INTRODUCTION

The concept of bonding to tooth structure was introduced by Buonocore 50 years ago when he introduced bonding acrylic resin to acid-etched enamel.1

Since then, the acid-etch technique has become the standard procedure for enamel surface preparation prior to restoration with resin composites. Furthermore, the total-etch technique and bonding to dentin were also adopted by the profession within the last decade. Bonding to enamel is primarily based on micromechanical interlocking of the bonding resin with created micro-pores on the enamel surface.2 Bonding to dentin is more complex, due to the inherent characteristics and morphological features of dentin with different chemical and histological composition.3 Dentin has an intricate structure with an enormous number of tubules running through it, carrying dentinal fluid from the pulp. This causes the bonding process to be more challenging and, unlike enamel, a primer is always necessary when bonding to dentin to form the hybrid layer.

The introduction of self-etching adhesives eliminated the use of a separate acid-etching step and significantly reduced post-operative sensitivity associated with composite restorations due to incorporation of the smear layer into the bonding interface.4 Self-etching adhesives use acidic monomers that simultaneously condition and prime enamel and dentin and provide vinyl groups for co-polymerization with resins. Their bonding mechanism is based mainly on hybridization (changing the chemical composition of the substrate surface by partially dissolving the surface layer of enamel and dentin, the resultant porosity is then filled by resin).5 In a recent study, De Munck and others compared the bonding effectiveness of one- and two-step self-etch adhesives to the more traditional three-step etch and rinse adhesive.6 The one-step adhesive scored the lowest bond strength, while the two-step self-etch adhesive resulted in bond strength close to that obtained with the three-step etch and rinse adhesive. In another recent study, the shear bond strength of six self-etching adhesives was compared to that of a conventional adhesive.7 Results revealed that the shear bond strength of three self-etch adhesives was similar to that of the conventional adhesive, while the remaining three self-etch adhesives resulted in significantly lower values. Therefore, it seems that not all self-etching adhesives have the same ability to bond to tooth structure.

GV Black introduced the principles of treatment of carious lesions, which consisted of removal of the carious structure, all demineralized dentin and unsupported enamel.8 However, a carious dentin lesion typically consists of a core of soft caries-infected substance surrounded with a peripheral layer of hard caries-affected dentin.9 While it is important to remove all caries-infected substance when preparing cavities, it is desirable to conserve hard affected dentin. Caries-affected dentin is characterized by mineral deposits in the dentinal tubules, which are acid-resistant and act as barriers that decrease infiltration of acid, bacteria and bacterial products.10 It is important to establish the bond strength of new self-etch adhesives to affected dentin.

The development of the microtensile bond strength (μTBS) test by Sano and others permitted measurement of the tensile strength of small specimens (0.5 mm² in cross section).11 Since then, a number of studies reported on bond strength to caries-affected dentin. Nakajima and others evaluated the μTBS of an etch and rinse adhesive to sound and caries-affected dentin under moist and dry conditions.12 Sound dentin showed better results under moist conditions than under dry conditions. However, under moist conditions, no significant difference in bond strength between sound and caries-affected dentin was found.12 This finding was confirmed in another similar study.13 However, Yoshiyama and others, who reported on the μTBS of a conventional adhesive and an experimental self-etch adhesive to sound, caries-affected and caries-infected dentin, found that the bond strengths of both adhesives to sound dentin were significantly higher than those to caries-affected dentin, and that the bond strengths to caries-affected dentin were significantly higher than those found for caries-infected dentin.14 The authors attributed these results to the presence of a more porous hybrid layer in the case of the caries-affected and infected dentin.14 In an attempt to improve bond strength to caries-affected dentin, Arrais and others evaluated the effect of additional and extended etching time.15 Mean μTBS values of two adhesives to caries-affected dentin were significantly lower than values obtained with sound dentin; the additional and extended etching improved bond strength to caries-affected dentin but not to sound dentin. Recently, Nakaronchai and others measured the μTBS of two dentin adhesives to caries-affected and sound dentin in primary teeth.16 Bond strength was significantly higher to caries-affected dentin than to sound dentin when the conventional adhesive was used, but with the self-etch adhesive, no significant difference in bond strength was found.

Because of the conflicting results reported in the above studies regarding bonding to caries-affected dentin, the aim of this study was to measure the μTBS of two recent self-etch adhesives (a two-step and a onestep) and a conventional adhesive to intact and caries-affected dentin with and without thermocycling.

