Effectsof high-speedcutting on dentinpermeabilityand bonding J. Tagami 1 L. Tad D.H. Pashley z• H. Hosoda~ H. Sano 1 tDepartment of Operative Dentistry Tokyo Medical and Dental University 1-5-45 Bunkyo-ku, Yushima Tokyo 113, Japan 2Department of Oral Biology/Physiology School of Dentistry Medical College of Georgia Augusta, GA 30912-1129
Received June 13, 1990 Accepted June 10, 1991 *To whom correspondence should be addressed This investigationwas supported, in part, by DE06427 from the NIDR and by the Medical College of Georgia Dental Research Center. Dent Mater 7:234-239, October, 1991
Abstract--The effects of high-speed cutting by use of a diamond bur with or without water coolantor sanding by 80-grit SiC paper on dentin permeability, before and after surface treatment, and dentin bonding of adhesive resins were compared. Three different bonding systems were used: Scotchbond DC, which requires no removal of smear layers, and ~o others, Clearfil Photobond and Superbond C&B, both of which remove smear layers (phosphoric acid gel or 10% citric acid containing3% ferric chloride, respectively). Creation of smear layers by bur cutting or sanding reduced dentin permeabilityto levels that were only 13% of the maximum permeabilityvalues. Scotchbond DC gave low but consistent bond strengths (3.7-6.1 MPa) to dentin covered with smear layers. Clearfil Photobond also produced consistentbond strengths (8.6-9.4 MPa). The increase in the permeability of dentin after phosphoricacid treatment was higherwhen the SiC paper was used (146%) than when the high-speed bur was used (87-90%). The smear layer and smear plugs produced by the diamond bur were more resistantto 10-3 treatment than were the SiC-created smear layers. The bond strengths of Superbond showed the highest bond strengths to the conditioned dentin when the high-speedcuttingwas used with water coolant (16.3 MPa), compared with the other two groups (12.212.5 MPa).
mear layers are created by the cutting of tooth substance with any dental instrument and are consid•to be a very important barrier to the diffusion of oral fluids and restorative materials to the pulp, because the smear layers occlude the dentinal tubules (Pashley et al., 1984). On the other hand, some investigators (Bowen, 1978; Br~_nstrSmet al. , 1979;Fusayamaet al. , 1979; Nakabayashi et al., 1982; Hosoda et al., 1989) recommend removal of the smear layers not only on enamel but also on dentin, since they interfere with bonding between adhesive resin materials and both dentin and enamel. Acidic solutions or chelating agents in newly developed dentin bonding systems were confirmed to remove smear layers, thus increasing dentin permeability (Tagami et al., 1990a). The quantity and quality of smear layers vary widely, depending upon whether the dentin is cut wet or dry, and the type of instrument employed (Gilboe et al., 1980). Generally speaking, a thicker layer of debris is created when cutting is done without water spray than with copious air-water spray. Furthermore, coarse diamond burs tend to produce thicker smear layers than carbide burs (Br~mstrSm et al., 1979; Shortall, 1981). A coarse diamond blade mounted on a metallurgical saw also creates thicker smear layers and tends to pack and burnish the debris into a smooth, highly glossy finish (Pashleyet al., 1981). Acid resistance of the smear layers may also be differentwhen the cutting method is varied. In most of the studies on the dentin bonding of resin materials, silicon carbide abrasive papers are used to prepare the dentinal surface, whereas different cutting instruments are routinely used in the clinic. The purpose of this study was to compare the effects of high-speed cutting by use of a diamond bur with or without water coolant and 80-grit SiC abrasive paper on dentin permeability after various surface treatments and dentin bonding of some adhesive resins to treated and untreated surfaces.
