Effect of surface treatments on push-out bond strength of calcium silicate-based cements to fiber posts

To evaluate the effect of surface treatments on the push-out bond strength of Biodentine (BD) and white mineral trioxide aggregate (WMTA) to fiber posts. Two brands of fiber posts were used: Reblida post; RP and RelyX post; RX. Each type of post (n = 80/group) was divided into four groups (n = 20/group) and exposed to surface treatment as follows: Control (no treatment), sandblasting (SB), hydrofluoric acid (HF), and TiF4 4 wt/v%. Each group was further subdivided into two subgroups (n = 10/subgroup) based on the type of CSCs used as follows: Subgroup A: BD and Subgroup B: WMTA. Push-out bond strength of BD and WMTA to glass fiber posts was assessed. Data were statistically analyzed using three-way ANOVA and Tukey’s test. A Weibull analysis was performed on the push-out bond strength data. BD showed higher bond strength than WMTA (P < 0.001). The push-out bond strength for posts treated with TiF4 4 wt/v% showed greater bond strength than the other surface treatments (P < 0.05). The BD/RP-TiF4 4 wt/v% showed the greater characteristic bond strength (σ0) (15.93) compared with the other groups. Surface treatments modified the surface topography of glass fiber posts. The BD/RP-TiF4 4 wt/v% showed greater bond strength compared with the other groups. The TiF4 4 wt/v% surface treatment enhanced the bond strength of BD and WMTA to glass fiber posts than the other treatments. Surface treatment of fiber post with TiF4 4 wt/v% could be used to improve the bond strength with calcium silicate-based cements.

One of the applications of CSCs is the treatment of non-vital immature permanent teeth [4]. This procedure is established by disinfection of root canal, placement of apical barrier together with root-filling materials [13]. The fracture resistance of simulated immature permanent teeth used BD with fiber post was compared with different canal filling materials [4]. It was suggested that BD combined with fiber post might reinforce the immature teeth with an apical BD plug [4]. Fiber posts improved the fracture resistance of the tooth because their flexural modulus mimic to that of dentin [4,7,[14][15][16][17]. This biomimetic characteristic enhances the stress distribution which decreases the incidence of root fracture, the most critical form of failure [18][19][20].
Surface treatments had been suggested to improve the adhesion properties between bi-materials interface, by providing micromechanical and chemical retention between different constituents [21,22]. Various surface treatments had been applied to enhance the bond strength between the fibre post and other restorative materials as composite core [22] and resin cements [19,23] including sandblasting, silane coupling agent, and acid etching agents [19,[22][23][24].
The adhesion between the BD and fiber post was not evaluated. It is important to enhance bonding between BD and fiber post to reinforce the root canal as in cases of non-vital immature permanent teeth [4]. Consequently, the aim of this study was to assess different surface treatments on the push-out bond strength of BD and WMTA to different systems of glass fiber posts. The null hypotheses tested of were that: (1) Different surface treatments would not affect the adhesion between fiber post and CSCs materials; (2) Different type of posts would not present difference on adhesion regardless the surface treatments, and (3) Different CSCs materials would not affect the adhesion of the fiber post regardless the surface treatments.

Methods
The sample size determination was performed for pushout bond strength test using GPower v3.1.3 software (University of Düsseldorf; Düsseldorf, Germany). A power analysis revealed that a sample size of 50 specimens per subgroup was found to meet the constraints of α = 0.05 and power = 0.95.
The specimens that were treated with HF and TiF 4 4 wt/v% were ultrasonically cleaned in distilled water for 1 min and then air-dried. Each group was further A sticky wax was used to position the post perpendicularly on a square plastic plate. Then, the post was surrounded by a cylindrical plastic matrix (12 mm height × 10 mm diameter) [22,24]. The cylinder was filled with the CSCs using a MAP system (MAP One, Produits Dentaires SA, Vevey, Switzerland). The specimens were stored at 37 °C and 100% humidity for 48 h. The specimens were exposed to 10,000 thermocycles in distilled water between 5 and 55 °C with 5-s transfer time and 30-s dwell time [4,7,26].
After that, the straight portion (10 mm) of each CSCspost system was sectioned using a low-speed diamond saw (Isomet 1000, Beuhler Ltd., Lake Bluff, IL, USA) that resulted in 5 discs (n = 50 discs/subgroup). The push-out bond strength of each disc was tested using a universal testing machine (Model TT-B, Instron Co., Canton, MA, USA) and loaded with a cylindrical plunger (1 mm diameter) at 0.05 mm/min cross-head speed. The pushout bond strength (MPa) was calculated by dividing the load at failure (Newtons) by the bonding area (mm 2 ). The bonding area was calculated using the following equation [22]: where r is the post radius, h is the thickness of each post section, and π is the constant 3.14.

