dental materials

Dental Materials 18 (2002) 311317 www.elsevier.com/locate/dental

Cytotoxicity of orthodontic bands, brackets and archwires in vitro

O. Mockers*, D. Deroze, J. Camps
Me diterrane e, 27 boulevard Jean Moulin, 13005 Marseille cedex 5, France U.F.R. Odontologie, Universite Received 28 March 2000; revised 12 February 2001; accepted 9 April 2001

Abstract Objectives: In orthodontic therapy, different materials are used and subjected to a damp oral environment, which can modify their properties. In order to evaluate the biocompatibility of metallic and non-metallic orthodontic appliances their in vitro cytotoxicity has been measured. Methods: Twenty-eight new and nine clinically used materials, including brackets, molar bands and archwires were used. The metallic materials were made of stainless steel, gold-plated steel, pure titanium, nickeltitanium, titaniummolybdenum and silver-based soldering alloy. The non-metallic materials were in polycarbonates and ceramics. After a release period of the material in the culture medium (0.1 mg/ ml) for 3 and 14 days, the viability of broblasts L929 cultivated with this medium was compared to negative control with MTT assay. Results: The results showed the non-cytotoxicity of the materials. The metallic and non-metallic materials were similar in terms of cytotoxicity. The cytotoxicity of clinically used samples was equivalent to that of the same non-used samples, except a cytotoxic sample, at 14 days, corresponding to a soldered and clinically used molar band. The 3 day results were different from the 14 day results in six cases out of 37. Signicance: In spite of the presence of one cytotoxic sample, the orthodontic materials can be considered as non cytotoxic. However, the practitioner should pay attention to the composition and the polish of soldering silver-based alloys containing copper and zinc in order to limit cytotoxic ion release. The cytotoxicity of the used sample related to ion release might be related to some clinical sub-acute effects related with orthodontic materials, thus a long term release period may be suitable to evaluate in vitro the sub-acute clinical effects of alloys. q 2002 Academy of Dental Materials. Published by Elsevier Science Ltd. All rights reserved.
Keywords: Cytotoxicity; Orthodontic bands; Orthodontic brackets; Archwires

1. Introduction The choice of materials available in orthodontics depends on factors such as physical, mechanical and biological properties [1]. The allergic effects of the orthodontic materials are well described nowadays. The nickel-based alloys, the latex-based elastic bands and the acrylic-resins are, by far, the most allergic orthodontic materials [25]. On the contrary, the cytotoxic effects of orthodontic materials have been less extensively studied with inconsistent results. In vitro experiments assessing the cytotoxicity of orthodontic materials by agar overlay showed the non cytotoxicity of orthodontic wires. The bands, however, were cytotoxic because of the silver- and copper-based brazing alloys used to manufacture the bands [6]. A more recent study, testing the in vitro cytotoxicity of orthodontic wires concluded that the wires may be considered as non cytotoxic [7].
* Corresponding author. Tel.: 133-4-91-78-46-70; fax: 133-4-91-69-8964.

Regarding the dental alloys, the cytotoxicity tests already allowed us to highlight unsuspected effects of some developing alloys. For example, zinc was identied as highly toxic within an alloy [8]. The materials used in xed orthodontic appliances are metallic (alloys) and non metallic (ceramics, composites and polycarbonates). Many alloys may be used in orthodontics, for example, stainless steel or nickeltitanium and lately titaniummolybdenum. These alloys are often placed in the oral cavity of patients for 12 years and undergo several corrosion phenomena [1,9,10]. In vitro experiments studying the corrosion of orthodontic appliances by using articial saliva showed iron, nickel molybdenum and chromium releases, which increased 57 times, compared to saline solutions usually used [10,11]. In vitro cytotoxicity tests are generally designed by International Standard Organization [12] to evaluate the acute cytotoxicity of a material but they also may be helpful to understand the pathogenicity of some subacute effects. In some cases, the corrosion products of the orthodontic appliances may even cause subacute effects such as glossitis,

0109-5641/02/$22.00 + 0.00 q 2002 Academy of Dental Materials. Published by Elsevier Science Ltd. All rights reserved. PII: S 0109-564 1(01)00055-0

312 Table 1 List of the tested orthotodontic molar bands Stainless steel molar bands Sample number 1 2 3 4 5 6 7 8 9b 10 b Trademark a Ormco Ormco R.M.O R.M.O Dentaurum Dentaurum G.A.C G.A.C G.A.C G.A.C

