Microleakage and SEM interfacial micromorphology of amalgam restorations using three adhesive systems

Transcription

1 Journal of Dentistry 28 (2000) Journal of Dentistry Microleakage and SEM interfacial micromorphology of amalgam restorations using three adhesive systems M. Toledano a, *, E. Osorio a, R. Osorio a, F. García-Godoy 1,b a Department of Dental Materials, University of Granada, Granada, Spain b Department of Restorative Dentistry, University of Texas Health Science Center at San Antonio, San Antonio, 7703 Floyd Curl Drive, San Antonio, TX , USA Received 22 June 1999; received in revised form 9 August 1999; accepted 3 February 2000 Abstract Objective: This study evaluated the microleakage and interfacial micromorphology of Class V cervical amalgam restorations lined with OptiBond, Ælitebond, or Panavia 21. Methods: Unlined amalgams served as control. Cavities were treated with each dentin bonding system according to the manufacturers instructions and restored with Tytin non-gamma 2 spherical amalgam. After one week of storage in tap water at 37 C, the specimens were thermocycled (1000 cycles, 6 60 C, 30 s dwell time). Microleakage was assessed by means of basic fuchsin dye penetration and recorded according to an ordinal scale. Results: None of the systems tested in this study completely eliminated microleakage. Kruskal Wallis one-way ANOVA and Mann Whitney U test found that on the occlusal wall, Panavia 21 and the control group had the least leakage P 0:05 : No statistically significant differences were found at dentin margins P 0:05 : Wilcoxon matched pairs signed rank test found that Panavia 21 and the control group had less leakage at the occlusal than at the dentin margins P 0:05 ; when Ælitebond and OptiBond groups were evaluated, microleakage at the enamel and at the dentin margins was similar for each group. With the adhesive systems, perhaps the hydrophilic bonding agents incorporated the dye during specimen immersion and/or sectioning. Conclusions: The use of adhesives may not be as worthy as resin cements for sealing and bonding amalgam restorations to enamel and dentin 2000 Published by Elsevier Science Ltd. All rights reserved. Keywords: Amalgams; Adhesives; Dentin adhesives; Microleakage 1. Introduction The marginal seal or adaptation of the restorative material to the cavity wall is crucial for the long-term performance of any restoration. Failure to prevent microleakage may produce post-operative pain, recurrent caries, marginal staining and possible pulpal pathology [1]. More teeth are restored with dental amalgam than with any other material. Amalgam affords no adhesion to the walls of the cavity preparation [2,3]. Therefore, poor adaptation and lack of adhesion to the dental structure are the two * Corresponding author. Facultad de Odontología, Universidad de Granada, Campus Universitario de Cartuja, Granada, Spain. Tel.: ; fax: addresses: toledano@platon.ugr.es (M. Toledano), (F. García-Godoy). 1 Also at the Department of Pediatric Dentistry, University of Texas Health Science Center, 7703 Floyd Curl Drive, San Antonio, TX , USA. Tel.: ; fax: disadvantages of amalgam [4], when compared to composite resin restorations. There is no chemical link between amalgam and tooth structure, and the mechanical retention for amalgam often results in further weakening of the remaining tooth structure [5]. In an attempt to diminish tooth reduction, to prevent the breakdown at the margins of restorations and to decrease the microleakage of amalgam restorations, amalgam adhesives have been used. High-copper spherical alloys (commonly used in clinical procedures) have excellent physical properties, but have been reported to allow more marginal leakage and have inferior microscopic adaptation to cavity walls compared to the traditional formulae [6,7]. The increase in marginal leakage may be partly the result of decreased corrosion. Therefore, the use of dentin bonding adhesives to reduce microleakage and to bond amalgam alloys has become a popular clinical practice in the restoration of teeth with extensive caries and resultant loss of sound supportive /00/$ - see front matter 2000 Published by Elsevier Science Ltd. All rights reserved. PII: S (00)

