REVIEW ARTICLE

Corrosion of orthodontic appliances—should

we care?
Kate House,a Friedrich Sernetz,b David Dymock,c Jonathan R. Sandy,d and Anthony J. Irelande
Bristol, United Kingdom, and Ispringen, Germany

Contemporary orthodontics relies on various bonded attachments, archwires, and other devices to achieve
tooth movement. These components are composed of varying materials with their own distinctive physical
and mechanical properties. The demands made on them are complex because they are placed under many
stresses in the oral environment. These include immersion in saliva and ingested fluids, temperature
fluctuations, and masticatory and appliance loading. The combination of these materials in close proximity
and in hostile conditions can result in corrosion. Our purpose in this article was to consider the literature to
date with regard to potential mechanical, clinical, and health implications of orthodontic corrosion. (Am J
Orthod Dentofacial Orthop 2008;133:584-92)

T
he corrosion of orthodontic appliances in the metal can form a protective surface layer (passivation),
oral environment has concerned clinicians for or until the cathodic reactant is consumed (eg, exhaus-
some time; this concern is focused around 2 tion of dissolved oxygen in solution). The level of
principal issues: whether corrosion products, if pro- corrosion of any metal depends on the chemistry of the
duced, are absorbed into the body and cause either solvent in which it is immersed.
localized or systemic effects; and what the effects of The stainless steel, cobalt-chromium, and titanium
corrosion are on the physical properties and the clinical alloys used in orthodontic appliances rely on the
performance of orthodontic appliances. Our purpose in formation of a passive surface oxide film to resist
this review article was to consider the evidence on these corrosion. This protective layer is not infallible; it is
issues. susceptible to both mechanical and chemical disrup-
Corrosion occurs from either loss of metal ions tion. Even without disruption, oxide films often slowly
directly into solution or progressive dissolution of a dissolve (passivation) only to reform (repassivation) as
surface film, usually an oxide or a sulphide. Whereas the metal surface is exposed to oxygen from the air or
some metals are noble and virtually inert, eg, gold and the surrounding medium. Acidic conditions and chlo-
platinum, this is not the case for the metals commonly ride ions can accelerate the passivation process. There-
used in orthodontics. fore, a diet rich in sodium chloride and acidic carbon-
Essentially, corrosion occurs from 2 simultaneous ated drinks provides a regular supply of corrosive
reactions: oxidation and reduction (redox). Using iron agents. Another contributor to acidic oral conditions is
in a weak acid as an example, the oxidation (anodic) fluoride-containing products, such as toothpaste and
reaction results in dissolution of the iron as ferrous ions mouthwash. Many laboratory studies have demon-
are produced (Fe ¡ Fe2⫹ ⫹ 2e⫺). Reduction occurs at strated that, in a fluoridated, acidic environment, the
the cathode, with hydrogen ions reduced to hydrogen corrosion susceptibility of certain metals, especially
gas (2H⫹ ⫹ 2e⫺ ¡ H2). This corrosion process titanium, is increased.1,2 In these circumstances, the
continues until the metal is totally consumed, unless the highly protective titanium-oxide film is breached, per-
a
Senior specialist registrar, Orthodontics, Bristol Dental Hospital, Bristol,
mitting corrosive attack of the underlying alloy. Schiff
United Kingdom. et al1 compared the corrosion resistance of 3 types of
b
Technical director, Dentaurum, Ispringen, Germany. orthodontic brackets (stainless steel, cobalt-chromium,
c
Senior lecturer, Oral Microbiology, Bristol Dental Hospital, Bristol, United
Kingdom.
and titanium) when placed in a reference solution of
d
Professor, Orthodontics, Bristol Dental Hospital, Bristol, United Kingdom. artificial saliva and 3 commercially available fluoride
e
Consultant orthodontist and senior lecturer, Bristol Dental Hospital, Bristol, mouthwashes (each with similar pH values of about pH
United Kingdom.
Reprint requests to: Anthony J. Ireland, Bristol Dental Hospital, Bristol, BS1 4.3). The corrosion potential and the corrosion current
2LY, United Kingdom; e-mail, Tony.ireland@tiscali.co.uk. density were measured over a 24-hour period, and the
Submitted, November 2006; revised and accepted, March 2007. polarization resistance values were calculated. The
0889-5406/$34.00
Copyright © 2008 by the American Association of Orthodontists. material with optimal electrochemical properties in
doi:10.1016/j.ajodo.2007.03.021 artificial saliva was titanium, followed by cobalt-chro-
584
American Journal of Orthodontics and Dentofacial Orthopedics House et al 585
Volume 133, Number 4

mium and stainless steel. All 3 mouthwashes had little scopic level, they can exhibit many pits and crevices.
effect on the cobalt-chromium brackets, but the stan- These features are thought to increase the susceptibility
nous fluoride in 1 mouthwash caused considerable to corrosion because of their ability to harbor plaque-
corrosion of the stainless steel and titanium brackets. forming microorganisms.6,7 These microorganisms
This was due to destruction of the protective oxide cause localized reduction in pH and depletion of
layer and was confirmed by scanning electron micros- oxygen, which in turn affect the passivation process.8
copy (SEM) and analysis of the released ions. Hunt et al,7 in their in-vitro study, demonstrated that
More relevant, perhaps, were the results of an polishing nickel-titanium (Ni-Ti) wires to a uniform
in-vivo study of the sensitivity of titanium brackets to finish reduced the corrosion rate. However, more recent
the corrosive influence of fluoride-containing tooth- evidence from Huang9 suggests that the link between
paste and tea.3 In this study, 18 patients undergoing the surface roughness of as-received archwires and the
fixed appliance therapy were bonded with titanium increased corrosion potential is not straightforward.
