Biological Trace Element Research

, Volume 137, Issue 2, pp 127–138

Trace Metal Release from Orthodontic Appliances by In Vivo Studies: A Systematic Literature Review

Authors

    • Department of Dentofacial Orthopaedics and OrthodonticsMedical University of Wrocław
  • Katarzyna Chojnacka
    • Institute of Inorganic Technology and Mineral FertilizersWrocław University of Technology
Article

DOI: 10.1007/s12011-009-8576-6

Cite this article as:
Mikulewicz, M. & Chojnacka, K. Biol Trace Elem Res (2010) 137: 127. doi:10.1007/s12011-009-8576-6

Abstract

The paper discusses various approaches used to investigate biocompatibility by the analysis of metals release by the materials of which orthodontic appliances are made. Analysis of various biomarkers of exposure: saliva, serum, mucosa cells, or urine is used in in vivo tests. In this work, the techniques, results, and conclusions of original papers were compared by the implementation of the concept of a systematic review. The aim of the present work was to report the state-of-the-art in the research on methods used to assess exposure to trace metals from orthodontic appliances. The PubMed search identified 35 studies, among which nine met the selection criteria. The general conclusion in the studies was that metal ions were released mostly in the initial stage of the treatment. However, the majority of studies included 1–2 months long period and did not reflect long-term changes nor the impact of the complete treatment, the duration of which is several years, on the whole organism and the overall accumulation of metals from orthodontic appliances. In studies which evaluated nickel concentrations in blood and urine, long-term metal release was detected and significant differences were found. It leads to the conclusion that nickel ions are released from orthodontic appliances in measurable amounts to human organism.

Keywords

Orthodontic applianceMetal releaseIn vivoSpectroscopic methods

Introduction

In the literature, a number of studies on identification of exposure and bioavailability of trace metals to human are found. These studies include analysis of various markers of exposure, both in vitro and in vivo. Many papers are of epidemiological character, and for this reason, populations of different sizes under different conditions are investigated. Frequently, it is very difficult to compare the results, because the studies were carried out under variable conditions and various research techniques were used. In medical sciences, which base practically only on clinical trials, a specific template of reporting results and writing reviews was elaborated. This facilitates comparisons of research original papers and enables to identify those with the highest level of evidence and eliminate unsound studies. For this reason, guidelines for so-called systematic reviews basing on meta-analysis were elaborated to facilitate drawing reasonable and general conclusions from clinical trials [1]. Systematic reviews report techniques used to prepare a review and experimental methods used by the authors of original papers. They also include the comparison and statistical elaboration of the results [2]. Specific guidelines of the preparation of systematic reviews were established and included in Quality of Reporting of Meta-analyses [3].

Recently, many papers investigating release of trace metals from orthodontic appliances have been published. Biocompatibility and resistance to biodeterioration of the materials was studied by various approaches. This results from the concern related with the possible release of trace elements during orthodontic treatment and their potential toxicity to patients. Wide range of appliances routinely applied during orthodontic treatment is made of alloys which contain cobalt, chromium, iron, nickel, titanium, among which of the major concern are nickel and chromium. The stainless steel brackets, wires, bands, auxiliaries as face bow, also elements of removable appliances, contain approximately 18% chromium, 8% nickel. Nickel–titanium archwires (NiTi wires) contain around 47–50% of NiTi [4, 5]. Nickel is known as strong immunologic sensitizer. Blanco-Dalmau et al. found 31.9% of the women and 20.7% of the men in a population of 403 showed a positive reaction to a patch test with nickel sulfate [6]. Both nickel and chromium can cause dermatitis, asthma, and do have mutagenic and cytotoxic effect [79]. Having that on mind, it is important to define amount of metal (nickel and chromium) released during orthodontic treatment and its influence on an organism. There are two ways to achieve it: by in vitro and in vivo studies. In accessible literature the majority of papers have established an in vitro approach which is not reliable to in vivo conditions.

