Relining effects on the pushout shear bond strength of glass fiber posts

Introduction: The correct use of glass fiber posts in endodontically treated teeth is essential for the clinical success of restorative treatment. Objective: This study evaluated the push-out shear bond strength of relined (R) or non-relined (NR) glass fiber posts, cemented with self-adhesive resin cement [RelyXTM U100 (U100)] and conventional resin cement [RelyXTM ARC (ARC)]. Material and method: Sixty human single-rooted teeth were endodontically treated and divided into ARC-NR; U100-NR; ARC-R; U100-R groups. The teeth were sectioned into cervical, middle and apical thirds, and subjected to the push-out test. Bond strength was analyzed by the Friedman test; cement and post types were compared by the Mann Whitney test. The pattern of failures was evaluated with digital camera through images at 200x magnification, and was classified as adhesive (at the cement/dentin or cement/post interface), cohesive (cement or post), and mixed failures. Result: In ARC-NR, bond strength values were higher in the cervical third; in U100-NR and ARC-R they were similar between the thirds. In U100-R, in the cervical and middle thirds the bond strength values were similar, and there was lower value in the apical third. For non-relined glass fiber posts, the highest mean bond strength values were observed with self-adhesive resin cement. Whereas, relined posts cemented with conventional resin cement had stronger cement layer in comparison with non-relined fiber posts. Conclusion: The post relining technique was efficient in ARC-R. ARC-NR and U100-R showed improved bond strength in the cervical region of canal walls. The main failures were adhesive at the cement-post interface. Descriptors: Dental bonding; dental materials; dentin; post and core technique; shear strength. Andrioli, Coutinho, Vasconcellos et al. Rev Odontol UNESP. 2016 July-Aug; 45(4): 227-233 228


INTRODUCTION
The application of intraradicular fiber posts is an alternative procedure for restoring endodontically treated teeth.These posts exhibit the following advantages: thy have esthetic properties; and are: resistant to corrosion; easy to remove if necessary; implanted in one or two visits 1 ; flexible and compatible with adhesive cement systems 2 .In addition, the modulus of elasticity of the posts is similar to that of root dentin 3 , distributing occlusal forces and reducing the risk of severe fractures 4 .Restoration failures usually result from endodontic complications or the cement system used 5,6 , emphasizing that adhesive processes in particular deserve attention.
Studies on the adhesive capacity of cement systems that have investigated the presence, location and thickness of the hybrid layer in root dentin have suggested that the clinical success of post bonding is associated with frictional rather than micromechanical retention or chemical bond to dentin 5,7,8 .
Whether a perfect fit between the post and canal walls is mandatory to ensure retention is a controversial question.Perez et al. 9 reported that the bond is not compromised when there is a thick cement layer around the fiber.However, other study associated fiber post displacement with the high thickness of the cement layer 10 .In general, cement thickness is not measured accurately, but D' Arcangelo et al. 11 found high bond strength values when using an appropriate oversized post space produced with a thickness of 0.1-0.3mm.The post relining technique was proposed for treating teeth with oval-shaped canals, irregular and/or retentive root canals, or those that underwent extensive endodontic preparation 10,12 .This technique is used in attempts to improve fitting quality, by allowing the formation of a more uniform and thinner cement layer.
Few studies have shown the efficiency of relined posts 13,14 , also known as anatomical posts 10,12 .Some researchers 7,8 have attributed the shear strength of intraradicular fiber post cementation not only to the action of adhesive agents but also to frictional forces; however, studies that investigate this relationship are scarce, especially as regards shaped glass fiber posts.
The bonding capacity of fiber posts is influenced by different factors, and luting agents are an important consideration.Two groups of adhesive resin cements widely used consist of a multiple-step technique and self-adhesive dual-cured luting agent, with significant difference in pre-treatment the of root canal 15 .A multiple-step technique is more sensitive and needs more clinical time; whereas, self-adhesive dual-cured luting agents require no pre-treatment 16 , and the bonding process between tooth and fiber post consists of a chemical reaction between phosphate methacrylates and hydroxyapatite 17 .
Another factor that influences the bond quality is the root canal depth, considering: the dentin morphology; dentinal tubule diameter per area; visualization during the clinical step; and proximity of photo-activating light 4,5 .
Against this backdrop, the present study used the push-out test to evaluate bond strength between the root dentin and relined or non-relined glass fiber posts, cemented with two types of resin cements.
The hypotheses were: glass fiber post relining would increase the bond strength to root canal walls; multistep etch-and-rinse resin cement would have a higher bonding capacity than self-adhesive cement; and the bond strength would be higher in the cervical third of dentin.

