Marginal microleakage of class V resin-based composite restorations bonded with six one-step self-etch systems

This study compared the microleakage of class V restorations bonded with various one-step self-etching adhesives. Seventy class V resin-based composite restorations were prepared on the buccal and lingual surfaces of 35 premolars, by using: Clearfil S3 Bond, G-Bond, iBond, One Coat 7.0, OptiBond All-In-One, or Xeno IV. The Adper Single Bond etch-and-rinse two-step adhesive was employed as a control. Specimens were thermocycled for 500 cycles in separate water baths at 5°C and 55°C and loaded under 40 to 70 N for 50,000 cycles. Marginal microleakage was measured based on the penetration of a tracer agent. Although the control showed no microleakage at the enamel margins, there were no differences between groups (p =  0.06). None of the adhesives avoided microleakage at the dentin margins, and they displayed similar performances (p  =  0.76). When both margins were compared, iBond presented higher microleakage (p < 0.05) at the enamel margins (median, 1.00; Q3–Q1, 1.25–0.00) compared to the dentin margins (median, 0.00; Q3–Q1, 0.25–0.00). The study adhesives showed similar abilities to seal the margins of class V restorations, except for iBond, which presented lower performance at the enamel margin. Descriptors: Dental Bonding; Stress, Mechanical; Dental Marginal Adaptation. Introduction The management of noncarious cervical lesions is an ongoing problem in preventive and restorative dentistry, due to their multifactorial etiology and increased prevalence.1 The use of direct resin-based composite as a conventional restorative procedure is still critical since this type of lesion contains simultaneously enamel and cementum/dentin contours. Therefore, class V restorations should employ bonding agents that allow effective interactions with different tissues, simple handling, and quick application, due to the relative difficulty in accessing these cavities and the presence of margins adjacent to the moist gingival tissue.2 One-step systems were developed to minimize the number of clinical steps, while incorporating the primer and adhesive into a single bottle. In addition to reducing the clinical time, these systems reduce the sensitivity of the technique and the risk of errors during application.3 With the onestep system, the infiltration of adhesive monomers occurs simultaneously with the self-etching process, decreasing the possibilities of discrepancies Declaration of Interests: The authors certify that they have no commercial or associative interest that represents a conflict of interest in connection with the manuscript. Submitted: Dec 06, 2012 Accepted for publication: Feb 08, 2013 Last revision: Mar 12, 2013 Marginal microleakage of class V resin-based composite restorations bonded with six one-step self-etch systems 226 Braz Oral Res., (São Paulo) 2013 May-Jun;27(3):225-30 between the processes and of unprotected collagen fibers.4 However, the clinical procedures are likely to be simplified at the expense of bonding performance.3-6 Due to their hydrophilic nature, one-step selfetching adhesives accumulate water upon solvent evaporation, degrading faster than hydrophobic multistep adhesives.5 Although strong one-step self-etching adhesives (pH  <  1) contain aggressive primers, their effectiveness of adhesion to the dental substrate may be harmed because esters, such as hydroxyethyl methacrylate, TEGDMA, and urethane dimethacrylate, are hydrolytically unstable.6 This effect may be reduced in mild systems (pH ~ 2); however, these adhesives may produce poorly defined etching patterns and very thin, oxygen-inhibited coatings, resulting in inadequately polymerized adhesive layers.7 Clinical microleakage is the major cause of failure for composite restorations and may lead to postoperative sensibility, marginal discoloration, secondary caries, or pulpal inflammation.8 A widely used in vitro method for determining the restoration durability involves the aging of class V restorations by thermal stressing.9 Changes in the elastic capacity of the restoration elements under occlusal loading may lead to damages in the margins.10 The assessment of tracer molecule microleakage through the adhesive interface is another commonly used method to evaluate the bonding effectiveness.8-10 One-step self-etch adhesives initially showed inferior results compared to multistep systems.3,4 However, since those studies were performed, improvements have been made in the chemical compositions of the systems and in our understanding of application techniques.11 Because few well-designed studies are available that liken one-step self-etching adhesives with