INTRODUCTION

Over the past 50 years, the practice of adhesive dentistry has changed dramatically. Even if some aspects of cemented prosthodontic restorations have been superseded by dental implantology, adhesive dentistry remains the core of the therapeutic arsenal of restorative dentistry. The principles of adhesion to enamel and dentin, set by the pioneering work of Buonocore,¹ have evolved over the years. Starting from the original total-etch multistep systems, we have witnessed the introduction of total-etch two-step systems, self-etch two-step systems, and finally, self-etch one-step systems, all in an effort to shorten and simplify the application procedure.^2-4 In the case of total-etch two-step systems, the problem of determining the appropriate dentin moisture required for successful bonding renders them technique sensitive.⁵ Concerns related to the continuous etching of the substratum by self-etch based systems have been raised,^6,7 along with concerns related to their ability to adequately precondition enamel. Moreover, incompatibility between self-etch based systems and chemically cured resin composites has been identified and the mechanism has been elucidated.^8-11 In spite of these concerns, the attractive aspect of a shorter and simpler application procedure has led to a relatively widespread use of self-etch based systems.^12-14 It should be emphasized, however, that total-etch multistep systems afford the most reliable performance and are considered the gold standard.^2,15

The evolution of adhesive systems was paralleled by the evolution of luting agents.^16,17 Major landmarks in this evolution were the introduction of zinc polycarboxylate cement¹⁸ (the first cement able to chemically bond to hard tooth tissues), glass ionomer cements,¹⁹ resin modified glass ionomer cements,²⁰ resin cements, adhesive resin cements, and finally, self-etch adhesive resin cements (SEARCs),^4,17 which have evolved along the principles of self-etch bonding systems. The latter seem to have all the required properties of an ideal luting agent: a resin component to reduce solubility and brittleness, adequate esthetic properties, the ability to precondition and chemically bond to hard tooth tissues, and a shorter and simplified application procedure.^21,22 However, as in the case of self-etch based adhesive systems, the adherence of SEARCs degrades over time.^23,24

The aim of this study was to assess the effect of water storage on the flexural strength (σ_f) of four commercially available SEARCs and on the dentin-metal shear bond strength (SBS) afforded by them. To simulate a clinically relevant dentin-SEARC-metal interfacial structure, we chose to lute Grade 3 commercially pure titanium (Ti)²⁵ rods to human dentin. The biocompatibility and corrosion resistance of Ti explain its increased use in dentistry. Several recent publications have investigated bonding to Ti.^26-29 The use of Ti for crown/bridge applications involves attachment to dentin, which was the rationale for selecting Ti for this study. The two null hypotheses tested were 1) there is no difference between the four selected SEARCs with regards to their σ_f and the SBS of dentin-Ti adhesive interfaces mediated by them, and 2) water storage does not affect the σ_f of the four selected SEARCs or the SBS of dentin-Ti adhesive interfaces mediated by them.

METHODS AND MATERIALS

The end surfaces of grade 3 commercially pure Ti rods (3 mm diameter and 20 mm height) were polished perpendicularly to the long axis on 800-grit silicon carbide (SiC) discs under water irrigation, followed by sandblasting (0.2 MPa pressure) with 50 μm alumina. Before luting, the Ti rods were sonicated in acetone for 3 minutes.

The four SEARCs selected for this study, along with pertaining information and group codes, are listed in Table 1. We used 200 caries-free human third molars (with bioethical approval from Université Paris-Descartes in conformity with Law Nr. 2004-800, August 6, 2004) within three months after extraction. The teeth were stored in an aqueous 1% chloramine-T solution at 4°C until used. After the occlusal enamel was flattened, the samples were embedded in self-curing acrylic resin (Plexil A6, Escil, Chassieu, France) in a cylindrical mold. The embedded samples were ground perpendicular to the long axis of the tooth, using 800 grit SiC paper under running tap water, to expose an approximately 3×3 mm area of coronal dentin. The opposite side of the resin cylinder was ground parallel to it (checked with a bubble level). Prepared specimens were stored in distilled water at 37°C until used. They were randomly allocated to the four experimental groups, one for each SEARC. The bonding protocol followed manufacturers' instructions.