MATERIAL AND METHOD
Sixty sound, single-rooted, extracted, human teeth were stored in 0.9% thymol until the experiment began.The inclusion criteria were absence of: restoration, caries or root cracks; previous endodontic treatments, posts or crowns; severe root curvatures; and root length of 15 ± 1 mm from the cement-enamel junction (CEJ).The research was approved by the Research Ethics Committee of the São Leopoldo Dental School (Protocol 2010/0068).

Endodontic Treatment
Teeth were sectioned transversally immediately below the CEJ by using a low-speed #7016 diamond saw (KG Sorensen, São Paulo, Brazil).The step-back technique was used for endodontic treatment, with stainless steel K-files (Dentsply-Maillefer, Ballaigues, Switzerland) up to file size #45.All enlargement procedures were followed by irrigation with 0.5% sodium hypochlorite.Prepared root canals were filled with gutta-percha cones by using the lateral condensation technique and AH 26 sealer (Dentsply, Rio de Janeiro, Brazil).The coronal access was sealed with Coltosol (Coltene-Whaledent, Altstätten, Switzerland) as temporary restorative material.The specimens were then stored in 100% relative humidity at 37 °C for 7 days.

Post-space Preparation
After 1 week, gutta-percha was removed from the canals with #1 Gates Glidden burs, leaving 4 mm of the apical seal.The post space was then prepared with low-speed drills.Drill #1 was used to prepare the canals of groups receiving non-relined posts (ARC-NR and U100-NR); and sequence drill #2 (Angelus, PR, Brazil); diamond bur #4137 (KG Sorensen, SP, Brazil); and drill #3 (Angelus, PR, Brazil) to prepare canals treated with relined posts (ARC-R and U100-R).
The specimens were prepared by a single operator using standardized procedures.After preparing the post spaces, the canals were irrigated with 10 ml of distilled water for cleaning and moisturizing, and then dried with paper points.

Post Preparation
For relining, posts were cleaned with 37% phosphoric acid (Condac, FGM, Brazil) for 1 min; rinsed before they received silane application for 1 min, and the Scotchbond Multi-Purpose Plus (SBMP) (3M-ESPE, Seefeld, Germany) adhesive system.Canal walls in turn, were lubricated with KY gel (Johnson & Johnson, SP, Brazil).Filtek Z250 (3M-ESPE, Seefeld, Germany).Composite resin was then condensed into the canal, and fiber post Exacto #1 (Angelus, PR, Brazil) inserted.The composite resin was photo-activated for 7 s (halogen light 600 mW/cm 2 , LH 600 Optilux 401, Demetron), and the post associated with the resin composite removed from the canal and fully photo-activated for 80s.The canal walls were rinsed copiously to remove the lubricant gel.
The relined and non-relined posts were cleaned by applying 37% phosphoric acid for 1 min; copiously rinsed with water; silane application for 1 min and activator application.The protocol used for post cementation was according to the cement used.
In groups ARC-R and ARC-NR, the canal walls were treated before cementation, according to the following steps: first, etching was performed with 37% phosphoric acid for 15s; rinsing, and removing the excess with paper points.After this, the activator was applied with microbrush for 15s; excess material was removed with paper points; primer was applied with microbrush for 15s; excess material was removed with paper points, and finally, the catalyst was applied with microbrush for 15s, and the excess material removed with paper points.
Whereas, in groups U100-R and U100-NR, the canal wall was not previously treated.The cementation protocol was the same for both cement groups.The cements were mixed for 10s and inserted into the root canals with a Centrix syringe.The posts were inserted with light hand pressure and excess luting material was removed.After 2 min, the cervical region of the root was photo-activated for 40s.
After storage in distilled water at 37 °C for 1 week, the roots were sectioned into three 1-mm-thick slabs.These were placed on a push-out jig in a mechanical testing machine (model DL2000, EMIC, São José dos Pinhais, Brazil).The load was applied at a crosshead speed of 0.5 mm/min until the post was dislodged.The bond strength value, in MPa, was calculated for each slab.
Comparisons between the cements (U100 and ARC) and between the techniques (using relined and non-relined posts) were made by the Mann-Whitney U test.The root regions (apical, middle and cervical) were compared using the Friedman test followed by the Dunn post-test.Statistical analysis were performed by using SigmaStat 2.0 program, at a significance level of 0.05.
After the push-out test, the failure patterns were evaluated by capturing images with a digital camera at 200x magnification (DIGLAB -BASF, Ludwigshafen, Germany).The failure patterns were classified as adhesive (failure at the cement/dentin or at the cement/post interface), cohesive (of the cement or of the post), and mixed failures (adhesive and cohesive).