O. Mockers et al. / Dental Materials 18 (2002) 311317

Condition New Used New Used New Used New Used New Used

metallic tastes, bleeding and inamed or hypertrophied gingivae which cannot be clinically distinguished from the gingivitis of a bacterial etiology [1015]. These subacute effects, due to corrosion products, may be linked to the cytotoxicity of the materials. The ion release time must be sufcient to measure this cytotoxicity but the norm ISO 10993-5 does not set any precise time. The objectives of this in vitro cytotoxicity study were: 1. to compare the cytotoxicity of metallic and non-metallic brackets; 2. to determine the effects of the oral corrosion on the cytotoxicity of metallic materials by comparing the toxicity of new molar bands to those worn in the oral cavity; 3. to compare the results after a 3 and 14 day release period since the norm ISO 10993-5 does not set any precise time; 4. to report on the results of the subacute effects observed during orthodontic treatments.

a Ormco (Glendora, USA); R.M.O (Denver, USA); Dentaurum (Ispringen, Germany); G.A.C (New York, USA); Nobil Metal (Villafranca d'Asti, Italia). b Soldered bands with silver-based soldering alloy of Nobil Metal.

Table 2 List of the tested brackets Brackets Sample number 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26

a

Trademark a Ormco Ormco R.M.O R.M.O Dentaurum Dentaurum G.A.C G.A.C R.M.O Dentaurum Dentaurum Dentaurum G.A.C G.A.C Leone Ormco

Composition Stainless steel (S.st) Stainless steel Stainless steel Stainless steel Stainless steel Stainless steel Stainless steel Stainless steel Gold plated steel Pure titanium Ceramics (Cer) Polycarbonates 1 S.st Polycarbonates 1 Cer Polycarbonates 1 S.st 1 Cer Polycarbonates 1 glass Polycarbonates

Condition New Used New Used New Used New Used New New New New New New New New

2. Materials and methods In this study, the cytotoxicity of 37 orthodontic samples of metallic and non-metallic composition was studied. The samples included molar bands, brackets and archwires of different brands. The metallic materials were of stainless steel, gold-plated stainless steel, nickeltitanium, titaniummolybdenum, pure titanium and silver-based soldering alloys. The non-metallic materials were made of polycarbonates and ceramics. Twenty-eight orthodontic samples were new and nine were clinically used for a period of 1525 months. The list of the samples is given in Tables 13. Pure copper (Engelhard Clal, Paris, France) as well as Rex-V (Jeneric Pentron, Wallingford, USA) were used as controls since their cytotoxicity and their mode of corrosion are well dened. Copper represented the positive control material (cytotoxic response) and Rex-V the negative control material (non-cytotoxic response) [7]. The samples (n 3 for each group) were sterilized according to the protocol dened by the ISO 10993-5 norm. Sterilization was carried out by autoclave and by Gamma rays for the polycarbonate brackets. The clinically used bands were sterilized the same way as the new materials. For the four clinically used brackets (samples 12, 14, 16 and 18), the bonding composite (which is known to be cytotoxic [16]) was removed by passing rapidly trough a ame and then by brushing before sterilization. A pilot study on new brackets showed that this treatment did not modify the outcome of the test. To meet ISO requirements, the ratio between the sample weight and the volume of the extraction solution was 0.1 mg/ml because the surface area of the samples was not easily measurable. The samples were placed at 378C, in closed waterproof test tubes with

Leone (Firenze, Italia).

Table 3 List of the tested orthodontic archwires Archwires Sample number 27 28 29 30 31 32 33 34 35 36 37

a

Trademark a Ormco Ormco R.M.O R.M.O Dentaurum Dentaurum G.A.C G.A.C 3M-Unitek Ormco Wilcock

Composition Stainless steel Nickeltitanium Stainless steel Nickeltitanium Stainless steel Nickeltitanium Stainless steel Nickeltitanium Gold plated steel Titaniummolybdenum Stainless steel

Condition New New New New New New New New New New New

3M Unitek (Saint Paul, USA); Wilcock (Australia).

