Shear bond strength of Scotchbond in vivo R.S. McGuckinI L. Tao2 W.O. Thompson3 D.H. Pashley21•

~Department of Prosthodontics School of Dentistry University of Texas Dental Branch Houston, Texas 2Department of Oral Biology School of Dentistry 3Research Computing and Statistics Medical College of Georgia Augusta, Georgia 30912-1129 Received January 31, 1990 Accepted June 13, 1990 fCorresponding author This investigationwas supported, in part, by NIDR grant DE06427, by 3M Dental Products, by the Medical College of Georgia Dental Research Center, and by an American Fund for Dental Health Teacher Training Fellowship to R.S.M. Dent Mater 7:50-53, January, 1991 Abslracl-The shear bond strengths of Scotchbond, HEMA/Scotchbond, and Scotchbond 2 were measured in vivo in dog canine and molar enamel and dentin. Dentin bond strengths were compared in superficial, middle, and deep dentin. The acid-etched enamel bond strengths of the three bonding systems ranged from 10 to 11 MPa and were not statistically different. Scotchbond/Siluxbonds to superficial and middle cuspid and molar dentin were 3 MPa and were not statistically different. HEMA-treateddentin did not consistently improve Scotchbond strengths to either tooth type at any dentin depth. Deep dentin from either tooth type yielded significantly lower bond strengths. Scotchbond 2/Silux shear bond strengths were significantly higher (6-8 MPa) in superficial and middle cuspid dentin but were not different from Scotchbond bonds made to deep cuspid dentin or to any depth of molar dentin. The observation that molar bond strengths are lower than those made to cuspid dentin indicates that there are important substrate differences between teeth as well as within dentin as a function of depth. The dog model may be useful for the screening of new dentin bonding systems prior to clinical trials.

lthough much has been learned about the bonding of adhesive resins to dentin through the use of in vitro laboratory testing (Komatsu and Finger, 1986; Retief et al., 1988), the methodology can be criticized on several grounds. In vitro studies generally employed extracted teeth of unknown age, stored under a wide variety of conditions which are often unphysiologic. The teeth are then embedded in plastic or plaster, which tends to dry the dentin and pulp chamber, thereby leading to bond strengths higher than would occur to wetter dentin (Kiyomura, 1987; Mitchem and Terkla, 1988). These teeth are then ground flat unknown distances into the dentin by use of abrasive wheels and papers that are seldom used in clinical dentistry (Pashley et al., 1988). Several reports have shown that bonds made to deep dentin are lower than those made to more superficial dentin (Causton, 1984; Stanford et al., 1985; Mitchem and Gronas, 1986; Suzuki and Finger, 1988; Tagami et al., 1990). Thus, the high variance reported by most investigators may be due, in part, to variations in dentin thickness. Most of these criticisms can be avoided if dentin bonding studies are conducted in vivo with clinically-appropriate burs and handpieces and careful measurements of dentin thickness. Microscopically, the density and diameter of dentinal tubules of dog and human teeth have been reported to be similar (Forssell-Ahlberg et al., 1975). Dog dentin has previously been used for measurement of in vivo bond strengths of adhesive resins (Pashley et al., 1988). The purpose of this study was to compare the in vivo bond strengths of Scotchbond, HEMA/Scotchbond, and Scotchbond 2 on dog cuspids and molars as a function of dentin depth.



