Bond strength of self- adhesive flowable resin composites to tooth structure: a systematic review

Aim: To review the current literature regarding the bond strength of self-adhesive flowable resin composites (SAFRCs) to tooth structure, comparing the outcomes with conventional flowable resin composites (CFRCs). Methods: PubMed/ Medline, EbscoHost and Scopus databases were screened (last update on November 2020) using related Medical Subject Headings (MeSH) and free terms. We included in vitro studies published in English language assessing the bond strength of SAFRCs and CFRCs to enamel and/or dentin from primary and/or permanent teeth. Results: In total, 23 articles were included. Unlike CFRCs, SAFRCs such as Vertise® Flow and FusioTM Liquid Dentin exhibited statistically lower bond strength to enamel and dentin from permanent teeth. There were limited studies comparing the enamel bond strength of CFRCs and SAFRCs (prior phosphoric acid etching and/or adhesive system use). Also, we found few studies that evaluated the bonding effectiveness of Constic® and other SAFRCs to primary teeth. Conclusions: Current SAFRCs showed low bond strength to permanent teeth, which impedes to recommend them as a reliable alternative to CFRCs. The bonding performance of Constic® on both hard dental tissues should be evaluated on future studies. Also, more evidence assessing the bond strength of SAFRCs to primary teeth and etched enamel is needed.


Introduction
The simplification of dental techniques represents one of the main goals and tendencies in current restorative dentistry. Interestingly, clinical studies have shown that dental restorations performed with simplified dental materials such as universal adhesive systems, self-adhesive resin cements and bulk-fill resin composites have an acceptable performance [1][2][3][4] . Recently, another simplified dental materials known as self-adhesive flowable resin composites (SAFRCs) were introduced into the market. SAFRCs are indicated for pit and fissure sealants, base/ liner and restorative material in small cavities [5][6][7] , the same clinical indications than conventional flowable resin composites (CFRCs). According to manufacturer's instructions, SAFRCs could be used without previous phosphoric acid etching and adhesive systems, especially for dentin bonding procedures [5][6][7] . This was possible by acidic functional monomers such as glycerol phosphate dimethacrylate (GPDM), 10-methacryloyloxi-decyl-dihydrogen-phosphate (10-MDP) and 4-methacryloxyethyl trimellitic acid (4-META) incorporated into Vertise® Flow, Constic® and Fusio™ Liquid Dentin, respectively. These functional monomers establish a chemical interaction with inorganic phase of hard dental tissues which theoretically would guarantee acceptable bond strength. In some cases, only previous phosphoric acid etching on uncut enamel surface is recommended to increase the bond strength of SAFRCs 5 , but findings from some in vitro studies using this approach are controversial [8][9] .
SAFRCs could represent a good alternative to perform dental restorative/preventive procedures because they would reduce clinical time, operative errors and post-operative sensitivity [5][6][7] . Nonetheless, the number of clinical trials assessing the performance of SAFRCs restorations or pit and fissure sealants are extremely limited and controversial [10][11][12] to contraindicate or recommend these novel dental materials. However, there are fairly available in vitro studies which evaluate microleakage, nanoleakage, solubility, water sorption and bond strength of SAFRCs [13][14][15][16] . This latter is one of the most important and critical features on self-adhesive materials due to it reflects the physico-chemical interaction with hard dental tissues, which could partially predict common clinical problems such as microleakage and retention loss. Until now, no consensus on the bonding effectiveness of SAFRCs has been established to determine if these novel dental materials could be used as a reliable alternative to conventional flowable resin composites (CFRCs). Therefore, a compilation of in vitro studies on this issue is urgently needed to indicate whether current SAFRCs should be used on future research or more technological developments are required. The aim of this study was to review the current literature regarding the bond strength of self-adhesive flowable resin composites to tooth structure, comparing the results with conventional flowable resin composites.

Materials and methods
The present systematic review was conducted following all parameters described in PRISMA guidelines (Preferred Reported Items for Systematic Reviews and Meta-anal-ysis) 17 . The research question was: Do SAFRCs exhibit comparable enamel and dentin bond strength to CFRCs?

Selection criteria
We included studies that used human enamel and/or dentin from primary and/or permanent teeth, independently if dental substrates were cut, grounded and/or laser ablated (patient). The studies had to evaluate SAFRCs (intervention) such as Vertise® Flow, Fusio™ Liquid Dentin and/or Constic® used with or without previous phosphoric acid etching and/or adhesive system. Also, CFRCs (control/comparison) used as pit and fissure sealant and/or restorative material bonded by etch-and-rinse adhesive systems (ERAs), self-etch adhesive systems (SEAs) or universal adhesive systems (UAs). All included studies had to compare the bond strength between SAFRCs and CFRCs to enamel and/or dentin (outcome). Reports not published in English language, literature reviews, clinical studies, case reports/case series, book chapters, congress abstracts, editor letters and studies which exclusively evaluated the bond strength of experimental SAFRCs were excluded from the analysis of the current systematic review.

