Journal of Oral Rehabilitation, 1992, Volume 19, pages 265-270
Microleakage studies comparing a one-visit indirect composite inlay system and a direct composite restorative technique A . M . C A S S I N and G . J . P E A R S O N * Department of E.xperimental Denttstry. School of Dentistry. University College London, and "Department of Biomaterials, The Institute of Dental Surgery, London, U. K.
Problems related to leakage have been observed with direct-filled composite restorations. The development of indirect system of fabrieation followed by eementation should theoretically reduee this phenomenon. The present study compares the leakage around directly placed composites with composite inlays prepared by a surgery-based system. The results indicate that the amount of leakage is reduced with the inlay system, but that it is not eliminated. Introduction
The aesthetic restoration of posterior teeth with composite materials and dentine adhesives is becoming increasingly popular in general dental practice, despite major limitations associated with the characteristics of the materials and techniques for placement. Shrinkage as a result of polymerization (Bowen et al., 1982) produces contraction forces which may compromise the bond at the tooth interface (Feilzer et al., 1987; Pearson & Hegarty, 1989). This leads to marginal gap formation, which is well doeumented (Phair & Fuller, 1985; Crim, 1987), and may allow bacterial access and subsequent pulpal pathology (Brannstrom & Nyborg, 1971; Brannstrom & Vojinovic, 1976). While the efficacy of the bond between composite and etched enamel has been thoroughly investigated, the corresponding long-term reliability of the bond to dentine has not been adequately demonstrated. Elimination of polymerization shrinkage would be a major factor in the reduction of microleakage. Three techniques have been suggested to reduce the effects of this shrinkage: (i) use of an incremental packing technique (Jensen and Chan, 1985); (ii) the 'waxing up' of a restoration in composite in the mouth and polymerization extraorally, and (iii) the complete extra-oral fabrication of a composite inlay which is then placed using a resin cement (Bishop 1989). This paper compares the marginal integrity of restorations produced through two of these three techniques by use of a dye penetration test, with and without a thermocycling regime designed to simulate intraoral temperature changes (Pearson & Longman, 1987).
Correspondence; Dr G.J. Pearson, Department of Biomaterials Institute of Dental Surgery, 2^6 Gray's Inn Road, London WCIX 8LD, U.K.
A.M. Cassin and G.J. Pearson
Materials and methods The two systems examined in this study were (i) the direct posterior composite Heliomolar* packed incrementally and (ii) the EOS (extra-oral system) indirect onevisit composite inlay*. Recently extracted permanent molars were collected and stored at 23°C in formol saline. After they had been cleaned and examined, only those that were free of earies and restorations and showed no evidenee of white spot lesions or eracks on the mesial and distal surfaces were selected for the study. Forty teeth were chosen, and Class II Mesio-occluso-distal (MOD) cavities, with one box finishing in enamel and the other extended into cementum, were prepared in each of the teeth using a tapered bur in an ultra high-speed hand-piece under water-spray. The cavities prepared for restoration with the EOS system were similar to conventional inlay cavities without cuspal reduction, but had a more exaggerated flare in the vertical walls (as recommended by the manufacturer). Those prepared for restoration with the Heliomolar system were minimal, undercut cavities. All cavo-surface margins were prepared to approximately 90°. All preparations were made as uniform as possible with regard to outline and depth. The preparations were washed with water and dried with eompressed air. An etchant gel (37% phosphoric acid) was applied to the internal enamel and 2 mm peripheral to the cavo-surface margins for 60s. The preparations were then washed with water for 30s to ensure the removal of excess gel. They were then dried with compressed air. The conventional composite restoration was paeked onto a ealcium hydroxide liner (Dycal)t in the cavity base (applied to the pulpal floor and axial walls). The composite was placed in small increments across the floor and up the walls of the cavity. Each layer was cured for 20 s. The final layer was shaped to approximately the original anatomy of the tooth, and was not polished. The entire restoration was then further polymerized according to the manufacturer's instructions (40 s from each interproximal aspect and 40s from the occiusal surface). Cavities for the EOS system were unlined, using only the dentine protector and cavity varnish supplied by the manufacturer. The inlay itself was constructed and fitted according to the manufacturer's instructions. The technique involved production of an additional silicone rubber impression of the inlay cavity. Following application of separating medium to the surface of the impression, a silicone die was constructed from it, lepresenting a positive likeness of the original cavity. The composite inlay was then built up incrementally and cured on this die. The inlay was tried in, the peripheral enamel etched with 37% phosphoric acid gel*, and the inlay was then cemented with a resin based dual-cure cement (Dual cement)*. Finger pressure was applied to seat the inlay. The root apices of all teeth were sealed with gutta-percha and coated with three eoats of varnish to within approximately 1 mm of the interface of the restoration and the tooth tissue. One group containing 10 EOS and 10 Heliomolar restorations was stored at 37°C for 24 h in a 10% solution of neutral red. A further group of nine EOS and 10 Heliomolar restorations was thermoeycled for the equivalent of 3 months
* Ivoelair-Vivadent, Schaan. Liechtenstein. t Caulk Deutsply. Delaware, U.S.A.