Failure mode analysis
The debonded specimens were observed by stereomicroscope (ZEISS, Stemi 2000-C, Oberkochen, Germany) at 50 × for analyzing the failure pattern. Failure mode was classified as Type 1; adhesive failure between the CSCs and the post, Type 2; cohesive failure within the post, Type 3; cohesive failure within the CSCs, and Type 4; mixed failure [23]. The failure mode was evaluated by the same operator.

Surface topography
A total of 20 specimens of each type of post (n = 5/group) were prepared and grouped as mentioned before. The specimens were sputter-coated with gold (Sputter Coater S 150A; Edwards, Crawley, England). A scanning electron microscope (JSM 5600 Lv JEOL, Tokyo, Japan) was used to characterize the surface topography of the specimens at magnifications of 500 × .

Statistical analysis
Normality of data distribution was tested using Kolmogorov-Smirnov and the equality of variances with the Levene's test. The data were normally distributed. Data of push-out bond strength were statistically analyzed (SPSS 22.0 software; IBM Corp., Armonk, NY, USA) using three-way ANOVA based on three factors (the type of post, type of treatment, and type of CSCs) and their interactions. Multiple comparisons were conducted by the Tukey's test. The level of statistical significance was set at P < 0.05. A Weibull analysis (SuperSMITH software; Fulton Findings, Torrance, CA, USA) was performed on the push-out bond strength data. Table 2 showed the results of 3-way ANOVA. The push-out bond strength was significantly influenced by the type of post, type of surface treatment, and type of CSCs (P < 0.001). The RP revealed higher bond strength to CSCs than RX (P < 0.001). Regarding the type of CSCs, the BD showed higher bond strength than WMTA (P < 0.001). There was no significant interaction between type of post and type of CSCs (P = 0.108) and between the type of CSCs and type of treatment (P = 0.394). There was a significant interaction between the type of post and type of treatment (P = 0.002).

Results
There was no significant interaction between type of post, type of CSCs, and type of treatment (P = 0.311). The mean and standard deviations of push-out bond strength (MPa) are presented in Table 3. The pushout bond strength for posts treated with TiF 4 4 wt/v% showed greater bond strength than the other surface treatments (P < 0.05). The improvement in the bond strength according to the surface treatments was as follows: TiF 4 4 wt/v% > HF > SB > C. Failure mode analysis of different groups is presented in Fig. 1 Table 4. The BD/RP-TiF 4 4 wt/v% showed the greater characteristic bond strength (σ 0 ) (15.93) than the other groups. The surface treatment with TiF 4 4 wt/v% had more reliability than the other treatments (Table 4 and Fig. 2). The Weibull plot for different groups is presented in Fig. 2.
Representative SEM photomicrographs for the surface microstructure of the different post systems with different treatments are shown in Fig. 3. The untreated RP showed exposed glass fibers with rather a rough surface than RX post that was typically more covered by the resin matrix (Fig. 3a, e; respectively). The sandblasted groups exhibited fractured glass fibers (Fig. 3b, f; respectively). For the RP post, the HF surface treatment caused cracks in the glass fiber (Fig. 3c). For the RX post, the glass fibers were exposed with HF surface treatment (Fig. 3g). The TiF 4 4 wt/v% surface treatments exposed the glass fibers of the RP and RX posts (Fig. 3d, h; respectively).

