Amalgam Repair: Quantitative Evaluation of Amalgam-resin and Resin-tooth Interfaces with Different Surface Treatments

The replacement of defective restorations represents the major part of restorative dentistry in today's general practice.1 The main reasons for replacement include secondary caries and marginal defects.2 Other reasons for replacement involve amalgam or cuspal fracture, inadequate marginal integrity or inadequate interproximal contact.2–3 

Occasionally, the criteria for replacing a restoration are subjective, where small deviations from ideal concepts determine the replacement, especially in cases where the restorations are clinically acceptable with localized defects.4 Thus, dentists are frequently faced with a clinical decision to replace or repair an amalgam restoration. When a restoration is replaced, there is a loss of healthy dental tissue, including tissue away from the localized defects, thus increasing the preparation and restoration size.5 The cost of replacing an existing restoration is at least the same as the original restoration and is probably more costly if indirect restorations are deemed necessary. The complete replacement of these large restorations is time-consuming, technically difficult and may be potentially damaging to the pulp.3 

The successful addition of new restorative materials to an existing restoration may be the most conservative course of treatment in certain situations. By removing part of the restoration to the full depth at the site of the defect, the clinician can make a firm diagnosis regarding its extent, as the defects often are well delineated. Provided that the main part of the restoration is satisfactory, the “exploratory” cavity preparation can then be filled with an appropriate material.6 Similarly, the presence of a narrow or stained non-carious marginal defect has been shown to be an insufficient reason for replacing amalgam restorations.7 The repair of an amalgam restoration can be a less invasive procedure than its complete removal and replacement.8 Repairing margin defects may increase the longevity of the restoration,9 eliminating areas that are difficult for the patient to properly cleanse, while conserving the remaining tooth structure.10 

Continuous developments in adhesive resin technology also improve the sealing quality of amalgam repairs, beginning with the introduction of the 4-META bonding agent (Amalgambond, Parkell, Farmingdale, NY, USA) in 1990. Studies related to the repair of amalgam and resin-based composite restorations have been published during recent decades.11–13 Many researchers have studied the repair strength of amalgam, with conflicting results ranging between 11%14 and 97%15 of the intact samples, with most studies approximating 50%.16 The conflicting nature of these conclusions cause suspicion about the clinical acceptance of amalgam repairs.13 

The wide range of results found in the literature can be attributed to various factors, such as time of repair, use of a mercury-rich interface between the repair surfaces, the effects of roughening the fractured segment, the type of alloy used and the use of adhesive resins. Each factor has been evaluated by researchers, with little agreement among studies.17 

This in vitro study assessed the repair quality of amalgam restorations at amalgam-resin and resintooth interfaces after partial removal of pre-existing amalgam using different 1) surface finishing methods and 2) adhesive systems, using quantitative microleakage assessment and SEM methods.

Fifty caries-free, readily available intact human molars were collected and stored in distilled water. The teeth were cleaned and polished with pumice and rubber cups for 10 seconds. Occlusal cavities were prepared on each tooth using a high-speed handpiece with air/water spray. A new bur was used for every five teeth. The average facio-lingual width of the cavities was approximately one-third of the intercuspal width, and the depth was 3 millimeters. The cavities were restored with high copper, microfine lathe-cut amalgam (Cavex Avalloy, Haarlem, Holland). The alloys were triturated according to the manufacturer's instructions using an amalgamator (SDI Ultramat 2, Victoria, Australia). A standard amalgam carrier and technique were used to transfer increments to the cavity preparation. After condensation, the amalgam was carved into the margins following the contours of the tooth but was not finished and polished. The restorations were stored in distilled water for 24 hours, after which thermal cycling in deionized water was performed at 5 ± 2°C to 55 ± 2°C for 1,000 cycles, with a dwell time of 30 seconds and a transfer time of 10 seconds.

Mesial and distal parts of the amalgam fillings were removed, leaving only the middle-third of the amalgam. One side of the cavity was finished with a coarse diamond bur (1.6 mm flat-ended cylinder, Ref: 835-012-4ML, Diatech, Alstatten, Switzerland), while the other part was finished with a fine diamond bur (Diatech). The samples were then randomly divided into five groups (n=10/group) to receive the following adhesive systems according to the manufacturer's directions (Table 1):

• Group 1: All Bond 3 (BISCO, Inc, Schaumburg, IL, USA) (dual cure, etch&rinse adhesive system)

• Group 2: Clearfil SE Bond+Alloy Primer (Kuraray, Okayama, Japan) (self-etch adhesive system with alloy primer)

• Group 3: Kuraray DC Bond (Kuraray) (dual-cure, self-etch adhesive system)

• Group 4: Xeno V (Dentsply DeTrey, Konstanz, Germany) (one-step, self-etch adhesive system)

• Group 5: XP Bond (Dentsply DeTrey) (etch&rinse, self-priming adhesive system)

