Effects of Thermocycling on theTensile Bond Strength Between

Resin Cement and DentinSurfaces After Temporary

Cement Application

Elisete Kazumi Watanabe, DOS'Hirofumi Yatani, DDS, PhD''Atsushi Yamashita, DDS,Kunio Ishikawa,Kazuomi Suzuki,

Purpose: In a previous study the authors found tiiat 2 of 3 tooth conditioners examinedwere able to reduce the negative effect of temporary cement on the bond strength betweenresin cement and teeth. The aim of this study was to evaluate the bond durability with theconditioners, as well as their capability for temporary cement dissolution. Materials andMethods: After eliminating the temporary cement with a curette from the bovine dentinsurface, a conditioner (ethyl dihydrogen phosphate, EP; or methacryloxyethyl diiiydrogenphosphate, MEP) was applied to the surface and a resin cement was adhered. Tensile bondstrength measurement, scanning electron microscopic observation, and energy-dispersiveanalysis were carried out without thermocycling and after thermocycling. Results: Thetensile bond strength gradually decreased with an increase in the numher of thermocycles.However, specimens treated with EP or MEP showed significantly higher tensile bondstrength values compared to those that did not receive conditioner application (P< 0.0001);EP showed significantly higher tensile bond strength values compared to the specimens thatdid not receive both temporary cement and conditioner application {P< 0.0001). Thescanning electron microscopic, energy-dispersive, and solubility analyses suggest that bothconditioners have the capability of dissolving temporary cement remnants. Conclusion:Both EP and MEP have potential value as conditioners to reduce the negative effect oftemporary cement on the bond strength between resin cement and dentin and to improvethe bond durability. Int J Prosthodont 1999:12:230-235.

Temporary cementation on dentin surfaces is an im-portant procedure to avoid pain and infection and

to restore function and esthetics.'"^ Although temporarycement has to be removed before tbe subsequent ad-hesion procedure, complete elimination of temporarycement from the dentin surface is diff icult.

^Doctoral Student, Department of Fixed Prosthodontics, OkayamaUniversity Dents! School, Okayama, lapan.''Associate Professor. Department of Fixed Prosthodontics.Okayama University Dentai Schooi, Okayama, Japan.^Professor, Department of Fixed Prosthodontics, OkayamaUniversity Dental School, Okayama, Japan.''Associate Professor, Department of Dental Materials, OkayamaUniversity Dental School, Okayama, Japan.^Professor, Department of Dental Materials, Okayama UniversityDental School, Okayama, Japan.

Reprint requests: Dr Elisete Kaiumi Watanabe, Department ofFixed Prosthodontics, Okayama University Dental Schooi, 2-S-1Shikata, Okayama 700-8525, Japan, Fax: t 81-86-235-6672.e-maii:ekw22l67@dent.okayama-u.ac,ip

Unfortunately, it has been demonstrated that residualtemporary cement on the dentin surface causes a de-crease in bond strength between the dentin and com-posite resin.''"'^ Therefore, many attempts have beenmade to eliminate the remaining temporary cement.Theefficacyofcleaningthe dental surface with pumicehas shown contradictory evidence,^-^-^ and the use ofsoap with pumice presented an inverse result, actuallydecreasing the bond strength.^" Etching twice sequen-tially showed some efficacy," but only slightly im-proved the contact angles of distilled water on thedentin surfaces,̂ ^ The use of a tooth conditioner thatcan eliminate the adverse effects of temporary cementapplication would be more practical for clinical use.

