In Vitro Cytotoxicity of Dental Bleaching Products and Gingiva Barriers

1. Introduction

Aesthetic demands in dental treatments are increasing, including dental bleaching which can lighten the tooth color by minimally invasive means. Discolorations can be of extrinsic or intrinsic origin. The former is caused by substances like tea, coffee or spices while the latter is triggered by aged dentin, leading to decreased translucency, or trauma with pulpal bleeding into the dentin. Intrinsic discoloration is preferably treated with internal bleaching but requires previous endodontic treatment. However, vital teeth are treated through external bleaching which can be categorized into In-Office bleaching or At-Home bleaching [1]. Bleaching agents mediate oxidizing reactions resulting in a brighter shade of the tooth [2]. Primarily, carbamide peroxide (CP) and hydrogen peroxide (HP) are used as bleaching agents, with HP as the main component mediating the chemical reaction where free radicals are formed among other products. CP is a complex which breaks down into urea and HP in a relation of seven to three and is recommended for At-Home treatments in concentrations between 10% and 15% due to its lower cytotoxicity and gradual HP release [3]. On the contrary, In-Office treatments use bleaching agents with higher concentrations, e.g. HP between 25% to 40%, for faster and stronger results [4].

However, the free radicals that are part of the bleaching process can also negatively affect the tooth structure or the surrounding soft tissue, causing sensitivity, irritation or burns. Especially, the In-Office bleaching products may cause those effects due to their high concentrations of HP [4]. Therefore, appropriate safety measures are necessary to prevent or minimize side effects, such as gingiva barriers, that are usually light-cured resin-based materials, which are applied to the gum margins around the teeth before the bleaching material is applied [5].

As with any product that comes into contact with cells, bleaching agents should be biocompatible and cause little to no cell damage. According to ISO 10993-5, assessing cytotoxicity by multiparametric strategic means is considered superior to a single type of evaluation because not a single testing method offers universally reliable results [6]. Therefore, this paper combined quantitative evaluation, offering viability data, and qualitative analysis, capturing cellular changes not resulting in cell death. The combination of two different endpoints offers more detailed conclusions about cellular reactions as recommended according to ISO 10993-5 [7]. This study was performed using three main material groups: In-Office bleaching materials, At-Home bleaching materials and gingiva barriers.

The objective of our study was to contribute our multiparametric approach according to ISO standards to the existing research about cytotoxicity of bleaching agents and to study the cellular reactions to gingiva barriers which has not been studied yet.

This study aims at answering two main questions:

• Are there differences in cytotoxicity between the tested material groups in the multiparametric analysis?
• Are there differences in cytotoxicity within the tested material groups in the multiparametric analysis?

2. Materials and Methods

2.1. Materials and Cells

Eleven different materials were tested which can be divided into three types: 1) At-Home Bleaching, 2) In-Office Bleaching, 3) Gingiva barrier (Table 1). Explants of normal human gingival tissue acquired through surgical periodontal operation were obtained from the Department of Oral Surgery and Implant Dentistry (Carolinum, Goethe University Frankfurt). The obtained tissue was stored overnight in Hanks balanced salt solution (Gibco-Life Technologies Ltd., Paisley, Scotland) in a refrigerator at 4 °C to remove blood and granulation tissue. They were supplemented with five milliliters of bicarbonate and an antibiotic additive and stored at 4°C (Gibco-Life Technologies Ltd., Paisley, Scotland) in order to reach conditions that were as antibacterial as possible. To prepare the human tissue for the tests, it was cut with a sharp scalpel (No. 15, Aesculap, Tuttlingen, Germany) into small, uniform pieces measuring 1 mm³. Afterwards, the explants were transferred to surface-treated 50 cm3 polystyrene culture bottles and 50 mm diameter polystyrene petri dishes (Falcon, Becton and Dickinson, Heidelberg, Germany). The tissue was dried for one to two minutes at room temperature. Then, five milliliters of the culture medium (BM Eagle–Basal Medium) and 10% calf serum (both materials: Gibco, Paisley, Scotland, UK) were added to each Petri dish. Since oral human tissue cannot be obtained under sterile conditions, penicillin was applied to every culture medium in order to reduce the germ content. The Petri dishes were stored in a gas incubator (Nr, Heraeus, Hanau, Germany) at 37 °C in a 4.5% CO2 atmosphere with a humidity of 95%. The culture medium was initially renewed every five to seven days and then after two to three days. The first proliferation of epithelial and fibroblast cells around the explants occurred after 18 to 24 days, and after two to three days, a cell monolayer could be observed. For this investigation, a pure fibroblast culture was obtained via trypsinization. The bleaching materials were applied in the center of sterile bases of glass slides under anti-bacterial conditions. Gingiva barriers were applied with curing with the aid of an Elipar II curing light (ESPE, Seefeld, Germany). The curing times were set according to the respective instructions for use as recommended by the manufacturer. These prepared glass slides were weighed and placed in the center of each petri dish. Afterwards, the culture medium was applied to petri dishes and five milliliters of this eluate with a concentration of 0.005mg/ml was applied to fibroblasts in Petri dishes that had been 24 hours old. They were between the 7th and 18th passage. Then, the Petri dishes were incubated at 37 °C in a 4.5% CO2 atmosphere for 48 hours. Finally, the cell cultures were fixed with 98% pure ethanol and stained with Pappenheim’s panoptic stain. Each Petri dish is treated as independent as no component is linked to the others.

