Adhesive sealing of dentin surfaces in vitro: A review

Abstract

Purpose

The purpose of this review is to describe the evolution of the use of dental adhesives to form a tight seal of freshly prepared dentin to protect the pulp from bacterial products, during the time between crown preparation and final cementum of full crowns. The evolution of these “immediate dentin sealants” follows the evolution of dental adhesives, in general. That is, they began with multiple-step, etch-and-rinse adhesives, and then switched to the use of simplified adhesives.

Methods

Literature was reviewed for evidence that bacteria or bacterial products diffusing across dentin can irritate pulpal tissues before and after smear layer removal. Smear layers can be solubilized by plaque organisms within 7–10 days if they are directly exposed to oral fluids. It is likely that smear layers covered by temporary restorations may last more than one month. As long as smear layers remain in place, they can partially seal dentin. Thus, many in vitro studies evaluating the sealing ability of adhesive resins use smear layer-covered dentin as a reference condition. Surprisingly, many adhesives do not seal dentin as well as do smear layers.

Results

Both in vitro and in vivo studies show that resin-covered dentin allows dentinal fluid to cross polymerized resins. The use of simplified single bottle adhesives to seal dentin was a step backwards. Currently, most authorities use either 3-step adhesives such as Scotchbond Multi-Purpose^a or OptiBond FL^b or two-step self-etching primer adhesives, such as Clearfil SE^c, Unifil Bond^d or AdheSE^e, respectfully.

Introduction

When indirect restorations are used to restore function, dentists must seal the exposed dentin with temporary materials during the interval required to fabricate and cement the final restoration. Full crown preparations expose up to 1 cm² of dentin that contains more than 3 million tubules/cm².¹ Such tubules represent millions of microscopic “pathways to the pulp” because they all terminate in the tooth pulp. Both enamel and cementum are impermeable and nerve-free. These peripheral seals have very low permeabilities. However, once these surface sealing hard tissues are removed from dentin surfaces, the exposed dentin becomes highly permeable and very sensitive to hydrodynamic stimuli². Because dentinal tubules contain collagen fibrils/fibers, constrictions etc. their functional diameter (0.1 µm) is far smaller than their 1.0 µm anatomical diameter³. This allows dentin to function like a 0.1 µm Millipore filter to prevent bacteria from invading the pulp via dentinal tubules. However, soluble bacterial products can permeate through dentin to the pulp where they provoke immunological reactions and pulpal inflammation that threaten pulpal health^4–6.

Prevention of pulpal inflammation

Current prosthodontists recommend conservative tooth preparations using copious air-water spray and intermittent cutting. This is followed by temporizing the preparations using a variety of temporary filling materials and crown formers as long as there are no pre-existing signs of pulp pathology. The rationale is that a healthy pulpodentin complex reacts to tooth preparation by the deposition of tertiary dentin under those tubules that were cut during cavity preparation that should wall off the prepared dentin and prevent bacterial invasion⁷.

On the other hand, old teeth have smaller pulps with fewer mesenchymal cells, and a poorer blood supply⁸. Tertiary dentin requires more than 30 days to begin to form and that dentin will not form if the pulp under the cut tubules produces an inflammatory response⁷. Old pulps contain pulp stones that interfere with endodontic treatment. If a temporary crown is lost and the cut dentin is exposed, bacterial products will begin to diffuse down the tubules toward the pulp. Pulpal cells will react to these bacterial antigens as if actual bacteria were invading the pulp. This will trigger neurogenic inflammation in the pulp leading to pulpal symptoms^9,10.

Those that advocate immediate dentin sealing (IDS) of freshly prepared dentin seek to protect the pulp from bacteria and bacterial products, using adhesive resins^11–24. These resins should prevent dentinal fluid from permeating from inside dentin, through polymerized resin to the surface. The resins should also prevent the inward diffusion of bacterial products through the polymerized resin. This involves acid-etching dentin with 37% phosphoric acid for 15 sec when using etch-and-rinse adhesives. When using self-etching adhesives, the two-bottle primer adhesives are preferred over single-bottle systems because the acidic primer is covered with a solvent-free adhesive, rich in dimethacrylates that create stronger resin films than monomethacrylates. Thicker resin seals are better than thinner coatings, because they are less likely to be lacerated by multiple setting of crowns during their final adjustments.

Combinations of resin adhesives covered by flowable composites provide tougher IDS than resin films alone. Optibond FL^b is an example of a 3-step, etch-and-rinse adhesive that uses an adhesive that contains 48% fumed SiO₂ and barium aluminoborosilicate Na₂SiF₆²⁵. Scotchbond Multi-Purpose^a adhesive contains few fillers and is often covered with a flowable composite to make the IDS stronger. The final adhesive material is covered with glycerin gel before polymerization to prevent the formation of an oxygen-inhibited layer. Such a layer of unpolymerized comonomers interferes with polymerization of impression material²⁶.

Advantages of IDS include eliminating post-cementation sensitivity²⁷, eliminating the potential risk of bacterial leakage and pulpal irritation. Local anesthesia is often not required for try-ins and occlusal adjustments. IDS should be considered for any patients at increased risk of pulpitis due to immunosuppression, aged pulps, previous history of tooth sensitivity, patients taking bisphosphonates, etc. who are at risk of developing bisphosphonate-induced osteonecrosis of the jaw²⁸. If such patients require root canal therapy and the roots are perforated during RCT, such extraction sites fail to heal properly. Abutment teeth in such patients should receive IDS as a precautionary measure.

The influence of the thickness of IDS materials has also been reported²⁹. That group found that thicker adhesives increased fracture resistance of IDS Empress 2 ceramic crowns.

The purpose of this review was to follow the evolution of the use of dental adhesives to create an immediate dentin seal to protect the pulpodentin complex from inflammatory insults, in vitro. This review will begin with etch-and-rinse adhesives, followed by self-etching primer/adhesives and finally single-bottle, simplified self-etching adhesives.

Historical review

The work of Bergenholtz^4,5 showed that bacterial products could diffuse across freshly prepared dentin to induce pulpal inflammation. This lead Pashley et al. in 1992 to propose sealing freshly prepared dentin with adhesive resins¹¹. This was endorsed by Davidson’s group in 1996¹², Paul and Schaerer 1997¹³, and Özturk et al.¹⁴. Prof. Tagami advocated “resin coating” of freshly cut dentin to prevent pulpal irritation^15–17, and to increase adhesion to dentin. Others have also stressed the importance of resin sealing^18–24. In their 1992 paper, Pashley et al.,¹¹, the authors mounted the crowns of extracted human third molars on plexiglass blocks and prepared full crown preparations (Fig. 1). They measured dentin permeability as a hydraulic conductance before and after sealing the preparations with Prisma Univeral Bond 2^f, Scotchbond 2^a, Superbond C&B^g, Amalgambond^h, Gluma Bondⁱ or Clearfil Photobond^c. All adhesives reduced dentin permeability in vitro by at least 50%, with Prisma Universal Bond 2^f sealing the best followed by Superbond C&B^g, Amalgambond^h and Scotchbond 2^a. The worst seals were produced by Gluma Bondⁱ. Bonds made with Scotchbond 2^a and Clearfil Photobond^c gave excellent initial seals, but began to leak after thermocycling the bonded dentin.

