The Journal of Adhesive Dentistry, 05/2013

Dear Readers, When I was a student, I was taught the basic rules about caries "therapy": extension for prevention and all the other rules for cavity preparations. Furthermore, we were instructed to remove ALL the decayed tissue, which meant that we excavated the diseased tissues until the dentin color became brighter (yellowish) and the surface became shiny and harder, hoping that our excavation ended in the zone of remineralization, which can be identified in histological preparations of carious teeth. The only exception to this rule was the pulpal or axial wall, where we were allowed to leave carious tissue behind. The rationale for this was to prevent pulp exposure. In these cases, we treated the dentin with calcium hydroxide. Years later I moved to Berlin, where I met a more radical faculty. We agreed to not tolerate any carious dentin tissue under restorations and therefore excavated to the bitter end, which meant that pulps were exposed and we ended up performing a substantial amount of direct pulp capping in the student courses. When we evaluated our students' work, we were penalized for our strategy with a high failure rate of the direct pulp cappings, which we associated with not strictly following the rule of placing the permanent restoration in the same session under rubber-dam. These facts forced us to adopt the philosophy described at the beginning of this editorial. Since then, much more than 10 years have passed, and now working and living in Gainesville, Florida, I met Dr. Saulo Geraldeli and a group like myself who are strong proponents of minimally invasive dentistry. We became involved in extensive and engaged discussions about how diseased, carious tissue should be assessed, quantified, and removed, and started to realize that the literature contains many studies which indicate that leaving affected dentin prior to placement of restorations, along with a reasonable preventive strategy to manage the caries process, can actually be done. However, all this can only work if the dentist is able to effectively, completely seal off the lesion in order to stop all nutrient supply to the few bacteria left behind. Moreover, if chlorhexidine is applied to the affected dentin, one can expect that this will help to further eliminate bacteria. This makes it clear how important good adhesive materials and protocols are in such an approach, with the potential to save a lot of dental hard tissue. The evidence for such an approach is not yet very strong, since there are no prospective randomized clinical trials available, but all indicators point in the same direction. Based on adhesion, we may soon face another paradigm shift in dentistry!

Purpose: To evaluate the potential for phosphoric acid solutions - common constituents of dental adhesive systems - of varying pH to solubilize dentin matrix components (DMCs) from human dentin. Materials and Methods: Human dentin chips were ground under liquid nitrogen to a powder (ca 100 µm) and incubated at 4°C with agitation in phosphoric acid of pH 1, 2, 3, 4, 5, and 6 (1 g/4 ml; n = 4) for six days with solution changes each day. Estimates of daily protein release were made by UV spectrophotometry at 280 nm. Extract solutions were dialyzed for 7 days in reverse osmosis water, lyophilized, and weighed. Non-collagenous proteins (NCPs) and glycosaminoglycans (GAGs) were quantitated by dye-binding assays. 1D-PAGE for preliminary protein characterization and sandwich ELISA for presence of TGF-ß1 were performed. The results were analyzed by ANOVA and regression (α = 0.05). Results: Protein release was drastically reduced after the first few days, with the highest amounts obtained from pH 1. There was no significant difference in the quantity of DMCs solubilized by the different pH levels, but there was a significant logarithmic relation between release and pH, suggesting that greater DMC solubilization occurs with higher hydrogen ion concentrations. Dye binding assays confirmed the release of NCPs and GAGs at all pH levels. There were only subtle differences in protein bands observed between the different pH levels (1D-PAGE). Significant levels of TGF-ß1 were identified from extraction at all pHs. Conclusion: Acids at pH levels relevant to those used in commercial dentin adhesives are capable of solubilizing human DMCs, with release being related to hydrogen ion concentration. Keywords: dental material, phosphoric acid, dentin matrix components, dentin, solubilization, protein, glycosaminoglycan

