Effect of luting materials and root third on glass fiber posts bonding strength after hydrothermal aging

TL;DR

This study examines how different dental cements and root sections affect the bond strength of glass fiber posts after aging, helping improve long-term dental treatments.

Contribution

The study identifies specific luting materials that provide more consistent bond strength in challenging root regions.

Findings

PV5 and RXU showed the highest and most consistent bond strength across root thirds.

Bond strength varied significantly among root thirds for some cements but not others.

Dual-cure resin cements with self-etch systems achieved more predictable adhesion in complex root areas.

Abstract

Teeth with extensive coronal destruction face the challenge of retaining the restorative materials; however, glass fiber posts cemented with resin materials can help to address the issue. The retention of these posts depends on the bond strength to the dentin structure and may vary significantly along the root thirds.

This study aimed to evaluate the effect of different luting materials and root thirds on the bonding strength of glass fiber posts to radicular dentin after hydrothermal aging.

Seventy bovine incisors were divided into 7 groups (n = 10), according to the luting materials: Panavia V5 (PV5), Rebilda DC (RDC), LuxaCore Z (LCZ), Allcem Core (ACC), RelyX Ultimate (RXU), RelyX U200 (RXU200) or Fuji Plus (FP). After posts cementation, the roots were thermocycled (10,000 cycles; 5–55°C), sectioned into thirds, and tested for push-out bond strength (2.0 kN; 0.5 mm/min). Failure patterns were classified using a stereomicroscope, according to the material and root third. Data were statistically analyzed (α = 0.05).

Bond strength significantly differed among root thirds for PV5, ACC, RXU, and RXU200 (p < 0.05), while no significant differences were found for RDC, LCZ, and FP (p ≥ 0.05). PV5 and RXU exhibited the highest overall bond strength values. PV5 presented means of 14.97 ± 3.68 MPa (cervical), 11.08 ± 3.12 MPa (middle), and 10.47 ± 2.45 MPa (apical), while RXU showed 16.97 ± 4.22 MPa, 16.48 ± 3.61 MPa, and 6.51 ± 1.26 MPa, respectively. Overall, PV5 and RXU demonstrated superior and more consistent bond strength across root thirds, whereas the other cements displayed lower and more heterogeneous results.

The luting material influenced the bond strength of glass fiber posts to intraradicular dentin. Dual-cure resin cements with self-etch adhesive systems that contain functional monomers, such as PV5 and RXU, achieved more predictable adhesion, particularly in deep and complex regions.

Analysis of bond strength and failure patterns supports clinical decision-making by identifying luting cements with superior retention, even in challenging regions, contributing to greater treatment longevity.

Figures

Incidence of failure for each experimental condition within each root third (A-C), and overall failure by root third (D).

Representative SEM images of the types of failures.A: type 1; B: type 2; C and D: type 3; E: type 4; and F: type 5. All images were taken using a scanning microscope at 40 × magnification.

Full text

Introduction

Teeth with extensive coronal structure loss — resulting from deep caries, large restorations, fractures, or non-conservative treatments — often require endodontic therapy, sometimes with a prosthetic purpose. In such cases, the lack of sufficient remaining tooth structure compromises the retention and support of the coronal restoration, making the restorative prognosis more challenging [1–3]. In this context, the use of intraradicular posts is often necessary [2,3]. Glass fiber posts are commonly used because of their Young’s modulus (30–50 GPa) that is closer to that of dentin (18.6 GPa), which favors stress distribution [4–6]. These types of posts are cemented with resin or resin-modified materials and depend on the adhesion of the cements to the intraradicular dentin [7–10]. This adhesion, in turn, depends on the quality of the two interfaces that make it up and that are directly connected [11,12].

