7

Long-Term Cytotoxicity ofDental Casting Alloys

ßobn C. Wataha. DMD,Petra E. Lockwood''Steven K. Nelson, DMD^Serge Bouiliaguet, DMD''

Purpose: Current in vitro cytotoxicity tests for dental casting alloys, may he inadequatebecau5e they use a short-term test (generally < 168 h) for a material that is present in themouth for years. A modified in vitro test was used to test the cytotoxicity of 8 dentalcasting alloys and 3 pure metals. This modified test attempted to simulate the long-terrrteffects of dental casting alloys in the mouth. iVlaterials and Methods: Alloys were testedin cell culture either immediately after polishing (initial) or after 10 months ofconditioning in a hiologic medium containing serum proteins. The cytotox ici ties of the 2conditions were then compared. Alloys were repolished to assess the reversibility of the10-month conditioning. Results: Alloys with little initial cytotoxicity showed no 10-monthcytotoxicity. Alloys with significant initial toxicity showed significantly less toxicity at 10months. No alloy increased in toxicity after conditioning. Repolishing multiple times didnot necessarily reestablish the initial cytotoxicity response. Conclusion: The resultsindicate that short-term cytotoxicity tests may not accurately measure the long-termcylotoxitity of these types of alloys. Int ¡ Prosthodont 1999;! 2:242-248.

In vitro tests are used to screen or evaluate new ma-terials for biocompatibility. These tests are used be-

cause they are faster, less expensive, less controver-sial, and offer more experimental control than animalor human tests.̂ In the past 20 years, many new alloyformulations that are fundamentally different from thetraditional gold-based alloys have been developed.

'Associate Professor of Dentistry, Department of Oral Rehabilita-tion, Medical Coiiege of Georgia School of Dentistry, Augusta,Georgia.''Research Assistant, Medical Coiiege of Georgia Schooi ofDentistry, Augusta, Georgia.'•Assistant Professor of Dentistry. Departmertion. Medical College of Georgia School o\

nt of Oral Rehahilila-f Dentistry. Augusta,

nt Professor of Dentistry.tion. Medical College of GeorgiGeorgia.''Aaociate Professor of Dentistry, Department of ResloraliveiJentistry, Uniyersily of Geneva Schooi of Denlisiry, Geneva,Switzerland,

Reprint requests: Dr lohn C. Wataha, Department of Oral He-habiiitalion, Medicai College of Georgia Schooi of Dentistry,Augusta, Georgia 30912-1260, fax.' + 17061 721-8349. e-maii:[email protected],edu

In vitro tests for cytotoxicity of dental casting alloyshave been used for the past 10 to 12 years to screennew alloy formulations and compare them with ex-isting formulations.^ In the absence of clinical expe-rience, the in vitro tests have offered a way to quicklycompare these new formulations with traditionalgold-based systems.'-^

However, the relevance of in vitro tests has beenquestioned because they often do not agree with ani-mal tests or clinical experience."* In spite of this fun-damental disadvantage, in vitro tests remain an Integralpart of any biocompatibility testing scheme because ofthe ethical and legal probiems associated with testinga new material in humans first.^'^ In vitro tests remainthe single best starting point for the evaluation of thebiocompatibility of a new biomaterial, and recent re-search has tried to improve their relevance.'

One important condition that limits the relevanceof in vitro tests is the duration of the exposure of thematerial to the cell cultures. Most "direct contact" invitro tests, which place the material directly adjacentto cells, are less than 168 hours long because of the

Tlie International locrral of Prostfiodontii 2 4 2 Volume 12, Nii

Long-Term C>1otoxicily of Dental Casiing AIIOV

a

Cells

Direct contact

t _̂ Assess^etfect

Figs 1a to 1c Materiais can betestedin vitrob/iaf piacing tfiematenai into direct contact witb cells for a specific lime and tbenassessing the eftect, ¡b) by eluting ttie matenai into a mediumwithout celis and tben testing the eluate on celis (indirect con-tact), or ¡c) by conditioning the matenai in a medium and tbentesting the material directly, t = time tor which material contactsceils.

