- Email: [email protected]
Rationale behind the design and comparative evaluation of an all-in-one self-etch model adhesive
journal of dentistry 37 (2009) 432–439
available at www.sciencedirect.com
journal homepage: www.intl.elsevierhealth.com/journals/jden
Rationale behind the design and comparative evaluation of an all-in-one self-etch model adhesive Masafumi Kanehira a, Werner J. Finger b,a,*, Hiroshi Ishihata c, Marcus Hoffmann d, Atsufumi Manabe e, Hidetoshi Shimauchi c, Masashi Komatsu a a
Graduate School of Dentistry, Department of Restorative Dentistry, Division of Operative Dentistry, Tohoku University, Sendai, Japan University of Cologne, Germany c Graduate School of Dentistry, Department of Restorative Dentistry, Division of Periodontology and Endodontology, Tohoku University, Sendai, Japan d Heraeus Kulzer GmbH, Research & Development, Wehrheim, Germany e Department of Clinical Cariology and Endodontology, Showa University, Tokyo, Japan b
article info
abstract
Article history:
Objective: To investigate and compare bonding and dentin sealing efficacy of a marketed all-
Received 20 November 2008
in-one and an experimental model adhesive with minimum effective amounts of acidic
Received in revised form
monomer and water.
27 January 2009
Materials and methods: Composition of model adhesive (NAD) in mass%: UDMA (45), 4-META
Accepted 28 January 2009
(20), H2O (7.5), and acetone (27.5). For characterization of a reasonable NAD application procedure shear bond strengths (SBS, n = 8) were determined on human enamel and dentin. Clearfil S3 Bond (TSB; Kuraray) served as reference. SBSs were evaluated after 10 min, 1 and 7
Keywords:
days, and 1 month, marginal adaptation (n = 8) was assessed in cylindrical butt-joint dentin
Self-etch adhesives
cavities. Diffusive and convective water fluxes through 1 mm thick adhesive-coated dentin
Enamel–dentin bonding
disks (n = 6) were qualitatively and quantitatively analyzed.
Bond strength
Results: SBSs proved that application of NAD in one coat with 20 s agitated dwell time was
Marginal adaptation
20 MPa, enamel SBSs (24 h) were 25 MPa, p > 0.05. Dentin SBSs for TSB and NAD were not
Phase separation
different ( p > 0.05) at the four stages (means: 18.9, 23.5, 25.4, and 23.6 MPa). Five and seven of
Permeability
the eight bonded restorations with TSB and NAD were gap-free ( p > 0.05). Dentin disks treated with EDTA from both sides or one side only were highly permeable for liquid, whereas adhesive-coated dentin disks showed no permeability at 0 and 2.5 kPa water pressure. Conclusions: Within the limitations of this study the model adhesive tested represents a promising basic composition for all-in-one adhesives, eliminating common problems encountered with single step adhesives such as phase separation and permeability. # 2009 Elsevier Ltd. All rights reserved.
1.
Introduction
All-in one or one-step self-etching adhesives are basically simple solutions, containing acidic methacrylate monomers,
either phosphate- or carboxylate-based, and water for ionization of the functional monomer to mediate dental hard tissue demineralization and simultaneous resin infiltration. Further, they contain hydrophobic dimethacrylate monomers
* Corresponding author at: Graduate School of Dentistry, Department of Restorative Dentistry, Division of Operative Dentistry, Tohoku University, 4-1 Seiryo-machi, Aoba-ku, Sendai 980-8575, Miyagi, Japan. E-mail address: [email protected] (W.J. Finger). 0300-5712/$ – see front matter # 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.jdent.2009.01.013
journal of dentistry 37 (2009) 432–439
to establish a cross-linked polymer structure. Co-solvents like ethanol or acetone are necessary to maintain the functional and cross-linking monomers in solution. Such organic solvents form azeotropic mixtures with water and enhance the surface dehydration when evaporated with compressed air. Finally, acid-compatible photoinitiators are added. Several marketed products contain additionally HEMA, an amphiphilic monomer that enhances component miscibility and resin penetration into the hard tissue.1 This complex mixture of diverse nature adhesive components is unequivocally the reason for a number of concerns expressed in current literature. Phase separation has been reported as a consequence of the immiscibility of water with hydrophobic monomers.1–5 Incomplete removal of water prior to adhesive polymerization may affect the curing performance of the adhesives. Several research reports confirm droplet and blister entrapment within the adhesive polymer layer increasing the convective and diffusive water flux from dentin, crossing the adhesive film and jeopardizing the adhesive strength.6–16 Further, the ester groups of the methacrylate cross-linking monomers, the amphiphilic and the acidic monomers of these adhesives are subjected to hydrolysis, leading to reduced shelf life.17–20 Water is essential for ionization of the functional monomers to render the adhesives acidic, however, water remaining upon polymerization leads to the described adverse effects. Water-free adhesives have consequently been proposed as alternatives.3 However, such adhesives would have to rely on the unpredictable amount of water available on or in the bonding substrate surface to gain the necessary demineralization capacity. Another gateway to overcome at least some of the waterrelated problems with all-in-one adhesives might be the reduction of the water concentration to a level where phase separation does not occur.21,22 With this approach it is however mandatory to keep a suitable balance between sufficient water for adequate ionization of the acidic monomer and a resin concentration high enough for good bonding efficacy to dentin. Therefore, in this study we investigated and compared the dentin-bonding efficacy of a simple HEMA-free model adhesive with minimum effective amounts of acidic monomer and water with a marketed all-in one adhesive. The efficacy parameters were bond strengths, marginal adaptation of bonded restorations in dentin cavities, and permeability of the adhesive films polymerized on dentin with or without hydrostatic pressure. The null hypothesis tested was that both adhesives are equally effectively bonding to and sealing dentin.
2.
Materials and methods
As follow-up23 a model adhesive (NAD) with the following components in mass percent was prepared: UDMA: urethanedimethacrylate (45), 4-META: 4-methacryloyloxyethyl trimellitic anhydride (20), water (7.5), acetone (27.5) and activated with camphorchinone (0.3) and coinitiator (0.35). The performance of this adhesive was compared with Clearfil S3 Bond (TSB.LOT: 00100B; Expiry: 2010-02; Kuraray Medical, Okayama, Japan). According to the manufacturer TSB includes bisphenol
433
glycidyl dimethacrylate (Bis-GMA), 2-hydroxyethyl methacrylate (2-HEMA), 10-methacryloyl-oxydecyl dihydrogen phosphate (10-MDP), silanated colloidal silica, water, and ethanol. In a preliminary trial, application details for NAD, namely the number of coating layers needed, the duration of the dwell time, and the effect of agitation are defined using shear bond strength to dentin as the target parameter.
2.1.
Phase separation
Three droplets of the experimental adhesive and of TSB, respectively, were dispensed in a white disposable mixing dish and left at ambient laboratory atmosphere under dimmed light for approximately 10 min. The adhesives were characterized free of phase separation when the solutions remained transparent and homogeneous during this time interval.
2.2.
Shear bond strength (SBS)
Human molars were immediately after extraction rinsed with water and immersed in 1% chloramine solution for a maximum of 6 months prior to experimental use. The teeth were placed with a sound proximal surface on the bottom of rubber molds and embedded with slow curing epoxy resin. Following over night curing peripheral enamel or dentin sites were exposed by wet grinding on SiC papers, grits 180–320, rinsed and air-dried prior to use. The adhesives were dispensed in a disposable dish and immediately applied with a microbrush either according to instructions for TSB, or with 1, 2 or 3 consecutive coats for 10, 20, 30 or 60 s. The effect of agitation of NAD on bond strength was tested with one layer applied and 20 s time. Following thorough air-drying for 10 s the adhesives were light-activated for 20 s with Translux Power Blue (Heraeus Kulzer, Hanau, Germany), clamped in the Ultradent bonding jig (Ultradent Products Inc., South Jordan, USA) under the proprietary 2 mm high and 2.3 mm wide cylindrical plastic mold for bulk-filling with Venus resin composite (Shade A2; Heraeus-Kulzer, Hanau, Germany), and 60 s light activation. From each of the two adhesives eight dentin bonded specimens were immersed into deionized water at 37 8C for 10 min, 24 h, 1 week, or 1 month prior to shear debonding (1 mm/min) with the Ultradent notched steel rod, mounted in a universal testing machine. For comparative enamel bond strengths eight specimens of each adhesive were tested after 24 h storage in 37 8C water. Data were analyzed by one-way ANOVA and Duncan’s post hoc test at a 95% confidence level ( p < 0.05). The modes of fracture were determined at 20-fold magnification and classified as adhesive, cohesive in resin or dentin, or mixed adhesive cohesive failures.