Search strategy and study selection
Different systematic searches were conducted by two trained and independent reviewers (C.M.T and S.M.P) until November 2020. We screened PubMed/Medline, EbscoHost and Scopus, using search strategies as follows; PubMed/Medline, ((((((((self-adhesive flowable composite resin) OR (self adhesive flowable resin composites)) OR (self-adhering flowable resin composite)) OR (self-adhering flowable composite resin) OR (vertise flow)) OR (fusio liquid dentin)) OR (constic)) AND (bond strength); Ebscohost, self-adhesive flowable composite resins OR self-adhesive flowable resin composite OR self-adhering flowable composite resin OR self-adhering flowable resin composite OR vertise flow OR fusio liquid dentin OR constic AND bond strength; Scopus, self-adhesive AND flowable AND resin AND composite OR self-adhering AND resin AND composite OR self-adhesive AND composite AND resin OR self-adhering AND flowable AND composite AND resin OR vertise AND flow OR fusio AND liquid AND dentin OR constic AND bond AND strength. Article titles were exported to Microsoft Excel® 2016 (Microsoft Corporation, Redmond, Washington, USA) to eliminate repeated hits in the same database and between them. Later, remaining titles and abstracts were screened in detail by two reviewers (C.M.T and S.M.P), excluding those that seem not to meet inclusion criteria. When abstracts presented limited information to be classified or seemed to meet all inclusion criteria, articles were downloaded for full-text reading. The titles were codified into 6 categories according to selection criteria, as follows: C1 (Articles not published in English language), C2 (clinical studies/case reports/ case series), C3 (Articles which did not compare the bond strength of SAFRCs with CFRCs), C4 (Studies that exclusively evaluated the bond strength of experimental SAFRCs), C5 (Others types of papers such as literature reviews, book chapters, congress abstracts and editor letters) and C6 (included studies). Finally, reference lists from selected studies were screened in detail to find possible articles which could meet inclusion criteria.

Data extraction
Data extraction was performed by two trained reviewers (C.M.T and S.M.P), using a standardized form containing information such as first author name, publication year, sample size (n), type of teeth, tested materials, type of materials, dental substrate (enamel, dentin or both), bonding test, aging technique, sample dimensions/load speed, failure mode analysis and predominant failure mode in SAFRCs. If relevant methodological information was missed from a study, we contacted the correspondence author via e-mail. If no answer was received after 2 weeks, we sent other mail, requesting the same methodological information. Finally, if no response was obtained four weeks following the first attempt, the article was included in the systematic review with not reported data (NR).

Data analysis
After methodological data extraction, meta-analysis was considered inappropriate due to great methodological divergences among included studies, especially in terms of bonding tests, load speed, adhesive systems and CFRCs. Nevertheless, means and standard deviations of bond strength values of SAFRCs and CFRCs groups from individual studies, were extracted and tabulated, indicating statistically significant differences (p≤0.05) among groups.

Risk of bias assessment
Risk of bias assessment was conducted in duplicate by two trained reviewers (C.M.T and S.M.P) and both analyses were later contrasted to find possible inconsistencies. To assess evidence quality, we employed an adapted instrument previously used in other systematic reviews about dental adhesion [18][19] . This instrument contains the following domains or items: randomization, sample size calculation, teeth free of caries, sample with similar dimensions, failure mode evaluation, manufacturer instructions, single operator and operator blinded. Each item was checked in individual studies, judging as "Yes" when reported in the methodology, but if not, the specific domain received "No". The number of positive responses obtained in each included study were counted to determine the overall risk of bias, as follows: high risk of bias (Yes:1 to 3), medium risk (Yes: 4 or 5) and low risk of bias (Yes: 6 to 8).       Table 2 summarizes the risk of bias of the included studies. Only one of the studies reported sample size was calculated, but none of the studies reported if operators were blinded. Most included studies (n=21) did not report in the methodology section whether experiments were conducted by a single operator. Conversely, aspects such as randomization, teeth free of caries, samples with similar dimensions and manufacturer instructions were reported. Overall, 20 studies scored medium risk of bias, two studies had low risk and other one scored high risk.