These two adhesive interfaces can be individually described and analyzed. The first is defined as the bond between the surface of the glass fiber posts and the luting material. The second is the adhesion between the intraradicular dentin and the luting material. There are many studies based on elements that can influence the longevity of adhesion [8,12–16]. It is worth highlighting the simplified post-and-core systems, as they are able to optimize clinical time by integrating the process of cementing the intraradicular post and the construction of the filling core into a single step, resulting in the formation of a “monoblock”. This approach eliminates the need for the conventional two-step protocol, in which the intraradicular post is first cemented, and the coronal core is subsequently built [7,17]. Although these systems reduce the number of adhesive interfaces, they may contribute to polymerization-shrinkage stresses inside the canal, especially due to the high inorganic filler content of these materials [7,18]. Considering the wide range of materials and techniques available for restoring teeth with extensive coronal destruction, it is important to understand the bond strength between the substrates involved and the factors that may influence it [10,19,20]. One of the methods accepted and recommended in the literature to mechanically analyze the adhesion of the dentin-cement-post complex is the push-out bond-strength test, whose acceptance is due to its practical and reliable methodology and its ability to evaluate different root thirds [18,21–23]. Therefore, it is essential to examine the limitations of different adhesive systems and luting materials used in restorations associated with intraradicular posts, considering adhesive challenges such as inadequate dentin retention, poor post adaptation, and polymerization-shrinkage stress.

Restoring endodontically treated teeth with extensive coronal structure loss presents clinical challenges that may compromise adhesion, particularly in deep regions such as the apical third. Proper control of intraradicular dentin hybridization, including moisture management after acid etching and removal of adhesive excess, is essential to achieving reliable bonding. Therefore, the aim of this study was to evaluate the bond strength and failure patterns of different luting systems used in restorations associated to glass fiber posts. The investigation focused on the bond strength of the materials in the cervical, middle, and apical thirds of the root after hydrothermal aging. The null hypothesis was that there would be no significant difference in bond strength among the luting systems across the different root thirds.

Materials and methods

This study was approved by the Animal Research Ethics Committee under local institutional protocol, registered under number 13/2019. The teeth were obtained through donation from slaughterhouse Terra do Boi, located in Caçu, Goiás, Brazil. As an establishment under state inspection (SIE No. 1459/14), the slaughterhouse follows strict sanitary control protocols, with oversight by a specialized technical team, ensuring animal welfare and the quality of by-products derived from slaughter. This regulation guarantees that the animals are free from clinical signs of disease and that the donated tissues pose no risk of contamination by toxic substances or infectious agents. After donation, the teeth were stored in 0.1% thymol solution for 24 h to ensure initial disinfection, in accordance with standard protocols for the preservation of biological samples [4,9,18]. Considering standardization criteria, seventy extracted single-rooted bovine teeth were selected. All specimens were intact, with a root length of at least 20–21 mm, no curvature, fractures, cracks, resorptions, dilacerations, or other root defects, and with complete root development. Measurements of the roots in the buccolingual and mesiodistal directions were made along the different thirds of the root (coronal, middle, and apical) using a digital caliper (DIN 862, Mitutoyo, Miyazaki, Japan), to ensure that standardized teeth dimensions were selected. Digital periapical radiographs were taken in the same directions in order to measure the intraradicular width, through readings in the ImageJ software version 1.51 (U.S. National Institutes of Health, Bethesda, Maryland, United States of America). The values for the width of the root canals were determined from the drill dimensions of the glass fiber posts (Whitepost DC1; FGM Produtos Odontológicos, Joinville, SC, Brazil) [4,18].

The coronal portion of all teeth was removed by transversal cut using a diamond saw under copious water irrigation (Isomet 2000; Buehler Ltd., Lake Buff, IL, United States of America), always adopting a 19 mm height for the remaining root from cervical to apical. Afterwards, the root canals were instrumented using the Bassi Logic rotary system (Easy, Belo Horizonte, MG, Brazil) up to a 40 file with 0.05 taper, under irrigation with 5.0 mL of sodium hypochlorite at 2.5% (Asfer, São Caetano do Sul, SP, Brazil), preserving 1.0 mm apical during instrumentation. Afterwards, EDTA 17% (Biodynamics, Ibiporã, PR, Brazil) was inserted into the canal for 60 s, followed by irrigation with 10 mL of saline solution. The canal was filled using the single cone technique with gutta-percha and AH plus endodontic cement (Dentsply Maillefer, Ballaigues, Switzerland). The endodontic access was sealed with resin modified glass ionomer material (Vitremer; 3M ESPE, St. Paul, MN, United States of America). Then, the roots were stored in artificial saliva for 7 days [4,24,25].

