Comparative Evaluation of Shear Bond Strength of Composite to Dentin in Presence of Calcium Releasing Liners: An In-Vitro Study

Saba Tohidkhah School of Dentistry, Tehran University of Medical Sciences Elham Ahmadi Department of Restorative Dentistry, Tehran University of Medical Science Mahdi Abbasi Department of Restorative Dentistry, Tehran University of Medical Science Reza Morvaridi Farimani Department of Orthodontics, School of Dentistry, Shahid Beheshti University of Medical Sciences, Ladan Ranjbar Omrani (  ladanomrani@yahoo.com ) Department of Restorative Dentistry, Tehran University of Medical Science


Introduction
Protective dental liners are applied in deep cavities to protect the pulp from different stimuli and facilitate the formation of reparative dentine [1]. These materials can seal dentinal tubules and protect the pulp from microorganisms' attacks and also have therapeutic effects and irritants as well as thermomechanical stimuli [2].Various materials have been used as cavity liners, including Resin Modified Glass Ionomer Cement, calcium silicate-based liners, calcium hydroxide liners, and bioactive glass-based cement [1,3,4] Traditionally, calcium hydroxide has been considered the gold standard of cavity lining materials for several years, in the case of direct and indirect pulp capping treatment procedure due to excellent antibacterial properties, alkaline pH, and its bioactivity in terms of formation of hard tissue barrier [5]. However, the unfavorable effects of calcium hydroxide, such as the weak physical properties, tunnel defects, high solubility, and gradual dissolution, led to a decline in its use as a liner with time [6], to overcome these undesirable drawbacks, several other materials have been introduced, light-activated calcium hydroxide lining materials were one of them which provides improved physical properties, decreased solubility in acids and water, although their bond to the substrate is weak and can shrink during polymerization. [ [3,7]]. Lime-Lite is a new resin-based light-activated calcium hydroxide-based material that could be applied as a liner and base [8], according to the manufacturer: this material contains hydroxyapatite (HA) and releases hydroxyl, fluoride, and calcium [9].
Tricalcium silica-based cements are other materials launched to the market to compensate for weaknesses of calcium hydroxide-based materials [10]. MTA, Biodentine, and Theracal LC are among the popular calcium silicate-based liners that can be applied as direct and indirect pulp capping materials [11]. Biodentine is a new liner of this group that can be used as a dentine replacement material under resin-composite restorations because it showed comparable mechanical characteristics to dentine [12][13][14].When compared with MTA, Theracal and Biodentine have the better sealing ability, higher compressive strength, lower setting time, biocompatible and bioactive properties [13,15,16].
For deeper restorations that are near to the pulp, without the pulp exposure, an RMGIC is one of the best choices. These materials are dimensionally very stable, release fluoride, bond to dentin and composite [3,17]. Furthermore, the adaptation of RMGIC over some liners has been advised; for instance, in order to overcome the drawbacks of calcium hydroxide liners, a protective layer of RMGI base over this liner is recommended, especially in deeper cavities. To illustrate, RMGIC can act as an insoluble barrier and prevent the microorganism from proceeding toward the pulp at the time when microleakage occurs [4].
Activa BioActive BASE/LINER was introduced by Pulpdent Corporation in 2014. This product is described as an enhanced RMGIC because, besides the properties of an RMGIC, it also has a modified resin matrix with improved physical characteristics [18,19]. Activa BioActive material reacts to pH changes in the mouth and starts to actively release high amounts of fluoride, calcium, and phosphate to maintain the chemical integrity of the tooth structure; this is how this material plays its bioactive role [20,21].
Although there are studies asserted that the internal adaptation of resin composite restorations when no liner was applied is more acceptable [22][23][24], we still need to use them in specific situations; therefore, there should be in vitro experimentations to evaluate their physical properties as a pulp capping material beneath composite restorations because the bond strength of liner and composite as a complex to dentin is of critical importance to the long-term success of a restoration.
For these reasons and also the limited information in the literature about lining materials behavior beneath composite restoration, this study was designed to evaluate the effect of these liners on the shear bond strength of liner-composite complex to dentin.

