Effect of Different Cleaning Regimens on the Adhesion of Resin to Saliva-Contaminated Ceramics.

Effect of Different Cleaning Regimens on the Adhesionof Resin to Saliva-Contaminated CeramicsAkın Aladag, DDS, PhD,1 Bahar Elter, DDS,2 Erhan Comlekoglu, DDS, PhD,1 Burcu Kanat, DDS, PhD,3

Mehmet Sonugelen, DDS, PhD,4 Atilla Kesercioglu, DDS, PhD,4 & Mutlu Ozcan, DDS, DMD, PhD5

1Associate Professor, Department of Prosthodontics, School of Dentistry, Ege University, Izmir, Turkey2Teaching assistant, Department of Prosthodontics, School of Dentistry, Ege University, Izmir, Turkey3Research assistant, Department of Prosthodontics, School of Dentistry, Ege University, Izmir, Turkey4Professor, Department of Prosthodontics, School of Dentistry, Ege University, Izmir, Turkey5Professor, Clinic for Fixed and Removable Prosthodontics and Dental Materials Science, Dental Materials Unit, Center for Dental and OralMedicine, University of Zurich, Zurich, Switzerland

Keywords

Zirconia; saliva contamination; cleaning paste;sodium hypochlorite; glass ceramics.

Correspondence

Akın Aladag, Department of Prosthodontics,School of Dentistry, Ege University, Bornova35100-Izmir, Turkey.E-mail: [email protected]

The authors deny any conflicts of interest.

Accepted December 27, 2013

doi: 10.1111/jopr.12170

AbstractPurpose: The aim of this study was to evaluate the influence of different cleaningregimens on the microshear bond strength (µSBS) of three different all-ceramicsurfaces after saliva contamination.Material and Methods: Cubic ceramic specimens (3 × 3 × 3 mm3) were pre-pared from three types of ceramics: zirconium dioxide (Z), leucite-reinforced glassceramic (E), lithium disilicate glass ceramic (EX; n = 12/subgroup). A total of 144composite resin cylinders (diameter: 1 mm, height: 3 mm) were prepared. Threehuman-saliva–contaminated surfaces of ceramic specimens were cleaned with eitherwater spray (WS), with 0.5% sodium hypochlorite solution (HC), or with a cleaningpaste (CP). Control surface (C) was not contaminated or cleaned. Composite cylinderswere bonded to each surface with a resin luting cement. All specimens were stored at37◦C in deionized water until fracture testing. µSBS tests were performed in a uni-versal testing machine (0.5 mm/min), and the results (MPa ± SD) were statisticallyanalyzed (two-way ANOVA, Bonferroni a = 0.05). Fractured surfaces were ana-lyzed to identify the failure types using an optical microscope at 50× magnification.Two representative specimens from all groups were examined with scanning electronmicroscopy.Results: µSBS test results were significantly affected by the saliva cleaning regimens(p = 0.01) and the ceramic types (p = 0.03). The interaction terms between theceramic type and saliva cleaning regimen were also significant (p < 0.05). Therewere no significant differences among the µSBS values (MPa ± SD) for the Z group(C = 17.5 ± 8.8; WS = 16.0 ± 4.9; HC = 17.6 ± 5.8; CP = 16.6 ± 7.5; p > 0.05).In the EX group, C resulted in significantly higher µSBS values (32.6 ± 7.4) than CP(17.4 ± 8.9), WS (15.6 ± 7.3), and HC (14.3 ± 4.5) (p < 0.05); however, C (20.4± 7.1) and HC (19.2 ± 7.5) showed higher µSBS values than CP (13.8 ± 4.8) andWS (10.9 ± 5.7) in the E group. Some cohesive failures within the luting resin wereobserved in the E and EX groups, whereas only adhesive failures were seen in zirconiagroups for all surface treatments.Conclusions: Different ceramic surface cleaning regimens after saliva contamina-tion of the zirconium dioxide revealed µSBS similar to the control group, whereas allsurface cleaning regimens tested significantly decreased the bond strength values inthe lithium disilicate glass ceramic. The leucite-reinforced glass-ceramic group bene-fited from 0.5% sodium hypochlorite solution cleaning with increased bond strengths.Clinical significance: Adhesive cementation of zirconia presents a clinically challeng-ing protocol, and the cementation surface contamination of the zirconia restorationsand the inadequate removal of the contaminants increase the risk of failure, as forall ceramic types. This study demonstrated that surface cleaning regimens should beapplied according to different ceramic properties.

1Journal of Prosthodontics 00 (2014) 1–10 C© 2014 by the American College of Prosthodontists

Effect of Cleaning Regimens on Saliva-Contaminated Ceramics Aladag et al

Cementation of all-ceramic restorations by adhesive lutingresins has become a routine procedure for clinical use, andproblems in bonding of luting resins to ceramics have beensignificantly solved in recent years. The success of long-term resin bond to all-ceramic restorations have been welldocumented.1,2

Zirconium dioxide (zirconia) polycrystalline materials, alsocalled “oxide ceramics,” have been used in a wide range forsingle and/or multi-unit fixed dental prostheses as well as im-plant abutments, due to their biocompatibility and relativelyhigh esthetic properties.3 Although adhesive cementation isnot a prerequisite for zirconia restorations,1 it is recommendedas in other all-ceramic materials in terms of decreasing therisk of secondary caries development by sealing the impuritiesaround the finish line and increasing the fracture resistance ofthe restored teeth and restorations.4 However, because zirconiadoes not contain a silicate phase, it cannot be adhesively lutedas in silica-based all-ceramics.2 Silica-based all-ceramics havethe advantage of successful bonding with self-, light-, or dual-curing resin-luting cements to teeth provided that the lutingsurfaces of such restorations are etched with hydrofluoric acidfollowed by silanization,5 which has been a widely acceptedprotocol; however, adhesive cementation of zirconia restora-tions has been best practiced by airborne particle abrasion ofthe intaglio surfaces of the zirconia followed by applicationof phosphate-bonded-monomer–containing composite resins.6