2 424 M. Toledano et al. / Journal of Dentistry 28 (2000) tooth structure [1,8 10]. The use of multipurpose adhesive materials is also recommended for bonding to amalgam, composite, and tooth structures [11]. Micromechanical and/or chemical means can provide retention and marginal seal. However, Jodakin [12] stated that as the aging process progresses, corrosion of the amalgam takes place at the interface between the cavity wall and restoration, and the gap is slowly reduced by corrosion products and microleakage is reduced. The presence of a thin layer of resin between the tooth structure and amalgam might prevent amalgam corrosion products from reaching the tooth surface and forming a seal. With the adhesive technique, sealing should be achieved initially by the cement or the dentin adhesive and not by subsequent corrosion; indeed, interfacial corrosion might not occur at all or in a different manner [4]. Adhesion to dentin is of primary importance for Class V restorations and in teeth with cervical erosions where no enamel is present for bonding to the gingival margin. In many of these teeth it is difficult to prepare cavities with an appropriate retention form [13]. The purpose of this study was to determine the microleakage and interfacial micromorphology of amalgam Class V restorations lined with two dentin bonding agents: Ælitebond (Bisco, Schaumburg, IL, USA), OptiBond (Kerr, Orange, CA, USA) and a multipurpose adhesive composite resin cement Panavia 21 (Kuraray, Osaka, Japan). These groups were also compared to a control unlined group. 2. Materials and methods 2.1. Specimen preparation Forty non-carious human molars that had been stored in saline solution for not longer than 3 months were selected as test specimens. After surface debridement with a hand scaling instrument and cleaning with a rubber cup and slurry of pumice, a standardized Class V cavity preparation was placed in the buccal and lingual surface at the cementoenamel junction. The preparations were made with a #329 carbide bur (Mid West Dental Products Corp., Des Plaines, IL, USA) in a high-speed handpiece to a uniform kidneyshaped outline using a template. The preparations measured 5 mm long, 3 mm wide and 2 mm deep with the occlusal margin in enamel and the gingival margin in dentin or cementum. The teeth were randomly divided into four groups of 10 teeth each. After completion of the preparations, the bonding agents were applied according to the manufacturers directions. Ælitebond: enamel and dentin were acid-etched with phosphoric acid (32%) for 15 s. Cavities were rinsed and blot-dried with a cotton pellet to remove excess water and to avoid desiccation. One coat of Ælitebond primer was applied; all surfaces were gently dried with an air syringe for 10 s to remove the excess solvent. A layer of Ælitebond bonding resin was brushed on the enamel and dentin and light-cured for 10 s. OptiBond: enamel and dentin were etched with 37.5% phosphoric acid gel and the enamel was etched for 30 s and the dentin for 15 s; cavities were rinsed and gently blot-dried using a cotton pellet to remove excess water and avoid desiccation. OptiBond Prime was scrubbed on the enamel and dentin for 30 s. Surfaces were gently dried for 10 s and light-cured for 20 s. OptiBond Dual Cure Resin Paste and Dual Cure 3A Activator were mixed for 15 s. A uniform layer of Dual Cure Adhesive was applied to the dentin and enamel surfaces and light-cured for 30 s. Curing light output was monitored with a Demetron Model 100 Curing Radiometer (Demetron Research Corporation, Danbury, CT, USA) to 400 mw/cm 2. Panavia 21: ED-Primer was applied over the enamel and dentin for 60 s. Surfaces were gently dried for 1 s. A fine coat of Panavia 21 was applied, having mixed, previously, the two pastes of the system for s. In this group, after placing the amalgam restoration, Oxyguard (included in the Panavia kit) was applied to all marginal areas with a brush tip and left in place for 3 min. This material excludes oxygen from the surface and allows the completion of polymerization of the surface layer of the resin. Control group: no dental adhesive was placed under the amalgam restoration. All cavities were filled with Tytin (Kerr Europe AG, Basel, Switzerland) amalgam. The amalgams were placed incrementally, using small condensers, condensing toward the cavity walls. The condensation was probably superior to that usually carried out clinically because of the perfect accessibility of the teeth in vitro. The amalgams were carved and burnished but not polished. All specimens were stored in tap water at 37 C for one week Microleakage evaluation All groups were prepared for microleakage evaluation by coating the complete tooth with one application of nail varnish, except for 1 mm around the restoration