Methods. A total of 88 Class I and 32 Class II restorations were made using Z100 (3M Dental Products Div.), Clearfil Ray-Posterior (Kuraray Co. Ltd.) and Prisma TPH (Caulk/Dentsply). Quadrant impressions were taken using a polyvinyl siloxane impression material, and stone casts were obtained every six months for indirect evaluations of anatomical form and marginal adaptation. The restorations were evaluated directly by three clinicians using the Ryge criteria.

Results. The authors statistically evaluated parameters at each recall period, using a ² test; compared baseline scores with the recall scores, using a McNemar test; and evaluated changes in the parameters, using the Cochran Q test. Color match, cavosurface margin discoloration and secondary caries remained unchanged at the end of the 24-month recall period for all materials (P .05). A total of 11.5 percent of the Z100 restorations and 16.7 percent of the Prisma TPH restorations showed a slight crevice along the margin (P .05). The surface texture was scored as Bravo only for Clearfil Ray-Posterior restorations at the end of the study (P .05). The authors found that the indirect evaluation results were different from the direct evaluations for Z100 and Prisma TPH.

Conclusions. According to these findings, all the materials used are suitable for posterior restorations. The difference between the two evaluation methods can be attributed to the level of sensitivity.

Clinical Implications. With the improvement in materials, careful case selection and application of the restorative techniques, posterior composites placed under appropriate conditions and monitored routinely can be expected to last 10 years or longer.

Acid-etching techniques have made it possible to place anterior restorations, which meet patients’ esthetic expectations and dentists’ high functionality requirements. Because of composites’ good mechanical properties, they were recommended for posterior use in the late 1960s; this inspired several clinical studies on conventional composites using the U.S. Public Health Service, or USPHS, criteria—also known as Ryge criteria—sas the main evaluation tool.^2–4 Investigations conducted in the 1970s and early 1980s found that the physical properties of conventional composite resins—especially wear, color stability and marginal leakage—were clinically unacceptable in posterior teeth.⁵ During the last decade, new formulations have been presented; the average filler size was reduced drastically, and submicrometer–particle-sized distributions of the fillers were used to optimize the filler load to improve the mechanical and wear characteristics. These improvements, coupled with better particle surface treatments for bonding to the resin component, have brought first-year average wear values to between 10 and 30 micrometers for most posterior composite resins.⁶

A goal in restorative dentistry is to make highly wear-resistant occluding surfaces that do not cause wear on opposing dental structures.⁷ Long-term quantitative in vivo wear measurements still are essential to determine the clinical performance of new posterior composite resin materials, and selection should be based primarily on these results. No laboratory test will predict the relative performance of any posterior composite resins.^8–10

We conducted this study to evaluate one posterior composite and two all-purpose composites in a 24-month clinical trial.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS EVALUATION RESULTS DISCUSSION CONCLUSION REFERENCES

 
We included in this study 38 patients (25 women and 13 men) who were referred to the Department of Restorative Dentistry, Ege University School of Dentistry, Izmir, Turkey, who demonstrated good oral hygiene and who had a cavity in a molar or premolar. We placed 120 restorations (approximately three restorations per patient) in 93 molars and 27 premolars. Seventy-five percent of the premolar cavities were Class II, and 90 percent of the molar cavities were Class I.

We used three light-cured composite resins to restore the teeth: Z100 (3M Dental Products Div.), Clearfil Ray-Posterior (Kuraray Co. Ltd.) and Prisma TPH (Caulk/Dentsply). Their properties are shown in Table 1.

All dentinal surfaces were covered with a glass ionomer base to protect the pulpal tissues from possible chemical irritation by either acid-etching agents or the composite resin monomers. All enamel margins and cavosurface margins were acid-etched and coated with the appropriate bonding agent.

The matrix and wedge for all Class II preparations were precontoured and translucent. The operator inserted the composite resins using an incremental technique. The first increment, 1 to 1.5 millimeters thick, was condensed in one corner of the proximal gingival floor and then cured with a visible light-curing unit for 40 seconds. The second increment was condensed in the opposite corner and light-cured on that side. Succeeding increments that were 1.5 to 2 mm thick were placed and sequentially light-cured until the proximal and occlusal contours were restored fully. After the matrices were removed for Class II restorations, the proximal limits were polymerized again for 20 seconds both lingually and buccally.

The occlusal surfaces were finished under water spray with flame diamond burs to remove gross excess filling material, followed by final polishing with a sharply tapered, pointed white polystone. Aluminum oxide disks were used for proximal finish. A smooth reflective surface was achieved using a fine aluminum oxide paste applied with a rubber cup.

