Effect of Two Different Adhesion Modes of a Universal Resin Cement on the Retention of Glass Fiber Posts Cemented to Root Canal Dentine: An In Vitro Study

Abstract

Purpose: The aim of this in vitro study was to investigate the adhesive bond strength of glass fiber posts when cemented with universal resin cement in two different adhesion modes: adhesive and self-adhesive. Methods: A total of 20 extracted single-root teeth were endodontically treated, decoronated and prepared to receive glass fiber posts (GFPs) with a diameter of 1.6 mm (RelyX fiber post 3D). Specimens were randomly divided into two groups: (G1) GFPs were cemented using RelyX Universal cement in self-adhesive mode, and (G2) GFPs were cemented using Scotch Bond Universal Plus and RelyX Universal cement (adhesive mode). Afterwards, the specimens were sliced at three root levels: coronal, middle and apical. Bond strength was measured using a push-out test. Data were analyzed with a two-way analysis of variance (ANOVA) test and independent sample T-test. Results: Bond strength was significantly influenced by the adhesive strategy (p < 0.025) and the position of the root third (p < 0.007). Microscopic analysis of failure mode revealed a higher prevalence of adhesive failures (cement–dentine). Conclusions: Glass fiber posts cemented with universal resin cement applied in adhesive mode showed significantly higher push-out bond strength than when applied in self-adhesive mode. In both study groups, the apical root regions exhibited the highest retention values, followed by the middle and coronal regions.

1. Introduction

Restoring endodontically treated teeth (ETT) can be difficult for dental clinicians. Compared to vital teeth, ETT exhibit different mechanical properties and are often more susceptible to fracture [1]. Anterior teeth and premolars often cope with a loss of ferrule [1]. To counteract this phenomenon, intra-canal posts were introduced. They enhance the retention of the restoration and minimize the stress transmitted through the tooth [2].

ETT with glass fiber posts (GFPs) have a reported failure rate from 2% to 40%. Many factors may have an influence on the complication rate and strength, such as the type of tooth and final restoration (direct vs. indirect), which explains the variation in results [3]. Post-related failures were only a minority within the overall failures of ETTs. Ferrari et al. [4] reported a 7–11% failure rate of fiber posts, which accounted only for 2.3% of the total number of failures with restored ETTs.

Not only is the adhesive bond between the GFP and the cement important but also the bond between the root dentine and the luting cement. The latter is considered the weakest link [5]. A systematic review of clinical trials by Marchionatti et al. [6] suggested that debonding of the cement from the dentine is the most common reason for failure and occurs mainly because proper adhesion at the cement–dentin interface is difficult to achieve [7].

Although bonding to root canal dentin follows a similar protocol as luting coronal restorations, the connection is usually weaker [8]. Several factors can interfere with the bonding procedure, such as poor moisture control, limited visibility and access inside the root canal, difficulties in light curing, and irregular structural features of the root canal. Also, remnants of sealer and gutta-percha from the endodontic treatment on the root canal wall may further jeopardize the bonding procedure [9].

While it may be possible to solve some of the challenges listed above, the excessive C-factor within the root canal can hardly be reduced. Consequently, the polymerization shrinkage can induce high stresses, which can destroy or weaken the bond between the root canal dentin and the luting material, thereby increasing the risk for microleakage and loss of retention [10].

In a systematic review, the bond strength between conventional adhesive resin and self-adhesive resin cement was investigated. Although the included articles showed high heterogeneity, the final analysis suggested that a self-adhesive resin cement could increase the retention of GFPs into root canals [11]. This was confirmed by more recent literature stating that self-adhesive resin cement tended to be the most effective luting agent in bonding between glass fiber posts and root canal dentin in short- and long-term aging conditions [12].

A multi-centered randomized clinical trial with 3 years of follow-up reported that both self-adhesive and adhesive resin cements can be successfully used to bond glass fiber posts [13]. However, there is little information about the efficacy of the new universal resin cements in bonding GFPs to dentin and the choice of their adhesion strategy to achieve the most favorable clinical outcome.

Therefore, the aim of this in vitro study is to investigate whether a universal resin cement in a self-adhesive approach will lead to a similar bond strength of glass fiber posts to root canal dentin as when used in a full-adhesive approach.

The first null hypothesis of this study states that the cementation approach of universal resin cement would not influence the retention of the glass fiber post. The second null hypothesis states that no significant difference in push-out bond strength would be detected in different root regions (coronal, middle and apical).

2. Materials and Methods

2.1. Specimen Preparation

Twenty human teeth were included, which were extracted for periodontal or orthodontic reasons. These single-rooted teeth (upper incisors and premolars) were caries-free and had a single root canal and mature apex. Calculus and soft tissue debris were removed by manual scaling. The teeth were carefully inspected to verify the absence of resorption, cracks and fracture lines and then stored in 0.9% physiologic saline at room temperature.