2.3.
Marginal gap formation
Proximal flat peripheral dentin surfaces of non-embedded molars were prepared on wet SiC paper, grit 600. Then, cylindrical cavities (4.5 mm in diameter and 1.5 mm deep) were prepared with a medium-grained diamond bur (approximately 10,000 rpm) into the coronal exposed dentin under copious water.
434
journal of dentistry 37 (2009) 432–439
The adhesives were applied to the rinsed and slightly airdried cavities in one coat for 20 s agitated dwell time, thoroughly air-dried for ten seconds, and light activated for 20 s. The cavities were then bulk filled with Venus resin composite, covered with a transparent matrix strip, light cured for 60 s, and immediately immersed into deionized water at room temperature for 15 min. Following removal of the excess resin composite on SiC paper, grits 600–4000, the restoration margins were inspected under a light microscope at 500-fold magnification. When a gap was detected the width was measured, using an ocular screw micrometer with a reading accuracy of 0.2 mm. For each adhesive eight specimens were prepared. The data were analyzed by Kruskal–Wallis ANOVA and Mann–Whitney’s post hoc test ( p < 0.05).
2.4.
Permeability of the adhesives
For this test human molars, frozen immediately after extraction, were used. The average age of the donors (7 male, 9 female) was 21 years, with a range from 17 through 28 years. All teeth were free of decay and restorations. Parallel, 1 mm thick coronal dentin slices were cut with a diamond wafer saw microtome (Model SP 1600, Leica Microsystems Nusssloch GmbH, Nussloch, Germany) under copious water-cooling, perpendicular to the vertical tooth axis between the occlusal enamel portion and the pulp horns. As a first reference, two dentin slices were cleaned from both sides with neutralized 0.5 M EDTA solution (pH 7.4) for 3 min each, to remove the cutting smear and to open the dentinal tubules, and rinsed with deionized water. As a second reference, another two tooth slices were cleaned with the EDTA solution for 3 min on the pulpal side only, leaving the cutting smear on the occlusal surface. For the final evaluation of adhesives permeability, six slices for each adhesive were treated with EDTA only on the pulpal side. The device used for determination of dentin permeability is fully described in a previous article.23 It is a split chamber column, modified from the model suggested by Pashley et al.24 (Fig. 1). Two cylindrical acrylic chambers are sealed with Orings on each side of a dentin slice and clamped in a metal frame. Each chamber has a liquid intake and a drainage hole. The chamber, fitting the occlusal side of the sandwiched dentin slice, is sealed with a clear glass cover slip. This chamber is filled with an aqueous solution of 0.02% luminol (5amino-2,3-dihydro-1,4-phthalazinedione) in 1% sodium hydroxide, whereas the opposite chamber is filled with the chemiluminescence activating liquid, an aqueous solution of 1% potassium ferricyanide and 0.3% hydrogen peroxide. When the activator containing side of the cell is pressurized the liquid may eventually pass through the dentinal tubules and produce a chemiluminescence reaction upon mixing with the illuminant reagent (Fig. 2). Luminol exhibits luminescence when activated with the oxidant hydrogen peroxide and a hydroxide salt in water as the activator. In the presence of potassium ferricyanide, the hydrogen peroxide is decomposed to form oxygen and water. When luminol reacts with the hydroxide salt, a dianion is formed. The oxygen produced from the hydrogen peroxide then reacts with the luminol dianion. The reaction product is a
Fig. 1 – Components of the split-chamber column used. The two chambers for the reagent solutions are clamped via Orings to the dentin slab, the detection cell is closed with a cover slip for registering of the photochemical signal, and the entire assembly is clamped between metal frames.
organic peroxide that is unstable and decomposes with the loss of nitrogen to produce 3-aminophthalic acid with electrons in an excited state. As the excited state relaxes to the ground state, the excess energy is liberated as photons, visible as blue light. This luminescence signal is recorded with a photodiode (S 6204; Hamamatsu Photonics, Hamamatsu City, Japan), installed 5 mm from the cover slip on the lower chamber. The entire equipment is set up in a lightproof box to prevent any outer light signal to interfere with the luminescence signal. The output voltage of the photodiode is recorded with an AD converter at 1 kHz and stored in the CPU unit controlling the system, from where the data at the end of the experimentation are transferred to a PC for further processing and analysis. The entire procedure is automated in a programmable sequencer. The non-covered dentin specimens were mounted between the split chambers that were filled with distilled water. Liquid pressure of 2.5 kPa was exerted for 120 s from the pulpal side, to flush the dentinal tubules free of extraneous liquid. Consecutively, the illuminant reagent was injected into the occlusal chamber, and the activating solution into the
Fig. 2 – Schematic presentation of the split-chamber device. The activator solution is pressurized, penetrates through the dentin disc, and a photochemical signal is generated upon contact with the luminol/hydroxide solution. The light signal is recorded with a photodetector.
journal of dentistry 37 (2009) 432–439
opposite one. The injection process lasted 45 s, followed by 30 s delay for stabilization of the atmospheric pressure on the luminal side, before the activating solution was pressurized with 2.5 kPa (25 cm H2O or 20 mmHg) for 2 min, and left without liquid pressure for another 2 min. The photochemical signal was recorded with the photodiode. Following this first run, a second wash cycle with water at 2.5 kPa was automatically initiated for 2 min before the chambers were filled with fresh reagent solutions, pressurized again as described above, and finally flushed with water. This entire procedure for determination of baseline permeability data was repeated twice on each dentin specimen. Six dentin specimens for each of the two adhesives were cleansed with EDTA on the pulpal side, rinsed and air-dried. The dentin adhesives, TSB or NAD were applied to the smearcovered occlusal surface, as described above and light-activated for 60 s under a continuous stream of nitrogen, to prevent oxygen from ambient air to interfere with polymerization by inhibition. The coated dentin slices were then mounted into the split-chamber device and investigated, using the same washing and reagent cycles as described for determination of the baseline permeability data. The permeability of each specimen was measured within 1 h after polymerization, after 1 day, 1 week, 1 months, and three months. During the time intervals between consecutive measurements the specimens were stored in deionized water. A special clamping device secured that the specimens were mounted reproducibly with the identical dentin/adhesive surface area exposed. For indirect observation of water crossing through the noncoated and coated dentin specimens an impression technique was used. Specimens, with a water-filled chamber connected to the pulpal side only, were clamped in a stand in the same position as during the photochemical test. Following water pressurizing at 2.5 kPa for 2 min, the free surface was dried with compressed air for 10 s, and immediately coated with freshly mixed hydrophobic condensation curing silicone impression material (Base: RTV 501, LOT 0002994233; Cat: Y, LOT 0002843600. Base-catalyst ratio by wt: 100/5. Dow Corning Toray Co., Ltd., Tokyo, Japan) delivered from a syringe. During impression coating and setting the 2.5 kPa water pressure was either maintained, to simulate the condition of convective flux, or the water pressure was released to zero, to simulate conditions of diffusive water flux. According to the manufacturer the working time of the impression material is 3 min, the setting time 7 min. The impression was left in place for 10 min at ambient temperature prior to removal. For the two reference conditions and for the adhesive covered dentin specimens, two samples each, subjected to the photochemical evaluation before, were randomly selected for this impression technique. The impressions were stored dry at ambient laboratory atmosphere for 24 h prior to mounting on aluminum holders and sputter coating with Pt for SEM inspection at 1000-fold magnification.
3.
Results
Macroscopically, both adhesives remained transparent during the 10 min storage time and were thus characterized as free of phase separation.
435
Fig. 3 – Shear bond strengths of the model adhesive on enamel (gray bars) and dentin (white bars) by agitated dwell time. The adhesives were applied with two consecutive coats. SBS testing after 24 h of water storage. One-way ANOVA revealed highly significant differences between the groups ( p < 0.001). Horizontal lines connect groups that are not significantly different ( p > 0.05).