MATERIALS AND METHODS
Unerupted, human third molars were
234 Tagami et al./ Effects of high-speed bur on dentin bonding
used within one month after extraction. They were stored in isotonic saline containing 0.2% sodium azide as a preservative to inhibit microbial growth. The methods used to prepare the crown segments were the same as those described previously (Pashley et a/.,1988a). That is, the occlusal enamel was removed when the dentin was cut with a lowspeed diamond saw (Isomet, Buehler, Ltd., Evanston, IL) near the DEJ, so that the superficial dentin would be exposed. All dentin surfaces were prepared within 0.7 mm of the DEJ. The roots were removed by a parallel cut adjacent to the cemento-enamel junction. The resulting crown segments were glued to pieces of Plexiglas (measuring 2 x 2 x 0.7 cm) with Histacryl Blue (B. Melsungen, Germany). A hole was drilled through the plastic, and an 18gauge stainless steel tube was inserted through the hole to the pulp chamber of the crown segments to permit measurement of dentin permeability as a hydraulic conductance. Dentin Permeability.--The smear layer created by the diamond saw was removed by use of 0.5 M EDTA (pH 7.4) for two rain, so that the permeability of each specimen would be maximized. The hydraulic conductance (Lp) of the smearlayer-free dentin surface was then measured by means of the method developed by Pashley et al. (1988a). The hydraulic conductanceofthe tooth was determined, and this value in each tooth was designated as 100% and used as a baseline for comparison with subsequent changes. In this way, each tooth served as its own control. Smear layers were created on dentin by use of: (1) three passes across the dentin surface at approximately 5 mm/s with a diamond bur (Two striper, Cat. #703-8F, Abrasive Technology, Inc., Premier Dental Products, Norristown, PA) rotating between 2-3 x 105 rpm in a high-speed handpiece with water spray; (2) the same as described above but without water spray; and (3) five strokes of 15-cm length on 80-grit SiC paper under tap water with hand pressure. Eighty-grit SiC was chosen since it produced a surface irregularity of den-
tin which was the most similar to that prepared by the diamond bur when observed with the naked eye. The dentin permeability of each specimen was then re-measured so that the effects of each type of smear layer on the hydraulic conductance ofdentin could be evaluated. These values were expressed as a percent ofthe maximum (EDTA-treated) values. The smear layers were then treated with one of two solutions: 37% phosphoric acid gel (Kuraray, Osaka, Japan) for one rain or 10% citric acid containing 3% ferric chloride (10-3 solutions) for 30 s (Sun Medical Co., Kyoto, Japan). The permeability of each specimen was remeasured, and the observed changes were expressed as a percent of the maximum values (i.e., EDTA treatment). There were ten specimens in each group. Bond Testing.--Flat superficial occlusal dentin surfaces located just below the DEJ were prepared as described above. Smear layers were created on these dentin surfaces by use of the same three methods as described above. There were ten specimens in each group. A nylon cylinder (Small Parts RSN4/2, Miami, FL, USA), 3 mm high with an ID of 3 mm, was positioned on the center of the prepared surface, and held in place with Scotch tape as a matrix retainer. Three kinds of resin material were used for the bonding test, according to the manufacturer's instructions: (1) Scotchbond DC (3M Dental Products, St. Paul, MN, USA) was spread on the tooth surface inside the nylon cylinder by means of a small brush, and then light-cured for 10 s (Visilux, 3M), after which two layers of Silux composite material (3M) were applied sequentially by a plastic condenser and light-cured for 30 s, respectively; (2) Clearfil PhotoBond (Kuraray, Osaka, Japan) was applied and cured similarly as for Scotchbond, but after being etched for one rain with the associated phosphoric acid gel, after which a resin composite, Photo Clearfil A (Kuraray), was placed and cured, exactly as was the Silux; and (3) a thin layer of SuperBond C&B (Sun Medical, Kyoto, Japan) was applied to 10-3treated dentin by means of a brush-on technique to dentin. Self-cured resin composite (Clearfil F II, Kuraray, Osaka, Japan) was bulk-filled into the nylon tube over the Superbond C&B. After the materials were allowed to polymerize for five min, the tape was
Fig. I. Surfaceappearanceof dentincut with a high.speeddiamondbur with copiousair-water spray. 150x.
Fig. 2. Surfaceappearanceof dentincut with a high-speeddiamond bur operated withoutwater spray. 150x.
removed, and the specimens were stored in distilled water at 37°C for 24 h. Shear force was applied by lead shot dropped into a container connected to the base of the nylon cylinder by a thin wire cable at a constant rate of 829 g/min as previously reported (Tao et al., 1988). The results obtained with this testing apparatus were not statistically significantly different from those obtained in an Instron
machine operated at a cross-head speed of 0.5 mm/min. Statistics.--The means and standard errors of the means were calculated for each group for both dentin permeability and shear bond strengths. The results were compared by a one-way analysis of variance and Duncan's multiple range test at the 5% level of confidence.