Indirect contact

Add conditioned medium

t I I ^ Assess"effect

Modified direct contact

various limitations involved in culturing cells forlonger periods of time (Fig la). ' " ' " These limitationsinclude microbial contamination, loss of potency ornutrition of the medium, or cell overgrowth. The rel-atively short contact times are not relevant to mate-rials such as dental casting alloys, which are presentin the mouth for years.

One alternative to direct contaa testing is indirectcontact testing (Fig 1 b|. In this strategy, the tested ma-terial is cultured with cell culture medium (but nocells) for a specific length of time, then the medium istransferred to the cells for toxicity testing in a secondstep. Using the indirea strategy, it is possible to "age"the material in a biologic medium and change the ex-traaing medium several times, testing its toxicity on cellcultures periodically. This strategy has been used tötestthe cytotoxicity of a variety of dental casting alloys.' 'However, the indirect method also has disadvantages.The medium must be changed on the cells. This pro-cedure can itself kill a percentage of the cells. Further,fhe indirect system is not a dynamic system that allowsthe material and cells to interact over time.

A second alternative to traditional direct contacttesting is to first condition the material in a biologicmedium, then use a direct contact test to evaluate thecytotoxicity of the material, not the conditioningmedium (Fig 1c). This modified direct test allows thematerial and cells to dynamically interact, limits thepractical problems of the indirect tests, and allows thematerial to be aged to give more relevant infomiation.

The purpose of the current study was to use a mod-ified direct test to evaluate the in vitro cytotoxicity ofa series of dental casting alloys and compare the results

with the traditional direct test. Thus, the current studyhad 3 specific aims: ll>to compare the cytotoxicity ofa series of dental casting alloys with or without 10months of conditioning in a biologic medium; (2) tocompare the mass lost from the alloys during the 10-month conditioning; and í5Jtogain insight into the re-versibility of the lO-month conditioning by repolish-ingthe alloys usingvarious means and rete sting themfor cytotoxicity.

Materials and Methods

Casting alloys were conditioned for 10 months in a bi-ologic medium, then tested using the modified directtest strategy (Fig 1c). The mass lost from the alloys dur-ing the 10-month conditioning was also measured toattempt to correlate any cytotoxicity with loss of mass.

A total of 8 types of dental casting alloys were used,representing the \vide variety of alloy formulationscurrently used in clinical practice (Table 1 ). The nobilityof these al loys ranged from 0 to 98 wt̂ -'o. By AmericanDental Association det"initions, 4 of the alloys were highnoble, 3 were noble, and one was predominantly basemetal.'- The alloys were obtained from Metalor. cast(n = 12), and polished using Tripoli and Rouge (Schein)on 2.5-cm rag wheels to a high luster. Fach sample wasa disk 5.5 mm in diameter and 2.5 mm thick with a 1.0-cm stem attached to one face of the disk. Previously es-tablished techniques were used to clean and disinfectthe alloys before cell culture testing so that chemicalor biologic agents would not affea the outcome ofthe cell culture tests.̂ Control materials included werepure palladium (Pd) as a nontoxic control, pure

". Number 3.1999 2 4 3 The International lournai of Prosthodontics

Lorg-Term Cytolo«lcity of Dental CaîtirR Alloys Wataha et al

Table 1 Compositions of Alloys (wt%)

Alloy Ag Ga Other Nobility

Au-PtAu-PdPd-CU'GaAu-Cu-AgPd-AgAu-Ag-CuAg-PdNi-Cr

tr = trace l<

13,724.525.065,0

t.ODUt%l.

