After three years, cumulative success rates in Thailand were 71 percent for single-surface ART restorations and 50 percent for sealants in permanent teeth.^2,6 Children responded well to being treated with the ART technique and showed little fear.⁷ Three-year findings in Zimbabwe showed cumulative success rates of 85 to 88 percent for single-surface ART restorations and 50 to 71 percent (for full and partial retention) for sealants in permanent teeth.³ The GIC sealants provided preventive dental care to the school student population and, again, the children were satisfied with the ART procedure.⁸ Three-year findings in China showed cumulative success rates of 92 percent for small occlusal restorations, 77 percent for large occlusal restorations and 72 percent for sealants (mainly in permanent teeth).⁵ The ART method also is suitable for use in more developed countries with apprehensive or disadvantaged patients or patients with disabilities, as local analgesia and dental handpieces may not be required.^9,10

However, there have been few reports of direct comparisons of GIC restorations placed using the ART approach with amalgam or other restoratives placed using conventional methods. One three-year study² of a GIC and an amalgam alloy placed in the permanent teeth of separate patient populations found a significantly higher cumulative survival for the single-surface amalgam restorations (85 percent) when compared with the single-surface GIC restorations (71 percent) (P < .001). Another study, which involved three restorative techniques and matched contralateral pairs of occlusal cavities in the permanent molars of children, found that irrespective of the technique, after two years the cumulative success rates were similar at 92.3 percent for the amalgam restorations and 95.8 percent for the GIC restorations.¹¹ In neither study were the GIC materials developed specifically for the ART approach.

More recently, several more viscous esthetic conventional GICs have been marketed specifically for use with the ART approach. However, there have been few reports of any direct comparisons of the clinical performance of these newer high-strength GICs with that of conventional amalgams. Yip and colleagues¹² conducted a 12-month controlled trial comparing two restorative techniques and the performance of two more viscous GICs and an amalgam alloy in the restoration of Class I and II cavities in primary molar teeth. They found that the GICs used in Class II restorations had a high failure rate when placed with the ART method, but not when they had been placed using rotary instrumentation.¹² Researchers who conducted another six-month controlled trial of one more-viscous GIC and an amalgam alloy reported no failures of the GIC occlusal restorations placed in permanent molar teeth when using the ART method.¹³ None of the occlusal amalgam restorations placed using rotary instrumentation in either of these two studies failed.

Therefore, the objective of our 12-month study was to evaluate clinically two newer more-viscous GICs placed using two cavity preparation methods, and one amalgam alloy placed using a conventional cavity preparation method, for the restoration of occlusal caries in permanent molar teeth, in a dental hospital setting. The null hypothesis we set out to test was that the deterioration and survival of the restorative materials are independent of the material and treatment method.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
We compared the handling, survival and deterioration of two encapsulated more-viscous esthetic conventional GICs, Fuji IX GP (GC International Corp., Tokyo) and Ketac-Molar Aplicap (3M ESPE, Seefeld, Germany), with those of an encapsulated high-copper-content admixed non–gamma 2 amalgam alloy, GK Amalgam Alloy (Advanced Technology & Materials Co. Ltd. Beijing), in a clinical trial over 12 months. The two GICs were marketed specifically for the ART approach and were available in several shades, of which we used A2 and A3.

We placed the three restorative materials in relatively small occlusal preparations in the vital permanent molar teeth (including third molars) of 68 volunteer adult patients who had given their informed consent to participate in the study. Subjects had to have at least two suitable carious teeth with opposing tooth contacts present. We obtained approval for the collaborative study from the ethics committee of The University of Hong Kong Faculty of Dentistry.

To reduce operative and assessment variables, the treatments and observations all were carried out in a dental hospital clinic at the Beijing Medical University School of Stomatology. The 149 restorations were placed sequentially by material and method of cavity preparation, using either ART hand instruments¹⁴ or conventional rotary instrumentation for the GIC cavity preparations, and conventional rotary instrumentation for the amalgam cavity preparations. Three dentists placed the restorations (92 percent were placed by two of the authors, W.G. and D.P.), with chairside assistance. Local analgesia, rubber dam isolation and bases were not used for any restorations. For the ART restorations, the dentists used only sharp ART hand instruments (Code S642017, GC International Corp.) to gain access and to excavate soft caries. The cavity preparations were not extended to remove any adjacent intact fissures. Moisture was controlled with cotton rolls. Neither oral evacuation nor the triple syringe was used.

For the GIC and amalgam restorations placed with the conventional method, the dentists used rotary instrumentation for cavity preparation. The amalgam cavity preparations extended into dentin and included the removal of any adjacent fissures. Moisture was controlled by means of oral evacuation and the triple syringe. The times taken for cavity preparation were recorded for the three treatments.

For the GIC restoration placement, the dentist first conditioned the cavities and adjacent occlusal fissures with polyacrylic acid for 15 seconds, then washed them with water-moistened cotton pellets before lightly drying them with fresh pellets. The dentists machine-mixed the encapsulated GICs and injected them into the cavities, then pressed them into any adjacent fissures using a gloved finger lightly lubricated with polyacrylic acid. Cavity varnish was applied after the initial set. The encapsulated amalgam alloy also was machine-mixed and then packed into the cavities using hand instruments; it was then burnished but not polished.