brackets on the left side of the mouth and stainless steel This in-vitro study investigated the variation in corro-
brackets on the right side. The authors noted that all sion potential of a number of commercially available
patients were right handed, and, therefore, a cross- Ni-Ti wires using a linear polarization test in acidic
mouth study might have been a more appropriate artificial saliva (pH 6.25). In addition, SEM and atomic
design. Fifteen patients were asked to brush with a force microscopy were used to assess the surface
fluoride gel (pH 3.2), and the other 3 used a fluoride- morphology and the roughness of the wires. Chemical
free paste (pH 9.1-9.7), for 3 minutes twice a day. The analysis of the passive film was performed by electron
patients also kept a dietary record, noting especially spectroscopy. The results showed that the passive films
foods containing fluoride— eg, tea. After 5.5 to 7 on all wires were essentially the same. However,
months, 2 titanium brackets and 1 stainless steel although the surface roughness of the wires differed
bracket were removed and examined for pitting and significantly, it did not correspond to the corrosion
roughness under SEM. Microscopic evaluation showed resistance. It was suggested that surface residual stress
no significant difference in the pits and crevices on the produced during the manufacturing process might be
surfaces of the brackets, whether or not fluoride paste more important than surface roughness in the suscep-
was used. The difference in the results between this tibility of the wires to corrosion.
in-vivo study and other laboratory studies might be the Crevice corrosion can also occur in removable
exposure times. Although the titanium-oxide protective appliances when wires or components of expansion
layer undergoes degradation at pH of 3 and below, screws enter the acrylic. A brown discoloration can
saliva, water, and food in the oral environment can appear beneath the acrylic surface in contact with the
dilute the fluoride ion concentration, keeping the pH metal. This is thought to be due to bacteria and a
above this critical level at which corrosion takes place. surface biofilm between the wire and the acrylic,
Therefore, clinically, the role of fluoride in the corro- leading to crevice corrosion of the metal.10-12
sion of orthodontic appliances might not be as impor- Galvanic corrosion occurs when 2 metals are joined
tant as suggested by the in-vitro studies. Even if it does together and placed in a conductive solution or an
occur to some extent, as shown in a laboratory study by electrolyte. The more electronegative of the metals
Strietzel,4 in which the corrosion of titanium by fluo- becomes the anode, and the more electropositive or the
ride-containing gels occurred at pH of less than 4, the noble metal becomes the cathode. Thus, the more
corresponding corrosion product, titanium tetrafluoride, electropositive metal corrodes preferentially. Essen-
is known to be an ideal medium for the remineralization tially, galvanic corrosion cells are created because of
of enamel.5 differences in electrochemical potential between the 2
types of metal or the same metal at different sites.
Types of corrosion—the chemical These galvanic cells can also be created under other
and physical processes circumstances, such as differential pH, differences in
Uniform attack is the most common form of corro- surface finish (roughness), and work hardening due to
sion; it affects all metals, although at differing rates. repeated bending.
The metal undergoes a redox reaction with the sur- In orthodontics, galvanic corrosion might occur
rounding environment, and it can be undetected until where 2 dissimilar metals are joined in the construction
much of the metal is affected. of a bracket or a posted archwire. In the case of
Pitting and crevice corrosion can form on the removable appliances, the 2 metals can also contribute
surfaces of as-received orthodontic wires and brackets, to galvanic corrosion, but the situation is exacerbated
because they are not perfectly smooth. At a micro- by a soldered joint. This is because the soldered joint is
586 House et al American Journal of Orthodontics and Dentofacial Orthopedics
April 2008

mechanically active, making it even more susceptible and metabolize metal from alloys, leading to corrosion;
to corrosion.13 With soldered wires, the greatest prob- and (2) the normal metabolic by-products of other
lem is the release of iron, zinc, copper, and, particu- microbial species can alter environmental conditions,
larly, cadmium ions; the last 3 are released from the making them more conducive to corrosion— eg, by
cadmium-containing silver solder.14 In another in-vitro increasing the local acidity levels. The corrosive effects
study with fibroblasts to assess potential cytotoxicity, of microbials have been demonstrated in restorative
various orthodontic components were tested, both new dentistry with dental alloys,19 particularly endodontic
and used. Only the used stainless steel molar band with silver points.20 The corrosion products themselves
its soldered buccal tube demonstrated potential cyto- might increase the resistance of some bacteria to
toxicity.15 This effect was possibly due to the silver and antibiotics. Certainly, it is known that the characteris-
copper brazing alloys used, as was noted previously by tics of some resistance systems in these organisms are
Grimsdottir et al.16 shared. An increase in metal resistance in 1 organism
Stainless steel is particularly susceptible to inter- can lead to increased antibiotic resistance, which might
granular corrosion during brazing and welding; this then be transferred to another bacterial species. There-
can occur at temperatures as low as 350°C. Heating fore, there is considerable potential for increased expo-
leads to the reaction of chromium with the carbon in the sure to metals and their corrosion products to result in
steel to form chromium carbide. Subsequent precipita- the spread of resistant genes between bacteria, includ-
tion of this carbide at grain boundaries and slip planes ing into pathogens of medical and dental significance.
