Can Silanization Increase the Retention of Glass-fiber posts? A Systematic Review and Meta-analysis of In Vitro Studies
The role of silanes in the bonding of resin luting agents to glass-fiber posts (GFPs) is a controversial topic, and the question still remains whether post silanization is able to improve the retention of GFPs luted into root canals. Thus, this study was designed to determine whether evidence exists to justify silanization of GFPs before cementation to increase their retention into root canals. In vitro studies that evaluated the retention of GFPs cemented into root canals or artificial substrates and that used silane coupling agents for pretreatment of the post were selected. Searches were carried out in PubMed and Scopus databases with no publication year or language limits. The last search was carried out in August 2014. Two distinct data analyses were carried out: 1) control group (no post pretreatment) vs silane only and 2) post pretreatment + silane vs silane only. Pooled-effect estimates were obtained by comparing the difference between each bond strength mean value and were expressed as the weighted mean difference between groups (p≤0.05). A total of 178 articles were found, and 23 were included in the review. The results were affected by the substrate into which the GFPs were luted (teeth or artificial devices). The analysis between control group and silane only for studies that used artificial devices favored the use of silane (p<0.0001), but considering studies that used teeth as substrate, no significant difference was observed (p=0.35). The analysis between silane only and pretreatment + silane did not show a significant difference between groups when artificial devices were used (p=0.71), whereas the analysis favored the use of post pretreatment + silane over silane (p<0.00001) only when the GFPs were luted into teeth. In conclusion, this review indicates that silanization improves the retention of GFPs luted into root canals provided that selective surface pretreatments are applied to the post before silanization.SUMMARY
INTRODUCTION
Glass-fiber posts (GFPs) have been developed to improve the optical effects of esthetic restorations1,2 and are widely used for restoring endodontically treated teeth with insufficient coronal structure to serve as a core for the restoration.3,4 The use of GFPs in cases in which the coronal tooth structure has been destroyed as a result of caries, trauma, or overaggressive endodontic procedures is gaining widespread acceptance among dental clinicians.5,6 Together with the increased use of prefabricated posts, particularly GFPs, an increase has also been observed in the number of studies on this subject available in the literature. These studies evaluate different cementation protocols, adhesive systems, and surface treatments for improving the bond between resin cements and GFPs. Yet the main reason for failure of GFPs is still debonding, which occurs mainly as a result of the difficulties clinicians face in achieving proper adhesion to the intraradicular dentin.7
Various surface pretreatments of GFPs have been tested in the literature. These pretreatments can be divided into 1) physical/chemical means intended to create surface irregularities and expose the inorganic glass fibers and 2) chemical treatments applied to improve micromechanical and/or chemical attachment to the post.8-12 Silanization is the most frequently used chemical pretreatment. Organo-silane coupling agents are bifunctional molecules in which one end of the molecule is capable of reacting with inorganic glass fiber and the other end with the resin cement.13 The role of silanes in the bonding of resin luting agents to GFPs is, however, a controversial topic.2 Some studies2,12,14 reported that silanization does not have a significant effect on the bond strength between resin cements and GFPs, whereas other studies15-17 reported improved bonding by silanization. It is also a possibility that increased exposure of the glass fibers to physical/chemical pretreatments could have a synergic effect with silanization, thereby improving the retention of GFPs.
Despite the large number of in vitro studies in the literature, the question still remains whether post silanization is able to improve the retention of GFPs luted into root canals. This question cannot be easily answered because of the large variability in methods and results among primary studies. Therefore, the aim of this study was to systematically review the literature to determine whether there is in vitro evidence to justify the use of silanes to improve the bond strength of GFPs to intraradicular dentin. The hypothesis tested was that application of silane does not improve the retention of GFPs.
METHODS
Search Strategy