However, air abrasion might induce surface defects on zirco-nia, reducing its strength.7 Alternative luting surface condi-tioning methods such as selective infiltration etching, zirconiaceramic powder coating, ceramic pearl layer application, glazelayer application or recently introduced nanostructured alumina(Al2O3) coating might help overcome the problems occurringdue to air abrasion.8,9 Luting of zirconia restorations by conven-tional bis-GMA containing resin cements, which do not con-tain the bifunctional phosphate-bonded monomer, alone havenot been successful in establishing a durable long-term bond tooxide ceramics.9,10

In addition to the abovementioned difficulties in achieving astrong bond between the zirconia restoration and resin lutingcements, the try-in procedure of all-ceramic and/or zirconiarestorations causes the intaglio bonding surface of the restora-tion to be generally contaminated with saliva, blood, or fittingindicator remnants such as silicone or try-in pastes, making theadhesive cementation of the zirconia restorations even moredifficult. Failure to remove fluids or try-in materials resultsin reduction of the bond strength.9,11 Thus, any inorganic ororganic contaminants should be eliminated before adhesive ce-mentation. Commonly used methods of decontaminating theluting surfaces of ceramic/zirconia restorations are scrubbingthe surface with acetone, application of 37% phosphoric acidfor 60 seconds once or for 30 seconds twice, immersing in70% isopropanol for 2 minutes, cleaning with 96% ethanolfor 15 seconds, airborne-particle abrasion, 2% chlorhexidine or5% sodium hypochlorite application, or water spray (WS).12,13

Recently, an alternative universal paste for extraoral cleaningof pretreated ceramic and metal restoration surfaces that havebeen contaminated during intraoral try-in has been developed.14

This cleaning paste (CP) consists of an alkaline suspension ofzirconium dioxide particles deemed to absorb the phosphate

contaminants to bond to them rather than to the surface of theceramic restoration, leaving behind a clean surface.13,15

Saliva contains organic materials such as salivary proteins,enzymatic molecules, bacteria, and food debris, and inorganiccompounds, such as mineral ions, in water solution.13 Adhesionof salivary proteins to dental materials and tooth surfaces (salivacontamination) results in formation of acquired enamel pelli-cle, which is free of bacteria of 10 to 20 nm thickness withina few minutes.16 When the protein transmission from salivaincreases, the thickness of the proteinaceous layer reaches 100to 1000 nm between 30 and 90 minutes.17 It is almost im-possible to avoid saliva contamination of all-ceramic and/orzirconia restorations during the try-in procedure. During ad-hesive cementation procedures, contaminant removal plays animportant role in the durable adhesion and clinical performanceof the restoration.9,13

Therefore, the objective of this in vitro study was to evaluatethe influence of various surface cleaning regimens on micros-hear bond strength (µSBS) of three all-ceramic surfaces aftersaliva contamination, thereby suggesting a preference for a pu-rifying procedure for all-ceramic surfaces. The null hypothesiswas that µSBS values obtained with different saliva decon-tamination methods would not be affected by surface cleaningregimens or different ceramic types.

Materials and methods

The brand names, types, manufacturers, chemical composi-tions, and batch numbers of the materials used in this study arelisted in Table 1.

Specimen preparation

Cubic specimens (3 × 3 × 3 mm3; N = 36) were prepared fromthree types of ceramics, namely, presintered yttrium-stabilizedzirconium dioxide (IPS e.max ZirCAD; Z), leucite-reinforcedglass-ceramic (IPS Empress CAD; E), and lithium disilicateglass-ceramic (IPS e.max CAD; EX; n = 12/subgroup) by sec-tioning under water cooling with a slow-speed diamond saw(Isomet R© 1000, Buehler, Lake Bluff, IL). The thickness of thesaw for all groups in addition to 20 to 25% sintering shrinkageof the zirconia for Z group was taken into consideration whenfabricating specimens in a 3 × 3 × 3 mm3 final dimension. Thepresintered zirconia specimens were then sintered (InFire HTCspeed; Sirona Dental Systems, GmbH, Bensheim, Germany)according to the manufacturers’ firing instructions, whereasthe lithium disilicate specimens were sintered in the furnace(Programat P300; Ivoclar Vivadent) to convert the crystallineintermediate (metasilicate) stage of the ceramic into the disili-cate phase and to improve the flexural strength of the ceramic.