has 2 effects: (1) the alloy becomes more brittle due to
slip interference, and (2) the alloy is less resistant to Manufacturing implications
corrosion, because the chromium was used up in the Manufacturers are well aware of the susceptibility
reaction to form the carbide, making less available to of orthodontic alloys to the various forms of corrosion
form the passive oxide layer.17 and have taken steps to combat this potentially destruc-
Fretting corrosion occurs in areas of metal contact tive process, including the following.
subject to sustained loads. An orthodontic example is
the archwire/bracket-slot interface. During the applica- 1. Alloy substitution or addition. The addition of
tion of a load, the 2 metals undergo a process of cold certain metals to an alloy can reduce its suscepti-
welding from the pressure at the interface between bility to corrosion. This fact has been used in the
them. Continued application of force at such an inter- production of Ni-Ti and stainless steel orthodontic
face eventually causes the welded junction to shear, components.
disrupting the protective surface oxide layers and leav- The corrosion resistance of Ni-Ti orthodontic
ing the metals susceptible to corrosion. components is due to the large amount of titanium,
Whenever an archwire is ligated to orthodontic usually from 48% to 54%. Titanium can form
brackets, the reactivity of the metal alloy increases at several oxide configurations (TiO, TiO2, and
sites of stress due to loading; this is called stress Ti2O5); titanium dioxide is the most stable and
corrosion. An electrochemical potential can therefore commonly formed oxide.
be created along the wire, with some sites acting as In the case of stainless steel alloys, the addition
anodes and other as cathodes, thus facilitating corro- of chromium and nickel imparts corrosion resis-
sion. tance. The chromium contributes to the surface
Metals generally have an increased tendency to oxide layer, which spontaneously undergoes passi-
fracture under repeated cyclic stressing (fatigue). This vation and repassivation in air and the oral envi-
phenomenon is accelerated if the alloy is also exposed ronment. The nickel aids corrosion resistance by
to a corrosive medium; this is called corrosion fatigue. competing with the chromium to form salts, making
For example, corrosion fatigue might occur when more chromium available for passivation.21 Oxy-
orthodontic wires are left in the oral environment for gen is necessary to initiate and maintain the oxide
long periods under load. However, in a study investi- film, whereas acidic conditions enhance its break-
gating corrosion fatigue of Ni-Ti, titanium molybde- down. The addition of molybdenum to American
num, and stainless steel wires, none showed increased Iron and Steel Institute 316L-type stainless steel has
corrosion as a result of mechanical and electrochemical been shown to reduce the amount of pitting and
stressing.18 crevice corrosion.8 Although stainless steel has a
Microbiologically influenced corrosion is also pos- passive oxide coating from the chromium, this layer
sible. Microorganisms and their by-products can affect is not as effective as that produced by titanium
metal alloys in 1 of 2 ways: (1) certain species absorb oxide in Ni-Ti components. Steel therefore has
American Journal of Orthodontics and Dentofacial Orthopedics House et al 587
Volume 133, Number 4

inferior corrosion resistance when compared with ceptibility. However, electropolishing can induce
Ni-Ti alloys. galvanic corrosion cells between polished (eg, tie
The use of brazing alloys in the fabrication of wings) and unpolished areas (eg, bracket slots).
orthodontic brackets can also lead to corrosion Another method of reducing corrosion of metals
through galvanic action. This can be dramatically during manufacture is to add a corrosion inhibitor
reduced by laser welding, rather than brazing, the to a solution into which the material is placed,
body of the bracket to its base.22 resulting in the formation of a protective layer or
2. Coatings. Orthodontic archwires and brackets can coating. A similar effect might also occur in the
be coated with either titanium nitride or an epoxy oral environment, with certain salivary proteins,
resin. The former is used to improve hardness and amylase, and ␥-globulin forming a biofilm that acts
reduce friction; the latter improves esthetics. An as a corrosion inhibitor.24-27 A retrieval study by
in-vitro study to compare these 2 coatings on Ni-Ti Eliades et al28 found that Ni-Ti wires become
wires with uncoated Ni-Ti, titanium, and stainless coated by a proteinacious film that masks surface
steel wires indicates that corrosion occurs readily in topography. The composition and thickness of this
both stainless steel and uncoated Ni-Ti wires.23 film depended on each patient’s oral conditions and
However, for Ni-Ti wires, the breakdown potentials the intraoral exposure time. The organic compo-
vary depending on the manufacturer. Although the nents of the film were found to be amides, alcohols,
nitride coating did not affect corrosion, the epoxy and carbonates, with other constituents comprising
coating improved corrosion resistance. Kim and crystalline precipitates of sodium chloride, potas-
Johnson23 did not consider the clinical observation sium chloride, and calcium phosphate. The authors
that the epoxy coating tends to wear off during use, hypothesized that the mineralized regions of the
exposing the underlying wire, and this would obvi- film might act as a protective layer, especially
ously affect corrosion behavior. under acidic conditions when the corrosion rates of
Some commercial brackets are available with a Ni-Ti and stainless steel wires have otherwise been
gold finish produced by either electrodeposition of shown to be increased. This might help to explain
gold or plasma-arc deposition of titanium nitride on why the in-vivo behavior of metallic appliances is
the metal surface. Although the effects of these often superior to the results predicted by laboratory
coatings are not fully understood, it is supposed that corrosion studies.