Fig. 1.

Schematic of the apparatus used to measure dentin permeability. N₂ gas passed into a pressure vessel containing a beaker of water. The water passes through the system at whatever pressure is used. The rate of movement of a tiny air bubble in a 25 µL micropipette quantitates how much water permeates across exposed dentin. The microsyringe adjusts the position of the air bubble for the next trial (from Pashley et al.¹¹, with permission).

Bouillaguet et al.²⁰ used extracted human third molar crowns flattened on the occlusal surface and glued to a plexiglass base penetrated by 18 ga. stainless steel tubing to permit measurement of dentin permeability. They measured the hydraulic conductance of acid-etched dentin before and after bonding with Scotchbond Multi-Purpose^a, Prime & Bond 2.0^f or All-Bond 2^j (Table 1). All of these etch-and-rinse adhesives reduced dentin permeability by 83%, 90% or 96.6%, respectively. Unfortunately, the authors only measured the initial reduction in permeability. When Gregorie et al.³⁰ evaluated the sealing ability of Optibond Solo Plus^b, Single Bond^a, Excite^e and Prime & Bond NT^f, all these etch-and-rinse adhesives only reduced dentin permeability to a residual value of 40%, while the self-etching adhesives Clearfil SE Bond^c, and Prompt-L Pop^a only reduced dentin permeability down to a residual value of 36 and 16%, respectively (Table 2). They also only measured initial permeability. Better initial permeability results were reported when Vaysman et al.³¹ using Clearfil SE Bond^c and Optibond Solo Plus^f to seal acid-etched dentin. Both adhesives sealed dentin 80–95%. All of the above authors remarked that no dental adhesive reduced dentin permeability 100%.

Table 1.

Effects of bonding on the permeability of human dentin.

Hydraulic conductance of dentin (µL min−1 cm−2 cm H2O−1)
Adhesive Systems	Smeared dentin	Etched dentin	Bonded dentin
SBMP Plusa	b1.1 ± 0.7×10−2 (24.4%)	a4.5 ± 3.0×10−2 (100%)	b0.78 ± 0.6×10−2 (17%)
P&B 2.0f	b1.0 ± 0.6×10−2 (20%)	a5.0 ± 3.2×10−2 (100%)	c0.5 ± 0.4×10−2 (10%)
All-Bond 2j	b1.2 ± 1.0×10−2 (23%)	a5.2 ± 3.9×10−2 (100%)	c0.24 ± 0.1×10−2 (3.4%)

Values are mean hydraulic conductances ± SD (n=12). Values in parentheses are the hydraulic conductance expressed as a percent of the control, acid-etched values that were defined as 100%. Groups identified by the same superscript are not significantly different (from Bouillaguet et al.²⁰, with permission).

Table 2.

Ability of various adhesives to seal dentin.

Bonding agent	Acid-etch permeability	Residual permeability after bonding
Excitee	150.13 ± 56.26a	40.21 ± 23.09a
Optibond Solo Plusb	175.77 ± 30.19a	38.53 ± 21.87a
Single Bonda	138.30 ± 27.08a	42.79 ± 13.61a
Prime & Bond NTf	137.82 ± 36.32a	41.87 ± 14.926a
Clearfil SE Bondc	--	36.42 ± 20.30ab
Prompt L-Popa	--	16.24 ± 15.66b

Values expressed in percentage decrease (−) or increase (+) with respect to baseline value (n=6). Mean ± SD, Duncan’s test (different superscript letters, a, b, or ab, indicate different groups (from Grégorie et al.³⁰, with permission).

Elgalaid et al.³² were the third group to attempt to seal full crown preparations with Prime & Bond NT^f or leave them covered with a smear layer as controls. They were surprised to discover that smear layers sealed dentin as well as the adhesive, but that either approach only reduced dentin permeability by 55–64% for up to 3 weeks (the longest time studied). Unfortunately, smear layers are very acid-labile and can solubilize in as little as one week when exposed³³. Carrilho et al.³⁴ repeated Elgalaid’s³² work in 2007, and obtained similar in vitro results. That is, smear layer/smear plugs seal dentin better than adhesive resins/resin tags.

Water movement across resins, in vivo

Unfortunately, few clinical studies of “immediate dentin sealing” have been reported. Hu and Zhu¹⁸ reported that the incidence of post-cementation hypersensitivity was significantly lower in teeth that received immediate dentin sealing with Prime & Bond NT^f compared to abutment teeth that were prepared and temporized without immediate dentin sealing. Tay³⁵ and others have prepared human teeth for full crowns and then taken polyvinyl siloxane impressions of smear layer-covered dentin (controls) or adhesive-coated dentin.

These impressions were poured up in epoxy resin for SEM examination of in vivo fluid transudation across dentin “sealed” with smear layers or adhesives, as evidenced by the presence of tiny bubbles of dentinal fluid on smear layer-covered or “adhesive-sealed” dentin (Fig. 2). All two-step, self-priming, total-etch adhesives tested (Single Bond^a, Prime & Bond NT^f, One-Step^j, Excite DSC^e) were covered with fluid droplets indicating that dentinal fluid had transudated across the polymerized adhesives in the 3–4 min required for self-polymerization of the polyvinyl siloxane impression material. Less fluid droplets came across smear-layer covered dentin than across polymerized resins.

Fig. 2.

SEM micrographs of epoxy resin replicas of crown preparations taken in vivo. (A) Smear layer(s) covered dentin showing how microdrops of dentinal fluid transudated across the smear layer and were trapped by the impression material in the 2–3 min required for the setting time. (B) Resin-covered crown preparation sealed with Single Bond^a, a 2-step one-bottle adhesive. Note the presence of extensive fluid mirodrops. (C) Higher magnification of (B) showing the presence of nanodroplets of dentinal fluid. (D) Resin-coated dentin after gentle clearing with pumice slurry. This produced cracks and partial removal of adhesive (arrow) exposing underlying dentin (from Tay et al.³⁵, with permission).