The use of dogs for these experiments was approved by our institu-


tional Committee on Animal Care and Use, under guidelines approved by the AAALAC. Six mongrel dogs of either gender, each weighing between 20 and 28 kg, were anesthetized with pentabarbital (30 mg/kg, iv). The dogs were artificially ventilated and their body temperature, blood pressure, and heart rate monitored at all times. Bonding was done on the buccal and lingual surfaces of maxillary and mandibular cuspids and on buccal and lingual surfaces of mandibular first molars. All three bonding systems were used in each dog. The bonding sites were prepared by use of a high-speed handpiece and fine-grit diamond burs from the same manufacturing lot (Two Striper 780.9F, Premier, Norristown, PA). The buccolingual width of each tooth was measured with a digital micrometer (Sylvae Ultra-Cal II, Fowler, Inc., Newton, MA) after a reference ledge was cut at the gingival third in the side of the tooth opposite the pre-selected bonding site. The enamel was then flattened until a 3-4-mm-diameter bonding area was available. The buccolingual thickness of the tooth was then remeasured, the site etched for 30 s with 37% phosphoric acid gel (Scotchgel, 3M Dental Products, St. Paul, MN), rinsed with water for 10 s, dried with an air blast for 10 s, and then bonded with one of the bonding agents followed by Sflux composite (3M Dental Products, St. Paul, MN), which was placed into a 3-mm I.D. x 3-mm cylindrical nylon matrix (RSN 2/2, Small Parts, Inc., Miami, FL) in two 1.5-mm increments, which were light-cured for 30 s each by means of a Visilux 2 curing light (3M Dental Products, St. Paul, MN). The nylon matrix material did not interact with either the bonding agents or the resin composite. Thirty minutes later, the enamel bonds were shear-stressed to failure by use of a 0.6-ram-diameter wire looped around the base of the nylon matrix, as previously described (Pashley et al.,

1988). The t e e t h w e r e not subm e r g e d in w a t e r prior to being tested, although little water sorption would have occurred within 30 min. Each dog's maxilla was firmly fLxed to the surgical table with 3-cmwide nylon straps. Rubber bite blocks stabilized the mandible to the maxillae. Another set of nylon straps fixed the mandible to the table. This provided a very effective but simple method of stabilizing the teeth for shear testing. After the enamel bonds were tested, the teeth were further reduced in a fiat plane parallel to the long axis of the pulp chamber, by use of a fine-grit diamond bur. As soon as the D E J was reached, the buccolingual thickness of the tooth was re-measured. Further reduction of dentin was accomplished until an area 3-4 mm in diameter was available for bonding. The tooth thickness was measured again and the bonding done as outlined above, according to the manufacturers' recommendations. Thirty minutes later, the shear bond s t r e n g t h was m e a s u r e d and the processes repeated on deeper dentin surfaces at least two more times. To ensure that each reduction in tooth depth was uniform, the surface was scored with a 0.3-mm depth-limiting diamond bur used in porcelain veneer preparations (LVS-2, Brasseler, Savannah, GA). The score lines were then removed with the fine-grit diamond bur, yielding reductions that averaged 0.5 mm. The same dentin bonding materials were used sequentially on the same teeth as the remaining dentin thickness was reduced. After the dog was killed, the teeth were removed, and the dentin thickness remaining between the last bonding site and the pulp chamber adjacent to the bonding site was measured with a pincer-type micrometer to facilitate calculation of the actual dentin thicknesses at each of the sequentially bonded surfaces. Dentin thicknesses were expressed in both absolute (i.e., ram) and relative (i.e., as a percent of the total dentin thickness) terms. Superficial dentin was defined as 83.4 _+ 7.9% _+ SEM) of the total dentin thickhess. Middle dentin was defined as 48.3 _ 9.2% of the total dentin thickness, and deep dentin was de-

fined as the inner 16.7 -+ 8.9% of the total dentin.

Statistics. - T h e following variables were evaluated: maxillary vs. mandibular teeth, right vs. left sides, buccal vs. lingual surfaces, cuspids vs. molars, superficial vs. middle and deep dentin, and Scotchbond LC vs. 35% H E M A / S c o t c h b o n d LC v s . Scotchbond 2. After the data were expressed in MPa, the means and standard error of the mean were calculated, after which the data were subjected to a multifactorial analysis of variance. Statistical significance was set at p ~< 0.05. P o s t hoc testing was done with Tukey's Multiple Range Test. Materials. - T h e materials tested are listed in Table 1. Light-cured (LC) Scotchbond was used directly on smear layers according to the manufacturer's instructions. The HEMA/ Scotchbond combination was also used directly on smear layers, as previously described (Pashley et al., 1988) for GLUMA/Scotchbond. It was painted on the smear layers with a sable brush, allowed to react for 60 s, and then air-dried for 10 s. The HEMA-treated surfaces were then immediately covered with a layer of LC Scotchbond, light-cured for 10 s, covered with Silux (which was lightcured for 30 s), and followed by a second increment of Silux, which was also cured for an additional 30 s. Scotchbond 2 was used according to the manufacturer's instructions. All dentin bonding systems were covered with Silux. RESULTS Table 2 s u m m a r i z e s the bond strengths obtained on cuspid and