they improve corrosion and wear resistance of the
bracket.24 From the study of Kim and Johnson23 on Mechanical implications of corrosion
wires, it is unlikely that titanium nitride coating will Ni-Ti wires with superelastic and shape-memory
give corrosion resistance to brackets. Hartung et properties have revolutionized modern orthodontics.
al,25 in an in-vivo experiment, found some evidence They have enabled treatment to be completed with
of corrosion on titanium-nitride-coated brackets. fewer archwire changes and have permitted patients to
3. Modification of the production process. Variations go for longer intervals between visits. The shape-
in manufacturing techniques and postmanufactur- memory effect and the superelastic properties of Ni-Ti
ing finishing and polishing operations can affect the wires are due to the well-documented austenitic-mar-
corrosion behavior of brackets. An in-vitro study tensitic phase transformations that occur with alter-
showed that brackets with essentially the same ations in stress or temperature.29 As previously men-
composition can have significantly different corro- tioned, acidic fluoridated conditions created by the
sion properties.26 The microstructure of an alloy regular use of fluoride prophylactic agents might cause
can affect corrosion, and the microstructure itself is increased corrosion of orthodontic wires; this in turn
affected by alloying, heat treatment, and cold work- can affect their mechanical properties. This hypothesis
ing. Cold working, for example, occurs during was tested in an in-vitro study by Walker et al.30
milling and cutting of the bracket slot; this in turn Sections of Ni-Ti and copper-Ni-Ti orthodontic wires
might induce galvanic couples between the worked with rectangular cross-sections were placed in 2 types
and the adjacent unworked areas.24 In addition, of high fluoride-ion concentration gels: Phos-flur gel
some manufacturers use different grades of stain- 1.1% sodium fluoride acidulated phosphate, 0.5% w/v
less steel for the mesh and the bracket base, and so fluoride, pH 5.1 (Colgate Oral Pharmaceuticals, New
introduce galvanic couples. Postmanufacturing sur- York, NY), and Prevident 5000 1.1% sodium fluoride
face finishing can also affect corrosion behavior. neutral agent, 0.5% w/v fluoride, pH 7 (Colgate Oral
Many manufacturers electropolish their brackets to Pharmaceuticals). Wire sections were placed in plastic
improve the appearance and reduce corrosion sus- vials with the fluoride gel at 37°C for 1.5 hours. The
588 House et al American Journal of Orthodontics and Dentofacial Orthopedics
April 2008

authors suggested that this would be equivalent to 3 mouthwash, and dietary fluoride could be assessed with
months of 1-minute daily topical application, although respect to the tensile strength of the Ni-Ti wires.
they did not disclose how this was determined. The Although corrosion has been implicated as a cause
principal outcome measures were elastic modulus and of wire fracture, other authors suggest that it might
yield strength. The results suggested that both fluoride have more to do with the surface finish of the wire
gels significantly decreased the unloading modulus and produced by the manufacturer. Schwaninger et al6
the yield strength when compared with the distilled tested the physical properties (bending and flexural)
water control. No significant effects were found for the and the surface topography of some Ni-Ti archwires
copper-Ni-Ti wires. This might suggest that, clinically, that had been stored in 1% sodium chloride solution for
topical fluoride agents can reduce the functional un- 11 months. Tests were performed at 2-month intervals,
loading mechanical properties of Ni-Ti wire and con- starting at month 1. They found no significant differ-
tribute to prolonged orthodontic treatment. ence in the physical properties, but, when the fracture
It is not unusual for superelastic Ni-Ti archwires to sites were studied with SEM, fracture initiation and
deform or fracture during clinical use. Few studies have propagation sites occurred at the surface pits. Also, the
investigated whether there is a relationship between wires that tended to fracture early had more surface
these events and the oral environment. Yokoyama et defects. The authors concluded that it was not the
al31,32 performed several laboratory studies on the role effects of corrosion that cause early fracture but, rather,
of hydrogen absorption in the fracture of Ni-Ti arch- the surface defects generated during manufacturing.
wires in saline and fluoridated environments. These Therefore, manufacturers should be encouraged to
environments might dissolve the protective oxide film improve the surface quality of their wires.
of Ni-Ti and allow adsorption of hydrogen, which has It would seem that the causes of archwire fracture
are multifactorial, with corrosion, surface finish, and
a high affinity with titanium. The subsequent formation
work hardening during treatment all as contributors.
of brittle hydrides, primarily titanium hydride, is
thought to increase the likelihood of wire fracture. In a Effects of recycling on mechanical properties
recent experiment, sections of superelastic Ni-Ti wire
Recycling orthodontic wires and brackets was once
were immersed in 0.2% acidulated phosphate fluoride
common, but this practice is no longer recommended in
(pH 5.0) for 24 hours. The martensitic transformation
some countries such as the United Kingdom (BOS,
temperatures of the wires were determined before the
Advice Sheet 16). A number of studies have attempted
study and after 24 hours. In addition, hardness and
to investigate the influence of repeated exposure and
tensile tests were conducted. The tensile strength of the
sterilization on mechanical properties.