The endodontic treatment was performed by one operator, using Protaper Next (Dentsply Maillefer, Ballaigues, Switzerland), until size X2 was reached. The working length was 0.5mm shorter than the visible apex and verified by a radiograph. While performing the instrumentation, a 30-gauge side vented needle was used to perform the irrigation, using 3% sodium hypochlorite. After instrumentation and thorough irrigation, the canal was rinsed with saline and dried with paper points (Dentsply, Maillefer, Switzerland). Subsequently, the canals were obturated with gutta-percha points (Dentsply, Maillifer, Switzerland) and sealer (AH Plus, Dentsply Sirona, Bensheim, Germany), using cold lateral condensation. The access cavity was sealed with a glass ionomer filling (GC Fuji IX, Tokyo, Japan) and stored in saline (NaCl 0.9%) at 37 °C for 1 week, allowing the endodontic sealer to set.

Before preparation of the post space, the teeth were decoronated 1 mm above the cemento–enamel junction using a handpiece and a diamond disc.

Gates glidden burs (#4 and #6, Dentsply, Maillefer, Switzerland) were used to remove the root canal filling to prepare space for the post up to a depth of 10 mm. Next, the canal walls were shaped using the dedicated low-speed drill (size 2, 0.8 mm apical diameter) (3M, Seefeld, Germany), which corresponded to the size of the glass fiber post selected for this study (RelyX^ΤΜ fiber post 3D, 3M, Seefeld, Germany). This type of fiber post has a length of 20mm and an 8% taper. The coronal part is equipped with macro retention rings to improve the retention of the composite build-up. The radio-opaque glass fibers have a parallel orientation within the composite resin matrix.

A microscope (x4 magnification, OPMI pico, Carl Zeiss, Oberkochen, Germany) was used to check the canal walls for remnants of gutta-percha or sealer. Any remaining debris was removed by a manual K-file. Finally, the canals were irrigated with 3% NaOCL followed by saline solution (NaCl 0.9%) and dried with paper points. The fit of the GFPs inside the root canal was checked, whereafter the post was cleaned with 70% denatured ethanol for 30 s and air dried. The GFP was then coated with a universal bonding agent (Scotchbond Universal Plus Adhesive, 3M, Seefeld, Germany) using a micro brush.

All samples were randomly allocated to one of the two groups:

Group I: Self-adhesive mode group: First, the universal resin cement (RelyX Universal, 3M, Seefeld, Germany) was injected into the prepared post space with a thin disposable tip until the cement was oozing out of the canal. The glass fiber post was placed in the canal (RelyX^ΤΜ fiber post 3D, 3M, Seefeld, Germany) with moderate finger pressure until it reached the predetermined length. A composite spatula was used to remove excess cement, and the cement was then polymerized by light curing (standard power: 100 mW/cm²) for 40 s.

Group II: The adhesive mode group: This followed the same protocol, but first, a micro brush was used to apply the universal adhesive (Scotch bond Universal Plus Adhesive, 3M, Seefeld, Germany) inside the canal by continuously rubbing the dentine walls for 20 s and then gently air dried, before putting the cement in the canal.

To precisely section the specimen, all test samples were imbedded in clear acrylic resin (Orthocryl, Dentarium, Ispringen, Germany) using a silicone mold. A dental surveyor was used to hold the specimen in the correct position while the acrylic set, to ensure that the sectioning of each specimen would be perpendicular to the length axis of the root.

The specimens were then sectioned into three 2 mm thick slices from the coronal, middle and apical regions with a 1 mm intermediate slice in between each section. Sectioning was performed with a low-speed diamond wafering blade (IsoMet™1000; Buehler Ltd., Lake Bluff, IL, USA). This resulted in 30 specimens per group, divided over 3 subgroups: coronal (#10), middle (#10) and apical (#10) (Figure 1).

2.2. Bond Strength

A push-out test was performed to determine the bond strength between the luting cement and GFPs, by applying a compressive force through a cylindrical plunger fixed to a universal testing machine (AG-X plus 50KN, Shimadzu, Kyoto, Japan). The samples were placed on a flat metal holder with a 2 mm hole at the center. The specimens were placed with the apical side facing upwards, and downward force was applied to dislodge the post towards the wider section of the root slice. The diameter of the cylindrical plunger’s tip used during the test was 0.5 mm for the cervical and middle slices and 0.25 mm for the apical ones. The plunger’s tip was placed on the center of the post, without touching the canal walls of the tooth.