The aim of the present work was to report the state-of-the-art in the research on methods used to assess exposure to trace metals from orthodontic appliances. Another objective was to implement a concept of a systematic review in papers which report and generalize results of research on exposure and bioavailability of metals to human in in vivo tests, using different techniques of metal analysis. In this work, various tests, which can be found potentially useful in monitoring of exposure from orthodontic appliances were discussed, pointing out on possible future trends in this topic. The paper is also an example of interdisciplinary approach towards discussion of problems on the boundary of human health and biological trace element research, reporting the problem of bioavailability and chronic exposure of children to metals released from orthodontic appliances.

Materials and Methods

In order to select the research papers describing release of metals from orthodontic appliances under in vivo conditions, we selected the following search criteria. The first search, which included the following keywords: orthodontic appliances, metal release, and in vivo, yielded 14 papers, among which ten in fact did not match inclusion criteria and concerned either different topic or were review papers. Since studies in vivo usually concern investigation of saliva, urine or serum, and the word ‘in vivo’ might not appear in the title or abstract of a paper, we have decided to increase the number of keywords. Finally, the following criteria were used to consider studies for this review: 1 orthodontic appliances, 2a metal release, 2b nickel, 3a in vivo, 3b salivary, 3c urinary, and 3d serum. The combinations of keywords and search results are presented in Table 1. To find articles which can match-mentioned criteria, we conducted a search PubMed database (from 1966 to September 2009). All the articles that seem to meet the inclusion criteria of the systematic review topic were selected, and the actual works collected. Eligibility of the selected studies was determined by reading the abstracts of papers identified by the search. The abstracts of related articles were reviewed to search for any similar studies that might have been missed. Following exclusion criteria were applied: other than English papers, exposure to trace metals caused by other factors, reviews, case reports, in vitro studies, or papers concerning other topic than release of metals from orthodontic appliances in vivo.
Table 1

PubMed Search Strategy

No.

Word or phrase

Results

1

Orthodontic appliances

16,514

2a

Metal release

51,748

2b

Nickel

23,809

3a

In vivo

486,721

3b

Salivary

46,825

3c

Urinary

543,040

3d

Serum

717,390

8

1 and (2a or 2b)

704

9

1 and (3a or 3b or 3c or 3d)

308

10

1 and (2a or 2b) and (3a or 3b or 3c or 3d)

49

11

1 and (2a or 2b) and (3a or 3b or 3c or 3d) not in vitro

35

Results

The PubMed search 35 chosen studies. No additional search was performed. From 35 identified studies, nine met the selection criteria. Excluded studies—with the reported reason of exclusion—are presented in Table 2. Included studies are listed in Table 3.
Table 2

Studies that Fulfilled Selection Criteria but were Excluded (listed as Further Reading)

Authors/date

Reason for exclusion

Baldwin et al., 1999

Different topic

Cortizo et al., 2004

Different topic

Elayyan et al., 2008

Different topic

Eliades et al., 2002b

Different topic

Eliades et al., 2001

Different topic

Eliades et al., 2000

Different topic

Evans et al., 1998

Different topic

Fricker, 1998

Different topic

Grimsdottir and Hensten-Pettersen, 1997

Different topic

Grosgogeat et al., 2003

In French

Ho and West, 1991

Different topic

Iijima et al., 2004

Different topic

Kapila et al., 1992

Different topic

Kebsch et al., 2007

Different topic

Lee et al., 2004

Different topic

Linder-Aronsonet al., 1996

Different topic

Pandis et al., 2005

Different topic

Ramalingam et al., 2008

Different topic

Rix et al., 2001

Different topic

Rondelli et al., 2000

Different topic

Rucker and Kusy, 2002

Different topic

Singh et al., 2008

Different topic

Tselepis et al., 1994

Different topic

Venza et al., 2002

Different topic

Venza et al., 2004

Different topic

Westphalen et al., 2008

Different topic

Table 3

Studies that Fulfilled Selection Criteria and were Included for Review

References

Materials and methods

Patients in group

Material

Metal

Methods

Statistics

Experimental

Control

Ağaoğlu et al. [15]

100 (in 5 subgroups)a

100 (in 5 subgroups)a

Saliva, serum

Cr, Ni

Electrothermal AAS

Mann–Whitney U test

Amini et al. [13]