Table 1:  Details Regarding the Self-etching Adhesive Resin Cements Used

View Fullscreen

After the respective SEARCs were applied to dentin, the resin cylinder containing the tooth was fixed in a special device (Figure 1a) to allow control of cement thickness and perpendicular alignment of the Ti rods. The Ti rod was brought into contact with the cement layer and placed under a 70 g load (Figure 1b). The excess cement was delicately removed immediately after assembly, before curing. In the dual-cure mode, the light source was a Demetron LC Curing Light (Kerr Corporation, Orange, CA, USA), activated for 40 seconds at four diametrically opposed locations around the sample, at the level of the adhesive interface, for a total of 160 seconds of curing time. Each bonded sample (Figure 1c) was maintained under load for 10 minutes in a dry place at room temperature, as per manufacturers' instructions. The specimens were then stored in tap water at 37°C for 1 hour, 1 day, 7 days, 30 days, and 60 days before SBS determination (n=10/cement /time period). The specimens were tested using a universal testing machine (LRX, JJ Lloyd Instruments, West Sussex, UK) at a crosshead speed of 0.5 mm/min. Fractured interfaces were viewed and photographed under an optical microscope (40×) to assess the locus and mode of failure.

View Figure

• Download as Powerpoint
• Download Figure

Using Teflon molds, 50 bars (2×2×25 mm) were made from each SEARC and stored in water at 37 °C for 1 hour, 1 day, 7 days, 30 days, and 60 days (n=10/material/time) before testing in three-point bending mode for σ_f determination. The specimens were tested using a three-point bending device (with a 20 mm span) and the universal testing machine operated at a cross head speed of 0.1 mm/min. The load at fracture, F, was used to calculate σ_f using the formula:

where L is the specimen span, h the specimen width, and c the specimen height.

The results of SBS and σ_f were statistically analyzed (PASW Statistics 18) using two-way ANOVA followed, if warranted, by Scheffé multiple means comparisons (α=0.05). Pearson's correlation coefficient was determined to assess correlation between SBS and σ_f.

RESULTS

The results of SBS and σ_f, along with the results of the statistical analysis, are summarized in Tables 2 and 3, respectively. The results showed that within each group SBS remained unaffected by water storage, with the exception of MXC, where a significant drop in SBS was detected after 1 day but no deterioration was detected thereafter. Comparing the SBS of the different groups at the same time interval showed that the RXU had significantly higher results, with the exception of the 7-day and 30-day results, which were not different from those of GC. The results obtained at 7 days, 30 days, and 60 days for MXC and SMC2 were significantly lower than the others but not different between themselves.

Table 2:  Results of Shear Bond Strength (SBS) Tests (Mean ± SD) and Statistical Analysis*

View Fullscreen

Table 3:  Results of Flexural Strength (σ_f) Tests (Mean ± SD) and Statistical Analysis*

View Fullscreen

The characterization of the mode of failure of the tested adhesive interfaces indicated that all failures were adhesive at the dentin surface; therefore, no further analysis was warranted.

The results showed that within each material group σ_f increased up to 7 days and thereafter remained either relatively constant over the 60 days (RXU, MXC, GC) or decreased (SMC2) (see Table 3 rows, small letter superscripts). Comparing the σ_f of the different groups at the same time interval (see Table 3 columns, capital letter superscripts) showed that even though there were some significant differences between the groups at 1 hour, 1 day, and 7 days, there were no differences between the materials investigated by 30 days. At 60 days, however, the results split again into three homogeneous groups with GC > RXU = MXC > SMC2. Overall, with regards to σ_f, after full maturation (7 days), only SMC2 was affected by long-term water storage (60 days).

Based on the results obtained, both null hypotheses were rejected.