Fig. 3 shows the box-and-whisker plots (median, interquartile distances, and extreme values) of the shear bond strengths for the model adhesive on enamel and dentin, applied in two layers and left for different agitated dwell times. At none of the dwell times significant differences were detected between enamel and dentin SBS. Ten seconds dwell time results in significantly lower SBS (mean 16.5 MPa) than longer application times. Between 20 and 60 s no SBS differences by dwell time are found (mean 21.7 MPa). The diagram in Fig. 4 illustrates the enamel and dentin SBSs of the model adhesive when applied in 1, 2 or 3 consecutive
Fig. 4 – Shear bond strengths of the model adhesive on enamel (gray bars) and dentin (white bars) by numbers of coats applied and 20 s agitated dwell time. SBS testing after 24 h of water storage. ANOVA revealed no significant differences between the groups ( p = 0.801).
436
journal of dentistry 37 (2009) 432–439
Fig. 5 – Shear bond strengths of the model adhesive on enamel and dentin when applied in one coat with (gray bars) or without (white bars) agitation during 20 s dwell time. SBS testing after 24 h of water storage. One-way ANOVA revealed significant differences between the groups ( p < 0.01). Same letters denote groups that are not statistically different ( p > 0.05).
Fig. 6 – Means and 95% intervals of confidence for shear bond strengths of TSB (open circles) and NAD (gray circles) tested after storage in water for 10 min through 1 month. One-way ANOVA proved significant differences ( p < 0.001). Identical small letters denote homogeneous groups according to Duncan’s multiple comparison testing ( p > 0.05).
layers and agitated during 20 s dwell time. There were no statistically significant differences between the groups ( p = 0.801). The mean bond strength was 21.4 MPa. Fig. 5 shows the effect of adhesive agitation during 20 s dwell time on SBSs to enamel and dentin. One coat of adhesive was applied. Gray bars are with, white bars without agitation. One-way ANOVA shows significant differences between the groups ( p = 0.007). Same lower-case letters denote groups that are not statistically different ( p > 0.05). Agitation gives significantly higher SBS on enamel, whereas agitation has no significantly different effect on dentin SBS. Bond strengths to enamel after 24 h storage in water were not significantly different ( p > 0.05), for TSB (25.5 2.5 MPa) and for NAD (22.4 3.7 MPa). The diagram in Fig. 6 illustrates the mean shear bond strengths to dentin of TSB (white circles) and NAD (gray circles) by storage time prior to debonding. The whiskers depict the 95% confidence intervals of the means. One-way ANOVA showed significant differences between the groups ( p < 0.001), same letters denote homogeneous data subsets ( p > 0.05). At each storage time the SBSs of both products were not significantly different ( p > 0.05). Table 1 summarizes the failure modes of the debonded specimens. No adhesive failures were found, most fractures were located in resin. Marginal adaptation of the bonded Venus restorations in the 4.5 mm wide dentin cavities is not significantly different for the two adhesives ( p = 0.328). With TSB five restorations are gap-free, the other three restorations show 0.6, 0.6, and 0.8 mm wide gaps. NAD resulted in seven gap-free restorations and 1 cavity margin with a 0.4 mm wide gap. Fig. 7 illustrates the characteristic chemiluminescence output in mV for specimens that were treated with EDTA on both cutting surfaces (upper trace). As a result of the diffusion process between the activator and reactant chamber during
injection of the reagents and prior to pressurizing the onset of the photo signal was larger than 100 mV for the non-coated dentin slices. Upon pressurizing a light signal proportional to the amount of trigger liquid reacting with the luminal containing reactant is generated, returning almost to the initial output after 120 s when the pressure is released to zero. The middle curve in the diagram reflects the permeability of dentin disks with cutting smear on the occlusal surface and without smear on the reverse side. The maximum output signal at the end of the 120 s pressurizing time is approximately 50% from specimens that are free of smear on both sides. At the end of the non-pressurized interval an appreciable photo signal is still recorded indicating a continuous reaction between the diffusing activator solution and the reactant. The signal tracing generated during the active pressure and the pressure-released intervals of adhesivecoated dentin disks is a straight line, parallel to the time axis of the diagram, at approximately 40 mV output. This small photo signal reflects the noise level of the instrument, and is hardly
Table 1 – Debonding failures of shear bond strength specimens (n = 8). Clearfil S3 Bond (TSB)
1h 24 h 1 week 1 month
Experimental adhesive (NAD)
CR
CD
RD
AD
CR
CD
RD
7 2 4 6
0 1 1 1
1 5 3 1
0 0 0 0
8 5 7 6
0 1 1 2
0 2 0 0
AD 0 0 0 0
CR: cohesive in resin; CT: cohesive in dentin; RD: cohesive in resin and dentin; AD: adhesive.
journal of dentistry 37 (2009) 432–439
437
SEMs of impressions made from smear-free and smearcovered dentin show numerous droplets (concave depressions) at the interface, both for the pressurized specimens and for specimens with passive water contact at the pulpal side (Fig. 8). Impressions made from smear-free pressurized samples displayed large confluent droplets. In contrast, no droplets were seen on dentin samples coated with TSB or NAD (Fig. 9), irrespective of the applied water pressure and time during which the specimens were stored in water. The rippled surface of the adhesive was produced under the stream of protective nitrogen directed to the surface.
4. Fig. 7 – Characteristic examples of the photochemical output in mV registered for dentin specimens, treated with EDTA from both surfaces (upper curve), cleaned with EDTA on the pulpal side only (middle curve), and coated with adhesive (lower trace, parallel to the time axis).
related to a chemiluminescence reaction. Throughout the course of the experiment with testing stages from 1 h to 3 months virtually the same linear tracings were found, when the permeability of the 6 specimens each, coated either with TSB or NAD was investigated, indicating that there was no activator liquid crossing the adhesive layer.
Discussion
Several commercially available all-in-one adhesives exhibit phase separation upon volatilization of organic solvent and simultaneous occurrence of water-rich blisters due to the immiscibility of hydrophobic and hydrophilic monomers.2–4 One might argue that such water-rich droplets are clinically not critical, since they are readily eliminated during preactivation air-drying. Nevertheless, even elimination of such aqueous blisters prior to polymerization is not the solution of the problem since reportedly the remaining homogeneous monomer-rich phase contains approximately 10% of water that is more difficult to remove than the water-rich phase.4 Other marketed adhesives contain HEMA, obviously added by their manufacturers as a good co-solvent to prevent phase
Fig. 8 – SEM micrographs: (A) specimen covered with smear layer, as produced during diamond wafer cutting. (B) Specimen treated with EDTA. (C) Vinyl polysiloxane impression of smear layer covered dentin slice under passive pulpal side water contact (0 kPa) and (D) at 2.5 kPa water pressure. The concave globular depressions in (C) and (D) represent water droplets that exuded through the dentin tubules and smear layer throughout the 3 min working time of the impression material.
438
journal of dentistry 37 (2009) 432–439
Fig. 9 – Representative SEM micrograph of impression from a dentin slice coated with adhesive. No differences were observed between the two adhesives, the times of water storage or the liquid pressure. No concave depressions were found, indicating no permeability of the adhesivecoated specimens during the 3 min working time of the silicone impression. The wavelike adhesive surface results from the stream of nitrogen blown to the surface before and during light-activation.