This systematic review was based on the guidelines of the Cochrane Handbook for Systematic Reviews of Interventions18 and followed the four-phase flow diagram based on the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) Statement.19 Two electronic databases (PubMed and Scopus) were searched to identify manuscripts that could meet the following inclusion criteria: in vitro studies that evaluated the retention (bond strength) of GFPs luted into root canals (human or bovine teeth) or into artificial devices that used silane coupling agents for pretreatment of the post. The following search strategies were used: (glass fib* post*) AND (silane*); (endodontically-treated teeth) AND (silane*).
Screening and Selection
No publication year or language limits were set. The last search was carried out in August 2014. Reference lists of included studies were hand searched for additional articles. Excluded from the study were investigations reporting in situ studies, literature reviews, types of posts other than GFPs, and studies that did not use silane coupling agents for post pretreatment. Two reviewers (APM and RSO) independently screened the titles identified in the searches. If the title indicated possible inclusion, the abstract was evaluated. After the abstracts were carefully appraised, the manuscripts considered eligible for the review and those with which there was some doubt were selected for full-text reading. In case of disagreement, a third reviewer (TPC) decided if the article should be included or not.
Data Collection
The two reviewers extracted all data simultaneously using a standardized outline. To make identification of variables found in the articles easier, the authors categorized similar information into groups (eg, post pretreatment used, bond strength mean reported in the articles). In case of measurement of bond strength values for different root thirds (push-out test, for instance), the arithmetic average of the values of the thirds was used. For studies that did not report bond strength means in tables, the authors were contacted via e-mail if data were missing or when more information was needed.
Statistical Analysis
Two distinct data analyses were carried out: 1) control group (untreated posts) vs silane only and 2) post pretreatment + silane vs silane only. Every possible comparison of bond strength between groups within the articles was simulated. Pooled-effect estimates were obtained by comparing the difference between each bond set of strength mean values and were expressed as the weighted mean difference between groups. A p-value < 0.05 was considered statistically significant (Z-test).
Statistical heterogeneity of the treatment effect among studies was assessed using the Cochran Q test, with a threshold p-value of 0.1, and the inconsistency I2 test, in which values greater than 50% were considered indicative of high heterogeneity.18 The analyses were carried out using a random-effects model. Taking into account that the analyses of substrate used in the test could present high heterogeneity, subgroup analyses considering artificial devices or teeth as distinct substrates were carried out to explore that influence on the results. All analyses were conducted using Review Manager Software, version 5.1 (The Nordic Cochrane Centre, The Cochrane Collaboration, Copenhagen, Denmark).
Assessment of Risk of Bias
Risk of bias of each included study was evaluated according to the description in the articles of the following parameters for the study quality assessment:20 randomization of teeth, use of teeth free of caries or restoration, materials used according to the manufacturers' instructions, use of teeth with similar dimensions, endodontic treatment performed by the same operator, description of sample size calculation, and blinding of the operator of the testing machine. If the authors reported the parameter, the article had a “Y” (yes) on that specific parameter; if it was not possible to find the information, the article received an “N” (no). Articles that reported one to three items were classified as having high risk of bias, those that reported four or five items were classified as having medium risk of bias, and those reporting six or seven items were classified as having low risk of bias. Only articles that used teeth as substrate for luting the GFPs had the risk of bias classified; the other studies had other parameters evaluated except those related to the teeth.
RESULTS
Figure 1 shows the flow diagram of the systematic review. A total of 178 articles were found, and 26 were eligible for full-text analysis. The hand searches revealed six more articles for full-text reading. From the 32 studies, 23 articles were included in the review.2,6,9,16,21-39 Nine studies were excluded for the following reasons: two did not test the bond strength;40,41 two were literature reviews;42,43 one study did not test the use of silane;44 three studies used quartz-fiber posts;45-47 and one did not present the group silane.48 In the included studies, the main outcomes evaluated were type of pretreatment, substrate used for luting the GFPs, bond strength test, and resin cement (Tables 1 through 3).