Subsequently, desobturation was initiated with a heated Rhein instrument, followed by Gates Glidden #2 and 3 and #2–4 Largo drills (Dentsply-Maillefer Instruments SA, Ballaigues, Switzerland), preserving a quantity of 5.0 mm of apical obturation. To remove the remaining filling material, an ultrasonic tip (E4D; Helse Ultrasonic, Santa Rosa de Viterbo, SP, Brazil) was used, activated by the device Ultrawave XS (Ultradent Products Inc, South Jordan, United States of America) at a frequency of 30 kHz and power of 25%, on the canal walls. This procedure was performed with the aid of a surgical microscope (MC-M1232, DF Vasconcellos, Valença, RJ, Brazil), with a 13 × magnification. Afterwards, all teeth were radiographed again in order to verify cleanliness and to certify the ideal intraradicular operation. Then, each root canal was prepared with the WhitePost DC1 drill (FGM Products Odontologist, Joinville, SC, Brazil) to adapt it to the glass fiber posts dimensions and taper [6,25].

Prior to cementation, the WhitePost DC 1 glass fiber posts (FGM Produtos Odontológicos, Joinville, SC, Brazil) were cleaned with 99% isopropyl alcohol and received a layer of silane (FGM Produtos Odontológicos, Joinville, SC, Brazil) for 60 s, followed by compressed air drying. Then, the roots were randomly distributed into 7 experimental groups (n = 10), according to the luting system to be used (Table 1) for the cementation of the glass fiber post. Randomization was performed using a simple draw method. For this, each specimen was sequentially numbered from 1 to 70 and then randomly assigned, one by one, to the 7 experimental groups (n = 10) in a rotating manner. This intercalated allocation process avoided block assignment and ensured balanced and unbiased distribution across groups. The radicular dentin of the teeth was treated according to the manufacturers’ directions of each luting material (Table 2).

Table 1: Material/manufacturer, code, cure mode, classification, adhesive systems and basic compositions of adhesive and cement materials.

Table 2: Luting procedures performed according to each proposed material.

After cementation of the posts, the entire sample (N = 70) was subjected to thermocycling: 10.000 cycles with immersion baths in distilled water at temperatures of 5–55ºC ± 2.0ºC; lasting 30 s for each temperature, with a 5.0 s delay between immersions in the different tanks. Subsequently, the roots were embedded in polystyrene-based resin (Maxi Rubber, Diadema, SP, Brazil) in 15 x 15 mm PVC cylinders, positioned vertically with the help of a dental surveyor (BioArt B2, São Carlos, SP, Brazil). The roots were sectioned in the apical, middle, and cervical thirds with thicknesses of 2.0 mm, respectively, from 1.0, 4.0, and 7.0 mm from the cervical to the apical region. The sectioned specimens were then polished using 1200 and 1500 grit silicon carbide paper (Norton, São Paulo, SP, Brazil) under running water to remove surface scratches and irregularities. Subsequently, the samples were cleaned with a soft brush and air jets to eliminate any debris prior to testing [26–28].

The push-out mechanical test was performed for each of the sections obtained using an electromechanical testing machine (EMIC DL 2000, São José dos Pinhais, PR, Brazil), with a 2.0 kN load cell, operating at constant speed of 0.5 mm/min. According to the root thirds, different metallic cylinders were used for complete extrusion of the glass fiber posts. The diameters of the cylinders were 1.3, 0.9 and 0.5 mm for the cervical, middle and apical thirds, respectively [26,28,29]. In all sections, the apical face was positioned to come into contact with the metallic cylinder of the push-out device, in such a way that the applied force was always applied from apical to cervical, providing adequate extrusion of the posts.

