Effect of cavity design on microleakage of resin-modified glass ionomer restorations

To investigate cavity sealing with resin-modified glass ionomer cements (RGIC) and a compomer relative to the design of Class V cavities crossing the CEJ.

Three types of cavities, 6 x 3 x 1.5 mm, were prepared buccally in human molars. Type 1: semicircular with 90 degrees cavo-surface angle; Type 2: as type 1, but with bevelled enamel margin; Type 3: V-shaped. The cavities were restored with the compomer Dyract (DYR), with one of the RGIC, Fuji II LC (FUJ), Photac-Fil (PHF) or Vitremer (VIT), or with the bonded resin composite system Gluma-CPS/Pekafill (GLP) as a reference. Six restorations in each group were finished immediately after curing and thermocycled 1500 between 5 degrees C and 55 degrees C. For Type 1 restorations, another set of six samples was stored in 23 degrees C water for 24 hours. After thermocycling and water storage, respectively, the specimens were immersed in 0.1% fuchsin. Two sections were made through the restoration to register microleakage on the four surfaces produced as depth of dye penetration coronally and apically by scores 0 (no) to 4 (beyond deepest part of cavity). The results were statistically analyzed by cavity type, site, and restorative material using Kruskal-Wallis ANOVA and Wilcoxon's two sample test at P = 0.05.

When compared with water storage, thermocycling had no effect on dye penetration with FUJ and PHF, yet more pronounced microleakage at the enamel site was noted with DYR and VIT, and apically with DYR and GLP. Type 1 restorations leaked more than type 2 and 3 cavities for FUJ, VIT and GLP.