molar enamel and the three levels of dentin with the three different bonding systems. The acid-etched enamel bonds ranged between 9.8 and 11.4 MPa (Table 2), and there were no statistically significant differences between the materials (p = 0.13) or between cuspid and molar enamel (p = 0.27, not shown). When S c o t c h b o n d / S i l u x was bonded to smear layers, there was no statistically sign~cant difference between the shear bond strengths of cuspids vs. molars at any given depth (Table 2A). One exception was the statistically significant lower bond strength of deep molar dentin (p < 0.05). Although Scotchbond bonds to cuspids and molars fell as the remaining dentin thickness decreased, the values were only statistically different (Table 2A) for deep molar dentin. Pre-treatment of the dentin with 35% HEMA increased the shear bond strengths to superficial cuspid dentin slightly, but the results were not statistically significant (Table 2B). It tended to lower the bond strengths to the other dentin substrates, but not significantly. Scotchbond 2 bond strengths were higher than the other two bonding systems at each depth in superficial and middle dentin in cuspids (p < 0.05) but not in molars (Table 2B). The Scotchbond 2 molar bond strength values were not statistically different from those of the other two bonding systems, although they were significantly higher in superficial and middle dentin of cuspids. DISCUSSION Generally speaking, the bonds made to smear layers with either Scotchbond or HEMA/Scotchbond were

TABLE1 BONDINGMATERIALS Part Numbers and Lot Numbers Material Part Number (1) Scotchbond Scotchbond LC* Resin 7533R Liquid 7533L (2) Scotchbond 2 Scotchprep* 7502P Light Cure Dental Adhesive* 7502A Scotchgel* 7430C Silux* 5502U (3) HEMA/Scotchbond Polyscience, Warrenton, PA 18975. Lot # 0755 used as 35% HEMA in water. * = 3M Dental Products Division, St. Paul, MN.

Lot 6J 6C 8AT 6AH 8CR26 8F2

Dental Materials~January 1991 51


A: Analysis of Dentin Depth R ± SEM (N) Dentin Depths* Scotchbond HEMA/Scotchbond Scotchbond 2 Superficial 3.2 ± 0.4 (15) I 5.4 ± 0.9 (15) 1 8.2 ± 0.9 (16) I Middle 3.6 -+- 0.7 (15) 3.7 ± 0.5 (17) I I 6.2 ± 0.8 (15) J Deep 2.9 ± 0.5 (18) ,, 2.2 ± 0.5 (17) 1 3.3 ± 0.7 (16) Superficial 3.4 ± 0.4 (7) 2.8 ± 0.9 (8) 3.3 ± 1.1 (9) Middle 3.2 __ 1.0 (6) 0.8 ± 0.3 (8) 3.0 ± 1.1 (8) Deep 1.3 ± 0.4 (8) 1.6 ± 0.4 (8) II 0.9 ± 0.3 (7) Enamel 10.8 ± 1.0 (11) 11.4 ___ 1.1 (12) 9.8 ± 3.1 (10) Groups joined by vertical lines in the same plane are not statistically different at p > 0.05. *Superficial dentin = 83.4 ± 7.9% of original dentin thickness. Middle dentin = 48.3 ___9.2% of original dentin thickness, and deep dentin = 16.7 __ 8.9% of original, total dentin thickness.


B: Analysis of Bonding Agents Dentin Depths Superficial Middle Deep Superficial Middle Deep Enamel

Scotchbond ....3.2 _ 0.4 (15) 3.6 __ 0.7 (15) 2.9 -- 0.5 (16)

HEMA/Scotchbond 5.4 ± 0.9 (15) 3.7 ± 0.5 (17) 2.2 ± 0.5 (17)

E --. SEM (N) Scotchbond 2 8.2 --_ 0.9 (16) 6.2 __ 0.8 (15) 3.3 --. 0.7 (16)

3.4 ± 0.4 (7) 3.2 ± 1.0 (6) 1.3 __ 0.4 (8)