85.951.5

2.072.0

56.03.0

6.99.8

11.86.5

8,55,5

tr1.54.5tr

2.0

2.0

38.478.5

61.55.0

23.0

trtr

2.01.7tr

Sn2.0

SniQ.OIrtrIrtrBtrBe 1.8Cr15.0

97-6-389,980.5 'S75.66 1 , 5 ^61,426.0 M0

copper (Cu) as a toxic control, pure nickel (Ni) as amoderately toxic control, and Teflon (DuPont) as a non-toxic control that was cleaned and disinfected but notpolished. The shapes of the control materials wereidentical to the shapes of the alloy samples.

Balb/c 3T3 mouse fibroblasts (CCL 153, clone A31.American Type Culture Collection) were used in theexperimentstotestalloycytotoxicity. These cells weremaintained in Dulbecco's Modified Eagle's Medium(Cibco BRL), 3% NuSerum (Collaborative Biomédical),gentamycin (10 pg/mL, Cibco BRL), penicillin (125units/mL, Cibco BRL), and streptomycin (125 jJg/mL,Cibco BRL). These cells were selected based on theircommon use in alloy cytotoxicity tests and their ap-proved status by the International StandardsOrganization (ISO).'^ Furthermore, these cells retain 2traits of importance to cells in vivo: they are contactinhibited as they become more dense in culture, andthey are nontumorogenic. After counting the cells witha hemocytometer, cells were plated into 24-well tis-sue culture plates at 12,500 cells/cm^ in 0,5 mL ofmedium. The alloy samples were added immediatelyso that the edges ofthe disk and polished face withoutthe stem were in contact with the medium (Fig 1 a). Thesurface area of the alloys in contact with the mediumwas 0.67 cm^, which gave a surface area-volumeratio of 1.3 cm^mL, This sud̂ ace area-volume ratio iswithin the range recommended by the ISO (0.5 to 0.6cm^mD.^'The cells and alloys were incubated at37°C for 72 hours, after which the cytotoxicity wasevaluated. Cells grew only around the alloy speci-men, and not underneath it (Fig 1 c).

Cytotoxicity was estimated by measuring the activ-ity ofthe mitochondria in the cells using a previously

established MTT technique.'" Active cells around thealloys converted the yellow, soluble MTT dye (Sigma)to a blue formazan compound insoluble in the cells.The amount of cell activity was determined by dis-solving the blue formazan into dimethyl sulfoxide andquantifying the amount of dye using a spectropho-tometer (Vmax, Molecular Devices). The overall ac-tivity ofthe cells around the alloys was then expressedas a percentage ofthe Teflon negative controls, 100%being defined as nontoxic. The MTT test has previouslybeen shown to be reproducible and a good estimatorof cell activity.'^

Alloys were tested using the above format with orwithout conditioning in cell culture medium (withserum as described above) for 10 months. A period of10 months was selected on the basis of elemental re-lease studies that showed that elemental release fromthese alloys was constant during this time.'^ For theconditioning of alloys, cell culture medium was placedaround the alloys in the format described above with-out cells (Fig 1 c). The medium was changed every 30days to avoid microbial overgrowth, excessive evap-oration, and drift of the pH of the medium. After the10-month conditioning was complete, the alloys wereplaced immediately into the cell culture test withoutany repolishing or cleaning. Thus, strict aseptic con-ditions had to be maintained during the conditioning.

The mass lost during the 10-month conditioningwas assessed by measuring the mass before and afterconditioning with a balance sensitive to 0.01 mg. Afterconditioning and cytotoxicity evaluation, the sampleswere rinsed twice with water and dried at 60°C for 48hours to constant mass, and the mass was remeasured.The change in mass of the Teflon controls was assumed

The irternalionEll (ournal of Prostliodorlii 244

Wataha et al Long-Term C>fotoxicit>' of Dental Casting Alloys

Fig 2 Comparison of inrtal (0 mo] cytotoxicity wrth cytotoxjcity after the materialhadbeenexposed to cell culture meflium for 10 months (10 mo), then repolisbedwith Rouge (PP], then with Tripoli and Rouge (TRP1). and then mth Tripoli andRouge again (TRP2]. The activity of the Balb/c fibroblasta was assessed after thematerial was in contact with the cells lor 72 hours. Cell activity was measured byWTT and expressed as a percentage of the Teflon negative controls. Lower-caseletters indicate statislicai differences among groups; within each group, columns withtbe same letters were not statistically different (ANOVA, Tiikey a = 0,05).