In a previous study, the authors evaluated the bondstrength of a resin cement after dentin conditioningwith 3 types of acids—-ethyl dihydrogen phosphate(EP), methacryloxyethyl dihydrogen phosphate (MEP),and 2-methacryloxyethyl hydrogen maléate (MEM)—to determine whether they could be used as a tooth

The International lourrial of Proslhodontic 2 3 0 Volume 12, Number 3. 1999

Watanabe et al

conditioner with the aim of avoiding the negative ef-fect of residual temporary cement on the bond strengthof resin cement to dentin. It was found that all condi-tioners had a positive effect on the bond strength ofresin cement to dentin.'^ In particular, EP and MEPpresented significantly higher tensile bond strengths,even when the values were compared with speci-mens to which no temporary cement had been ap-plied. When the bond strength values to substratescontaminated by temporary cement were comparedto those of zinc-phosphate cements, the valuesachieved with resin cement were still superior in ten-sion to results generated with zinc-phosphate cementand clean dentin surfaces.'•" However, the bond dura-bility is unknown and should therefore be evaluatedprior to clinical use. A thermocycling test can be usedfor that purpose. ' ' " ' ̂

The aims of this study were: í / ^o investigate theeffects of thermocycling stress on the bond strengthof resin cement to dentin treated with EP or MEP; and(2) to evaluate the capabilit>' of these conditioners fortemporary cement dissolution. The null hypothesis es-tablished was that the tensile bond strengths of spec-imens treated with EP and MEP after removing tem-porary cement with a curette would be the same,even after thermocycling tests, as those of specimensin which EP and MEP had not been applied.

Materials and Methods

Conditioner Preparation

The EP and MEP were obtained commercially (JohokuChemical) and used without further purification. Bothchemicals were dissolved in distilled water with theminimum volume of ethanol necessary to make the so-lution clear, resulting in 10 wt% and 20 wt% solutions.

Tensile Bond Strength Measurement andSEM Observation of Fractured Surfaces

Bovine mandibular incisors, frozen immediately afterextraction, had the radicular portion cut and the pulpremoved. The vestibular area of the coronal portionwas then ground by a diamond disk to expose thedentin. The exposed area was polished flat with wa-terproof 600-grit silicon carbide abrasive paper(Nihon Kenshi) under running water.

The specimens were randomly divided into 4groups (Eig 1]: (1) specimens without application oftemporary cement and conditioner as a control (Cgroup); (•2Jspecimens in which only temporary cementwas applied (TC group); (3) specimens in which tem-porary cement application was followed by EP con-ditioner application (EP group); and f-ij specimens in

Thermocycling and Bond Strength Between Resin Cemehi and Dentjn

emporary cement appticalionRemoval of temporaiy cement niith excavator

ED primer applicatonAdhesion to stainless steel rod ivith Panavia 21

Fig 1 Research design. C = controi group; TC = temporary ce-ment-apptied group: EP = ethyl öihydrogen ptiosphate-treatedgroup after temporary cement appiication; MEP = melhacry-loxyethyl dihydrogen phcjsphate-Irealed group after temporary ce-ment application; EDS = energy-dispersive x-ray spectroscopy;SEM = scanning electron microscopy; TBS = tensile bond strength.

which temporary cement application was followed byMEP conditioner application (MEP group).

For the temporary cement-treated specimens, a10 mm X 5 mm x 2 mm acrylic resin plate (UnifastII, GC] was cemented to the polished dentin surfacewith HY-Bond Temporary Cement Hard (Shofu). Thiscement contains zinc oxide, an aqueous solution ofpolyacr>'lic acid, and HY compound (tannin, ZnE^,SrE,, and ZnO mixture), and is widely used becauseof its hardness.'^'^ After 10 minutes the specimenswere immersed in distilled water at 37°C for 48 hours.The acrylic resin plates were then removed and theremaining temporary cement on the dentin surfacewas removed with a curette.

Among the conditioner-treated specimens, the fol-lowing combinations of conditioner type, concentra-tion, and period of conditioner application were used:í í n O % EP for 30 seconds and I2l 20% MEP for 60seconds. The conditioners' concentration and appli-cation period were determined based on the results ofa previous investigation in which the selected condi-tioners presented the higher bond strengths.'^ Afterconditioner application, specimens in the EP groupwere rinsed with running water for 10 seconds and air

. .2,Number3,T999 231 The InlernaHonal |ouma! of Pro5thodonlic5

Thermocycling and Bond Strength Between Resin Cement and Dentin Watanabe et al

dried for 10 seconds, whereas specimens in the MEPgroup were air dried without water rinsing since thisconditioner presents a methacrylate radical and thuscan be classified as a self-etching primer.

