SUMMARY
This study investigated the bonding performance of three universal adhesive systems applied using etch-and-rinse (ER) or self-etch (SE) strategies on natural dentin caries.
Sixty human third molars were selected for this study: 30 naturally carious (CAD) and 30 sound (SD) teeth. The dentin surfaces were exposed, and teeth were randomly assigned to each evaluated adhesive system: Scotchbond Universal (SBU), Futurabond U (FBU), and Prime&Bond Elect (PBE) and an adhesive strategy: ER or SE. The adhesive systems were applied following the manufacturer's instructions, and the teeth were restored using a resin composite (Filtek Supreme Ultra, 3M). After 24 hours (distilled water at 37°C), samples were sectioned and evaluated using microtensile bond strength analysis (μTBS), micro-Raman spectroscopy to evaluate the degree of conversion within the hybrid layer (DC), and scanning electronic microscopy (SEM) to describe the morphology of the hybrid layer. The μTBS and DC data were analyzed using three-way analysis of variance and Tukey's test for means comparison (α=0.05). The SEM images were analyzed qualitatively.
Reduced μTBS values were observed when comparing CAD with SD, regardless of adhesive system or strategy (p<0.0001). SBU showed statistically higher μTBS for both dentin substrates and strategies (p<0.0001). Furthermore, SBU showed greater integrity of the hybrid layer and resin tag formation compared with FBU and PBE. Mean μTBS values for FBU were higher for SD in the SE mode, whereas higher mean μTBS values were observed for CAD in the ER mode, both compared with PBE (p<0.001).
Bonding performance is reduced on a caries-affected substrate. The ER strategy was not able to improve the bonding performance on natural CAD for universal adhesive systems. Improved bonding performance was obtained when using the Scotchbond Universal system.
INTRODUCTION
In contemporary dentistry, minimally invasive approaches have been consistently developed and reinforced for restorative procedures, with one technique advocating the removal of the outer layer of highly infected, denatured caries-infected dentin.1,2 The remaining caries-affected dentin tissue above the pulpal floor is to be preserved because it is a substrate that is suitable for remineralization and could prevent disease progression, reducing the unnecessary exposure of dental pulp tissues.1,2
Several chemical, biological, and physical modifications can occur as a result of the caries process, leading to a clinically distinct situation, which can affect adhesion to caries-affected dentin (CAD).3,4 As a result, lower bond strength is found, compared with bonding procedures to sound dentin (SD),5-7 and a poorly formed hybrid layer7 can also be observed. However, most of the adhesion studies have focused on investigating SD only, which differs completely from the carious-dentin substrate. Furthermore, some studies are performed in artificially induced lesions,8-11 which is a less challenging substrate than the one resulting from clinical caries.
In an attempt to simulate the caries process, several artificially induced caries protocols have been described.12,13 These methods have a very important advantage in that they provide a standardized substrate, which is difficult to achieve with the natural lesions. However, it is known that no artificial carious lesion process is able to completely simulate the cascade of events involved in the natural caries process.14,15 Additionally, the chemical and histologic characteristics promoted by the natural carious process (water content, blocked dentin tubules due to the deposition of intratubular dentin, among other factors) cannot be simulated.16-18 Therefore, controversial results are found in the literature regarding different caries induction, bonding, or adhesive strategies.4,6,7,9,10,13 This creates difficulty for the clinician when deciding which adhesive system or strategy to use with this sorely modified substrate.
More recently, new “universal” or “multimode” adhesives have been launched in the market.19 Among the main advantages of these systems is the ability to use either a one-step self-etch approach or phosphoric acid (like an etch-and-rinse system) can be used.20-23 Therefore, the use of universal systems seems favorable due to the broad versatility and facility of use by the clinician.
Unfortunately, there is a lack of dental literature regarding the bonding performance to natural dentin caries, especially when using universal or multimode adhesive systems.19-23 It is very important to investigate the performance of universal systems when bonding to natural dentin caries, as this substrate represents one of the major reasons for restorative procedures.24,25 Unfortunately, the performance of these universal systems has been tested using SD22,23,26 or artificially induced caries-affected dentin.9,10,13
The aim of this study was to evaluate the bonding performance of three universal adhesive systems on natural CAD, using both etch-and-rinse and self-etch strategies, compared with sound dentin. The null hypotheses tested were as follows: 1) there is no difference in the bonding performance to natural CAD or SD; 2) the bonding performance is the same for all universal adhesive systems evaluated; and 3) there is no difference between the self-etch and etch-and-rinse strategies on the bonding performance to the tested substrates.
