INTRODUCTION

Resin-based luting agents were introduced to overcome the inherent problems of zinc phosphate cements and provide better handling and esthetic properties.¹ To promote adhesion to tooth structures, conventional resin cements use etch-and-rinse or self-etching adhesive systems.² Dentin is the main substrate available for adhesion in prosthetic procedures, especially in vital teeth. Dentin is a hydrated hard tissue in the vital state, when there is an outward flow of dentinal fluid through the dentinal tubules with a positive pulpal pressure, estimated to be approximately 15 cm H₂O.³ Water presents deleterious effects for adhesive procedures, such as the plasticization of the polymer chains, leading to compromised mechanical properties and hydrolytic degradation of resin and collagen fibrils.^4-8

The etch-and-rinse multistep adhesive technique has been considered to be complex and sensitive.⁹ This technique produces a complete smear layer and smear plug removal, increasing outward flow of dentinal fluid.¹⁰ In addition, an incompletely infiltrated hybrid layer has been reported.^11,12 Some adhesive systems behave as semipermeable membranes¹³ and can allow outward fluid flow through the dentin-adhesive interface, even after polymerization.^14-16 These disadvantages probably account for a higher incidence of postoperative sensitivity after bonding procedures,¹⁷ pulpal damage,¹⁸ and premature degradation of the resin-dentin interface.^7,8,19

Self-etching adhesive systems were developed in an attempt to reduce the technique sensitivity of etch-and-rinse systems. Self-etching monomers simultaneously etch and infiltrate dentin, providing micromechanical retention after polymerization.²⁰ The maintenance of smear plugs when self-etching adhesive systems are used minimizes moisture contamination by dentinal fluid transudation when compared with the use of etch-and-rinse adhesives.²¹ Simplified single-step all-in-one self-etching adhesives have been reported to allow water diffusion through the adhesive interface even after polymerization.^13,14,16,22

A new type of luting material that does not require any pretreatment of the tooth surface with adhesive systems has been developed, the so-called self-adhesive cement.^23-25 This material aims to combine the favorable properties of conventional (zinc phosphate, glass ionomer, and polycarboxylate cements) and resin-luting agents.²⁵ After the first self-adhesive cement was developed (RelyX Unicem; 3M ESPE, St Paul, MN), it rapidly gained popularity among clinicians because of its simplified “mistake-free” application technique. Application on smear layer-covered substrates maintains dentin permeability in very low levels,¹⁵ contributing to reduced postoperative sensitivity and lower susceptibility to moisture degradation.²⁶ Its interaction with subjacent dentin and enamel has been suggested to occur through formation of a hybridized complex and chemical interaction with hydroxyapatite.²⁷ However, limited information is available with regard to the bonding mechanism and nanoleakage of such products when they are applied to teeth under simulated pulpal pressure.^25,28-30 Within this context, prolonged storage time with simulated pulpal pressure may compromise the sealing ability and mechanical properties of the bonded interface over time.

The aim of this study was to evaluate the effects of simulated pulpal pressure (SPP) on bond strength and nanoleakage in resin-dentin interfaces produced by different cementation strategies after 12 months. The research hypothesis was that bond strength provided by self-adhesive resin cements in indirect restorations is not as susceptible to the outward flow of dentinal fluid as regular resin-luting agents (resin cements/adhesive systems). In addition, it was anticipated that the changes in the nanoleakage patterns at the adhesive interface as a consequence of SPP are similar for multistep and self-adhesive resin cements.

METHODS AND MATERIALS

Tooth Preparation

One hundred sixty-eight recently extracted caries-free third molars stored in 0.1% thymol (Symrise GmbH, Holzminden, Germany) solution at 4°C for no longer than 3 months were used in this study. Teeth were obtained by protocols that were approved (SISNEP/384) by the review board of Guarulhos University (Guarulhos, São Paulo, Brazil). After disinfection and removal of soft tissues, flat middle depth coronal dentin surfaces were exposed with 600-grit SiC paper (3M of Brazil Ltd, Sumare, Brazil) under running water to create a standardized smear layer. Teeth had their roots removed using a diamond saw (ISOMET, Buehler, Lake Bluff, IL) 2 mm below the cement-enamel junction. Pulpal tissue was gently removed so as not to damage the predentin region.