Chersoni et al.³⁶ extended the in vivo work of Tay et al.³⁵ by making replicas of resin-covered dentin in vivo and also conducting fluid filtration studies across crown segments in vitro (Fig. 3). When self-etch adhesives Adper Prompt^a, Xeno III^f, iBondⁱ and One-Up Bond F^k were used to seal dentin, Adper Prompt^a and Xeno III^f could not seal dentin as well as did smear layer-covered dentin (Table 3). One-Up Bond F^k sealed dentin as well as did smear layers. When the two-step self-etching system Unifil Bond^d was used to seal dentin, it reduced dentin permeability to 2.1% of smear layer covered dentin values (Table 3)! This was due to the fact that the acidic primer layer was covered by a solvent-free hydrophobic layer (Fig. 3). Unifil Bond^d produced the best seal of dentin, in part, because it was a self-etching system, and because it was covered with a neat hydrophobic resin that absorbed little water³⁰.

Fig. 3.

SEMs of resin replicas of crown preparations of vital teeth covered with (A) smear layer. Note presence of a few microdroplets of transudated dentinal fluid trapped by the impression material. (B) Unifil Bond^d bonded surface after removal of O₂-inhibited resin. No fluid droplets were seen when using this two bottle, self-etching primer adhesive (from Chersoni et al.³⁶, with permission).

Table 3.

Fluid conductance across dentin before and after bonding.

	% Fluid flow compared to acid-etched controls
Adhesives self-etch	Smear layer covered	Bonded dentin
Adper Prompta	17.3 ± 4.5A	28.3 ± 4.4A
Xeno IIIf	12.4 ± 6.2A	24.2 ± 2.9A,B
iBondi	15.1 ± 5.6A	18.7 ± 3.3B
One-Up Bond Fk	14.0 ± 3.1A	14.9 ± 5.0B
Two-step
Unifil Bondd	18.2 ± 5.0A	2.1 ± 2.1C

Groups identified by different superscript letters are significantly different (p<0.05) (from Chersoni et al.³⁶, with permission).

Clearly, not all polymerized resins should be used for immediate dentin sealing, for if dentinal fluid can transudate across the “resin-sealed dentin,” those water droplets might interfere with polymerization of impression materials. Ghiggi et al.³⁷ sealed dentin with Clearfil SE Bond^c (CSE) alone or CSE covered with a glycerin jelly to exclude oxygen and polymerized through jelly. Other specimens were treated with CSE that was polymerized in air, but the oxygen-inhibited layer was removed with alcohol, or the CSE was covered with Protect Liner F^c (PLF), a flowable composite alone or the flowable was covered with a glycerin jelly or the polymerized PLF was scrubbed with alcohol before taking impressions with Express XT^a or Impregum^a. Their results showed that small amounts of impression material remained attached to dentin sealed with CSE alone or PLF, but were not found on CSE covered with jelly or scrubbed with alcohol. Magne and Nielsen²⁴ reported that only CSE used with Extrude^b generated ideal impressions. They did not recommend Impregum^a for taking impressions of immediate dentin seals.

Immediate dentin seals, if treated properly, can serve as a foundation for indirect composite inlays. Duarte et al.³⁸ found that immediate dentin sealing with Adper Single Bond^a gave microtensile bond strengths of 51.1 MPa, while Adper Prompt L-Pop^a only gave 1.7 MPa bond strength. If dentin is temporized without immediate dentin sealing, the subsequent use of Optibond FL^b or CSE^c for indirect restorations falls 11.6 MPa or 1.81 MPa, respectively³⁹. When those same adhesives were used for immediate dentin sealing and the final restoration delayed 1–12 weeks, the microtensile bond strengths exceeded 45 MPa for both adhesives³⁹. Dillenburg et al.⁴⁰ reported that immediate dentin seals should be cleaned of temporary cement with aluminum oxide powder and 37% phosphoric acid treatment, followed by a second layer of the same adhesive. These microtensile bond strengths were similar to the controls that were never temporized.

Water movement across polymerized resins, in vitro

Özok et al.⁴¹ prepared class II cavities (Fig. 4) in crown segments in vitro, and measured the permeability of smear layer-covered dentin. Then the cavities were bonded with Scotchbond 1^a or Prompt L-Pop^a following the manufacturer’s instructions, under 15 cm H₂O of pulpal pressure or 0 cm H₂O pulpal pressure. Scotchbond 1^a reduced dentin permeability 48% when bonded under 15 cm H₂O pressure, but 88% when no pressure was applied (Table 4). In contrast, Prompt L-Pop^a reduced dentin permeability 88% when 15 cm H₂O was applied during bonding, but 95% when no pulpal pressure was applied.

Fig. 4.

Schematic of the apparatus used for measuring the dentin permeability of class V cavities before and after restoration with bonded composites (from Özok et al.⁴¹, with permission).

Table 4.

Percent reduction in dentin permeability in class V composite restorations bonded with Scotchbond or Prompt L-Pop with or without pulpal pressure (15 cm H₂O).

	Percent reduction in permeability
Adhesive	with pulp pressure	without pulpal pressure
Scotchbond-1a	48.2 ± 49.2a	88.3 ± 19.1b
Prompt L-Popa	88.4 ± 17.3b	94.6 ± 7.8b

Groups identified by different superscript letters were significantly different (p<0.05) (from Özok et al.⁴¹, with permission).

Itthagarun et al.⁴² created exposed, flat dentin surfaces in human dentin and then bonded them with Prompt-L-Pop^a, Etch & Prime 3.0^m, One-Up Bond F^k, Reactmer Bond^l or Unifil Bond^d. Unifil Bond^d lowered dentin permeability 98% to a residual post-bonded permeability of only 2% of smear layer values (similar to the results of Chersoni et al.³⁶ above), while Prompt L-Pop^a and Etch & Prime 3.0^m didn’t seal dentin any better than did the smear layer (Table 5). One-Up Bond F^k and Reactmer Bond^l were better than Prompt L-Pop^a and Etch & Prime^m but not as good as Unifil Bond^d (Table 5).

Table 5.

Fluid flow across bonded dentin as a percent of maximum acid-etched values.

	Fluid flow (% of maximum) measured
Adhesives	across smear layers	across bonded dentin
Single-step
Prompt L-Popa	18.9 ± 5.3A	19.2 ± 3.6A
Etch & Prime 3.0m	18.9 ± 4.4A	17.4 ± 5.8A,B
One-Up Bond Fk	14.8 ± 4.4A	7.7 ± 4.9C
Reactmer Bondl	13.1 ± 3.9A	8.9 ± 5.7B,C
Two-step
Unifil Bondd	18.2 ± 5.0A	2.1 ± 2.1D

Groups identified by different superscript letters are significantly different (p<0.05) (from Itthagarun et al.⁴², with permission).