Dental Materials~October 1991 235
levels that were only 1-3% of the maximum values. Dentin permeability increased after the smear-layer-covered surfaces were treated with either the phosphoric-acid-etching gel or 10-3 solution. However, the percenta~gechanges in permeability were much greater for the etching gel (87-146%) than the 10-3 solution(35-60%), despite the differences in cutting method. Dentin prepared with 80-grit SiC paper showed significantly larger (p < 0.05) percent changes ofpermeability after the treatments with acidic solutions, compared with dentin prepared by use of the high-speed diamond bur with or without water. Bonding Test.--Shear bond strengths of
F/g. 3. Surface appearance of dentin prepared with 80-grit SiC abrasive paper. 150x.
the adhesive resin materials are shown in Table 2. Bond strengths ofScotchbond DC were not significantly different among the three cutting methods, although the mean values varied from 3.7 to 6.1 MPa. The bond strengths of Clearfil Photobond were statistically significantly higher (p < 0.05) than those of Scotchbond (4-6 MPa), but there was no significant difference in the degrees of shear bond strength of Clearfil Photobond among the three cutting methods. SuperBond C&B produced the highest shear bond strength among the three materials (p < 0.01). Dentin surfaces prepared with the diamond bur cooled with water produced significantlyhigher (i.e., 16.3 MPa, Table 2) bond strengths (p < 0.05) than those produced by the diamond bur without water or 80-grit SiC paper. S E M Observation.--The permeability
F/g. 4. Appearance of dentin prepared with diamond bur with water-spray after being acid-etched with 37% phosphoric acid for one min. Note linear zones of dentin where tubule orifices remained occluded. 1000x.
S E M Observation.----Scanning electron
microscopy (JA-840, JEOL Ltd., Tokyo, Japan) was used for examination of the dentin surfaces prepared by the various cutting methods and surface treatments applied in the permeability and bonding studies. All specimens were gold-coated and observed at magnifications varying from x150 to x3000.
Dentin Permeability.--The permeabil-
ity of the dentin treated with EDTA was defined as 100% (not shown). Re-measurement of the permeability after creation of smearlayers yielded the results shown in Table 1. Smear layers produced with a highspeed diamondbur with or without water coolant and 80-grit SiC paper dramatically decreased dentin permeability to
2~ Tagami et al. /Effects of high-speed bur on dentin bonding
data were confirmed by SEM observation of the dentin surfaces before and after the various treatments. There were no apparent differences between the SEM appearance of smear layers created by the diamond bur operated with water spray vs. no water spray and smear layers created by 80-grit SiC abrasive paper (Figs. 1-3). Treatment of these surfaces with phosphoric acid removed the smear layer and most of the smear plugs and enlarged the tubule orifices (Figs. 4-6). Even ai~r the acid treatment, there were linear zones or bands of dentin exlu'biting occluded tubules observed in all groups. More of these bands ofoccluded tubules appeared in the specimens cut with the high-speed bur (Fig. 4) than in those abraded with
SiC paper (Fig. 6). Treatment with 10-3 solution removed all of the smear layer but lei~ far more residual smear plugs (Fig. 7) compared with the specimens treated with phosphoric acid (Fig. 5). Residual smear plugs were also observed more frequently in the bur-cut surfaces than in the sanded surfaces (compare Figs. 8 and 9). The bands of occluded tubules along the abrasive scratches appeared the same in those specimens treated with 10-3 as in the cases treated with phosphoric acid.
H',?;O, i ,.,,Ji !,hbu t, (
r , c:
Most of the latest generation of adhesive resin materials require removal of the smear layer on dentin by use of acidic solutions or chelating agent in order for bond strength to be maximized. However, removal of the smear layer and the underlying smear plugs increases dentin permeability, thereby jeopardizing the pulp if the bonding is not perfect or if the bonds fail. Ideally, a dentin conditioner should remove only the smear layer but not the smear plugs (Pashley, 1990). The solubility of the smearlayer to pre-treatment solutions should be different depending upon the cutting method, since the smear layer thickness and particle size vary (Pashley et al., 1988b). The smear layers created by diamond burs with or without water spray were very similar in appearance to those createdby 80-grit SiC paper (Figs. 1-3). That observation tended to validate the use of 80-grit SiC'paper. All smear layers were observed to decrease dentin permeability. However, the smear layer produced by use of 80-grit SiC abrasive paper was more easily removed by the acidic solutions than that produced by a high-speed diamond bur with or without water spray. The observation that these samples had a mean dentin permeability that was 146% of that of the EDTAtreated surfaces was due to the fact that all specimens were sanded thinner when they were prepared with a new smear layer. The fact that the bur-cut specimen permeabilities were only 87-90% of the EDTA value indicates that they remained partially occluded with respect to the sanded specimens. Even on the same surface of a tooth, adjacent areas with open and occluded tubules were observed after surface treatment with acidic solutions. This may have been due to the fact that smear layers in the area of occluded tubules were more acid-resistant than
10Kt, I , ',:'::1,, 000'
Fig. 5. SEM appearanceof dentinsurface preparedwith diamondbur withoutwater sprayafter beingetchedwith 37% phosphoricacid (one min). Manytubulescontain remnantsof smear plugs. Linearband of de~tin near the bottom of the picturecontainswhat appearsto be occludedtubules, lOOOx.