to be zero. The recorded increase in weight of Teflonwas used to adjust the change in mass ofthe aiioy sam-ples to compensate for any medium residue that wouidincrease the mass and mask the true ioss of mass fromtheailoys. Differences in mass ioss from the various al-ioys were assessed using anaiysis of variance ¡ANOVA)and Tukey muitipie comparison intervais (a = 0.05).The mass iost from each ailoy reiative to tbe Tetloncontrois was assessed using a 2-sided (test (a = 0.05).

After the 10-month toxicity test and mass meas-urements were complete, the alioys were repoiishedusing Rouge oniy. The ailoys were then cieaned,disinfected, and retested for cytotoxicity. The ailoyswere polished using both Tripoii and Rouge (TRPl).and were cieaned, polished, and retested. Finally, theaiioys were repoiished again using Tripoli and Rouge(TRP2), and were cieaned, polished, and retested. Thecytotoxicity of tbe ailoys after the various repolish-ing procedures was compared to the cytotoxicity be-fore and after conditioning using ANOVA and Tukeymultiple comparison inten/ais (a = 0.05),

Results

Fffect of 10-Month Conditioning onCytotoxicity and Mass

Conditioning the alloys for 10 months in ceil culturemediumchanged the cytotoxicity of some but not ailof the aiioys (Fig 2). The Pd-Ag, Au-Cu-Ag, Ag-Pd. andNi-Cr aiioys did not change significantiy in cytotoxi-city after conditioning. iHowever, the Au-Pt, Au-Pd,Pd-Cu-Ga, and Au-Ag-Cu aiioys all decreased in cy-totoxicity (or increased in cell activit>') after poiishing(statisticaiiy significant, P< 0.05). The amount of in-crease in ceii activity ranged from neariy 70% for theAu-Ag-Cu alioy to 10% for the Pd-Cu-Ca alioy. Forthe controi materiais, the Pd showed a smaii decreasein ceii activity, the Ni a iarge increase, and the Cu re-mained severeiy cylotoxic. In no case did an alloy be-come more c>'totoxic after conditioning.

The aiioys lost mass as expected during the 10-month conditioning in ceii cuiture medium (Fig 3),

Number 3,1999 245 The Intematiorîal loumal of Prosthodontics

Long-Term Cytotosicityof Denfal Casting Alloys

Fig 3 Total mass lost from eachalloy during the 10-month expo-sure to cell culture medium. Massloss was measured by a differencein mass before and affer exposureand was normalized to the surfacearea of the alloys exposed to thecell culture medium. The mass lossfor the Teflon was arbitrarily set alzero. The mass loss for Cu was349 jjg/cm^, • - significant differ-ence Irom Teflon ((tests a =0.05];lower-case letters indicate differ-ences between alloys (ANOVA,Tukeya = 0.05¡; columns with thesame letters were not stafisticallydifferent.

The mass loss ranged from < 10|jg/cm-forlhe Ni-Crto nearly 350 [jg/cm- for Cu. Of the alloys, the largestmass lost was from iheAu-Pt alloy. The mass lost fromthe alloys did nol correlate well with the cytotoxic-ity. For example, the Au-Ag-Cu alloy showed rela-tively low ma55 l055 (Fig 3¡, bul a significant toxicityinitially (Fig 2). The Au-Pt alloy showed an initial tox-icity of about 20% (80% cell activity), which wasmuch less than the Au-Ag-Cu alloy, but the Au-Ptalloy released more mass.