Teeth were assigned to four experimental groups (with or without SPP and immediate or 12 months), which were distributed into 24 experimental subgroups (n=7) according to the luting technique, pulpal pressure, and storage time. Three self-adhesive cements, RelyX Unicem (UN), RelyX U100 (UC), Clearfil SA Luting (SA), and two conventional luting agents, one that uses a two-step etch-and-rinse adhesive (RelyX ARC [RX]) and one that uses a one-step self-etching adhesive (Panavia F [PF]), were used in this study. An additional group included the use of a two-step self-etching primer adhesive system (Clearfil SE Bond) prior to the application of Panavia F (PS). Luting agents were mixed and placed according to manufacturers' instructions (Table 1).

Table 1:  Cements, Lot Number, Manufacturers, Delivery System, Composition, and Application Technique^a

View Fullscreen

To simulate pulpal pressure on the dentin surface, each tooth was bonded to a Plexiglass platform (3 cm × 3 cm × 0.3 cm) penetrated by an 18-gauge stainless-steel tube and fixed with cyanoacrylate adhesive (Loctite Super Bonder Gel; Henkel, Düsseldorf, Germany). The pulp chamber was filled with distilled water via polyethylene tubing connected to a syringe barrel with 10 mL of distilled water and suspended 15 cm from the tooth crown. Thus, each specimen was connected to a hydraulic pressure device that delivered 15 cm water pressure.³¹ The teeth were kept under hydrostatic pressure for 48 hours or 12 months, starting 24 hours before luting procedures.

RESULTS

μTBS

The mean μTBS values are presented in Table 2. Three-way ANOVA revealed a significant difference for the factor “cements” (p=0.00001), for the factor “pulpal pressure,” and for the interaction “cements × pulpal pressure × storage time” (p=0.00028). For RX, SPP significantly reduced μTBS both at 24 hours and 12 months. However, the negative effects of SPP were more pronounced after 12 months. Storage in water for 12 months without SPP did not influence negatively μTBS values of RX. SPP and long-term storage did not affect μTBS values of PS and UN. When SPP was applied, a considerable reduction in μTBS of PF was recorded at both storage times. Even though the self-adhesive cement UC was positively influenced by SPP at 24 hours, the association of long-term storage and SPP was catastrophic to UC and SA, which presented 100% of pretest failures.

Table 2:  Mean Bond Strength Values in MPa (Standard Deviation) for the Different Resin Cements Applied on Dentin With and Without Simulated Pulpal Pressure and Stored in Water for 24 Hours and 12 Months^a

View Fullscreen

Table 3:  Three-Way Analysis of Variance^a

View Fullscreen

Without SPP at 24 hours, RX and PS presented the highest bond strength values (Table 2). PF, UN, UC, and SA did not present significant differences. With SPP at 24 hours, the highest μTBS values were produced by RX, PS, and UC, which did not differ among them. However, no difference in μTBS values was observed among UC, UN, and SA. PF showed the lowest μTBS but with no significant difference from UN and SA. Without SPP at 12 months, RX presented the highest bond strength values, followed by PS. The lowest μTBS at 12 months without SPP were recorded for PF, UN, UC, and SA, which did not differ among them. After 12 months of storage with SPP, the highest μTBS values were recorded for PS. RX, UN, and PF did not differ from each other after 12 months with SPP. UC and SA did not withstand the challenge and could not be tested.

Distribution of failure modes is presented in Figure 1. Cohesive failure in resin cement and adhesive failure were predominantly observed in RX at 24 hours and PS at both times, regardless of the application of pulpal pressure. At 12 months, RX presented predominantly adhesive failure (AD). On the other hand, the predominant failure mode for all other groups was adhesive between dentin and the luting agent.

View Figure

• Download as Powerpoint
• Download Figure

Nanoleakage

SPP increased nanoleakage in all groups, except for the self-adhesive cements UN and UC at 24 hours. The influence of storage time was material dependent. When RX was applied without SPP, silver deposition was observed in some regions within the hybrid layer (Figure 2). RX with SPP revealed silver deposition within the entire thickness of the hybrid layer and tags (Figure 2). After 12 months, increased silver deposition was observed for RX, increasing considerably more when it was submitted to SSP (Figure 2). PS with or without SPP presented small silver deposition at the bottom of the hybrid layer (Figure 3). After 12 months, no increase in silver deposition was observed for PS without SPP, and a moderate increase was observed with SSP. Some regions of the interface produced by PF without SPP presented adhesive layer, hybrid layer, and tags completely impregnated by silver (Figure 4). PF with SPP presented gaps between the cement layer and dentin (Figure 4). After 12 months, PF with and without SPP presented higher silver deposition than observed at 24 hours.