Purpose: Differently prepared hydroxyapatite (HAp) nanoparticles were incorporated into the adhesive solution of a commercial adhesive system in order to evaluate the effect on microtensile bond strength to dentin. Materials and Methods: HAp nanoparticles (20 to 70 nm) were prepared by different processes (biomimetic and hydrothermal) and incorporated into the adhesive of the Adper Scotchbond Multi-Purpose (SBMP) system at various concentrations. Control (unfilled) and experimental groups (filled) were applied onto flat mid-coronal human dentin. Composite crowns were built up and cut into beams with a cross-sectional area of 0.65 ± 0.05 mm2. Specimens were fractured in tension and examined with a scanning electron microscope (SEM) for fractographic analysis. Microtensile bond strength (µTBS) data were analyzed using a two-way ANOVA and modified LSD test at a = 0.05. Analysis of the nanofiller distribution and ultramorphological characterization of the interface was performed by transmission electron microscopy (TEM). Results: HAp nanoparticle incorporation into the adhesive of SBMP significantly influenced µTBS to dentin depending on the fillers and the concentration used. A significant increase of the mechanical strength was obtained for the adhesives containing 1% (wt/vol) biomimetic and 5% hydrothermal silanized HAp particles, while the other particle fractions did not influence µTBS significantly. 10% (wt/vol) HAp particles significantly lowered the µTBS irrespective of the particle type used. TEM micrographs revealed nanoparticle dispersion through the adhesive layer but no deposition on or penetration into the hybrid layer. Conclusions: HAp nanoparticle incorporation into SBMP increased bond strength to dentin by cohesively reinforcing the interface adhesive layer. At a concentration of 10% (wt/vol), nanofiller incorporation had a negative effect on bond strength. Keywords: HAp, biomimetic and hydrothermal HAp preparation, silanization, adhesion, nanoparticle filler, dentin bonding, tensile bond strength

Purpose: To measure dimensional changes due to hygroscopic expansion and their effect on interface gaps and sealing in four light-cured restorative materials using an original confocal microscopic methodology. Materials and Methods: The materials tested were an ormocer (Admira [Voco]), a compomer (Dyract AP [Dentsply]), a hybrid composite (Spectrum [Dentsply]), and a nanohybrid composite (Esthet·X [Dentsply]). Water sorption was evaluated by weighing material disks after immersion. Hygroscopic expansion was measured from volumetric variations of material fillings in cylindrical cavities in dentin slices; the interfacial gap size was obtained from the same cavities using a novel confocal microscopic method. Microleakage was evaluated in cavities prepared in extracted third molars. Measurements followed water immersion for 24 h, 1 week, 4 weeks, and 8 weeks. A factorial ANOVA, the Student Newman Keuls test for post-hoc comparisons, the Student's t-test, and the Pearson test were used for the statistical analysis (p 0.05). Results: Positive correlations were found among water sorption, hygroscopic expansion, and sealing. Hygroscopic expansion reduced post-polymerization interfacial gaps and improved cavity sealing. Dyract AP and Admira showed the highest water sorption, hygroscopic expansion, and gap size reduction. Conclusions: 1. The proposed methodology is valid to measure hygroscopic expansion and interfacial gap. 2. Water sorption and hygroscopic expansion are positively correlated, and hygroscopic expansion, gap size, and sealing are also positively correlated. 3. The adhesive influences the interfacial gap size and its variation after hygroscopic expansion. 4. Hygroscopic expansion reduces the interfacial gaps generated by polymerization shrinkage and improves cavity sealing. Keywords: restorative material, adhesive, hygroscopic expansion, interfacial gap, sealing, confocal microscopy

Purpose: To evaluate the activity of a methacryloyloxydodecylpyridinium bromide (MDPB)-containing self-etching primer (Clearfil Protect Bond) against Streptococcus mutans and its ability to reduce biofilm formation on standardized experimental Class I restorations in vitro. Materials and Methods: Forty experimental Class I round restorations were prepared on enamel-dentin slabs using different adhesive strategies: group 1 = MDPB-containing adhesive system (Clearfil Protect Bond); group 2 = MDPB-free self-etching adhesive system (Clearfil SE Bond); group 3: MDPB-containing self-etching primer in combination with a fluoride-free bonding agent; group 4: MDPB-free self-etching primer in combination with a fluoride-containing bonding agent; group 5: a three-step etch-and-rinse adhesive system (Adper Scotchbond Multi Purpose). A Streptococcus mutans biofilm was grown for 48 h on the restoration surfaces and subsequently evaluated using scanning electron microscopy on three different areas: enamel, composite, and interface surfaces. Statistical analysis was performed by multiple ANOVA after data transformation. Results: Specimens in groups 2, 4 and 5 showed greater biofilm formation than those in groups 1 and 3 (p 0.001) on all investigated substrates (enamel, composite, and interface areas). Conclusions: Specimens prepared with an MDPB-containing primer exhibited significant decreases in biofilm formation on Class I restorations in vitro. Further in vitro and in vivo studies are required to clarify the role of quaternary ammonium compounds in reducing bacterial biofilm formation on restoration surfaces. Keywords: dentin bonding agents, MDPB monomer, antibacterial agents, biofilms, Streptococcus mutans