separation. One of the disadvantages with HEMA containing adhesive compounds is difficult removal of solvents due to HEMAs polar characteristics, i.e. hydrogen bonding capability. With such adhesives droplets are found in the polymerized adhesive film on dentin as an expression of osmotic effects.2,7,9,13,16 Therefore, it was considered reasonable to design all-inone adhesives with minimal effective contents of water and functional acidic monomer. The present data indicate that a simple light-curing adhesive with 7.5% of water, 20% 4-META, and 45% UDMA dissolved in acetone19 might be a useful approach to overcome some of the problems mentioned. The model adhesive shows no phase separation, and one coat applied for 20 s agitated dwell time is clinically practicable. Its final bond strength to dentin and enamel and the marginal adaptation in cylindrical butt-joint dentin cavities are comparable with the commercial Clearfil S3 Bond. Therefore, the null hypothesis is accepted that both adhesives bond equally effective to dentin and enamel. The smear-layer-free specimens and the samples with intact smear-layer on the occlusal side were clearly permeable for water. In contrast, the permeability tests did not reveal liquid crossing through the adhesive-coated dentin slices for the experimental or for the HEMA-rich commercial adhesive TSB. This observation may be related to the specimen production procedure, where light activation following thorough and strong air-drying was carried out under a stream of protective nitrogen directed to the surface. Further, the adhesives were applied on air-dried dentin and cured immediately after the 20 s dwell time without interference of water. Van Landuyt et al.2,13 pointed out that osmosis occurs when the light-cured adhesive is still permeable enough to
transmit small molecules and when the osmotic pressure is higher at the free, oxygen-inhibited side than at the dentin side of the adhesive. In their study the m-tensile bond strength of HEMA-rich adhesive on dentin was significantly lower when cured after 20 min delay than when light-cured immediately. This confirms that oxygen inhibition and thus the high osmotic pressure is stopped when the free surface is protected from oxygen diffusion either by polymerizing a lining composite on top or by polymerization under an inert atmosphere. Addition of small amounts of HEMA to conventional water-rich adhesives does not prevent phase separation,4,25 whereas large amounts of HEMA render the adhesive polymer more hydrophilic and prone to osmosis, and the resulting polymer less cross-linked and therefore mechanically weaker.12,14,25,26 No permeability of the adhesive-coated dentin slices was recorded even after 3 months storage in water. It may be speculated whether the relatively low water pressure (2.5 kPa) exerted and/or the short pressurizing time used (2 min) were sufficient to make liquid cross from the pulpal specimen side through the adhesive. On the other hand, osmotic blistering in the adhesive film occurs reportedly fast and has been demonstrated already after short-term storage of specimens in water.25 At this point the validity of this new chemiluminescence method to assess adhesive permeability is still not fully clarified, and further research is necessary to evaluate whether, e.g. curing under nitrogen polymerizes the adhesive to a degree that water crossing is prevented without giving evidence for what might happen at the resin–dentin interface. In conclusion, the similarity in performance of Clearfil S3 Bond and the outlined model adhesive indicates that the simple composition of the experimental solution may be a promising base for further development of all-in-one self-etch adhesives. The model adhesive shows no phase separation, bonding strength to dentin and enamel is reasonably high, marginal adaptation of bonded resin composite restorations in cylindrical dentin cavities is very satisfactory, and within the limitations of the permeability test used there are no signs of water crossing immediately after light activation and throughout the 3-month follow-up tests. Studies currently under progress are designed to elucidate possible effects of adjusting of the activating system, combination of the carboxylic functional monomer with phosphate ester monomers, and addition of filler fractions on the physico-mechanical resistance and long-term efficacy of such model adhesives.
Acknowledgment Dr. Masae Furukawa from Showa University in Tokyo is gratefully acknowledged for her laboratory assistance in performing part of the bond strength tests.
references
1. Van Meerbeek B, Van Landuyt K, De Munck J, Hashimoto M, Peumans M, Lambrechts P, et al. Technique sensitivity of contemporary adhesives. Dental Materials Journal 2005;24: 1–13.
journal of dentistry 37 (2009) 432–439
2. Van Landuyt K, De Munck J, Snauwaert J, Coutinho E, Poitevin A, Yoshida K, et al. Monomer-solvent phase separation in one-step self-etch adhesives. Journal of Dental Research 2005;84:183–8. 3. Van Landuyt KL, Mine A, De Munck J, Coutinho E, Peumans M, Jaecques S, et al. Technique sensitivity of water-free onestep adhesives. Dental Materials 2008;24:1258–67. 4. Finger WJ, Shao B, Hoffmann M, Kanehira M, Endo T, Komatsu M. Does application of phase-separated selfetching adhesives affect bond strength? Journal of Adhesive Dentistry 2007;9:169–73. 5. Gaintantzopoulou M, Rahiotis C, Eliades G. Molecular characterization of one-step self-etching adhesives placed on dentin and inert surfaces. Journal of Adhesive Dentistry 2008;10:83–93. 6. Tay FR, Pashley DH, Yoshiyama M. Two modes of nanoleakage expression in single-step adhesives. Journal of Dental Research 2002;81:472–6. 7. Tay FR, King NM, Chan KM, Pashley DH. How can nanoleakage occur in self-etching adhesive systems that demineralize and infiltrate simultaneously? Journal of Adhesive Dentistry 2002;4:255–69. 8. Tay FR, Pashley DH. Water-treeing: a potential mechanism for degradation of dentin adhesives. American Journal of Dentistry 2003;16:6–12. 9. Hashimoto M, Tay FR, Ito S, Sano H, Kaga M, Pashley DH. Permeability of adhesive resin films. Journal of Biomedical Materials Research 2005;74:699–705. 10. Malacarne J, Carvalho RM, de Goes MF, Svizero N, Pashley DH, Tay FR, et al. Water sorption/solubility of dental adhesive resins. Dental Materials 2006;22:973–80. 11. Hashimoto M, Tay FR, Sano H, Kaga M, Pashley DH. Diffusion-induced water movement within resin–dentin bonds during bonding. Journal of Biomedical Materials Research 2006;79:453–8. 12. Hashimoto M, Fujita S, Kaga M, Yawaka Y. Effect of water on bonding of one-bottle self-etching adhesives. Dental Materials Journal 2008;27:172–8. 13. Van Landuyt KL, Snauwaert J, De Munck J, Coutinho E, Poitevin A, Yoshida Y, et al. Origin of interfacial droplets with one-step adhesives. Journal of Dental Research 2007;86:739–44. 14. Sauro S, Pashley DH, Montanari M, Chersoni S, Carvalho RM, Toledano M, et al. Effect of simulated pulpal pressure on
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
439
dentin permeability and adhesion of self-etch adhesives. Dental Materials 2007;23:705–13. Donmez N, Belli S, Pashley DH, Tay FR. Ultrastructural correlates of in vivo/in vitro bond degradation in self-etch adhesives. Journal of Dental Research 2005;84:355–9. Tay FR, Lai CNS, Chersoni S, Pashley DH, Mak YF, Suppa P, et al. Osmotic blistering in enamel bonded with one-step self-etch adhesives. Journal of Dental Research 2004;83: 290–5. Inoue S, Koshiro K, Yoshida Y, De Munck J, Nagakane K, Suzuki K, et al. Hydrolytic stability of self-etch adhesives bonded to dentin. Journal of Dental Research 2005;84: 1160–4. Moszner N, Salz U, Zimmermann J. Chemical aspects of selfetching enamel–dentin adhesives: a systematic review. Dental Materials 2005;21:895–910. Fujita K, Ma S, Li R, Ikemi T, Nishiyama N. Effect of solvent type on the degradation of 4-MET. Dental Materials Journal 2007;26:792–9. Da Silveira Lima G, Ogliari FA, Oliveira da Silva E, Ely C, Demarco FF, Carren˜o NLV, et al. Influence of water concentration in an experimental self-etching primer on the bond strength to dentin. Journal of Adhesive Dentistry 2008;10:167–72. Hiraishi N, Nishiyama N, Ikemura K, Yau JYY, King NM, Tagami J, et al. Water concentration in self-etching primers affect their aggressiveness and bonding efficacy to dentin. Journal of Dental Research 2005;84:653–8. Furukawa M, Shigetani Y, Finger WJ, Hoffmann M, Kanehira M, Endo T, et al. All-in-one self-etch model adhesives: HEMA-free and without phase separation. Journal of Dentistry 2008;36:402–8. Ishihata H, Kanehira M, Nagai T, Finger WJ, Shimauchi H, Komatsu M. Effect of desensitizing agents on permeability of dentin. American Journal of Dentistry 2009, in press. Pashley DH, Stewart FP, Galloway SE. Effects of air-drying in vitro on human dentine permeability. Archives of Oral Biology 1984;29:379–83. Van Landuyt KL, Snauweart J, De Munck J, Peumans M, Lambrechts P, Van Meerbeek B. The role of HEMA in onestep self-etch adhesives. Dental Materials 2008;24:1412–9. Asmussen E, Peutzfeldt A. Influence of selected components on crosslink density in polymer structures. European Journal of Oral Sciences 2001;109:282–5.
available at www.sciencedirect.com
journal homepage: www.intl.elsevierhealth.com/journals/jden
Rationale behind the design and comparative evaluation of an all-in-one self-etch model adhesive Masafumi Kanehira a, Werner J. Finger b,a,*, Hiroshi Ishihata c, Marcus Hoffmann d, Atsufumi Manabe e, Hidetoshi Shimauchi c, Masashi Komatsu a a
Graduate School of Dentistry, Department of Restorative Dentistry, Division of Operative Dentistry, Tohoku University, Sendai, Japan University of Cologne, Germany c Graduate School of Dentistry, Department of Restorative Dentistry, Division of Periodontology and Endodontology, Tohoku University, Sendai, Japan d Heraeus Kulzer GmbH, Research & Development, Wehrheim, Germany e Department of Clinical Cariology and Endodontology, Showa University, Tokyo, Japan b
article info
abstract
Article history:
Objective: To investigate and compare bonding and dentin sealing efficacy of a marketed all-
Received 20 November 2008
in-one and an experimental model adhesive with minimum effective amounts of acidic
Received in revised form
monomer and water.