All groups received ED Primer (from the Panavia 21set, Kuraray) application for 60 seconds (as indicatedby the manufacturer), and were then air dried for 10 sec-onds. A 4-mm diameter stainless steel rod (SUS-304)sandblasted with 50-|jm aluminum oxide was ce-mented with Panavia 21 resin cement on the substratesfor tensi Ie bond strength (TBS) measurement. After ce-menting the specimens were stored in distil led water at37°C for 24 hours. Some specimens were submitted tothe TBS test at this point without thermocycl ing. The re-maining specimens were thermocycled 5,000,10,000,and 20,000 times before testing tbe bond strength be-tween 4''C and 60°C at 60-second intervals. The spec-imens were positioned in a jig for TBS measurement;values were obtained with a universal testing unit (DCS-2000, Shimadzu) at a cross-head speed of 2 mm/min.^''A total of 8 specimens were measured in each group.

After TBS measurements the fractured dentin sur-faces of all groups, without thermocyciing and after20,000 thermocycles, were gold coated in an ioncoater (IB-5, Eiko Engineering) and observed with scan-ning electron microscopy (SEM; DS-720, Topcon),

Energy-Dispersive Analysis

The quantitative energy-dispersive (EDS) analyses werecarried out in each group both on the bonding dentinsurface before applying a dentin primer and on the frac-tured dentin surface after TBS tests (Fig 1). The EDS an-alyzer (Voyager, Noran Instruments) was used for tbefollowing conditions: (1) accelerating voltage 15 keV,acquisition time 60 seconds, and magnification of 55Xon the adhesive dentin surface; and (2^ accelerating volt-age 15 keV, acquisition time 120 seconds, and mag-nification of 1,00OX on the fractured dentin surface. Forthe analysis of each dentin surface 2 specimens fromeach group were used.For each specimen 3 measure-ments of 6 different regions of the substrate were ob-tained. On the fractured dentin surface all regions withdifferent fracture patterns in tbe same specimen weremeasured. Tbe molar ratio of zinc to phosphorus wasused as an index of the remaining temporary cementsince the temporary cement contains zinc.

Soluhility Evaluation

Six blocks of square plate of a temporary cement (H Y-Bond Temporary Cement Hard), 1 cm per side and 1mm thick, were dried at 37°C for 30 minutes and thenin a vacuum oven at 60°C for 15 hours. The weightof each block was measured and then 3 samples were

immersed in 1 0 mL of either EP or MEP conditioner.After 1 hour of water incubation at 37''C, the tempo-rary cement blocks were removed, dried in the man-ner described above, and weighed. The mean per-centage of weigbt loss was calcuLjted for eachconditioner to evaluate their capability tor temporarycement dissolution.

Statistical Analysis

A 2-way factorial analysis of variance (ANOVA) wasemployed to analyze the effects of thermocyciingand conditioner type on the TBS values. Levels of sig-nificance were set at a = 0.05. Scheffe's comparisontest was employed to evaluate differences amonggroups at a significance level of 0.1 Oto avoid excessType II error.

Results

Tensile Bond Strength Measurement

The ANOVA showed that tbe type of conditioner andthe number of tbermocycles significantly influencedthe TBS values (P < 0.001 ). Mean TBS values gradu-ally decreasetd as the number of thermocycles in-creased, despite tbe use of a conditioner (Fig 2). Thevalue of tbe EP group was significantly higher than thatofttieTC group forevery number of thermocycles ex-amined (P< 0.1 ~ f < 0.0001 ) and significantly higherthan that of the C group before thermocyciing {P< 0.1)and after 10,000 thermocycles (P< 0.0005); whereastbe value oí the MEP group was significantly higherthan that of the TC group only before thermocyciing(P< 0.05) and after 5,000 thermocycles (P< 0.1) andsignificantly higher than that of tbe C group only after20,000 thermocycles (P < 0.05).