margins. The apices of the teeth were sealed with zinc oxide eugenol cement (IRM, Caulk/Dentsply, Milford, DE, USA). The specimens were then subjected to 1000 temperature cycles, as suggested by Crim and García-Godoy [14]; each cycle consisted of 30 s at 6 C and 30 s at 60 C. Following thermocycling, the teeth were placed in a 2% basic fuchsin dye solution (Fisher Scientific, Fair Lawn, NY, USA) for 24 h at room temperature, rinsed with water, lightly brushed with a pumice slurry to remove superficial dye and embedded in phenolic rings with epoxy resin. Each tooth was sectioned longitudinally in a bucco-lingual direction at three different locations along each restoration with a slow-speed watercooled diamond saw (Silverstone-Taylor Hard Tissue Microtome, Scientific Fabrication, Littleton, CO, USA). The sections produced from the sectioning were then separated, and the cut surfaces corresponding to the most mesial, central (mesial and distal) and most distal portion of the

3 M. Toledano et al. / Journal of Dentistry 28 (2000) Table 1 Microleakage scores on occlusal and gingival walls (on occlusal leakage, scores for the Control and Panavia 21 groups were statistically significantly different from Ælitebond and OptiBond P 0:05 : On gingival scores, groups were not statistically different. (Total number of specimens varied due to samples damaged during sectioning) tooth restoration interface were examined at the occlusal and gingival margins using a stereomicroscope (Olympus Co., Tokyo, Japan) at 16 magnification. As Mixson et al. [15] reported that microleakage may be more extreme at mesial and distal margins of a restoration, these areas were chosen for evaluation. Examination of the specimens was carried out at random, and all investigators were unaware of the exact nature of the restorative treatment being evaluated. If conflicts in scores occurred, a consensus was obtained between observers. The staining along the tooth restoration interface was recorded according to the following criteria: 0 ˆ no dye penetration; 1 ˆ partial dye penetration along the occlusal or gingival wall; 2 ˆ dye penetration along the occlusal or gingival wall, but not including the axial wall; 3 ˆ dye penetration to and along the axial wall. The worst value (maximum amount of leakage) recorded as the reading for each margin was selected for the analysis. The occlusal and gingival scores for each group of restorations were compared using non-parametric statistical tests, Kruskal Wallis one-way analysis of variance and Mann Whitney U, to identify any statistically significant p 0:01 differences between the materials. Combined occlusal and gingival mean scores within each restoration were compared using the Wilcoxon matched pairs signed rank test SEM evaluation Occlusal Gingival Total Control Ælitebond OptiBond Panavia Two specimens of each group were used to evaluate with the scanning electron microscope (SEM) the presence or absence of marginal gaps along the entire tooth-restoration interface. After thermocycling the teeth with restorations, they were sectioned bucco-lingually through the center of each restoration. One sectioned surface produced from each of every restoration was then gently polished using 600-grit silicon carbide abrasive paper under hand pressure and water to remove surface debris. After acid-etching the surfaces with 10% polyacrylic acid for 10 s to remove the smear layer, the specimens were washed with deionized water and gently dried with oil-free compressed air. The sections were then immediately replicated by making an impression with a vinyl polysiloxane material (Extrude, Kerr, Orange, CA, USA) of the surface, and an epoxy resin die (Buehler, Lake Bluff, IL, USA) was constructed 24 h later and allowed to cure for 5 days. All tooth sections and replicas were then prepared for SEM examination by thorough desiccation in a Sampla Dry Keeper (Samplatec Co., Japan) for 48 h before sputter coating with gold palladium by means of a Polaron Equipment Limited SEM Coating Unit E 500 (Comercial Assens-Llofrin SA, Barcelona, Spain). Examination of all sections under a SEM (ZEISS, Model DSM-950, Germany) was performed at an accelerating voltage of 20 kv, and a working distance of 14 mm, micrographs were taken at 500 and 1000 operating magnifications. Replicas were also examined under SEM as control for any artifact formation. 