Hybrid Layer as a Dentin-Bonding Mechanism Nohito Nakahayashi, Ph.D.* Mitsuo Nakamura, Ph.D.' Nohoroir Yasudu. Ph.D.
'
A number of mechanisms (both mechanical and chemical)have been proposed as the cause of dentin adhesion. Extensive research in Japan during the past 10 years has shown that strong, long-livedbonds between resin and living dentin will form when a monomer such as 4-META, which contains both hydrophilic and hydrophobic chemical groups, penetrates the tissue and polymerizes in situ. This resin-impregnation creates a transitional 'hybrid" layer. that is neither resin nor tooth, but a hybrid of the two. The thin layer of resin-reinforced dentin locks the two dissimilar substances together on a molecular level, sealing the surface against leakage and imparting a high degree of acid resistance.
0
nce Dr. Buonocore pointed the way, it became apparent that strong, durable bonds to etched enamel were relatively easy to create.' However, similar success bonding to dentin has remained elusive. The problem. of course,is the fundamental difference in the nature of the two tissues. Enamel is largely inert, composed primarily of hydroxyapatitewith very low water content. Dentin, on the other hand, is lMng tissue. Its chemicalstructure involvesboth inorganic and organic materials, and it features a highly complex physical structure that varies with the depth of the tissue. Depending on the particular tooth, the age of the patient, and the depth of the preparation, the substrate surface can consist of widely varying proportions of intertubular dentin, peritubular dentin, and sclerotic dentin. The dentin may be very de=...or highly porous. Near the dentin-enamel junction (DEJ) only l percent of the surface area consistsoftubules. N e a r the pulp, tubules can comprise as much as 22 percent of the surface area.aFurthermore,those tubules communicatewithaveryvul~erablepulpthatreactsadvemely to mechanical,t.hen.1~4biological, or chemical insult.
ing to the type and depth of dentin, approximately 13 percent of overall dentinal volume is fluid.3 Is it possible to chemically bond to dentin? Twenty years ago the research community believed it was, and the first two generations of dentin 'adhesives" claimed to create chemical bonds to the dentin. Unfortunately, these agents performed much better in the laboratory than in the mouth. With the exception of covalent bonds, chemical bonds are vulnerableto degradationin the oral environment, which results in leakage, discoloration, and secondary caries. Based on research conducted over the past decade at the Institute of Medical and Dental Engineering. the authors have seen little to indicate that durable graft polymerization (i.e., chemical adhesion)to ground dentin can be achieved. First of all, it is extremely ditllcult to achieve effective chemical reaction when one of the materials is a solid. because the frequencyof molecular collisions is very low. (Most chemical reactions are designed to take place in SoMion.) Furthermore. there are a number of limiting criteria that any chemical reaction involving dentin would have to meet: 1. It would have to occur at body temperature. 2. It must not involve monomem or catalysts that might injure the pulp. 3.The reaction must be completed within 10 min-
Andtomakematterswwse.thedentlnissaturatedwith agygenandwater,Thoughwatercontentvstriesaccord-
Utes.
HYBRID h 1963 Masuhara et al reported that when M-nbutyl bosane EBB)was used as an initiator, methyl methanylate (&lMf% c h a w grafted to the collagen in wet ivory and produced an egcellent bond.' The 133
THL3 to enamel. A number of new organic molecules were studicd.7'Though virtually all of them significantly improved the bond (Table 1). by far the most stable bonds whcre achieved when small amounts of 4met hacryloxyet hyl trimellitate anhydride (4-META)were added to the monomer.H Microscopic examination of enamel that had been bonded with this new copolymer yielded a surprising finding.According to the generally-accepted, tag-theory of enamel bonding, pretreatment using phosphoric acid creates mechanical undercuts by selectively attacking the interprismatic material. The resin adhesive then locks into these undercuts to create mechanical retention. In other words. even the best monomers could not penetrate beyond the depth of the etch. A n yet, that is precisely what examination of the 4-META-resin-totooth interface suggested was happening. The resin appeared to penetrate beyond the etch and encapsulate the prisms (Figs. 1 and 2). Chemical analysis of the interface showed that the tags themselves were pure resin, but at the end of the tags there was a thin zone, where the resin had impregnated the interprismatic material and formed a new material that was part tooth and part resin. We called this zone a resin-reinforced tissue the "hybrid layer." Because no such layer was found when we examined the MMA-TE3B-to-enamel interface, we hypothesized that this hybrid was the reason for the dramatic difference in bond stability. The hybrid layer is highly acid resistant. It remains intact even after both enamel and dentin have been dissolved in HCl. This suggested the intriguing possibility of simultaneously bonding to the tooth and rendering it caries-resistantwithout the use of fluoride.