View Figure

• Download as Powerpoint
• Download Figure

View Figure

• Download as Powerpoint
• Download Figure

View Figure

• Download as Powerpoint
• Download Figure

The self-adhesive cements presented lower silver impregnation than the other cement systems at 24 hours (Figures 5 through 7). UN, UC, and SA presented almost no or very little silver deposition at the cement-dentin interface after 24 hours, with or without SPP (Figures 5 through 7). After 12 months of storage without SPP, a small increase in silver deposition was observed for SA and UC. However, when they were stored with SPP, gaps were frequently observed. The cement-dentin interface could not be observed for SA and UC after 12 months with SPP, because specimens did not survive SEM/TEM preparation procedures.

View Figure

• Download as Powerpoint
• Download Figure

View Figure

• Download as Powerpoint
• Download Figure

View Figure

• Download as Powerpoint
• Download Figure

DISCUSSION

In this study, the etch-and-rinse group presented significantly decreased μTBS when it was submitted to SPP either after 24 hours or 12 months. This finding can be attributed to the diffusion of water into the hybrid layer, as noted by the silver deposition within the hybrid layer (Figure 2). As a consequence, the mechanical properties of the polymer within dentin are compromised by swelling and decreased frictional force between the polymer chains, a process known as plasticization.^5,19,35

Even in the absence of SPP, all conventional cement systems presented a certain degree of nanoleakage, which was located mainly within the hybrid layer (Figures 2 through 4). The degree of silver deposition depends on the adhesive used, application mode, and composition. The presence of water on the adhesive composition plays an important role in both total- and self-etching techniques, as water-based adhesives used in total-etch systems can solvate dried matrices and reexpand dentin collagen.³⁶ Probably, water was not eliminated during the adhesive procedures, producing the nanoleakage patterns observed in Figures 2A and 2A′. Besides the presence of water in their composition, the two-step etch-and-rinse adhesive produces a semipermeable membrane due its high concentration of hydrophilic monomers and solvents.^13,14 Hydrophilic resin monomers attract water molecules and permit their movement from dentin across the adhesive layer through water channels.^16,37 Such a statement is confirmed in Figures 2B and 2D, which show an evident increase in nanoleakage, probably due to fluid flow produced by SPP. This water-filled channel has been considered to be a site of hydrolytic degradation and crack propagation during μTBS testing.^38,39

Another phenomenon that contributes to reduction in μTBS values is water diffusion from the underlying hydrated dentin structure across the polymerized hydrophilic adhesive layer via an osmotic gradient during the slow setting process of these resin cements.¹³ According to Hiraishi and others,¹⁵ continuous water uptake via the adhesive layer could result in an unstable porous region, increasing the degradation along the interface between the adhesive and resin cement. As a result, this bonding interface turns into a weak link when pulpal pressure is simulated.^2,40 Porous regions were observed in fractured specimens of RX with SPP, suggesting the presence of water channels (Figure 2E,F,G). At higher magnification, nonuniform globular structures were identified, which might suggest poor polymerization (Figure 2E,F,G).¹ In addition, backscattered SEM micrographs of PF subjected to SPP after 24 hours and 12 months revealed silver deposition between the adhesive cement and resin (Figure 4), suggesting permeability within this system.

When SPP was applied for 24 hours or 12 months, PF presented a significant reduction in μTBS values. Previous studies have reported a negative influence of pulpal pressure for Panavia F.^2,15 This performance is probably associated with a high concentration of hydrophilic and ionic resin monomers in ED Primer, resulting in the formation of a highly permeable layer after polymerization.^13,41 Because of its high permeability,¹⁵ primed dentin allows water to diffuse from dentin across the hybrid layer and form water droplets in the interface (Figure 4), resulting in low μTBS. Hydrophilic monomers, such as HEMA, can attract water, which impairs the cement polymerization, reducing mechanical properties.⁴¹ This statement can be confirmed by the high amount of silver deposition (Figure 4), even without SPP. As previously demonstrated,⁴² for an ideal bonding performance, the water concentration must be sufficient to provide adequate ionization of the acidic monomers but without lowering the resin concentration too much to optimize their bonding efficacy to dentin. The high water content at the interface during polymerization setting probably contributed to the reduction in monomer concentration when PF was submitted to SPP. This fact can explain the lower μTBS associated with gap formation at the interface. The failure modes for PF were exclusively adhesive between resin cement and dentin. The fracture of PF groups with SPP occurred frequently on top of the hybrid layer, where it was not possible to visualize the tubular lumen, suggesting a poorly cured adhesive layer. Besides, it could be speculated that inhibition of the polymerization of the luting agent (Panavia F) could occur due to the presence of acidic monomers within the ED Primer composition.⁴¹ However, as light activation was performed immediately upon luting, this effect is probably negligible.²⁸