wires was found to increase slightly with immersion Lee and Chang33 tested tensile strength, friction,
times of up to 3 hours and then decreased rapidly with bending fatigue, microscopic surface appearance, and
longer immersion (1250-600 MPa), after which they surface roughness of Ni-Ti wires that had been exposed
remained constant. Fracture of the alloy tended to occur to artificial saliva for 4 weeks and a similar group that
before the martensitic transformations, when the wires was then sterilized at 121°C at 15 to 20 psi for 20
had been immersed for more than 6 hours, with minutes. Although there was no difference in tensile
fractographs suggesting that fracture began at the outer properties or bending fatigue, there were increases in
wire surface. SEM confirmed that surface corrosion had surface roughness and coefficient of friction in the
occurred when the samples immersed for 24 hours were recycled group. Mayhew and Kusy34 failed to detect a
compared with the as-received wires. Greater hydrogen significant difference in the mechanical properties or
adsorption had taken place with increased immersion. It the surface topography of Ni-Ti wires subjected to
was concluded that, when the amount of hydrogen various forms of sterilization when compared with
adsorbed exceeds 200 mass ppm, the tensile strength of untreated wires. Perhaps recycling brackets with brazed
the immersed alloy is reduced to the critical stress level bases might be more problematic, since recycling can
of the martensite transformations. Hydrogen embrittle- comprise heat, chemical, and mechanical processes,
ment as a corrosion process appears to be 1 reason for which could lead to accelerated crevice corrosion of the
fracture of titanium and its alloys in a fluoridated brazed joint.26
environment. However, the results of in-vitro studies
should be treated with caution. It would be helpful to Health implications
repeat this experiment with wires that had been sub- Should we be concerned about the release and
jected to the oral environment, so that the effects of potential absorption of nickel and other corrosion
regular low-dose fluoride products such as toothpaste, products? Orthodontic appliances differ from other
American Journal of Orthodontics and Dentofacial Orthopedics House et al 589
Volume 133, Number 4

medical uses of nickel alloys because they are not when absorbed, it is not clear whether this readily
implanted; rather, they are placed in the oral environ- occurs with orthodontic appliances. It is worth consid-
ment. Although these appliances might not seem to be ering how much, if any, nickel and chromium are
in such an intimate relationship with body tissues, the released during orthodontic treatment. Nickel and chro-
oral environment is considered hostile and potentially mium are consumed in our diets, with average values of
corrosive. There has been much interest about whether 200 to 300 ␮g per day for nickel and 280 ␮g per day for
detectable levels of nickel are released during orth- chromium. Significant exposure to nickel and chro-
odontic treatment, and, particularly, whether released mium can occur from the atmosphere, drinking water,
nickel is detectable in saliva or serum, and whether it clothing fasteners, and jewelry.42 When considering the
has any health effects. role of dental rather than just orthodontic alloys, nickel
It was suggested that nickel can have carcinogenic, release has been reported to be about 4.2 ␮g per day.43
mutagenic, cytotoxic, and allergenic effects.8 Thus, If an assessment of heavy metal loading, and therefore
there has been a move to publicize possible adverse recommended safe levels, is based on our estimated
reactions to nickel and to emphasize patient awareness intake of dietary elements, then the additional loading
of its potential dangers. A well-known example of this from orthodontic corrosion products is likely to be
is Proposition 65, the Safe Drinking Water and Toxic small. However, due to our limited knowledge of the
Enforcement Act, in California. It lists nickel, nickel physical and chemical states of the corrosion products
compounds, and chromium as chemicals known to released from dental materials in the oral environ-
cause cancer, birth defects, and other reproductive ment— eg, valency state, particulate matter form, and
harm.35 hapten binding—recommendations can be used only as
However, Tomakidi et al36 published an extensive rough guidelines at best.44
study on the cytotoxicity and genotoxicity of the There are 3 ways of investigating metal ion release:
corrosion eluates from orthodontic materials, using in vitro, retrieval (ex-vivo investigation of in-vivo aged
monolayer cultures of immortalized human gingival samples), and in vivo. As with many of the corrosion
keratinocytes. The test materials included nickel-free studies previously cited, most are in vitro, and this
wires and brackets, nickel-containing stainless steel makes the results and conclusions potentially irrelevant
bands and brackets, and expansion screws made of to the clinical situation. Nevertheless, these studies
titanium. Each was placed in artificial saliva, according must be considered, because there are few alternatives.
to International Standards Organization 10271, for up Some studies have suggested that no nickel is
to 14 days, and cell cultures were then exposed to released during intraoral placement of orthodontic ap-
eluates with the highest ion concentrations. None of pliances. Eliades et al45 investigated nickel released
the eluates had acute cytotoxicity, and an assessment from stainless steel and Ni-Ti wires retrieved after
of genotoxicity also showed no apparent DNA damage. clinical use and compared them with as-received wires.
Perhaps the most common adverse effect that orth- The test wires were all of 0.016 ⫻ 0.022-in cross-
odontists encounter, or are consulted about, is nickel sections and had been ligated to stainless steel brackets.
hypersensitivity. There have been many articles on the In total, 20 stainless steel and 25 Ni-Ti wires were
subject, often isolated case reports. They indicate that, retrieved; the intraoral service period was 1.5 to 12
in some instances, nickel-containing orthodontic appli- months. SEM and energy-dispersive x-ray microanaly-
ances have caused gingival hyperplasia, labial desqua- sis were used to assess the elemental composition of the
mation, angular cheilitis, swelling, and burning sensa- wires. The authors found no significant difference
tions affecting the oral mucosa.37-39 This inflammatory between the retrieved and the as-received wires with
response is considered an example of type IV hyper- respect to nickel-content ratios.