30

30

Saliva

Co, Cr, Ni

AAS with a graphite furnace

Student t test

Eliades et al. [12]

17

7

Saliva

Cr, Fe, Ni

ICP-AES

Two-way ANOVA, Tukey's test

Faccioni et al. [10]

55

30

Mucosa oral cells

Co, Ni

ICP-MS

Student t test

Kerosuo et al. [16]

47a

47a

Saliva

Cr, Ni

Electrothermal AAS

Wilcoxon’s matched-pairs

Kocadereli et al. [17]

45

15

Saliva

Cr, Ni

AAS

Wilcoxon matched-pairs

Matos de Souza et al. [14]

30a

30a

Saliva

Co, Fe, Ni

AAS with a graphite furnace

Kruskal–Wallis

Menezes et al. [18]

21a

21a

Urine

Ni

AAS

Student t test

Petoumeno et al. [11]

18a

18a

Saliva

Ni

ICP-MS

Wilcoxon's matched-pairs

aThe material was sampled from the same patients before and after placement of orthodontic appliance

Table 3 gives a brief description of research methodology: the size of the studied groups (experimental and control), biomarker of exposure selected, metals investigated, and description of statistical methods used in the elaboration of the results. In the studies, two methodological concepts were employed. According to the first, the patients were divided into subgroups: control (consisting of 7–30 individuals which did not undergo orthodontic treatment) and experimental (7–55 patients). Another approach was to investigate the composition of the material collected from patients before the treatment and then during the therapy. In this case groups, 18 of 100 (in five groups) patients were chosen for the research. The material investigated included first of all saliva (in seven out of nine papers) and also there were single reports found on using serum, urine, and mucosa oral cells. Metals, the concentration of which was determined are as follows: Ni (all papers), Cr (five out of nine), Co (three studies), and Fe (two studies). Titanium was not investigated, although it is the component of NiTi archwires. Analytical techniques used in the tests were ICP-MS [10, 11], ICP-AES [12] and AAS with graphite furnace [13, 14] or electrothermal AAS [15, 16] or AAS [17, 18] (Table 3). All the techniques (with the exception of ICP-AES) were found useful. The latter was found to have detection limit close to the concentration of the analyte. More popular were non-parametric tan parametric tests. The latter require normal distribution of the results. For matched-pairs, either non-parametric Wilcoxon's test was used (three works) or Kruskal–Wallis (one work), or parametric student's t test (one matched-pairs study). Student's t test was also applied (in two works) to test differences for unlinked variables with normal distribution. In one work also 2-way ANOVA, Tukey's test was employed.

Below, detailed information on the results and conclusions stated in the papers chosen for the present systematic review is provided.

Ağaoğlu et al. (2001) [15] evaluated saliva and serum samples for the concentration of chromium and nickel ions from patients with fixed orthodontic appliances. The group of 100 patients was divided into five subgroups—the material was sampled from the same patients before and after placement of orthodontic appliance (1 week, 1 month, 1 year and 2 years). In the conclusions, the authors stated that in saliva samples nickel and chromium reached the highest levels in the first month after placement of the appliance and decreased to the level before orthodontic treatment. In the serum—nickel/chromium levels reached their highest peak after 2 years after appliance placement [15].

Amini et al. evaluated metal levels (nickel, chromium, and cobalt) in oral mucosa cells in a group of 30 patients, and a control group of 30 patients. The metal content determinations were carried out with use of atomic absorption spectrometry with a graphite frunace. Authors did not find statistically significant differences in chromium and cobalt levels in oral mucosa cells between the test group and the control at p < 0.05 level throughout, although the mean values differed for Ni (0.44 in the control group and 0.84 ng/ml in the group of patients, p = 0.1). The nickel level was higher in the test group comparing to control (21.74 and 12.26 ng/ml, respectively) [13].