Magne²³ recommended the etch-and-rinse adhesive, OptiBond FL^b, to seal dentin. After acid-etching and priming of dentin, the final adhesive layer is solvent-free and relatively hydrophobic, like the adhesive of Unifil Bond^d and creates an excellent seal. Indeed, Scotchbond Multi-Purpose’s^a adhesive produces good dentin sealing because it is also solvent-free. Clearfil SE Bond^c adhesive is another example of a solvent-free, relatively hydrophobic resin blend that seals dentin well.

Grégorie et al.⁴³ bonded smear layer-covered dentin in vitro with ten self-etching adhesives. Smear layers reduced dentin permeability 40–50% compared to acid-etched values. However, when they bonded the 4000 grit SiC smear layers with Xeno III^f, AdhSE^e, Adper Prompt L-Pop^a, Etch & Prime^m or One-Up Bond F^k, the post-adhesion permeability fell 50–60% compared to the initial acid-etched value (Table 6). Optibond Solo Plus^b, Prime & Bond NT^f-treated dentin reduced dentin permeability 45.6 to 42.3%. Prime & Bond^f non-rinse conditioner and 1 layer of Prompt L-Pop^a only reduced dentin permeability 16% (Table 6).

Table 6.

Dentin permeability of specimens after bonding with self-etching adhesives, compared to acid-etched controls.

Adhesives	Reductions in permeability after bonding
Xeno IIIf	−68.9 ± 15.7%a
AdhSEe	−58.2 ± 26.8%a,b
Adper Prompt L-Popa	−51.0 ± 33.2%a,b
Etch & Prime 3.0m	−56.8 ± 26.3%a,b
One-Up Bondk	−60.5 ± 27.3%a,b
Optibond Solo Plus SEb	−45.6 ± 34.4%a,b
Prime & Bond NT controlf	−42.3 ± 5.1%b
Clearfil SE Bondc	−35.5 ± 14.7%b,c
Prime & Bond NRC NTf	−16.4 ± 33.3%c
Prompt L-Popa	−16.3 ± 13.3%c

Values are mean ± SE, N=10. Groups identified by different superscript letters are significantly different (p<0.05). Smear layer was created using 4000 grit SiC paper. Only 1 layer of adhesive was used with Prompt L-Pop, while multiple layers were used with Adper Prompt L-Pop. Prime & Bond NRC NT is a no rinse conditioner that does not require water rinsing, used with Prime & Bond NT adhesive (from Gregorié et al.⁴³, with permission).

Use of potassium oxalate to seal tubules with crystals

Yiu et al.⁴⁴ reported that the in vitro dentin permeability of smear layer-covered dentin varied from 9–11% of acid-etched values. When acid-etched dentin was treated with BisBlock^j (2.8% oxalic acid), the permeability fell to between 9.4- 11.9% of acid-etched values, much like the smear layer covered dentin (Table 7). When acid-etched dentin was bonded with One-Step^j, Single Bond^a, OptiBond Solo Plus^b or Prime & Bond NT^f, the permeability only fell to 31–34% of acid-etched values that were all significantly higher than the smear layer or BisBlock^j values (Table 7). However, when BisBlock^j was used to pretreat dentin prior to bonding with One-Step^j or Single Bond^a, the dentin permeability fell to values that were only 2 or 6%, respectively, far lower permeability values than were produced by smear layers or by BisBlock^j alone, indicating that a double seal by calcium oxalate crystals and resin tags lowers dentin permeability. One-Step^j and Single Bond^a reduced dentin permeability almost to zero. In contrast, when BisBlock^j-treated dentin was bonded with OptiBond Solo Plus^b or Prime & Bond NT^f, the permeability of dentin only fell to 27–28%, about like the use of these adhesives without BisBlock^j (Table 7). Further studies were done to try to determine why OptiBond Solo Plus^b and Prime & Bond^f were not as effective as One Step^j and Single Bond^a. Table 7 shows that the pH of OptiBond Solo Plus^b and Prime & Bond NT^f were 2.8 and 2.7, respectively. Clearly, OptiBond Solo Plus^b and Prime and Bond NT^f were more acidic than the other two adhesives. When the adhesives were analyzed for ionic fluoride, One-Step^j and Single Bond^a contained 70 and 130 ppm F^-, while OptiBond Solo Plus^b contained 4527 ppm F^- and Prime & Bond NT^f contained 3641 ppm F^- (Table 7). Further, TEM studies of ammoniacal silver nitrate immersed specimens revealed the presence of globular deposits into dentinal tubules that may be CaF₂-phosphate complexes. These were not found in One-Step^j or Single Bond^a bonded specimens.

Table 7.

Effects of adhesives on fluid conductance of dentin.

	Percent of fluid flow at 20 cm H2O pressure
Adhesive	Smear layer covered	Oxalate Tx only	Oxalate Tx plus adhesive	Adhesive only	pH	F− ppm
One Stepj	11.1 ± 2.0b	9.4 ± 1.9b	2.0 ± 0.9a	33.6 ± 8.9c	4.6	70
Single Bonda	10.8 ± 1.8b	9.9 ± 2.5b	6.0 ± 1.7a,b	34.2 ± 11.3c	3.6	130
OptiBond Solo Plusb	10.7 ± 5.9b	11.9 ± 5.6b	28.3 ± 12.2c	31.4 ± 4.7c	2.8	4527
Prime & Bond NTf	8.7 ± 1.7b	10.9 ± 4.7b	27.5 ± 10.8c	31.1 ± 6.3c	2.7	3641

Fluid flow from acid-etched dentin was assigned a value of 100%. Each tooth served as its own control. Groups identified by different superscript letters are significantly different (p<0.05) (from Yiu et al.⁴⁴, with permission).

When low viscosity polyvinyl siloxane impressions were made of the BisBlock^j pretreated, resin-bonded surfaces, epoxy resin replicas were made for SEM examination. Figure 5 shows only a few droplets of water on the surface of One-Step^j or Single Bond^a bonded dentin, but extensive amounts of water had filtered across the polymerized resin of Prime & Bond NT^f or Optibond Solo Plus^b in the 3–4 min required for setting of the impression material.

Fig. 5.

SEM micrograph of epoxy resin replicas taken from acid-etched dentin treated with 2.7% potassium oxalate (pH 2.7) prior to bonding with adhesives. (A) One Step^j bonded dentin with only a few microdrops of fluid transudation. (B) Single Bond^a bonded dentin showed few microdrops (pointer) and nanodrops (triangle). (C) Prime & Bond NT^f bonded dentin showed massive numbers of microdrops that had transudated across the polymerized adhesives. (D) Optibond Solo Plus^b bonded dentin contained so many microdroplets that they coalesced into pools of fluid (from Yiu et al.⁴⁴, with permission).