Fffect of Repolisbing on Cylotoxicity After10-Month Conditioning

Repolishing after the 10-month conditioning affectedcytotoxicity, but in several cases the initial cytotoxic-¡ty could not be reestablished by repolishing (Fig 2]. Forexample, the Au-Pt alloy showed an initial cell activ-ity of about 78% of the Teflon controls. After repol-ishing with Rouge, the cell activity was significantlyhigher than initially, at 110% (P< 0.05). Further pol-ishing with Tripoli and Rouge (TRP1 and TRP2¡ re-duced the cell activity significantly, to 88% to 90%, butnot to the original level of 78%. An even more dramaticexample is the Au-Ag-Cu alloy, which was severelytoxic initially but retained a 90% cell activity compareidto Teflon after all of the repolishing. For other alloys,such as the Pd-Cu-Ca alloy, the Tripoli and Rouge pol-ishing restored the initial cytotoxicity. Still other alloys.

such as the Pd-Ag alloy, showed statistically similar cy-totoxicitiesaftereven a Rouge polish. In general, alloysfor which the 10-month conditioning had little effecton cytotoxicity were restored to their original cyto-toxicity by polishing. Alloys for which the 10-monthconditioning significantly decreased toxicity oftencould not be restored to their original cytotoxic be-havior, Forthe pure metal controls, repolishing had theexpected effects. The Pd was essentially not toxic be-fore and after repolishing. The Ni samples were cyto-toxic initially and less toxic after the 10-month condi-tioning (Fig 2], but gradually increased in cytotoxicity(decreased in cell activity) after repolishing. The Cusamples were uniformly toxic for all conditions.

Discussion

The current study estabi ished that a 10-month condi-tioning of common dental casting alloys can signifi-cantly change their cytotoxicity in a direct contact for-mat (Fig 2). Hlowever, the effect was not uniform for allcompositions. Interestingly, no alloy became moretoxic from the 10-month conditioning. This result agreeswith reports of decreasing element release from thesealloys' ̂ as well as from other types of al loys,' ''-' ̂ Thus,it appears that for some dental casting alloy composi-tions, element release decreases from initially highervalues. The cytotoxicity of the alloys may or may notdecrease because of the redtiction in element release.

Tlie Inferrational lournal of Prosthodontii 246

Wataha et al Long-Term Cvtoloxicity oí Dental Casting Alloys

but it will not increase. Iftheelement release was lowenough to avoid toxicity initially, then it did not appearto increase sufficiently after conditioning to increase thecytotoxicity. The decreasing cytotoxicity obser\ ed afterconditioning also agrees with reports tbat some alloysbecome increasinglv noble at the surface.'^ '^ The in-creasing nobility probably limits further release of ele-ments and in turn reduces the c>1otoNÍcit>'. The low cy-totoxicity of all of these alloys after 10 months ofconditioning correlates with the in vivo experiencethat all have been used successfully in the mouth.

From an in vitro testing standpoint the results of thecurrent study show that without conditioning, directcytotoxicit>' tests may give more oiotoxic resultsthan conditioned alloys. This fact does not mean thata preconditioned evaluation is not appropriate, butit simply should be kept in mind when interpretingthese types of results and deciding whether a new for-mulation should be "screened out." On the otherhand, tests that use a 10-month conditioning areinherently cumbersome because of the lengthy con-ditioning time; tests that could accelerate this condi-tioning would be desirable.

Mass loss from the alloys in the current study wasquite iow (Fig 3) relative to metals like Cu, which areknown to corrode signit"icantly.-° This lack of mass losscorrelates with the observation that most alloys showno overt signs of corrosion in vitro or in vivo. The lackof con-elation between mass lost and initial cytotoxi-city is not surprising since cytotoxicity is a function ofexactly what is released, how much, and over v\ hat pe-riod of time.'^ However, mass loss is probably stiil auseful indicator of the risk that the alloy poses to thebody, and it is much simpler than biologic testing.