Infiltration of hydrophobic monomers from Clearfil SE Bond into primed dentin along with direct light activation of the adhesive system probably resulted in a better monomer conversion and stronger network polymer within the hybrid and adhesive layers, resulting in the significantly higher μTBS of the PS group in comparison to the values of PF group and in the highest μTBS after 12 months with SPP. This hydrophobic layer can reduce permeability of dentinal fluids between dentin and the luting agent and result in better long-term performance.^2,43,44 In addition, the presence of smear plugs is very important to reduce significantly the rate of fluid flow through the interface even in the presence of intrapulpal pressure, in vivo and in vitro.^14,16,45

The self-adhesive resin cements maintained their μTBS when submitted to SPP for 24 hours. Three different self-adhesive materials were used in this study. The self-adhesive cements RelyX U100 (UC) and RelyX Unicem (UN) were developed by the same manufacturer and are marketed under the same name in some countries. According to the manufacturer, the only difference between these products is the delivery system. While UN requires an activator, triturator, and applicator, UC can be hand mixed. Another self-adhesive cement used was Clearfil SA Luting, which needs to be hand mixed. When SPP is applied, water transudation increases the acidic monomers' aggressiveness, improving smear layer dissolution and dentin demineralization.⁴² It also optimizes the acid-basic reactions, allowing better setting.²⁶ The favorable μTBS and very low silver impregnation (Figures 5 through 7) observed for self-adhesive resin cements have been attributed to the micromechanical retention and chemical interaction between monomer acidic phosphate groups and dentin-enamel hydroxyapatite.^15,25,26 This possible improvement produced by water was evidenced by the μTBS values and favorable nanoleakage patterns observed when UC and UN were initially submitted to SPP (Figures 5 and 6).

However, despite the favorable μTBS results observed at 24 hours with and without SPP, and after 12 months without SPP, long-term storage with SPP resulted in debonding of the indirect restorations cemented with UC and SA prior to testing. The low initial pH of UN and UC (pH<2 in the first minute, according to the manufacturer) and SA (pH 2-3 in the first minute, according to the manufacturer), produces almost no demineralization and hybrid layer formation on dentin surface.^18,28,46-48 The present findings demonstrated that this interaction was not strong enough to resist 12 months of storage with SPP for UC and SA.

During polymerization of the self-adhesive resin cements (UN and UC), an increase in pH from 1 to 7 is observed as a consequence of the reaction between phosphate groups and both alkaline filler particles and hydroxyapatite from enamel and dentin, to neutralize resin acidity.^49,50 The presence of calcium hydroxide can also act in pH neutralization. The pH neutralization results in water formation and a more hydrophilic cement, which enhances the cement's wetting ability on the dentin surface and the cement tolerance to water. Water is crucial for self-adhesive luting agents to release hydrogen ions required for smear layer demineralization⁵¹ and is also reused in the reaction between multifunctional acidic phosphate monomers and alkaline filler particles. Such a phenomenon is speculated to be responsible for a change in the nature of the cement from hydrophilic to hydrophobic, which is thought to improve adhesive stability. This hydrophobic characteristic is desired for bonded interfaces, because it can prevent hydrolytic degradation and consequently improve the long-term durability of indirect restorations. After 12 months of storage with SPP, UN demonstrated bond strength stability and very low silver deposition throughout the study.

Except for the materials that presented a high number of pretest failures, all groups presented acceptable bond strengths throughout the evaluation period. However, the groups that presented the highest bond strength values are more indicated for cementation of indirect restoration in less-retentive cavity preparations.

CONCLUSION

Based on the results of this study, it was concluded that the hydrostatic pulpal pressure effect was material dependent. Storage in water for 12 months without simulated pulpal pressure did not affect the bond strength for all materials. However, long-term storage with simulated pulpal pressure influenced the performance of the etch-and-rinse adhesive system RelyX and the self-adhesive cements U100 and Clearfil SA luting regarding μTBS and nanoleakage patterns. Panavia F+ Clearfil SE Bond and Unicem presented bond strength stability and low silver deposition throughout the study regardless of the application of simulated pulpal pressure.