Clearly, low pH and high fluoride concentrations somehow interfere with the sealing properties of OptiBond Solo Plus^b and Prime & Bond NT^f, but not with One Step^j or Single Bond^a. When less acidic, low fluoride adhesives are combined with pretreatment by oxalic acid, one obtains double sealing of dentinal tubules⁴⁵.

Fluid movement across polymerized resins

Sauro et al.⁴⁶ extended the work of Tay et al.³⁵ and Chersoni et al.³⁶ by quantitating the number of water droplets on various adhesive resins bonded to dentin (Fig. 6), and to then correlate it with actual fluid flow transudation across the bonded dentin using the hydraulic conductance technique (Fig. 7). They published a highly significant positive linear regression equation showing an R² = 0.96 of the relationship between the number of fluid droplets on resin surfaces and the permeability of the adhesives (Fig. 7).

Fig. 6.

SEM microscopy impression replicas of bonded dentin before and 3 min after application of a physiologic pulpal pressure (20 cm H₂O). (A) Dentin bonded with One Up Bond F Plus^k with no pulpal pressure. (B) Same specimen after application of pulpal pressure. (C) Dentin specimens bonded with Clearfil S^3c with pressure. (D) Specimen bonded with G-Bond^d with pressure. (E) Dentin specimens bonded with Clearfil Protect Bond^c, a two-bottle self-etching adhesive where the primed dentin is covered by a solvent-free adhesive (from Sauro et al.⁴⁶, with permission). Abbreviations: CPB = Clearfil Protect Bond^c; G = G-Bond^a; CS3 = Clearfil S³-Bond^c, OUB-F = One-Up Bond F Plus^k.

Fig. 7.

Regression analysis between adhesive permeability (abscissa) and number of microdroplets of fluid on the bonded surface (ordinate) after application of 20 cm H₂O pressure. A significant (p<0.001) positive correlation (R² = 0.96) was seen between the two variables (from Sauro et al.⁴⁶, with permission).

Two years later, Sauro et al.⁴⁷ measured droplet formation by tandem confocal scanning microscopy (TSM) dentin sealed with G-Bond^d, DC-Bond^c, Single Bond^a, OptiBond FL^b or Filtek Silorane^a. The best sealing was obtained using G-Bond^d. This was the exact opposite of their results obtained in 2007. The authors explain that the polymerized G-Bond^d films looked frothy, but did not allow much fluid transudation. OptiBond FL^b sealed dentin gave the next to the best seal. DC Bond^c and Single Bond^a gave the worst seals.

Grégorie et al.⁴⁸ tested the moisture-sensitivity of Prompt L-Pop^a versus Scotchbond 1XT^a by measuring the ability of these resins to seal dentin compared to their acid-etched values. Scotchbond 1XT^a was more sensitive to overly dry dentin than was Prompt L-Pop^a. Scotchbond 1XT^a reduced dentin permeability by 55%, while Prompt L-Pop^a reduced it by 62%, although these values were not significantly different. These relatively poor seals suggest that the ability of adhesive resins to seal dentin had not improved much over the 2000–2009 period.

Use of resin-modified glass ionomers

Rusin et al.⁴⁹ evaluated the resin-modified glass ionomer (Vitrebond Plus^a) for its ability to seal dentin, when applied to acid-etched or smear layer-covered dentin. When applied to smear layers, Vitrebond Plus^a reduced dentin permeability 98.9 ± 1.1%, because it was a second layer on top of a smear layer that contained smear plugs in the tubules. This is the best seal that we have seen published in the literature. When Vitrebond Plus^a was placed on acid-etched dentin it created a hybrid layer complete with resin tags (Fig. 8). The Vitremer Plus^a matrix separated from the filler particles creating Vitremer Plus matrix resin tags about 2 µm long. Vitrebond Plus^a reduced the permeability of acid-etched dentin by 87.7 ± 18.6%. This means that some specimens produced perfect seals, while others only sealed dentin 69%.

Fig. 8.

SEM of Vitrebond Plus^a light-cured glass ionomer liner/base applied to acid-etched dentin. Resin tags of the Vitrebond^a matrix separated from the GIC fillers, extended 2 µm into open tubules (from Rusin et al.⁴⁹, with permission).

Use of 2-bottle self-etching primer adhesives

Clearly, most resins do not seal dentin as well as do enamel or cementum. Two step, self-etching adhesives tend to seal better than do etch-and-rinse adhesives because the mildly (pH 2–2.4) acidic versions barely etch through smear layers (Fig. 9). As many smear plugs are more than 2 µm long and are covered by 1 µm thick smear layers, the mild etching of self-etching adhesives fails to remove the smear plugs, preventing dentinal fluid from contaminating dentin surfaces⁴⁴. Self-etching adhesives only contain 25–35% water, while etching dentin with phosphoric acid and rinsing with water, leaves demineralized dentin floating in 70 vol% water⁴¹. Thus, it is easier to displace 25–35% water from thin hybrid layers than trying to displace 70 vol% water using etch-and-rinse adhesives.

Fig. 9.

Schematic illustrating the smear layer/smear plug complex on the left. When a weakly acidic primer is applied, it loosens up the smear layer and the smear plug and infiltrates them with resin (Violet). The presence of residual smear plugs prevents outward seepage of dentinal fluid during bonding. After application of the water-free, neutral adhesive (purple), the bond is well-sealed (from Pashley⁵⁹, with permission).

For some etch-and-rinse adhesives (One-Step^j and Single Bond^a), surface water can be controlled by applying oxalic acid to acid-etched dentin prior to bonding. Oxalates do not work with more acidic, high fluoride-containing products like OptiBond Solo^b or Prime & Bond NT^f44.

The two-step self-etching primer/adhesives like Unifil Bond^d, Clearfil SE Bond^c or Clearfil Protect SE^c or AdheSE^e, confine their water and acidic monomers to their primers and cover the primed dentin with solvent-free, relatively hydrophobic resins that seal dentin much better than do simplified, single bottle adhesives (e.g. G-Bondⁱ, iBond^j, Xeno III^f, etc.). If we assume that higher resin-dentin bond strengths represent improved dentin sealing, then there are several papers that show that multiple resin coats improve dentin bonding.