27 January 2009
Materials and methods: Composition of model adhesive (NAD) in mass%: UDMA (45), 4-META
Accepted 28 January 2009
(20), H2O (7.5), and acetone (27.5). For characterization of a reasonable NAD application procedure shear bond strengths (SBS, n = 8) were determined on human enamel and dentin. Clearfil S3 Bond (TSB; Kuraray) served as reference. SBSs were evaluated after 10 min, 1 and 7
Keywords:
days, and 1 month, marginal adaptation (n = 8) was assessed in cylindrical butt-joint dentin
Self-etch adhesives
cavities. Diffusive and convective water fluxes through 1 mm thick adhesive-coated dentin
Enamel–dentin bonding
disks (n = 6) were qualitatively and quantitatively analyzed.
Bond strength
Results: SBSs proved that application of NAD in one coat with 20 s agitated dwell time was
Marginal adaptation
20 MPa, enamel SBSs (24 h) were 25 MPa, p > 0.05. Dentin SBSs for TSB and NAD were not
Phase separation
different ( p > 0.05) at the four stages (means: 18.9, 23.5, 25.4, and 23.6 MPa). Five and seven of
Permeability
the eight bonded restorations with TSB and NAD were gap-free ( p > 0.05). Dentin disks treated with EDTA from both sides or one side only were highly permeable for liquid, whereas adhesive-coated dentin disks showed no permeability at 0 and 2.5 kPa water pressure. Conclusions: Within the limitations of this study the model adhesive tested represents a promising basic composition for all-in-one adhesives, eliminating common problems encountered with single step adhesives such as phase separation and permeability. # 2009 Elsevier Ltd. All rights reserved.
1.
Introduction
All-in one or one-step self-etching adhesives are basically simple solutions, containing acidic methacrylate monomers,
either phosphate- or carboxylate-based, and water for ionization of the functional monomer to mediate dental hard tissue demineralization and simultaneous resin infiltration. Further, they contain hydrophobic dimethacrylate monomers
* Corresponding author at: Graduate School of Dentistry, Department of Restorative Dentistry, Division of Operative Dentistry, Tohoku University, 4-1 Seiryo-machi, Aoba-ku, Sendai 980-8575, Miyagi, Japan. E-mail address: [email protected] (W.J. Finger). 0300-5712/$ – see front matter # 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.jdent.2009.01.013
journal of dentistry 37 (2009) 432–439
to establish a cross-linked polymer structure. Co-solvents like ethanol or acetone are necessary to maintain the functional and cross-linking monomers in solution. Such organic solvents form azeotropic mixtures with water and enhance the surface dehydration when evaporated with compressed air. Finally, acid-compatible photoinitiators are added. Several marketed products contain additionally HEMA, an amphiphilic monomer that enhances component miscibility and resin penetration into the hard tissue.1 This complex mixture of diverse nature adhesive components is unequivocally the reason for a number of concerns expressed in current literature. Phase separation has been reported as a consequence of the immiscibility of water with hydrophobic monomers.1–5 Incomplete removal of water prior to adhesive polymerization may affect the curing performance of the adhesives. Several research reports confirm droplet and blister entrapment within the adhesive polymer layer increasing the convective and diffusive water flux from dentin, crossing the adhesive film and jeopardizing the adhesive strength.6–16 Further, the ester groups of the methacrylate cross-linking monomers, the amphiphilic and the acidic monomers of these adhesives are subjected to hydrolysis, leading to reduced shelf life.17–20 Water is essential for ionization of the functional monomers to render the adhesives acidic, however, water remaining upon polymerization leads to the described adverse effects. Water-free adhesives have consequently been proposed as alternatives.3 However, such adhesives would have to rely on the unpredictable amount of water available on or in the bonding substrate surface to gain the necessary demineralization capacity. Another gateway to overcome at least some of the waterrelated problems with all-in-one adhesives might be the reduction of the water concentration to a level where phase separation does not occur.21,22 With this approach it is however mandatory to keep a suitable balance between sufficient water for adequate ionization of the acidic monomer and a resin concentration high enough for good bonding efficacy to dentin. Therefore, in this study we investigated and compared the dentin-bonding efficacy of a simple HEMA-free model adhesive with minimum effective amounts of acidic monomer and water with a marketed all-in one adhesive. The efficacy parameters were bond strengths, marginal adaptation of bonded restorations in dentin cavities, and permeability of the adhesive films polymerized on dentin with or without hydrostatic pressure. The null hypothesis tested was that both adhesives are equally effectively bonding to and sealing dentin.
2.
Materials and methods
As follow-up23 a model adhesive (NAD) with the following components in mass percent was prepared: UDMA: urethanedimethacrylate (45), 4-META: 4-methacryloyloxyethyl trimellitic anhydride (20), water (7.5), acetone (27.5) and activated with camphorchinone (0.3) and coinitiator (0.35). The performance of this adhesive was compared with Clearfil S3 Bond (TSB.LOT: 00100B; Expiry: 2010-02; Kuraray Medical, Okayama, Japan). According to the manufacturer TSB includes bisphenol
433
glycidyl dimethacrylate (Bis-GMA), 2-hydroxyethyl methacrylate (2-HEMA), 10-methacryloyl-oxydecyl dihydrogen phosphate (10-MDP), silanated colloidal silica, water, and ethanol. In a preliminary trial, application details for NAD, namely the number of coating layers needed, the duration of the dwell time, and the effect of agitation are defined using shear bond strength to dentin as the target parameter.
2.1.
Phase separation
Three droplets of the experimental adhesive and of TSB, respectively, were dispensed in a white disposable mixing dish and left at ambient laboratory atmosphere under dimmed light for approximately 10 min. The adhesives were characterized free of phase separation when the solutions remained transparent and homogeneous during this time interval.
2.2.
Shear bond strength (SBS)
Human molars were immediately after extraction rinsed with water and immersed in 1% chloramine solution for a maximum of 6 months prior to experimental use. The teeth were placed with a sound proximal surface on the bottom of rubber molds and embedded with slow curing epoxy resin. Following over night curing peripheral enamel or dentin sites were exposed by wet grinding on SiC papers, grits 180–320, rinsed and air-dried prior to use. The adhesives were dispensed in a disposable dish and immediately applied with a microbrush either according to instructions for TSB, or with 1, 2 or 3 consecutive coats for 10, 20, 30 or 60 s. The effect of agitation of NAD on bond strength was tested with one layer applied and 20 s time. Following thorough air-drying for 10 s the adhesives were light-activated for 20 s with Translux Power Blue (Heraeus Kulzer, Hanau, Germany), clamped in the Ultradent bonding jig (Ultradent Products Inc., South Jordan, USA) under the proprietary 2 mm high and 2.3 mm wide cylindrical plastic mold for bulk-filling with Venus resin composite (Shade A2; Heraeus-Kulzer, Hanau, Germany), and 60 s light activation. From each of the two adhesives eight dentin bonded specimens were immersed into deionized water at 37 8C for 10 min, 24 h, 1 week, or 1 month prior to shear debonding (1 mm/min) with the Ultradent notched steel rod, mounted in a universal testing machine. For comparative enamel bond strengths eight specimens of each adhesive were tested after 24 h storage in 37 8C water. Data were analyzed by one-way ANOVA and Duncan’s post hoc test at a 95% confidence level ( p < 0.05). The modes of fracture were determined at 20-fold magnification and classified as adhesive, cohesive in resin or dentin, or mixed adhesive cohesive failures.