	EVALUATION

TOP ABSTRACT MATERIALS AND METHODS EVALUATION RESULTS DISCUSSION CONCLUSION REFERENCES

 
We used both direct and indirect evaluation methods.

Direct evaluation method. We evaluated all restorations using the modified Ryge criteria,^4,12–14 which included the evaluation of color-matching ability, marginal adaptation, anatomical form, cavosurface marginal discoloration, secondary caries and surface texture. For each of the criteria, we used a score of Alfa to indicate the highest degree of clinical acceptability, used scores of Bravo and Charlie to indicate progressively lessening degrees of clinical acceptability, and used Delta to indicate that the restoration was not clinically acceptable (Table 2).

Indirect evaluation method. We made quadrant impressions of each restoration, using a polyvinyl siloxane impression material that was carried intraorally with a dispenser in a disposable quadrant tray. We also fabricated two stone models, using diestone for all samples at the baseline and at each six-month recall.

We evaluated each restoration for anatomical form and cavosurface marginal adaptation according to the modified Ryge criteria (Table 2).

We performed statistical evaluations of the parameters for each recall period, using a ² test. We compared the baseline scores with those of the recalls, using a McNemar test and evaluated the changes in the parameters during 24 months with Cochran Q test.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS EVALUATION RESULTS DISCUSSION CONCLUSION REFERENCES

 
Of the 120 restorations placed, 107 were evaluated at the 24-month recall appointment. Several patients were unavailable for recall appointments because they had moved. One patient’s restoration was broken because of a hard particle present in her food, and we excluded her tooth from the study.

The vitality of the restored teeth had not changed during the 24-month recall period.

After the 24-month recall period, we classified color match, cavosurface margin discoloration and anatomical form for all restorations as excellent (Figure 1, page 199) (P .05). The vitality of the restored teeth had not changed during the 24-month recall period. According to the periapical radiographs taken, there was no evidence of secondary caries in restorations using any of the three materials (P .05). Relative to marginal adaptation, 11.5 percent of the Z100 and 16.7 percent of the Prisma TPH restorations showed evidence of a slight crevice along the marginal interface (Figure 2, page 199) (P .05). The surfaces of all Prisma TPH and Z100 restorations except one were smooth (P .05). The surface texture was scored as Bravo for all Clearfil Ray-Posterior restorations at the end of the 24-month recall period (P .05). There was no statistically significant difference between Class I and Class II restorations (P .05). Similar results were obtained when the findings of premolar and molar restorations were compared (P .05).

View larger version (98K): [in this window] [in a new window]   Figure 1. Occlusal restorations of Z100 (3M Dental Products Div.) (arrows) in the mandibular left molars at the 24-month recall appointment.

View larger version (92K): [in this window] [in a new window]   Figure 2. A crevice along the margin of Class I Prisma TPH (Caulk/Dentsply) restoration in the mandibular right first molar after 24 months (thick arrow). The distoproximal Class II restoration of Z100 (3M Dental Products Div.) in the mandibular right second molar remained unchanged (thin arrow).

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS EVALUATION RESULTS DISCUSSION CONCLUSION REFERENCES

The surfaces of all Prisma TPH and Z100 restorations, except one, were smooth. The surface of the one rough restoration was scored Bravo from the 12-month recall appointment through the 24-month recall appointment. The direct evaluations for surface texture of 35 of 40 Clearfil Ray-Posterior restorations at the six-month recall appointment were scored Bravo. At the 12-month recall appointment, the surfaces of all but one of the remaining 36 Clearfil Ray-Posterior restorations were slightly rough and, thus, were scored Bravo. The surface of the one Clearfil Ray-Posterior restoration scored Alfa was smooth during the 24-month study period because the antagonist tooth was extracted five months after the restoration was placed. The slight roughness that indicated the Bravo scores was perceptible only with a sharp explorer, and none of the patients complained about the situation. The roughness could be attributed to the particle size and to the use of quartz as a filler particle in this material. The surface of the composite resin restoration, on examination with a sharp explorer, would characteristically show a gray discoloration from the metal instrument. This has been noted in other composite restorative materials that use quartz as filler particles.¹⁵

Through indirect evaluation of anatomical form, we found a different result from the direct evaluation for Z100 and Prisma TPH restoration groups. A total of 88.5 percent of the Z100 restorations and 91.6 percent of the Prisma TPH restorations were scored as Alfa (P .05) relative to anatomical form. For the cavosurface margin adaptation, only 36.1 percent of the Prisma TPH restorations were scored as Alfa at the end of the 24-month recall period. The percentage of Alfa scored margins was 65.7 percent for Z100 restorations and 72.2 percent in Clearfil Ray-Posterior restorations groups at the end of the 24-month recall period (P .05).