Attempts to seal dentin using multiple coatings of adhesives has resulted in mixed success. Pashley et al.⁵⁰ compared applications of one versus two coats of Prompt L-Pop^a. Using one layer of adhesive, as recommended, produced microtensile bond strength of only 14.2 ± 7.2 MPa (n=24). Application of two layers produced a µTBS of 29.7 ± 5.7 (n=23, p<0.001). Transmission electron microscopy of the bonded specimens revealed a thin layer of polymerized resin on top of the hybrid layer. There was about 10–15 µm of oxygen-inhibited adhesive on top of the 5–7 µm polymerized resin. This layer was responsible for the low bond strength of single layers of adhesives.

Use of multiple layers of resin

In 2004, Hashimoto et al.⁵¹ tested the effects of multiple applications of OptiBond Solo Plus^b or Single Bond^a on both microtensile bond strength and silver nanoleakage. The results showed a progressive increase in bond strength and decrease in nanoleakage with each additional layer, up to 4 coats. Solvent evaporation after each coating increased the thickness of the adhesive layer. This result was confirmed by Ito et al.⁵² using the single bottle, self-etch adhesives Xeno III^f or iBondⁱ. The results are summarized in Table 8.

Table 8.

Effects of multiple layers of two self-etch adhesives on microtensile bond strength.

	Layers
Groups	1	2	3	4	5	1 + SBMP
Xeno IIIf (evaporation)	7.2 ± 6.3a	22.6 ± 9.2a,b	30.0 ± 9.9b	43.5 ± 7.7c	41.4 ± 8.0c	27.8 ± 5.9b
Xeno IIIf (evaporation + 1c)	7.2 ± 6.3a	47.3 ± 19.2b	65.4 ± 20.4d	81.8 ± 20.8d	68.9 ± 22.6d	64.7 ± 21.2d
iBondi (evaporation)	12.2 ± 7.5A	18.5 ± 6.2A	30.6 ± 7.0B	34.2 ± 6.0B,C	51.6 ± 14.8B,C	62.3 ± 23.0C
iBondi (evaporation + 1c)	12.2 ± 7.1A	39.6 ± 15.2B	42.7 ± 12.3B,C	36.6 ± 7.7B	35.8 ± 13.4B	58.0 ± 23.5C

Values are mean ± SD in MPa, N=16. Groups identified horizontally by different superscript letters are significantly different (p<0.05) (from Ito et al.⁵², with permission). +lc = with light-curing of each layer.

The results revealed that the bond strengths with Xeno III^f increased with the number of coatings, just as in the Hashimoto et al.⁵¹, but the bond strengths were significantly higher after each coating, if the layer was light-cured (Table 8). The reverse was true with iBondⁱ. Using iBondⁱ, simply evaporating the solvent of each layer increased bond strengths faster than if each layer was evaporated and light-cured. While the results were very interesting, the technique requiring multiple solvent evaporations and multiple light-curing is so laborious that most clinicians would not adopt it.

There was one final group in the Ito et al.⁵² study, where after application of the first layer of self-etching adhesive, the “primed” dentin surface was covered with a layer of solvent-free Scotchbond Multi-Purpose^a (SBMP) adhesive, which was then light-cured. In the group 1 specimens, where only the solvent in the first layer was evaporated, but not light-cured, application of one layer of SBMP^a and its light-curing gave a bond strength of 27.8 ± 5.9 MPa. In group 2, where the single layer of Xeno III^f was both evaporated and light-cured prior to application of SBMP^a, the bond strength increased 232% to 64.7 ± 21.2 MPa (Table 8, 1 + SBMP column). Using iBondⁱ in groups 3 and 4, the bond strengths were above 55 MPs regardless of whether the single layer of iBondⁱ was only evaporated or was evaporated and light-cured prior to application of SBMP^a adhesive. Silver nanoleakage studies showed that self-etching single-bottle adhesives exhibit much more nanoleakage than does SBMP^a adhesive because self-etch adhesives contain residual water, while SBMP^a adhesives are water-free (Figs. 10A and 10B).

Fig. 10.

TEM micrographs of (A) dentin bonded with a coating of iBondⁱ, followed by solvent evaporation and light-curing. The bonded surface was then coated with Scotchbond Multi-Purpose^a adhesive that was light-cured. After soaking in 50% ammoniacal silver nitrate overnight, the specimens were processed for transmission electron microscopy (TEM). Note the moderate silver nanoleakage in the bottom layer of iBondⁱ adhesive (A) and the near absence of nanoleakage in the SBMP^a adhesive (R). (B) Dentin bonded with one layer of Xeno III^f, then the solvent was evaporated and the surface light-cured. The bonded surface was covered with one coat of water-free SBMP^a adhesive (R). Note there is much less silver nanoleakage in (R) than in (A) (from Ito et al.⁵², with permission).

The results of the Ito et al.⁵² study clearly show the benefits of covering self-etching adhesives with solvent-free, dimethacrylate-rich adhesives. This concept was reinforced by the paper by Brackett et al.⁵³. In that paper, the microtensile bond strengths of three single-bottle self-etching adhesives Prompt L-Pop^a, iBondⁱ and Xeno III^f were compared when they were used according to the manufacturer’s instructions and light-cured, or when the cured adhesive was covered with an additional layer of a solvent-free, neutral, dimethacrylyte-rich adhesive. The results are shown in Table 9, confirming that the use of solvent-free adhesives over simplified self-etch adhesives improves bond strengths.

Table 9.

Microtensile bond strengths of dentin specimens bonded with simplified, all-in-one adhesives alone, or plus an additional layer of hydrophobic resin from the same manufacturer.

Adhesive	Microtensile bond strength (MPa)
iBondi	41.0 ± 17.7a
iBond + Gluma Solid Bond Si	53.5 ± 19.0b
Prompt L-Popa	16.4 ± 9.9c
Prompt L-Pop + Scotchbond Multi-Purpose adhesivea	56.4 ± 21.8b
Xeno IIIf	10.3 ± 8.3c
Xeno III + experimental adhesivef	25.8 ± 8.0c

Values are mean ± SD, N=5. Groups identified by different superscript letters are significantly different (p<0.05) (from Brackett et al.⁵³, with permission).

The question of whether resin-dentin bond strengths can be increased using multiple layers of resin, seems to have been answered in the affirmative. However, Reis et al.⁵⁴ asked whether the durability of one-step self-etching adhesives would be improved by double application or by an extra layer of hydrophobic dentin. Both the double layer and the hydrophobic layer either improved the immediate bond strength, or did not differ from the results obtained with the manufacturer’s directions (Table 10). Resin-dentin bond strengths after 6 months of water storage were higher when the additional layer of hydrophobic resin was used compared to the results obtained with the manufacturer’s directions. These results should be repeated but measuring dentin permeability instead of bond strength, and extended to 12 months of water storage. Reis et al.⁵⁵ repeated their 2008 study⁵⁴ in a clinical trial of retention of resin composites bonded with the Clearfil S^3c or iBondⁱ using either the manufacturer’s directions or an additional hydrophobic layer (in this study, the solvent-free adhesive of Scotchbond Multi-Purpose^a Adhesive system) bonded to noncarious cervical lesions for 18 months. The results showed higher 18-month retention rates for both adhesives if they were covered with an extra layer of hydrophobic resin, especially iBondⁱ (p<0.05)⁵⁵.