The maximum bond strength was measured by displacing the glass fiber post from the root canal. The maximum force to displace the glass fiber posts was obtained in Newton (N) and converted into megapascal (MPa) [26,28]. Before performing out the bond strength test, the root diameters of the top and bottom surfaces of all the specimens were measured under 20 × magnification using a stereomicroscope (Leica Microsystems-M80, Wentzler, Germany). Then, the adhesive area (mm^2^) was calculated for conversion of N values into MPa. The formula used was: π ⋅ (R+r) ⋅ [h^2^+(R-r)^2^]^0.5^, where “R” and “r” represent, respectively, the radius of the top and bottom portions and “h” represents the thickness of each slice in millimeters. The values obtained in N for each mechanical test were divided by the adhesive area values of each specimen.

The bond strength data (MPa) were submitted to normality (Shapiro-Wilk; p ≥ 0.05) and homogeneity of variance (Levene; p ≥ 0.05) tests. The bond strength produced in the cervical, middle and apical thirds (n = 10) was statistically analyzed for each material separately, in order to compare the intra-group results in the different root thirds. In addition, separate analyzes of the root thirds (cervical, middle or apical third) were performed in order to compare the materials with each other. For this, one-way analyses of variance were used (α = 0.05), considering Material, with the results of the different root thirds or Root third, with the results of the different materials for each region as isolated factors. Tukey’s HSD post-hoc tests were used (α = 0.05) when a significant effect was verified. For the analyses, the SPSS version 24.0 program (Statistical Package for Statistical Science Inc., Chicago, IL, United States of America) was used.

The failure pattern was analyzed by a single calibrated operator using 40 × magnification in a stereomicroscope (Leica Microsystems-M80, Wentzler, Germany). The failure patterns were organized and tabulated, according to the materials and root thirds. Percentage calculations were then carried out to determine the incidence of each type of failure for each experimental condition. Scanning electron microscopy (SEM) analysis was then carried out using a JSM 5600LV microscope (JEOL Ltd, Rio de Janeiro, RJ, Brazil), operating at 15 kV, with 40 × magnification, to obtain representative images of each failure mode. The failure modes were classified according to Leandrin et al. [4] and Ramos et al. [30] as type 1 (adhesive failure between luting material and dentin), type 2 (adhesive failure between luting material and glass fiber posts), type 3 failure (cement cohesive) and type 4 (mixed failure, when adhesive and cohesive failures occur concomitantly). However, in this study, cohesive failures of the glass fiber posts were also observed, being classified as type 5 [4,23,30].

Results

One-way ANOVAs revealed statistically significant differences (p < 0.05) in bond strength among the materials across each root third (Table 3; horizontal comparisons). When the materials were evaluated separately, statistically significant differences (p < 0.05) were found only among the root thirds of PV5, ACC, RXU, and RXU200 (Table 3; vertical comparisons).

Table 3: Means and standard deviations (±) of bond strength values (MPa) for the different materials within each root third and for each material among the root thirds, along with the results of Tukey’s HSD post-hoc test.

Comparing the materials, PV5 and RXU produced the highest bond strength in the cervical root third (p < 0.05). For the middle and apical thirds, the findings were more heterogeneous among the materials.

When the root thirds were compared, all materials showed a similar pattern, with significantly higher bond strength in the cervical third compared with the apical third (p < 0.05), and statistically similar values between the middle and apical thirds (p ≥ 0.05), except for RXU, which produced lower results in the apical third compared with the middle third (p < 0.05). No significant differences (p ≥ 0.05) were observed for PV5, RXU, and RXU200 between the cervical and middles thirds.

After the push-out mechanical strength test, a qualitative analysis of the specimens was carried out. The data were tabulated and analyzed by percentages, and the results were presented graphically (Fig 1) to show the frequency of failure patterns for each experimental condition and root third. Scanning electron microscopy (SEM) analysis provided representative images of each failure pattern, as shown in Fig 2. Images C and D show the apical and cervical thirds with cement remnants, respectively, indicating the presence of cohesive failure of the cement.

Incidence of failure for each experimental condition within each root third (A-C), and overall failure by root third (D).

Representative SEM images of the types of failures.A: type 1; B: type 2; C and D: type 3; E: type 4; and F: type 5. All images were taken using a scanning microscope at 40 × magnification.

Discussion

The bond strength of glass fiber posts to radicular dentin in different thirds by means of using the luting materials PV5, RDC, LCZ, ACC, RXU, RXU200 and FP, after hydrothermal aging, varied significantly, rejecting the inferred null hypothesis.