METHODS AND MATERIALS
Tooth Selection and Preparation
Sixty human third molars (30 natural CAD and 30 SD) were obtained with informed consent from donors (20-35 years old) under Local Ethics Committee number 1750969. Teeth were stored at 4°C in 0.5% chloramine T for up to 1 month before use. For carious teeth, the inclusion criteria required teeth diagnosed with active lesions using tactile, visual, and radiographic analysis, described as a 5 on the International Caries Detection and Assessment System (ICDAS) scale.27,28 The samples of carious teeth were prepared by exposing a flat midcoronal carious dentin surface using a slow speed diamond saw (Isomet, Buehler, Lake Bluff, IL, USA) under water cooling. One operator was calibrated to achieve appropriate dentin for the lesion walls29 determined by the tactile dentin texture, ensuring that only carious tissue was removed until firm dentin (resistant to hand excavator) was left on the pulpal wall and that the periphery of the lesion contained clean to hard dentin (similar to sound dentin)29 using a hand excavator. The exposed dentin was then finished using 600-grit silicon carbide paper under irrigation with distilled water to standardize the dentin roughness and smear layer thickness. All teeth were thoroughly rinsed with water, followed by the removal of the surrounding enamel using a diamond bur in a high-speed handpiece (#2135, KG Sorensen; São Paulo, SP, Brazil) under water irrigation. All exposed dentin surfaces were further finished using a 600-grit silicon carbide (SiC) abrasive paper for 60 seconds.
Experimental Design
All teeth were assigned to an experimental group based on the type of substrate (CAD or SD) and then three levels for adhesive system: Scotchbond Universal (3M Oral Care, St. Paul, MN, USA, also known as Single Bond Universal in some countries), Futurabond Universal (Voco, Cuxhaven, Germany), and Prime&Bond Elect (Dentsply Caulk, Konstanz, Germany); and two levels for adhesive strategy: etch-and-rinse or self-etch, totaling 12 experimental groups (n=5). Detailed information about the composition of adhesive systems, batch numbers, and application modes are described in Table 1.
Bonding and Restorative Procedure
One single calibrated operator performed all adhesive protocols according to each adhesive system and the respective manufacturer's instructions (Table 1). After bonding, the dentin surfaces were restored using a resin composite build-up (Filtek Supreme Ultra, 3M Oral Care). The build-ups were 4-5 mm in height, with three to four increments of resin composite, individually light-cured for 40 seconds using a monowave LED curing unit (Radii Cal, SDI, Bayswater, Victoria, Australia; 1200 mW/cm2). All the restored teeth were stored in distilled water at 37°C for 24 hours prior to preparation.
Specimen Preparation
After 24 hours of storage, teeth were longitudinally sectioned in both the mesio-distal and bucco-lingual directions across the bonded interface in a cutting machine (Buehler, Lake Bluff, IL, USA), resulting in resin-dentin sticks with a 1-mm2 cross section. The number of premature failures per tooth during specimen preparation was recorded. The resin-dentin sticks originating from areas immediately above the pulp chamber covered by CAD were selected for testing (selected by dentin color, under 40× magnification). Selected sticks were randomly allocated to three different tests: two sticks for degree of conversion analysis, two for hybrid layer morphology observation, and all other remaining sticks were used for microtensile bond strength testing.
Microtensile Bond Strength (μTBS)
The cross-sectional area of each stick was confirmed using a digital caliper (Absolute Digimatic, Mitutoyo, Tokyo, Japan) with a precision of 0.01 mm. Each bonded stick was attached to a jig for microtensile testing using cyanoacrylate resin (Super Bonder Gel, Loctite, São Paulo, Brazil) and subjected to a tensile force using a universal testing machine (Kratos, São Paulo, SP, Brazil) at a crosshead speed of 0.5 mm/min. The failure modes were evaluated under stereomicroscopy at 40× magnification and classified as cohesive, adhesive, or mixed.