Eliades et al. [12] evaluated saliva of patients with orthodontic appliances for metal traces (nickel, chromium, and iron). A test group consisted of 17 patients undergoing orthodontic treatment for over a year. A control group consisted of seven untreated individuals. Salivary samples obtained from patients were analyzed by ICP-AES (no information on the type of nebulization was provided). Authors concluded that there was no significant difference between patient and control group with respect to salivary metal level, regardless of element. The concentration of metals in saliva was in some cases below detection limit of the instrument [12].

Faccioni et al. [10] evaluated metal levels (nickel and cobalt) in oral mucosa cells in the group of 55 patients with orthodontic appliances (20 brackets, 4–8 bands each). Control group—30 patients without appliances. Epithelial cells of buccal mucosa from each patient were collected. The collected cells went through potential DNA damage evaluation test and measurement of metal content (ICP-MS) was undertaken. Authors reported increase of cobalt and nickel level in mucosa cells in a group undergoing orthodontic treatment; 2.8-fold and 3.4-fold higher, respectively. The potential genotoxic effects of metals on mucosa cells was evaluated with use of comet test (strand breaks, alkali-labile sites, incomplete excision repair sites). Authors noted that higher frequencies of cells with comets and apoptosis were found in the patient group. The parameters characterizing DNA breaks were higher in the cells of the orthodontic patients than in the controls. It was found that the presence of Ni and Co released from orthodontic appliances induced DNA damage and reduced cellular viability of mucosa cells. Metals were released during the first 4–5 months [10].

Kerosuo et al. (1997) [16] also investigated the level of nickel and chromium in the saliva of patients which underwent treatment with the use of fixed appliances. The experimental group consisted of 47 patients. There was no separate control group. Saliva was sampled before insertion of the orthodontic braces, after 1–2 days, 1 week, and 1 month after the orthodontic treatment began. The conclusion was that concentration of nickel and chromium in saliva has not significantly increased as the result of orthodontic therapy.

The topic of the subsequent study by Kocadereli et al. [17] was to evaluate chromium and nickel levels in the saliva of patients. In the experimental group, 45 patients were included and the control group consisted of 15 individuals. Saliva samples were collected from patients before appliance placement and after (1 week, 1 month, and 2 months). In the results, authors underlined that the release of chromium ad nickel ions was not significantly different during the first 2 months of treatment [17].

Matos de Souza et al. [14] evaluated saliva of patients with orthodontic appliances for metal traces (nickel, chromium, and iron). A group of 30 patients underwent the experiment (time, 60 days). There was no control group; however, saliva of each patient was sampled before the experiment. The collected saliva went through a measurement of metal concentration (ICP-MS). Large variation between individuals was encountered which was explained by the effect of diet, period of day, psychic conditions etc. Metals were released from various alloys in similar quantities—no statistically significant differences were found. Authors stated that ion concentrations (nickel and iron) increased immediately after appliance placement (the peak concentration was detected after 10 min, 1 day) [14].

Menezes et al. [18] controlled nickel level by another biomarker of exposure—urine. Samples from the experimental group (21 patients) were collected before appliance placement and two months after. In conclusions, the authors stated that urinary nickel levels increased significantly after 2 months.

Petoumeno et al. [11] evaluated nickel concentration in saliva samples from patients with orthodontic appliances. The studied group consisted of 18 patients from which saliva was taken for the analysis before and after placement of brackets and bands. Sampling was carried out after 2 weeks—before placement of NiTi archwires, immediately after, 4 and 8 weeks after placing the wires. Nickel release occurred immediately after placement of bands and brackets and after placement of NiTi archwires and decreased after 10 weeks [11].

Table 4 reports the levels of trace metals (Co, Cr, Fe, and Ni) released to the saliva in the group of patients undergoing orthodontic treatment and in the control group together with statistical interpretation of significance of differences between the groups. The final conclusions in the papers are coherent, although the concentrations of metals in the saliva differed between the studies. This is the consequence of using different sample preparation techniques (filtration, digestion) as well as different analytical methods. In all the studies, except of [12, 17], increased level of all metals in the studied samples was confirmed. Amini et al. [13], Matos de Souza [14], and Faccioni et al. [10] reported 2–3 times higher level of cobalt and nickel ions and 20% higher level of chromium in the saliva of patients than in the control group. The differences were found statistically significant for: Co in saliva [10], Cr in serum and saliva [15] and Ni in saliva and mucosa oral cells [10, 11, 13, 15]. In the work of Amini et al. [13] the differences for Co and Cr were not significant, although p level was low (0.09 and 0.10). Increased number of the individuals in the groups could confirm statistical significance of differences between the groups. Another conclusion stated Eliades et al. [12] who found that in six out of four data sets, the concentrations were below detection limit of the analytical instrument.
Table 4