.idhwoii w,i\ srronc .ind rxtrrrriely durable. As 01 ,J.iniIan' l!W 1 . \miples ol this system have k*enaging in water for 15 yrars at 37 C and the h i d is still I I inct lon,d
Linfortunately. the system did not adhere togroiind dentin. In 1969 this MMA-THB system was introduced as a n orthodontic- adiiesivc called "Direct E3onding-Systern' for enumel bonding.' " It was one of the first dental adhesives commercially introduced alter Huonocore proposed thc concept of 'wid-etching enamel. While exploring ways to improve the MMA-TBB system at the Institute for Medical and Dental Engineering. we discovered that the addition of monomers that contained both hydrophobic groups (like the phenyl group) and hydrophilic groups (like the carboxyl group) dramatically enhanced the bonding stability of MMA-
Additwe -- -
None 5% HPPM' 3%HNPM' 5% 4-META'
Etchant' .~-
Bond MPo Strengih Reference . .
10-3 10-3 10-3 EDTA 3 - 2 10 - 3 1-1 EDTA 3 2 10-3
-
596Phenylc1
133 104 105 209 173 163 157 105
2
2
t ?:
.t i: .t 5
49 38 40 28 49 02 24 1.8
10 18 19
20 19 10 21
22
-10 - 3 fOr 3Qs: EDTA fW bos.
tSeeFie1.
A
B Flgam I. SEM ofencapsulated dprisms. A. ?he 4-META-basedadhcslvc penetrates beyond the etch and diffuses into the tnkrpdemtk Ussue. 7he hybrid It forrns when It combtnes Wlth this tlssue Is hlghly acid mistant.The honeycomblike pattern Is t h c v matahl that mnatncd aReh theenamel rodsdimolvedin HCl. B.When 4-METAwasomitted hmn the formula. the xxstn -not penetrate the intqi-bmatictissucto forma hybrid.” R curcdrestn. E: enamel.
134
f iybrid L;\yrr as a I)entin-Bonding Mechanism
DENTIN HYBRID LAYER It was discovered that the 4-META/MMA-TBB system formed tenacious bonds not only to enamel. but also to cast metals, amalgam, acrylic. and composite. However, initial results applying the new 4-METAJ MMA-TBB system to dentin were disappointing.’ Fusayama et a1 proposed that phosphoric acid was a n appropriated pretreatment for dentinaswell as But when phosphoric acid was used to prepare the dentin. the 4-META-based resins produced very low bond strengths. SEM examination found no evidence of the expected hybrid layer formation. It appeared that the acid was not only decalcifying the dentin, but was also denaturing the collagen, decreasing its diffusivity and preventing monomer contalning 4-META and 4MET from penetrating the tissue to form a hybrid. Building on the suggestion of Dr. Raphael Bowen. a technique was developed involvinga short pretreatment using an aqueous solution of 10 percent citric acid and 3 percent ferric chloride. l o In contrast to the phosphoric acid, a short 10-second treatment with the 10-3 solution permitted the operator to dissolve a thin layer of calcium without damaging the collagen. The strength of the 4-META/MMA-TBB system to bovine dentin prepared with the 10-3solution increased by a factor of 200 percent to 18 MPa. Interestingly, when the ferric chloride was eliminated from the etching formula, and the dentin was prepared wlth a 10 percent solution of citric acid, the bond strength plummetted to just 6 MPa, which was no better than when the surface was prepared with phosphoric acid. The resin tag length in both the 6 MPa and the 18 Mpa samples was identical. The only observed dtfference was that the 18 MFb samples showed formation of a win impregnated zone -a hybrid layer. The 6 MPa sample showed no hybrid layer (Fig. 3).The ferric
I
Elchlnp
Resln
u Generally accepted concept
m
Resin
C Hybrd
Hybrld concept
proms a.
Traclimnd envnel bonding theory postulates that resin tags mecbnkaIiy lock into undercuts in the etched e n d . The hybrid conceptproposesthat bondingstability will incnase stgnlacantty if the rrsfn tags arc supplemented by a thin u 3 of~resin-ninfomdtissue & y M d layer).
