Bonding to dentine: an update on universal adhesives

A brief history of bonding to dentine

In the past, dentine bonding agents were classified into generations.⁶ However, this means of identifying different groups of bonding agents generally fell into disarray because of confusion regarding which ‘generation’ each type of bonding agent fitted into. Until recently, the classification has therefore been to simply subdivide resin-based dentine bonding agents into etch and rinse materials (also known as total etch materials) and self-etch materials, with some workers classifying these according to the number of steps involved in their placement (one or two), or by their pH.^3,7

The year 1955 heralded what we now realize to be a game-changing breakthrough in restorative dentistry, namely the genesis of adhesive (and, therefore, more minimally invasive) dentistry by enabling clinicians to bond to enamel, when this was first described by Buonocore.⁸ This also has facilitated the development of resin composite materials, with these materials becoming increasingly used worldwide,¹ principally because of patient concerns regarding mercury in dental amalgam, the Minamata Agreement of 2013 that recommended reduction in the use of dental amalgam, and increasing patient requests to receive tooth-coloured restorations in their posterior teeth.⁹

Unlike enamel, dentine is a vital substrate with circa 20% organic material and 10% water by weight. These factors make bonding to dentine more challenging than to enamel. Also involved in this challenge is the smear layer on the dentine surface, this having been defined as a ‘layer which appears on the surface of teeth that have undergone dental instrumentation procedures, such as cutting with a dental bur: it cannot be rinsed away, but it can be dissolved by dilute organic acids, such as phosphoric acid’.¹⁰ This crystalline, uniform layer comprises debris consisting of hydroxyapatite particles, altered collagen and bacteria. Early resin-based dentine bonding agents were little more than wetting agents that penetrated the smear layer. When it was realized that the smear layer was not well adherent to the underlying dentine, it became obvious that it was necessary to treat it in some way. Until recently, therefore, there were two distinct types of resin-based dentine adhesive.

Etch and rinse (or total etch) materials

The etch and rinse (or total etch) materials are for which the smear layer was dissolved by the application of 30–35% phosphoric acid, followed by washing, drying and the application of a primer and the bonding agent. These materials, with a four-stage clinical application protocol, could be considered technique sensitive because of the risk of a failure if any stage was not carried out optimally.³ However, some materials of this type, such as Optibond FL (Kerr, Orange, CA, USA) remain in use, given that they are considered by some clinicians to be the gold standard to which other bonding agents must be compared.¹¹ This material has been shown to provide excellent retention at 13 years in non-carious cervical lesions (NCCL).¹² It may also be relevant to note that, according to two reviews, Optibond FL and Clearfil SE (Kuraray, Tokyo, Japan) have been considered to be the gold standard of etch and rinse and self-etch adhesives, respectively.^13,14

What happens to the smear layer with etch and rinse adhesives? The smear layer is dissolved and then washed away, with the acid also dissolving the most superficial 1–5 microns of dentine, removing hydroxyapatite, leaving only collagen fibres. When the resin bonding agent is applied to the decalcified dentine surface, a layer is formed on the decalcified dentine surface, with the resin polymerizing around the collagen fibres: this was termed the hybrid layer by Nakabayashi (Figure 1).¹⁵

Self-etch adhesives

Because of the clinical technique sensitivity associated with etch and rinse materials, and the belief by manufacturers that clinicians seek materials that are easy and fast to use, the group of bonding agents known as self-etch adhesives were presented to clinicians in the early 1990s. (In this regard, it has been considered that a material that is easy to use will produce optimal results.¹⁶) In these self-etch bonding systems, the bonding resin is combined with an acid (generally phosphoric acid) in order to reduce the pH to a level that is said to be sufficiently low to carry out the etching step at the same time as producing bonding. These materials were, therefore, significantly simpler to use than the etch and rinse materials and could be presented in only one bottle, thereby facilitating their use. The so-called self-etch adhesives may be considered to have advantages when compared to etch and rinse types, for example:³

No post-conditioning rinsing, hence less operator sensitivity;

Less sensitive to the degree of wetness/dryness of the dentine surface;

Single unit dose packaging possible, hence reduced risk of cross infection, and simple to use;

Simultaneous demineralization and resin-infiltration meant that it was not possible to over-etch the dentine, leading to reduced levels of post-operative sensitivity.

What happens to the smear layer with self-etch adhesives? The acidic resin does not remove the smear layer: the slightly decalcified surface leads to the formation of a hybrid layer, with the smear layer becoming incorporated into the hybrid layer (Figure 2).

Among the disadvantages of self-etch adhesives listed in 2004 was: ‘adhesion to enamel requires further long-term evaluation in some systems’.³ Hence, the development of the concept of selective enamel etching.

