INTRODUCTION

Resin composite materials are widely used to conservatively restore enamel and dentin after caries or trauma and/or to satisfy esthetic demands. The use of dental adhesive is a necessary step to effectively bond resin composites to dental hard tissues. Dental adhesives rely on micromechanical retention of a resin polymer to enamel and dentin substrates. Such a mechanism occurs through a surface conditioning step, resulting in the superficial tissue demineralization and infiltration of resin monomers, which form resin microtags and a hybrid layer.^1,2 Dental enamel is a highly mineralized tissue composed of 96 wt% mineral, ∼3 wt% water, and only ∼1 wt% residual biomacromolecules.³ On the other hand, dentin presents an intricate mineralized organic matrix comprised of ∼70 wt% mineral, ∼20 wt% organic components (mainly type I collagen), and ∼10 wt% water.⁴ While clinically successful, the adhesion of hydrophobic and hydrophilic blends of resins to the dentin organic matrix is based on a complex and technique-sensitive mechanism contributing to the short service life of resin composite restorations.^2,5 Because dentin is the bulk tissue, the complex composition and structure of dentin serve as important constraints for technological advances in adhesive dentistry.

Phosphoric acid (PA) in the form of liquid or gel, and with concentrations ranging from 30% to 40%, has been almost exclusively used as a surface conditioner of enamel and dentin prior to the placement of dental adhesives.¹ PA surface treatment increases surface wettability, roughness, and hardness of enamel and dentin.⁶ Adequate resin bond strength to PA-etched enamel is highly predictable.^7,8 In dentin, PA demineralizes the peritubular and intertubular dentin, exposing a matrix rich in type I collagen fibrils.^2,9 The high acidity of PA may induce structural changes in dentin collagen⁶ and activates the inactive proforms of endogenous dentin proteases associated with resin-dentin degradation.¹⁰ It has been shown that the depth of dentin demineralization does not correlate with bonding effectiveness,¹⁰ leading to strategies such as short application times,⁶ lower acid concentrations,¹¹ and use of alternative dental surface conditioners.¹²

Glycolic acid (GA) is the smallest of a group of naturally occurring organic acids known as α-hydroxy acids. GA is widely used in dermatology to promote skin chemical peeling,¹³ while poly(lactic-co-glycolic acid) has been used to promote wound healing in skin and bone.¹⁴ GA is colorless, odorless, and water soluble^15,16 and is reported to elevate collagen synthesis and fibroblast proliferation in in vivo and in vitro studies.^16-18 Because of such characteristics, GA is potentially attractive for dental applications as a surface conditioner during restorative procedures. Such application is explored for the first time in this study to promote surface demineralization of enamel and dentin for resin adhesion to dental surfaces.

The aim of this study was to determine the effectiveness of GA as a dental surface conditioner for adhesion of commercially available adhesive systems to enamel and dentin. The enamel and dentin surface demineralization pattern was assessed by microhardness measurement and ultrastructural surface morphology using scanning electron microscopy (SEM). Microtensile bond strength (μTBS) to enamel and dentin was evaluated using three different adhesive systems and two application modes (rubbing and no rubbing). The null hypothesis tested was that surface conditioning of enamel and dentin using GA would result in similar surface etching pattern, surface microhardness, and adhesive bond strength to standard phosphoric acid etching.

METHODS AND MATERIALS

Study Design

The microhardness test was proposed to determine the demineralization effect of experimental etchants in enamel and dentin. The pattern of surface etching was investigated under SEM. The effectiveness of GA etching for resin adhesion was investigated by μTBS test on enamel and dentin using different etchant application modes (rubbing and no rubbing) and three adhesives (Adper Single Bond [SB; 3M ESPE, St Paul, MN, USA], One Step Plus [OSP; Bisco Inc, Schaumburg, IL, USA], and Scotchbond Universal Adhesive [SBU; 3M ESPE]). The morphology of the adhesive interfaces was assessed by SEM. Preparation of acid solutions, composition of adhesives, application modes, and manufacturers are detailed in Table 1.