RESULT
Table 2 shows the push-out shear bond strength values of the posts, comparing the cements, post types (R and NR) and root regions.Differences in cement (U100 vs ARC) were observed only in middle and apical root regions and non-relined posts, with higher bond strength values exhibited for U100.Differences between the techniques (R vs NR) were also observed only in the middle and apical thirds when ARC cement was used, with higher bond strength values shown in ARC-R than in ARC-NR.Push-out shear bond strength differed among the root regions only in ARC-NR and U100-R.In ARC-NR, bond strength values were higher in the cervical than in the other root regions, whereas in U-100-R, the Dunn test showed no differences between the pairs.As regards the bond failures, a visual observation of the specimens detected 180 failures, 103 (57.2%) adhesive at the cement/dentin interface, 7 (3.88%)adhesive at the cement/post interface, 12 (6.66%)cohesive of the cement, 1 (0.55%) cohesive of the post and 57 (31.66%) mixed failures, which represented different failure modes (Figure 1).

DISCUSSION
Establishing adequate shear bond strength between the post and the root canal is fundamental for the clinical success of restorative treatment of endodontically treated teeth 6 .Thus, the shear bond strength of glass fiber posts has been extensively studied.
Adhesive techniques and the root canal environment are highly variable and susceptible to the action of multiple factors that cannot be controlled.As a result, the shear bond strength data showed high variance, and were therefore analyzed by non-parametric procedures, as adopted in an earlier study 18 .
Grandini et al. 10 reported that the technique of post relining with composite resin is easy to apply.However, in the present study several difficulties were encountered during the restoration steps.Firstly, to remove undercuts in the canal walls; secondly, to pull the post/resin set out of the root canal; and finally, to readapt the post during cementation.Polymerization shrinkage of the composite resin used for relining the post did not facilitate removal of the relined posts before the complementary photo-activation session.Nevertheless, this technique has a number of advantages.The relined post is tightly inserted into the root canal, so that cement pressure against the canal walls improves cement-bond contact, avoiding water absorption 13 and reducing porosity at the bond interface 13,19 .This prevents bubble formation 11 that can inhibit polymerization, if there are trapped bubbles in the cement layer 19 .
Studies using scanning electron microscopy (SEM) have shown that fiber post cementation produces resin-dentin interdiffusion zones.These zones consist of lateral resin projections together with discontinuous gaps along the interface between the cement-adhesive system throughout the root canal, which are probably produced by polymerization shrinkage 18 .The hygroscopic expansion of the cements produced by water sorption is believed to reduce the size of the gaps produced by shrinkage and improve marginal sealing 20 .These findings indicate that fiber post resistance to dislodgement is mainly related to frictional retention and this was corroborated by the similarity among specimens treated with fiber post application with the use of different adhesive systems 7 .This suggested that fiber post resistance to displacement was proportional to the area of contact with dentin; that is, with the intertubular dentin area 14 .
Following the above mentioned line of reasoning, we raised the hypothesis that the push-out shear bond strength would be increased by post relining due to the rise in frictional retention.Indeed, another study showed that post relining enhanced shear bond strength, irrespective of cement type 14 .However, the hypothesis was only partially accepted since the effects of relining were observed only when RelyX ARC cement was used, as found by Faria-e-Silva et al. 13 .The ARC-R treatment produced a thinner, more uniform and stronger cement layer compared with that of ARC-NR, the latter exhibiting low bond strength values in the middle and apical third of roots.
No effects of relining were detected in treatments when RelyX TM U100 cement was used.Thus, the complex technique of glass fiber post relining promoted no benefits that justified its use in conjunction with this type of resin cement.
The second hypothesis, in which the multistep etch-and-rinse system (SBMP + RelyX TM ARC) produce higher post shear bond strength than self-adhesive resin cement (RelyX TM U100) was rejected.As observed in a previous study, the bond strength provided by post relining was not associated with the type of cement system, since the non-relined posts achieved higher bond strength values when they were treated with self-adhesive resin cement (U-100) 6 .
Self-adhesive cements are generally better than multistep etch-and-rinse cements because of their good chemical interaction with calcium in hydroxyapatite, which improves the mechanical features of the interface produced 21 .They appear to exhibit low polymerization shrinkage due to their viscoelasticity, increasing shear bond strength by maintaining close contact with root canal walls 22 .Self-adhesive cements are easy to apply, and do not require dentin etching, or the use of a bonding agent, which makes cementation protocols less complex.The phosphoric acids of self-adhesive resin cements are responsible for etching the demineralized substrate, and simultaneously allowing resin cement infiltration into the dentin 7 .