The advent of selective enamel etching

While laboratory testing allows variables to be investigated, it may be considered that the ultimate test of adhesives is their clinical performance in a non-retentive class V cavity. Results of a 5-year landmark study from Peumans and colleagues¹⁷ involving the clinical use of the (so-called) self-etch adhesive, Clearfil SE (Kuraray), in 100 non-retentive class V cavities, restored with resin composite, has indicated that, when the enamel margins were ‘selectively etched’ with phosphoric acid, the enamel cavity margins had advantages in terms of integrity and lack of staining. Accordingly, the concept of selective enamel etching was born.¹⁷ When these workers extended their study to 13 years,¹⁸ the results, with a 62% recall rate and 96% retention rate, continued to indicate fewer small marginal defects on enamel in the selective enamel etch group (16% cf 4% in the self/non-etch group). The authors commented that ‘the limited micromechanical retention of the enamel surface conditioned by the self-etching primer (pH = 2) and the stable chemical bond between 10-methacryloyloxydecyl dihydrogen phosphate (10-MDP) and hydroxyapatite (HA), seems to be strong and durable enough to have a clinically acceptable enamel bond in 95% of restorations after 13 years of clinical functioning. The authors added that ‘the only remarkable (not significant) difference in the present clinical trial was that small marginal defects and/or superficial discoloration at the enamel side were observed more frequently in the solely self-etch group. This difference between both groups, also noticed at previous recalls, has remained quite stable during the last 5 years’. In this regard, the so-called self-etch adhesives were introduced because of dentists' requests for ease of use and speed, with this not involving the use of phosphoric acid. However, with the benefit of hindsight, it could be considered that clinicians were naïve to expect materials with a pH of circa 2.5 to provide similar marginal integrity or low incidence/absence of marginal discolouration on enamel margins as could be obtained when these were etched using phosphoric acid (pH circa 0.5), remembering that pH operates on a logarithmic scale.

Universal adhesives

It should be added that the two groups of bonding agents discussed above were type specific, in other words, for the etch and rinse materials to be clinically effective, the dentine surface must be etched, while for the self-etch materials, the dentine surface must not be etched, but merely be physically clean. In this regard, the present authors suggest cleaning with pumice and water. A new group of adhesives, the universal adhesives (UAs), has therefore been introduced to overcome the type specificity associated with previous adhesives in regard to mode of etching.

A UA has been defined as one that:¹⁹

In addition:

So, while the etch and rinse and self-etch bonding agents were type specific, the UAs are indeed universal, insofar that they can be used in whichever etching mode a clinician feels is appropriate and for other clinical applications. Factors involved in the decision process regarding etch mode may include:

The rise and rise of 10-MDP

Another difference between UAs and self-etch adhesives are the functional phosphate and/or carboxylate monomers incorporated into their constituents that facilitate chemical bonding to the calcium in HA. A factor that the majority of this new group of bonding agents has in common is the resin 10-MDP,²³ with this monomer first appearing in the adhesive, Clearfil New Bond (Kuraray), in the early 1980s (Figure 3). Van Meerbeck and colleagues²⁴ have considered that 10-MDP is one of the most effective monomers to strongly ionically bond to HA, forming stable 10-MDP-Ca salts, with these authors adding that this chemically stable bond has been shown to contribute to bond durability. This has been demonstrated clinically in the 13-year study by Peumans et al¹⁸ when using Clearfil SE Bond (Kuraray), which includes 10-MDP, as adhesive for restoration of class V resin composite restorations, with very positive results,¹⁸ this having been attributed by the authors to the presence of 10-MDP in the formulation, given that the restorations without selective enamel etching were still retained despite marginal discrepancies.

It has been considered by Perdigão²⁵ that, without calcium, it is unclear how 10-MDP adhesives are able to bond ionically to etched dentine. They therefore stated that 10-MDP-containing adhesives may need to be applied only on unetched dentine, additionally using a selective enamel-etching technique.

The length of the long hydrophobic 10-carbon chain has also been reported to contribute to its bonding ability.²⁶ However, in this regard, longer chains in monomers such as 10-MDP are more hydrophobic, which could enhance the chemical interaction with calcium and reduce their degradation.²⁷

In summary, therefore, it may be considered that the incorporation of 10-MDP into an adhesive's formulation provides a ‘belt and braces’ approach, this uniquely providing an ionic bond to calcium, ie to HA in dentine, via the hydrophilic group incorporated in the molecule, along with a micromechanical bond via the hybrid layer. However, although Van Meerbeck and colleagues,²⁴ in their comprehensive review of dental adhesives, termed 10-MDP as ‘today's most effective functional monomer’, they stated that it was not the perfect functional monomer because it was sensitive to hydrolytic degradation at the two ester groups linking both functional groups to the central spacer group. The 10-MDP patent expired in 2003, with the result that manufacturers other than the patent holders, Kuraray, were able to start adding this phosphate monomer to their new adhesives.