etch-and-rinse control systems,11,12 the aim of this study was to compare the marginal microleakage of class V resin-based composite restorations bonded with these systems in vitro. Methodology Thirty-five sound human premolars were carefully scaled of calculus, soft tissue, and other debris by using a dental curette and pumice/water slurry. The teeth were stored at room temperature in 0.1% thymol solution to avoid bacterial development. The source of the teeth (no. 17/2007) and the research protocol (no. 06273/06) were approved by the Ethics Committee of the State University of Ponta Grossa. Seventy class V cavities (3.0  mm mesial-distal, 2.0 mm occlusal-gingival, and 1.8 mm deep) were made on the buccal and lingual surfaces, with the occlusal margin in enamel and the gingival margin in cementum (dentin). The preparations were made with carbide bur #6 (Beavers Dental, Morrisburg, Canada) and diamond bur #3131 (KG Sorensen Ind., São Paulo, Brazil) in a high-speed hand piece cooled with an air-water spray and using a digital caliper (Mitutoyo, Kanagawa, Japan).9 The burs were discarded after each preparation. The specimens were randomly divided into seven groups (n = 10) according to the one-step self-etching system used: • Clearfil S3 Bond (Kuraray Medical Inc., Okayama, Japan), • iBond (Haraeus Kulzer, Armonk, USA), • G-Bond (GC Corp., Tokyo, Japan), • One Coat 7.0 (Coltène/Whaledent, Altastätten, Switzerland), • OptiBond All-In-One (Kerr Corp., Orange, USA), and • Xeno IV (Dentsply/Caulk, Milford, USA). The Adper Single Bond 2 (3M-ESPE, Saint Paul, USA) etch-and-rinse two-step adhesive system was used as a control (Table 1). After bonding, the cavities were each filled with three increments of the microhybrid composite resin Synergy D6 (Coltène/Whaledent, Altastätten, Switzerland) and light-activated for 40 s with 600 mW/cm2 by a dental lamp Demetron Optilux 400 (Kerr Corp., Orange, USA). The light intensity was verified every five activations. The teeth were stored at 37°C for 24 h, and the restoration surfaces were finished by using diamond PRO discs (Dental MFG, Aurora, USA). Teeth were submitted to thermal stressing for 500 cycles in separate distilled water baths at 5°C and 55°C, with dwell and transference times of 15 and 5 s, respectively (El Quip, São Carlos, Brazil).9 The roots of the teeth were embedded in epoxy resin and Sánchez-Ayala A, Farias-Neto A, Vilanova LSR, Gomes JC, Gomes OMM 227 Braz Oral Res., (São Paulo) 2013 May-Jun;27(3):225-30 microscope (Olympus American Ind., Hamburg, Germany) at 40× magnification. The section with the greatest microleakage was selected for analysis. Marginal microleakage was scored as follows: • 0: absence of penetration of tracer agent through the enamel or dentin margin; • 1: penetration to one-third of the cavity; • 2: penetration to two-thirds of the cavity; and • 3: penetration to more than two-thirds and in the axial wall of the cavity.9 All of the procedures were performed by a single calibrated examiner. Data were explored with the SPSS statistical program (IBM, Armonk, EUA). Analysis of variance on ranks test was used to compare the microleakage scores of each adhesive system for the enamel and dentin margins independently. The Wilcoxon signed rank test was applied to compare the microleakage scores between the enamsubmitted to 50,000 mechanical cycles at 2 Hz each. Teeth were loaded perpendicularly on the center of the occlusal surface by using a steel ball tip 5.1 mm diameter (El Quip, São Carlos, Brazil). Each cycle involved a load variation from 40 to 70 N, according to the lower and higher deflections of a spring system.9 The entire tooth surface was sealed with two coats of commercial nail varnish, with a 1-mmwide border being left around the restoration margins. The teeth were immediately immersed in 50% AgNO3 solution for 2 h in a dark room. After washing, the specimens were placed in a developer solution and exposed to fluorescent light for 6 h (Eastman Kodak Company, Rochester, USA).9 The specimens were sectioned longitudinally through the center of the restoration, from the facial to the lingual surface. Two sections were obtained from each restoration. Microleakage at the enamel and dentin margins was analyzed with an optical Table 1 Characteristics of the adhesive systems. Adhesive system Composition Application mode Adper Single Bond 2 Etchant: 35% H3PO4; Adhesive: dimethacrylates, HEMA, polyalkenoid acid copolymer, 5 nm silane treated colloidal silica, ethanol, water, photoinitiator; pH 3.3; Batch: 4BC Etch for 15 s; rinse for 10 s; blot excess water with a cotton pellet (glistening surface w/o pooling of water); apply 2 consecutive coats of adhesive, by brushing gently on the cavity surface; evaporate solvent by gentle air stream for 5 s; light-cure for 10 s Clearfil S3 Bond MDP, bis-GMA, HEMA, hydrophobic dimethacrylate, photoinitiators, ethanol, water, silanated colloidal silica; pH 2.7; Batch: 41132 Dry; apply 1 layer of adhesive onto surface cavity w/o agitation for 20 s; evaporate solvent by high-pressure air flow for 5 s; light-cure for 10 s G-Bond 4-MET, phosphoric ester-monomer, UDMA, TEGDMA, acetone, water, stabilizer, silica filler, water, photoinitiator; pH 2.0; Batch: 0609051 Shake the bottle; dry; apply 1 layer of adhesive onto surface cavity w/o agitation for 10 s; dry thoroughly under maximum air