Table 10.

Microtensile bond strength of self-etching adhesives to dentin measured immediately or after 6 months, using manufacturer’s instructions or double application or an extra layer of hydrophobic resin.^*

	Microtensile bond strength to dentin
Adhesive	Time	Manufacturer Directions	Double application	Hydrophobic layer
Prompt L-Popa	Immediately	24.3 ± 3.1b	45.3 ± 6.4a	29.5 ± 4.9b
	6 month	16.9 ± 4.1c	17.2 ± 4.6b	29.1 ± 3.6b
Xeno IIIf	Immediately	30.0 ± 1.2a	33.8 ± 3.4b	49.8 ± 3.7a
	6 month	25.2 ± 3.9b	30.3 ± 4.2b	43.2 ± 4.8a
iBondi	Immediately	19.1 ± 2.4b	29.2 ± 2.2a	37.6 ± 3.4a,b
	6 month	18.1 ± 4.8b	26.4 ± 3.9b	28.2 ± 3.1b

Values are means ± SD, in MPa.

^*Hydrophobic layer was the adhesive from Clearfil SE Bond^c. Groups identified by different superscript letters are significantly different at p<0.05 (from Reis et al.⁵⁴, with permission).

Evaluation of sealing using silver nitrate

Another approach to evaluating the sealing ability of various adhesives to dentin is to measure silver nanoleakage⁵⁶. This is done by immersing resin-bonded dentin into 50 wt% ammoniacal silver nitrate for 24 hrs, followed by rinsing off excess AgNO₃ and exposing the specimens to photodeveloper and/or bright light to reduce the silver ions to metallic silver grains that cannot diffuse after reduction.

Silver nitrate uptake discloses the presence and distribution of water-filled channels within the hybrid layer and overlying adhesive layer. Figure 11 shows two transmission electron micrographs of resin-dentin bonds made with single-bottle, simplified self-etching adhesive. These adhesives contain 25–35% water to ionize their acidic monomers. Although clinicians attempt to evaporate the residual water in the adhesives prior to light-curing, 5–15 sec is insufficient to remove all water⁵⁷. Some of that residual water resides in the small lateral branches of dentinal tubules (Fig. 11, pointer) that are sometimes referred to as “water-trees”⁵⁸. Water-trees sometimes seem to originate from hybrid layers (often from one or two dentinal tubules), that extend up into the overlying adhesive layer (Fig. 11). Water-trees are seldom seen in dentin bonds made with 2-step self-etching adhesives like Clearfil SE Bond^c because the water in the acidic primer is sealed by the water-free hydrophobic adhesive like Clearfil SE Bond^c adhesive or Scotchbond Multi-Purpose^a adhesive⁵⁰.

Fig. 11.

Transmission electron micrographs of dentin bonded with an all-in-one adhesive, incubated in 37°C water for 24 h, followed by immersion in 50% ammoniacal silver nitrate overnight to show the presence and distribution of residual water in hybrid layers (H) and adhesive layer (A) by silver grains. Silver filled branching channels are called “water-trees” (pointers). Spot-like silver deposits are simply called nanoleakage (from Pashley⁵⁹, with permission).

Role of matrix metalloproteinase enzymes in bond degradation

When using etch-and-rinse adhesives, the etchant is 37% phosphoric acid. This acid is strong enough to solubilize and extract all extrafibrillar and intrafibrillar apatite crystallites from the top 8–10 µm of acid-etched dentin. Since the mineral content of dentin is about 70 vol%, the removal of all mineral from the dentin matrix is replaced by 70% water⁶⁰. It is very difficult to replace 70% water with adhesive monomers in the 30–60 sec most clinicians use to infiltrate adhesives into dentin. The result of incomplete resin-infiltration is incomplete water removal. Silver ions saturate any residual water-filled spaces^56,60,61.

During acid-etching, not only is the dentin matrix exposed, but the matrix metalloproteinases (MMPs) bound to the matrix is are uncovered and activated by acid-etching⁶². MMPs slowly degrade collagen by adding water across specific peptide bonds in collagen⁶². The more residual water in resin-dentin bonds, the greater the rate of hydrolysis. Collagen degradation not only lowers resin-dentin bond strength, but slowly increases dentin permeability by loosening resin tags and destroying the hybrid layer⁶⁰. Thus, for many reasons, it is important to minimize residual water in resin-dentin bonds.

One approach to minimizing residual water at bonded interfaces is to treat the 8–10 µm demineralized layer with 10% NaOCl gel⁶³. This treatment dissolved some of the collagen fibrils in the hybrid layer and lowered the bond strength of Single Bond^a by 69% and Prime & Bond NT^f by 69%⁶³. Others have shown that a solution of 5% NaOCl is more effective than gels. The use of 5% NaOCl as an endodontic irrigant to remove exposed collagen is well known⁶⁴. It is used, in part, to disinfect root canals of contaminating bacteria as well as remove exposed collagen so that endodontic seals are placed on stiff mineralized dentin rather than soft, demineralized dentin. Sodium hypochlorite has a pH of 12. It is a strong oxidizing agent. Its action lowers the bond strength of free-radical polymerizing resins. This oxidized state can be reversed within 2–5 min by treating NaOCl-treated dentin with 10% sodium ascorbate^65,66. The only disadvantage is that a 3 GPa adhesive layer comes in direct contact with 18 GPa underlying mineralized dentin, since most of the 8–10 µm thick demineralized layer is removed by NaOCl.

Are extra steps of pretreatment able to improve resin sealing of dentin as was seen in the Yiu et al.⁴⁴ study with oxalate? Oxalates can be used with some etch-and-rinse adhesives but not with others.

Multi-step approach to sealing dentin

Abu-Nawareg⁶⁷ compared the dentin sealing ability of three adhesive systems: Adper Single Bond Plus^a, AdheSE^e, and G-Bond^d representing an etch-and-rinse adhesive, a two-step self-etching primer adhesive, and a single bottle, simplified self-etching adhesive.