METHODS AND MATERIALS

Thirty extracted permanent molars with occlusal caries, previously stored in 0.5 chloramines solution, were used (Figure 1). The crown portions were horizontally-sectioned at a level 1 mm below the dentino-enamel junction using a slow-speed Isomet saw (Buehler, Lake Bluff, IL, USA) under constant water cooling to expose the flat dentin surface. Both the caries-affected and sound dentin were present on the same surface (Figure 2). The exposed dentin surfaces were examined visually and their surface texture was tested with a small excavator. Surfaces with soft, discolored dentin were considered caries-infected and were excluded. Hard brown dentin was considered as caries-affected, while hard light-yellow dentin was considered to be sound. Dentin surfaces were further examined using a microscope (Wild M3Z Heerbrugg, Heerbrugg, Switzerland) to eliminate specimens that had either enamel or pulpal tissues. The teeth were then randomly assigned to three groups of 10 teeth each, according to the bonding agent used. The bonding agents were applied to the dentin surfaces according to the manufacturers' instructions (Table 1).

View Figure

• Download as Powerpoint
• Download Figure

View Figure

• Download as Powerpoint
• Download Figure

Table 1 Adhesives Investigated in This Study

View Fullscreen

For each tooth, a composite core buildup was made in two increments, each 1.5 mm thick, using Herculite XRV, shade A2 (Kerr Corp, Orange, CA, USA). Light curing was performed with an LED light-curing unit for 40 seconds for each increment (Ultralume5, Ultradent, South Jordan, UT, USA). The specimens were stored in water at 37°C for 24 hours. The teeth were then cut longitudinally into five or six sections perpendicular to the tooth/adhesive interface, with each section being 1 mm thick and 6 mm long using the Isomet saw under water cooling (Figure 3). The sections were left attached to the remainder of the tooth for further sectioning to obtain sticks 1 x 1 mm thick and 6 mm long (Figure 4). The specimens were examined under a microscope using 40x magnification in order to assign them to either caries-affected groups or sound dentin groups. Each group consisted of 50 rods and was divided into two equal subgroups (n=25). One subgroup was tested directly for bond strength, while the other was first thermocycled (Haake, Germany) for 3,000 cycles in water baths maintained at 5°C and 55°C, with a dwell time in each bath of 30 seconds and a 10-second transfer period between the two baths. The specimens were glued to the jig of a microtensile testing machine (BISCO Inc, Schaumburg, IL, USA) (Figure 5) using cyanoacrylate cement (Zapit, Dental Ventures of America, Corona, CA, USA). Tensile load was applied until specimen failure. Failure load was recorded for each specimen and μTBS was then calculated. Means and standard deviations were determined for each group and data were analyzed statistically using ANOVA and Tukey's tests.

View Figure

• Download as Powerpoint
• Download Figure

View Figure

• Download as Powerpoint
• Download Figure

View Figure

• Download as Powerpoint
• Download Figure

RESULTS

Table 2 records means and standard deviations for μTBS values obtained with the 3 bonding agents with and without thermocycling and the means are represented in Figure 6. One-way ANOVA revealed significant differences in mean μTBS values among the groups (p<.0001). Further statistical analysis with Tukey's post hoc test revealed that, when sound dentin specimens were tested without thermocycling, mean μTBS values were not significantly different for all three adhesives. However, with thermocycling, the mean value obtained with SE (22.3 MPa) was significantly higher (p<.016) than the value obtained with SBMP (15.7 MPa) but not significantly different (p=.21) from the value obtained with XEIV (18.3 MPa). Mean μTBS values obtained with XEIV and SBMP were not significantly different when sound dentin specimens were either thermocycled (p=.472) or not thermocycled (p=.93).

Table 2 Mean μTBS and the Standard Deviation in MPa

View Fullscreen

View Figure

• Download as Powerpoint
• Download Figure

For the caries-affected dentin specimens, SE resulted in a mean μTBS value (20.7 MPa) that was significantly higher (p<.015) than the value obtained with XEIV (15.5 MPa) when the specimens were not subjected to thermocycling. However, this mean value was not significantly different (p=.47) from that obtained with SBMP (18.6 MPa). When the caries-affected specimens were subjected to thermocycling, SE also resulted in the highest mean μTBS value (20.2 MPa) that was significantly higher (p<.019) than the value obtained with XEIV (14.8 MPa); however, it was not significantly different (p=.159) from the mean μTBS value obtained with SBMP (16.62 MPa). Mean μTBS values obtained with XEIV and SBMP were not significantly different when caries-affected dentin specimens were either thermocycled (p=.623) or not thermocycled (p=.207).