The observation in the current study that repolishingcould not restore the original cytotoxicity was inter-esting and surprising (Fig 2). One must consider whatthis result means about the surface of the alloy after 10months in a biologic environment. Several possibilitiesexist, noneof whicb can be confirmed from the currentdata. It is possible that the original surtace of the alloysis unique and was irreversibly changed by exposure tothe biologic environment. In this case, repolishingwould not restore the surface. However, it is also pos-sible that the original surface was not unique, but wasmodified by the biologic conditioning to a depth greaterthan that reached by the repolishing techniques. Thissecond possibility seems unlikely given the low massloss from the alloys and the multiple and aggressive pol-ishings that were used. Previous reports have shownthat 96 hours of exposure of similar alloys to cell cul-ture medium alters the surface to only 100 Â,̂ ^ but nodata is available on how deeply a 10-month condi-tioning would affect the alloy. Additional data areneeded to address this question.

Table 2 Cytotoxicity of Alloys (% ot Teflon Control):Direct vs Indirect Method After 10 Months in CellCulture

Alloy

Au-RAu-PdPd-Cu-GaPd-AgAu-Cu-AgAu-Ag-CuAg-PdNi-CrPdNiCu

Direct

106(15)109i3)107(7)104(12)as (7)77(8)97(5)

104(7)130 <7)104(9)

5(2)

Indirect"

32(13)101 (10)105(13)92(12)54(17)35(27)

101 (14)99(11)

100(11)71 (26)

- -9(4)

Statistical difference*

YesYesNoyesYesYesNo S BNoNo ' i mYesNo

Past reports using high-noble and noble dentalcasting alloys have shown unequivocally the repro-ducibility of alloy c>totoxicity after repolishing withRouge.^ Thus, the current results were surprising.However, the difference in the current work was the10-month exposure to the cell culture medium, whichwas only 71 hours in the previous study. The currentstudy cannot define the mechanisms for these differ-ences, but the practical consequence Is that oneshould not assume reproducibility by polishing.

The direct contact testing strategy in the currentstudy was not always in agreement with indirect testson the same alloys done with identical cells, incuba-tion conditions, and serum-containing cell culturemedium.'^ As Table 2 shows, a comparison of the in-direct tests of the previous studies with those of the cur-rent study at 10 months indicates signit'icant discrep-ancies. Forexample, the Au-Pt alloy had a cell activityof only 32%ofTetlon when tested in the indirect for-mat at 10 months. When tested directl\' the activity wassignitlcantly higher (106%). For each discrepancy, theindirecttest gave more cytotoxic results. Although thecauses of these discrepancies are not known, theyprobably are a consequence of procedural differencesbetween the formats. For example, in the direct method(Fig 1 a) tbe cells may have time to divide and estab-I Ish themselves before element release reaches higherlevels. Previous reports have shown that the cell den-sit>'directly affects toxicity.'' In the indirect format (Fig1 b), the full brunt of the element release Is presentedto the ceils at the outset Other variables may also playa role in these differences. The differences in Table 2present the problem of which testing format Is more ap-propriate. The results ofthe current study cannot pre-sent evidence that one format is better than the other.However, the direct format has the advantages of al-lowing a dynamic exchange between the material andthe cells and not requiring changing the medium onthe cells, which can lead to artifacts. Furthermore, it

.2 , Number 3,1999 247 The Intemaiionai loumal of Prosthodontics

Long-Term Cytotoxicity of Dental Casting Alloys Wataha et al

is intuitively more relevant to use a test that places thematerial in direct contact with cells because it dupli-cates the in vivo conditions.

The current study has shown that a modification ofthe direct contact tests for in vitro cytotoxicity signif-icantly changes the results obtained. Conditioning ofdental casting alloys for 10 months in ceil culturemedium generally decreased their cytotoxicity. Themass lost from these al loys during this period was gen-eral ly 5 to 50 pg/cm^, far lower than metals such asCu, which are known to have significant intraoralcorrosion. Interestingly, repolishing the alloys after the10-month conditioning sometimes did not restore theoriginal cytotoxic response, indicating that the surfaceof the alloys may be permanently altered in someway by the biologic conditioning. Finally, the directtesting format used in the current study did not alwaysagree with the indirect format that has been advocatedby the ISO as an option for cytotoxicity testing.