Citation: Operative Dentistry 40, 6; 10.2341/14-330-O



Results of the meta-analyses are presented in Figures 2 and 3. The analysis between control group (untreated posts) and silane only for studies that used artificial devices (Figure 2) favored the use of silane (p<0.0001), with I2 = 94%. Considering studies that used teeth as substrate, no significant difference was observed between groups (p=0.35; I2=87%). The analysis between silane only vs pretreatment + silane (Figure 3) did not show a significant difference between groups when artificial devices were used (p=0.71; I2=81%), whereas the analysis favored the use of post pretreatment + silane (p<0.00001; I2=94%) over silane only when the GFPs were luted into teeth. The articles by Bitter and others21,22 were not included in the analyses because the data necessary for analysis were not obtained after an attempt at e-mail contact with the authors.



Citation: Operative Dentistry 40, 6; 10.2341/14-330-O



Citation: Operative Dentistry 40, 6; 10.2341/14-330-O
Table 3 shows that a wide variety of resin cements were used in the selected studies, with varied results reported. One study27 reported that post silanization improved the adhesion of GFPs luted with self-adhesive resin cements, and 12 studies6,22,23,26,28-32,34-36 showed a positive effect of silane on the bond strength of posts luted with regular resin cements. Yet other studies showed no significant improvement in the retention of GFPs by silanization using self-adhesive,24 regular resin cements,2,9,21 or both.39 It was not possible to observe any interaction among resin cements, post silanization, or other post treatments.
Risk of Bias
The nine articles2,27,30,31,33-35,37,39 that used teeth as substrate had the risk of bias classified as high. From the studies that used artificial devices as substrate, 11 articles6,9,21-24,26,28,32,36,38 reported that the materials were used according to manufacturers' instructions, and none reported sample size calculation or whether blinding of the operator of the testing machine was used (Table 4).