Table 1 Materials, Application Mode, and Manufacturer (Batch No.) of Acid Solutions and Adhesive Systems

View Fullscreen

Extracted human central incisors and molars were selected, cleaned, and kept frozen (−20°C) until use. Studies on enamel and dentin were carried out on the polished enamel surface of central incisors and the occlusal dentin surface of third molars, respectively.

RESULTS

The microhardness results are shown in Table 2. Statistically significant differences were observed among groups for enamel and dentin (p<0.001). Surface treatment with either PA or GA resulted in a statistically significant decrease in the enamel microhardness (p<0.001 and p=0.022, respectively), while GA was significantly less aggressive than PA (p<0.001). Both acids reduced the dentin surface microhardness when compared to the control, with no statistically significant differences between PA and GA (p=0.810).

Table 2 Dentin and Enamel Knoop Microhardness Values (KHN) (Mean [Standard Deviations]) for the Etching Protocols^a

View Fullscreen

The enamel and dentin surfaces etched with PA and GA revealed a similar etching pattern. It is evident that both etchants exposed enamel prism rods as a result of the preferential interprismatic apatite dissolution (Figure 1a-d). Both etchants removed the smear layer, dissolved intratubular dentin, and exposed collagen fibrils, which is more evident within the dentin tubules (Figure 1e-h). No visual variations could be observed between active and passive applications.

View Figure

• Download as Powerpoint
• Download Figure

The enamel and dentin μTBS results are depicted in Tables 3 and 4, respectively. On enamel, there were no statistically significant interactions among factors (adhesives vs acids vs application modes, p>0.05) and between acids (p=0.602). Statistically significant differences were found between application modes (p=0.003) and adhesive systems (p<0.001). Rubbing the etchant yielded significantly higher bond strength when compared to use of the no rubbing mode (p=0.0030). The enamel bond strength of SBU was significantly higher than that of SB and OSP (p<0.001), with no statistically significant different between SB and OSP (p=0.228). On dentin, significant interactions were observed among factors (adhesives vs acids vs application modes, p<0.05), except for adhesive vs acid (p=0.882). Overall, SBU presented the highest bond strength values, with no statistical difference compared to OSP (p=0.110). The dentin bond strength of the three adhesive systems was differently affected by the acid and application mode, as detailed in Table 4. Most notably, GA significantly increased the bond strength of SB and OSP when compared to PA under rubbing mode (p>0.05), and the bond strength of SBU was not significantly affected by the etchant (p<0.05).

Table 3 Results of the Microtensile Bond Strength (μTBS) to Enamel (Means [Standard Deviations]) and Failure Mode of the Experimental Groups^a

View Fullscreen

Table 4: Results of the Microtensile Bond Strength (μTBS) to Dentin (Means [Standard Deviations]) and Failure Mode of the Experimental Groups^a

View Fullscreen

Statistically significant differences in failure mode distribution were observed for enamel (p=0.0106) and dentin (p<0.0001) adhesive interfaces. No cohesive failures were observed for enamel. Adhesive failure was predominant for SB and OSP, and there was an increase in mixed failures for SBU, for both GA and PA (Table 3). In dentin interfaces, all failure modes were observed, but mixed and cohesive modes in resin failures were predominant.

Representative SEM images of the resin-enamel interface are shown in Figure 2. The interfaces of all adhesives exhibited similar morphology, with the presence of resin tags and micro-tags.

View Figure

• Download as Powerpoint
• Download Figure

No visual differences were observed between etchant application modes (rubbing or no rubbing). However, the adhesive layers were thicker for the SBU when compared to the SB and OSP. Representative SEM images of the resin-dentin interfaces are shown in Figure 3. The presence of resin tags in dentin tubules was seen in all groups. An apparently thicker hybrid layer could be noted in PA when compared to GA.

View Figure

• Download as Powerpoint
• Download Figure

DISCUSSION

In addition to being simple and clinically acceptable, acid etching is a required step in bonding techniques. The rationale for the current study is based on exploring alternative material with which to etch enamel and dentin using a less aggressive and potentially more biocompatible acid. The treatment with GA etched the surfaces of enamel and dentin, leading to significant changes in the surface microhardness and etching pattern. Such changes favored bonding of the three different commercially available dentin adhesives to enamel and dentin, with bond strength values comparable to those of PA-treated surfaces. Thus, the null hypothesis was partially rejected.