3. Results 3.1. Microleakage None of the adhesives tested in this study completely eliminated microleakage. The mean values of the leakage scores are presented in Table 1. On the occlusal wall, the Kruskal Wallis test revealed statistical differences between groups p ˆ 0:0002 ; the resin-lined restorations with Panavia 21 and the control group appeared to leak less than the other two groups with Ælitebond and OptiBond. On the gingival wall, differences were not statistically significant P ˆ 0:16 : When comparing occlusal and gingival leakage in each group, the statistical analyses only found significant difference with Panavia 21 and the control group, in which gingival leakage scores were higher than occlusal. When Ælitebond and OptiBond groups were evaluated, microleakage at enamel and dentin margins was similar for each group SEM results Micrographs 1 4 show the main SEM findings; which were taken from actual teeth sections. OptiBond and Ælitebond specimens demonstrated bond failures both between the enamel and the adhesive system (Fig. 1) and/or between the amalgam and the adhesive system. This was not evidenced in the Panavia 21 group (Fig. 2). SEM observation of the adhesive interface of the specimens showed that at a high magnification, none of the samples showed a completely bonded interface. In all cases, some bonding failures or separations could be observed; this gap was consistently present at the axial wall of the cavities. The interface between the adhesive systems and dentin showed bonded sites but some separations were also seen along the margins of the restorations (Figs. 3 and 4). The thickness of the adhesive resin was not uniform along the amalgam interface and it was thinner in the Ælitebond

4 426 M. Toledano et al. / Journal of Dentistry 28 (2000) Fig. 1. Failure at the enamel/adhesive surface (SEM 500). E: enamel, AD: adhesive (Ælitebond), A: amalgam. group. The Panavia 21 group showed the thickest resin layer. 4. Discussion Fig. 3. Interfaces between the amalgam Panavia 21 dentin (SEM at 1000 ). A, amalgam; P, Panavia 21; D, dentin. Bond failures alternate with zones of perfect seal. In the present study, two dentin adhesives (Ælitebond and OptiBond) and a multipurpose adhesive cement (Panavia 21) have been evaluated. Their use has been previously recommended as a routine clinical procedure for amalgam restorations [10,11]. Panavia seems to have had less leakage than the other groups at dentin margins, although the difference was not statistically significant; on occlusal walls, when Ælitebond and OptiBond were used, the degree of microleakage was higher than that obtained with Panavia 21. Before applying Ælitebond or OptiBond, the preparation must be etched and a primer must be placed to promote the adhesion to the dental structure, and then a thin layer of the bonding agent is applied and cured followed by the restoration placement. When Panavia 21 is used, acid-etching is not required, but the conditioning of dentin and the etching of enamel were performed using the primer, a phosphoric acid ester monomer 10-MDP and 5-NMSA (N-methacryloyl-5-aminosalicylic acid). The inclusion of phosphate esters in the primer may permit hydrolysis from the phosphoric acid that could decalcify dentin or enamel [16]. 5-NMSA may also increase micromechanical bonding; although ionic bonding to Ca 2 is probably a less important factor in adhesion, ionic interactions may contribute to bonding by NMSA [17]. Conditioning with these priming agents demineralizes the dentin surface to a certain depth leaving behind a collagen-rich meshwork [18]. Watanabe et al. [19] stated that dentin bonding systems which do not require acid-etching will not cause discrepancy between the depth of demineralization and the depth of resin infiltration, as both the processes occur together. Acidic conditioners penetrate through the smear layer and demineralize the dentin leaving a collagen-rich mesh work, that the adhesive hydrophilic monomer is impregnating at the same time; forming a uniform interface, the formation of this interface will explain the lower degree of microleakage obtained with Panavia 21 when compared to OptiBond and Ælitebond, because these two bonding agents require acid-etching, rinsing, and drying for the removal of the bur-prepared smear layer before the primer is applied. Gwinnett et al. Fig. 2. Panavia 21 bonded to enamel and amalgam (SEM at 1000 ). E, enamel; P, Panavia 21; A, amalgam. Gap-free zone between the enamel- Panavia 21-amalgam interfaces. Fig. 4. Interfaces amalgam Ælitebond dentin (SEM at 1000 ). D, dentin; A, amalgam. There is a gap formation along the interface.