Surfaces of all specimens were wet-polished first with 800-grit followed by 1200-grit silicon carbide paper for 120 seconds.All specimens were ultrasonically cleaned (Quantrex 90 WT;L&R Manufacturing, Inc., Kearny, NJ) in 99% ethanol for 15minutes. All conditioning, cleaning, and bonding procedureswere carried out by the same operator. Although airborne-particle abrasion with 50 µm Al2O3 at 2.5 bar pressure wasapplied on the all specimens for 15 seconds from a distance of10 mm,15,18 glass-ceramic specimens were further etched with

2 Journal of Prosthodontics 00 (2014) 1–10 C© 2014 by the American College of Prosthodontists

Aladag et al Effect of Cleaning Regimens on Saliva-Contaminated Ceramics

Table 1 Chemical compositions of the materials used

Product Type Chemical composition Manufacturer Batch number

IPS e.max ZirCAD (Z) Zirconium dioxide framework 87% to 95% ZrO2, 4% to 6%Y2O3, 1% to 5% HfO2, 0% to1% Al2O3,< 0.2% otheroxides

Ivoclar Vivadent; Schaan,Liechtenstein

S06538

IPS e.max CAD (EX) Lithium disilicate veneeringceramic block

57% to 80% SiO2, 11% to 19%Li2O2, 0% to 13% K2O, 0% to11% P2O5, 0% to 8% ZrO2,

0% to 5% Al2O3,0% to 5%MgO, 0% to 8% coloringoxides


S08650

IPS Empress CAD (E) Leucite glass ceramic block 60% to 65% SiO2, 16% to 20%Al2O3, 10% to 14% K2O,3.5% to 6.5% Na2O, 0.5% to7% other oxides, 0.2% to 1%pigments


P50773

Ivoclean Extraoral cleaning paste 10% to 15 zirconium oxide,65% to 80% water, 80% to10% polyethylene glycol, <1sodium hydroxide, 4% to 5%pigments, additives


P66483

MonoBond Plus Universal primer mediating abond between metal,glass/oxide ceramics andresin

50% to 100% ethanol, <2.5%3-trimethoxysilylpropylmeth-acrylate, <2.5%methacrylated phosphoricacid ester


P41360

Variolink II Dual-curing resin cement Ivoclar Vivadent; Schaan,Liechtenstein

632325

Heliobond Bis-GMA, triethyleneglycoldimethacrylate,initiators, stabilizers

Base Bis-GMA, TEGDMA, urethanedimethacrylate (UDMA),inorganic filler, ytterbiumtrifluoride, initiator, stabilizer


R71123

Catalyst Bis-GMA, TEGDMA, UDMA,inorganic filler, ytterbiumtrifluoride, benzoyl peroxide,stabilizer

Tetric EvoFlow Composite resin Bis-GMA, UDMA, decandiold-imethacryate, barium glassfiller, ytterbium trifluoride,highly dispered silica, mixedoxide, prepolymers, additives,catalysts, stabilizers, andpigments


R82387

a 9% hydrofluoric acid (Ultradent Porcelain Etch; UltradentProducts Inc., South Jordan, UT) for 60 seconds and 20 sec-onds for leucite (E) and lithium disilicate (EX) glass-ceramics,respectively. All specimens were further ultrasonically cleanedin 99% ethanol for 10 minutes.

Surface cleaning regimens

Three surfaces of each cubic ceramic specimen were contami-nated with fresh human saliva obtained from a healthy female

donor who had not consumed any food or drinks 1.5 hoursbefore sample collection18 by a cotton pellet. One surface ofeach cubic ceramic specimen was used as control. Becausefour of the six surfaces of each cubic specimen were used forexperimental and control groups for standardization, contam-ination of the surfaces to be treated was avoided using apolyethylene mold with a square (3 × 3 mm2) window thatfit precisely to the specimens. The saliva was applied, left for10 minutes, and then the saliva-contaminated surfaces for allspecimens were randomly divided into three groups and cleaned

Journal of Prosthodontics 00 (2014) 1–10 C© 2014 by the American College of Prosthodontists 3


with different regimens of WS, 0.5% sodium hypochlorite so-lution (HC), and a CP.

Water spray: The saliva-contaminated surfaces of all speci-mens were washed with WS for 20 seconds, and then dried for20 seconds.

Sodium hypochlorite solution (HC): The second saliva-contaminated surface of all specimens were cleaned with 0.5%sodium hypochlorite solution for 20 seconds, washed and rinsedfor 10 seconds, and dried for 5 seconds.

Cleaning paste: The third saliva-contaminated surface of allspecimens was rinsed with WS and dried with oil-free air. ACP (IvocleanTM; Ivoclar Vivadent) was applied for 20 secondsusing a microbrush, then thoroughly rinsed and dried.

Control surface (C): The untreated fourth surfaces of all spec-imens served as the control group.

Bonding procedures

The composite resin material (Tetric EvoFlow; Ivoclar Vi-vadent; N = 144) was filled in a glass cylinder tube with aninner diameter of 1 mm and photopolymerized (Bluephase G2;Ivoclar Vivadent) from all sides for 40 seconds. After removalfrom the tube, the cylindrical specimens were sectioned to a 3mm height under water cooling with a slow-speed diamond saw.The composite resin specimens were ultrasonically cleaned inethanol for 15 minutes. Polyethylene plates with a hole (di-ameter: 1 mm) in the center were used for preparation of thebonding area. The ceramic surfaces, which were cleaned withdifferent regimens, were silanated (Monobond Plus; IvoclarVivadent) using a microbrush, left 60 seconds for its reac-tion, and dried for 10 seconds. The bonding agent (Heliobond;Ivoclar Vivadent) was applied with a microbrush and gently airthinned. The resin luting cement including base and catalystpastes (Variolink II; Ivoclar Vivadent) was mixed according tomanufacturer’s instructions and applied to the ceramic surface,which was visible from the polyethylene mold. Cleaned ceramicsurfaces and conditioned composite specimens were bonded,excess resin was removed using a microbrush, and the surfaceswere photopolymerized for 40 seconds from each aspect ofthe specimens. After polymerization, polyethylene molds weregently removed, and the specimens were stored in deionizedwater at 37◦C for at least 48 hours until fracture testing.