The USPHS criteria for direct evaluation remains the preferred system for evaluating important characteristics of dental restorations like color matching, secondary caries, cavosurface margin discoloration and postoperative sensitivity. Regarding the loss of anatomical form, the direct evaluation method is able to generate data that are of clinical significance and are compatible with clinical treatment.^8–10 Measurement of wear and evaluation of anatomical form and marginal integrity, in particular, benefit from the use of indirect techniques using photographs or casts. The evaluation of wear is improved by the use of casts no matter what method of analysis is used.

A restoration ideally should have a wear rate that is comparable to that of the tissues it replaces.

Indirect cast evaluations and photographs can achieve high levels of interevaluator agreement. They are more sensitive than direct clinical evaluation and may be capable of making consistent discriminations within groups that would receive the same ratings using the USPHS method.⁹

The direct clinical evaluation based on the Ryge criteria uses standard criteria for the classification of the degree of wear, but they are conceptual rather than physical and are more for the determination of appropriate clinical action than for the measurement of quantitative changes. They cannot quantitatively measure actual material loss from the occlusal surface and are not able to detect the cavosurface angle until it is exposed by as much as 150 to 175 µm.^{9,10,12,16,17} Situations in which the margin is exposed by only 100 to 150 µm would remain undetected. Additionally, wear amounting to only 50 µm would be sufficient to convert many of the Alfa ratings to Bravo.

The results of the direct clinical evaluation method suggested that the wear rates accelerated with time, while the indirect results suggested that wear rates decreased with time. The differences in these results can be attributed to the level of sensitivity associated with the two evaluation methods.^{3,9,10,12,18–20}

A restoration ideally should have a wear rate that is comparable to that of the tissues it replaces. Optimally, restorative materials and enamel should have identical wear rates.⁷ Most composite resins, however, wear at higher rates than does enamel. Depending on the amount of material loss over time, a distinct step between enamel and composite resin can develop.²¹ According to Flynn,²² the wear rate tended to be higher during the first six months, lower during the second six months, and somewhat lower and constant for the remainder of the 30-month study. These findings were in concordance with our results.

Posterior composite resins filled with zirconium silicate and especially quartz cause more antagonistic enamel wear than composite resins containing microfilled or barium glass fillers. This increased wear probably is caused by the hard filler particles.^23,24 In its pure form, quartz is highly resistant to hydrolytic decomposition. Compared with certain glass fillers, quartz can be used to generate an ideal match to the polymer matrix. Furthermore, quartz-filled composite resins, regardless of their particle size, are difficult or impossible to polish, and deterioration of the surface texture occurs within the first six months.^23–25 There is a strong relationship between the filler’s hardness and the amount of wear observed. In two studies, Suzuki and colleagues^23,24 showed that posterior composite resins filled with zirconium silicate or quartz caused more antagonistic enamel wear than did composite resins containing microfilled or barium glass fillers.

Careful case selection and meticulous attention to the restorative technique are recognized as essential requirements for achieving longer-term clinical success with posterior composite restorations. Since a sound posterior composite technique is regarded as technically demanding and time-consuming compared with the "forgiving" nature of the amalgam technique, caution should be exercised in applying these findings to general dental practice.

	CONCLUSION

TOP ABSTRACT MATERIALS AND METHODS EVALUATION RESULTS DISCUSSION CONCLUSION REFERENCES

 
The three light-cured composite resins used in this study showed satisfactory results at the 24-month recall appointment relative to all criteria except, in the case of Clearfil Ray-Posterior, the surface texture (P .005). The surface roughness of this material was found to not be clinically important, as the patients did not have any complaints and some were not even aware of it.

We conclude that all three of the materials can be used for posterior restorations. Evaluations still need to be carried out to reveal longer-term clinical performance of such materials.

Additionally, with the improvement in materials, careful case selection and application of the restorative techniques, posterior composites placed under appropriate conditions and monitored routinely can be expected to last 10 years or longer.

View larger version (136K): [in this window] [in a new window]   Dr. Türkün is an assistant professor, Department of Restorative Dentistry and Endodontics, School of Dentistry, Ege University, 35100, Bornova, Izmir, Turkey, e-mail "sebnemturkun{at}hotmail.com". Address reprint requests to Dr. Türkün.

View larger version (143K): [in this window] [in a new window]   Dr. Aktener is a professor, Department of Restorative Dentistry and Endodontics, School of Dentistry, Ege University, Izmir, Turkey.