In order to reduce the bias of the samples, the intraradicular preparation for the glass fiber posts was standardized, so that they were juxtaposed to the walls of the root canal. The adequate fit of the posts ensures better mechanical imbrication and smaller cementation thickness, reducing the incidence of voids and cracks, as well as reduces the stress of polymerization shrinkage, which predisposes to bonding failures [1,7,31]. Grandini et al. [13] report that the greater the thickness of the luting agent, the greater the number of voids and, consequently, the greater the fragility of the adhesion; aspect corroborated by several authors [16,19,29]. The analysis of the bond strength through the mechanical push-out test favors the analysis by root thirds in a more simplified and efficient way. This mechanical test generates transversal force with homogeneous distribution causing parallel failures in the adhesive interface (dentin-cement-post), with a lower incidence of premature failures [8,19].

The absolute numerical difference in the values of intraradicular bond strength between the root thirds was notorious for most of the luting materials evaluated, which generally behaved in a decreasing manner, from cervical to apical. This corroborates other studies [15,16] that studied intraradicular cementation and adhesion, attributing the different results between the cervical and apical thirds to factors such as luting agent selection, sensitivity of the adhesive technique, dentinal morphological difference, anatomical aspects, difficulty to access the apical third during the adhesive procedures such as cleaning, bonding agent application and light curing. The apical third has a lower density of dentinal tubules and a smaller amount of intertubular dentin, which reduces bond strength to adhesive and luting systems [19,32]. The technical sensitivity of intraradicular dentin hybridization directly influences the bonding efficiency. Among the various adhesive systems, the total-etch has superior technical sensitivity, both in terms of acid etching time, as well as washing quality and substrate moisture control [33]. Despite this, from the statistical observations, little variation was observed between the middle and apical thirds, with the exception of the RXU adhesive resin material. Still, there were no statistically significant differences, regardless of the root third, between the FP resin modified glass ionomer cement and RDC and LCZ post-and-core materials, contrary to the findings of other studies [15,16]. These results are likely related to methodological differences, particularly the standardized intraradicular preparation protocol adopted in this study, which aimed to ensure the closest possible adaptation between the fiber post and the root canal walls. Proper juxtaposition of the post promotes better mechanical interlocking and a thinner cement line, reducing void formation, adhesive failures, and polymerization shrinkage stress, all of which directly impact bonding performance [16,18,19].

Luting materials can be classified according to their chemical composition, curing mode (chemical, dual or light curing), interaction with the dental substrate (chemical, mechanical or micromechanical) and application mechanism/interaction with various substrates (adhesive, self-adhesive). Materials for cementing glass fiber posts should preferably be dual-curing or chemically curing, especially due to the difficulty of light penetration in the apical region of the root [10]. Resin-modified glass ionomer cements have good adhesion to the dentin substrate because they have an intrinsic chemical bond that interacts with dentin calcium, as well as a mechanical bond and adhesion due to the resin monomers incorporated into their composition. Resin cements may have functionalized monomers in their composition to promote chemical bonding to the dentin substrate and to the glass fiber posts [10,34]. Thus, intraradicular bond strength results can be higher when compared to total-etch adhesive systems and conventional resin cements. When comparing the luting agents in relation to the evaluated root thirds, we observed that the PV5 adhesive resin cement, whose application system consists of the use of a self-etching adhesive system with dual polymerization, produced outstanding performance regardless of the evaluated root thirds.