SEM Images of Fractured Dentin Surfyce

The C group showed adhesive failure between thedentin and resin cement, with dentinal tubule aper-tures visible. In addition, a smear resulting from sili-con carbide polishing was observed over the wholesutface. The TC group showed adhesive failure be-tween the dentin and resin cement. Few dentinaltubules were observed and a granular substance waspresent on the dentin surface. In contrast, the EPgroup showed a mixed adhesive-cohesive failuresurface, tbe adhesive failure between dentin andresin and the cohesive failure of the resin. The MEPgroup showed an adhesive failure surface betweenthe dentin and resin with some remnant substances.Approximately half of the dentinal tubules wereclosed and half were open (Fig 3).

The International loLrnal of Prosttiodonlics 2 3 2 Volume T2, Numb.-.

Watanabeet al Thermocycling and Bond Strength Between Resin Cement and Dentin

Fig 2 Relationship between ten-sile bond stfengtfi of resin cementto bovine dentin and number ofthermocyctes. C = control group;TC = temporary cement-appliedgroup; EP = ethyl dihydrogenphosphate-treated group attertemporary cement application;MEP = methacryloxyelhyl dihy-drogen phosphate-treated groupafter temporary cement appli-cation: vertical lines = standarddeviation.

. .2 , Number 3,1999 233 ThelniemationalloumafofProsthodonuQ

Thermocycling and Bond Strength Between Resin Cement and Dentin Watanabe et al

There were no relevant differences between theSEM images of the fractured dentin surface in speci-mens without thermocycling and after 2(],000 ther-mocycles.

Energy-Dispersive Analysis

The zinc-phosphate r,atio of specimens in theTC groupwas lower compared with that of specimens in the EPor MEP groups, showing that the application of a toothconditioner reduced the amount of remaining zinc onthe dentin surface. The amount of remaining zinc wassmaller in the EP group than in the MEP group. Thezinc-phosphate ratios were 1.41 for the TC group,0.42 for the EP group, and 0.54 for the MEP group.

The EDS qualitative analysis ofthe fractured dentinsurfaces revealed peaks of zinc in theTC group thatwere not present in the C, EP, or MEP groups. Also,a high peak of silica was observed on the regions ofcohesive failure only in the EP group.

Solubility Evaluation

Calculation of the percentage of temporary cementdissolved by immersion into each conditioner solu-tion revealed that EP dissolved a larger amount oftemporary cement when compared with MEP (31%± 2% vs 9% ± 2%1.

Discussion

This study clearly demonstrated that the tooth condi-tioners tested are useful in reducing the negativeeffects of temporary cement on the bond strength ofadhesive resin cement to dentin. Although the meanTBS values decreased with an increase in the numberof thermocycles, even in the specimens treated witha conditioner, the TBS values of the conditioner-treated groups were significantly higher than those ofthe TC group (Scheffe's comparison test, P< 0.0001for both conditioners! and the C group (Scht'ffe's com-parison test, P< 0.0001 for EP). In a previous study hy-brid layers and resin tags that were absent in the TCgroup were observed in the EP and MEP groups.'^These might account in part for the higher TBS valueseven after thermocycling stress.^'"^^ The decrease inTBS with the increase in the number of thermocyclesmight be a result of deterioration of the interfacecaused by the high temperature and decrease in thenumber of hydrogen bonds.^ ^

It should be mentioned that the teeth storagemethod in a freezer was previously confirmed to notinfluence the bond strength when compared withfreshly extracted bovine teeth.^'' The teeth were frozen

immediately after extraction to avoid the deteriorationofthe organic compounds included in dentin.

