Influence of the methodology and evaluation criteria on determining microleakage in dentin-restorative interfaces.

This in vitro study compared microleakage along the dentin-restorative interface using a spectrophotometer protocol and two conventional single-surface methods (scores and percentages), using an organic dye (0.5% buffered methylene blue) or a tracer (50% silver nitrate). Occluso-proximal preparations with gingival margins in dentin were made in 40 human teeth. The teeth were divided into four groups (n=10) according to the solution dyes and adhesive system used: group 1, single bond/methylene blue; group 2, single bond/silver nitrate; group 3, Clearfil SE Bond/methylene blue; and group 4, Clearfil SE Bond/silver nitrate. The dye penetration measurements were made in all groups, using scores and percentages. Groups 1 and 3 were also assessed by UV spectrophotometer. For percentage measurement, the data were submitted to ANOVA and Tukey's test. For the material factor, there was a statistically significant difference between groups 1 and 3. For the dye factor, there was a statistically significant difference between groups 3 and 4. The score results were submitted to Kruskall-Wallis test and showed differences between groups 1 and 3 and groups 2 and 3. For spectrophotometer measurement, no significant difference was observed between groups 1 and 3. The results of dye penetration suggest that there was a difference between dyes and measurement methods, and this should change the interpretation of microleakage tests.