O. Mockers et al. / Dental Materials 18 (2002) 311317

313

E-MEM (Eagle's minimum essential medium, Gibco, Cergy-Pontoise, France) supplemented with 10% NuSerum (Serovial, Neuf-Brisah, France), penicillin (125 units/ml), streptomycin (125 mg/ml) and glutamine (2 mmol/l). Two release periods were tested because the norm ISO 10993-5 does not determine any precise time: 3 and 14 days. Mouse broblast cells L929 (NCTC, CCL1, Flow Laboratories, Paisley, UK) at 1 10 5 cells/cm 2 were seeded in 96 well-plates of 12 columns. In our study, four 96 well-plates were used in triplicates to test all the samples at 3 and 14 days. Following 24 h of incubation, the pure culture medium was replaced by the culture medium containing the release products. At the same time, for each plate, the culture medium of the control column, cultivated with plain medium, was also renewed. An additional incubation period of 24 h was carried out before performing the cytotoxicity test. Cytotoxicity was determined by the use of a mitochondrial coloring of living cells (MTT assay). After incubation, the 100 ml/well of extraction liquid were replaced by 100 ml/well of a 0.5% solution of 3-(4,5-dimethylthiazol2-yl)-2, (-diphenyl tetrazolium bromide) (Sigma Chemical Company, St Louis, USA). After incubation for 3 h at 378C, the solution was removed and the intracellular formazan crystals were solubilized with 100 ml/well of the solubilization solution of the MTT kit [Triton X-100 plus HCl 0.1 N (10%) in isopropanol (90%)] (Sigma). The amount of the coloring produced was determined by a spectrophotometer of a wavelength of 570 nm (MR 700, Dynamytech Lab. Inc., Europe). The optic density (OD) of each well was proportional to the amount of coloring. The degree of cytotoxicity for each sample was determined according to the reference column represented by the cells with a pure culture medium. Because of interactions between the two factors (exposure time and materials) the use of a two-way analysis of variance (ANOVA) was not possible. In order to compare the cytotoxicity of metallic versus non metallic materials (purpose 1) six ANOVA were used to seek differences among the materials for the bands (samples 110), the brackets (samples 1126) and the wires (samples 2737) at 3 and 14 days. A one-way ANOVA was used to compare the cytotoxicity of each of the nine new materials to the same nine worn materials (samples 118) (purpose 2). A one-way ANOVA was performed for each of the 37 materials to compare the 3 vs. 14 days results (purpose 3). When differences were found, multiple comparisons were performed (NewmanKeuls test) to identify the source of the differences. Statistical signicance was dened at P , 0.05. 3. Results 3.1. 3 Day results (samples 137) Copper, the positive control, showed a severe cytotoxicity (7 ^ 1% cell viability) and Rex-V, the negative control, was non-cytotoxic (99 ^ 1% cell viability). There was a statistically signicant difference among the

Table 4 Cytotoxicity (mean value (SD)) of molar bands at 3 days. The subgroups with the same letter are not statistically different Molar bands at 3 days Sample number Control 1 2 3 4 5 6 7 8 9 10 Mean (SD) 100 (9) 100 (11) 118 (10) 116 (8) 113 (7) 92 (12) 121 (19) 136 (19) 124 (15) 125 (12) 138 (41) Subgroups ab ab bc b b ab bc c bc bc c

Table 5 Cytotoxicity (mean value (SD)) of brackets at 3 days. The subgroups with the same letter are not statistically different Brackets at 3 days Sample number Control 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 Mean (SD) 100 (9) 124 (28) 114 (32) 105 (19) 89 (12) 106 (17) 110 (18) 117 (12) 104 (17) 125 (15) 130 (15) 120 (21) 110 (16) 104 (5) 109 (6) 110 (3) 112 (7) Subgroups ab bc bc ab a ab abc abc ab bc bc bc abc ab abc abc abc

molar bands (P , 0.001) and the NewmanKeuls test showed three different subgroups (Table 4). Only samples 7 and 10 were different from the control and presented a higher cell viability (respectively 136 ^ 19 and 138 ^ 41% cell viability). There was a statistically signicant difference among the brackets (P , 0.001) (Table 5) as well as among the archwires (P , 0.001) (Table 6) and the Newman Keuls test showed three different subgroups in both cases, but none of the samples was statistically different from control. 3.2. 14 Day results (samples 137) Copper, the positive control, showed a severe cytotoxicity (4 ^ 1% cell viability) and Rex-V, the negative control, was non-cytotoxic (96 ^ 1% cell viability). There was a statistically signicant difference among the