All of the cavities were restored with resin composite (TPH Spectrum, Dentsply DeTrey) and light polymerized with a halogen light of 500 mW/cm² intensity (Hi-Lux Ultra, Benlioglu, Ankara, Turkey). The specimens were again thermocycled as previously mentioned (1,000×) and immersed in 0.5 basic fuchsin solution (Wako Pure Chemical Industries, Osaka, Japan) for 24 hours. After a thorough rinsing with distilled water, the samples were air dried and embedded in epoxy resin (Struers, Copenhagen, Denmark). The specimens were sectioned mesiodistally with a slow speed diamond saw (Isomet, Buehler, Lake Bluff, IL, USA) and digitally photographed. The extent of dye penetration on the tooth-sealant interface was measured in length by image analysis software (ImageJ, Scion Image) for both the course-finished and fine-finished surfaces at both the resin-tooth and resin-amalgam interfaces. The data were statistically analyzed with one-way ANOVA and post hoc Tukey tests (α=0.05).

A total of 90 sections were examined for the quantitative evaluation of dye penetration. All the groups exhibited microleakage between the amalgam-resin interface and the tooth-resin interface. Statistical analysis of the dye penetration of both groups, with respect to different regions of the cavities, is presented in Table 2.

One-way ANOVA revealed that the difference between the experimental regions (resin-tooth or resinamalgam), with respect to adhesive systems, was statistically significant (p<0.05). The post hoc Tukey multiple comparison tests revealed the following results:

The amalgam-resin interface finished with coarsebur: All Bond 3 and XP Bond (etch&rinse adhesive systems) exhibited statistically significant results compared with the other adhesive systems (p<0.05). There was no statistically significant difference between the DC Bond, Xeno V and SE Bond + Alloy Primer adhesive systems (p>0.05) (Table 2).

Tooth-resin interface finished with coarse bur: The Xeno V adhesive system presented the worst microleakage, with statistically different results (p<0.05) from the other materials; however, there were no statistically significant differences between All Bond 3, SE Bond + alloy primer, DC Bond and XP Bond (p>0.05) (Table 2).

Amalgam-resin interface finished with fine-bur: This region showed similar results compared with groups finished with a coarse bur. All Bond 3 and XP Bond exhibited better microleakage values when compared with the other adhesive systems (p<0.05) (Table 2).

Tooth-resin interface finished with fine bur: All Bond 3 and XP Bond showed statistically significant results from Xeno V (p<0.05) but similar results with SE + alloy primer and DC Bond (Table 2).

Differences between the regions in each material were analyzed by one-way ANOVA. The results revealed that there were no statistically significant differences between each region of the restoration for only All Bond 3 (p>0.05), meaning that there was no difference among each region when using All Bond 3, while this material also presented the least microleakage. Post hoc Tukey tests were performed for multiple comparisons (Table 3).

The results revealed that there was no statistical difference between the microleakage values of surfaces with either coarse or fine finish (p>0.05). However, amalgam-resin surfaces exhibited statistically more microleakage than tooth-resin surfaces for the other adhesive systems (p<0.05) (DC Bond, XP Bond, Xeno V and Clearfil SE + Alloy primer).

Amalgam comprises about 40% of the restorations being replaced, with a median age of 12 to 15 years for the replaced restorations.18 A successful technique for the repair of existing amalgam restorations would certainly be advantageous for the clinician interested in conservative dental treatment. These types of repairs and esthetic procedures may be selected in cases when it is desirable to reduce further trauma to the tooth or avoid additional stress on the patient.8 

Recently, three treatment options versus total replacement were described.6 These options may increase the longevity of the defective restoration and preserve tooth structure. The alternative options for treating defective amalgam restorations include repairing, sealing and refurbishing. 1) Repairing consists of removal of part of the restoration and any defective tissue adjacent to the defective area and restoration of the removed site. 2) Sealing includes application of a resin-based sealant in a small, defective site or deficient margin (up to .2 mm). 3) Refurbishing involves removal of surface defects or excess from amalgam restorations using finishing burs.

Sardenberg and others reported that replacing ditched amalgam restorations with other similar restorations resulted in significant dental structure loss, while maintaining or replacing them by resin restoration did not result in significant loss.19 The microleakage of amalgam restoration repaired with amalgam was beyond the scope of this study. Although amalgam-to-amalgam repair is a widely utilized technique, previous studies report that only 40%–70%20–22 bond strength is achieved with this technique. The intro and these are both correct. The ratio ranges from 11%–97% with a median of 50%. With the improvement of composite materials, “minimally invasive dentistry” has achieved more attention. In the current study, the repaired cavities were minimally invasive, though designed for composite repair. It was the intention of the authors of the current study to evaluate the “minimally invasive repair” of amalgam restorations.