Acknowledgments

The authors (hank Melaior Corporation, the BiocompatibilityCroup at tlie Medicai College of Georgia, and J. M. Meyer at theUniversity of Geneva for Lheir support of thii work.

References

1. Hanks CT, Wataha JC, Sun ZL. in vitro modeis of biocompati-bility: A review. Dent Mater 1996;12:186-193.

2. Craig RG, HanksCT. Cytotoxicity of experimental casting alloysevaluated by cell culture test!, i Dem Res 1990;69:I,539-1,542.

3. Craig RG (ed). Restorative Dental Maierials. St Louis: Mosby,1997:383^07.

4. Wennberg A, Mjör IA, Hensten-Pettersen A. Biological evalua-tion of dental restorative materials: A comparisori of d i fièrent testmethods. J Biomcd Mater Res 1983;17:23-36.

5. Schmalz C. Modern concepts in biocompatibility tesling ofrestorative materials. Trans Acad Dent Mater 1996,9:170-179.

6. Wataha ;C, Hanks CT. Biorompatibility testing: What can we an-ticipate? Proceedings of the Third international Congress onDental Materials, 4-8 Nov 1997, Honolulu, Hawaii. Trans AcadDent Mater 1997:109-120.

7. Niemi L, Hersten-Pettersen A. in vifracytotnfticiry of Ag-Pd-Cjcasting alloys. JBiomed Mater Res 19B5;19:549-5&1.

8. Bumgardner |D, Lucas LC, Tilden AB. Tosicity of copper-baseddental alloys in cell culture. ) Biomed Mater Res 1989;23:1,103-1,114.

9. Wataha |C, Craig RG, Hanks CT, Precision of and new methodsfor testing in vííro alloy taxicity. Dent Mater 1992;8:65-71.

10. BersteiriA, Benauerl, Marx R, Geurtsen W. Human cell culturestudies with dental metallic materials. Biomaterials 1992;13:98-100.

11. Wataha JC, Lockwood PE, Nelson SK, Rakich DR. in vitro cyto-toxicity of dental casting alloys over 8 months. ) Oral Rehabil(forthcoming).

12. Council ori Dental Materials, Instruments, and Equipment.Classification system for casting alloys. | Am Dent Assoc 1984;109:766.

13. International Standards Organizatiori. Biological evaluation ofmedical devices, part S: Tests forcytotoxicity—/n Wiromethods.Geneva: ISO publication 10993-5:]992{E1.

14. PearseAG. Histochemistry: Theoretical and Applipd Baltimore:Williams & Wilkins, 1972:921-961.

15. Wataha JC, Hanks CT, Craig RG. The in vitro effects of metalcations on eukaryotic cell metabolism. | Biorred Mater Res 1991 ;25:1,133-1,149.

16. Wataha JC, Lockwood PE. Release of elements from dental cast-ing alloys into cell-culture med ¡um over 10 rnontbs. Dent Mater1998:14:158-163.

1 7. Cerstorfer |G, Sauer KH, Passler K. ion release from Ni-Cr-Moand Co-Cr-Mo alloys. Int | Prosthodont 1991 ;4:152-158.

18. Goehl ichV,MarekM. Corrosion behavior of Pd-Cu and Pd-Coalloys in synthetic saliva. Dent Mater 199O;6:1O3-110.

19. Wataha ]C, Malcolm CT. Effect of alloy surface composition onrelease of elements from dental casting alloys. J Oral Rehabil1996;23:583-589.

20. lohansson Bl, Lemons JE, HaoSQ. Corrosion of dental copper,nickel and gold alloys in artificial saliva and saline solutions.Dent Mater l989;5:324-328.

2 1 . Wataha JG, Hanks CT, Craig RG. The effect of cell monolayerdensity on the cytotoxicity of metal ions which are released fromdental alloys. Dent Mater 1993;9:1 72-1 76.

The international Joumsi of Prosthodontics 248 Voiume 12, Nur