Purpose: To evaluate the bonding of simplified adhesive systems to sound and caries-affected dentin of primary teeth with microtensile (µTBS) and nanoleakage (NL) tests. Materials and Methods: Occlusal cavities were prepared in 36 sound second primary molars. Half of the specimens were submitted to pH cycling to simulate caries-affected dentin. Teeth were randomly restored with one of three materials: the etch-and-rinse adhesive system Adper Single Bond 2 (SB), the two-step self-etching adhesive system Adper SE Plus (SE), and the one-step self-etching adhesive system Adper Easy One (EASY). After storage for 24 h, specimens with cross-sectional areas of 0.8 mm2 were prepared for microtensile testing (1 mm/min). One stick from each tooth was immersed in silver nitrate solution (24 h) and allowed to develop for 8 h in order to score the nanoleakage with SEM. The fracture pattern was evaluated using a stereomicroscope (400X). The µTBS means were analyzed by two-way ANOVA and Tukey's post-hoc test. For NL, the Kruskal- Wallis and Mann-Whitney tests were used (α 0.05). Results: SB (35.5 ± 3.5) showed the highest µTBS value to sound dentin, followed by EASY (26.3 ± 1.9) and SE (18.2 ± 6.5) (p 0.05). No difference among materials was observed for caries-affected dentin (SB: 17.8 ± 4.2; SE: 13.9 ± 3.2; EASY: 14.4 ± 4.2, p > 0.05). For all groups, adhesive/mixed fracture prevailed. Cariesaffected dentin promoted silver nitrate uptake into the adhesive interface; however, with SE, the nanoleakage was more pronounced than in the other adhesive systems, even in sound dentin. Conclusion: Caries-affected dentin negatively influences the bond strength and nanoleakage of the two-step etch-and-rinse and one-step self-etching adhesive systems tested in primary teeth. Keywords: microtensile, bond strength, caries-affected dentin, nanoleakage, primary teeth

Purpose: To evaluate the tensile bond strength at repaired interfaces of aged dental composites, either dimethacrylate- or silorane-based, when subjected to different surface treatments. Materials and Methods: The composites used were Filtek P60 (methacrylate-based, 3M ESPE) and Filtek P90 (silorane-based, 3M ESPE), of which 50 slabs were stored for 6 months at 37°C. The surface of adhesion was abraded with a 600-grit silicone paper and the slabs repaired with the respective composite, according to the following surface treatment protocols: G1: no treatment; G2: adhesive application; G3: silane + adhesive; G4: sandblasting (Al2O3) + adhesive; G5: sandblasting (Al2O3) + silane + adhesive. After 24-h storage in distilled water at 37°C, tensile bond strength (TBS) was determined in a universal testing machine (Instron 4411) at a crosshead speed of 0.5 mm/min. The original data were submitted to two-way ANOVA and Tukey's test (α = 5%). Results: The methacrylate-based composite presented a statistically significantly higher repair potential than did the silorane-based resin (p = 0.0002). Of the surface treatments for the silorane-based composite, aluminum-oxide air abrasion and adhesive (18.5 ± 3.3MPa) provided higher bond strength than only adhesive application or the control group without surface treatment. For Filtek P60, the control without treatment presented lower repair strength than all other groups with surface treatments, which were statistically similar to each other. The interaction between the factors resin composite and surface treatment was significant (p = 0.002). Conclusion: For aged silorane-based materials, repairs were considered successful after sandblasting (Al2O3) and adhesive application. For methacrylate resin, repair was successful with all surface treatments tested. Keywords: resin composites, silorane, adhesive dentistry, bond strength