The components of some universal bonding agents are presented in Table 1.^21,22 From this, it is apparent that all UAs contain 10-MDP and some have an added resin to facilitate bonding, such as G Premio Bond with 4-META (developed by Nakaybashi), Optibond Universal (GPDM, Kerr) Prime&Bond Active (PENTA, Dentsply Sirona, Weybridge, UK) and others have a polyalkenoate derivative such as PAC (Scotchbond Universal and Scotchbond Universal Plus, both 3M, MN, USA), and MCAP (Adhese Universal, Ivoclar-Vivadent, Schaan, Liechtenstein).

So, just how good are universal adhesives?

As with any recently introduced material, there is a relative paucity of evidence relating to these new materials, although this is building. The most researched UA is Scotchbond Universal Adhesive (SBU, 3M), given that this was the first of this group of materials to be released commercially, but other UAs have recently been examined.

UAs may be classified, similarly to self-etch adhesives, according to their pH, according to the classification by Miyazaki et al²⁸ and van Meerbeck and colleagues²⁴ in which the material was classified as shown in Table 2.

Laboratory studies

A number of laboratory studies were presented previously,⁴ namely those by Loguerico and co-workers,²⁹ da Rosa and co-workers,³⁰ Munoz and co-workers,³¹ Chen and co-workers³² and others,^33,34,35,36 with generally positive results concerning the in vitro performance of UAs. Three substantial review articles will be used to update readers on the current laboratory performance of UAs.

A recent systematic review, by Cuevas-Suarez and colleagues,³⁷ of shear bond strength testing to dentine, has updated a 2015 review by the same group. They identified 9284 publications, 81 of which were read in full and 57 finally accepted for the review. Their results indicated that:

Cuevas-Suarez and colleagues³⁷ finally stated that ‘the new meta-analyses performed in this update demonstrated that the stability of multimode adhesives' bond strength to dentine depends largely on their pH’. On enamel, irrespective of the pH of the adhesive, bond strength was improved by prior phosphoric acid etching. On the other hand, dentine bond strength of mild UAs was not dependent on the adhesive strategy used, and these adhesives seemed to be the materials with better stability. Summarizing the results of their review, Cuevas-Suarez and colleagues³⁷ provided the following helpful recommendations to clinicians:

Nagarkar and co-workers²¹ reviewed the laboratory testing and clinical performance of UAs. They identified 1250 articles, including 285 for final data analysis, of which, 12 were clinical studies (vide infra). They pointed out that chemical bonding promoted by 10-MDP was more effective and stable in water than that provided by other functional monomers, but concluded that, overall, for durable bonding with UAs, laboratory studies recommend the use of a selective enamel-etching strategy for permanent teeth, and an etch and rinse strategy for primary teeth. They concluded that ‘available laboratory and clinical evidence does not support the claim that UAs can be used with any adhesive strategy’.

Van Meerbeck and colleagues²⁴ reviewed the pros and cons of UAs. They considered that, while the etch and rinse strategy was ‘undoubtedly’ the best bonding strategy to enamel, the resultant thick and hydroxyapatite-free hybrid layer formed on dentine was highly sensitive to degradation with time.

Finally, recent laboratory studies include the work by Lago and co-workers³⁹ who compared the shear bond strength of six UAs to dentine, using Clearfil SE Bond (Kuraray) as control. The results indicated highest bond strength values for Scotchbond Universal (3M) (33.9MPa), but this was not significantly different to Clearfil Universal (Kuraray) and Tetric N-Bond (Ivoclar-Vivadent). All six UAs provided superior bond strength values to the Clearfil SE control.

In summary, therefore, laboratory studies appear to confirm that the bond strengths obtained by UAs are generally an improvement over those previously attained, with a selective enamel etch strategy being preferred.

Clinical studies

While laboratory data may provide an indicator regarding the performance of a given material, there is no substitute for clinical evaluation, given the tenuous association between in vitro and in vivo studies.⁴⁰ The previous paper⁴ cited Loguerico and colleagues,⁴¹ Perdigão and colleagues⁴² and Lawson et al,⁴³ in demonstrating the short-term clinical success of Scotchbond Universal (SBU). Recently published clinical research on UAs, with evaluation periods of greater than 2 years, includes the following:

The recall rate was 86%, with 19 restorations (of 153 assessed) lost, and the results indicated that the SBU adhesive performed worse in self-etch mode. Significant differences were also observed for both marginal staining and adaptation, with the clinical behaviour of this UA being better in etch and rinse mode when compared with the self-etch category. The authors therefore recommended that, if a self-etch strategy was used, selective enamel etching be employed.