In Situ Degree of Conversion Within the Hybrid Layer (DC)
The degree of conversion (on prepared resin-dentin sticks) was analyzed using micro-Raman spectroscopy30 (Horiba Scientific Xplora, Villeneuve d'Ascq, France) with a 785-nm diode laser through a ×100/0.9 NA air objective. The Raman signal was acquired with 600 lines/mm grafting, centered between 800 and 1800 cm−1, and the setting parameters were as follows: 100 mW, spatial resolution of 3 μm, spectral resolution of 5 cm−1, accumulation time of 30 seconds, with five co-additions. Spectra were obtained at the dentin-adhesive interface at three random sites (per stick) within the intertubular-infiltrated dentin. Spectra of uncured adhesives were taken as references. The ratio of double-bonds of uncured and after curing (monomer to polymer) in the adhesive was calculated using the following formula: DC (%) = (1 − [R cured/R uncured]) × 100, where R is the ratio of aliphatic and aromatic peak intensities at 1639 cm−1 and 1609 cm−1 in cured and uncured adhesives, respectively.
Morphologic Analysis of the Hybrid Layer
Specimens were polished using wet SiC paper (grits #1500, 2000, and 2500). After ultrasonic cleaning, specimens were demineralized in HCl (6 N) for 30 seconds and deproteinized in 1% NaOCl for 10 minutes to reveal the hybrid layer. Specimens were dehydrated in ascending grades of ethanol: 25% (20 minutes), 50% (20 minutes), 75% (20 minutes), 95% (30 minutes), and 100% (60 minutes).31 Following preparation, the specimens were mounted and sputter coated with gold-palladium in a vacuum evaporator (SCD 050, Balzers Union, Balzers, Liechtenstein), and the entire surface was examined using a scanning electron microscope (Vega, Tescan, Warrendale, PA, USA). Three photomicrographs of representative surface areas were taken at 5000× magnification.
Statistical Analysis
The values obtained for specimens from the same experimental unit were averaged for statistical purposes for μTBS (MPa) and DC (%). Sticks with premature and cohesive failures were not included in the calculation of mean value for the tooth due to their low frequency in the experiment. The Kolmogorov-Smirnov test was used to assess whether the data from these tests followed a normal distribution. Bartlett's test was performed to determine whether the assumption of equal variances was valid (data not shown). After observing normality and equality of the variances, the data from μTBS (MPa) and DC (%) were subjected to a three-way analysis of variance (type of dentin, adhesive system, and adhesive strategy). Tukey's test was used for pairwise comparisons for all analyses (α=0.05).
RESULTS
Microtensile Bond Strength
Approximately 13-15 bonded sticks were obtained per tooth, including the pretest failures. The mean cross-sectional area was 0.94 ± 0.8 mm2, and no differences were detected among the experimental groups (p>0.05). The failure modes are shown in Table 2. The number of dentin cohesive failures increased in CAD compared with SD.
Means and SDs obtained from μTBS for all experimental groups are shown in Table 3. The cross-product interaction was statistically significant (p<0.0001). The higher mean μTBS values were observed for sound dentin compared with caries-affected dentin, regardless of the adhesive system or strategy (p<0.0001).
When Scotchbond Universal (SBU) was applied in the self-etch or etch-and-rinse mode, the highest mean μTBS values were observed in CAD and SD, which were statistically higher than those obtained for Futurabond U (FBU), and Prime&Bond Elect (PBE) for both strategies in both substrates (p<0.001). When FBU was compared with PBE, the former presented higher mean μTBS values for SD in the self-etch mode and higher mean μTBS values for CAD in the etch-and-rinse and self-etch mode compared with PBE (p<0.001).
When both strategies were compared, the only significant difference was observed for SBU in SD. Higher mean μTBS values were achieved when SBU was applied using the etch-and-rinse mode compared with the self-etch mode (p<0.001).
In Situ Degree of Conversion Within the Hybrid Layer (DC)
The means and SDs obtained from the degree of conversion within the hybrid layer are shown in Table 4. The cross-product interaction was statistically significant (p<0.001). Higher mean DC values were observed for SD compared with CAD for both SBU and FBU (p<0.001). However, no significant difference was observed for PBE regardless of the type of dentin or adhesive strategy (p>0.05).