sensitivity. The incidence of adverse reactions in orth- Other in-vitro and some in-vivo studies confirmed
odontic patients was estimated at 1:100, with 85% of the release of nickel and chromium ions into saliva after
these incidents attributable to contact dermatitis, mostly fixed orthodontic appliance placement, although the
involving the extraoral components of headgear.40,41 levels were low, far lower than normal dietary in-
However, diagnosing nickel hypersensitivity affect- take.40,42,46-48 Grimsdottir et al16 analyzed the quanti-
ing the oral mucosa is more difficult than on the skin. In ties of nickel and chromium released into physiologic
the mouth, for example, nickel lesions can be easily saline solution over 14 days in a laboratory study.
confused with those caused by mechanical injury or Interestingly, although they also found that negligible
poor oral hygiene. levels of nickel and chromium were released from the
Although there is evidence that nickel and its archwires, high levels were released from a headgear
compounds can, at certain concentrations, cause harm facebow. Facebows contain silver solder, which is
590 House et al American Journal of Orthodontics and Dentofacial Orthopedics
April 2008

thought to be capable of inducing the formation of high nickel content, had some local hypersensitivity
galvanic couples, leading to the release of nickel and reactions. A more recent retrospective study investi-
other metal ions.49 However, the skin’s reaction to gated the roles of age, previous allergic history, and
headgear is totally different from the reaction in the oral time of exposure to fixed appliances in the etiology of
cavity because of the absence of Langerhans cells. It nickel hypersensitivity.53 Some of the 48 patients (ages,
could be concluded from the high release of nickel from 10-44 years) in the study exhibited clinical manifesta-
a soldered facebow in this study that the laser-welded tions of nickel hypersensitivity. The results demon-
facebow might be a better alternative. strated that the clinical signs of hypersensitivity were
Even if nickel and chromium are released during independent of the length of time the subject was
orthodontic treatment, is it taken up by the saliva and, exposed to the orthodontic appliance. Also, the patients
more importantly, by the bloodstream? This was the with these signs were significantly younger than those
aim of a cross-sectional study that examined saliva and without signs. The most telling finding was that previ-
serum samples from 100 patients.48 The subjects (ages, ous allergic history was the most important factor in
12-33 years) all had fixed appliances. Saliva and blood characterizing nickel hypersensitivity and not the pres-
samples were collected before the appliances were ence of orthodontic appliances. Another study has
placed and then 1 week, 1 month, 1 year, and 2 years indicated that nickel-containing orthodontic appliances
later. The results showed, in the saliva samples, a have little or no effect on the oral and gingival health of
detectable increase in nickel and chromium levels in the nickel-sensitive patients. This might be because higher
first month, when compared with the initial and 1-week concentrations of contact allergens are required to elicit
samples. The levels then decreased by a statistically a reaction on oral mucosa than on skin.54
significant amount at 2 years. To place these results into The relationship between sensitization by a poten-
context, the amounts in saliva were 0.53 to 1.53 ppb of tial allergen at an early age and the reaction after a later
chromium and 4.12 to 11.53 ppb of nickel. These are new exposure is not simple. Early contact with sus-
within the normal ranges and far below average daily pected allergens can actually result in a diminished
dietary intake. The serum samples showed a statisti- chance of allergic reaction later in life. Perhaps orth-
cally significant increase in the amount of chromium at odontic treatment with nickel-containing components
2 years when compared with the other time periods. No before sensitization to nickel (eg, ear piercing) might
differences were found in the amount of serum nickel lower the incidence of nickel hypersensitivity.41,55
throughout the study. Interestingly, there was also no
correlation between the saliva and the serum levels of CONCLUSIONS
nickel over the different periods. This suggests that Although corrosion of orthodontic devices occurs,
nickel can be detected in saliva but is not absorbed into it does not appear to result in significant destruction of
the bloodstream. A criticism of this study is that the the metallic components or have significant detrimental
samples were taken from various patients during the effects on mechanical properties. Exceptions to this
study period and not from the same cohort. Because of might be soldered joints on removable appliances and
other factors— eg, previous allergic history and differ- facebows, and the brazed joints of some stainless steel
ences in dietary intake of nickel and chromium—it brackets.
would seem that following a cohort of patients through- The literature suggests that metal ions are released
out their treatment would have given more valid results. during orthodontic treatment, but the level is far lower
At these low levels of nickel and chromium in than that ingested in a routine daily diet. Some patients
saliva, what is the evidence for cytotoxic and hyper- may well demonstrate nickel hypersensitivity when
sensitivity reactions? The cytotoxic effects of various exposed to nickel-containing alloys; previous nickel
metallic orthodontic devices, including molar bands, sensitivity and the patient’s age are the best indicators.
brackets, and archwires, have been assessed by using This relationship, however, is not entirely clear, and
mouse fibroblasts as previously described.16,50 The there are even indications that orthodontic treatment
brackets and archwires showed no cytotoxic effect.51 can improve the immune system’s tolerance to nickel in
The lack of cytotoxic effect of Nitinol on human sensitive people.
fibroblasts was also demonstrated by Ryhanen et al.52 The impact of corrosion on orthodontic treatment
It has been suggested that nickel-containing jewelry and the health of our patients is not well understood.
worn from a young age can induce sensitization of Based on the best current evidence, it does not appear
prospective orthodontic patients. A clinical study by to be a process that should cause concern. Future work
Greppi et al51 indicated that most nickel-sensitive in more clinically relevant situations will lead to a
people, receiving intraoral exposure from wires with a better understanding of the clinical effects of corrosion.