In summary therefore, there is a strong body of evidence that indicates that recently developed UAs provide clinical effectiveness as good as, or better, than previous ‘gold standard’ adhesives, and that selective etching of the enamel is desirable, given that the results presented above indicate improved retention rates of class V restorations when the margins are etched, and reduced levels of discolouration around the margins of all restorations. The present authors therefore strongly recommend this procedure. Does that statement apply to all UAs? It is the authors' view that, in view of the similarities between many of the UAs (Table 1^21,22), and the fact that their pH values tend to lie between 1.5 and 3, it is prudent to suggest that this is carried out if the clinician wishes to limit marginal staining over time.

Handling evaluations

There have been four ‘handling’ evaluations of UAs^51,52,53,54 by a UK-based practice-based research group, the PREP (Product Research and Evaluation by Practitioners) Panel. The first⁵¹ involved SBU soon after its release, with 12 evaluators placing a total of 875 restorations (class 1:172; class II:189; class III: 134; class IV: 178; class V: 182; other: 20) and rating the material on visual analogue scales. The results indicated that the bonding agent used prior to the evaluation scored 4.0 out of 5.0 (where 5 indicated optimal ease of use), with SBU scoring 4.9. More recently, G-Premio Bond (GC, Leuven, Belgium), Prime&Bond Active (Dentsply Sirona) and Zipbond (SDI, Melbourne, Australia) have been evaluated,^52,53,54 with ease of use scores being 4.9, 4.8 and 4.9, respectively. It may therefore be concluded that the UAs tested, and possibly UAs in general, score highly for ease of use.

Finally, it has been considered that patient- and operator-related factors may have a higher impact on restoration longevity than the actual adhesive employed,⁵⁵ therefore, with regard to the clinical use of UAs, it is worth adding that, while isolation with rubber dam is optimal (although not universally used⁵⁶) and moisture control (by whatever means) is essential, it may be considered that a further advantage is the reduced number of steps, and concomitant reduced technique sensitivity of these new adhesive systems.

Recent innovations

A radiopaque dentine adhesive

3M have recently introduced Scotchbond Universal Plus, which bears similarities to Scotchbond Universal, except that a radiopaque resin has been incorporated, substituting the BisGMA in the original version, and the silane has been improved, along with the adhesive's ability to bond to carious dentine, according to manufacturer's data. The addition of the radiopaque resin helps clinicians avoid any doubt regarding whether a radiolucent layer at the base of a posterior composite restoration is a pool of (non-radiopaque) resin or residual/secondary caries (Figure 4).

Universal adhesives as zirconia/ceramic primers?

It is beyond the scope of this article to address these topics in detail. However, in short, given that the resin 10-MDP has a demonstrable adhesion to zirconia by ionic and hydrogen bonding,⁵⁷ UAs containing this resin may act as zirconia primers. On the other hand, the bond strength has been shown to reduce following 6 months' storage in water⁵⁸ and a further study questioned whether UAs containing 10-MDP produced a significant effect regarding bonding to zirconia.⁵⁹ It would therefore appear that more work is required to solve one of dental materials' most problematical problems, namely, how to achieve a reliable bond of resin to zirconia.

Two UAs contain a silane in their list of components, namely Clearfil Universal (Kuraray) and Scotchbond Universal/Universal Plus (3M). These UAs could therefore be used as primers for materials such as glass ceramics, alumina ceramics and lithium disilicate, especially when used in conjunction with the resin luting material from the same manufacturer.

MMPs: friend or foe

Matrix metalloproteinases (MMPs) are proteolytic enzymes, found in dentine, that can lead to the destruction of collagen (ie the organic matrix) in dentine when the pH is low. Destruction of the collagen in the hybrid layer can result, potentially leading to a reduction in bond strength, and which could be extrapolated to failure of adhesive restorations. As a result, MMP inhibitors have been proposed by some workers, adding an extra step in a dentine bonding protocol, such as the application of chlorhexidine. Most of the work on this subject has been carried out in vitro; however, a meta-analysis has failed to endorse the use of MMP inhibitors.⁶⁰ Clinicians may therefore breathe a sigh of relief that their dentine bonding has not been made more difficult!

If bonding is as good as this, what are the clinical applications?

Figures 5–8 present a variety of the clinical applications that could be appropriate to today's universal adhesives.

Conclusions

In summary, universal adhesives hold promise and:

However, as with any new material or technique, more long-term clinical evaluations (alongside those referenced above) are needed to adequately demonstrate the value of these universal adhesives.