314

O. Mockers et al. / Dental Materials 18 (2002) 311317 Table 8 Cytotoxicity (mean value (SD)) of brackets at 14 days. The subgroups with the same letter are not statistically different Brackets at 14 days Mean (SD) 100 (9) 111 (8) 106 (9) 100 (17) 88 (10) 108 (21) 111 (21) 114 (21) 120 (13) 117 (14) 118 (18) 106 (11) Subgroups abc bc abc abc a abc bc bc c bc bc abc Sample number Control 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 Mean (SD) 100 (9) 113 (16) 93 (13) 120 (27) 128 (30) 135 (48) 102 (20) 133 (35) 109 (22) 105 (25) 82 (15) 148 (23) 119 (23) 106 (26) 96 (12) 126 (15) 117 (23) Subgroups abc abc ab abc bc cd abc cd abc abc a cd abc abc abc abc abc

Table 6 Cytotoxicity (mean value (SD)) of archwires at 3 days. The subgroups with the same letter are not statistically different Archwires at 3 days Sample number Control 27 28 29 30 31 32 33 34 35 36 37

Table 7 Cytotoxicity (mean value (SD)) of molar bands at 14 days. The subgroups with the same letter are not statistically different Molar bands at 14 days Sample number Control 1 2 3 4 5 6 7 8 9 10 Mean (SD) 100 (9) 91 (23) 92 (13) 121 (32) 122 (41) 119 (7) 125 (26) 120 (30) 93 (15) 118 (12) 43 (22) Subgroups ab b b ab ab ab a ab b ab c

Table 9 Cytotoxicity (mean value (SD)) of archwires at 14 days. The subgroups with the same letter are not statistically different Archwires at 14 days Sample number Control 27 28 29 30 31 32 33 34 35 36 37 Mean (SD) 100 (9) 114 (25) 118 (20) 103 (26) 109 (15) 99 (18) 118 (21) 128 (27) 127 (19) 117 (14) 100 (14) 89 (8) Subgroups abc abc bc abc abc abc bc c c bc abc a

molar bands (P , 0.001) and the NewmanKeuls test showed three different subgroups (Table 7). Only sample 10 presented a severe cytotoxicity (43% of cell viability). There was a statistically signicant difference among the brackets (P , 0.001) (Table 8) and among the archwires (P , 0.001) (Table 9). The NewmanKeuls test respectively showed four different subgroups for the brackets and three subgroups for the archwires, but none of the samples was statistically different from control. 3.3. Comparison of metallic vs. non-metallic brackets (samples 1126) At 3 days, only one material (sample 14), out of the 16 brackets, did not belong to subgroup b and presented a lower cell viability. At 14 days, only sample 20 did not belong to subgroup c and also presented a lower cell viability. Thus, it can be concluded that stainless-steel (samples 1118), goldplated steel (sample 19), and pure titanium brackets (sample 20) were similar in terms of cell viability to ceramic (sample

21) and polycarbonate based materials (samples 2226) at 3 days (Table 5) and 14 days (Table 7). 3.4. Comparison of new molar bands and brackets to those worn in the oral cavity (samples 118) The ANOVA did not show any difference between the new materials and their worn counterparts, apart from one case at 14 days: sample 10 (43% cell viability), the worn molar band, was statistically more cytotoxic than sample 9 (118% cell viability), the counterpart new molar band. 3.5. Comparison of 3 vs. 14 days results (samples 137) Thirty one materials did not show any difference between the 3 and 14 day results. Six materials showed a statistically different cytotoxicity after the 3 and 14 day release period.