Discussion
The present study evaluated the effect of surface treatment on bond strength of CSCs materials to fiber posts. The findings showed that the adhesion was considerably affected by the type of post, type of surface treatments, and type of CSCs. Accordingly, the null hypothesis was rejected.
It is significant to enhance bonding between CSCs and fiber posts to reinforce the root canals for treating nonvital immature permanent teeth [4]. In the present study, a cylindrical plastic matrix was used instead of a root specimen for testing the bond strength as the purpose of the study was to assess the bond strength between the posts and CSBs. There are limitations of using root dentin to test adhesion with fiber post including variation in hardness, elastic modulus, flexural strength, geometric parameters among different teeth that may affect the bond strength measurements [27,28]. Consequently, specimens from the post and CSBs were prepared to test the adhesion between them without another factor that may affect the outcomes. All tested groups were subjected to thermocycling to simulate the clinical conditions that might provide a possible prediction of bonding durability [23]. It was postulated that approximately 10,000 thermal cycles correspond to 1 year of clinical function [26]. In the present study, TiF 4 4 wt/v% was used because it was shown in a previous study that this concentration enhanced the bond strength of resin cement to a fiber post [23]. Similarly, surface treatment of RP and RX posts with TiF 4 4 wt/v% for 4 min revealed greater bond strength to CSCs than the other treatments. This could be explained that TiF 4 4 wt/v% treatment might remove the surface layer of resin of the post which allows further areas for micromechanical retention with the CSCs [22,23]. The TiF 4 4 wt/v% surface treatments exposed the glass fibers of the posts (Fig. 3d, h; respectively). In addition, surface treatments of fiber posts with TiF 4 4 wt/v% showed higher percentages of mixed failure modes (Type 4) than the other types of surface treatment.
The surface treatment with HF showed higher bond strength of CSCs to fiber posts than SB and C groups. The HF surface treatment might modify the outer surface layer of the fiber post without compromising the strength properties of the post [23,25]. It was observed that HF surface treatment caused some cracks in RP post and the glass fibers were exposed in RX post (Fig. 3c, g; respectively). It was reported that HF surface treatment had a destructive effect on the surface integrity of the fiber post [29,30].
Roughening the surface of fiber post with tribochemical silica coating might enhance the bond strength with the other bonded materials due to mechanical retention [23,31]. Surface roughness with sandblasting increased the surface area of exposed glass fibers for bonding with the CSCs and accordingly enhancing the bond strength [23,31]. Fiber posts that did not receive surface treatments showed the lowest bond strength with CSCs compared with the other groups. This finding indicates the importance of surface modifications of fiber posts to enhance the adhesion with CSCs. The RP showed higher bond strength with CSCs than the RX post. This finding could be attributed to the different surface topography between RP and RX posts [22]. The surface topography of RP showed a relatively uneven surface with some uncovered glass fibers as shown in the control group (Fig. 3a) which provides more surface areas for mechanical retention with CSCs. The RP is composed of 70% glass fiber, 10% filler, and 20% urethane dimethacrylate [23]. However, the surface of the RX post was typically more enclosed by the resin matrix (Fig. 3e). The RX post is composed of glass fiber reinforced composite and methacrylate resin [22]. The differences in the composition of posts might influence their bond strength with the CSCs.
The BD revealed higher bond strength with RP and RX posts than WMTA. This enhancement in the bond strength might be due to the improved physical properties and the integrity of BD [5,32] that might improve the bonding with the glass fiber posts. It had been reported that BD had a higher resistance to dislodgement than WMTA [32]. It could be postulated that BD had better mechanical adhesion with the posts after surface treatment that interlocks mechanically within the surface irregularities. CSCs materials should have adequate adaptation and a consistent adhesion to the post surface to reinforce the roots for treating non-vital immature permanent teeth [4]. The Weibull analysis was conducted on the push-out bond strength data. The BD/RP-TiF 4 4 wt/v% showed higher characteristic bond strength (σ 0 ) than the other groups, comparable to the bond strength data. It was observed that TiF 4 4 wt/v% surface treatment had more reliability than the other treatments. It could be postulated that the clinical relevance of fiber post surface treatment with TiF 4 4 wt/v% might improve the bond strength with CSCs. One of the limitations of the present study is to mimic the clinical conditions in practice for restoring endodontic treated teeth with CSBs and glass fiber post. In addition, other parameters existing in the oral environment such as fatigue loading, temperature changes, and constant moisture could influence the Fig. 3 Representative SEM photomicrographs (×500) for the surface microstructure of a-d RP and e-h RX posts; respectively with different treatments. White arrows showed fractured glass fibers, yellow arrows showed the cracks in the glass fibers, and the black arrows showed the exposed glass fibers performance of glass fiber post and CSBs. Due to the limitations of in vitro studies, further clinical studies are essential to assess the performance of treated glass fiber posts with the CSCs to restore endodontic treated teeth to give reliable recommendations for dental practitioners.

Conclusions
BD revealed higher bond strength to fiber posts than WMTA. The TiF 4 4 wt/v% surface treatment enhanced the bond strength of BD and WMTA to glass fiber posts than the other treatments. The RP post improved the bond strength to CSCs than RX posts.