B
A
135
'
A number of mechanisms (both mechanical and chemical)have been proposed as the cause of dentin adhesion. Extensive research in Japan during the past 10 years has shown that strong, long-livedbonds between resin and living dentin will form when a monomer such as 4-META, which contains both hydrophilic and hydrophobic chemical groups, penetrates the tissue and polymerizes in situ. This resin-impregnation creates a transitional 'hybrid" layer. that is neither resin nor tooth, but a hybrid of the two. The thin layer of resin-reinforced dentin locks the two dissimilar substances together on a molecular level, sealing the surface against leakage and imparting a high degree of acid resistance.
0
nce Dr. Buonocore pointed the way, it became apparent that strong, durable bonds to etched enamel were relatively easy to create.' However, similar success bonding to dentin has remained elusive. The problem. of course,is the fundamental difference in the nature of the two tissues. Enamel is largely inert, composed primarily of hydroxyapatitewith very low water content. Dentin, on the other hand, is lMng tissue. Its chemicalstructure involvesboth inorganic and organic materials, and it features a highly complex physical structure that varies with the depth of the tissue. Depending on the particular tooth, the age of the patient, and the depth of the preparation, the substrate surface can consist of widely varying proportions of intertubular dentin, peritubular dentin, and sclerotic dentin. The dentin may be very de=...or highly porous. Near the dentin-enamel junction (DEJ) only l percent of the surface area consistsoftubules. N e a r the pulp, tubules can comprise as much as 22 percent of the surface area.aFurthermore,those tubules communicatewithaveryvul~erablepulpthatreactsadvemely to mechanical,t.hen.1~4biological, or chemical insult.
ing to the type and depth of dentin, approximately 13 percent of overall dentinal volume is fluid.3 Is it possible to chemically bond to dentin? Twenty years ago the research community believed it was, and the first two generations of dentin 'adhesives" claimed to create chemical bonds to the dentin. Unfortunately, these agents performed much better in the laboratory than in the mouth. With the exception of covalent bonds, chemical bonds are vulnerableto degradationin the oral environment, which results in leakage, discoloration, and secondary caries. Based on research conducted over the past decade at the Institute of Medical and Dental Engineering. the authors have seen little to indicate that durable graft polymerization (i.e., chemical adhesion)to ground dentin can be achieved. First of all, it is extremely ditllcult to achieve effective chemical reaction when one of the materials is a solid. because the frequencyof molecular collisions is very low. (Most chemical reactions are designed to take place in SoMion.) Furthermore. there are a number of limiting criteria that any chemical reaction involving dentin would have to meet: 1. It would have to occur at body temperature. 2. It must not involve monomem or catalysts that might injure the pulp. 3.The reaction must be completed within 10 min-
Andtomakematterswwse.thedentlnissaturatedwith agygenandwater,Thoughwatercontentvstriesaccord-
Utes.
HYBRID h 1963 Masuhara et al reported that when M-nbutyl bosane EBB)was used as an initiator, methyl methanylate (&lMf% c h a w grafted to the collagen in wet ivory and produced an egcellent bond.' The 133
THL3 to enamel. A number of new organic molecules were studicd.7'Though virtually all of them significantly improved the bond (Table 1). by far the most stable bonds whcre achieved when small amounts of 4met hacryloxyet hyl trimellitate anhydride (4-META)were added to the monomer.H Microscopic examination of enamel that had been bonded with this new copolymer yielded a surprising finding.According to the generally-accepted, tag-theory of enamel bonding, pretreatment using phosphoric acid creates mechanical undercuts by selectively attacking the interprismatic material. The resin adhesive then locks into these undercuts to create mechanical retention. In other words. even the best monomers could not penetrate beyond the depth of the etch. A n yet, that is precisely what examination of the 4-META-resin-totooth interface suggested was happening. The resin appeared to penetrate beyond the etch and encapsulate the prisms (Figs. 1 and 2). Chemical analysis of the interface showed that the tags themselves were pure resin, but at the end of the tags there was a thin zone, where the resin had impregnated the interprismatic material and formed a new material that was part tooth and part resin. We called this zone a resin-reinforced tissue the "hybrid layer." Because no such layer was found when we examined the MMA-TE3B-to-enamel interface, we hypothesized that this hybrid was the reason for the dramatic difference in bond stability. The hybrid layer is highly acid resistant. It remains intact even after both enamel and dentin have been dissolved in HCl. This suggested the intriguing possibility of simultaneously bonding to the tooth and rendering it caries-resistantwithout the use of fluoride.