Introduction
The management of noncarious cervical lesions is an ongoing problem in preventive and restorative dentistry, due to their multifactorial etiology and increased prevalence. 1 The use of direct resin-based composite as a conventional restorative procedure is still critical since this type of lesion contains simultaneously enamel and cementum/dentin contours.Therefore, class V restorations should employ bonding agents that allow effective interactions with different tissues, simple handling, and quick application, due to the relative difficulty in accessing these cavities and the presence of margins adjacent to the moist gingival tissue. 2 One-step systems were developed to minimize the number of clinical steps, while incorporating the primer and adhesive into a single bottle.In addition to reducing the clinical time, these systems reduce the sensitivity of the technique and the risk of errors during application. 3With the onestep system, the infiltration of adhesive monomers occurs simultaneously with the self-etching process, decreasing the possibilities of discrepancies Declaration of Interests: The authors certify that they have no commercial or associative interest that represents a conflict of interest in connection with the manuscript.
between the processes and of unprotected collagen fibers. 4][5][6] Due to their hydrophilic nature, one-step selfetching adhesives accumulate water upon solvent evaporation, degrading faster than hydrophobic multistep adhesives. 5Although strong one-step self-etching adhesives (pH < 1) contain aggressive primers, their effectiveness of adhesion to the dental substrate may be harmed because esters, such as hydroxyethyl methacrylate, TEGDMA, and urethane dimethacrylate, are hydrolytically unstable. 6his effect may be reduced in mild systems (pH ~ 2); however, these adhesives may produce poorly defined etching patterns and very thin, oxygen-inhibited coatings, resulting in inadequately polymerized adhesive layers. 7linical microleakage is the major cause of failure for composite restorations and may lead to postoperative sensibility, marginal discoloration, secondary caries, or pulpal inflammation. 8A widely used in vitro method for determining the restoration durability involves the aging of class V restorations by thermal stressing. 9Changes in the elastic capacity of the restoration elements under occlusal loading may lead to damages in the margins. 10[10] One-step self-etch adhesives initially showed inferior results compared to multistep systems. 3,4owever, since those studies were performed, improvements have been made in the chemical compositions of the systems and in our understanding of application techniques. 11Because few well-designed studies are available that liken one-step self-etching adhesives with etch-and-rinse control systems, 11,12 the aim of this study was to compare the marginal microleakage of class V resin-based composite restorations bonded with these systems in vitro.

Methodology
Thirty-five sound human premolars were carefully scaled of calculus, soft tissue, and other debris by using a dental curette and pumice/water slurry.
The teeth were stored at room temperature in 0.1% thymol solution to avoid bacterial development.The source of the teeth (no.17/2007) and the research protocol (no.06273/06) were approved by the Ethics Committee of the State University of Ponta Grossa.
Seventy class V cavities (3.0 mm mesial-distal, 2.0 mm occlusal-gingival, and 1.8 mm deep) were made on the buccal and lingual surfaces, with the occlusal margin in enamel and the gingival margin in cementum (dentin).The preparations were made with carbide bur #6 (Beavers Dental, Morrisburg, Canada) and diamond bur #3131 (KG Sorensen Ind., São Paulo, Brazil) in a high-speed hand piece cooled with an air-water spray and using a digital caliper (Mitutoyo, Kanagawa, Japan). 9The burs were discarded after each preparation.
The Adper Single Bond 2 (3M-ESPE, Saint Paul, USA) etch-and-rinse two-step adhesive system was used as a control (Table 1).