American Journal of Orthodontics and Dentofacial Orthopedics House et al 591
Volume 133, Number 4

This review of the corrosion of orthodontic prod- 18. Widu F, Drescher D, Junker R, Bourauel C. Corrosion and
ucts was originally requested by ISO TC106/SC1 biocompatibility of orthodontic wires. J Mater Sci Mater Med
1999;10:275-81.
working group 13 during the writing of the ISO 19. Palaghias G. Oral corrosion and corrosion inhibition processes.
standard, “Dentistry—wires for use in orthodontics.” Swed Dent J 1985;30(suppl):39-65.
We thank all committee members for their helpful 20. Margelos J, Eliades G, Palaghias G. Corrosion of endodontic
comments during the writing and for permission to use silver points in vivo. J Endod 1991;17:282-7.
the material gathered. 21. Toms AP. The corrosion of orthodontic wire. Eur J Orthod
1988;10:87-97.
22. Müller WD, Dorow S, Finke CH, Petzold D, Lange KP.
REFERENCES Elektrochemische charakterisierung unterschiedlicher verbind-
ungen zwischen edelstahlkrone und klammerdraht (electrochem-
1. Schiff N, Dalard F, Lissac M, Morgon L, Grosgogeat B.
ical characterization of different joinings between stainless steel
Corrosion resistance on three orthodontic brackets: a compara-
crowns and clasps). Deut Zahnaerztl Z 2004;59:535-9.
tive study of three fluoride mouthwashes. Eur J Orthod 2005;27:
23. Kim H, Johnson JW. Corrosion of stainless steel, nickel-tita-
541-9.
nium, coated nickel-titanium, and titanium orthodontic wires.
2. Toumelin-Chemla F, Rouelle F, Burdairon G. Corrosive proper-
Angle Orthod 1999;69:39-44.
ties of fluoride-containing odontologic gels against titanium. J
24. Von Fraunhofer JA. Corrosion inhibition by salivary proteins and
Dent 1996;24:109-15.
enzymes. J Dent Res 1992;71:526.
3. Harzer W, Schroter A, Gedrange T, Muschter F. Sensitivity of
25. Hartung M, Fischer J, Grabowski R, Behrend D. Long term
titanium brackets to the corrosive influence of fluoride-contain-
ing toothpaste and tea. Angle Orthod 2001;71:318-23. corrosion behaviour of TiN-coated brackets in vivo (abstract
4. Strietzel R. Einfluss von fluoridhaltigen zahnpasten auf zahno- 252). Poster presented at the 75th Congress of the European
berflächen (influence of fluoride-containing toothpastes on tooth Orthodontic Society; 23rd-26th June 1999; Strasbourg, Ger-
surfaces). Zahnä Welt 1994;103:82-4. many.
5. Büyükyilmaz T, Tangurson V, Ogaard B, Arends J, Ruben J, 26. Maijer R, Smith DC. Biodegradation of the orthodontic bracket
Rolla G. The effect of titanium tetrafluoride (TiF4) application system. Am J Orthod 1986;90:195-8.
around orthodontic brackets. Am J Orthod Dentofacial Orthop 27. Brown S, Merritt K. Fretting corrosion in saline and serum.
1994;105:293-6. J Biomed Mater Res 1981;15:479-88.
6. Schwaninger B, Sarkar NK, Foster BE. Effect of long-term 28. Eliades T, Eliades G, Athanasiou AE, Bradley TG. Surface
immersion corrosion on the flexural properties of nitinol. Am J characterisation of retrieved NiTi orthodontic archwires. Eur
Orthod 1982;82:45-9. J Orthod 2000;22:317-26.
7. Hunt NP, Cunningham SJ, Golden CG, Sheriff M. An investi- 29. Kusy RP. A review of contemporary archwires: their properties
gation into the effects on polishing on surface hardness and and characteristics. Angle Orthod 1997;67:197-207.
corrosion of orthodontic archwires. Angle Orthod 1999;69: 30. Walker MP, White RJ, Kula KS. Effect of fluoride prophylactic
433-40. agents on the mechanical properties of nickel-titanium-based orth-
8. Eliades T, Athanasiou AE. In vivo aging of orthodontic alloys: odontic wires. Am J Orthod Dentofacial Orthop 2005;127:662-9.
implications for corrosion potential, nickel release, and biocom- 31. Yokoyama K, Hamada K, Moriyama K, Asaoka K. Degradation
patibility. Angle Orthod 2002;72:222-37. and fracture of Ni-Ti superelastic wire in an oral cavity.
9. Huang H. Variation in corrosion resistance of nickel titanium Biomaterials 2001;22:2257-62.
wires from different manufacturers. Angle Orthod 2005;75: 32. Yokoyama K, Kaneko K, Moriyama K, Asaoka K, Sakai J,
661-5. Nagumo M. Hydrogen embrittlement of NiTi superelastic alloy
10. Sernetz F. Biocompatibility of metallic orthodontic appliances. in fluoride solution. J Biomed Mater Res 2003;65:182-7.
In: Sernetz F, editor. Materiali ortodontici e biocompatibilita 33. Lee SH, Chang Y. Effects of recycling on the mechanical
(materials and biocompatibility in Orthodontics). Milan, Italy: properties and the surface topography of nickel-titanium alloy
Società Italiana Di Ortodonzia. Syllabus 7,1997, p. 4-24. wires. Am J Orthod Dentofacial Orthop 2001;120:654-63.