2.8 __ 0.9 (18) 0.8 __ 0.3 (8) 1.6 ± 0.4 (8)

3.3 -+ 1.1 (9) 3.0 -+ 1.1 (8) 0.9 _+ 0.3 (7)

10.8 __ 1.0 (11)

11.4 ± 1.1 (12)

9.8 _+_3.1 (10)

Groups joined by horizontal lines in the same plane are not statistically different at p < 0.05.

lower than those made to dentin surfaces treated for removal of smear layers (/~e., Scotchbond 2). That was expected from previous studies (see Tao et al., 1988; Tao and Pashley, 1988) as well as from other literature reports. The results were complex because of differences between cuspids and molars and as a function of dentin depth. The Scotchbond bond strengths obtained in the present study on superficial and middle dentin of both cuspids and molars were higher than those we previously reported in vivo (Pashley et al., 1988). In that report, we combined both cuspid and molar data because at that time, with that variance, they were not different. In the p r e s e n t report, the bond strengths to deep dentin of molars were lower than the combined mean values that we previously reported. We had expected that HEMA/ Scotchbond would increase the shear bond strength of Scotchbond/Silux 34 times in a manner analogous to the increase in bond strengths produced by Gluma/Scotchbond (Pashley et al., 1988). This did not occur (Table 2), suggesting that the 5% glutaraldehyde in Gluma may be more important than the 35% HEMA that was

common to both the Gluma used in our previous study and the HEMA used in this study. The maximum value for Scotchbond 2, which was obtained on superficial dentin of cuspids, was only 8.2 MPa. This is much lower (i.e., from one-third to one-half) than the bond strengths obtained by other investigators using human or bovine teeth in vitro. Even the maximum Scotchbond bond strengths that we obtained in vivo (3.6 MPa for this study and 2.2 MPa in our previous in vivo study, respectively) are much lower than the 5-6-MPa values that we obtained using human molars in vitro (Tao et al., 1988; Tao and Pashley, 1988). It may be that the in vivo system produces lower bond strengths because of wetter dentin or self-contamination of its dentin with plasma proteins in dentinal fluid. Alternatively, dentin in vivo may be "contaminated" with pieces of cut odontoblast processes. We have previously reported that dentin becomes more permeable as dentin thickness is reduced (Outhwaite et al., 1976; Reeder et ~ , 1978; Fogel et al., 1988). We also reported that dog molar dentin is almost 3.5 times more permeable than cuspid


dentin (Pashley et al., 1981). The lower bond strengths obtained with molars may be due to differences in surface wetness. Thus, when the smear layers are removed, molar dentin becomes wetter than cuspid dentin as pulpal fluid slowly seeps to the surface. The enamel bonds obtained on dog teeth were lower than we expected from our previous studies of enamel bond strengths in human teeth (Tao et al., 1988). We have no explanation for that result. No previous reports are available on the bond strength of resins to dog enamel. Jemt et al. (1986) reported that the tensile strengths of polycarboxylate cement and glass-ionomer cements to human enamel were lower in vivo than in vitro. I n vivo models for bonding studies are becoming more popular. Tyler et al. (1987) reported lower in vivo tensile bond strengths of glass-ionomer cements to monkey dentin than when the studies were repeated in vitro on the same teeth. Gray and Burgess (1989) reported that there were no differences in the bond strength of Scotchbond 2 or Gluma made to goat incisor dentin in vivo vs. in vitro. However, they obtained higher

shear bond strengths on the incisal halves of the incisors relative to the gingival halves. The fact that we observed that molar bond strengths were lower than cuspid bond strengths indicates that there are important substrate differences that must be considered in the evaluation of dentin bonding systems. Differences apparently exist between teeth and even within teeth. The dog model may be useful for the screening of new dentin bonding systems prior to clinical trials.

ACKNOWLEDGMENTS The authors wish to thank Shirley Johnston for secretarial support and Dr. Robert Erickson, 3M Dental Products, for demonstrating the proper bonding techniques for Scotchbond 2. REFERENCES CAUSTON,B. (1984): Improved Bonding of Composite Restoratives to Dentin, Br Dent J 156:93-95.