DISCUSSION

A recently-published study reported mean μTBS for a composite material bonded to the sound dentin of molars with three adhesives (three-step etch and rinse, a two-step and one-step self-etching adhesives) when specimens were subjected to 20,000 thermocycles.17 For the control groups (without thermocycling), mean values ranging from 14.7 to 23.8 MPa were reported, while with thermocycling, the range was from 12.1 to 27.0 MPa. These mean values are comparable to values reported in this study for sound dentin (21.4–24.3 MPa without thermocycling and 15.7–22.3 MPa with thermocycling). However, while the three adhesives used in this study belonged to the same types (three-step etch and rinse, a two-step and one-step self-etching adhesives) as the ones used in the above study, they were not the same brands. Nonetheless, there is good resemblance for the location from which the dentinal specimens were obtained (mid-coronal with perpendicular orientation of the dentinal tubules) in the two studies.17

The μTBS test allows multiple specimens to be obtained from the same tooth. In the current study, sound and caries–affected dentin specimens were obtained from the same tooth. This is important, as it helps to reduce variability among the test groups, and hence, ensure fair comparisons.

Preparation of the μTBS specimens involves sectioning the specimens several times after the bonding procedure is completed, which may lead to pretesting failure due to the vibrations created during sectioning. In this study, pretesting failure occurred with a few specimens from the XEIV group only. No specimens from either the SBMP or SE groups underwent pre-testing failure.

Dentin adhesives tend to function well in bond-strength tests when tested shortly after application.18–19 Clinically, the oral environment represents a challenge to durability of bond strength because of temperature changes, masticatory load-cycling, water sorption and pH fluctuations. Although thermal cycling represents only one of these challenges, the other factors may have detrimental effects on the μTBS. The effect of thermal cycling on the bond strength of four self-etch adhesives was studied by Miyazaki and others.20 They found that up to 30,000 cycles did not have a significant effect on the bond strength of three self-etch adhesives. These results are in agreement with the findings reported in this study. However, the effect of thermal cycling on the bond strength of three one-bottle (etch and rinse) adhesive systems (Single Bond, Bond 1, One Step) and one self-etching adhesive system (Clearfil Liner Bond 2V) was also determined.21 The authors concluded that 5,000 cycles dramatically reduced the bond strength values of all adhesives. In contrast, in another study, thermal cycling did not affect the μTBS values of a conventional adhesive, significantly increased the bond strength of another and decreased the bond strength of a third when the specimens were subjected to 500 thermal cycles. The authors stated that thermal cycling effects are variable and adhesive-specific.22

SBMP is a conventional (three-step etch and rinse) adhesive, with the first step involving etching and rinsing to remove the smear layer and expose a microporous network of collagen that is nearly deprived of hydroxyapatite.23 Priming follows, which enables the applied resin layer to wet the etched surface. The adhesive, which is a Bis-GMA and HEMA resin combined with an initiation system, seals the dentinal surface and provides an interface for bonding to composites. Clearfil SE Bond is a 2-step self-etching adhesive characterized by a relatively mild pH (pH =1.9). When the primer is applied, it partially demineralizes dentin to a depth of 1 μm, which is sufficient to obtain mechanical interlocking through hybridization.24 As no rinsing of the primer is needed, it remains diffused throughout the dentin tissues, preventing collapse of the collagen network. Xeno IV is a one-step self-etching adhesive with an acidic monomer ingredient. It partially demineralizes the subsurface of the dentin, modifies the smear layer and allows infiltration of the adhesive resin through the residual smear layer into the underlying dentin. Xeno IV resulted in inferior μTBS values when used to bond caries-affected dentin perhaps due to the fact that it is more hydrophilic than the other two adhesives.25

In general, there is variability among dentin bond strength values reported by various studies. This may be attributed to different testing methods and conditions, the varying nature of dentin as a substrate and the composite-adhesive used.6–7 Adhesion of resin composite to caries–affected dentin was thought to be inferior to that of sound dentin; 26–27 however, the results of this study revealed no significant difference in μTBS between caries-affected and sound dentin for both the conventional and two-step self-etch adhesives. These results are in agreement with those of two previous studies.1328

CONCLUSIONS

Under the conditions of this study, the microtensile bond strength of composite to sound dentin was not significantly different from that of caries-affected dentin when conventional and two-step self–etching adhesives were used.

A one-step and two-step self-etching adhesive resulted in μTBS values that were not significantly different when used to bond composite to sound dentin, while a highly significant difference was detected when the self-etching adhesives were used to bond to caries-affected dentin with the two-step self-etching adhesive, giving higher values.

Thermocycling significantly decreased μTBS values when sound dentin was bonded to composite with a conventional adhesive. However, no significant effect on bond strength was detected when bonding sound or caries-affected dentin with two other self-etching adhesives.