5 M. Toledano et al. / Journal of Dentistry 28 (2000) [8] showed good results of Panavia-EX (chemically similar to Panavia 21) used as a bonding agent to prevent the microleakage. The phosphate ester monomer, 10-MDP has been developed for bonding to mineralized dental tissues and restorative materials. The mechanism of adhesion to tooth structure involves a negatively charged phosphate ester (with a long hydrophobic carbon chain and a short hydrophilic phosphate group) that has strong ionic interactions with the positively charged calcium ions on the tooth and alloy surface [20]. Following manufacturer s instructions when using Opti- Bond, air must be applied after rinsing the etchant. This conventional air drying of dentin may result in the collapse of the collagen of the demineralized layer and this may impair the penetration of the priming agent or the resin monomers creating an incomplete resin infiltration with the exposed and non-infiltrated collagen, contributing to a so-called hybridoid region [21]. Tay et al. [21] also suggested that the incomplete resin infiltration within the demineralized intertubular matrix results in a weak collagen-rich zone that is susceptible to hydrolysis and microleakage. However, HEMA-containing priming agents, such as OptiBond, tend to minimize the collapse of the demineralized surface; previous reports on the ability of HEMA to enhance the penetration capability of dentin substrates described the use of aqueous solutions of HEMA [22]. The MMEP or MEP (methacryloxyethyl phtalate) is an aliphatic radical similar to HEMA and a carboxyl group connected to a benzene ring [23]. However, it is speculated that water could be an essential component in some current adhesive systems, like OptiBond, allowing desiccated collagen to be re-wetted prior to infiltration by bifunctional methacrylate components. Ælitebond is an alcohol-based, water-free, single component dentin adhesive primer system containing diarylsulfonedimethacrylate (DSDM). The wet-bonding technique is required for its application, so that it may also reduce the collapse of the demineralized layer, thus improving the penetration and attachment of the bonding resin [24]. In the wet-bonding technique, loosely arranged collagen fibers probably provide little resistance to the infiltration of the resin dissolved in either acetone or ethanol; at the same time, the acetone or ethanol displace the water within the interfibrillar channels into which the resin infiltrates and it is supposed to form a close physical relationship with the much more loosely arranged tissue microstructure [21]. Panavia 21 showed less microleakage on occlusal than on the gingival wall, perhaps because of the lower calcium concentrations in the dentin and the moisture present in dentin tubules that may adversely affect the wetting of dentin [21]. When using OptiBond or Ælitebond as bonding agents, there was no difference between occlusal and gingival walls and microleakage scores on enamel were higher for these two groups than for Panavia 21, and it may be explained because following the protocol developed by Buonocore [25], the etched enamel should be completely dry and uncontaminated; the introduction of wet bonding to dentin, created a dilemma [26]. Since it may not always be practical to dry dentin by blot drying (wet-bonding technique) and produce enough drying on etched enamel; it has been also found that contact of etched enamel with dentin primers may decrease the bond strength to the enamel by 31 44% [27,28]. These adhesive systems have similar bond strengths to the enamel and dentin, because of a decrease in the bonding to the enamel that even leaked more than the unlined group. When the specimens were observed by SEM, some adhesives failures at the enamel resin junction were found (Fig. 1), although Panavia 21 seemed to bond much better to the enamel surface (Fig. 2). The mechanism of adhesion to cast metal surfaces consists