After bonding, the cavities were each filled with three increments of the microhybrid composite resin Synergy D6 (Coltène/Whaledent, Altastätten, Switzerland) and light-activated for 40 s with 600 mW/cm² by a dental lamp Demetron Optilux 400 (Kerr Corp., Orange, USA).The light intensity was verified every five activations.The teeth were stored at 37°C for 24 h, and the restoration surfaces were finished by using diamond PRO discs (Dental MFG, Aurora, USA).
Teeth were submitted to thermal stressing for 500 cycles in separate distilled water baths at 5°C and 55°C, with dwell and transference times of 15 and 5 s, respectively (El Quip, São Carlos, Brazil). 9The roots of the teeth were embedded in epoxy resin and Braz Oral Res., (São Paulo) 2013 May-Jun;27(3):225-30 microscope (Olympus American Ind., Hamburg, Germany) at 40× magnification.The section with the greatest microleakage was selected for analysis.Marginal microleakage was scored as follows: • 0: absence of penetration of tracer agent through the enamel or dentin margin; • 1: penetration to one-third of the cavity; • 2: penetration to two-thirds of the cavity; and • 3: penetration to more than two-thirds and in the axial wall of the cavity. 9l of the procedures were performed by a single calibrated examiner.Data were explored with the SPSS statistical program (IBM, Armonk, EUA).Analysis of variance on ranks test was used to compare the microleakage scores of each adhesive system for the enamel and dentin margins independently.The Wilcoxon signed rank test was applied to compare the microleakage scores between the enam-submitted to 50,000 mechanical cycles at 2 Hz each.Teeth were loaded perpendicularly on the center of the occlusal surface by using a steel ball tip 5.1 mm diameter (El Quip, São Carlos, Brazil).Each cycle involved a load variation from 40 to 70 N, according to the lower and higher deflections of a spring system. 9he entire tooth surface was sealed with two coats of commercial nail varnish, with a 1-mmwide border being left around the restoration margins.The teeth were immediately immersed in 50% AgNO 3 solution for 2 h in a dark room.After washing, the specimens were placed in a developer solution and exposed to fluorescent light for 6 h (Eastman Kodak Company, Rochester, USA). 9 The specimens were sectioned longitudinally through the center of the restoration, from the facial to the lingual surface.Two sections were obtained from each restoration.Microleakage at the enamel and dentin margins was analyzed with an optical el and dentin margins for each adhesive (α = 0.05).

Results
The frequencies of microleakage scores at the enamel and dentin margins for each adhesive are shown in Table 2. Resin-based composite restorations bonded with Adper Single Bond 2 showed absence of microleakage, and the highest microleakage values were displayed by iBond.Nevertheless, no significant difference was found (p = 0.06).All of the systems appeared to be incapable of sealing the restoration and, thus, preventing microleakage through the dentin margins.At the dentin margins, One Coat 7.0 presented the lowest microleakage values, whereas Adper Single Bond 2 and G-Bond showed the highest microleakage values, however no significant difference was also found (p = 0.76).When comparing the enamel and dentin margins for each adhesive system, iBond presented higher microleakage (p < 0.05) for the enamel than for the dentin margins.

Discussion
The results from this study are consistent with those of Owens et al., 11 who found no difference be-tween iBond or OptiBond Solo Plus, a two-step etchand-rinse system.Yazici et al. 12 also observed similar microleakage at the enamel margins among Single Bond, Adper Easy One, and G-Bond.Geerts et al. 8 showed no difference between Single Bond or Ad-heSE One, and Manuja and Nagpal 13 also found similar performance between OptiBond All-In-One or iBond.In contrast, Vinay and Shivanna 14 found that Clearfil S 3 Bond presented better sealing compared to Single Bond, iBond  , and G-Bond.Deliperi et al. 15 found higher microleakage in restorations adhered with iBond compared to those treated with Prime & Bond NT, a two-step etch-and-rinse system.The latter system also presented lower microleakage at the dentin margins than Clearfil S 3 Bond. 16Yaman et al. 17 found lower microleakage for XP Bond (a twostep etch-and-rinse system) compared to One Coat 7.0.Owens and Johnson 18 found lower performance at the enamel margins and higher sealing at the dentin margins using Xeno IV and Clearfil S 3 Bond, respectively, as compared to iBond and G-Bond.