Posterior bitewing radiographs and color transparencies (x2 magnification) were taken before the cavity preparations were done, and more photographs were taken afterward. Addition-cured silicone impressions and photographs also were taken after placement of the restorations and at the six- and 12-month recalls, for indirect clinical assessments. Two dentists (R.J.S. and W.G.) compared the color transparencies against standard sets depicting progressive deterioration of restoration characteristics for marginal staining and discrepancies, surface staining and tarnishing, and color changes.¹⁵ They gave "good" restorations a score from 0 to 2, "adequate" restorations a score from 3 to 8 and "unsatisfactory" restorations a score from 9 to 12, which approximated the alfa, bravo and charlie categories of the Ryge direct clinical observation method.¹⁶ They also used the transparencies to assess the GICs’ surface porosity (a "yes" or "no" rating) and sealant retention ("complete," "partial, repair not required," "partial, repair required" or "all lost"). Die stone replicas were poured from the impressions and compared against a semiquantitative standard set of Rhein-berger ivorine teeth (Ivoclar-Vivadent, Schaan, Lichtenstein) for occlusal wear (measured in micrometers) of the GIC restorations relative to the cavity margins and contiguous tooth surfaces.¹⁷ The method was not sensitive enough to assess the very low wear associated with the amalgam alloy. Two dentists (W.G. and D.P.) made direct clinical observations (registered with "yes" or "no" ratings) of restoration bulk fractures, pain of pulpal origin and caries, as shown by cavitation and softening to probing. It was not possible to distinguish either directly or indirectly between the two GICs or their treatment modes.

We entered all observations into a database, checked them for errors and then analyzed them using Instat 3.0 and Prism 2.0 (GraphPad Software Inc., San Diego, Calif.) and SPSSPC 8.0 (SPSS Inc., Chicago) statistical software, with the probability level set at 5 percent for significance. Tests included ² one-way analysis of variance, or ANOVA; nonparametric Mann-Whitney U; and Friedman statistics. Two of the researchers (R.J.S. and W.G.) repeated the indirect assessments for 30 restorations after several months to test examiner reliability. agreements for eight restoration characteristics were high and ranged from 0.71 to 0.95.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
General observations. The number of subjects assessed at the 12-month recall appointment was 46, with 98 restorations, for recall rates of 68 percent and 66 percent, respectively (Tables 1 and 2). Subjects attended for recall visits at mean (± standard deviation, or SD) times of 171 ± 27 days and 359 ± 28 days.

The mean (SD) times for the relatively small cavity preparations were 2.1 ± 0.3 minutes for the ART approach and 0.9 ± 0.3 minutes for the conventional method when GICs were used, and 1.1 ± 0.3 minutes for the conventional method when amalgam was used. There were significant differences between the ART method and the conventional method in cavity preparation times (P < .001). One capsule of material usually was sufficient for a restoration. The restorative treatments generally were uneventful, and none of the subjects reported immediate postoperative problems.

Restoration and sealant performance. There were no instances of recurrent marginal caries or other causes of restoration failure for any material. However, where GIC had covered occlusal pits and fissures adjacent to the restorations, there was a rapid early loss of material, which was progressive (P < .001) (Table 3). After 12 months, sealants appeared to have been completely lost in two instances for Fuji IX GP (6.1 percent) and in one instance for Ketac-Molar Aplicap (3.3 percent). There were no significant differences in regard to sealant losses between the two cements at any observation time (Mann-Whitney U-statistic = 462.50 to 1,372.00, P = .97 to 1.00), and there were no instances of fissure caries detected where sealant had been lost.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

We used encapsulated materials to reduce operator variability in mixing and to obtain optimal material properties. However, some occasional problems were encountered with the GIC capsules, especially for Fuji IX GP, requiring these mixes to be discarded. The GK amalgam alloy was the easiest material to use.

Restoration and sealant performance. The ART approach increasingly is being advocated for use in pediatric and general clinical practice, and not merely for field use.^10,19 In our study, we attempted to simulate some of the field conditions for the GIC restorations placed using the ART approach, to facilitate comparison of their clinical performance with that of the GIC and amalgam restorations placed using conventional rotary instrumentation. However, with no failed GIC and amalgam restorations after 12 months, it was not possible for us to compare survival rates for the two different treatment methods. But the 100 percent success rates for the GIC and amalgam restorations in this study may be compared with the results after 12 months from the following similar studies.