.idhwoii w,i\ srronc .ind rxtrrrriely durable. As 01 ,J.iniIan' l!W 1 . \miples ol this system have k*enaging in water for 15 yrars at 37 C and the h i d is still I I inct lon,d
Linfortunately. the system did not adhere togroiind dentin. In 1969 this MMA-THB system was introduced as a n orthodontic- adiiesivc called "Direct E3onding-Systern' for enumel bonding.' " It was one of the first dental adhesives commercially introduced alter Huonocore proposed thc concept of 'wid-etching enamel. While exploring ways to improve the MMA-TBB system at the Institute for Medical and Dental Engineering. we discovered that the addition of monomers that contained both hydrophobic groups (like the phenyl group) and hydrophilic groups (like the carboxyl group) dramatically enhanced the bonding stability of MMA-
Additwe -- -
None 5% HPPM' 3%HNPM' 5% 4-META'
Etchant' .~-
Bond MPo Strengih Reference . .
10-3 10-3 10-3 EDTA 3 - 2 10 - 3 1-1 EDTA 3 2 10-3
-
596Phenylc1
133 104 105 209 173 163 157 105
2
2
t ?:
.t i: .t 5
49 38 40 28 49 02 24 1.8
10 18 19
20 19 10 21
22
-10 - 3 fOr 3Qs: EDTA fW bos.
tSeeFie1.
A
B Flgam I. SEM ofencapsulated dprisms. A. ?he 4-META-basedadhcslvc penetrates beyond the etch and diffuses into the tnkrpdemtk Ussue. 7he hybrid It forrns when It combtnes Wlth this tlssue Is hlghly acid mistant.The honeycomblike pattern Is t h c v matahl that mnatncd aReh theenamel rodsdimolvedin HCl. B.When 4-METAwasomitted hmn the formula. the xxstn -not penetrate the intqi-bmatictissucto forma hybrid.” R curcdrestn. E: enamel.
134
f iybrid L;\yrr as a I)entin-Bonding Mechanism
DENTIN HYBRID LAYER It was discovered that the 4-META/MMA-TBB system formed tenacious bonds not only to enamel. but also to cast metals, amalgam, acrylic. and composite. However, initial results applying the new 4-METAJ MMA-TBB system to dentin were disappointing.’ Fusayama et a1 proposed that phosphoric acid was a n appropriated pretreatment for dentinaswell as But when phosphoric acid was used to prepare the dentin. the 4-META-based resins produced very low bond strengths. SEM examination found no evidence of the expected hybrid layer formation. It appeared that the acid was not only decalcifying the dentin, but was also denaturing the collagen, decreasing its diffusivity and preventing monomer contalning 4-META and 4MET from penetrating the tissue to form a hybrid. Building on the suggestion of Dr. Raphael Bowen. a technique was developed involvinga short pretreatment using an aqueous solution of 10 percent citric acid and 3 percent ferric chloride. l o In contrast to the phosphoric acid, a short 10-second treatment with the 10-3 solution permitted the operator to dissolve a thin layer of calcium without damaging the collagen. The strength of the 4-META/MMA-TBB system to bovine dentin prepared with the 10-3solution increased by a factor of 200 percent to 18 MPa. Interestingly, when the ferric chloride was eliminated from the etching formula, and the dentin was prepared wlth a 10 percent solution of citric acid, the bond strength plummetted to just 6 MPa, which was no better than when the surface was prepared with phosphoric acid. The resin tag length in both the 6 MPa and the 18 Mpa samples was identical. The only observed dtfference was that the 18 MFb samples showed formation of a win impregnated zone -a hybrid layer. The 6 MPa sample showed no hybrid layer (Fig. 3).The ferric
I
Elchlnp
Resln
u Generally accepted concept
m
Resin
C Hybrd
Hybrld concept
proms a.
Traclimnd envnel bonding theory postulates that resin tags mecbnkaIiy lock into undercuts in the etched e n d . The hybrid conceptproposesthat bondingstability will incnase stgnlacantty if the rrsfn tags arc supplemented by a thin u 3 of~resin-ninfomdtissue & y M d layer).
B
A
135