The performance of Clearfil S 3 Bond may be explained by its alcohol solvent.It has a higher vapor pressure and a low dielectric constant, and produc-  es hydrogen bridges with water, resulting in a better evaporation rate. 3,9,19Other solvents such as acetone promote higher dehydration and collapse of the collagen network at the dentin margin. 18Water sorption by the osmotic gradient or phase-separation of system components after acetone evaporation may lead to dentin and enamel blisters.This adhesive also includes a proprietary "Molecular Dispersion Technology," which enables the formation of a two-phase, liquid hydrophilic/hydrophobic component homogenous state at the molecular level.It may result in the reduction and/or loss of water droplets at the interface. 14urthermore, the methacryloxy decyl phosphoric acid (MDP) monomer can generate decalcification and bond to ionic calcium and hydroxyapatite. 6-Bond decalcifies the dentin surface only slightly and produces almost no exposure of the collagen fibers. 14Consequently, an extremely thin adhesive interface is formed, explaining the microleakage results.However, in the "nano interaction zone", the 4-MET monomer may react with hydroxyapatite to form an insoluble calcium compound, resulting in a better seal that is less likely to be enzymatically deteriorated. 7he low flow rate of the adhesive due to the 4-MET component can result in less water uptake when the adhesive is stored in water. 5Its properties may also be improved by applying a strong, continuous air blast to remove the solvent prior to polymerization. 4owever, the drying procedures associated with iBond have been regarded as labor-intensive and may not be realistic for some practitioners in a clinical setting. 14These procedures include multiple applications on prepared surfaces and relatively long waiting periods. 18The lower performance of this adhesive at the enamel margin may be explained by the lack of an apparent etching pattern, the minimal exposure of crystallites in irregular areas on the unground enamel, and the islands of superficially dissolved tissue within areas without dissolution displayed on ground enamel after its application. 4dhesives containing 4-MET also may show lower enamel-bonding capacities than MDP-containing adhesives.Therefore, enamel pre-etching with phosphoric acid is recommended to improve adhesion. 20imited evidence is available regarding microleakage with the use of the HEMA-based adhesives One Coat 7.0 and Optibond All-In-One.HEMA improves the miscibility of hydrophobic and hydrophilic adhesive components. 5It can replace the solvent and keep the less-miscible dimethacrylate monomers in solution, whereas the water molecules in a hydrophilic adhesive remain hydrogen-bonded to the HEMA monomers. 21Nevertheless, void formation by the increment on hydrophilicity and HEMA presence has frequently been discussed. 7Optibond All-In-One has two solvents, ethanol and acetone, with different vapor pressures, which also might increase the void formation. 19A higher concentration of solvent would be necessary for HEMA to be omitted. 5he solvent-monomer balance might be disturbed by evaporation or lack thereof; water droplets and any nonevaporated water remaining can also contribute to void formation; 22 a strong air blowing was shown to have a significant effect on the degree of solvent evaporation and removing water droplets. 4The application of two adhesive layers onto the surface cavity could help to improve the results. 7dper Single Bond 2 is able to wet and impregnate the etched enamel in an efficient manner comparable to those of three-step systems.However, its efficiency on dentin is lower because the adhesive incompletely diffuses under wet-bonding conditions, and a porous collagen network remains. 6,19Phaseseparation also occurs in the interphase region between the hydrophilic primers and hydrophobic resins, resulting in water absorption.This effect may be justified by the incorporation of a high-molecular-weight polyalkenoic acid and the presence of HEMA. 3 Nevertheless, the incorporation of alcohol as a cosolvent may help to explain its performance. 14n addition to the compressive/tensile stresses produced during occlusal loading, 10 thermocycling results in contraction/expansion stresses and accelerated chemical degradation. 3However, the use of an extremely large number of cycles may be unnecessary, because the stress emerges early on the adhesive interface components.Thus, the use of 500 cycles can induce aging. 23In this study, the use of a dwell time of 15 s was considered clinically relevant, because a vital tooth will not tolerate extreme temperatures for extended periods. 24The viscosity, surface tension, functional monomers, pH, water con-centration, and cohesive strength of adhesives may affect the bonding and explain the microleakage of the adhesives studied. 6,13The absence of these variables can be considered as study limitations.Other features, such as the type of composite, cavity, and dye tracer, analyzed tooth section, and number of sample, may also have influenced the results. 3

Conclusions
No significant differences between the adhesive systems were encountered at the enamel and dentin margins.Comparing all surfaces, iBond showed significantly greater microleakage values than the other adhesives at the enamel margins.

Table 1 -
Characteristics of the adhesive systems.

Table 2 -
Frequency of microleakage scores at the enamel and dentin margins.