of mechanical retention as well as chemical interaction mainly due to hydrogen bonding, and presumably the mechanism of adhesion to amalgam is similar [4]. Staninec and Holt [4] reported that when amalgam bonds to resin, it appears as a mechanical interlocking and adhesion mediated by Van der Waals forces between the adhesive resin and amalgam. It seems that the phosphate ester monomer (10- MDP) component of Panavia 21, covalently links to the resin through a long carbon chain and bonds to the positively charged metallic ions on the alloy surface [29]. However, the existence of a true chemical bond between amalgam and adhesive resins has not been verified [11]. One other consideration is that Panavia 21 is a self-cure resin, which permits the amalgam to be incorporated into its body while the amalgam is being condensed, and the resin is still unset. The development of a large cement tag formation in the amalgam structure, when it is condensed in the cavity while the adhesive cement is still plastic, demonstrates that mechanical interlocking occurs [30 32]. Some studies have indicated that significant amounts of the adhesive can be incorporated into the amalgam during condensation [33]. Fig. 3 shows that Panavia 21 and amalgam are thoroughly mixed. When using OptiBond or Ælitebond the adhesive is light-cured before the amalgam is condensed onto the cavities and mechanical interlocking will not be so effective (Fig. 4). Caution should be used when applying these results to clinical practice. In microleakage studies using a dye-penetration method, leakage is always undertaken. Although thermocycling was used to stress and age the restored teeth, no load cycling was applied and the studies investigating the effect of repeated load cycling on resin-bonded restorations [34,35] concluded that the tooth reinforcement by the adhesive restoration was lost after a number of load cycles that were within the physiologic chewing range. It must also be considered that the total storage time used in this study was not long and longevity of the adhesive bond strength remains an important question. One other aspect to consider is that when dealing with hydrophilic bonding agents, the dye penetration score may very well be staining of the bonding agent and not microleakage per se [30].

6 428 M. Toledano et al. / Journal of Dentistry 28 (2000) Conclusion The use of adhesives may not be as worthy as the autocured multipurpose adhesive resin cements for sealing and bonding amalgam restorations to enamel and dentin. Acknowledgements This study was supported by a Grant #MAT from Comisión Interministerial de Ciencia y Tecnología (Spain). References [1] Berry FA, Tjan AHL. Microleakage of amalgam restorations lined with dentin adhesives. American Journal of Dentistry 1994;7: [2] Andrews JT, Hembree JH. In vitro evaluation of marginal leakage of corrosion-resistant amalgam alloy. Journal of Dentistry for Children 1975;42: [3] Derkson GD, Pashley DH, Derkson M. Microleakage measurement of selected restorative materials: a new in vitro method. Journal of Prosthetic Dentistry 1986;56: [4] Staninec M, Holt M. Bonding of amalgam to tooth structure: tensile adhesion and microleakage tests. Journal of Prosthetic Dentistry 1988;59: [5] Mondelli J, Steagall L, Ishikiriama A, Lima MF, Soares FB. Fracture strength of human teeth with cavity preparation. Journal of Prosthetic Dentistry 1980;43: [6] Ben-Amar A, Liberman R, Bar D, Gordon M, Judes H. Marginal microleakage: the effect of the number of cavity-varnish layer and the type of amalgam used. Dental Materials 1986;2:45 7. [7] Fayyad MA, Ball PC. Cavity sealing ability of lathe cut, blend and spherical amalgam alloys: a laboratory study. Operative Dentistry 1984;9: [8] Gwinnett AJ, Baratieri LN, Monteiro S, Ritter AV. Adhesive restorations with amalgam: guideliness for the clinician. Quintessence International 1994;25: [9] Berry FA, Parker SD, Rice D, Muñoz CA. Microleakage of amalgam