Means and SDs (%) Obtained for Degree of Conversion Within the Hybrid Layer for All Experimental Groupsa

When SBU was applied in either the self-etch or etch-and-rinse modes, the highest mean DC values were observed for SD, which were statistically higher compared with those obtained from FBU (only in the self-etch strategy) and PBE for both strategies and both substrates (p<0.001). However, when comparing CAD results, the highest mean DC values were achieved with PBE, which were statistically higher than those obtained from FBU (p<0.001). The results from SBU in CAD showed intermediary values.
Morphologic Analysis of the Hybrid Layer
Photomicrographs (5000×) obtained from the morphologic analysis of all resin-dentin interfaces for both SD and CAD by scanning electronic microscopy (SEM) are shown in Figure 1. For SD, a hybrid layer with more integrity was observed with a greater number and longer resin tags compared with CAD. For all systems, these features were more likely to occur in the etch-and-rinse technique compared with the self-etch technique. CAD images presented hybrid layers with scarce and incomplete tag formation for all adhesive systems.
Photomicrographs (magnification ×5000) obtained by scanning electron microscopy of all experimental groups. In the photomicrographs, it is possible to observe the variability of the hybrid layer formation (HL) between the SD and CAD. Note in SD more regularity and integrity of the hybrid layers, showing higher quantity and formation of longer resin tags (hands), mainly when the adhesives systems were applied in etch-and-rinse mode. The HLs promoted on CAD were evidenced by more porous zones, with incomplete resin tag formation. Observe in SBU, the HL showed better resin infiltration promoting HLs with more integrity than in FBU and PBE, regardless of substrate and adhesive strategy. Contrarily, PBE exhibited more disorganized HLs, for SD when applied in the self-etch mode. It is possible to identify confluence of tubules demonstrating poor adhesive penetration and collagen fibril encapsulation, in CAD note the very short and scarce resin tags present for the etch-and-rinse mode, and in the self-etch mode, there is almost total absence of hybridization signals (arrows). For FBU it is possible to observe intermediary hybridization performance, where in CAD, there was slight resin tag formation (arrows) in the self-etch mode and relatively longer tags in the etch-and-rinse mode. Abbreviations: C, composite resin; CAD, caries-affected dentin; HL, hybrid layer; SD, sound dentin.
Photomicrographs (magnification ×5000) obtained by scanning electron microscopy of all experimental groups. In the photomicrographs, it is possible to observe the variability of the hybrid layer formation (HL) between the SD and CAD. Note in SD more regularity and integrity of the hybrid layers, showing higher quantity and formation of longer resin tags (hands), mainly when the adhesives systems were applied in etch-and-rinse mode. The HLs promoted on CAD were evidenced by more porous zones, with incomplete resin tag formation. Observe in SBU, the HL showed better resin infiltration promoting HLs with more integrity than in FBU and PBE, regardless of substrate and adhesive strategy. Contrarily, PBE exhibited more disorganized HLs, for SD when applied in the self-etch mode. It is possible to identify confluence of tubules demonstrating poor adhesive penetration and collagen fibril encapsulation, in CAD note the very short and scarce resin tags present for the etch-and-rinse mode, and in the self-etch mode, there is almost total absence of hybridization signals (arrows). For FBU it is possible to observe intermediary hybridization performance, where in CAD, there was slight resin tag formation (arrows) in the self-etch mode and relatively longer tags in the etch-and-rinse mode. Abbreviations: C, composite resin; CAD, caries-affected dentin; HL, hybrid layer; SD, sound dentin.
When comparing the adhesive systems, SBU demonstrated greater frequency of tag formation for both strategies and substrates of FBU and PBE. FBU demonstrated more regular resin tag formation compared with PBE. When evaluating the self-etch mode, resin tag formation was more evident for SD, whereas PBE showed more collapsed collagen fibrils with more porosity signals. FBU demonstrated few and short needlelike resin tags with the CAD substrate, although with a higher intensity than PBE. When using the etch-and-rinse mode for the CAD substrate, PBE demonstrated minimal resin tag formation, while resin tags were mostly absent when using the self-etch mode.
