A comparative evaluation of salivary and plasma bisphenol-A levels in patients released from four different orthodontic appliances: ultra-performance liquid chromatography–mass spectrometry study. A non-randomized clinical trial

Abstract

Objective

To evaluate the amount of bisphenol-A (BPA) released from different orthodontic appliances into saliva and plasma and to compare BPA levels between orthodontic appliance groups.

Methods

This non-randomized clinical trial included 78 patients, and they were divided into four groups: fixed appliance, temporary anchorage devices (TADs), removable appliance, and functional appliance groups. Saliva and plasma samples were collected before treatment/appliance wear (T0) and one day (T1), one week (T2), and one month (T3) after treatment/appliance wear. Saliva and plasma BPA levels were analysed using Ultra-performance Liquid Chromatography–Mass Spectrometry (UPLC-MS/MS). Data was analysed using analysis of variance (ANOVA) and Tukey HSD multiple comparison tests with Bonferroni correction. Correlations between saliva and plasma BPA levels at different time points were evaluated using Pearson correlation analysis.

Results

BPA levels ranged from 0.07–7.28 ng/ml in saliva samples and from 0.17–12.51 ng/ml in plasma samples in all groups. Within-group comparisons showed no significant changes over time (p > 0.0125) in each group. Between-group comparisons showed significantly higher saliva BPA levels in the fixed appliance group compared to the other groups (p < 0.017), with no significant differences in plasma BPA levels between groups (p < 0.0175). Positive correlations were observed between saliva BPA measurements, whereas positive correlation was observed only between T0 and T3 plasma PBA levels.

Conclusion

This research obviously demonstrated that the orthodontic appliances we investigated release BPA into the saliva. However, the BPA levels detected in both saliva and plasma are consistently within the acceptable safety limits. However, precautions should be taken to minimize BPA release during orthodontic treatment.

Bisphenol-A (BPA), identified by CAS number 80-05-7, is an organic and synthetic chemical manufactured for industrial and professional laboratory use [1]. Although initially synthesized as a synthetic oestrogen in the 1940s, it has been used in the manufacture of plastics for its ability to impart hardness and transparency [2, 3]. Today, BPA is widely used to manufacture polycarbonate and epoxy resins. BPA was first reported to leach from plastic products and to be an endocrine-disruptor with oestrogenic activity by Krishnan et al. [4] Following this finding, further research has shown that BPA leaches from various products, particularly food packaging. Exposure to BPA has been detected in various body fluids such as blood, saliva, and urine. Studies on health issues related to this exposure have reported that BPA may be associated with endocrine disruption, including reproductive health, obesity, and thyroid disorders [5].

In 1996, Olea et al. [6] reported that BPA was released from dental resins for the first time, leading to further research into dental treatment outcomes. Studies have also shown that BPA can be released as a monomer from orthodontic materials [7]. Low levels of BPA release from orthodontic appliances have been detected in vitro [8]. It is known that BPA derivatives such as Bis-GMA and Bis-DMA are used in the manufacture of dental materials used in orthodontic treatment. It is thought that un-polymerized monomers and/or end monomers containing open bonds undergo hydrolysis reactions to convert to BPA [9]. Some researchers [10] suggest that the BPA released from orthodontic appliances may be below the daily intake dose but should still be considered. However, other researchers [11] have found that the BPA released from orthodontic appliances has no cytotoxic or oestrogenic adverse effects.

In this context, it is crucial to determine the amounts of BPA released from materials used during orthodontic treatment and to compare different treatment methods. However, there is no comprehensive study in the current literature on this topic. This study aimed to evaluate the amount of bisphenol-A (BPA) released into saliva and plasma from different orthodontic appliances and to compare the BPA levels between different orthodontic appliance groups. Specific objective of the present study is to determine BPA levels in the saliva and blood plasma samples of the subjects underwent four different treatment/appliance wear including fixed orthodontic appliances and temporary anchorage device (TADs), removable and functional appliances, and to compare BPA levels between the appliance groups.