Loading was performed at a speed of 1 mm/min until failure. The peak force required for debonding was registered by the machine’s software (Trapezium) in Newton (N). The push-out strength (measured in MPa) was computed by dividing the force (N) by the area (A) of the bonded interface, i.e., bond strength (MPa) = maximum load (N)/bonded surface area (mm²). The bonded surface area was calculated through the formula: A = π (R + r) [h²+ (R − r)²]^1/2, R represents the coronal post radius, r represents the apical post radius and h is the thickness of the slice in mm (Figure 2).

To measure the post radius, a reflex camera with macrolens (Nikon D90, Nikkor 105 mm, Tokyo, Japan) was used to take high-resolution pictures of both sides of each slice, resulting in 120 pictures in total. The images were analyzed in imageJ software to measure both radii, (R) and (r). Push-out bond strength was calculated for each of the three groups (coronal, middle and apical).

To determine the failure mode of the debonded posts, new images were captured of each slice using a reflex camera with a macroscopic lens. Four failure modes were observed (Figure 3): (i) adhesive failures between dentin and cement; (ii) adhesive failures between post and cement; (iii) mixed failures; and (iv) cohesive failures inside the post.

2.3. Statistical Analysis

Sample size calculation was based on the study by Oskoee et al. [14], where the self-adhesive cement showed a mean bond strength of 7.95 MPa (SD 3.31) and the use of the self-adhesive cement with a universal bonding showed a mean bond strength of 13.45 MPa (SD 4.7). For 95% power, a sample size of 9 was adequate.

The two-way analysis of variance (ANOVA) test was used to determine the effects of the cementation approach and the position of the slice within the root, as well as the interaction between these two factors on the bond strength. An independent sample T-test was used to compare the two groups. Fisher–Freeman–Halton exact test was used to compare the differences in fractures between the groups. The significance level was set at 0.05 (p < 0.05). Data were analyzed with SPSS 28 (SPSS Inc., Chicago, IL, USA).

3. Results

In this study, the push-out bond strength of two cementation approaches is compared between two groups containing 10 teeth each. The surface area was highest in the coronal sections and lowest in the apical sections (

Table 1. Mean values and SD of surface area in the coronal, middle and apical sections regardless of the cementation method used.

	Surface Area [mm2]
Coronal	9.55 ± 0.81
Middle	7.22 ± 0.57
Apical	5.26 ± 1.14

). The bond strength and SD were measured for both groups in each of the three sections and as an average of the three sections, as shown in

Table 2. Push-out bond strength following the application of two different cementation approaches. Data are shown in each section.

	GP1—Self Adhesive		GP2—Adhesive		p-Value
	Mean Bond Strength [Mpa]	SD [Mpa]	Mean Bond Strength [Mpa]	SD [Mpa]
Coronal	13.04	2.62	17.65	4.68	0.014
Middle	16.29	5.35	20.07	6.79	0.184
Apical	20.47	7.58	23.78	9.77	0.409

. The bond strength was lowest in the coronal sections and highest in the apical sections, regardless of the cementation approach used (Table 2).

The two-way ANOVA results show that both “the method of cementation” and “the position of the root third” had significant effects on bond strength (p = 0.025 and p = 0.007, respectively). The interaction between these two factors was not significant (p = 0.95). The push-out bond strength was numerically higher at all root levels when adhesive mode was applied, compared with self-adhesive mode. However, the mean comparison using independent sample T-test at each section separately shows that the adhesive mode significantly increased bond strength only in the coronal section (p = 0.014). The mean difference was not statistically significant at the middle and apical sections (p = 0.184 and p = 0.409, respectively). It is worth mentioning that bond strength was the lowest in the coronal sections and highest in the apical sections in both groups.

The dispersions of the different failure modes are reported in

Table 3. Failure mode analyses expressed in percentage.

Failure Mode [%]
	Adhesive Dentine/Cement	Adhesive Post/Cement	Mixed	Cohesive
Self-adhesive	90	3.3	6.6	0
Adhesive	73.3	3.3	6.6	10

. In both groups, the most recurrent type of failure was the adhesive between dentin and cement. Mixed failures were the second most frequent group. Fisher–Freeman–Halton exact test demonstrated no statistically significant difference between the two groups (p = 0.248).

4. Discussion

The purpose of this study was to compare the bond strength of glass fiber posts luted to root canal dentin when using a universal resin cement in two different modes: adhesive and self-adhesive. The push-out test showed that the bond strength of the universal cement was significantly higher when used in the adhesive mode in combination with the universal bonding agent (group 2) than when used in the self-adhesive mode (group 1). Therefore, the first null hypothesis was rejected. This confirms the findings by Oskoee et al. [14] and Kosan et al. [15], who reported that the use of universal adhesive in self-etch promoted the bond strength to dentin. The mild etching of the adhesive created a thin zone of demineralized dentine and made the penetration of resin tags in the dentinal tubules possible.