Materials and Methods
This in-vitro experimental study assessed caries-free human molars (maxillary and mandibular) that had been extracted for purposes other than this study, such as periodontal disease or According to a previous study [7], the minimum sample size for shear bond strength (SBS) testing was calculated to be 10 in each study group, using one-way ANOVA Power Analysis (SPSS 26), assuming alpha=0.05, beta=0.2, mean standard deviation of 1.89, and the effect size of 0.487.
For this purpose, 70 sound human molars that had been extracted within the past three months were disinfected using a 0.5% chloramine T solution at 4°C. Teeth were then kept in distilled water at 37°C for two weeks. The enamel in the occlusal surface of teeth was removed entirely with 180-grit sandpaper under water cooling with the circular polishing machine (Malekteb, Tehran, Iran) until the 7mm-diameter flat dentine was exposed. Molars were then embedded in Brass molds of 2.5× 3.5 cm filled with self-curing acrylic resin. 600-grit sandpaper attached to the polishing machine used to standardized the smear layer under water lubrication. Specimens were randomly divided into 7 groups. Table 1 shows an overview of the study groups, materials used, and application modes according to their instructions for use. The samples were stored in distilled water at 37°C until the experiments finished.
The polyethylene tube was used to standardize the area and volume of liners (1.5 mm height and 1.5 mm internal diameter) and composite restorative materials (4 mm height and 4 mm internal diameter). The schematic picture of sample preparation is shown in figure 1.
After the application of respective lining materials, dentine surfaces in all study groups were then etched with 37% phosphoric acid gel for 15 seconds and rinsed. Two layers of total-etch tetric N bond (Ivoclar/Vivadent, Schaan, Liechtenstein) were applied over the liner and surrounding dentine surface with a bristle brush; It was rubbed for 15s followed by gentle air drying for approximately 5s and cured for 10s by a light-curing unit (Bluephase C8; Ivoclar Vivadent, USA) with a light intensity of 800 mW/cm 2 at a standardized distance of 1 mm for 40 seconds.
The polyethylene tube (4 mm height and 4 mm internal diameter) was placed over the lining material and filled with the resin composite (Gradia (GC, Japan)) using the incremental technique (two 2-mm increments), and each increment was cured with a curing unit for 40 seconds. The polyethylene tubes were removed with a sharp knife after the completion of the resin composite build-up.
The specimens then underwent thermocycling (Delta Tpo2, Nemo, Mashhad, Iran) for 5000 cycles between 5-55°C with a dwell and transfer time of 30 s. Shear bond strength testing was performed in a universal testing machine Zwick/ Roell Z050 (Zwick/Roell, Ulm, Germany) with a 50-kg load cell at a crosshead speed of 1mm/min until bond failure occurred.
The fractured test specimens were examined under a stereomicroscope (SZX 16; Olympus, Tokyo, Japan) at a magnification of × 25 to analyze the nature of the failure. The modes of failures were classified into three groups; cohesive (fracture entirely within the liner or resin composite), adhesive (fracture at the interface of the material and dentin), and mixed (a combination of cohesive and adhesive failure) [25].
The data of SBS test were analyzed using one-way ANOVA, and pairwise comparisons of the groups were performed using the Tamhane post-hoc test. All statistical analyses were accomplished with SPSS version 22 at a 0.05 level of significance.

Results
According to the Kolmogorov-Smirnov test, all data presented a normal distribution (p > 0.05). Table 2 shows the shear bond strength scores and standard deviations of the seven groups.
Among the groups which received lining materials, GT showed the highest shear bond strength (11.29 ± 1.89), followed by GR (8.19±4.71), and GB showed the lowest shear bond strength According to pairwise comparisons of the groups (table 2), the shear bond strength results among several groups are statistically significant (P<0.05), among all 7 groups only GT, did not significantly differ (P>0.05) from the GC (12.30 ± 1.54), other groups have a significant difference with the GC (P<0.05). Also noteworthy is the fact that GT had a significant difference with GD, GL, GA, and GB. On the other hand, the difference in shear bond strength of GL and GB was only significant with GT and GC (P<0.05). Another fact that needs to be highlighted is that the SBS of GR was significantly higher than GB and GD, groups that showed the lowest results of shear bond strength.
The distribution of failure modes (Fig2) of the specimens after the shear bond strength evaluation was characterized as adhesive, cohesive, or mixed. The observed modes of failure were predominantly cohesive in the respective pulp capping materials of GD, GT, and GB, while GA was showed more adhesive modes of failure. Mixed modes of failure were mostly shown in GL and GR specimens. No cohesive failure in resin composite was observed.