2.3.
Marginal gap formation
Proximal flat peripheral dentin surfaces of non-embedded molars were prepared on wet SiC paper, grit 600. Then, cylindrical cavities (4.5 mm in diameter and 1.5 mm deep) were prepared with a medium-grained diamond bur (approximately 10,000 rpm) into the coronal exposed dentin under copious water.
434
journal of dentistry 37 (2009) 432–439
The adhesives were applied to the rinsed and slightly airdried cavities in one coat for 20 s agitated dwell time, thoroughly air-dried for ten seconds, and light activated for 20 s. The cavities were then bulk filled with Venus resin composite, covered with a transparent matrix strip, light cured for 60 s, and immediately immersed into deionized water at room temperature for 15 min. Following removal of the excess resin composite on SiC paper, grits 600–4000, the restoration margins were inspected under a light microscope at 500-fold magnification. When a gap was detected the width was measured, using an ocular screw micrometer with a reading accuracy of 0.2 mm. For each adhesive eight specimens were prepared. The data were analyzed by Kruskal–Wallis ANOVA and Mann–Whitney’s post hoc test ( p < 0.05).
2.4.
Permeability of the adhesives
For this test human molars, frozen immediately after extraction, were used. The average age of the donors (7 male, 9 female) was 21 years, with a range from 17 through 28 years. All teeth were free of decay and restorations. Parallel, 1 mm thick coronal dentin slices were cut with a diamond wafer saw microtome (Model SP 1600, Leica Microsystems Nusssloch GmbH, Nussloch, Germany) under copious water-cooling, perpendicular to the vertical tooth axis between the occlusal enamel portion and the pulp horns. As a first reference, two dentin slices were cleaned from both sides with neutralized 0.5 M EDTA solution (pH 7.4) for 3 min each, to remove the cutting smear and to open the dentinal tubules, and rinsed with deionized water. As a second reference, another two tooth slices were cleaned with the EDTA solution for 3 min on the pulpal side only, leaving the cutting smear on the occlusal surface. For the final evaluation of adhesives permeability, six slices for each adhesive were treated with EDTA only on the pulpal side. The device used for determination of dentin permeability is fully described in a previous article.23 It is a split chamber column, modified from the model suggested by Pashley et al.24 (Fig. 1). Two cylindrical acrylic chambers are sealed with Orings on each side of a dentin slice and clamped in a metal frame. Each chamber has a liquid intake and a drainage hole. The chamber, fitting the occlusal side of the sandwiched dentin slice, is sealed with a clear glass cover slip. This chamber is filled with an aqueous solution of 0.02% luminol (5amino-2,3-dihydro-1,4-phthalazinedione) in 1% sodium hydroxide, whereas the opposite chamber is filled with the chemiluminescence activating liquid, an aqueous solution of 1% potassium ferricyanide and 0.3% hydrogen peroxide. When the activator containing side of the cell is pressurized the liquid may eventually pass through the dentinal tubules and produce a chemiluminescence reaction upon mixing with the illuminant reagent (Fig. 2). Luminol exhibits luminescence when activated with the oxidant hydrogen peroxide and a hydroxide salt in water as the activator. In the presence of potassium ferricyanide, the hydrogen peroxide is decomposed to form oxygen and water. When luminol reacts with the hydroxide salt, a dianion is formed. The oxygen produced from the hydrogen peroxide then reacts with the luminol dianion. The reaction product is a
Fig. 1 – Components of the split-chamber column used. The two chambers for the reagent solutions are clamped via Orings to the dentin slab, the detection cell is closed with a cover slip for registering of the photochemical signal, and the entire assembly is clamped between metal frames.
organic peroxide that is unstable and decomposes with the loss of nitrogen to produce 3-aminophthalic acid with electrons in an excited state. As the excited state relaxes to the ground state, the excess energy is liberated as photons, visible as blue light. This luminescence signal is recorded with a photodiode (S 6204; Hamamatsu Photonics, Hamamatsu City, Japan), installed 5 mm from the cover slip on the lower chamber. The entire equipment is set up in a lightproof box to prevent any outer light signal to interfere with the luminescence signal. The output voltage of the photodiode is recorded with an AD converter at 1 kHz and stored in the CPU unit controlling the system, from where the data at the end of the experimentation are transferred to a PC for further processing and analysis. The entire procedure is automated in a programmable sequencer. The non-covered dentin specimens were mounted between the split chambers that were filled with distilled water. Liquid pressure of 2.5 kPa was exerted for 120 s from the pulpal side, to flush the dentinal tubules free of extraneous liquid. Consecutively, the illuminant reagent was injected into the occlusal chamber, and the activating solution into the
Fig. 2 – Schematic presentation of the split-chamber device. The activator solution is pressurized, penetrates through the dentin disc, and a photochemical signal is generated upon contact with the luminol/hydroxide solution. The light signal is recorded with a photodetector.
journal of dentistry 37 (2009) 432–439
opposite one. The injection process lasted 45 s, followed by 30 s delay for stabilization of the atmospheric pressure on the luminal side, before the activating solution was pressurized with 2.5 kPa (25 cm H2O or 20 mmHg) for 2 min, and left without liquid pressure for another 2 min. The photochemical signal was recorded with the photodiode. Following this first run, a second wash cycle with water at 2.5 kPa was automatically initiated for 2 min before the chambers were filled with fresh reagent solutions, pressurized again as described above, and finally flushed with water. This entire procedure for determination of baseline permeability data was repeated twice on each dentin specimen. Six dentin specimens for each of the two adhesives were cleansed with EDTA on the pulpal side, rinsed and air-dried. The dentin adhesives, TSB or NAD were applied to the smearcovered occlusal surface, as described above and light-activated for 60 s under a continuous stream of nitrogen, to prevent oxygen from ambient air to interfere with polymerization by inhibition. The coated dentin slices were then mounted into the split-chamber device and investigated, using the same washing and reagent cycles as described for determination of the baseline permeability data. The permeability of each specimen was measured within 1 h after polymerization, after 1 day, 1 week, 1 months, and three months. During the time intervals between consecutive measurements the specimens were stored in deionized water. A special clamping device secured that the specimens were mounted reproducibly with the identical dentin/adhesive surface area exposed. For indirect observation of water crossing through the noncoated and coated dentin specimens an impression technique was used. Specimens, with a water-filled chamber connected to the pulpal side only, were clamped in a stand in the same position as during the photochemical test. Following water pressurizing at 2.5 kPa for 2 min, the free surface was dried with compressed air for 10 s, and immediately coated with freshly mixed hydrophobic condensation curing silicone impression material (Base: RTV 501, LOT 0002994233; Cat: Y, LOT 0002843600. Base-catalyst ratio by wt: 100/5. Dow Corning Toray Co., Ltd., Tokyo, Japan) delivered from a syringe. During impression coating and setting the 2.5 kPa water pressure was either maintained, to simulate the condition of convective flux, or the water pressure was released to zero, to simulate conditions of diffusive water flux. According to the manufacturer the working time of the impression material is 3 min, the setting time 7 min. The impression was left in place for 10 min at ambient temperature prior to removal. For the two reference conditions and for the adhesive covered dentin specimens, two samples each, subjected to the photochemical evaluation before, were randomly selected for this impression technique. The impressions were stored dry at ambient laboratory atmosphere for 24 h prior to mounting on aluminum holders and sputter coating with Pt for SEM inspection at 1000-fold magnification.
3.
Results
Macroscopically, both adhesives remained transparent during the 10 min storage time and were thus characterized as free of phase separation.
435
Fig. 3 – Shear bond strengths of the model adhesive on enamel (gray bars) and dentin (white bars) by agitated dwell time. The adhesives were applied with two consecutive coats. SBS testing after 24 h of water storage. One-way ANOVA revealed highly significant differences between the groups ( p < 0.001). Horizontal lines connect groups that are not significantly different ( p > 0.05).
