cycling (2000 cycles, 5 degrees C/55 degrees C), whereas the other 36 were not. All the teeth were then immersed in 1% methylene blue dye solution for 48 hours. Microleakage score, margin location (enamel/cementum), thermal cycling history, and preparation location (buccal/lingual) were analyzed using a multivariate model (alpha = .05). Multivariate analysis was performed using a polychotomous logistic regression. RESULTS: The preparation location had no significant effect on dye penetration. The margin location (enamel or cementum) and the thermal cycling had a significant effect on microleakage (odds ratios ORs] = 17.6 and 8.04, respectively). In comparing the 3 resin-based luting agents (Variolink II, Panavia F, and Resinomer) to Fuji Plus, Panavia F exhibited the lowest significant overall microleakage (OR = 0.09), followed by Variolink II (significant OR equal to 0.43), whereas Resinomer demonstrated the greatest significant overall microleakage (OR = 1.35). CONCLUSION: Within the experimental conditions of this in vitro study, thermal cycling significantly increased microleakage (OR = 8.04). The overall microleakage at the enamel margins was significantly lower than the overall microleakage at the cementum margins for the 4 luting agents tested (OR = 17.6).