µSBS test and failure analysis

The specimens were inserted in a custom-made specimenholder (Fig 1) and µSBS tests were performed in a univer-sal testing machine (Autograph AG-50 kNG; Shimadzu, Ky-oto, Japan) at a 0.5 mm/min crosshead speed. The shear forcewas applied to the ceramic-composite interface until fractureoccurred. After µSBS tests, all debonded specimens were ana-lyzed to identify the failure types using an optical microscope(MP 320; Carl Zeiss, Jena, Germany) at 50× magnification. Af-ter all specimens were evaluated, failure modes were classifiedeither as adhesive failure between the ceramic and resin (A) orcohesive failure within the luting resin or in the composite resin(C).19

Two representative fractured surfaces from each group andfailure type were dried by vacuum dessication and carboncoated. A scanning electron microscope operating at 20 kV

Figure 1 Custom-made specimen holder for the application of µSBStest.

(JSM-5200; JEOL, Tokyo, Japan) was used to observe the fail-ure modes of the debonded ceramic specimens after µSBStesting.

Statistical analysis

The obtained results (MPa ± SD) were statistically analyzed(SPSS 16.0 for Windows; SPSS Inc., Chicago, IL) within ce-ramic groups for comparison of cleaning regimens using two-way ANOVA, Bonferroni correction. Note that p < 0.05 wereconsidered to be statistically significant in all tests. Weibullstatistical analysis (MINITAB Version 14; State College, PA)was applied to describe SBS analysis and to evaluate the re-liability of cleaning regimens of saliva from different ceram-ics. The adjusted Anderson-Darling goodness-of-fit with 95%CI estimation method of maximum likelihood was carriedout for mechanical test results data to determine the Weibulldistribution.

Results

µSBS results were significantly affected by the saliva clean-ing regimens (p = 0.01) and the ceramic types (p = 0.03;two-way ANOVA). The interaction terms between the ceramictype and saliva decontamination method were also statisticallysignificant (p < 0.05).

There were no significant differences among the µSBS val-ues for cleaning regimens in the Z group (p ≥ 0.05). Inthe EX group, C resulted in significantly higher µSBS val-ues than CP, WS, or HC (p < 0.05); however, C and HCshowed higher µSBS values than CP and WS in the E group(Table 2).

Estimates of the parameters of the Weibull distributionfor the different ceramic groups in different decontamina-tion methods are presented in Table 3. The test to estimateshape parameters (p ≥ 0.05) indicated no statistical differ-ences in Weibull shape parameters among the ceramic groups,whereas the test for equal scale parameters (p < 0.05) indi-cated that ceramic types in C and HC groups had differentscale parameters for the µSBS mechanical test. According tothe analysis with the Weibull distribution, probability plot and



Table 2 µSBS (MPa) results of three ceramics after application of water spray (WS), sodium hypochlorite solution (HC), and cleaning paste (CP)cleaning regimens following saliva contamination and noncontaminated control group. Same superscript small letters indicate no significant differencein the row, same superscript capital letters indicate no significant difference in the column. Results of two-way analysis of µSBS of ceramic typesand saliva cleaning regimen (p < 0.05)

Ceramic types Water spray Sodium hypochloriteControl (C) (WS) solution (HC) Cleaning paste (CP)

E 20.4 ± 7.1b,A 10.9 ± 5.7a,A 19.2 ± 7.5b,A 13.8 ± 4.8a,A

EX 32.6 ± 7.4b,B 15.6 ± 7.3a,A 14.3 ± 4.5a,A 17.4 ± 8.9a,A

Z 17.5 ± 8.8a,A 16.0 ± 4.9a,A 17.6 ± 5.8a,A 16.6 ± 7.5a,A

Table 3 Relation of the scale and shape parameters among the saliva contamination methods. Different superscript letters indicate significantdifferences among groups

Scale (MPa) 95% CI (scale; N) Shape 95% CI (shape) AD

C E 23a (19, 27) 3.5a (2.2, 5.6) 1.380EX 36b (31, 40) 4.8b (3.2, 7.3) 1.518Z 20a (15, 26) 2.3a (1.4, 3.6) 1.387

WS E 12c (9, 16) 2.1c (1.3, 3.4) 1.522EX 18c (14, 23) 2.4c (1.5, 3.9) 1.433Z 18c (15, 21) 3.8c (2.4, 5.9) 1.459

HC E 22d (17, 27) 2.8d (1.9, 4.1) 1.488EX 16e (14, 18) 3.9d (2.5, 6.2) 1.416Z 20f (16, 23) 3.3d (2.2, 5.0) 1.416

CP E 15g (13, 18) 3.7e (2.2, 6.1) 1.735EX 20g (15, 26) 2.2e (1.4, 3.4) 1.339Z 19g (15, 24) 2.5e (1.6, 3.7) 1.378

Table 4 Failure types per test group. Adhesive failure at the interface between ceramic and luting resin (A); cohesive failure within the resin only (C)

Hypochlorite Water Cleaningsolution (0.5%) spray paste Control

A C A C A C A C

Leucite-reinforced glass ceramic (E; n: 12) 10 2 12 0 12 0 10 2Lithium disilicate glass ceramic (EX; n: 12) 10 2 12 0 11 1 8 4Zirconium dioxide (Z; n: 12) 12 0 12 0 12 0 12 0

parametric survival plot graphics of µSBS test can be seen inFigure 2.