Purpose: To investigate the chemical polymerization kinetics of commercial dual-curing adhesive systems when used solely or in conjunction with chemically-curing resin cement. Materials and Methods: Four adhesive systems comprising simplified-step adhesives and activators (Prime&Bond NT with Self Cure Activator, Excite DSC, AQ Bond Plus, All-Bond SE) were used. The pH values of the adhesives and adhesive/activator blends were measured. Differential scanning calorimetry (DSC) was used to investigate the extent of the chemical polymerization of the adhesives when used alone or directly intermixed with a chemically-cured resin cement (C&B Cement) for 60 min (n = 5). The data derived from the DSC analysis were statistically compared using one-way ANOVA and the Games-Howell post-hoc test (α = 0.05). Results: All the adhesives were highly acidic; when they were blended with the respective activators, their pH values increased. Neither the adhesive/activator blends nor the adhesive alone/cement mixtures showed any detectable heat generation. The Prime&Bond NT/activator showed delayed heat generation only when intermixed with the catalyst/base paste. The other three adhesive systems produced similar exotherms when intermixed with the catalyst paste alone or with the catalyst/base paste (p > 0.05), but at significantly different maximum rates of polymerization (p 0.05). Significantly shorter induction periods resulted when AQ Bond Plus and All- Bond SE were intermixed with the catalyst/base paste rather than with the catalyst paste alone (p = 0.004). Conclusion: The chemical polymerization occurring at the adhesive system/resin cement interface appears highly dependent on the adhesive system used and may be considerably delayed. Keywords: activator, adhesive, polymerization, resin cement

Purpose: To evaluate the microtensile bond strength (MTBS) of ceramic cemented to dentin varying the resin cement and ceramic shades. Materials and Methods: Two VITA VM7 ceramic shades (Base Dentine 0M1 and Base Dentine 5M3) were used. A spectrophotometer was used to determine the percentage translucency of ceramic (thickness: 2.5 mm). For the MTBS test, 80 molar dentin surfaces were etched and an adhesive was applied. Forty blocks (7.2 x 7.2 x 2.5 mm) of each ceramic shade were produced and the ceramic surface was etched (10% hydrofluoric acid) for 60 s, followed by the application of silane and resin cement (A3 yellow and transparent). The blocks were cemented to dentin using either A3 or transparent cement. Specimens were photoactivated for 20 s or 40 s, stored in distilled water (37°C/24 h), and sectioned. Eight experimental groups were obtained (n = 10). Specimens were tested for MTSB using a universal testing machine. Data were statistically analyzed using ANOVA and Tukey's post-hoc tests (α = 0.05). Results: The percentage translucency of 0M1 and 5M3 ceramics were 10.06 (± 0.25)% and 1.34 (± 0.02)%, respectively. The lowest MTBS was observed for the ceramic shade 5M3. For the 0M1 ceramic, the A3 yellow cement that was photocured for 20 s exhibited the lowest MTBS, while the transparent cement that was photocured for 40 s presented the highest MTBS. Conclusions: For the 2.5-mm-thick 5M3 ceramic restorations, the MTBS of ceramic cemented to dentin significantly increased. The dual-curing cement Variolink II photocured for 40 s is not recommended for cementing the Base Dentine 5M3 feldspathic ceramic to dentin. Keywords: cement shade, feldspathic ceramic, dentin, translucency, microtensile bond strength

Purpose: To determine the bond strength between zirconia and porcelain with varying numbers of veneer firing cycles. Materials and Methods: Fifty specimens of zirconia veneered with feldspathic ceramic were submitted to one (1-firing), two (2-firings), three (3-firings), four (4-firings), or five (5-firings) firing cycles to sinter the porcelain. After the respective number of firings, the specimens were embedded into acrylic resin and sectioned into bars with a 1-mm2 cross-sectional area. The microbars were bonded to a special device and attached to a universal testing machine (Emic DL 1000). Microtensile bond strength testing (MTBS) was performed at 0.5 mm/min. The maximum load for fracture was recorded (N) and the microtensile bond strength was calculated in MPa. Data were analyzed using one-way ANOVA and Tukey's test (α = 0.05). The Weibull modulus and characteristic strength was also calculated for each experimental group. Results: Specimens submitted to a single firing cycle presented the lowest bond strength values (14.1 MPa), two firing cycles provided intermediate bond strength values (15 MPa) and the other groups presented equivalently high values (18.1 - 18.4 MPa). The Weibull modulus did not change between the groups. Conclusion: More than three firing cycles of a veneer ceramic provided higher bond strengths between zirconia and the veneering ceramic. Keywords: all-ceramic, zirconia, layering veneer, microtensile bond strength