Fig. 3 shows the box-and-whisker plots (median, interquartile distances, and extreme values) of the shear bond strengths for the model adhesive on enamel and dentin, applied in two layers and left for different agitated dwell times. At none of the dwell times significant differences were detected between enamel and dentin SBS. Ten seconds dwell time results in significantly lower SBS (mean 16.5 MPa) than longer application times. Between 20 and 60 s no SBS differences by dwell time are found (mean 21.7 MPa). The diagram in Fig. 4 illustrates the enamel and dentin SBSs of the model adhesive when applied in 1, 2 or 3 consecutive
Fig. 4 – Shear bond strengths of the model adhesive on enamel (gray bars) and dentin (white bars) by numbers of coats applied and 20 s agitated dwell time. SBS testing after 24 h of water storage. ANOVA revealed no significant differences between the groups ( p = 0.801).
436
journal of dentistry 37 (2009) 432–439
Fig. 5 – Shear bond strengths of the model adhesive on enamel and dentin when applied in one coat with (gray bars) or without (white bars) agitation during 20 s dwell time. SBS testing after 24 h of water storage. One-way ANOVA revealed significant differences between the groups ( p < 0.01). Same letters denote groups that are not statistically different ( p > 0.05).
Fig. 6 – Means and 95% intervals of confidence for shear bond strengths of TSB (open circles) and NAD (gray circles) tested after storage in water for 10 min through 1 month. One-way ANOVA proved significant differences ( p < 0.001). Identical small letters denote homogeneous groups according to Duncan’s multiple comparison testing ( p > 0.05).
layers and agitated during 20 s dwell time. There were no statistically significant differences between the groups ( p = 0.801). The mean bond strength was 21.4 MPa. Fig. 5 shows the effect of adhesive agitation during 20 s dwell time on SBSs to enamel and dentin. One coat of adhesive was applied. Gray bars are with, white bars without agitation. One-way ANOVA shows significant differences between the groups ( p = 0.007). Same lower-case letters denote groups that are not statistically different ( p > 0.05). Agitation gives significantly higher SBS on enamel, whereas agitation has no significantly different effect on dentin SBS. Bond strengths to enamel after 24 h storage in water were not significantly different ( p > 0.05), for TSB (25.5 2.5 MPa) and for NAD (22.4 3.7 MPa). The diagram in Fig. 6 illustrates the mean shear bond strengths to dentin of TSB (white circles) and NAD (gray circles) by storage time prior to debonding. The whiskers depict the 95% confidence intervals of the means. One-way ANOVA showed significant differences between the groups ( p < 0.001), same letters denote homogeneous data subsets ( p > 0.05). At each storage time the SBSs of both products were not significantly different ( p > 0.05). Table 1 summarizes the failure modes of the debonded specimens. No adhesive failures were found, most fractures were located in resin. Marginal adaptation of the bonded Venus restorations in the 4.5 mm wide dentin cavities is not significantly different for the two adhesives ( p = 0.328). With TSB five restorations are gap-free, the other three restorations show 0.6, 0.6, and 0.8 mm wide gaps. NAD resulted in seven gap-free restorations and 1 cavity margin with a 0.4 mm wide gap. Fig. 7 illustrates the characteristic chemiluminescence output in mV for specimens that were treated with EDTA on both cutting surfaces (upper trace). As a result of the diffusion process between the activator and reactant chamber during
injection of the reagents and prior to pressurizing the onset of the photo signal was larger than 100 mV for the non-coated dentin slices. Upon pressurizing a light signal proportional to the amount of trigger liquid reacting with the luminal containing reactant is generated, returning almost to the initial output after 120 s when the pressure is released to zero. The middle curve in the diagram reflects the permeability of dentin disks with cutting smear on the occlusal surface and without smear on the reverse side. The maximum output signal at the end of the 120 s pressurizing time is approximately 50% from specimens that are free of smear on both sides. At the end of the non-pressurized interval an appreciable photo signal is still recorded indicating a continuous reaction between the diffusing activator solution and the reactant. The signal tracing generated during the active pressure and the pressure-released intervals of adhesivecoated dentin disks is a straight line, parallel to the time axis of the diagram, at approximately 40 mV output. This small photo signal reflects the noise level of the instrument, and is hardly
Table 1 – Debonding failures of shear bond strength specimens (n = 8). Clearfil S3 Bond (TSB)
1h 24 h 1 week 1 month
Experimental adhesive (NAD)
CR
CD
RD
AD
CR
CD
RD
7 2 4 6
0 1 1 1
1 5 3 1
0 0 0 0
8 5 7 6
0 1 1 2
0 2 0 0
AD 0 0 0 0
CR: cohesive in resin; CT: cohesive in dentin; RD: cohesive in resin and dentin; AD: adhesive.
journal of dentistry 37 (2009) 432–439
437
SEMs of impressions made from smear-free and smearcovered dentin show numerous droplets (concave depressions) at the interface, both for the pressurized specimens and for specimens with passive water contact at the pulpal side (Fig. 8). Impressions made from smear-free pressurized samples displayed large confluent droplets. In contrast, no droplets were seen on dentin samples coated with TSB or NAD (Fig. 9), irrespective of the applied water pressure and time during which the specimens were stored in water. The rippled surface of the adhesive was produced under the stream of protective nitrogen directed to the surface.
4. Fig. 7 – Characteristic examples of the photochemical output in mV registered for dentin specimens, treated with EDTA from both surfaces (upper curve), cleaned with EDTA on the pulpal side only (middle curve), and coated with adhesive (lower trace, parallel to the time axis).
related to a chemiluminescence reaction. Throughout the course of the experiment with testing stages from 1 h to 3 months virtually the same linear tracings were found, when the permeability of the 6 specimens each, coated either with TSB or NAD was investigated, indicating that there was no activator liquid crossing the adhesive layer.
Discussion
Several commercially available all-in-one adhesives exhibit phase separation upon volatilization of organic solvent and simultaneous occurrence of water-rich blisters due to the immiscibility of hydrophobic and hydrophilic monomers.2–4 One might argue that such water-rich droplets are clinically not critical, since they are readily eliminated during preactivation air-drying. Nevertheless, even elimination of such aqueous blisters prior to polymerization is not the solution of the problem since reportedly the remaining homogeneous monomer-rich phase contains approximately 10% of water that is more difficult to remove than the water-rich phase.4 Other marketed adhesives contain HEMA, obviously added by their manufacturers as a good co-solvent to prevent phase
Fig. 8 – SEM micrographs: (A) specimen covered with smear layer, as produced during diamond wafer cutting. (B) Specimen treated with EDTA. (C) Vinyl polysiloxane impression of smear layer covered dentin slice under passive pulpal side water contact (0 kPa) and (D) at 2.5 kPa water pressure. The concave globular depressions in (C) and (D) represent water droplets that exuded through the dentin tubules and smear layer throughout the 3 min working time of the impression material.
438
journal of dentistry 37 (2009) 432–439
Fig. 9 – Representative SEM micrograph of impression from a dentin slice coated with adhesive. No differences were observed between the two adhesives, the times of water storage or the liquid pressure. No concave depressions were found, indicating no permeability of the adhesivecoated specimens during the 3 min working time of the silicone impression. The wavelike adhesive surface results from the stream of nitrogen blown to the surface before and during light-activation.