In contrast to our findings, other studies have shown poor bonding behavior of self-adhesive resin cement.This is explained by the deficient dentin hybridization 18,23,24 and incomplete smear layer removal promoted by its weak-acid monomers and hydrolysis of adhesive components 7 .Given these discrepancies, the bonding behavior of cements deserves further studies, especially since the high variability of the root canal environment hinders the application of cementation techniques.Shear bond strength in the root thirds tended to decrease with root canal depth (Table 2), corroborating several studies 24,25 .Therefore, the third hypothesis tested in the present study, which was that post shear bond strength would be higher in the cervical than in the other root regions, was accepted.Indeed, the highest bond strength values were found in the cervical root region, except for ARC-R and U100-NR, in which no statistical differences were found between the root thirds, as also reported by Faria-e-Silva et al. 13 .
The cervical third shows a peculiar dentin morphology, which justifies the higher shear bond strength values in this root region.Compared with the middle and apical thirds, it exhibits higher density of larger-diameter tubules per area 4,5,24 , facilitating the penetration of resin cement 4,5 .With easy visualization and accessibility, procedures such as cleaning, application of adhesive agents and evaporation of materials are efficient 19 .The proximity of photo-activating light allows greater cement conversion, increasing the push-out shear bond strength and release of polymerization stress 21 through the free surface area 26 .
The middle third showed intermediary push-out shear bond strength, sometimes approaching that of the cervical third and sometimes that of the apical third, as previously reported in another study 11 .This reinforces the transitional nature of this root region, reflected in its morphology 4 .
The apical third, in the deepest canal region, showed lower shear bond strength values in groups ARC-NR and U100-R, in agreement with earlier studies 5,24 .Unlike the cervical third, the apical root region is not well visualized and is difficult to reach.With inefficient cleaning, residues of gutta percha and other products may not be completely removed 19 .Moisture control is also difficult in the apical third, hindering the functioning of adhesive systems 5,6,8 .This region receives a low incidence of photo-activating light, thereby decreasing cement conversion 26 .Given these features, the dual resin cements are used to treat this root region rather than chemically and light-activated types.
Another characteristic of the apical third is the low density of small-diameter tubules, which prevents resin tag formation 4 and thus reduces the bond strength.The shrinkage stress produced at this site during polymerization cannot be dissipated, due to root anatomy, even though resin conversion is not complete.Cement shrinkage stress results from the C-factor (ratio of the bonded to the unbonded surface areas of the cavity), which is likely to exceed 200 in root canals 5 .This causes loss of integrity and gap formation at the adhesive interface 5,8,18 .Of several studies reviewed, only those of Muniz, Mathias 27 and Bitter et al. 28 found higher push-out shear bond strength in the apical third.The authors reported that the dentin area had a higher influence on this variable than the tubule density.
The last stage of the present study identified the main failure modes at the adhesive interface.Adhesive failure at the cement-dentin interface was the most common defect (57%), as shown in Figure 1A, followed by mixed fracture, which represented adhesive fracture associated with cohesive failure of the cement (Figure 1B).These findings confirm the difficulty in establishing good interaction between the cement and dentin substrate, probably because of the great variability in root canal morphology and features of resin system bonding, as described in other studies 13,25 .Although some studies 10,13,18 have discussed the post relining technique, further studies are needed to safely apply this procedure.In practical terms, the post lining technique coupled with the use of multistep etch-and-rinse cements is more complex than the technique with the use of RelyX TM U100 self-adhesive cement, thus the former requires more skilled operators, material resources and longer clinical time.

CONCLUSION
The authors were able to conclude that: 1) Post relining improved the bond strength when used with conventional resin cement in comparison non-relined fiber posts, especially in middle and apical thirds; 2) With non-relined glass fiber posts, self-adhesive resin cement showed better bond strength than conventional resin cement; 3) The cervical root third showed the highest bond strength values.

Figure 1 .
Figure 1.(A) Adhesive failure at cement/dentin interface (cd): cement deposition was observed around the entire post; (B) Mixed failure, adhesive at cement/post interface (ap) and cohesive of the cement (cc); (C) Cohesive failure of the cement (cc); P = post; C= cement/resin; D = dentin.

Table 1 .
Chemical composition of resin cements used

Table 2 .
Push-out shear bond strength of posts in the different root regions and according to cement and post type *Contrasts between the root regions were determined by means of the Dunn test (p<0.05).