Purpose: To evaluate the effect of post-silanization heat treatment of a silane agent and rinsing with hot water of silanized CAD/CAM feldspathic ceramic surfaces on the microtensile bond strength between resin cement and the ceramic, before and after mechanical cycling. Materials and Methods: Blocks measuring 10 x 5.7 x 3.25 mm3 were produced from feldspathic ceramic cubes (VITA Mark II, VITA Zanhfabrik). Each ceramic block was duplicated in composite resin using a template made of polyvinylsiloxane impression material. Afterwards, ceramic and corresponding resin composite blocks were ultrasonically cleaned and randomly divided according to the 5 strategies used for conditioning the ceramic surface (n = 10): GHF: etching with hydrofluoric acid 10% + rinsing with water at room temperature + silanization at 20°C; G20: silanization; G77: silanization + oven drying at 77°C; G20r: silanization + hot water rinsing; G77r: silanization + oven drying at 77°C + hot water rinsing. The resin and ceramic blocks were cemented using a dualcuring resin cement. Every group was divided in two subgroups: aging condition (mechanical cycling, designated as a) or non-aging (designated as n). All the bonded assemblies were sectioned into microsticks for microtensile bond strength (µTBS) testing. The failure mode of the tested specimens was assessed and µTBS data were statistically analyzed in two ways: first 2-way ANOVA (GHF, G20 and G77 in non-aging/aging conditions) and 3-way ANOVA (temperature x rinsing x aging factors, excluding GHF), followed by Tukey's test (p = 0.05). Results: The 2-way ANOVA revealed that the µTBS was significantly affected by the surface treatment (p 0.001) but not by aging (p = 0.68), and Tukey's test showed that G77-n/G77-a (18.0 MPa) > GHF-n/GHF-a (12.2 MPa) > G20-n/G20-a (9.1 MPa). The 3-way ANOVA revealed that the µTBS was significantly affected by the "heat treatment" and "rinsing" factors (p 0.001), but not affected by "aging" (p = 0.36). The rinsing procedure decreased, while oven drying increased the bond strengths. Group G77, in both non-aging and aging conditions (18.6-17.4 MPa), had the highest bond values. Failure modes were mainly mixed for all groups. Conclusion: Oven drying at 77°C improved the bond strength between the resin cement and feldspathic ceramic, but hot water rinsing reduced the bond strength and should not be recommended. Keywords: adhesion, microtensile bond strength, ceramic, heat treatment, silane

Purpose: To compare the percentage of collagen exposed in dentin root thirds after two irrigation protocols with manual or rotary instrumentation using two methacrylate resin-based sealers. Materials and Methods: Forty-eight single-root human teeth were prepared with manual (n = 24) or nickeltitanium ProFile rotary (n = 24) instrumentation, using 5% NaOCl between instruments and 5 ml 17% EDTA as final irrigant or 20% citric acid + 2% chlorhexidine (CHX) between instruments and as the final irrigant. RealSeal or EndoREZ were used as filling materials. One 1-mm slice per third was abraded and stained with Masson's trichrome method. Mean exposed collagen values were obtained in four areas from each section (at 60X magnification) and a complete factorial ANOVA was used to analyze the influence of the study variables. Non-parametric Mann-Whitney's test was used to compare groups. Differences with p 0.05 were considered significant. Results: A significantly higher percentage of collagen was exposed in all thirds with the use of the 20% citric acid + 2% CHX protocol with rotary vs manual instrumentation, but percent collagen exposed did not differ as a function of the filling material. After the 5% NaOCl + 17% EDTA protocol, the percentage of collagen exposed did not differ between rotary and manual instrumentation but was higher with the use of RealSeal. Conclusion: The highest percentage exposure of collagen was with 20% citric acid + 2% CHX using rotary instrumentation, regardless of the filling material. Keywords: citric acid, chlorhexidine, collagen exposure, EDTA, Masson's trichrome, methacrylate resin-based sealers