Concentration of Metals in Evaluated Samples of Saliva, Serum, and Urine ng/ml

References

Material

Concentration of metal ions

Co

Cr

Fe

Ni

Experiment

Control

Experiment

Control

Experiment

Control

Experiment

Control

Ağaoğlu et al. [15]

Saliva

0.53–4.45b

0.76b

4.12–11.53b

4.45b

Ağaoğlu et al. [15]

Serum

6.16–10.98b

6.21b

7.87–10.27

8.36

Amini et al. [13]

Saliva

0.84

0.44

4.24

3.46

21.74b

12.26b

Eliades et al. [12]

Saliva

  

27c

20c

14–17

21c

10c

11–18

Faccioni et al. [10]

Mucosa oral cells

0.568b

0.202b

    

2.521b

0.725b

Kerosuo et al. [16]

Saliva

  

68–86

61

  

50–65

55

Kocadereli et al. [17]

Saliva

0.49–1.98

2.20–3.43

0.49–0.67

1.16–1.46

Matos de Souza et al. [14]

Saliva

0.3–1.7

0.6

28–104

94

1.7–16

5.3

Menezes et al. [18]

Urine

19.89

17.67

Petoumeno et al. [11]

Saliva

28–78b

34b

aThe material was sampled from the same patients before and after placement of orthodontic appliance

bThe differences between the groups were statistically significant

cBelow detection limit

Discussion

The majority of studies which investigate metals release from orthodontic appliances, the aim of which is to test biocompatibility of the materials, report concentrations of Co, Cr, Ni, and Fe in solutions of NaCl and artificial saliva (in vitro tests) or in saliva, blood, and urine (in vivo tests). In vitro tests are quick, simple, enable to assure controlled laboratory conditions, however, do not exactly reflect phenomena occurring in oral environment. The studies in vivo which use saliva as the biomarker of exposure and the first diluent usually report momentary, static (not dynamic) level, frequently after rinsing with water. Consequently, other factors which change the momentary concentration of metals are not included, but are related with eating or drinking (acidic conditions) and teeth brushing (toothpaste) [4, 13].

In vivo tests confirmed release of Ni and Cr and agree that the quantity of released metals is not proportional to their content in an alloy [13, 14]. Problematic is that in some cases the levels of metals are below lower detection limit of analytical instruments, meaning that are not measurable [12] (Table 4). Even if the concentrations are measurable, the authors of the majority of papers conclude that the doses of metals released from orthodontic appliances were far below the dietary intake and did not reach toxic doses [10]. WHO guideline values for drinking water (2008) currently accept the concentration for Cr 50 ng/ml and for Ni 70 ng/ml, which is an order of magnitude higher than in saliva of patients (Table 4) [19].

Recently, some problems related with the interpretation of the results of metal ions concentration in saliva were signalized [14, 20]. Fors et al. investigated the concentration of metals in both saliva and in plaque. The results in saliva were very similar and did not differ statistically (0.004 and 0.005 µg/g, respectively, p = 0.6) in the control group and in the group of patients, while high and statistical differences were found between metals content in plaque (1.03 and 0.45 μg/g, respectively, p = 0.001) [20]. This means that microorganisms in the biofilm biotransformed metals into ions, solubilized and bioaccumulated or biosorbed them in the biomass. This also shows that even if the concentration of metal ions in saliva remains unchanged after placement of orthodontic appliances, metals can be bioavailable and can be absorbed through intake of plaque which contains metals, which is actually not measured by widely accepted procedures.