separation. One of the disadvantages with HEMA containing adhesive compounds is difficult removal of solvents due to HEMAs polar characteristics, i.e. hydrogen bonding capability. With such adhesives droplets are found in the polymerized adhesive film on dentin as an expression of osmotic effects.2,7,9,13,16 Therefore, it was considered reasonable to design all-inone adhesives with minimal effective contents of water and functional acidic monomer. The present data indicate that a simple light-curing adhesive with 7.5% of water, 20% 4-META, and 45% UDMA dissolved in acetone19 might be a useful approach to overcome some of the problems mentioned. The model adhesive shows no phase separation, and one coat applied for 20 s agitated dwell time is clinically practicable. Its final bond strength to dentin and enamel and the marginal adaptation in cylindrical butt-joint dentin cavities are comparable with the commercial Clearfil S3 Bond. Therefore, the null hypothesis is accepted that both adhesives bond equally effective to dentin and enamel. The smear-layer-free specimens and the samples with intact smear-layer on the occlusal side were clearly permeable for water. In contrast, the permeability tests did not reveal liquid crossing through the adhesive-coated dentin slices for the experimental or for the HEMA-rich commercial adhesive TSB. This observation may be related to the specimen production procedure, where light activation following thorough and strong air-drying was carried out under a stream of protective nitrogen directed to the surface. Further, the adhesives were applied on air-dried dentin and cured immediately after the 20 s dwell time without interference of water. Van Landuyt et al.2,13 pointed out that osmosis occurs when the light-cured adhesive is still permeable enough to
transmit small molecules and when the osmotic pressure is higher at the free, oxygen-inhibited side than at the dentin side of the adhesive. In their study the m-tensile bond strength of HEMA-rich adhesive on dentin was significantly lower when cured after 20 min delay than when light-cured immediately. This confirms that oxygen inhibition and thus the high osmotic pressure is stopped when the free surface is protected from oxygen diffusion either by polymerizing a lining composite on top or by polymerization under an inert atmosphere. Addition of small amounts of HEMA to conventional water-rich adhesives does not prevent phase separation,4,25 whereas large amounts of HEMA render the adhesive polymer more hydrophilic and prone to osmosis, and the resulting polymer less cross-linked and therefore mechanically weaker.12,14,25,26 No permeability of the adhesive-coated dentin slices was recorded even after 3 months storage in water. It may be speculated whether the relatively low water pressure (2.5 kPa) exerted and/or the short pressurizing time used (2 min) were sufficient to make liquid cross from the pulpal specimen side through the adhesive. On the other hand, osmotic blistering in the adhesive film occurs reportedly fast and has been demonstrated already after short-term storage of specimens in water.25 At this point the validity of this new chemiluminescence method to assess adhesive permeability is still not fully clarified, and further research is necessary to evaluate whether, e.g. curing under nitrogen polymerizes the adhesive to a degree that water crossing is prevented without giving evidence for what might happen at the resin–dentin interface. In conclusion, the similarity in performance of Clearfil S3 Bond and the outlined model adhesive indicates that the simple composition of the experimental solution may be a promising base for further development of all-in-one self-etch adhesives. The model adhesive shows no phase separation, bonding strength to dentin and enamel is reasonably high, marginal adaptation of bonded resin composite restorations in cylindrical dentin cavities is very satisfactory, and within the limitations of the permeability test used there are no signs of water crossing immediately after light activation and throughout the 3-month follow-up tests. Studies currently under progress are designed to elucidate possible effects of adjusting of the activating system, combination of the carboxylic functional monomer with phosphate ester monomers, and addition of filler fractions on the physico-mechanical resistance and long-term efficacy of such model adhesives.
Acknowledgment Dr. Masae Furukawa from Showa University in Tokyo is gratefully acknowledged for her laboratory assistance in performing part of the bond strength tests.
references
1. Van Meerbeek B, Van Landuyt K, De Munck J, Hashimoto M, Peumans M, Lambrechts P, et al. Technique sensitivity of contemporary adhesives. Dental Materials Journal 2005;24: 1–13.
journal of dentistry 37 (2009) 432–439
2. Van Landuyt K, De Munck J, Snauwaert J, Coutinho E, Poitevin A, Yoshida K, et al. Monomer-solvent phase separation in one-step self-etch adhesives. Journal of Dental Research 2005;84:183–8. 3. Van Landuyt KL, Mine A, De Munck J, Coutinho E, Peumans M, Jaecques S, et al. Technique sensitivity of water-free onestep adhesives. Dental Materials 2008;24:1258–67. 4. Finger WJ, Shao B, Hoffmann M, Kanehira M, Endo T, Komatsu M. Does application of phase-separated selfetching adhesives affect bond strength? Journal of Adhesive Dentistry 2007;9:169–73. 5. Gaintantzopoulou M, Rahiotis C, Eliades G. Molecular characterization of one-step self-etching adhesives placed on dentin and inert surfaces. Journal of Adhesive Dentistry 2008;10:83–93. 6. Tay FR, Pashley DH, Yoshiyama M. Two modes of nanoleakage expression in single-step adhesives. Journal of Dental Research 2002;81:472–6. 7. Tay FR, King NM, Chan KM, Pashley DH. How can nanoleakage occur in self-etching adhesive systems that demineralize and infiltrate simultaneously? Journal of Adhesive Dentistry 2002;4:255–69. 8. Tay FR, Pashley DH. Water-treeing: a potential mechanism for degradation of dentin adhesives. American Journal of Dentistry 2003;16:6–12. 9. Hashimoto M, Tay FR, Ito S, Sano H, Kaga M, Pashley DH. Permeability of adhesive resin films. Journal of Biomedical Materials Research 2005;74:699–705. 10. Malacarne J, Carvalho RM, de Goes MF, Svizero N, Pashley DH, Tay FR, et al. Water sorption/solubility of dental adhesive resins. Dental Materials 2006;22:973–80. 11. Hashimoto M, Tay FR, Sano H, Kaga M, Pashley DH. Diffusion-induced water movement within resin–dentin bonds during bonding. Journal of Biomedical Materials Research 2006;79:453–8. 12. Hashimoto M, Fujita S, Kaga M, Yawaka Y. Effect of water on bonding of one-bottle self-etching adhesives. Dental Materials Journal 2008;27:172–8. 13. Van Landuyt KL, Snauwaert J, De Munck J, Coutinho E, Poitevin A, Yoshida Y, et al. Origin of interfacial droplets with one-step adhesives. Journal of Dental Research 2007;86:739–44. 14. Sauro S, Pashley DH, Montanari M, Chersoni S, Carvalho RM, Toledano M, et al. Effect of simulated pulpal pressure on
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
439
dentin permeability and adhesion of self-etch adhesives. Dental Materials 2007;23:705–13. Donmez N, Belli S, Pashley DH, Tay FR. Ultrastructural correlates of in vivo/in vitro bond degradation in self-etch adhesives. Journal of Dental Research 2005;84:355–9. Tay FR, Lai CNS, Chersoni S, Pashley DH, Mak YF, Suppa P, et al. Osmotic blistering in enamel bonded with one-step self-etch adhesives. Journal of Dental Research 2004;83: 290–5. Inoue S, Koshiro K, Yoshida Y, De Munck J, Nagakane K, Suzuki K, et al. Hydrolytic stability of self-etch adhesives bonded to dentin. Journal of Dental Research 2005;84: 1160–4. Moszner N, Salz U, Zimmermann J. Chemical aspects of selfetching enamel–dentin adhesives: a systematic review. Dental Materials 2005;21:895–910. Fujita K, Ma S, Li R, Ikemi T, Nishiyama N. Effect of solvent type on the degradation of 4-MET. Dental Materials Journal 2007;26:792–9. Da Silveira Lima G, Ogliari FA, Oliveira da Silva E, Ely C, Demarco FF, Carren˜o NLV, et al. Influence of water concentration in an experimental self-etching primer on the bond strength to dentin. Journal of Adhesive Dentistry 2008;10:167–72. Hiraishi N, Nishiyama N, Ikemura K, Yau JYY, King NM, Tagami J, et al. Water concentration in self-etching primers affect their aggressiveness and bonding efficacy to dentin. Journal of Dental Research 2005;84:653–8. Furukawa M, Shigetani Y, Finger WJ, Hoffmann M, Kanehira M, Endo T, et al. All-in-one self-etch model adhesives: HEMA-free and without phase separation. Journal of Dentistry 2008;36:402–8. Ishihata H, Kanehira M, Nagai T, Finger WJ, Shimauchi H, Komatsu M. Effect of desensitizing agents on permeability of dentin. American Journal of Dentistry 2009, in press. Pashley DH, Stewart FP, Galloway SE. Effects of air-drying in vitro on human dentine permeability. Archives of Oral Biology 1984;29:379–83. Van Landuyt KL, Snauweart J, De Munck J, Peumans M, Lambrechts P, Van Meerbeek B. The role of HEMA in onestep self-etch adhesives. Dental Materials 2008;24:1412–9. Asmussen E, Peutzfeldt A. Influence of selected components on crosslink density in polymer structures. European Journal of Oral Sciences 2001;109:282–5.