Discussion
According to our knowledge, this is the first systematic review that critically approaches the bonding performance of SAFRCs on permanent and primary teeth, comparing the outcomes with CFRCs associated to different adhesive systems. There were considerable variations in enamel bond strength values among included studies, probably due to methodological divergences such as type of teeth, specimen preparation technique, bonding test, enamel treatment, bonding area and load speed which made it impossible to conduct a meta-analysis. The results of this systematic review revealed low enamel bond strength of SAFRCs (Table 3), which was in agreement with previous studies testing the same SAFRCs 40,41 or self-etching sealants exhibiting similar chemical composition 42,43 . These findings may be explained due to enamel is a very complex and mineralized dental structure 44 , which requires a surface treatment prior to composite resin restorations or resin-based sealant placement. Phosphoric acid etching is the most used strategy to promote micro morphological alterations on enamel surface, leading to an effective resin interlocking and enhanced bond strength 45,46 . Some included studies revealed that SAFRCs applied under etched enamel exhibited higher bond strength values compared to SAFRCs used in self-etch mode 9,29 . However, only two studies aimed to compare the findings between resin-based sealants and SAFRCs 24,29 , highlighting the need for future research using this approach to confirm if SAFRCs applied on etched enamel could show the same bonding performance than resin-based sealants. Functional monomers such as GPDM and 4-META incorporated into Vertise® Flow and Fusio™ Liquid Dentin, respectively, are highly acidic but do not promote the same enamel demineralization pattern that phosphoric acid etching 47 . Therefore, the bonding effectiveness of these functional monomers relies largely on the chemical interaction with dental HAp, which is lower and less stable compared to that promoted by 10-MDP monomer 48 . These facts may explain why Vertise® Flow and Fusio™ Liquid Dentin used without prior phosphoric acid etching performed significantly worse than CFRCs 8,9,20,21,30,36 Other strategies to improve the bonding effectiveness of adhesive restorations involve lasers such as erbium:yttrium aluminum garnet laser (Er:YAG) or neodymium-doped yttrium aluminum garnet (Nd:YAG) 49 Regarding dentin bond strength of SAFRCs, considerable mean variations were also found among included studies, probably due to the same reasons explained for enamel bonding tests. SAFRCs exhibited statistically lower dentin bond strength in contrast to CFRCs (  8,9,22,23,28,30,31,33,[34][35][36][37][38][39] . This indicates a deficient and non-stable chemical interaction between functional monomers incorporated into SAFRCs and dentin microstructure. These hypotheses were also demonstrated by a chemical study 48 as well as on Transmission Electron Microscopy (TEM) 53 and Scanning Electron Microscopy (SEM) studies assessing Vertise® Flow 54 . This self-adhesive flowable resin composite followed by Fusio™ Liquid Dentin were the most evaluated materials, especially in primary and permanent dentin, showing similar bond strength values 16,35,36 . In contrast to GPDM and 4-META monomers, 10-MDP monomer promotes a superficial demineralization of dentin collagen fibers and enables a stable ionic interaction between phosphate group and remaining calcium ions of HAp 55 , leading to satisfactory dentin bonding performance as demonstrated in other dental materials 56,57 . Nonetheless, two included articles 35,37 tested Constic®, a 10-MDP containing SAFRCs which revealed deficient dentin bond strength values, being comparable 35 or lower 37 than other SAFRCs that do not incorporate this phosphate monomer. This raises the suggestion that 10-MDP monomer by itself did not guarantee acceptable bonding performance of this SAFRC. There are other material-dependent factors such as water content, purity and functional monomer concentration which may negatively impact the bond strength of self-adhesive dental materials 58,59 .
Self-adhesive resin cements 60 and SAFRCs are flowable materials that present similar chemical composition. Self-adhesive resin cements also incorporate silanized inorganic fillers, methacrylate monomers and an activator-initiator system. In addition, self-adhesive resin cements contain functional monomers such as 10-MDP, 4-META, Dipentaerythritol penta-acrylate monophosphate (Penta-P) or others 60 . One study 27 included in this systematic review additionally compared the bond strength of Ver-tise® Flow and some self-adhesive resin cements, showing similar bonding performance to dentin. However, it is not possible to indicate SAFRCs as alternatives for metallic crowns, posts, inlays, onlays, or ceramic crowns cementation because dualcured luting materials are desired for these clinical applications 60 . SAFRCs are not even recommended as light-cured resin cements because film thickness is not suitable for that purpose and limited color availability 5 . Besides bond strength of SAFRCs to hard dental tissues, other relevant aspects such as color stability 61 , water sorption, solubility 13 , nanoleakage 14 , microleakage 16,62 , polymerization stress, gap formation 63 need further research.
Main strengths of this systematic review were extensive searches on different databases, strict selection criteria, risk of bias assessment and data extraction. Conversely, one limitation was that most included studies evaluated the immediate bond strength of SAFRCs to hard dental tissues. This is not clinically relevant due to mechanical loading, chemical and hydrolytic degradation of laboratorial samples are important issues to predict the possible mechanical performance of adhesive restorations. In addition, the findings should be carefully interpreted due to most included evidence showing medium risk of bias (table 2) which appears to be usual in systematic reviews of in vitro studies on dental adhesion 18,19 . The lack of methodological homogeneity was another limitation that made it impossible to conduct a meta-analysis. Based on the results of the current systematic review, it is possible to affirm that chemical changes on SAFRCs as well as additional studies are required to consider these dental materials as a possible alternative in restorative and preventive dentistry. For a while, the use of phosphoric acid on enamel is essential on resin-based sealants placement. Also, the acid etching, especially in enamel and adhesive systems in both hard dental tissues remains as mandatory steps for successful restorative treatments involving resin composites 8,9,[20][21][22][23]26,27,28,31,33,36,37 .
Self-adhesive flowable resin composites, such as Vertise® Flow and Fusio™ Liquid Dentin used in self-etch mode exhibited lower bond strength to enamel and dentin from permanent teeth, compared to conventional flowable resin composites.
The bonding performance of Constic® on both hard dental tissues should be evaluated on future studies. The evidence is still limited to support that self-adhesive flowable resin composites applied under etched enamel exhibit comparable bond strength to resin-based sealants. The number of studies assessing the bond strength of self-adhesive flowable resin composites to primary teeth are limited.