restorations using dentin bonding system primers. American Journal of Dentistry 1996;9: [10] Christensen GJ. Should we be bonding all tooth restorations? Journal of the American Dental Association 1994;125: [11] Pilo R, Brosh T, Shapinko E, Dodiuk H. Long-term durability of adhesive system bonded to fresh amalgam. Journal of Prosthetic Dentistry 1996;76: [12] Jodakin A. Experimental microleakage around aging dental amalgam restorations: a review. Journal of Oral Rehabilitation 1981;8: [13] Fuks AB, Hirschfeld Z, Grajower R. Marginal leakage of cervical resin restoration with a bonding agent. Journal of Prosthetic Dentistry 1985;54: [14] Crim GA, García-Godoy F. Microleakage: effect of cycling and storage duration. Journal of Prosthetic Dentistry 1987;57: [15] Mixson J, Eick JD, Chappell RP, Tira DE, Moore DL. Comparison to two-surface and multi-surface scoring methodologies for in vitro microleakage studies. Dental Materials 1991;7: [16] Eliades GC, Vougiouklakis GJ. 31 P-NMR study of P-based dental adhesives and electron probe microanalysis of simulated interfaces with dentin. Dental Materials 1989;5: [17] Perdigão J, Swift EJ. Adhesion of a total-etch phosphate ester bonding agent. American Journal of Dentistry 1994;7: [18] Van Meerbeek B, Inokoshi S, Braem M. Morphological aspects of the resin dentin interdiffusion zone with different adhesive systems. Journal of Dental Research 1992;71: [19] Watanabe I, Nakabayashi N, Pashley DH. Bonding to ground dentin by a phenyl-p self-etching primer. Journal of Dental Research 1994;73: [20] Kubo S, Finger WJ, Muller M. Principles and mechanisms of bonding with dentin adhesive materials. Journal of Esthetic Dentistry 1991;3:62 9. [21] Tay FR, Gwinnett AJ, Pang KM, Wei SHI. Resin permeation into acid-conditioned, moist, and dry dentin: a paradigm using water-free adhesive primers. Journal of Dental Research 1996;75: [22] Nakabayashi N, Watanabe A, Gendusa NJ. Dentin adhesion of modified 4-META/MMA-TBB resin: function of HEMA. Dental Materials 1992;8: [23] Chigira H, Koike T, Hasegawa T, Itoh K, Wakumoto S, Hayakawa T. Effect of the self etching dentin primers on the bonding efficacy of a dentin adhesive. Dental Materials 1989;8: [24] Titley K, Chernecky R, Maric B, Valiquette N, Smith D. The morphology of the demineralized layer in primed dentin. American Journal of Dentistry 1994;7:22 6. [25] Buonocore MG. Simple method of increasing the adhesion of acrylic filling materials to enamel surfaces. Journal of Dental Research 1955;34: [26] Gwinnett AJ. Dentin bond strength after air drying and rewetting. American Journal of Dentistry 1994;7: [27] Hadavi F, Hey JH, Ambrose ER, Louie PW, Shinkewski DJ. The effect of dentin primer on the shear bond strength between composite resin and enamel. Operative Dentistry 1993;18:61 5. [28] Fortin D, Perdigão J, Swift EJ. Microleakage of three new dentin adhesives. American Journal of Dentistry 1994;7: [29] Souza M, Retief DH, Russell CM. Laboratory evaluation of phosphate ester bonding agents. American Journal of Dentistry 1994;7: [30] García-Godoy F, Finger WJ. Reliability of microleakage evaluation using dentin bonding agents. Journal of Dental Research 1993;72:308 (Abstr 1636). [31] Nobilo MA, Consani S, De Goes MF. Influence of adhesive system on the bonding of dental amalgam to glass ionomer cement. Journal of Prosthetic Dentistry 1997;77: [32] Eakle WS, Staninec M, Lacy AM. Effect of bonded amalgam on the fracture resistance of teeth. Journal of Prosthetic Dentistry 1992;68: [33] Charlton DG, Murchison DF, Moore BK. Incorporation of adhesive liners in amalgam: effect on compressive strength and creep. American Journal of Dentistry 1991;4: [34] Fissore B, Nicholls JI, Yuodelis RA. Load fatigue of teeth restored by a dentin bonding agent and a posterior composite resin. Journal of Prosthetic Dentistry 1991;65:80 5. [35] Bonilla E, White SN. Fatigue of resin-bonded amalgam restorations. Operative Dentistry 1996;21:122 6.