O. Mockers et al. / Dental Materials 18 (2002) 311317

315

Samples 5, 14 and 21 showed a higher cell viability after 14 days, but on the contrary, samples 8, 9 and 20 showed a lower cell viability after 14 days. 4. Discussion The rst objective of this study was to evaluate and compare the cytotoxicity of metallic and non-metallic brackets. The metallic brackets (samples 1120) were, in term of cytotoxicity, similar to the non-metallic brackets represented by ceramics (sample 21) and polycarbonates (samples 2226). The lack of difference among samples 2226 showed that the presence, inside the polycarbonate brackets, of a stainless steel slot (samples 22, 24), particles of ceramics (samples 23, 24) and glass bers (sample 25) did not modify the outcome of the test. To the authors' knowledge, no other experiment testing the cytotoxicity of polycarbonate and ceramic brackets has been published to this day. In this study, the various alloys and the ceramic and polycarbonate materials recurrently presented percentages of cell viability superior to 100%. These high percentages, not different from control, might have been explained by some retardation in growth of the control if the control medium had not been changed. In fact, all of the media were simultaneously changed, thus cell proliferation might be related to release of substances stimulating cell growth. Modications of cell multiplication, cell adhesion and cell migration have already been reported during in vitro cytotoxicity tests for non-cytotoxic alloys or metals [17]. In our study, the 16 brackets did not show any cytotoxicity. The stainless steel brackets (samples 1118) did not show lower values than those in gold-plated steel (sample 19) or in pure titanium (sample 20). Gold and titanium metals are highly biocompatible compared to others metals or alloys [18], but ionic releases, which are very low according to the norm ISO 10993, did not allow us to discriminate among the different alloys despite a 14 day release study. In relation to the metallic archwires (samples 2737), stainless steel, nickeltitanium (NiTi) and titaniummolybdenum archwires were not cytotoxic. This conforms to the results of two previous studies concerning the biocompatibility of NiTi. In the rst study, L929 cells were placed in contact with NiTi for 24 h [19]. The second experiment used human broblasts and osteoclasts in incubation with NiTi for 10 days [20]. These two studies concluded that NiTi had a good biocompatibility and a behavior very similar to the stainless steel, even if initially NiTi released 10 times more nickel than stainless steel. In these two studies, NiTi was tested directly in contact with the cells because these NiTi wires corresponded to a material used in orthopedic or craniofacial surgery. In the present study, because the orthodontic materials are placed distant from the gingival cells or cells of the oral mucosa, a culture medium was used as an extraction vehicle. The medium may act as saliva which plays an active role through the corrosion phenomena that

occur on the surface of the alloys and then a passive role by transportation of the release products to the cellular surfaces. Another experiment already observed the cytotoxic effects of orthodontic archwires [7]. It was, as with this study, an extraction test on cultures of L929 cells followed by a MTT assay, but with only one 3 day release period. The results showed that NiTi or titaniummolybdenum alloy, like stainless steel, did not inuence cellular proliferation. The strongest inhibition was shown by nickelchromiumcobalt alloy. This study concluded that those different alloys had various degrees of cytotoxicity, but all these orthodontic wires could be considered non cytotoxic since, within the norm ISO 10993, the cellular growth inhibition percentages were very low. The second purpose of our work was to compare new and clinically used molar bands. In our study only sample 10, out of the 10 molar bands, turned out to be cytotoxic. It was a soldered and clinically used band. A previous experiment, which evaluated the cytotoxicity of orthodontic bands, showed less constant results: only one out of the ve tested bands was not considered cytotoxic [6]. However, this previous experiment used an agar overlay test, which is less clinically relevant, and only provided semi-quantitative results. This cytotoxicity can be explained by the presence of soldering and the clinical use of this sample. Actually, the same band, soldered but new (sample 9) and the same band without solder (samples 7 and 8) did not show any cytotoxicity. This silver solder was composed of 59% Ag, 16% Cu and 25% Zn. The copper and zinc ions are highly cytotoxic and the copper is easily corroded [6,21]. That is why many experiments concerning alloys on cellular cultures use copper as a positive control [7]. The corrosion tests generally showed that copper and zinc ions dominate the extract solutions, even in small concentrations in the alloy, and also that the release of dental alloys was not proportional to the atomic percentage of the metals consisting of it [8]. Alloys containing more than 25% of copper should not be used as dental alloys. Moreover, the multi-composition of alloys (which can be up to 11 different metals) reduces the corrosion resistance by increasing the galvanic couples [18]. The copper and zinc ions released during the release phase could be then the cause of this cytotoxicity. However, the same sample tested over 3 days did not show any cytotoxicity. As 11 days separate the two protocols, ionic release, in more considerable quantities and which may be of a different nature, is likely to be the cause of this cytotoxicity. The corrosion phenomena occurring in the oral cavity can lead to dissolution of solders, separation of orthodontic components and damage to the material's surface [10,22]. The results of an experiment studying the cytotoxicity of orthodontic magnets showed a difference in cytotoxicity between the new and recycled magnets [22]. The cleaning and the polishing of alloys reduced the ionic release and more selectively the copper and zinc ions. A new polished alloy released 25 times less than the same oxidized alloy by the corrosion phenomena [23].

