Purpose: The purpose of this in vitro study was to evaluate microleakage around Class V glass-ceramic restorations of different coefficients of thermal expansion after prolonged thermal cycling.
Materials and methods: One hundred and twenty noncarious extracted human premolars (patient age range 12-20 years) were randomly assigned to three groups. Standard Class V preparations were cut in the buccal surface using customised Cerana burs, size no. 3. Glass-ceramic inserts from two manufacturers (Cerana, Nordiska Dental AB, Helsingborg, Sweden; Beta-Quartz, Hager & Werken GmbH, Duisburg, Germany) were used to restore the cavities and were luted with a hybrid, high-viscous composite (Tetric Ceram, Ivoclar Vivadent, Schaan, Liechtenstein) and a bonding agent (Excite, Ivoclar Vivadent, Schaan, Liechtenstein). A control group, without inserts, was bulk-filled with the same composite used as the luting agent. In accordance with American Dental Association guidelines, half of the preparation was in enamel, half in dentine/cementum and had a mesio-distal width of 3 mm, an occluso-gingival height of 3 mm, and a depth of 2 mm. All margins had butt joints. Sixty teeth, selected at random, were not thermal cycled; the remaining 60 teeth were thermal cycled 4000 times between water baths held at 5 degrees C and 55 degrees C and the specimens prepared and examined for microleakage using 2.0% Procion Red (ICI, Slough, UK) dye, buffered at pH7, as a marker. The results were analysed using the Kruskal-Wallis test (ANOVA) at a 95% significance level.
Results: At the occlusal margins there was no significant difference in microleakage between the three groups (P>0.5) without thermal cycling. After thermal cycling, microleakage at the occlusal margins was significantly less around cavities restored with Cerana glass-ceramic inserts versus Beta-Quartz and Tetric Ceram (P<0.05 in both cases). At the gingival margins, there was no significant difference in microleakge between the groups before thermal cycling (P>0.5). After thermal cycling, there was significantly less microleakage between Cerana inserts and Tetric Ceram (P<0.05). Comparisons between non-thermal cycled and thermal cycled groups showed there was no significant difference with the Cerana inserts (P=0.5590).
Clinical implications: The results indicate that, after thermal cycling, restorations restored with Cerana glass-ceramic inserts, which have a coefficient of thermal expansion approximating to that of enamel, show a decrease in marginal microleakage, compared with Beta-Quartz glass-ceramic inserts and Tetric Ceram resin-based composite material. Restorative materials, which have a coefficient of thermal expansion approximating to that of enamel, would seem to be the materials of choice in reducing the problem of marginal microleakege.