The fractured dentin surfaces for the EP and MEPgroups revealed adhesive or cohesive failure of theresin. The temporary cement remnants observed inthe TC group were not detected on the SEM imagesin either of the conditioner-treated surfaces. This in-dicates the conditioners' efficacy in eliminating thetemporary cement. The adhesive fracture patternsbetween dentin and resin—with many pluggedtubules and some regions of cohesive failure of theresin—that were observed in the conditioner-treatedsurfaces indicate a probable improvement in adhe-sion compared to the C group, and might be a desir-able failure pattern, probably preventing penetrationof toxins into the dentinal tubules.^^'^^

The ability of the tooth conditioners to reduce thenegative effect of temporary cement on the TBS val-ues was caused at least in part by their capacity to dis-solve temporary cement. The temporary cementmight be a weak layer,̂ •'̂ In addition, the polymer-ization ofthe primer and adhesive resin may be in-hibited by components of temporary cement such asthe tannin-fluoride preparation (HY-agent¡ presentin the HY-Bond carboxylate temporary cement. Theremaining cement may also reduce the infiltration ofadhesive substances into the dentinal tubules,^^'^^The EP dissolved a larger amount of temporary ce-ment than MEP and the EP group showed signifi-cantly higher TBS values compared to the MEP andC groups. It should be noted that the dentin surfacewas rinsed with running water for 10 seconds afterapplication of EP since EP is not classified as self-etching primer. Asa result, any remaining temporarycement on the dentin surface would have been ef-fectively washed away. This possibility might be sup-ported by the EDS results, which showed that the ratioof zinc to phosphate, an index ofthe remaining tem-porary cement, was smaller in the specimens treatedwith EP than in those treated with MEP.

Although the higher bond strength and durability ofthe specimens treated with EP should be noted, it is aserious shortcoming for its clinical use that EP is anetching substance that does not have a methacrylateradical and therefore requires an additional rinse step.In contrast, MEP does not require the rinse step sinceMEP has a methacrylate radical and thus is classifiedas a self-etching primer. As a result, temporary cementdissolved by MEP stays in the adhesive layer, andconsequently the ability to reduce the negative effectof temporary cement on the bond strength might nothave been as strong compared to EP, However, the im-portance of the significant difference between the EPand MEP groups may be brought into question by the

Tiie International Journal of Proslhodontics 2 3 4 Volume 12, Numb

Thermocycling and Bond Strength Between Resin Cement and Dentin

low TBS values recorded. It is difticult, therefore, to de-cide whether EP or MEP should be used clinically.Controlled clinical trials must be carried out in tlie fu-ture to clarify the clinical efficacy of the conditionersand to determine which conditioner is superior.

Acknowledgment

This investigation was îupported in part bv a Grant in Aid forScientific Research from the Ministrv of Education, Science, Sports,and Culture, ]apan.

References

1. Vahidi F. The provisional restoration. Dent Clin North Am 1987;31:363-381.

2. Christensen C|. Provisional restorations ror fixed prosihodonLcs.I Am Dent As«ic 1996;! 27249-253.

3. Hombnxik DS. Provipontand Provilink. Maximizing aestheticsand function when fabricating provisional restorations. Signature1995:10-16.

4. Hansen EK, .^smüssen E. influenceottemporary filling materialson effect of dentin-bonding agents. Scand J Dent Res 1987,-95:516-520.

5. Woody TL, Daviî RD. The effect of eugenol-containrng andeugenol-lree iemporary cements on microleakage in resinbonded restorations, Oper Dent 1992:17:175-180,

6. Watanabe EK, Suzuki K, Yamashila A, Shigeta N, Imai M, YataniH, Nakai H. InrTuericeof (emporary cement on the adhesivenessofresin cement lo dentin |in lapanese!. lapan] Dent rVlater 1996;15:187-191.

7. Watanabe EK, Vamashita ,\. Imai M, Yatani H, Suzuki K.Temporary cement remains as an adhesion inhibiting factor inthe interface between resin cements and bovine dentin. Int JProsthodont 1997,10:440-452.