316

O. Mockers et al. / Dental Materials 18 (2002) 311317

In vitro experiments studying the cytotoxicity of orthodontic bonding resins on human oral broblasts concluded a lack of cytotoxicity of these materials despite the presence of residual monomers, certain polymers and chemical sterilizing agents [16]. On the contrary, resinous monomers have already been shown to be cytotoxic [12]. Thus, because of the lack of agreement among the studies, the worn samples 2, 4, 6, 8 and 10 were red because of the presence of bonding resin on the molar bands. Firing was necessary to crack the bonding material which was immediately removed by brushing. A pilot study did not show any inuence on their cytotoxicity of ring and cleaning of the molar bands. The third objective of this study consisted of comparing the results of cytotoxicity after 3 and 14 days release. The conditions of ionic release, dened by the norm ISO 10993, are not to be less than 24 h at 378C. However, this norm does not dene any precise time or limits of release. Clearly, the time function has a direct inuence on the release process [8,24]. It has been shown that a long exposure to non cytotoxic concentrations of resinous monomers modies the proliferation as well as the activities of human monocytes [25]. Most of the experiments on culture media studying the release of dental alloys have a release time of 3 days [7,8,23]. Sometimes the release process time can increase to ve at 7 days [8,16]. Nonetheless, these time periods cannot represent the behavior of orthodontic materials that are placed in the oral cavity for 12 years [24]. In our study, the choice of two release periods separated by 11 days was justied by the lack of consensus on this subject and by an expected stronger discrimination when interpreting the results. The results of our study showed various results according to the material. Only six materials out of 37 showed a statistically different cytotoxicity after 3 and 14 day release periods: three cases of increase and three cases of decrease of cell viability after 14 days. Because of the discrepancy among the results of this study, and the impossibility to reproduce clinically relevant release times, it can be recommended to use longer release times. The last purpose was to report the results on the subacute effects observed during orthodontic treatments. Only sample 10 presented a cytotoxicity at 14 days, which was really different from control and might be clinically relevant. A long 14 day release period probably allowed sufcient ion release to explain the lack of cytotoxicity at 3 days [8]. The solutions used in the corrosion tests or electrochemical tests might have allowed a stronger ionic release, but cannot be used in cytotoxicity tests because they would not allow cell survival and growth. Studying a correlation between ionic release in a specic solution and in vitro cytotoxicity might be of some interest. Soldering silver-based alloy, used in orthodontics, and containing copper and zinc, presented a severe cytotoxicity after corrosion in the oral cavity. In orthodontics, the practitioner should pay attention to the composition of soldering argent-based alloy and more particularly to the quantities

of copper and zinc. Orthodontic soldered appliances should be well polished to limit ionic release to its minimum. The clinical relevance of the in vitro cytotoxicity studies has already been debated. Some of the tested materials are very recent but some have been clinically used for many years without any problem. The lack of in vitro cytotoxicity of the tested materials corresponds to the good clinical history of these materials and shows a good correlation between the lack of in vitro cytotoxicity and the lack of clinical problems. The correlation is more difcult to establish in the case of a positive response, i.e. when the materials are cytotoxic in vitro. Only sample 10 was cytotoxic, very likely because of the presence of soldering. This cannot be related to any acute cytotoxic effect since no acute clinical effect has been reported with orthodontic materials. However, this may related to some subacute clinical symptoms sometimes related, such as glossitis, metallic tastes, bleeding and inamed or hypertrophied gingivae which can not be clinically distinguished from the gingivitis of a bacterial etiology [1015]. Under these conditions, the use of a long term release period may be more suitable since orthodontic materials remain in contact with living tissues for years. This enlarges the eld of application of in vitro cytotoxicity studies from acute to subacute effects but requires further investigations. 5. Conclusions 1. The metallic materials were similar in terms of cytotoxicity to the polycarbonate and ceramic-based materials. The orthodontic bands, brackets and archwires tested in vitro can be considered non cytotoxic. 2. The clinical use of the molar bands in a damp and oral environment modied their cytotoxicity in one case. 3. The outcome of the study may be modied by the release period 4. Other studies are needed to correlate subacute clinical symptoms and in vitro cytotoxicity.