Mainly adhesive failures were observed for C, WS, HC,and CP groups for all-ceramic groups; however, some cohe-sive failures within the luting resin were observed in E andEX groups, while no cohesive failures were seen in zirconiagroups for C, WS, HC, and CP surface treatments (Table 4).For the control group of zirconia specimens, a clean surfacetexture of the polycrystalline structure was observed (Fig 3),while lower amounts of surface contaminants were present inthe experimental group (Fig 4). The etched surfaces in the glassceramic (E) group were observed as wide grooves (Fig 5),and HC groups revealed similar clean surfaces (Fig 6). In theEX control group (Fig 7), the etched surfaces revealed smallergrooves with homogenous irregularities while the experimen-tal groups, especially CP, presented remnants on the surface(Fig 8).

Discussion

The performance of different ceramic surface cleaning regi-mens with leucite-reinforced glass-ceramic, lithium disilicatebased ceramic, and zirconia specimens after human salivarycontamination has been investigated in this study. A significanteffect of the decontamination methods was observed, comparedto the control group, and ceramic types affected the µSBS val-ues. Thus, the null hypothesis was rejected.

Cleanliness of the bonding surfaces has an influence on thesuccess of the durable bond strength.9,20 Therefore, removalof the saliva contaminants that occur during the try-in proce-dures from ceramic inner surfaces before adhesion plays animportant role in the longevity of the restorations.21 Solid sub-strates such as enamel, dentine and different types of restora-tive materials (glass ceramics and zirconia) in the oral cavityhave different polarities and physical properties, which affect



Figure 2 Survival and probability plots of different cleaning methods.

the adhesion of the salivary proteins.22 When the restorationcomes into contact with saliva during the try-in phase, animmediate proteinaceous layer composed of adsorbed proteins,several enzymes, glycoproteins, and other macromolecules isformed.23,24 Through the physicochemical interactions betweenthe solid substrates and saliva, supramolecular protein assem-blies (proline-rich proteins, PRPs) may precipitate and adhereto the substrates by forces with different ranges.25,26 It has alsobeen demonstrated through X-ray photoelectron spectroscopic(XPS) analysis that adsorption of the salivary proteins to thezirconia surface increases the carbon, nitrogen, and silica lev-els, leading to an alteration in the composition of the surface.20

Therefore, immediate elimination of these precipitates is cru-cial for adhesion.

Although standard airborne particle abrasion was not usedas a ceramic decontamination method in this study, all-ceramicsurfaces were air abraded with 50 µm Al2O3 at 2.5 bar pressure,simulating the as-arrived restoration from the laboratory to theclinic, and the contamination was then conducted mimickingthe final intraoral try-in of the restoration before cementation,and the most practical and commonly used decontaminationmethods were then applied. As has been previously reported,airborne-particle abrasion produces the surface roughness nec-essary for adhesive luting. On the other hand, it might increase



the possibility of contamination and prevent adequate decon-tamination because of the created pits and irregularities on theluting surfaces of the restorations.27 After the contaminationof this roughened surface by saliva, further application of airabrasion for cleaning of the adherent surface of the restorationmight induce surface defects on zirconia, reducing its strength.7

Various methods, either tribochemical silica coating followedby the application of the trialkoxysilane coupling agents28,29 orairborne Al2O3 particle abrasion followed by phosphate estermonomer-containing primers,30 have been used to obtain goodadhesion between zirconia and resin cement. While air abra-sion roughened and increased the wettability of zirconia,31 theprimers/silanes that consist of adhesive functional monomerapplications such as methacrylate phosphoric ester32 and 10-methacryloyloxydecyl dihydrogen phosphate obtain the chem-ical bonding.33 The organophosphate monomers contain bothpolymerizable functional groups, which copolymerize with thematrix of the acrylate-based dental resin cements, or com-posites and phosphoric acid (phosphate ester) groups, whichbond chemically to metal oxides at the zirconia ceramic sur-face with Van der Waals forces or hydrogen bond.28 For thisintermolecular binding to be effective, the surfaces shouldbe extremely clean. Besides these advantages of phosphate-bonded monomers, it was found that application of either phos-phate ester containing primer or resin cement might be suc-cessful to improve the bond strength with respect to using bothof them.34,36 In this study, the zirconia surfaces were condi-tioned with a methacrylated phosphoric acid ester containinguniversal primer (Monobond Plus) before bonding with a resincement.

Saliva contamination could be cleaned with several regimens;however, limited literature has been found about the most ef-fective method to remove the remnants from the inner surfaceof the various types of ceramics. In a previous study wherethe effect of contaminations (saliva and disclosing agent) andcleaning regimens on bonding to zirconia had been investigatedusing XPS chemical analysis and tensile strength testing, con-tamination with saliva significantly decreased the bond strengthof zirconia, and airborne-particle abrasion was the most effec-tive cleaning regimen with respect to water rinsing, isopropanol,and phosphoric acid gel.27 Phosphoric acid has previously beenproposed as a zirconia surface cleaning agent with the assump-tion that it was a good organic solvent; however, an earlierstudy demonstrated that H3PO4 led to a decrease in the bondstrength.4 In this study, 0.5% sodium hypochlorite solutionand a CP were preferred because of the lack of informationabout the surface cleaning of saliva-contaminated ceramics us-ing these materials. Sodium hypochlorite is a nonspecific pro-teolytic agent capable of removing the organic material andproteins from the smear layer.37 The universal paste used inthis study is an alkaline suspension of zirconium dioxide par-ticles in water, and it has been reported to absorb the phos-phate contaminants in the media, leaving a clean surface.14 Allzirconia groups exhibited insignificant differences in terms ofbond strength values for various surface cleaning regimens;however, SEM images (Figs 3 and 4) revealed more or lessamounts of contaminated surfaces for zirconia specimens. Thismight be an explanation for high initial bond strength valuesfor all bonded specimens because an adhesive, from a chemi-

Figure 3 SEM image (5000×) of the zirconia material in the controlgroup. A clean polycrystalline structure of the zirconia can be observed.