DISCUSSION

The aim of this study was to assess the effect of water storage, between 1 hour and 60 days, on both the SBS afforded by the four SEARCs and their σ_f. The intent was to monitor the effect of water storage on a relevant mechanical property of the luting agent (σ_f) and investigate if it correlates with the adherence. Although significant differences were identified between the SBS obtained with the four SEARCs, the results of the study showed that water storage affected only one material (SMC2), and the effect was evident only between the results obtained at 1 hour and all the others. The conditions of this in vitro study are, however, different from those found in vivo, where adhesive interfaces are challenged not only by the presence of water at 37°C, but also by the presence of enzymes and exposure to cyclic loading.³⁰ Another possible explanation for the identified long-term stability of the SEARCs investigated is the absence of hydroxyl-ethyl methacrylate from their composition, a component that could render a material susceptible to water degradation.³¹

The acidity (pH) of the two SEARCs that showed better long-term SBS, that is, RXU and GC, have a lower pH (<2) than the other two materials, MXC and SMC2 (pH > 2.2). This difference in pH may have been responsible for an increase in the rugosity/surface area of dentin, leading to higher SBS. Moreover, the presence of urethanes and/or triethylene glycol dimethacrylate in MXC and SMC2 could render these materials more susceptible to water sorption and degradation.³⁰ The large variability of the results did not permit the identification of a significant effect of water storage; however, the 1-hour values for MXC and SMC2 were twice those recorded after 60 days.

The analysis of the fractured adhesive interfaces revealed that all of them failed adhesively at the dentin surface. Our results differ from those reported by Pisani-Proenca, ² who observed a 30% mixed failure in the case of RXU and MXC. The difference between the results may be attributable to the different dentin surface preparations: we used 800 grit SiC while they used 600 grit SiC. A rougher surface may encourage a deeper penetration of the luting agent and provide a larger surface area for bonding, which, in turn, could lead to the mixed failures reported. The use of 800 grit SiC in this study parallels the study by De Munck and others,²² who reported a majority of adhesive failures as well. They also showed that a pH <2 is not necessarily sufficient to adequately demineralize dentin and lead to the formation of a well-defined hybrid layer.²² Moreover, the same authors speculated that the high viscosity of the materials does not allow for close wetting of and penetration into dentin. It has been shown that the application of force could counteract these problems.³³ Use of phosphoric acid^22,32 or polyacrylic acid³⁴ preconditioning before the application of SEARC has also been proposed, but this step seems to defeat the advantages afforded by the use of a SEARC. It has been reported that the use of polyacrylic acid doubled the performance of RXU, a fact explained by the possible interaction between the polyacrylic acid infiltrated into dentin and the fluoro-alumino silicate glass filler present in RXU.^34,35

The results obtained for σ_f showed an increase from 1 hour to 7 days for all the materials, paralleling the completion of their setting reaction and equilibration with water. It is interesting to note that although there was no statistically significant difference between the four SEARCs at 39 days, at 60 days one SEARC (SMC2) exhibited a significant decrease in σ_f. Most studies report mechanical properties within 30 days but it may be advisable to monitor these properties for longer periods to better predict their long-term performance. The SEARCs tested contained groups capable of forming hydrogen bonds, which may be responsible for the relatively high σ_f determined.³⁶ The results of this study did not show any correlation between the results of SBS and those of σ_f, which is probably due to the fact that all the failures were adhesive at the dentin SEARC interface. It is possible that correlations would be found in systems where the mechanical properties of the adhesive are actually challenged by the presence of a strong adherence.

CONCLUSIONS

This study identified significant differences in dentin-Ti SBS afforded by the four SEARCs tested along with significant differences in their σ_f values. The adherence afforded by Rely X Unicem and G-Cem was significantly higher than that of Maxcem and SmartCem2. Under the experimental conditions of this study, 60 days of water storage significantly affected the σ_f of SmartCem2 but did not affect the SBS of SEARC mediated dentin-Ti adhesive interfaces (Maxcem showed a significant drop after 1 day but remained constant thereafter). The advantages offered by SEARCs are appealing and their clinical usage is increasing. However, the selection and use of SEARCs, as with that of any novel materials, should be approached with caution and grounded on evidence-based information.