Acknowledgements The authors would like to thank Dr Corinne Tardieu for her assistance in the laboratory. References
[1] Toms AP. The corrosion of orthodontic wire. Eur J Orthod 1988;10:8797. [2] Bass JK, Fine H, Cisneros G. Nickel hypersensitivity in the orthodontic patient. Am J Orthod 1993;103:2805. [3] Nagel ML. A Latex glove alert. Chem Health Safety 1997;4:148. [4] Staerkjaer L, Menne T. Nickel allergy and orthodontic treatment. Eur J Orthod 1990;12:2849.

O. Mockers et al. / Dental Materials 18 (2002) 311317 [5] Wiltshire WA, Ferreira MR, Ligthelm AJ. Allergies to dental materials. Quintessence Int 1996;27:51320. [6] Grimsdottir MR, Hensten-Pettersen A, Kullman A. Cytotoxic effect of orthodontic appliances. Eur J Orthod 1992;14:4753. [7] Rose E, Jonas I, Kappert HF. Investigation into the biological assessment of orthodontic wires. J Orofac Orthop 1998;59:25364. [8] Wataha JC, Malcom C, Hanks CT. Correlation between cytotoxicity and the element released by dental casting alloys. Int J Prosthod 1995;8:914. [9] Barrett RD, Bishara SE, Quinn JK. Biodegradation of orthodontic appliances, Part I. Biodegradation of nickel and chromium in vitro. Am J Orthod 1993;103:814. [10] Brune D. Metal release from dental materials. Biomat 1986;7:163 75. [11] Kerosuo E, Moe G, Kleven E. In vitro release of nickel and chromium from different types of simulated orthodontic appliances. Ang Orthod 1995;2:1116. [12] International Organisation for Standardization, ISO 10993-5, Biological evaluation of medical devicesPart 5: Tests for cytotoxicity: in ve 20, Switzerland, 1992. vitro methods, CH-1211 Gene [13] Grill V, Sandrucci MA, Basa M, Dorigo E, Bareggi R. The inuence of dental metal alloys on cell proliferation and bronectin arrangement in human broblast cultures. Archs Oral Biol 1997;42:6417. [14] Kois JC. The restorativeperiodontal interface: biological parameters. Periodontology 2000;11(1996):2938. [15] Hamula DW, Hamula W, Sernetz F. Pure titanium orthodontic brackets. J Clin Orthod 1996;30:1404.

317

rkman L, Ekstrand J. In vitro cytotoxicity of [16] Tang TH, Liu Y, Bjo orthodontic bonding resins on human oral broblasts. Am J Orthod Dentofac Orthop 1999;116:1328. [17] Wataha JC, Hanks CT, Sun Z. Effect on cell line on in vitro metal ion cytotoxicity. Dent Mater 1994;10:15661. [18] Craig RG, Hanks CT. Cytotoxicity of experimental casting alloys evaluated by cell culture tests. J Dent Res 1990;69:153942. [19] Dai K, Chu Y. Studies and applications of NiTi shape memory alloys in the medical eld of China. J Biomed Engin 1996;6:23340. [20] Ryhanen J, Niemi E, Serlo W, Sandwik P, Salo T. Biocompatibility of nickeltitanium shape memory metal and its corrosion behavior human cell cultures. J Biomed Mat Res 1997;15-3:4517. [21] Wataha JC, Hanks CT, Craig RG. In vitro effects of metal ions on cellular metabolism and the correlation between these effects and the uptake of the ions. J Biomed Mater Res 1994;28:42733. [22] Bondemark L, Kurol J, Wennberg A. Biocompatibility of new, clinically used and recycled orthodontic samarium-cobalt magnets. Am J Orthod 1994;105:56874. [23] Wataha JC, Craig RG, Hanks CT. The effects of cleaning on the kinetics of in vitro metal release from dental casting alloys. J Dent Res 1992;71:141722. [24] Wataha JC, Lockwood PE. Release of elements from dental casting alloys into cell-culture medium over 10 months. Dent Mater 1998;14:15863. [25] Bouillaguet S, Wataha JC, Virgilito M, Gonzalez L, Rakich DR, Meyer JM. Effect of sub-lethal concentrations of HEMA on THP-1 human monocyte-macrophages, in vitro. Dent Mater 2000;16:2137.