Fig. 6. SEM appearanceof dentinwhichhad beensandedwith 80-gdt SIC abrasivepaperand thenetchedwith 37% phosphoricacid for one min. More of the tubulesremainopen, and the tubulesare larger in diameter, lO00x,
those in the area of open tubules. Bands of occluded tubules were seen along the scratches produced by both cutting methods (Figs. 4-6), although they were more frequently seen in specimens prepared with burs. This might have been caused by the differenceofsmearing force with individual diamond or abrasive particles. Treatment of smear layers with phosphoric acid seemedto maximize
permeability and remove most traces of smear layers in all three groups. The diameters of the tubules in 10-3-treated 80-grit SiC-treated surfaces were somewhat larger than those in the other groups. Phosphoric acid treatment has been shown to maximize permeabili~ when compared with many other pretreatment solutions for dentin bonding (Tagami et al., 1990a).
Dental 'Mateviala /October 19917 . l ~ i
Fig. 7. SEM appearance of dentin prepared with a diamond bur with air.water spray, then acid-etched with 10-3 solution. This treatment led to removal of most of the smear layer. There was less loss of peritubular dentin in this group than in the others. 3000x.
Fig. 8. SEM appearance of dentin prepared by use of a diamond bur without water, after being acid-etched with 103 solution. Althoughthere was an almost complete removal of the smear layer and a good deal of loss of peritubular dentin, many of the tubules remain occluded with residual smear plug debris. 3000x.
Although the bond strengths of Scotchbond to smear layers created with 80-grit SiC were lower than those made to bur-created smear layers, a statistically significant differencewas not found. This confirms the similar bond strengths of Scotchbond DC to bur- and SiCpaper-created smear layers reported by Tao et al. (1988).
No significant differences among the shear bond strengths within the Clearfil Photobond group were found in this study. It appeared that differences in smear layers were not important when phosphoric acid was used. Clearfil Photobond was developed as both an enamel and dentin bonding agent and is applied to all cavity walls treated with
238 Tagami et al./ Effects of high-speed bur on dentin bonding
phosphoric acid in clinical practice. It was confirmed that this bonding agent produced consistent bond strengths regardless of how the smear layers were created (Table 2). Some dentin bonding systems are more sensitive to dentin wetness than others (Prati et al., 1991). Scotchbond DC is relatively insensitive to dentin wetness, because it is designed to be placed directly on a smear layer. The smear layer/smear plugs occlude the underlying dentin and permit the surface to be drier than would be possible if the smear layer/plugs were removed. Those bonding systems that remove the smear layer and use HEMA to prime the dentin surface tend to be more sensitive to dentin wetness and produce low bond strengths in deep dentin, which is wetter than superficial dentin. Clearfil Photobond is relatively insensitive to changes in dentin wetness (Pashley, unpublished observations), as is Superbond (Tao et al., 1991). Sanded specimens treated with 10-3 solution were twice as permeable as those cut with burs, yet the bond strengths of Clearfil Photobond and SuperBond to sanded and cut dentin were similar. The fact that the highest Superbond bond strengths were found with dentin prepared by means of a diamond bur with water spray is encouraging, because it closely simulates the conditions of clinical practice. Although there were no SEM or permeability differences between dentin prepared with the diamond bur in the presence and that prepared in the absence of water spray, there may be, in the resulting dentin surface, subtle differences that are, at present, undetectable and responsible for the higher bond strengths obtained with SuperBond. Smear plugs created by the 80-grit SiC abrasive paper seemed to be more acidlabile to both phosphoric and 10-3 acidic solutions than were those created with burs. Perhaps they are shorter. Such speculations can be tested only by further research. ACKNOWLEDGMENT