Material and methods

This non-randomized clinical trial was approved by the Clinical Research Ethics Committee of Faculty of Medicine, Atatürk University (Approval No: 2018/16). Patients and/or parents who volunteered to participate in the study were informed about BPA and the research methodology and gave their informed consent to participate in the study. Helsinki Declaration guidelines are followed in this non-randomized clinical trial.

The sample size of this study was calculated with a statistical power of at least 95% and a type 1 error of 5%, based on a similar study [12]. A minimum of 17 individuals were required to be included in each group.

Initial sample size was included 78 subjects who volunteered to enrol in this study. Volunteers were divided into four different groups according to their treatment method. These groups consisted of:

• Fixed appliance group, 20 volunteers (8 male, 12 female, mean age: 17.8 years; range: 12 –22 years)

• Temporary anchorage device (TADs) appliance group, 20 volunteers (3 male, 17 female, mean age: 17 years; range: 11–20 years)

• Removable appliance group, 19 volunteers (6 male, 14 female, mean age: 12.94 years; range: 8–20 years)

• Functional appliance group, 19 volunteers (10 male, 9 female, mean age: 12.48 years; range: 10–15 years)

The following criteria have been established for the inclusion of volunteers in the study:

1. No systemic disease,

2. No previous orthodontic treatment

3. Not chronically exposed to plastic materials,

4. No composite restorations,

5. Not taking medications regularly or recently,

6. Patients in the fixed orthodontic appliance group must have a non-extraction orthodontic treatment plan.

Primarily outcome variable of the present study was the BPA levels in saliva and blood plasma samples. Saliva and plasma samples were collected before treatment/appliance wear (T0) and one day (T1), one week (T2), and one month (T3) after treatment/appliance wear. Saliva and plasma BPA levels were analysed using Ultra-performance Liquid Chromatography–Mass Spectrometry (UPLC-MS/MS).

Clinical procedures

In the fixed appliances group, the procedures were meticulously performed with isolation ensured by high-power dental suction. The sequence of the procedure was as follows: first, etching (Pulpdent Etch Royale, Watertown, USA, for 40 seconds), then bond was applied, and each tooth was light-cured for 20 seconds (Woodpecker LED D, Guangxi, China). Metal brackets (Unitech™ Gemini Metal Brackets, Minnesota, USA) were bonded to the teeth by applying bond and composite (3M Unitek Transbond XT, Minnesota, USA) to the base of the brackets. The excess composite was carefully cleaned, and each bracket was light-cured for 20 seconds. Treatment of both jaws was started in the same session. A metal ligature wire was used to attach the arch wire to the brackets.

The removable appliance group did not differ in terms of working principle. All appliances were worn in the maxilla. In the functional appliance group, the activator was worn for at least 18 hours daily. Clear acrylic (Orthocryl, Dentaurum, Ispringen, Germany) was used to prepare the appliances, and polymerization was carried out at 45–50°C, under 2.5 atmospheres of pressure, for 30 minutes.

The TADs appliance group consisted of a mixture of a modified Nance appliance and a modified Keles Slider appliance (Fig. 1). A mini-screw (Lomas, 2 × 7 mm, Mondeal, Mühlheim, Germany) was placed in the third rugae area and a button-shaped orthodontic acrylic was prepared around it. Light-cured resin (3M Unitek Transbond Plus, Minnesota, USA) was applied to the bands. In the mini-screw space created in the acrylic button, a bond and then a tube of compomer (3M Unitek Transbond LR, Minnesota, USA) were applied to the mini-screw space created in the acrylic button. After placement, the excess resin was removed and all resin-containing surfaces were light-cured for 20 seconds each.

Unstimulated saliva samples were collected with propylene tubes (ISO Lab, Eschau, Germany, Product code: 078.02.002), and blood samples were collected with gel-containing vacuum tubes (BD Vacutainer SST II Advance Serum Separator Tubes, Dubai, UAE) with a volume of 5 ml each. After resting for 30 minutes at room temperature, the tubes were centrifuged at 3000 rpm for 10 minutes to precipitate the samples. The supernatant was then stored at −80°C until the day of analysis. Saliva and plasma samples were collected before the treatment/appliance wear (T0), one day (T1), one week (T2), and one month (T3) after the treatment/appliance wear.