Discussion
Since the pulp capping materials may affect the durability and condition of the tooth-restoration interface, the bond strength of these liners to dentine and restorative materials and their solubility during the etching process is essential for the success of restorations and also to maintain pulp vitality. This bond strength can be measured by various bond strength testing methods [2,26]. Therefore, the present study was planned with the aim of evaluating and comparing the SBS of the liner and composite complex of six types of liners to the underlined tooth dentin structure.
Shear bond strength is the most commonly used technique to gauge bond strength and is used in approximately 30% of scientific papers worked on bond strength [22]. It is a simple, quick, and reliable method [23]. No further processing on the specimens is required after the bonding procedure [24]. In this study, the macro shear bond strength test was used since we wanted to evaluate a complex of liner and composite bond strength to dentin, which could not be evaluated by micro tests.
The results of SBS performed in the current study revealed that the highest bond strength was in GC (12.30±1.54), followed by GT (11.29 ± 1.89 MPa) and GR (8.19±4.71 MPa), which have Biodentine exhibited significantly lower bond strength values as compared to other groups [27,28]. In our study, the composite was placed over Biodentine after its initial setting time (12 min) as recommended by the manufacturer. According to previous studies [16,29], it takes approximately 12 minutes for Biodentine to reach its initial setting time, while it needs as long as 14 days to get enough bulk strength to withstand the polymerization stresses. Consequently, the setting reaction of Biodentine may affect the bond strength between Biodentine and restorative materials. Furthermore, the exact mechanism of Biodentine's bonding to dentine is still not thoroughly clear. A combination of the chemical bond and a micromechanical anchorage providing by the infiltration of cement tags into the dentinal tubules are believed to be responsible for this bonding. The alkaline pH of Biodentine may cause the organic collagen component of the interfacial dentin to denature and become permeable, leading to intra-tubular tag formation combined with an interfacial mineral interaction layer called mineral infiltration zone. Thus, in our study, placing composite before the complete formation and maturation of mechanical bond of Biodentine to dentin (after 12 minutes) could have resulted in lower SBS values [27,28].
The other factor that can be attributed to the lower bond strength achieved in our study may be due to the etching procedure over Biodentin. There are many studies confirmed that etched Biodentine showed structural and chemical changes compared to non-etched Biodentine, which may affect the material's microhardness [11,[29][30][31]. However, all those studies mentioned that the etching process could negatively affect the shear bond strength of Biodentine to composite, while in this study, the SBS of the complex of liner-composite to dentin was measured. Thus, the solubility of water-based Biodentin under etching may contaminate the dentine surface bonding to composite leading to the low SBS of liner-composite complex to dentin. It seems that this process intensifies when the structure of Biodentin is not in its full maturation. This can also affirm the adhesive mode of failure found in Biodentin in the present study; because it can be due to the low bond of composite to dentin surface surrounding the liner.
To overcome the limitation of poor bonding of calcium silicate materials in final restorations, TheraCal LC was introduced as a material for vital pulp therapy. In our study, TheraCal LC exhibited significantly higher bond strength values compared to other materials. This could be due to the fact that TheraCal is a resin-based light-curable liner with a high filler load (30%-50%) that can achieve sufficient early cohesive force upon photo-activation [25]; therefore, curing contraction of the overlying composite cannot cause stresses in the structure of TheraCal. The higher SBS of TheraCal may also be attributed to the hydrophilic resin-based methacrylate monomers of its structure that increase chemical adhesion to dentin and forms a strong interface between TheraCal and the bonding surface [32,33]. Furthermore, Theracal has shown lower solubility than Biodentine and calcium-hydroxide-based liners [3,34,35]; thus, it seems that the contamination of the dentine surface due to the etching process, which is supposed to have a role in the bonding of composite to dentin, is less likely to occur.
Theoretically, composites' bonding mechanism to calcium in liners is assumed to be comparable to the bonding mechanism that they have with the calcium of the tooth structure [12]. Compared to other pulp capping materials (except Biodentine), it has been proven that TheraCal LC releases a higher concentration of calcium ions, especially compared to Dycal and Lime-Lite [36]. Although the amount of calcium released by lime-lite, a light-cured resin-based calcium hydroxide liner, was less than chemical-cured formula (Dycal) [37], the improved mechanical properties and presence of resin monomers in the structure of lime-lite led to higher SBS results than Dycal.
As mentioned, water resorption and solubility are two main physical properties of pulp capping material because degradation of the lining materials leads to the restoration's debonding and failure. In the study conducted by Gandolfi et al. [38], TheraCal showed lower solubility than Dycal, and Biodentine had the highest rates. It is perhaps because both Dycal and Biodentine are going through aging and etching stages; with both liners' high solubility properties in the water, lower SBS amounts were expected. Biodentine is prepared by mixing the mineral powder with water-based liquid, which is required to evaporate. Within the 12-minute setting time in our study and the 9 minutes in Gandolfi's, the water cannot be dried thoroughly, leading to a higher solubility rate. Moreover, they reported Lime-lite as the less soluble material in water and acetic acid, among others. However, less solubility of Light Cured Calcium Hydroxide liners in water due to resin-based structure [3,37] can justify better SBS results of lime-lite compared to Dycal; however, the difference was not significant.