The bond strength differed significantly depending on the section within the root. For both approaches, the highest values were found at the apical third, followed by the middle and coronal sections. Therefore, the second null hypothesis was rejected. These results are consistent with another study found in the literature, where the highest push-out bond strength was found in the apical part of the root [16].

According to Gwinnett et al. [17], resin tags are responsible for about 30% of the total bond strength. Therefore, a lower bond strength would be expected at the apical third of the root, due to the lower density of dentinal tubules. However, this was not observed in the current study. In general, universal cements rely on the chemical bond with dental hard tissue through the functional monomers which form insoluble calcium salts with hydroxyapatite resulting in a stable bond [18,19]. This indicates that the presence of solid dentine may positively influence the adhesion strategy of these universal cements and that dentin depth and tubular density have less impact on universal cements [14].

Initially, the high hydrophilicity and the low pH of the self-adhesive cements are imperative to obtain ample wetting of the dentine, but can also be responsible for an increased endurance to moisture, which can be an explanation of the better bonding strengths of the self-adhesive cements to the apical root region [18].

Resin cements employed in this paper exhibited chemical and physical polymerization, which corresponds directly to the intensity of the light source. The high bond strength for the apical samples can indicate an effective polymerization of the cement, regardless of the space from the light source. As stated by the manufacturer, the glass fibers incorporated in the post system can conduct light through the post. This could be an explanation for the effective polymerization of the universal cement in the apical third of the canal. However, this finding cannot be verified since the specimens were not shielded from light sources during sectioning or while performing the push-out test. According to the literature, the efficacy of glass fibers to transmit light is still debatable [16]. Some studies also suggest that the apical part of the root is more acidic; this can possibly elevate the availability of calcium for chemical adhesion and compensate for the effect of the light curing distance [18,20].

Another determining factor is the polymerization stress. Machry et al. [21] reported that, when the cement layer is thinner, there is less chance of incorporating microporosities and polymerization shrinkage. In the apical portion of the root, there is less room for the cement; thus, a thinner cement layer can be achieved regardless of the protocol, as in the coronal part, a higher volume of cement is present, causing increased polymerization shrinkage and stress at the adhesive interface. This increase in polymerization shrinkage will jeopardize the adhesion and can cause the reduced push-out bone strength, especially because of the high C-factor within the root canal [16].

Since the light curing was performed from the coronal part of the post, it is safe to assume that the cement has been dual cured in the coronal part and mostly self-cured in the apical part. The shrinkage strain depends on the curing mode used in dual cure cements. In a study by Kitzmüller et al. [22], the shrinkage strain doubled when light polymerization was also involved, in comparison to only self-curing.

Also, the formation of locking areas, where closer contact between the post and the canal walls exists, can occur more easily at the apical part of the root. This might also attribute to the higher retentive values, as the retention of the post in this region occurs by mechanical overlapping and tenso-friction with the dentine walls, in addition to the adhesive bonding [23]. Therefore, preserving the canal shape and minimal widening during preparation is recommended to ensure excellent post adaptation to the canal walls.

A recent review showed that in short-term aging conditions, self-adhesive and self- etch tend to be the most effective luting agents for bonding GFPs, with self-adhesive resin cement being the most effective in long-term aging conditions. In this study, no aging was applied, confirming the findings of the review where the self-etch adhesive tends to perform equally with the self-adhesive. This study only showed a significant difference in the coronal region between the two adhesive strategies [12].

Although the literature suggests that the self-adhesive resin cement is the most effective, Andrews et al. concluded that different relyX cements showed better shear bond strength with the use of a bonding agent [24].

The failure mode analysis revealed that the adhesive failure between the cement and the root canal was the most common fracture mode. This is consistent with previous studies [24,25,26] and demonstrates that, even with the recent advances in adhesive materials, the adhesion between the dentin and resin cement is still the weakest link in the dentin–adhesive–post connection.

The results of this research should be considered with reservation. Laboratory studies have intrinsic limitations when trying to simulate in vivo conditions. For example, the test specimens were not restored coronally and no aging (thermal or mechanical) was applied. Thermocycling decreases the flexural strength of fiber posts, while mechanical fatigue enhances microleakage in all types of post. These factors may limit the direct transferal of the study results to the clinical situation. The more oval shape of the premolar’s root canal might have reduced the adaptation of the post, resulting in a wider cement spacer and lower push-out values.

5. Conclusions

Within the limitations of the current study, it was concluded that the push-out bond strength of the cemented GFP is higher when the universal resin cement is applied in adhesive mode with a separate universal bonding agent than in self-adhesive mode, although this is only significant in the coronal part. The highest bond strengths were achieved in the apical third of the root.