In the majority of studies on in vivo tests on identification of metals release from orthodontic appliances, saliva [1117], serum [15], mucosa oral cells [10], and urine [18] were selected as the biomarkers of exposure, although these media are usually used in monitoring momentary, not chronic exposure to trace metals. This creates a lack of cumulative data [12]. Metals release from orthodontic appliances could be treated as chronic exposure to toxicants. Perhaps it would be appropriate to investigate not only release of metals to saliva which is diluent, but also sites of metals accumulation in an organism of human. In toxicology and environmental sciences, markers of exposure to toxic metals, where the concentration of toxicant is determined in in vivo tests, generally include sites of elimination or accumulation. Frequently, it is impossible to conduct measurements at the site of accumulation (internal organs—liver and kidneys) by non-invasive means. Therefore, it is recommended to use excretion routes to develop biological markers of exposure [21]. Also, if using in vivo tests on human, it is necessary to identify other sources of exposure to elements released from orthodontic appliances. It would be advantageous that the subjects would fill a questionnaire which would include questions about other potential sources of metals potentially released from orthodontic appliances in their organism.

The main route of metals elimination is urine, because of hydrophilic nature, and small size of metal ions and their compounds. Resulting from binding of metal cations to thiol groups of cysteine in proteins, another route of excretion is via keratin materials: hair and nails [21]. The follicle of hair is perfused and chemicals from blood become deposited partly in hair shaft during the growth of hair, however detailed mechanisms are not fully understood, yet [22], as well as kinetics of incorporation of trace elements in hair [23]. Although hair mineral analysis is recommended by The Environmental Protection Agency and International Atomic Energy Agency as one of the most important biomarkers of chronic exposure of human to metals [24], the technique needs to be improved.

Conclusions

The general conclusion in the studies is that metal ions are released only in the initial stage of the treatment. However, the majority of studies covered 1–2 months long period and did not reflect long-term changes nor the impact of the complete treatment, the duration of which is several years, on the whole organisms and the overall accumulation of metals from orthodontic appliances. In studies which evaluated nickel concentrations in blood and urine significant differences were found. It leads to the conclusion that nickel ions are released in measurable amount to human organism. We point out on the lack of in vivo long-term studies which monitor chronic exposure over several years of therapy. The size of the experimental groups in in vivo studies was rather small (7–47 subjects). Among the reasons for the limited number of patients in the experimental groups the following can be mentioned: invasiveness related with collection of samples (blood), inconvenience (urine and saliva). Consequently, many patients denied to volunteer in the experiment. Usually procedures of evaluation are carried out in external, highly specialized laboratories, since expected concentrations of analytes are very low (nanogram/milliliter) and for this reason required the use of instruments with very low detection limits (e.g., ICP-MS).

Many studies concluded that the dose of metals released by orthodontic appliances was far below the toxic levels. The concentrations of metal ions reported for saliva were far below the maximum acceptable concentrations of these metals in drinking water. This shows that the exposure from orthodontic appliances might be negligible as compared with others (e.g., drinking water). This conclusion was similar in in vivo as well as in in vitro reports. The concentrations of metals released to saline solutions [7] or artificial saliva [25] in in vitro studies were of the same magnitude as in in vivo investigations discussed in the present paper. Also, the conclusions concerning low dosages of metals (below daily dietary intake) were consistent with in vivo reports

Nowadays, advanced techniques are frequently used in the assessment of degree of release of various substances by materials places in oral cavity in order to test their biodeterioration and biocompatibility [26, 27]. Consequently, the most useful future course of in vivo tests of determination of exposure to metals from orthodontic appliances seem non-invasive analyses of urine, hair or nails, or blood (which is invasive) [28]. In the literature many papers reporting tests on release of metals from orthodontic wires and brackets in vivo by using saliva, urine, or blood can be found. No research on using hair nor nail analysis has been reported, although they can be found to be useful in biomonitoring of exposure from dental appliances in the future.

Reassuming, to achieve reliable and comparable results, in vivo tests should be carried out under strictly controlled conditions so the results would be reproducible and meaningful. There is a need to elaborate certain rules and standards for in vivo studies in this area of interest.

Copyright information

© Springer Science+Business Media, LLC 2009