DISCUSSION
This is the first study that evaluated the bonding performance of universal adhesive systems on natural dentin caries. Although other studies have evaluated the bonding performance of universal systems on CAD, some in primary dentin,9,13 and others in permanent dentin,10,32 none of them evaluated natural caries lesions. However, studies involving natural caries lesions have an important disadvantage, which is the difficulty in standardizing the substrate for all evaluated teeth, and this, perhaps, is a potential limitation for studies using this experimental design. Several artificially induced caries protocols have been developed (using acid-gels,33 pH cycling,34 and microbiological agents35) with the intent of inducing changes to the dental substrate and allowing an easier understanding of the results.14 However, the chemical caries induction methods have demonstrated controversial results due to the difference between the levels of demineralization and the duration of the demineralization and remineralization cycles not being standardized in pH cyclic protocols.14 Additionally, the microbiological protocols provide softness, color alterations, and the presence of distinct zones,14,35 but there are no standardized protocols for this method36 and the few studies in the literature vary with the type of teeth evaluated, creating a situation where the studies are not comparable with each other.14,33,37,38
The events involved in the natural caries process promote a more dynamic environment. Natural caries is a diffusion-controlled process, involving not only the chemical dissolution of the inorganic material but also the exposure and the degradation of the organic matrix.39,40 The different bacterial populations involved in the caries process drop the pH at different times,35 creating a more challenging condition than the laboratory protocols. In addition to demineralization, the exposed organic matrix is degraded by the proteinases from microorganisms of the carious process and the host matrix metalloproteinase activated by bacterial acids, quantitatively and qualitatively affecting the structural substrate.15,41 It is likely that these events combine to promote a distinct clinical condition that is, probably, more difficult to be completely simulated by artificial methods.
The current results showed that the bond strength was reduced when the bonding substrate was caries-affected dentin, regardless of the adhesive system (or technique) used. Thus, the first null hypothesis was rejected. These results were expected because the caries process is associated with the dynamic events of mineral loss and gradual denaturation of collagen fibrils.7,42 These events can result in increased porosity in the intertubular dentin,7 which can contribute to poor hybridization of the adhesive system for CAD,18 as evidenced by SEM analysis for all CAD groups. Additionally, the reduced biomechanical properties for CAD compared with SD7,42,43 can also be associated with reduced bond strength values for CAD. Unlike the intertubular dentin, the presence of an acid-resistant intratubular mineral deposit18 can make resin tag formation difficult. According to SEM analysis, this tubule blockage in CAD could have interfered with acid-etching (ER strategy) and monomer penetration (ER and SE strategies) compared with SD.
These factors, such as a decrease in the biomechanical properties of dentin and lack of tag formation due to intratubular deposits, could support the lower bonding performance for CAD. Curiously, even when the current study was performed in natural CAD, the results agreed with other studies that used artificial caries protocols in permanent or primary teeth.6,9,13,44,45 It is known that, although the artificial protocols are not able to completely simulate the cascade of events involved in the natural caries process, they do promote isolated alterations in the dentin substrate. According to Marquezan and others,14 the pH-cycling protocol seems to resemble a natural affected caries dentin layer after caries removal. On the other hand, the microbiological method seems to more adequately simulate a dentin caries lesion with an evident infected layer, simulating a lesion prior to caries removal, although the artificial lesions are softer compared with natural lesions. The selection method of caries induction depends on what the researcher desires to study.14 Isolated alterations would lead to better understanding the caries process and the potential effect on the bonding performance to a carious substrate. Thus, even when isolated, the alterations would impair the bonding performance compared with SD, consequently explaining the reason why different experimental designs obtained similar findings in bond strength tests.6,9,13,44,45
Our study also evaluated the degree of conversion within the hybrid layer and found reduced bond strength values for CAD, likely due to the lower degree of conversion inside the hybrid layer of two of three adhesive systems. The probable reason for this result is that the higher water content of CAD43 could compromise the photopolymerization of the adhesive systems. Additionally, the phase separation phenomenon could be more prevalent for CAD compared with SD, because hydrophilic monomers tend to penetrate into the wet substrate, whereas hydrophobic monomers penetrate less,18,46 which could affect the hybrid layer integrity. Furthermore, the phase separation phenomenon could affect the polymerization and, associated with the higher water content, influence the degree of conversion. However, the degree of conversion for PBE was not affected by the substrate.
