RESUMO
PURPOSE: This study's purpose was to evaluate microleakage of two types of glass ionomer (GI) and composite restorations with saliva contamination at different stages of restorative procedures. METHODS: Extracted teeth with Class V cavities were restored with a conventional GI, nanofilled RMGI, or total-etched composite. The preparations were contaminated with saliva before the adhesive/primer application or before the restoration placement (N=10). The restored teeth were thermocycled (1000X), stained (basic fuchsin), and sectioned. Microleakage distance was measured and subjected to analysis of variance and Duncan's post-hoc test (P=.05). RESULTS: For the enamel margin, no significant difference was found between the conventional GI and composite restoration, with or without saliva contamination (P>.05). The nanofilled RMGI with contamination before restoration had the highest microleakage. For the cementum margin, composite had significantly more microleakage than both types of GI restorations, regardless of saliva contamination. CONCLUSIONS: Conventional and resin-modified glass ionomer restorations had less cementum microleakage, while the composite had less enamel microleakage. Saliva contamination did not affect microleakage of the conventional GI at either enamel or cementum margins. The nanofilled RMGI system was not sensitive to saliva contamination at the gingival margin, but had increased microleakage at the enamel margin, especially after the primer application.