Laboratory studies

Thawed samples were vortexed for 1 minute. Then, 250 μl of plasma was taken, and 750 μl of methanol was added to precipitate plasma proteins, and 500 μl of saliva was taken, and 500 μl of methanol was added to precipitate saliva proteins. The prepared samples were centrifuged at 9000 rpm for 5 minutes at 20–25°C. After centrifugation, 350 μl of the supernatant was removed from the samples, filtered (PTFE.22 μm, Jaytee, Kent, United Kingdom), and transferred to a vial (Agilent 2ml clear screw cap vial part no.; 5182-714, Agilent Technologies, California, USA).

Analysis was performed by UPLC-MS/MS [13]. A reverse-phase C8 analytical column (Zorbax Eclipse XDB-C8, Agilent Technologies, Loveland, CO, USA; 4.6 mm, 150 mm, 5 microns) was used for chromatographic separations at 25°C. The mobile phase consisted of a mixture of methanol and ammonium acetate in channel A and acetonitrile and ammonium acetate in channel B, with gradient elution for separation. A jet stream electrospray ionization source, operated in negative ion mode, was used for mass spectrometric analysis. Ions with an m/z ratio of 227.1→212.0 were selected for analysis in Multiple Reaction Monitoring (MRM) mode. A retention time of 6.80 minutes was determined for bisphenol A. The method’s limit of detection (LOD) value was calculated based on the value when the signal/noise ratio was 3/1, and the limit of quantification (LOQ) value was calculated based on the value when the signal/noise ratio was 10/1. This study determined the LOQ value as 6.25 ng/ml and the LOD value as two ng/ml. The linearity range of the study was determined as 6.25–100 ng/ml. The calibration curve was obtained by linearly plotting the peak areas of bisphenol-A solutions prepared at concentrations of 6.25, 12.50, 25.00, 50.00, and 100 ng/ml against the concentration. The LC-MS/MS chromatography of the bisphenol-A standard is shown in Fig. 2. The mass spectrum for BPA is shown in Fig. 3. Mass Hunter QQQ Quantitative Analysis Software 6460 UPLC/MS-MS Version B.07 package program was used for the analytical calculations. The levels of BPA detected in the plasma and saliva samples were measured in ng/ml.

Statistical analysis

The Monte Carlo Chi-squared test was used to compare sex and groups. A one-way ANOVA test was used to compare age and groups. Changes in mean BPA levels of saliva and plasma were separately compared between examination times with repeated measures analysis of variance (ANOVA) with Bonferroni correction in each group, and a significance value of P < 0.0125 was considered significant. Between-group comparisons and group × time interaction in plasma and saliva samples were analysed separately by multiple analysis of variance (ANOVA) with Bonferroni correction, and a significance value of P < 0.017 was considered significant. In addition, between- and within-group comparisons and their interactions were performed while adjusting for the role of patient sex and age (2-way repeated measures ANCOVA). When a statistically significant difference was observed between groups or measurement times, this difference was compared with the Tukey HSD multiple comparison test to determine which groups or times were statistically significant. The correlation between salivary and plasma BPA levels at different time points was evaluated using Pearson correlation analysis. SPSS® 25.00 software package was used for data analysis.

Results

Final sample size was also included 78 subjects; 20 subjects (8 male and 12 female) in fixed appliance group; 20 subjects (3 male and 17 female) in TADs appliance group, 19 subjects (6 male and 14 female) in removable appliance group and 19 subjects (10 male and 9 female) in functional appliance group.

BPA levels ranged from 0.07–7.28 ng/ml in the saliva samples and from 0.17–12.51 ng/ml in the plasma samples. There was no significant difference in the gender distribution between the groups (p = 0.134). However, when the age distribution was evaluated, it was found that the fixed appliance and TADs groups consisted of significantly older individuals compared to the removable and functional appliance groups (p = 0.00).