In vivo studies related to the repair of amalgam indicate a significant impact on the improvement of clinical performance of amalgam restorations with minimal intervention.3–4,23 

An important factor in the quality of amalgam repair is the interfacial bond between the joined surfaces. The interfacial bond has been studied in terms of microleakage and bond strength. Variables that were investigated include a clean, uncontaminated substrate; roughening the amalgam surface; providing additional undercuts and using repair material that is different from the substrate; and the use of adhesive systems designed for metallic surfaces. Most studies on the quality of repaired amalgam evaluate the flexural or bond strength and microleakage of the repaired amalgam.13 

Microleakage tests are useful methods for evaluating the sealing performance of adhesive systems. Among the different methods employed, dye penetration measurement on sections of restored teeth is the most commonly used technique. In the current study, image analysis was performed to obtain quantitative results, instead of a conventional subjective scoring. The relative merit of this objective approach, when compared with a subjective scoring system, was to discard the need for scoring by separate evaluators and for consensus scoring in borderline cases, as well as to reduce the need for statistical procedures with regards to interexaminer reliability.24 

As a part of this research, both halves of the specimens were measured for dye penetration with 50 specimens and 90 sections. This would have tended to nearly double the number of mutually exclusive specimens in the sample and exaggerate the statistical power to detect differences. However, the quantitative Image J analysis showed that the two sections of the same tooth demonstrated different microleakage values. The first intention was to investigate only one section, but after realizing that the two sections show different extents of dye penetration, it was decided to include both sections. This difference might be attributed to the thickness of the low-speed blade.

Research relating to the use of adhesive resins to facilitate amalgam repair has consistently found either no improvement in repair strength21 or a decrease in repair strength.26 Leelawat and others were the only investigators to find a significant increase in strength when utilizing resin adhesives.14 

The effect of roughening the surface of existing amalgams is another factor that has been studied. Jessup and Vandewalle reported improved bond strengths after roughening the surface of samples with a carbide bur prior to repair.17 Hadavi and others reported similar results after sample preparation with diamond and carbide burs.25 However, the results of the current in vitro study could not correlate microleakage and surface roughness of the joined surfaces.

Roberts and others found that the use of a bonding agent did not measurably improve the degree of protection against microleakage provided by the resin composite.27 Unlike the current findings, Ozer and others found a significant improvement in microleakage, especially at the amalgam-resin interface, when using adhesive systems.13 It is interesting to note that the results of the current study highlighted the amalgam-resin interface as the weakest part of the repaired restorative complex.

As a large number of improved resin brands are being released on the market, it is important for dentists to be aware of the probable longevity and likely modes of failure when repairing existing amalgam restorations. It seems reasonable to suggest an efficient adhesive system and a surface treatment to optimize the repair procedure.

Numerous bonding systems are available that fall under two main categories: etch&rinse and self-etch adhesives. Etch&rinse adhesive systems use phosphoric acid-etching of enamel and dentin before bonding, which is thought to be beneficial, as it is considered that the demineralized and uninfiltrated dentin zone becomes the weak point of the bond, owing to the hydrolytic degradation of collagen over time.28–29 Although the reduction or absence of demineralized dentin for the simplified “self-etching” systems was believed to primarily prevent hydrolytic degradation of bonded structures, it recently has been shown that hydrolysis of collagen may well take place in the longterm. 30 This is clinically significant, since the degradation between resin-dentin bonds may enable access for microleakage.

Self-etch adhesive systems have become increasingly popular in the last decade. The combination of etchant and primer into one system is advantageous in that it reduces the application time and technique-related sensitivity.31 On the other hand, there is an ongoing debate on the efficacy of bonding to enamel with self-etch adhesives.32 Some authors support the manufacturer's recommendations that the adjunctive use of phosphoric acid etching is necessary when bonding to uncut enamel,32 while others argue that the bond strengths of self-etch adhesives are equal to that of total-etch adhesives on unground enamel.33 To date, no correlation has been established between microleakage scores and the bond strengths of dental adhesive systems.34 

In fact, all of the specimens in the current study exhibited some microleakage, both at the amalgam-resin interface and the resin-tooth interface. However, etch&rinse systems performed significantly better than other two-step or one-step self-etch adhesive systems. The results also demonstrate that the accompanying alloy primers did not significantly improve sealing in the restorative complex.

All-bond 3 is a hydrophobic, radiopaque-filled bonding resin that is also HEMA-free. The absence of HEMA in the resin layer is a desirable feature that makes the bonding layer less prone to water sorption. Water sorption can be a critical cause in degradation of the bond, while also promoting further decay of the tooth structure. Hydrophobic adhesives are expected to be more durable than current formulas that are on the market.

The better performance of etch&rinse systems may also be related to micromechanical interlocking of the resin system to acid-etched surfaces. However, it should be noted that additional roughening with coarse burs did not facilitate bonding, both in amalgam and tooth surfaces.

In terms of preventing microleakage, etch&rinse adhesives may be preferred for amalgam repair. The use of coarse versus fine diamonds for preparation did not impact microleakage.