RXU and PV5 exhibited high bond strength values, particularly in the cervical and middle thirds. This performance can be attributed to the presence of the functional monomer 10-MDP, which is contained in the PV5 cement and in the adhesive system used with RXU. Chemically, 10-MDP establishes stable ionic bonds with calcium in hydroxyapatite, forming durable MDP-Ca salts [9,35]. Mechanically, demineralized dentin allows monomer infiltration, promoting hybrid layer formation and resin tag development, which together produce micromechanical interlocking and strengthen the adhesion interface [9,35]. In addition, effective copolymerization between the resin cement and the adhesive system may have contributed to favorable mechanical behavior across the root thirds. RXU demonstrated superior values in the cervical and middle thirds, locations where clinical control of dentin moisture and adhesive application is more predictable [12,16,19]. However, the increased technique sensitivity and anatomical constraints of the apical third may limit adhesive performance, which is consistent with previous findings [18,20]. PV5 demonstrated high bond strength along the entire canal, including the apical third. This result may be associated with its self-etch adhesive system and dual-curing mechanism, which reduce technique sensitivity by eliminating prior phosphoric acid etching. In addition, the incorporation of 10-MDP in its formulation likely contributed to stable bonding throughout the root length [8,9,35,36]. The Ambar Universal adhesive system of the ACC material, despite having 10-MDP in its composition, is light-curing; a fact that may have directly influenced the results of this material, especially in the middle and apical thirds, even though it was used in the total-etch technique, due to the difficulty for the adequate photopolymerization of the bonding agent in the apical area. It was not the scope of this study to compare the viscosity and filler content of different luting agents, but the post-and-core materials have a higher filler concentration, providing greater viscosity [7]. ACC is a resin material that is included in this classification and, according to the study of Barreto et al. [2], also has higher viscosity, due to its higher filler concentration in comparison to the self-adhesive resin cement RXU200. Thus, despite the greater adhesion expected to the ACC material, as it has 10-MDP as a component of its Ambar Universal adhesive system, it produced similar bond strength to the self-adhesive resin cement RXU200. In the current investigation, etching with phosphoric acid was used prior to the application of Ambar Universal adhesive system and cementation with ACC material, unlike the study by Barreto et al. [2], which used this bonding agent in the self-etching protocol. However, there was similarity between the results of these previously cited authors and the present study. It is necessary to highlight that there is a scarcity of studies with the ACC material, limiting the search for justifications despite the found results.

When comparing the bond strength results of the different root thirds for each material, RDC, LCZ and FP did not produce statistically significant differences among the root thirds. The homogeneity results among the different root thirds presented by the LCZ can be justified by the use of its self-cure activator in combination with its bonding agent, capable of maintaining the bond strength at deeper root thirds, regardless of the limited light conditions in the apical area [37].

The resin-modified glass-ionomer luting material FP did not show high numerical values of bond strength when compared to some resin cements, such as PV5, regardless of the root third, or RXU, in the cervical and middle thirds, but showed comparable results to ACC, RDC, RXU200 and LCZ in the cervical and middle thirds, and to ACC, RDC, RXU200, LCZ and RXU in the apical third. The homogeneity of the results obtained along the root thirds may be due to its composition, which favors chemical adhesion, in addition to its chemical cure, not being influenced by the adhesive technique, as well as light conditions for photopolymerization. According to Cury et al. [38], during its curing process, the resin-modified glass-ionomer cement initially consumes all the water in its composition. This initial reaction causes material shrinkage; however, as curing progresses, a hygroscopic expansion occurs — an increase in volume due to the absorption of additional moisture from the dentinal tubules [39,40]. This expansion provides a better bond strength with dentin along the entire length of the root canal [33]; a fact that justifies the homogeneous results produced by FP [39,40].

The RDC material also showed homogeneity among the root thirds. Similar results were previously reported [17], suggesting that its self-etching adhesive system (Futurabond U) and the degree of resin cement conversion may be related to such findings. The self-etching adhesive system can interact with dentin in two ways, micromechanically and chemically. The micromechanical interaction occurs due to the polymerization of monomers that infiltrate the dentin, and this system does not completely remove the smear layer as it has a medium aggressive pH. Therefore, it causes partial dentinal demineralization, exposing a network of collagen fibers that will be infiltrated by the monomers. On the other hand, chemical adhesion occurs through ionic bonds between conventional adhesive monomers and the residual calcium of hydroxyapatite. In addition, this post-and-core system presents lower technical application sensitivity. Another extremely important factor is that this material is a dual-curing composite, which compensates the lower light penetration in the apical third and favors its degree of conversion along the entire length of the canal [17]. Considering the post-and-core materials RDC and LCZ are compared, it can be observed that both produced uniform adhesion in all root thirds. On the other hand, LCZ produced high bond strength results in the apical third, fact that might be related to the radicular dentin pre-preparation protocol recommended by its manufacturer, in which phosphoric acid etching is used.