Within-group comparisons

In saliva samples, BPA levels ranged from 0.24–7.28 ng/ml in the fixed appliance group, 0.07–5.46 ng/ml in the TADs appliance group, 0.08–3.69 ng/ml in the removable appliance group, and 0.16–2.89 ng/ml in the functional appliance group. In the plasma samples, BPA levels ranged from 0.33–9.76 ng/ml in the fixed appliance group, 0.17–7.58 ng/ml in the TADs appliance group, 0.18–12.51 ng/ml in the removable appliance group, and 0.24–6.61 ng/ml in the functional appliance group. The average BPA levels in saliva and plasma at different time points according to different appliance groups are shown in Tables 1 and 2, respectively. Average saliva and plasma BPA levels did not show significant changes in all groups (p > 0.0125) (Table 3). Gender and age covariates did not show a significant effect (Table 6).

Between-group comparisons

Saliva BPA levels

Salivary BPA levels differed significantly between the groups (p < 0.0175). However, no statistically significant difference was observed between the measurement times and the interaction of group × measurement time (p > 0.017) (Table 4). Multiple Tukey HSD comparison tests showed that the average saliva BPA level in the fixed appliance group was significantly higher than in other groups (Table 1). The mean salivary BPA level in the TADs appliance group was significantly higher than that in the removable appliance group. No statistically significant difference in mean salivary BPA levels was found between the functional and TADs appliance groups. Similarly, no statistically significant difference in mean saliva BPA levels was found between the functional and removable appliance groups. Gender and age covariates did not significantly affect the BPA levels (Table 6).

BPA levels in plasma

No statistically significant difference was found in plasma BPA levels between groups, between the measurement times, or in group × measurement time interaction (Table 5). Gender and age covariates did not show a significant effect (Table 6).

Correlations

Among the salivary measurements, a strong positive correlation was observed between T0 Saliva and T1 Saliva BPA levels (r = 0.475, p < 0.01) and between T1 Saliva and T2 Saliva BPA levels (r = 0.482, p < 0.01). In addition, moderate positive correlations were found between T0 Saliva and T3 Saliva BPA levels (r = 0.310, p < 0.01) and between T1 Saliva and T3 Saliva BPA levels (r = 0.308, p < 0.01). A positive and statistically significant correlation was also found between T0 Plasma and T3 Plasma BPA levels (r = 0.286, p < 0.05) (Table 7).

Discussion

BPA, a chemical that can be released from polymers, is associated with various health problems and is one of the major concerns today [5]. Debates continue about the safety of its use in various areas of daily life. Its use in baby products has been legally banned in many countries [14]. One of the key issues in these discussions is the use of dental resins [5]. It is known that BPA derivatives are present in the structure of dental polymers [9]. BPA release from dental materials has been studied mainly in in-vitro studies.

In orthodontics, most studies on BPA have been conducted in vitro [7, 10, 11, 15–20]. In in-vivo studies, BPA levels have been studied only in saliva [12, 21, 22]. No studies have investigated BPA levels in blood samples.

In our study, saliva and blood samples were taken before appliance wear to determine pre-treatment BPA levels. Significant amounts of BPA were detected in saliva and plasma samples taken before each treatment (T0 time), indicating that patients are exposed to BPA through diet or environmental factors. The BPA levels obtained at T0 time showed individual variations, which are thought to be due to patients from different socio-cultural and socio-economic backgrounds.

The primary source of BPA exposure is diet, but studies have also reported exposure to BPA from environmental sources [5]. Previous studies [5] have attempted to determine the amount of daily BPA exposure, but different BPA levels have been reported due to the subjective nature of the assessment. The amount of daily BPA exposure is influenced by external factors such as occupation, living environment, and dietary habits. Daily BPA exposure varies from person to person, depending on different lifestyles [23].

Due to the association of BPA with systemic diseases and the potential effects of medications on BPA metabolism, individuals who regularly took medications and did not have systemic diseases were included in the study [24]. To minimize environmental exposure to BPA, individuals who had prolonged contact with plastic products (including dental resins) were excluded from the study. It has been reported that an increase in the number of resin-coated surfaces increases BPA exposure [25]. Therefore, the study did not include individuals with planned treatment for extraction or missing teeth. In addition, patients who had undergone previous treatment were excluded to ensure objective assessment, as they had previously been exposed to orthodontic materials. In in-vitro studies, the release of BPA from elastic ligatures and chains has been reported [20]. Therefore, wire ligatures were used in the study instead. The type of light source used to cure the resin and the distance between the light source and the resin has been reported to influence BPA release [26]. Therefore, LED light was used at the closest distance in this study.