As another reason, since Dycal lacks the resin content in its structure, unlike the TheraCal, its bond to resin composite is totally micromechanical, which means that penetration and interlocking of the adhesive systems into the surface irregularities play the main role in bonding [39].
In this present study, RMGIC (8.19±4.71 MPa) got the second rank, among other materials, and did not significantly differ with Theracal LC. Mehra et al. [14] conducted the study to measure the SBS of RMGIC with two different time intervals (immediately and 7days) compared to Theracal LC. They found that RMGIC showed higher SBS at a 7-day period (26.51±1.05 MPa), even higher than Theracal. This difference in results could be because In the beginning, the RMGIC gets its high early strength by the polymerization of the methacrylate groups. After that, it is the acid-base reaction that helps the RMGIC to complete its setting and reach final strength [37]. Still, the fact that how long this acid-base reaction takes to complete is not precisely determined.
ACTIVA is considered a dual-cured resin-modified glass ionomer (RMGI) and composed of bioactive glass as a filler and diurethane and methacrylate-based monomers with a modified polyacrylic acid and polybutadiene-modified diurethane dimethacrylate. [40,41]. ACTIVA has desirable properties of glass ionomer cements plus resin-based materials simultaneously, such as water-friendly, releasing, and recharging fluoride and calcium phosphate ions, aesthetics, durability, and boosted physical properties [18].
It seems that Activa can release an equal amount of Ca and OH ions as TheraCal LC; therefore, it can be concluded that the ability of these two materials in stimulating the biomineralization of the tooth structure can be at the same level [42]; however, according to our results, SBS of Activa (5.60±1.93 MPa) was significantly lower than TheraCal LC (11.29 ± 1.89 MPa) and also RMGIC (8.19±4.71 MPa).
Although the structure of Activa BioActive products is similar to RMGICs, the laboratory and clinical findings report that the manufacturer's claim about the self-adhesive ability of this material has not been approved [43]. This assumption is in accordance with the study done by Dijken et al. [44] in which a 1-year clinical evaluation of Activa™ bioactive posterior restorations indicated that; When applied without using the adhesive as instructed by the manufacturer, Activa led to a non-acceptable very high failure rate after a year. It needs to be mentioned that, to the best of our knowledge, there is no data available about the solubility of Activa Bioactive in water and acid; however, based on its composition, it can be inferred that Activa might have low solubility in water and acid; thus, acid etch might have a less destructive effect on its structure. Therefore, the lower values of Activa in the present study could be due to the absence of the dentine pretreatment procedure.
After the shear bond strength testing, failure modes were evaluated under a stereomicroscope and recorded as cohesive, adhesive, and mixed modes. Adhesive failure, which was predominantly found in Activa bioactive specimens, is an indicator of the lack of a strong bond at the lining material-composite/dentin interface due to not using an adhesive. In comparison, cohesive failures are more acceptable than adhesive failures because cohesive failure may be due to the material's low internal resistance or the bond strength being more higher than the material's internal resistance [45]. In our study, Dycal, Theracal, and Biodentine showed more cohesive failure. Since Dycal and Biodentine are self-cured materials, this type of failure mode can be related to their low early bulk strength. TheraCal had more cohesive and a few mixed modes, which could also be due to its low bulk strength. On the other hand, As Mehra et al. [14] mentioned, this pattern might relate to the cohesive strength of material rather than the real weak bond strength in the interfaces.
The current study evaluated the SBS of the composite-liner complex to dentin in an in-vitro situation; however, the oral cavity situation is different from the in-vitro condition, leading to different material behaviors. Therefore, further in-vivo investigations are required to obtain additional data. In order to evaluate the effect of the setting time on the adhesion of the Biodentine to the bonded surfaces or materials future studies should consider measuring the mechanical properties with dedicating more time before the application of final restorative material. As the type of the bonding material can have an influence on the SBS of composite to dentin, the behavior of these materials should be compared using other kinds of adhesives.

Conclusion
Within the limitation of this study, it can be concluded that: 1. The bond strength of resin-composite to dentine can be affected differently using various types of liners compared to not using them.
2. Although all of the pulp capping materials used in this study failed to fulfill the criterion of desirable bond strength (17 MPa), TheraCal LC showed maximum Shear Bond Strength when bonded with composite, and Biodentine showed the lowest shear bond strength values after its initial setting time (12 min). However, due to the appropriate features of Biodentin in place of using that a waiting period can be helpful in obtaining high SBS values.

Data Availability GR
Apply the resin composite after the application of adhesive on the prepared dentin surface.

Control
(no liner)