In this study, the decrease in enamel microhardness mean values was significantly greater for PA than for GA, while similar decreases in the microhardness of dentin were observed for both acids. The lower pH of PA (pH=0.12) compared to GA (pH=1.2) may have accounted for a more aggressive pattern of demineralization of the highly mineralized enamel. Furthermore, the pK_a of PA is only slightly superior to that of GA (2.16 and 3.83, respectively); thus, the acid dissociation is relatively strong for both solutions. Similar superficial demineralization patterns were observed for both acid solutions (Figure 1), although the interfacial microscopy data support a deeper demineralization of dentin by PA (Figure 3). Moreover, bond strength values for GA etching in enamel and dentin were comparable to those obtained with PA etching, with or without rubbing (Tables 3 and 4), suggesting that the depth of demineralization did not have a significant effect on the bond strength of the investigated dental adhesives. Taken together, these findings indicate that GA can be used to etch enamel and dentin in a clinically relevant condition while being relatively less aggressive than PA.

In addition to removal of minerals, PA can also interfere with dentin organic components. It has the potential to activate endogenous gelatinolytic/collagenolytic enzymes (matrix metalloproteinases and cysteine cathepsins) able to cleave collagen at the bonded interface over time, accelerating the degradation of the adhesive interface.^23-25 Such activation is of particular concern in dentin and the pH of the etching solution.²⁶ The effect of GA on the activation of endogenous proteases needs to be further investigated.

The findings demonstrate that acid rubbing during the etching procedure can significantly affect the adhesive bond strength to enamel for both PA and GA. Previous studies^27-30 have shown that the active application mode of adhesive itself may improve the immediate and long-term bonding performance of self-etching systems to dentin. Active application of the adhesive improves performance of universal adhesives to enamel when compared to passive mode, with results similar to those associated with the etch-and-rinse technique.³⁰ In the present study, SEM analyses showed no apparent morphological differences in PA- or GA-etched surfaces with or without rubbing that could explain the outcomes. It may be possible that rubbing of the acid on enamel may have disturbed the exposed hydroxyapatite crystals and chemical composition and weakened the micromechanical retention of the enamel microtags, apparently more significantly than was seen with the GA-etched surfaces. Further characterization of mineral composition at the enamel surfaces and enamel microtag bonding mechanisms of GA to enamel could clarify the effect of application mode.

Hence, adhesive failures were predominant with SB and OSP, implying that the weak link was the bond between the resin and enamel. SBU exhibited predominantly mixed failures (Figure 3), with similar distribution between acid and acid application mode. Although mode of failure in dentin showed variation among the three adhesives, there was no difference in the failure pattern between the GA and PA groups (Table 4).

The enamel and dentin bond strength results indicate that GA is effective in promoting adhesion to dental tissues, regardless of the application mode and adhesive chemistry. It is important to emphasize that the complex composition of dentin warrants a sophisticated bonding mechanism and protocol, in which the water/solvent evaporation after demineralization is crucial and is frequently not done completely.³¹ Since GA has lower depth of demineralization than PA, it is plausible that deep areas of residual water/solvent are not as frequent as in PA-treated dentin. In addition, the formation of an apparently thinner hybrid layer was evident when using the GA (Figure 3). However, hybrid layer thickness does not correlate with bond strength or better bonding quality,³² as was also observed in the dentin bond strength data. It can be speculated that the less aggressive GA associated with the formation of thinner hybrid layers might reduce the discrepancies between demineralization and adhesive infiltration, potentially resulting in less exposed partially demineralized dentin areas prone to degradation. Long-term data and permeability studies to evaluate the hybrid layer overtime are necessary to confirm such an assumption.

CONCLUSIONS

The demineralization with either PA or GA significantly decreased enamel and dentin surface microhardness and produced similar surface demineralization morphology. The etching with GA in enamel and dentin resulted in immediate bond strength values comparable to those of PA using three different adhesive systems and adhesion strategies. GA and PA exhibited similar etching effectiveness at clinically relevant application times.