The authors are grateful to Shirley Johnston for her excellent secretarial support. REFERENCES
BRANNSTROM,M.; GLANTZ,P.-O.; and NORDENVALL, H.-J. (1979): The Effect of Some Cleaning Solutions on
the Morphology of Dentin Prepared in Different Ways, An in vivo Study, J Dent Child 46:19-23. BOWEN,R.L. (1978): Adhesive Bonding of Various Materials to Hard Tooth Tissue---Solubility of Dental Smear Layer in Dilute Acid Buffers, Int Dent J 28:97-107. BOYER,D.B. and SVARE,C.W. (1981): The Effect of Rotary Instrumentation on the Permeability ofDentin, JDent Res 60:966-971. FUSAYAMA,T.; NAKAMURA,M.; KUROSAKI, N.; and IWAKU,M. (1979): Non-pressure Adhesion of a New Adhesive Restorative Resin, J Dent Res 58:1364-1370. GILBOE,D.B.; SVARE,C.W.;THAYER,H.E.; and DRENNON,D.G. (1980): Dentinal Smearing: An Investigation of the Phenomenon, J Prosthet Dent 44:310316. HOSODA,H.; HIRASAWA,K.; and FUJITANI, M. (1989): New Tooth Surface Conditioning Treatment for Adhesive Composite Restorations, Jpn J Conserv Dent 32:421-433. NAKABAYASHI, N.; KOJIMA, K.; and MASUHARA,E. (1982): The Promotion of Adhesion by the Infiltration of Monomers into Tooth Substrates, J Biorned Mater Res 16:265-273. PASHLEY,D.H. (1990): The Interactions of Dental Materials with Dentin, Trans Acad Dent Mater 3:55-73. PASHLEY,D.H.; DERKSON,G.D.; TAO,L.; DERKSON, M.; and KALATHOOR,S. (1988a): The Effects of a Multi-Step Dentin Bonding System on Dentin Permeability, Dent Mater 4:60-63. PASHLEY,D.H.; MICHELICH,V.; and KEHL, T. (1981): Dentin Permeability: Effects of Smear Layer Removal, J Prosthet Dent 46:531 -537. PASHLEY,D.H.; TAO,L.; BOYD,L.; KING, G.E.; and HORNER,J.A. (1988b): Scanning Electron Microscopy of the Substructure of Smear Layers in Human Dentine, Arch Oral Biol 33:265-270. PASHLEY,E.L.; TAO,L.; DERKSON,G.; and PASHLEY,D.H. (1989): Dentin Permeability and Bond Strengths After Various Surface Treatments, Dent Mater 5:375-378. PRATI,C.; Erickson, R.; TAO,G.; SIMPSON, G.; and PASHLEY,D.H. (1991): Measurement ofDentin Permeability and Wetness by Use of the Periotron Device, Dent Mater 7:269-274. SHORTALL,A.C. (1981): Cavity Cleansers in Restorative Dentistry, Br Dent J
Fig. 9. SEM appearance of dentin surface prepared by use of 80-grit SiC abrasive paper with water, The surface was subsequently treated with 10-3 solution, which led to a loss of smear layer, peritubular dentin, and smear plugs. 3000x.
TABLE 1 PERCENT CHANGES OF DENTIN PERMEABILITY AFTER VARIOUS CUTTINGS AND TREATMENTS WITH PHOSPHORIC ACID OR 10-3 SOLUTION (X+ S--EM) Cutting Method
High-speed diamond bur with water
1.7 + 0.3
86.9 + 6.6
35.8 + 3.6 I
High-speed diamond bur without water
1.0 + 0.3
90.3 + 6.4
35.1 + 8.2
80-grit silicon carbide paper
3.0 + 0.3
146.2 + 9.8
60.1 + 4.7
Groups connected by the same vertical line are not significantlydifferent at p = 0.05. N = 10 in each group. TABLE 2
SHEAR BOND STRENGTHS (MPa) OF RESIN MATERIALSTO DENTIN AFTER VARIOUS CUTTINGS AND TREATMENT WITH THE ASSOCIATED ACIDIC SOLUTIONS (-X.t_SEM)
High-speed diamond bur with water
6.1 + 0.8
9.1 + 1.1
16.3 + 0.4
High-speed diamond bur without water
5.1 + 0.5
8.6 + 0.6
12.2 + 0.8
80-grit silicon carbide paper
3.7 + 0.5
9.4 + 0.9
12.5 + 0.7 I
Groups connected with the same line are not significantly different at p = 0.05. N = 10 in each group.
TAGAMI,J.; SUGIZAKI,J.; and HOSODA,H. (1990a): Effect of Various Pretreatments for Dentin Bonding on Dentin Permeability, Jpn J Dent Mater 9:240-246. TAGAMI,J.; TAO, L.; and PASHLEY,D.H. (1990b): Correlation Among Dentin Depth, Permeability, and Bond Strength of Adhesive Resins, Dent Mater 6:45-50.
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. TAO, L.; TAGAMI,J.; and PASHLEY,D.H. (1991): Pulpal Pressure and Bond Strengths of Superbond and Gluma, Am J Dent 4:73-76.
Dental Materials/October 1991 239