Several studies [5], have found BPA in various body fluids. In-vivo assessments have been performed on mouthwash, saliva, and urine. However, the chemicals in mouthwash fluids may affect the release of BPA, so this sampling method was not considered appropriate for the study [12, 21, 22, 27]. However, unstimulated saliva has been reported to reflect the major condition of the oral cavity and provide a more accurate composition for analysis [28]. Therefore, unstimulated saliva samples were collected in our study. It is known that orally ingested BPA can undergo first-pass metabolism in the liver and convert to its passive form [29, 30]. BPA not conjugated in the liver enters the circulation and may have endocrine-disrupting effects on tissues and organs. Therefore, blood samples were collected to determine the amount of BPA in the plasma.

Studies on BPA have used several methods, including ELISA, HPLC, LC-MS/MS, and GC-MS/MS, to analyse saliva and plasma samples. Some researchers have reported that ELISA may not be suitable for determining BPA levels in human samples [31]. LC-MS/MS and GC-MS/MS methods can provide more sensitive measurements than HPLC. Among the analytical methods used, mass spectrometry, especially isotope dilution mass spectrometry, is considered to be the most reliable method for measuring trace levels of BPA in biological samples [5]. Due to its ability to perform more sensitive analysis, the samples were analysed by UPLC-MS/MS. However, due to various reasons related to the exporting company, the BPA-D16 isotope could not be obtained during the study, so the isotope dilution method could not be applied.

Transbond Tx composite has previously been evaluated in in-vitro studies. Some authors [17, 20]reported BPA release from the composite and an increase over time, while others [18, 19]reported no BPA release. Previous studies have reported a statistically significant increase in BPA levels in saliva samples taken half an hour after treatment compared to baseline. However, there was no statistically significant difference from baseline at later time points [12, 22]. Also, it has been shown that exposure to BPA during bonding can be reduced with protective measures [27, 32]. Because the amount of BPA released into the oral cavity during the first half hour varies with environmental conditions, our study evaluated changes associated with the device on the first day, when the oral cavity was more stable. The other time intervals chosen for the present study are consistent with those in other in-vivo studies [12, 21]. Manoj et al. [12] conducted a study comparing the release of BPA into saliva from two different composites (light-cured, chemically cured). HPLC then analysed the samples. Before the treatment, no BPA was detected in saliva samples from either group. In the chemically cured composite group, BPA concentrations decreased from 19.6 μg/ml–1.2 μg/ml over time, while in the light-cured composite group, they decreased from 11.2 μg/ml–0.6 μg/ml. The light-cured composite used in the study of Manoj et al. [12], however, unlike their study, samples were not collected until 30 minutes after placement. When comparing the results obtained with light-cured composites between the two studies, the salivary BPA concentrations in our study (Table 1) were much lower. This difference could be due to the sensitivity of the analytical method. The analysis method used in this study is capable of more sensitive measurements. In a study conducted by Raghavan et al. [21] on 45 patients, the amounts of BPA released from different retention appliances were compared using HPLC. In the heat-cured acrylic group, the highest BPA concentration detected was 0.00091 ppm, while in the chemically-cured acrylic group, the highest BPA concentration detected was 0.06031 ppm. Higher levels of BPA were observed when comparing acrylic appliance groups with Raghavan et al. [21] This difference could be due to differences in the sensitivity of the analytical methods, the types of acrylics used, and the processing of the acrylics. The acrylics used in this study vary according to their polymerization reactions, resulting in different processes in the preparation of the appliances. It has been reported that an increase in temperature increases the release of BPA [33–35]. In the Raghavan et al. [21]study, the acrylic was exposed to higher temperatures and extended periods in hot water. This is likely to have resulted in more BPA being released into the water used during the procedure, resulting in less BPA being detected in the saliva samples collected from the patients. There has been no study to compare our findings of plasma BPA levels. This is the first study to investigate the amount of BPA released into plasma from different orthodontic appliances. Although the amount of BPA released from different dental materials has been investigated in studies focussing on blood, this study is the first to detect BPA in plasma [36, 37].