Self-adhesive resin cements are known for their low technique sensitivity and ease of handling and clinical application. When comparing RXU200 and RXU, both manufactured by the same company, it becomes clear that their intended clinical applications differ. The RXU adhesive resin cement requires prior dentin conditioning, which facilitates hybrid layer formation and results in improved bond strength, as confirmed by the findings of this study [6,8,18]. In contrast, the RXU200 self-adhesive resin cement aims to simplify the technique by eliminating the etching step. Because it does not promote smear layer removal or deep dentin demineralization, RXU200 does not form a true hybrid layer or resin tags. Its acidic monomers interact superficially with the hard tissues, mainly through the smear layer. This interaction is less effective due to the limited demineralization capability caused by a rapid local pH increase. These characteristics likely explain the lower bond strength values observed for RXU200, especially in the apical third, where anatomical and clinical challenges such as limited access and moisture control further compromise adhesive performance [6,8,9,18,19].

The results observed regarding failure patterns showed that the highest number of failures occurred at the intraradicular cement-dentin adhesive interface. This analysis is relevant as it helps identify the region of greatest susceptibility to failure within the bonding interfaces. Factors such as the complexity of the radicular dentin substrate, with significant morphological variations from the cervical to the apical third, including lower dentinal tubule density, especially in the middle and apical thirds, hinder the formation of the hybrid layer and resin tags. In other words, they compromise the infiltration of adhesive monomers between collagen fibrils, promoting adhesive failures in this region [8,12,16,19].

Additionally, the stress generated by polymerization shrinkage of the cement may concentrate at the cement-dentin interface, especially when there is an uneven thickness of the cement line [16,19]. Although a rigorous intraradicular preparation protocol was adopted in this study to ensure a thin and standardized cement layer, the bond strength results indicate that cement distribution within the canal was not entirely uniform. This is evidenced by the differences observed among the values recorded for each root third, suggesting a likely thickening in the apical region, which consistently showed the lowest bond strength values across all groups. Greater polymerization shrinkage in these deeper regions may generate increased stress on the dentinal walls, compromising the adhesion at the cement-dentin interface and contributing to the higher incidence of adhesive failures in this third, predominantly failures at the dentin-cement interface [18,19]. The findings of Amiri et al.[19] support these observations, indicating that such conditions can significantly compromise bond strength in the apical region. In contrast, the cement-post interface showed lower susceptibility to failure, possibly due to the greater chemical compatibility between the resin-based components of the cement and the resin matrix of the glass fiber post, along with the homogeneous surface of the post, which facilitates micromechanical adhesion along its entire length [41]. These findings are in line with previous studies [18,21,42–44] that identify the dentin-cement interface as the most vulnerable point in glass fiber post adhesive cementation.

The present study has limitations such as the analysis of material viscosity and penetrability and its role in intraradicular adhesion. Additional and longitudinal studies evaluating the influence of factors such as cementation line thickness, histological and anatomical variations should be conducted to elucidate the direct correlation of these bond strength results with the success and prognosis of restorative therapy in endodontically treated teeth with extensive coronal destruction.

Conclusions

The choice of luting material system significantly influences the bond strength of glass fiber posts to intraradicular dentin, especially in challenging regions such as the apical third. In clinical scenarios requiring effective adhesion in deep and complex root canals, the use of dual-cure self-etch adhesive and resin cement systems containing functional monomers such as 10-MDP (e.g., PV5 and RXU) may provide more predictable bonding outcomes. In contrast, etch-and-rinse systems such as LCZ or light-dependent adhesive systems like that of ACC exhibited greater technique sensitivity and lower performance in difficult-to-access areas. Moreover, the resin-modified glass-ionomer cement FP demonstrated consistent performance across root thirds, suggesting it may be a viable option when simplified protocols and reduced technique sensitivity are desired. Therefore, the selection of luting materials should consider not only ease of application but also the adhesive strategy and anatomical challenges of each root third to optimize retention and restoration longevity.

Supporting information

S1 DataRaw data repository.(XLSX)