FOGEL, H.M.; MARSHALL, F.J.; and PASHLEY, D.H. (1988): Effects of Distance from the Pulp and Thickness on the Hydraulic Conductance of Human Radicular Dentin, J Dent Res 67:13811385. FORSSELL-AHLBERG, K.; BP~NNSTROM, M.; and EDWALL,L. (1975): The Diameter and Number of Dentinal Tu-

bules in Rat, Cat, Dog and Monkey, Acta Odontol Scand 33:243-250. GRAY,S. and BURGESS,J. (1989): In vivo and in vitro Comparison of Dentin Bonding Agents, J Dent Res (Spec Iss)

68: 375, Abst. No. 1547.

JEMT, W.; ST~LBLAD, P.A.; and 0]LO, G. (1986): Adhesion of Polycarboxylate-based Dental Cements to Enamel: An in vivo Study, J Dent Res 65: 885-887. KrYOMURA,M. (1987): Bonding Strength to Bovine Dentin with 4-META/MMATBB Resin, J Jpn Soc Dent Mater & Devices 6:860-872. KOMATSU, M. and FINGER, W. (1986): Correlation of Early Bond Strength with Marginal Gaps, Dent Mater 2:257262. MITCHEM,J.C. and GRONAS,D.G. (1986): Effects of Time after Extraction and Depth of Dentin on Resin Dentin Adhesives, J A m Dent Assoc 113:285-287. MITCHEM,J.C. and TERKLA,L.G. (1988): Bonding of Resin Dentin Adhesives under Simulated Physiological Conditions, Dent Mater 4:351-353. OUTHWAITE, W.C.; LIVINGSTON, M.J.; and PASHLEY, D.H. (1976): Effect of Changes in Surface Area, Thickness, Temperature and Post-extraction Time on Human Dentine Permeability, Arch Oral Biol 21:599-603. PASHLEY, D.H.; NELSON, n.; and PASHLEY, E.L. (1981): In vivo Fluid Movement Across Dentine in the Dog, Arch Oral Biol 26:707-710. PASHLEY, E.L.; TAO,L.; MACKERT,J.R.; and PASHLEY, D.H. (1988): Comparison of in vivo vs. in vitro Bonding of

Composite Resin to the Dentin of Canine Teeth, J Dent Res 67:467--470. REEDER, 0.W.; WALTON, R.E.; LIVINGSTON, M.J.; and PASHLEY,D.H. (1978): Dentin Permeability: Determinants of Hydraulic Conductance, J Dent Res 57:187-193. RETIEF, D.H.; O'BRIEN, J.A.; SMITH, L.A.; and MARCHMAN,J.I. (1988): In vitro Investigation and Evaluation of Dentin Bonding Agents, A m J Dent (Spec Iss) 1:176-183. STANFORD, J.W.; SABRI, Z.; and JOSE, S. (1985): A Comparison of the Effectiveness of Dentin Bonding Agents, Int Dent J 35:139-144. SUZUKI, T. and FINGER, W.J. (1988): Dental Adhesives: Site of Dentin versus Bonding of Composite Resins, Dent Mater 4:379-383. TAGAMI,J.; TAO, L.; and PASHLEY,D.H. (1990): Correlation among Dentin Depth, Permeability, and Bond Strength of Adhesive Resins, Dent Mater 6:45-50. TAO, L. and PASHLEY,D.H. (1988): Shear Bond Strengths to Dentin: Effects of Surface Treatments, Depth and Position, Dent Mater 4:371-378. TAO, L.; PASHLEY, D.H.; and BOYD, L. (1988): Effect of Different Types of Smear Layers on Dentin and Enamel Shear Bond Strengths, Dent Mater 4: 208-216. TYLER, M.; CHARBENEAU, G.; DENNISON, J.; HEYS, D.; and FITZGERALD,M. (1987): In vivo and in vitro Tensile Bond Strengths of a Glassionomer Cement, J Dent Res (Spec Iss) 66:112, Abst. No. 48.

Dental Materials~January 1991 53

Shear bond strength of Scotchbond in vivo.

The shear bond strengths of Scotchbond, HEMA/Scotchbond, and Scotchbond 2 were measured in vivo in dog canine and molar enamel and dentin. Dentin bond...
460KB Sizes 0 Downloads 0 Views