DISCUSSION
This systematic review is the first to summarize the in vitro data on the influence of silanization on the retention of GFPs into root canals. Several materials, surface treatments, and cementation strategies have been tested in the literature in an endeavor to increase the retention of GFPs into root canals. Analysis of all available data together could clarify the role of silane with regard to the performance of luted GFPs and give support for the clinician in terms of evidence-based decision making. The hypothesis tested that application of silane does not improve the retention of GFPs was rejected.
Several surface pretreatments for posts have been tested to improve the bonding between GFPs and resin cements.10-12,26,41,49-51 Pretreatment procedures aim to generally improve the adhesion to GFPs by facilitating chemical and/or mechanical interaction between the different substrates at the bonded interface. The results of the present study indicate that silanization improves the retention of GFPs only when appropriate surface pretreatment of the post is performed before application of silane. This finding is explained by the fact that the glass fibers in untreated posts are covered by the highly cross-linked, low-reactive epoxy resin. Application of surface pretreatments might expose the glass fibers, allowing more effective formation of siloxane bonds between silane and glass. The rough surface left by the surface pretreatments may also aid in improving micromechanical retention at the post-resin cement interface.46,52
Previous studies16,17 have clearly indicated the positive effect that silanization might have on the bond strength between GFPs and methacrylate-based materials. However, the question that remained unanswered was whether post silanization would have a role in improving its retention into root canals. In this study, investigations that did not lute the GFPs into dental root canals or artificial root canals were excluded, since the retention analysis was the main focus here. It was noted that silanization alone is not sufficient to improve the retention of GFPs luted into root canals, whereas the combination of surface pretreatment + silanization was able to improve the retention into root canals.
Post debonding is the main reason for clinical failure of GFP-retained restorations.7 This clinical failure type might result from poor interaction between resin cement and intraradicular dentin and/or poor interaction of resin cement and post. The findings of the present study indicate that when the posts were luted into natural root canals, the combination of post pretreatment + silanization significantly improved the post retention. This result is explained by a better interaction between resin cement and post surface leading to a situation in which the mechanical stresses during testing concentrate at the interface between the resin cement and root dentin only. In such a scenario, the better mechanical keying at the post-cement interface does not contribute significantly toward stress concentration and/or magnification during the test, leading to higher bond strength values.
In contrast to the findings from studies performed using extracted teeth, no significant improvement in the retention of GFPs was observed for the combination of post pretreatment + silane when the posts were luted into artificial devices. When artificial devices are used, there is no dentin-resin cement interface; in other words, the resin cements used to lute the posts do not interact with dental hard tissues but rather with synthetic materials such as methacrylate-based composites. In such a scenario, the interaction of the cement with the artificial devices is expected to be improved as compared with that associated with dentin, which is acknowledged to be the weakest link in adhesive bonding. In addition, the use of artificial devices usually does not have the same limitations that are imposed upon extracted teeth, such as great variability in root canal diameter and resin cement film thickness between specimens. Therefore, it is suggested that the use of artificial devices to lute GFPs should be restricted to situations in which the post-cement interface is the main focus of the investigation.
Among the surface pretreatments tested in the included studies, sandblasting stands out as the pretreatment most often used. A total of 80% of comparisons carried out here on the effect of surface pretreatments on the retention of GFPs into artificial devices, and ~62% of the comparisons on the retention of GFPs into root canals, used sandblasting as pretreatment. As an overall result, the present findings indicate a positive effect of surface pretreatments before silanization; however, this result should be mainly concentrated at the combination of sandblasting + silanization on the retention of GFPs, because most studies only tested that specific combination. That notwithstanding, surface pretreatments that only selectively expose the glass fiber by chemical means could be considered the ideal situation to enhance the silanization effect. Sandblasting is known not to be selective in exposing the glass and may cause structure damage to the post, although there is no evidence regarding whether this could affect the mechanical stability of post-and-core restored teeth.
Different mechanical tests to measure the bond strength and a wide variety of adhesives and resin cements are reported in the in vitro literature, resulting in a tough scenario for one seeking comparisons between the results of different studies. Authors sometimes do not follow the manufacturers' directions in applying materials, underscoring the problem of comparing studies in the literature. Systematic reviews have the advantage of analyzing the literature data together, but they also suffer from the limitation that the methods employed in distinct studies differ to extents that often are difficult to predict. With that in mind, we have used a tool to assess the risk of bias of each study.
The results indicate that all selected studies present a high risk of bias, demonstrating that variables that could influence the results of the studies were not controlled by researchers, favoring the high heterogeneity of the findings in the present study. However, the risk of bias assessment can be subjective and should be interpreted as such. Heterogeneity among the studies was in fact expected, since it is known that laboratory analyses have intrinsic variability related to experimental setups, procedures for specimen preparation, and the mechanical tests themselves.
The results of the present review should be interpreted with caution considering that laboratory studies have intrinsic limitations in terms of simulating in vivo conditions. However, the improved retention of GFPs by a combination of post pretreatment and silanization is of particular interest, bearing in mind that it could affect the clinical survival of GFP-retained restorations. Additionally, clinicians should be aware of the beneficial effects that post silanization might have on the clinical performance of restoration, particularly because post silanization is a procedure that might be overlooked in the clinical practice if it is regarded as being of minor significance. Furthermore, it is important to know if the posts are commercially available in a pre-silanized or pretreatment form by the manufacturer. For this reason, following the manufacturers' recommendations when preparing the GFPs before luting is necessary. Regardless of the results presented here, well-designed randomized controlled clinical trials (RCTs) with long follow-up periods would provide the ultimate answer as to whether use of a silane coupling agent could result in improved clinical success rates for GFP-retained restorations. However, it is known that RCTs cannot be used indiscriminately to support all clinical decisions. Therefore, the overall results of the present study favor the combination of post surface pretreatment and silanization for the retention of GFPs.
CONCLUSIONS
Analysis of the in vitro literature indicates that silanization improves the retention of GFPs luted into root canals provided that selective surface pretreatments are applied to the post before silanization.

Flow diagram of the systematic review according to the PRISMA Statement.

Forest plot for the analysis between control group (untreated posts) and silane only. Studies that used artificial devices favored the use of silane, whereas studies that used teeth as a substrate for luting the posts reflected no significant difference.

Forest plot for the analysis between silane only vs pretreatment + silane. No significant difference between groups was observed when artificial devices were used, whereas the analysis favored the use of post pretreatment + silane over silane only when the posts were luted into teeth.
Contributor Notes
Rafael Ratto de Moraes, DDS, MSc, PhD, Federal University of Pelotas, Graduate Program in Dentistry, Pelotas, Brazil