Introduction
Material & Methods
Results
Conclusions

Introduction

Because some physical properties of certain glass ceramics are similar to that of enamel they could theoretically be suitable for restorative purposes.

In this study a newly developed glass ceramic (IPS Empress, Ivoclar) was evaluated as an adhesive inlay restorative material. The glass ceramic was compared to amalgam and human enamel using an in vitro test, which closely simulated the in vivo situation.

Material & Methods

24 lower extracted, caries-free human molars were randomly divided into 4 equal groups

• In group 1, the occlusal contact enamel surfaces were polished using flexible discs.
• In group 2, conventional MOD cavities were prepared and filled with Dispersalloy amalgam.
• In group 3, inlay cavities were restored by adhesively cementing of newly formulated pressure cast polished glass ceramic.
• In group 4, the inlay cavities were restored exactly as in group 3 except that the pressured glass ceramic was glazed.

All Mod cavities were prepared using diamond burs. The cervical margin of one of the two proximal boxes was placed in dentin, the other proximal box was placed 1mm above the cemento enamel junction.

The Amalgam was condesed, then finished and polished with super fine diamonds, rubber polishing cups and flexible discs.

The lab-made ceramic inlays were cast at a pressure of 3-4 Mpa according to the procedure patented (IPS Empress, Ivoclar) by Wohlwend. The inner surface of the inlays were etched for 2 minutes using Stripit and silanized with Silanit. The enamel was etched for 60 seconds. The cementing composite, Dual Cement, was light and chemically cured. The glass ceramic inlays of group 3 were contoured and polished using fine and super fine diamonds and flexible discs. The outer surfaces of the inlays of group 4 were covered with a glaze.

The restorations and the enamel surfaces were stressed using the method developed by Krejci et al. The were chemically disintegrated using 75% aqueous ethanol. The were also subjected to a toothbrush/toothpaste-abrasion, and finally occlusally loaded and simultaneously thermally stressed. Palatal enamel cusps of similar size and shape of 24 extracted upper molars were sued as opposing cusps. Wear and measurements were quantitated at time intervals simulating 6 months, 1 year, 2.7 years and 5 years of function in vivo.

Occlusal mechanical loading and thermal cycling were done simultaneously in a computer controlled masticator using a chewing force of 49.0 Newtons with a rounded saw-tooth profile. The chewing rate was 1.7 Hertz with a cuspal contact time of 0.36 seconds. The lateral excursion was 0.2mm and the impact distance 2mm. Water temperatures of 5 and 55 degrees centigrade constituted a thermal cycle, each lasting two minutes.

Wear of the enamel fossae, of the restorations and their opposing enamel cusps was quantitated after each stress period using a 3D-Scanner developed by Roulet et al. \

The restoration margins were quantitatively examined in a SEM at 200 magnification using epoxy replicas of the occlusal and proximal surfaces of each restoration before and after each stressed phase. The initial width of the luting cements was also measured.

Surface changes of opposing enamel cusps were evaluated from the epoxy replicas and compared with the wear of the restoration.

Results

The quantitative wear measurements of enamel cusps opposing the restorations in um as a function of years in vivo equivalents are summarized in table 1.

The cuspal wear caused by the polished and glazed glass ceramics was similar. Although glazed glass ceramic induced initially a slightly higher wear rate, both ceramic wear rated decreased with time, showing a slightly higher wear rate for the glazed ceramic. The cusps opposing amalgam showed the least wear.

Both glass ceramic inlays were less destructive than human enamel. In contrast, both ceramics were more destructive than amalgam.

However, the wear rates for all groups at the end of the test were not statistically significant.

Table 2 shows the wear rates in the occlusal contact area as a function of time. Both polished and glazed glass ceramics were more wear resistant than enamel and their wear rates were almost three times less than that of amalgam.

The difference of all groups at five years in vivo equivalent was significant, but there was no significant difference between the polished and the glazed glass ceramic.

Table 3 summarizes the wear of enamel cusps caused by opposing restorations added to the wear of the restorations. This sum is called "total loss of vertical distance". For both glass ceramic groups the total loss of vertical distance was less than the cumulative wear of enamel. In amalgam, the total loss of vertical distance exceeded 200 microns after 5 years of simulated service. The least loss of vertical distance was recorded in the polished glass ceramic.

The contact prints of the occlusal contact surfaces of the polished glass ceramic group were mostly sharp and rougher than the polished surrounding ceramic after the test. The surfaces of the opposing cusps were flat with minimal disintegration along the border. The wear patterns were smooth.

The contact surfaces of the glazed glass ceramic inlays after stress on the right were rougher than the contact surface of the glaze.

The contact print of the amalgam occlusal surface was smooth. The boundary between the contact surface and the polished occlusal surface was sharp and slightly raised. The opposing cusps were only slightly worn, but were partly covered with an amalgam smear layer.

The contact surfaces of enamel were shallow with sharp borders showing some disintegration in the middle of the contact area. The surfaces of the opposing enamel cusps of enamel were flat with minimal disintegration along the border.

In table 4 the marginal adaptation of the polished ceramic inlays in percentage of "excellent margin" at the tooth-cement interface for the total restoration margin before stressing decreased to 68% after stressing. The initial marginal adaptation of the cervical margins in dentin was poor and decreased after stressing to less than 20% of "excellent margin". Although the initial marginal adaptation for the cervical margins in enamel was good, it dropped to 28% of "excellent margin" after stressing. The occlusal margins in contrast to the two proximal boxes remained good.

Table 5 summarizes the marginal adaptation at the cement-ceramic interface for different tooth sites. For all sites, excluding the total proximal part in dentin, which dropped to 75%, 80% or more "excellent margin" remained for the 5 years in vivo equivalent.

Table 6 shows the percentage of "excellent margin" at the tooth-amalgam interface for different tooth sites. Initial marginal adaptation at all sites was good indicating the high standards of margin finishing. The percentage of "excellent margin" for the total restoration margin was 93%m before stressing, but decreased to less than 20% after stressing. It was generally noticed that the occlusal site had the worst marginal adaptation at 5 years simulation.

Table 7 compared the percentage of "excellent margin" along the circumference of the tooth-restoration interface in the amalgam and pressed glass ceramic groups. Initially the marginal adaptation was good but that of the glass ceramic inlays decreased constantly during the test. This was in contrast to the decrease of the marginal adaptation of amalgam which dropped markedly during the first part of the test.

The final difference in marginal adaptation between amalgam and ceramic was statistically significant.

Table 8 shows the box plot of the widths of the luting cement for the polished glass ceramic group at three different tooth sites. The mean width for all sites was about 50 um indicating the excellent fit of the pressed glass ceramic inlays.