This study found no significant difference in the gender distribution between the groups. However, when the age distribution was analysed, it was found that the fixed appliance and TADs groups consisted of significantly older individuals than the removable and functional appliance groups. This difference may be related to the treatment options based on age. In particular, removable and functional appliances are treatment modalities typically used at younger ages, whereas TADs appliances (Nance or Keles slider appliances) are more commonly preferred at older ages. The study also found that sex and age as covariates had no effect within or between groups. A literature review showed that no studies had evaluated the effects of age and sex on BPA exposure.

No statistically significant difference was found in the average BPA levels in saliva samples across different time points. However, when comparing the groups, it was observed that the mean BPA level in the fixed treatment group was significantly higher than in the other groups. Conversely, the removable appliance group had significantly lower BPA levels than the TADs and fixed appliance groups. This result is primarily due to the different levels of BPA detected in these groups at T0 (Tables 1, 2). In addition, resins containing significantly different BPA derivatives are used directly in the fixed treatment and TADs groups, whereas only acrylic is used in the removable appliance group. The difference in resin usage between the fixed and TADs groups and the acrylic usage for the activator and removable appliance should be considered as potential contributors to the variation observed between these groups. In addition, the lack of a significant difference between the removable and functional appliance groups in the study suggests that the increased amount of acrylic does not lead to a significant change in salivary BPA levels. Although there were significant differences in mean salivary BPA levels, evaluation of mean plasma BPA levels obtained from different groups did not reveal significant differences over time. This finding suggests a more static movement of BPA in plasma than in saliva. The amount of BPA detected is lower because the mouth is constantly rinsed with saliva. In contrast, BPA, which can bypass the body’s detoxification mechanisms and enter the plasma, fluctuates in more balanced amounts. Correlation analyses revealed significant relationships between saliva samples at different time points (Table 7). In particular, strong positive correlations were observed between consecutive time points (T0-T1 and T1-T2; 𝑟 = 0.475 and r = 0.482, p<0.01), suggesting that salivary BPA levels show consistent temporal changes. In contrast, in most cases, plasma samples showed no significant correlations, except for a statistically significant relationship between T0Plasma and T3Plasma (r = 0.286, p<0.05). These results suggest that saliva could serve as a reliable matrix for monitoring BPA levels, although a comprehensive assessment may require the combined analysis of saliva and plasma. Although the primary concern is that BPA in saliva may enter the bloodstream, it should also be considered that the amount of BPA in saliva may change due to the secretion of BPA from blood into saliva.

This non-randomized clinical trial has limitations. The blood sampling was done without clinical need, and it was done for scientific purposes and for to draw attention to plasma BPA levels. Blood sampling may raise some ethical concerns that blood sampling might be painful and frightening to the patients. However, firstly, the blood sampling was done by expert nurses with great attention and measures to minimize the risks and burdens. Secondly and most importantly, this clinical trial was approved by the Clinical Research Ethics Committee of the Faculty of Medicine before the study started, and the study was carried out on a group of as many patients as possible in limited numbers following the guidelines of the Helsinki Declaration. Thirdly, patients and/or parents who volunteered to participate in the study were informed of the research methodology, and informed consent was obtained from the patients/parents before the blood and saliva sampling. Finally, this study is not the first study carried out with blood sampling done without clinical need and only for research purpose. Several studies with similar design and purpose were found in the literature [38, 39].

Conclusions

Our findings showed that orthodontic appliances release BPA into the saliva, but the levels of BPA detected in saliva and plasma remain within acceptable limits. However, precautions are needed to minimize BPA exposure during orthodontic treatments.

Acknowledgements

The data of the fixed appliance group were presented as an oral presentation at the 17th International Digital Congress of the Turkish Orthodontic Society.

The authors would like to thank Professor Dr Hamdullah KILIC and other Eastern Anatolia High Technology Application and Research Centre staff, as well as Dr Osman Nuri Aslan, for their assistance in developing and validating the methods used in this study.

Conflict of interest

None.

Funding

This study was supported by the Atatürk University Scientific Research Projects Coordination Unit [TDH-2019-7003].

Data Availability

The data underlying this article will be shred on reasonable request to the corresponding author.

References