Comparative microleakage studies
in vivo (Pearson & Longman, 1987). These restorations were then placed in a 10% solution of neutral red dye for 24h at 37°C (only nine EOS examples were available, since the tenth inlay could not be adapted to the cavity). All teeth were seetioned longitudinally in a buceo-lingual plane using a slowcutting diamond wheel*. A maximum of five sections was cut from each tooth, each seetion having up to four interfaces (two eavo-surface margins on each of two faces). The first and last sections were cut within 1 mm of the mesial and distal boxes so that microleakage around the base of the box could be examined. Table I hsts the number of teeth, seetions, interfaces and boxes examined. Dye penetration was recorded at each interface and classified as follows: 0 = no dye penetration; 1 = dye penetration, but not to the full depth of the eavity; 2 = dye penetration to the full depth of the cavity; 3 = dye penetration to the full depth and along the base of the cavity; 4 = dye penetration into the dentine and pulp (Fig. 1). Dye penetration at the base of the interproximal boxes was scored as follows: 0 = no dye penetration; or 1 = dye penetration along the base of the cavity. Statistical analysis of the results was performed using a Chi-square test. Results
The data for dye penetration at the oeclusal interface gaps are shown in Table 2. In non-thermocycled teeth there were more interface gaps with no penetration in the EOS than in the Heliomolar restorations, although the eombined numbers of gaps in which dye had penetrated to the full depth of the cavity, along the base of the cavity or beyond the cavity and into the dentine were similar for both materials. Overall, there was less penetration in thermoeycled teeth, but again there were more gaps with no penetration at all in the EOS restorations. The reduction in penetration in thermoeycled compared to non-thermocycled teeth in both EOS and Heliomolar restorations was highly significant statistically (EOS, P « 0-001; Heliomolar, P « 0-001; the numbers in categories 3 and 4 were combined to obtain adequate expected values, Siegel, 1956). A partieularly striking feature in non-thermocycled teeth restored with Heliomolar was the large number of gaps with
Table 1. Number of teeth, sections, interface gaps and boxes examined in the dye penetration study Heliomolar
Before thermoeycling Number of teeth Number of seetions Number of gaps Number of boxes
10 60 208 20
10 56 192 20
After thermocycling Number of teeth Number of sections Number of gaps Number of boxes
10 100 200 20
9 91 182 18
* Testbourne Ltd, Basingstoke, Hants, U.K.
A.M. Cassin and G.J. Pearson Inlay
Fig. I. Diagram illustrating the scoring system for dye penetration along the interface between the restoration and tooth tissue; 0 = no penetration; 1 = dye penetration down the side of the cavity; 2 = dye penetration to the full depth of the cavity; 3 = dye penetration to the full depth of the cavity and along the cavity floor; 4 = dye penetration into the dentine and toward the pulp.
dye extended along the occiusal floor of the eavity. Statistical analysis also confirmed the differences between EOS and Heliomolar restorations when comparing nonthermoeycled ( f «0-001) and thermoeycled (/'