Figure 4 SEM image (5000×) of the zirconia groups after the cleaningpaste (Ivoclean) application. The remnants on the contaminated surfaceare indicated with arrows.

cal aspect, attaches well to even contaminated and/or modifiedsurfaces initially, but the deterioration of the bond is compro-mised in the short run. Thus, long-term clinical results shouldbe obtained for a valid cleaning and adhesion protocol to bedefined. Therefore, the abovementioned phosphate absorbingmechanism of the universal CP used in this study might nothave completely affected the removal of PRPs in the saliva,and thereby might have blocked the removal of these proteinsfrom the air-abraded and saliva-contaminated zirconia surface,leaving behind a possible film layer. The differences among thetested decontamination methods for the zirconia surface wereinsignificant in this study. Therefore, even the most practicalWS cleaning of the zirconia surface can be recommended forchairside cleaning procedures.

In this study, some commonly used glass ceramics besideszirconia were tested in terms of chairside cleaning after salivacontamination. When glass ceramics are concerned, it can beassumed that re-etching of the previously HF acid-etched sur-faces would work well for cleaning. Klosa et al15 reported thatre-etching with hydrofluoric acid was the most successfulmethod for cleaning the saliva-contaminated lithium dis-ilicate ceramics (IPS e.max Press) with respect to 37%



Figure 5 SEM image (3000×) of the leucite-reinforced ceramic in thecontrol group. The wider porous structure was observed after etchingwith HF acid when compared to etched lithium disilicate ceramic.

phosphoric acid, 5% hydrofluoric acid, 96% isopropanol, oran air-polishing device with sodium bicarbonate; however, re-peated HF acid etching of the glass-ceramic surfaces may leadto weakening of the ceramic structure, thus decreasing bondstrengths.38 While cleaning with 0.5% sodium hypochloritesolution affected the µSBS of saliva-contaminated, leucite-reinforced glass-ceramic groups, all the decontamination meth-ods applied in this study were found to be ineffective forthe lithium disilicate surface cleaning, µSBS tests also sup-ported these findings. As lithium disilicate ceramics containquartz, lithium dioxide, phosphor oxide, alumina, potassiumoxide, and other components, the block form of lithium disil-icate glass-ceramic material, suitable for CAD/CAM applica-tions, has a two-crystalline phase (metasilicate and fully crys-tallized lithium disilicate forms) microstructure (Li2Si2O5).14

When the glass ceramics are etched with hydrofluoric acid,a chemical reaction: 4HF+SiO2 —> SiF4+2H2O occurs, be-cause the affinity of fluoride to silicon is higher than to oxy-gen, and microgrooves and irregularities can be formed onthe glass-ceramic surface to promote micromechanical adhe-sion with resin cement; however, the etchability of leucite-reinforced and lithium disilicate ceramics differs due to theirchemical compositions.40 The same concentration and appli-cation time for HF acid results in higher surface roughnessin the lithium disilicate based glass ceramics when comparedwith leucite-reinforced glass ceramic.41,42 Although the speci-mens in both glass-ceramic groups were HF acid etched ac-cording to manufacturer’s instructions (E: 60 seconds, EX:20 seconds), different etching patterns were observed in theEX and E groups, with the E group revealing a wider porousstructure easing the removal of the contaminants (Fig 5) whilein the EX group the grooves were more dense, not allowingfor the removal of remnants and saliva (Fig 7) when SEMimages were observed. The increased µSBS values in theHC group in E and decreased values in EX may stem fromthe abovementioned surface mechanism. A possible explana-tion for the decreased µSBS and increased adhesive failurecould be that an invisible thin residual organic film mighthave covered the fitting surfaces of the restoration blocking

Figure 6 SEM image (3000×) of the HF acid etched leucite-reinforcedglass ceramic after cleaning with 0.5% sodium hypochlorite solution.The wider grooves are seen as clean surfaces.

Figure 7 SEM image (3000×) of the HF acid etched lithium disilicateceramic in the control group with more dense grooves than leucite-basedglass ceramic.

the penetration of the silane and luting cement into themicroporosities.

In correlation with µSBS values, mainly adhesive failureswere observed in glass-ceramic groups for CP, WS, and HCcleaning regimens, which related to lower bonding due to notbeing effective enough to remove the saliva contaminations.Cohesive failures within the cement observed in C indicateda cleaner surface, which leads to higher µSBS values. SEMimages of this study revealed that surface contaminants werepresent in the experimental zirconia groups. A clean polycrys-talline structure of the zirconia of the control group was ob-served (Fig 3), while the CP was unable to eliminate the sur-face contaminants completely (Fig 4). Cleaning with a 0.5%hypochlorite solution proved to be effective in the E group(Fig 6), revealing similar images in the control group (Fig 5) assupported by µSBS values. In SEM images of the EX controlgroup, acid-etched porous structure was observed (Fig 7), whilea clearly visible remnant layer was observed in EX-CP group,verified by µSBS (Fig 8).