PBE was the only evaluated adhesive system with acetone in its composition as a solvent. It is known that acetone demonstrates a higher vapor pressure compared with ethanol and/or water (as used for SBU and FBU), which may facilitate evaporation.47 Solvent volatilization can also facilitate the polymerization reaction, because it reduces the distance among monomers, increasing the degree of conversion.48 Although adhesive systems containing acetone as a solvent may demonstrate these advantages, a rapid acetone evaporation might not allow enough time for the monomers to infiltrate into dentin. It has been previously reported in the literature that adhesive systems with similar compositions to PBE did not promote uniform adhesive layer thickness across the interface (confocal microscopy), which requires twice the number of applications than recommended by the manufacturer, to obtain an acceptable resin-dentin bond strength.26,49 Certainly, this could explain the reduced bond strength values observed for PBE and poor hybrid layer formation, mainly for CAD compared with FBU and SBU. The second null hypothesis was rejected, because different adhesive systems resulted in varied bonding performances.
Among the evaluated universal adhesive systems, SBU exhibited the highest bond strength results, regardless of the dentin substrate and technique. Universal adhesive systems have a composition similar to those of one-step self-etch adhesive systems, and most universal systems also contain specific carboxylate and/or phosphate monomers that can bond ionically to the Ca2+ in hydroxyapatite.19 SBU contains the phosphate acid monomer, 10-MDP (10-methacryloyloxydecyl dihydrogen phosphate), which chemically bonds to hydroxyapatite, forming hydrolytically stable calcium salts in the form of “nano-layering” on hydroxyapatite.50,51 Additionally, SBU contains a polyalkenoic acid copolymer in its composition, which interacts with apatite substrates following the same adhesion-decalcification reaction.52,53 Thus, for SBU, both bonding mechanisms promoted micromechanical retention by diffusion of resin monomers and chemical adhesion. In the CAD substrate, even though it is partially demineralized, there is a reduced content of insoluble minerals in the dentinal tubules, which could be speculated to interact with the mineral content, explaining the morphology of the hybrid layer (SEM) obtained by SBU. FBU and PBE do not contain any of these compounds in their formulations; perhaps, the absence of these functional carboxylic or phosphate derivatives of methacrylate could explain the lower bond strength values compared with SBU. However, further studies are needed to determine whether the presence of carboxylate and/or phosphate monomers are able to maintain the bond strength values when subjected to storage and cycling methods, mainly for CAD.
The third null hypothesis was partially rejected, because the results showed that there is not a significant difference between etch-and-rinse and self-etch approaches for CAD compared with SD, indicating that the application of phosphoric acid is an unnecessary step during the bonding procedures for this substrate. This could be explained by the fact that CAD contains more residual mineral β-tricalcium phosphate18 (whitlockite) in the dentinal tubules, which is less soluble than hydroxyapatite at a pH lower than 5.554 compared with SD. Even though the adhesive strategy for universal systems is still controversial in the literature,9,21,22 the bond strength should not be compromised by the strategy used,55 because the ionization process resulting from self-etching acidic monomers should promote similar values compared with the etch-and-rinse mode. However, future studies need to be done evaluating which adhesive strategy is better in CAD after long-term exposure to water.
CONCLUSION
The use of the Scotchbond Universal system seems to be an interesting alternative when bonding to CAD, because it can lead to greater bond strengths and adequate hybrid layer formation compared with other adhesive systems tested. The application of phosphoric acid seems to be an unnecessary step during the bonding procedure to CAD when using universal adhesive systems.
Acknowledgments
The authors thank the support by Foundation for Research and Scientific Development of Maranhão (FAPEMA), the Research Funding supported by University Ceuma, and the interdisciplinary laboratory CLABMU of State University of Ponta Grossa for technique support. This study was partially supported by the National Council for Scientific and Technological Development (CNPq) under grant 305588/2014-1. This study was development during the Visiting Professor Scholarship of Prof Dr Alessandro D. Loguercio in the Ceuma University (São Luiz, MA, Brazil, 2014/2015).
Regulatory Statement
This study was conducted in accordance with all the provisions of the local human subjects oversight committee guidelines and policies of approval of the Ethics Committee of University Ceuma. The approval code for this study is 1.750.969.
Conflict of Interest
The authors of this manuscript certify that they have no proprietary, financial, or other personal interest of any nature or kind in any product, service, and/or company that is presented in this article.