11. Fitjer LC, Jonas IE, Kappert HF. Corrosion susceptibility of 34. Mayhew MJ, Kusy RP. Effects of sterilization on the mechanical
lingual wire extensions in removable appliances. An in-vitro properties and the surface topography of nickel-titanium arch
study. J Orofac Orthop 2002;63:212-26. wires. Am J Orthod Dentofacial Orthop 1988;93:232-6.
12. Kusy RP, Ambrose WW, LaVanier LA, Newman JG, Whitley 35. Turpin DL. California proposition may help patients in search of
JQ. Analyses of rampant corrosion in stainless-steel retainers of better oral health. Am J Orthod Dentofacial Orthop 2001;120:97-9.
orthodontic patients. J Biomed Mater Res 2002;62:106-18. 36. Tomakidi P, Koke U, Kern R, Erdinger L, Krüger H, Kohl A, et
13. Grimsdottir MR, Gjerdet NR, Hensten-Pettersen A. Composition al. Assessment of acute cyto- and genotoxicity of corrosion
and in vitro corrosion of orthodontic appliances. Am J Orthod eluates obtained from orthodontic materials using monolayer
1992;101:525-32. cultures of immortalized human gingival keratinocytes. J Orofac
14. Berge M, Gjerdet NR, Erichsen ES. Corrosion of silver soldered Orthop 2000;61:2-19.
orthodontic wires. Acta Odontol Scand 1982;40;75-9. 37. Dunlap CL, Vincent SK, Barker BF. Allergic reaction to orth-
15. Mockers O, Deroze D, Camps J. Cytotoxicity of orthodontic odontic wire: report of case. J Am Dent Assoc 1989;118:449-50.
bands, brackets and archwires in vitro. Dent Mater 2002;18: 38. Lindsten R, Kurol J. Orthodontic appliances in relation to nickel
311-7. hypersensitivity: a review. J Orofac Orthop 1997;58:100-8.
16. Grimsdottir MR, Hensten-Pettersen A, Kullman A. Cytotoxic 39. Starkjaer L, Menne T. Nickel allergy and orthodontic treatment.
effect of orthodontic appliances. Eur J Orthod 1992;14:47-53. Eur J Orthod 1990;12:284-9.
17. Matasa CG. Attachment corrosion and its testing. J Clin Orthod 40. Hensten-Petersen A. Casting alloys: side effects. Adv Dent Res
1995;29:16-23. 1992;6:38-43.
592 House et al American Journal of Orthodontics and Dentofacial Orthopedics
April 2008

41. Greig DGM. Contact dermatitis reaction to a metal buckle on a 49. Agaoglu G, Arun T, Izgu B, Yarat A. Nickel and chromium
cervical headgear. Br Dent J 1983;155:61-2. levels in the saliva and serum of patients with fixed orthodontic
42. Barrett RD, Bishara SE, Quinn JK. Biodegradation of orthodon- appliances. Angle Orthod 2001;71:375-9.
tic appliances. Part I. Biodegradation of nickel and chromium in 50. Grimsdottir MR, Hensten-Pettersen A. Cytotoxic and antibacte-
vitro. Am J Orthod Dentofacial Orthop 1993;103:8-14. rial effects of orthodontic appliances. Scand J Dent Res 1993;
43. Park HY, Shearer TR. In vitro release of nickel and chromium from 101:229-31.
simulated orthodontic appliances. Am J Orthod 1983;84:156-9. 51. Greppi L, Smith DC, Woodside DG. Nickel hypersensitivity
44. Brune D. Metal release from dental biomaterials. Biomaterials 1986;7: reactions in orthodontic patients. J Dent Res 1991;70:361.
163-75. 52. Ryhanen J, Niemi E, Serlo W, Niemela E, Sandvik P, Pernu H,
45. Eliades T, Zinelis S, Papadopoulos MA, Eliades G, Athanasiou et al. Biocompatility of nickel titanium shape memory metal and
AE. Nickel content of as-received and retrieved NiTi and its corrosion behaviour in human cell cultures. J Biomed Mater
stainless steel archwires: assessing the nickel release hypothesis. Res 1996;35:451-7.
Angle Orthod 2004;74:151-4. 53. Genelhu MCL, Marigo M, Alves-Oliveira LF, Malaquias L,
46. Huang T, Yen C, Kao C. Comparison of ion release from new Gomez RS. Characterization of nickel-induced allergic contact
and recycled orthodontic brackets. Am J Orthod Dentofacial stomatitis associated with fixed orthodontic appliances. Am J
Orthop 2001;120:68-75. Orthod Dentofacial Orthop 2005;128:378-81.
47. Kerosuo H, Moe G, Hensten-Pettersen A. Salivary nickel and 54. Bass JK. Nickel hypersensitivity in the orthodontic patient. Am J
chromium in subjects with different types of fixed orthodontic Orthod 1993;103:280-5.
appliances. Am J Orthod Dentofacial Orthop 1997;111:595-8. 55. Janson GR, Dainesi EA, Coonsolaro A, Woodside DG, de Freitas
48. Kerosuo H, Moe G, Kleven E. In vitro release of nickel and MR. Nickel hypersensitivity reaction before, during, and after
chromium from different types of simulated orthodontic appli- orthodontic therapy. Am J Orthod Dentofacial Orthop 1998;113:
ances. Angle Orthod 1995;65:111-6. 655-60.