Figure 8 SEM image (3000×) of the lithium disilicate ceramic with visi-ble remnant layer followed by the cleaning paste application.

Although aging procedures may be required for in vitro adhe-sion testing of dental substrates and inorganic materials (i.e., ce-ramics, composites), various low-temperature aging treatmentshave proven not to negatively affect the flexural strengths ofY-TZP ceramic specimens, one of the materials tested in thisstudy.43-45 In another study, zirconia specimens were placedin saline solution at 508◦C and 958◦C for 3 years and in dis-tilled water at 1218◦C for 2000 hours, and they did not ex-hibit any changes in flexural strength, even after 30 months oflow-temperature degradation treatment.46 Therefore, becauseno dental substrate (dentin or enamel) was used, and the aimwas to test the effectiveness of surface cleaning regimens be-fore luting, no artificial aging procedures were performed inthis study.

One limitation of this study was that a chemical analysissuch as XPS was not used to observe to what extent the con-tamination reached the irregularities and contaminated ceramicsurfaces and the amount of the removal of the contaminant.Rather, the main objective focused on the effectiveness of thenewly developed universal CP in terms of bond strength andelectron microscopic observation of any contaminant present onthe ceramic surfaces. Further studies on the cleaning effective-ness of the most commonly preferred surface decontaminantson conditioned real ceramic restorations ready for adhesivecementation, but intraorally tried for final control should beconducted, although standardization of contaminants are dif-ficult for clinical studies, and production of duplicate ceramicrestorations may not be practical and economical.

Conclusions

From this study, the following could be concluded:

1. Different ceramic surface cleaning regimens after salivacontamination of the zirconium dioxide revealed similarµSBS as the control group. Cleaning the surface withWS can be recommended as a practical clinical method.

2. The use of hypochloride or CP on the acid-etched lithiumdisilicate glass-ceramic surfaces after saliva contamina-tion could not restore the decreased µSBS values.

3. The use of 0.5% hypochloride on acid-etched leuciteglass-ceramic surfaces recovered the decreased µSBSvalues.

Acknowledgments

The authors of the study would like to thank Prof. Dr. TijenPamir from the Restorative Dentistry Department, Ege Uni-versity, School of Dentistry for conducting the SBS tests andProf. Dr. Bilge Hakan Sen from the Endodontics Department,Ege University, School of Dentistry for her assistance with theSEM analyses.

References

1. Blatz MB, Sadan A, Kern M: Resin-ceramic bonding: a review ofthe literature. J Prosthet Dent 2003;89:268-274

2. Thompson JY, Stoner BR, Piascik JR, et al:Adhesion/cementation to zirconia and other non-silicateceramics: where are we now? Dent Mater 2011;27:71-82

3. Denry I, Kelly JR: State of the art of zirconia for dentalapplications. Dent Mater 2008;24:299-307

4. Yang B, Wolfart S, Scharnberg M, et al: influence ofcontamination on zirconia ceramic bonding. J Dent Res2007;86:749-753

5. Ozcan M, Dundar M, Comlekoglu ME: Adhesion concepts indentistry: tooth and material aspects. J Adhes Sci Technol2012;24:2661-2681

6. Amaral R, Ozcan M, Valandro LF, et al: Effect of conditioningmethods on the microtensile bond strength of phosphatemonomer-based cement on zirconia ceramic in dry and agedconditions. J Biomed Mater Res B Appl Biomater 2008;85:1-9

7. Zhang Y, Lawn BR, Malament KA, et al: Damage accumulationand fatigue life of particle- abraded ceramics. Int J Prosthodont2006;19:442-448

8. Cura C, Ozcan M, Isık G, et al: Comparison of alternativeadhesive cementation concepts for zirconia ceramic: glaze layervs zirconia primer. J Adhes Dent 2012;14:75-82

9. Zhang S, Kocjan A, Lehmann F, et al: Influence of contaminationon resin bond strength to nanostructured alumina-coatedzirconia ceramic. Eur J Oral Sci 2010;118:396-403

10. Piwowarczyk A, Lauer HC, Sorensen JA: In vitro shear bondstrength of cementing agents to fixed prosthodontic restorativematerials. J Prosthet Dent 2004;92:265-273

11. Eiriksson SO, Pereira PN, Swift EJ Jr, et al: Effects of salivacontamination on resin-resin bond strength. Dent Mater2004;20:37-44

12. Aboush YE: Removing saliva contamination from porcelainveneers before bonding. J Prosthet Dent 1998;80:649-653

13. Yang B, Lange-Jansen HC, Scharnberg M, et al: Influence ofsaliva contamination on zirconia ceramic bonding. Dent Mater2008;24:508-513

14. Scientific Documentation Ivoclean, IvoclarVivadent, Schaan,Leichtenstein, 2011;3-10.

15. Klosa K, Wolfart S, Lehmann F, et al: The effect of storageconditions, contamination modes and cleaning procedures on theresin bond strength to lithium disilicate ceramic. J Adhes Dent2009;11:127-135

16. Lendenmann U, Grogan J, Oppenheim FG: Saliva and dentalpellicle-a review. Adv Dent Res 2000;14:22-28



17. Hannig M: Ultrastructural investigation of pelliclemorphogenesis at two different intraoral sites during a 24-hperiod. Clin Oral Investig 1999;3:88-95