Single Bond Plus^a controls (no pretreatments) showed the highest residual permeabilities, 25.3 ± 0.7% at day 1 and 40.4 ± 1.4% (p<0.05) after 2 months in a simulated body fluid at 37°C. When Single Bond Plus^a specimens were pretreated with oxalate (BisBlock^j) just before bonding, their 1 day permeabilities were only 11.7 ± 0.8% and their 2 month values were 17.4 ± 0.8% (p<0.05). Sequential pretreatment of etched dentin with both NaOCl and oxalate produced a significant (p<0.05) further reduction in residual permeability to 5.7 ± 0.6 and 7.6 ± 0.5% after 1 day or 2 months, respectively (Fig. 12).

Fig. 12.

Permeability of human dentin after bonding with Single Bond^a, AdhSE^e or G-Bond^d using the manufacturer’s instructions, or after pretreatment with 2.7% potassium oxalate (pH 2.7), or after sequential pretreatment with 5.25% NaOCl followed by oxalate, after 1 day or 2 months (from Abu-Nawareg⁶⁷, Adhesive sealing of dentin, Cairo University, PhD Thesis, 2011, with permission).

Control specimens bonded with AdheSE^e showed the greatest sealing of dentin by exhibiting residual post-bonded permeabilities of only 11.0 ± 0.6 and 15.1 ± 0.7% after 1 day or 2 months, respectively (Fig. 12). When AdheSE^e bonded specimens were pretreated with oxalate (BisBlock^j), their dentin permeability fell (p<0.05) to 2.6 ± 0.6 and 5.0 ± 0.4% after 1 day or 2 months storage, respectively. Sequential pretreatment of dentin with NaOCl, plus oxalate produced the lowest permeability recorded in that study, 0.80 ± 0.2 and 1.2 ± 0.4%, respectively, after 1 day or 2 months.

Control specimens bonded with G-Bond^d produced intermediate residual permeabilities of 16.2 ± 0.8 and 20.5 ± 0.7% after 1 day or 2 months, respectively. Oxalate pretreated specimens bonded with G-Bond^d showed significantly (p<0.05) higher permeabilities 22.1 ± 0.9 (day 1) versus 30.8 ± 0.8% (60 days), respectively (Fig. 12). When G-Bonded^d specimens were subsequently pretreated with NaOCl and oxalate, their permeabilities were significantly higher from the oxalate alone values (Fig. 12).

The superior sealing ability of AdheSE^e, a two-bottle, self-etching primer adhesive, mimics the excellent results obtained by others^36,37 using Unifil Bond^d. In such two bottle systems, the water-containing acidic primer is scrubbed on the dentin for 20 sec, dried, and then covered with a water-free, dimethacrylate-containing adhesive that provides an excellent seal. Other two-bottle self-etching adhesives such as Clearfil SE Bond^c or Clearfil Protect SE^c or Optibond Solo^b should work equally well.

The use of Adper Prompt L-Pop^a for dentin sealing of crown preparations is not recommended because it etches far more deeply and removes smear plugs as well as smear layers, allowing dentinal fluid access to the bonded interface. As the depth of etch is nearly as deep as that produced by 37% phosphoric acid, there is concern that adhesive monomers may not infiltrate to the full depth of the etch⁵⁸, allowing dentin proteases to degrade the resin-dentin bond⁶².

Recent expanded applications of resin sealing of dentin

In the U.S., Pashley’s group began sealing the coronal top of root canal fillings with unfilled adhesives so that the pink gutta percha could be visualized through an adhesive intracoronal seal⁶⁸ in case the root canal filling needed to be removed sometime later. Others used Vitrebond^a, but it is not transparent⁷⁰. Intracanal seals protect root canal fillings during the time interval between completion of root canal treatment and permanent restoration of the tooth.

In Japan, Tagami’s group were also advocating intracoronal seals to prevent microleakage in endodontically-treated teeth^69,71. Imazato et al.⁷² recommended the use of antibacterial monomers. Nikaido et al.⁷³ reported that the bond strength of resin cements to resin-coated dentin could be improved by covering the adhesive with a flowable composite. Magne’s latest recommendations for “immediate dentin sealing” also recommend covering resin seals with flowable composites⁷⁴. To treat exposed root surfaces in elderly patients with cervical caries, Daneshmehr et al.⁷⁵ and Tajima et al.⁷¹ demonstrated that coating root surfaces with adhesive resins decreased biofilm adherence compared to untreated root surfaces. Others have incorporated antimicrobial quaternary ammonium methacrylates like 12-methacryloyloxydodecyl pyridinium bromide (MDPB) into adhesives⁷². The use of MDPB is indicated when bonding to dentin containing bacteria⁷². More recently, MDPB has been shown to be a potent inhibitor of dentin MMPs and cathepsin K⁷⁷. Thus, resins can be bioengineered to have therapeutic effects⁷⁶. By increasing the concentration of acidic monomers in adhesive blends, Brambilla et al.⁷⁸ were able to inhibit S. mutans colonization.

Summary

The use of adhesive resins for immediate sealing of dentin (ISD) began in 1992, using early generation dentin bonding agents alone. As bonding agents improved, in vitro studies demonstrated that Scotchbond Multi-Purpose^a (SBMP) adhesive or Optibond FL^b sealed dentin better than most other adhesives and so they were widely employed in ISD. When self-etching adhesives were developed, the best dentin seals were produced by Unifil Bond^d36 or Clearfil SE Bond^c24,37,39 (CSE). These adhesive systems seal the acid-etched dentin with a solvent-free adhesive that is rich in dimethacrylates that have been shown to minimize water sorption into the bonded assembly^75,76. Others empirically found that covering SBMP^a or CSE^c with a flowable composite such as Protect Liner F^c provides additional sealing benefits^24,37,38, while various pretreatments can improve resin sealing of dentin⁶⁷, though they require extra bonding steps. More clinical studies need to be done to determine if IDS prevents later adverse pulpal responses to indirect restorative procedures.

Magne’s latest recommendations⁷⁴ include the use of flowable composite over unfilled adhesives as an alternative to the use of filled adhesives, and to cover the resin sealed preparation with glycerine gel to prevent forming a layer of oxygen inhibited resin at the surface that interferes with polymerization of impression material. More controlled clinical trials need to be done to statistically test whether post-restorative pulpal health is significantly better in teeth protected by IDS, compared to the use of conventional temporization.

CLINICAL SIGNIFICANCE.

Although newly developed adhesive resins have attempted to improve dentin sealing, many such attempts failed. The use of two-step self-etching primer adhesives that combine the use of an acidic primer with a solvent-free adhesive layer give excellent sealing in vitro. Alternatively, the use of three-step, etch-and-rinse adhesive systems like Scotchbond Multi-Purpose^a or OptiBond FL^b that also use solvent-free adhesives seal dentin very well, in vitro.