8. Schwartz R, Davis R, Hilton TJ. Effect of temporary cements onthe bond strength of a resin cement. Am | Dent 1992;5:147-150.

9. Paul 5|, Schärer P. Effect of provisional cements or the bondstrength of various adhesive bonding systems on dentine. J OralRehabiM997;24:B-14,

10. Bachmann M, Paul S|, LOthyH, Schärer P. Effect of cleaning den-tine with soap and pumice on shear bond strength of dentine-bonding agents, | Orai Rehabil 1997;24:433--J38.

11. XieJ, Powers ]M, McCuckin RS. In wiro bond strength of twoadhesives to enamei and dentin under normal and contaminatedconditions. Dent Mater l993;9-295-299,

12. Terata R. Characterization of enamel and dent n sjh^aces afterremoval of temporary cement—Study on removal of temporarycement. DentMater 11993;! 2:18-28.

13. Watanabe EK, Yamashila A, Yatani H, Ishikawa K, Suzuki K.Improvement m (he tensile bond strength between resin ce-ment and dentin surfaces after temporary cement application.Int I Prosthodoni 1998;l 1:203-2l 1.

14. Ayad MF, Rosenliel SE, Salama M. Infiuence of tooth surfaceroughness and t>pe of cement on retention of complete castcrowns. I Prosthet Dent 1997r77:n6- l 21.

15. Ohno H, Araki Ï , Endo K, Yamane Y, Kawashima I. Evaluationof water durability at adhesion interfaces by peeling test of resinfilm. Dent Mater 11996;15:183-Í 92.

16. Smith RM, Barrett MG, Gardner WA, Marîhail T, McLean MJ,McMichael DW, et al. Effect of environmental stress and surfacetreatment on resin-to-metal bonding. Ami Dent 1993,-6:1 ! 1-115.

17. Davidson CL, Abdaila Al, De Gee A]. An investigation into thequality of dentine bonding systems for accomplishing a durablebond. I Oral Rehabil I993;30r291-300.

18. Tanizaki K. Effect o i tannin-fluoride preparation on the reduc-tion of secondary dental caries |in lapanesel. Japan J Cor\servDent 197S;21 279-296.

19. Yamaga M, Koide T, Hieda T, Daito M. Obturation of dentinaltubuleî with tannin-ñuoride preparation (HY agent) incorporatedinto giass ionomer cement. | Osaka Dent Univ 1993:27:77-87.

20. Suzuki K, Munechika T, Matsukawa S, Hone K. Studies on polv-mers having adhesiveness to tooth substance ¡part 21—SiTithesisand adhesiveness of rnethacrylate monomer with a-amino acidused as skeleton |in |apanese|. Japan i Dent Mater 198037:

21. Nakabayashi N, Ashiaawa M, Nakamura M. Identification of aresin-dentin hybrid layer in vital human dentin created m \ivo:Durable bonding to vital dentin. Quintessence Int 1992;23:135-141.

22. Nakabayashi N, Watanabe A, Ikeda W. Intra-orai bonding of 4-META/MMA-TBB resin to vital human denfin. Am [ Dent1995:8:37.-I2.

23. Uno S, Finger W|. Function of the hybrid zone as a stress-absorbing layer in resin-dentin bonding. Quintessence Int1995;26:733-738.

24. Tonami K, Takabashi H, NisbimuraE. Effect of frozen storage andboil ing on tensile strength of bovine dentin. Dent Mater J1995;1S2O5-2I1.

25. Iwaku M, Nakamichi I, Nakamura K, Horie K, Suizu S, EusayamaT. Tags penetrating dentin of a nesv adhesive resin. Bull TokyoMed Dent Univ 1981^8:45-51.

26. Staninec M, Marshall GW ]r, Kawakami M, Lowe A. Bondstrength, interfaciai characterization, and fracture surface analy-sis for a new stress-breaking bonding agent. ) Prosthet Dent1995;74:469-475.

!- . . :_• Í2, Number 3,1999 235 The Intemaiional Journal of Prosthodontii