Comparison of a GIC placed using the ART approach with an amalgam alloy placed using mobile dental equipment, in the permanent posterior teeth of separate patient populations, showed successes for single-surface restorations of 92.9 percent and 97.7 percent, respectively (Y. Songpaisan, unpublished data, 1993). A controlled trial of occlusal restorations in permanent teeth, comparing two metal-modified GICs placed using the ART technique with an amalgam alloy placed conventionally, yielded success rates of 95.9 percent and 98.9 percent, respectively.²⁰ One other controlled trial of matched occlusal restorations in permanent teeth comparing a GIC with an amalgam alloy placed using three restorative techniques resulted in success rates of 98.0 percent and 96.5 percent, respectively, irrespective of the technique used (L. Forsten, written communication, July 3, 1996). The GICs used in these studies were not marketed specifically for the ART approach, and the more viscous GICs used in our study performed better.

In our study, there was a high early loss of GIC material from the sealed occlusal fissures adjacent to the restorations, but with 95.2 percent of the sealants still partially or completely retained after 12 months (Table 3). This finding is consistent with other reports of early loss of sealant material, with relatively few teeth having completely sealed pits and fissures after six months. After 12 months, other ART studies have reported partial and complete occlusal sealant retention in permanent teeth of 73.7 percent (with 0.8 percent of teeth experiencing fissure caries),⁸ 75.0 percent (with no fissure caries mentioned),²¹ 78.0 percent (with no fissure caries mentioned)⁷ and 83.0 percent (with 5.3 percent of teeth experiencing fissure caries).⁹ In our study, we found no instances of fissure caries where sealant was missing. An in vitro study has shown that more-viscous GICs used with the ART approach will penetrate occlusal fissures successfully when placed using the "finger-press" method,²² and remnants of the sealants may remain in the depths of the fissures after superficial losses to confer some protection against caries.²³ GICs alone are unlikely to prevent or arrest caries progression in people at high risk of experiencing caries²⁴ and, if caries removal is incomplete, then the sealing ability of the GIC is extremely important,²⁵ especially in the permanent dentition.

There are relatively few references in the literature to modes of deterioration of ART restorations and sealants placed in permanent teeth.^3,11,20 In the present study, progressive restoration deterioration was characterized for the amalgams by minor surface tarnishing and marginal discrepancies (Table 4) and, for the GICs, by significant (but improved) body color matching changes (Table 5) and high occlusal wear (Table 6). The two GICs showed a marked color shift from baseline, becoming darker to more closely match the restored teeth within six months. However, there were few "good" or ideal shade matches, although this is not critical for restorations in molar teeth. Similar color shifts have been reported for other conventional GICs,^9,13 but the long-term clinical relevance is not known.

Both of the encapsulated GICs also showed relatively high cumulative net occlusal wear rates from baseline, resulting in only minor instances of surface and marginal staining and marginal discrepancies. The mean (SD) cumulative net occlusal wear at 12 months for encapsulated Fuji IX GP (77.4 ± 47.0 micrometers) was slightly higher than that reported in another study for hand-mixed Fuji IX placed in permanent molar teeth (61.4 ± 38.9 µm).⁹ There was a wide variation in wear between individual restorations for both of the GICs, which, together with their large mean values, would place the cements into the American Dental Association’s restrictive category of posterior restoration use.²⁶ There was no indication that the wear rates had decreased significantly after six months (Table 6). Although no GIC restorations were replaced because of excessive occlusal wear in our study, this was the cause of failure for 3.0 percent of restorations composed of two metal-modified GICs,²⁰ and for 0.9 percent of restorations composed of another GIC⁸ over 12 months in other clinical studies. As with the short-term marginal fracture studies of amalgam alloys, the 12-month GIC wear data may provide an indication of the future longevity of occlusal restorations.

The survival of the restorative materials was independent of the material and treatment method, and partly supported the null hypothesis. However, both GICs showed relatively high losses of material from surface deterioration. Although the early 12-month findings are promising, further improvements in the mechanical properties of the more-viscous GICs are required for their optimal long-term survival in load-bearing situations in adults.

	CONCLUSIONS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION CONCLUSIONS REFERENCES

 
From the initial findings of this 12-month clinical study of two more-viscous GICs and one amalgam alloy placed in relatively small occlusal cavities, prepared in permanent molar teeth using two instrumentation methods, we made the following conclusions:

– The encapsulated materials generally were easy to use, although some GIC capsules were faulty.

– The restorative procedures, using either the ART or conventional instrumentation, also were generally uneventful. However, cavity preparation with the ART approach took approximately twice the time as it did with the conventional method.

– There were no GIC or amalgam failures, but three (4.8 percent) of the sealants adjacent to the GIC restorations appeared to be completely missing, without fissure caries being detected.

– The relatively high wear rates of the GICs resulted in only occasional minor surface and marginal staining and marginal discrepancies. For the amalgam restorations, minor surface tarnishing and marginal discrepancies were more frequent, more obvious and more progressive. The mean (SD) cumulative net occlusal wear was 77.4 ± 47.0 µm for Fuji IX GP and 82.5 ± 50.9 µm for Ketac-Molar Aplicap, without any significant difference (P > .05).

– Both GICs became darker with time to more closely match the restored teeth, but there were few ideal shade matches.

– The two more-viscous conventional GICs appear initially suitable for restricted clinical use in the occlusal surfaces of permanent molar teeth in adults.