18. Quaas AC, Yang B, Kern M: Panavia F 2.0 bonding tocontaminated zirconia ceramic after different cleaningprocedures. Dent Mater 2007;23:506-512

19. Fabianelli A, Pollington S, Papacchini F, et al: The effect ofdifferent surface treatments on bond strength between leucitereinforced feldspathic ceramic and composite resin. J Dent2010;38:39-43

20. Phark JH, Duarte Jr S, Kahn H, et al: Influence of contaminationand cleaning on bond strength to modified zirconia. Dent Mater2009;25:1541-1550

21. Marshall SJ, Bayne SC, Baier R, et al: A review of adhesionscience. Dent Mater 2010;26:11-16.

22. Hannig C, Hannig M: The oral cavity—a key system tounderstand substratum-dependent bioadhesion on solid surfacesin man. Clin Oral Invest 2009;13:123-139

23. Hannig C, Hannig M, Attin T: Enzymes in the acquired enamelpellicle. Eur J Oral Sci 2005;113:2-13

24. Vitkov L, Hannig M, Nekrashevych Y, et al: Supramolecularpellicle precursors. Eur J Oral Sci 2004;112:320-325

25. Hannig M, Hannig C: Does a biofilm free of bacteria, exist insitu? J Parodontol Implantol Orale 2007;26:187-200

26. Schenkels LC, Veerman EC, Nieuw Amerongen AV:Biochemical composition of human saliva in relation to othermucosal fluids. Crit Rev Oral Biol Med 1995;6:161-175

27. Yang B, Scharnberg M, Wolfart S, et al: Influence ofcontamination on bonding to zirconia ceramic. J Biomed MaterRes B Appl Biomater 2007;81:283-290

28. Yang B, Barloi A, Kern M: Influence of air-abrasion on zirconiaceramic bonding using an adhesive composite resin. Dent Mater2010;26:44-50

29. Ozcan M, Cura C, Valandro LF: Early bond strength of two resincements to Y-TZP ceramic using MPS or MPS/4- META silanes.Odontology 2011;99:62-67

30. Tanaka R, Fujishima A, Shibata Y, et al: Cooperation ofphosphate monomer and silica modification on zirconia. J DentRes 2008;87:666–670

31. Lee TH, Ahn JS, Shim JS: Influence of cement thickness onresin-zirconia microtensile bond strength. J Adv Prosthodont2011;3:119–125

32. Oyague RC, Monticelli F, Toledano M, et al: Effect of wateraging on microtensile bond strength of dual cured resin cementsto pre-treated sintered zirconium-oxide ceramics. Dent Mater2009;25:392-399

33. Aboushelib MN, Matinlinna JP, Salameh Z, et al: Innovations inbonding to zirconia-based materials. Part I. Dent Mater2008;24:1268-1272

34. Chen C, Kleverlaan CJ, Feilzer AJ: Effect of an experimentalzirconia-silica coating technique on micro tensile bond strengthof zirconia in different priming conditions. Dent Mater2012;28:127-134

35. Blatz MB, Sadan A, Martin J, et al: In vitro evaluation of shearbond strengths of resin to densely-sintered high- purityzirconium-oxide ceramic after long-term storage and thermalcycling. J Prosthet Dent 2004;91:356-362

36. Magne P, Paranhos MP, Burnett LH Jr: New zirconia primerimproves bond strength of resin-based cements. Dent Mater2010;26:345-352

37. Inaba D, Ruben J, Takagi O, et al: Effect of sodium hypochloritetreatment on remineralization of human root dentine in vitro.Caries Res 1996;30:218-224

38. Naves LZ, Soares CJ, Moraes RR, et al: Surface/interfacemorphology and bond strength to glass ceramic etched fordifferent periods. Oper Dent 2010;35:420-427

39. Comlekoglu ME, Dundar M, Gungor MA, et al: Preliminaryevaluation of titanium tetrafluoride as an alternative ceramicetchant to hydrofluoric acid. J Adhes Dent 2009;11:447-453

40. Santos GC Jr, Santos MJ, Rizkalla AS: Adhesive cementation ofetchable ceramic esthetic restorations. J Can Dent Assoc2009;75:379–384

41. Barghi N, Fischer DE, Vatani L: Effects of porcelain leucitecontent, types of etchants, and etching time onporcelain-composite bond. J Esthet Restor Dent 2006;18:47-52

42. Borges GA, Sophr AM, de Goes MF, et al: Effect of etching andairborne particle abrasion on the microstructure of differentdental ceramics. J Prosthet Dent 2003;89:479-488

43. Probster L, Diehl J: Slip-casting alumina ceramics for crown andbridge restorations. Quintessence Int 1992;23:25-31

44. Papanagiotou HP, Morgano SM, Giordano RA, et al: In vitroevaluation of low-temperature aging effects and finishingprocedures on the flexural strength and structural stabilityof Y-TZP dental ceramics. J Prosthet Dent 2006;96:154-164

45. Ardlin BI: Transformation-toughened zirconia for dental inlays,crowns and bridges: chemical stability and effect oflow-temperature aging on flexural strength and surface structure.Dent Mater 2002;18:59-595

46. Shimizu K, Oka M, Kumar P, et al: Time-dependent changes inthe mechanical properties of zirconia ceramic. J Biomed MaterRes 1993;27:729-734


Effect of Different Cleaning Regimens on the Adhesion of Resin to Saliva-Contaminated Ceramics.

Documents

Transcript of Effect of Different Cleaning Regimens on the Adhesion of Resin to Saliva-Contaminated Ceramics.