2.2. The Multiparametric Strategies Tested

2.2.1. Quantitative Evaluation

A total of 144 petri dishes, 12 per group and 12 for the control group were used for the quantitative assessment. Each sample was evaluated as being “viable”, “dead,” or “debris” after 48 hours using a cell counter (Cell-Counter CASY DT, OLS GmbH and Co KG, Bremen, Germany). The values were adjusted after establishing the initial settings for fibroblasts: viable: 12.8–100 µm; dead: 7.6–12.8 µm; debris: 3.3–7.7 µm. The values for “viable”, “dead,” and “debris” fibroblasts are based on manufacturer recommendations.

2.2.2. Qualitative Evaluation

The samples were examined at 100-fold magnification under a contrasting phase microscope (Leica, Bensheim, Germany) to identify physiological and pathological cellular changes. Predefined evaluation criteria included the general morphology, reactions and growth of the fibroblasts were evaluated, along with any vacuolization, detachment, and cell lysis that may have occurred. To document these cellular changes, photos of the cell cultures were taken. Two trained observers assessed the fibroblasts.

2.2.3. Reactivity Index

Based on the qualitative evaluation, the reactivity index was assessed according to ISO 10993-5 which defines grade two or greater as cytotoxic [7]. Grade zero is defined as no reactivity based on observed discrete intracytoplasmatic granules with no cell lysis or reduction in cell growth. A slight reactivity is defined at grade one when no more than 20% of the cells were round and loosely attached without intracytoplasmatic granules or changes in morphology and sporadically present lysed cells and only slight growth inhibition. Grade two is described as a mild reactivity with no more than 50% of round cells or cells devoid of intracytoplasmatic granules or extensive cell lysis and no more than 50% growth inhibition. Moderate reactivity is grade three where no more than 70% of the cell layers contained rounded cells or had been lysed and cell layers were not completely destroyed but more than 50% growth inhibition was observed. Grade four is defined as severe reactivity based on nearly complete or complete destruction of the cell layers.

2.3. Statistical Analysis

The results obtained for the viable cell counts were tested if they were normally distributed using the Kolmogorov-Smirnov test, which showed significant deviations from a normal distribution. Therefore, we decided to use the Kruskal–Wallis-test for statistical evaluation, which revealed significant differences in the results. By performing Conover–Iman post hoc pairwise comparisons with Bonferroni–Holm corrections (with an adjusted significance level of alpha ≤ 0.05), we determined which compared pairs of groups showed significant differences while considering the risk of error. Statistical analysis was performed with SPSS and Matlab (Version 20.0)

3. Results

3.1. Quantitative Evaluation

There were no statistically significant differences when comparing At-Home and In-Office bleaching products in regard to cell viability (Table 2 and Table 3). However, products of both groups showed statistically significant fewer viable cells compared to the gingiva barriers and the cell control. The fewest cell count for viable cells could be observed within the In-Office bleaching products compared to the cell control and the gingiva barriers.

Opalescence 15 (p = 0.03), 20 (p < 0.001) and Perfect Bleach 16 (p = 0.03) showed statistically significant fewer viable cells than the cell control. Further, for Opalescence 20 statistically significant differences could be observed compared to Opalescence 10 (p= 0.015) and Perfect Bleach 10 (p = 0.01). Contrary, no statistically significant differences could be observed for the gingiva barriers compared to each other or the cell control.

3.2. Qualitative Evaluation

Based on the morphological assessment of the cells, all tested In-Office bleaching products had moderate reactivity indexes (3.0) which is considered cytotoxic according to ISO 10993-5. However, the reactivity indexes of those were not statistically significantly higher compared to cell control (1.0). The At-Home bleaching products Opalescence 15 (3.70) and 20 (3.20) were evaluated as the products with the highest reactivity indexes and showed statistically significant differences compared to the cell control while the other At-Home bleaching products were observed as slight to mild reactive (0.70 – 1.8) and were per definition not cytotoxic [7]. In contrast, all gingiva barriers showed none to slight reactivity (1.0).

According to the observed results we can answer our former study questions:

3.

There are differences in cytotoxicity between the tested material groups in the multiparametric analysis.

4.

There are differences in cytotoxicity within the tested material groups in the multiparametric analysis.

4. Discussion

With high demand for teeth whitening, dental bleaching is becoming increasingly popular and numerous products have been introduced to the market. However, iatrogenic side effects can appear alongside the procedure, e.g. accidental contact with surrounding gingival tissue. Possible signs of cytotoxicity like burning, blanching or necrosis can be the consequence which have been studied before [4,16,17]. The higher the concentration of the bleaching agent and the longer the exposure time, the higher the cytotoxicity [4,18,19].

Our study has observed similar outcomes in regard to cytotoxicity of bleaching products on fibroblasts. For the In-Office bleaching products, fewer viable cells could be observed compared to the cell control. This is in accordance with other studies that revealed cytotoxicity of bleaching agents with high concentrations [20]. The cause of bleaching agents leading to cell death can be explained through the reactive oxygen species (ROS), caused by HP, which are able to penetrate through soft tissue where they cause oxidative stress. This reaction does not only cause apoptosis but can also cause mild necrosis through upregulation of inflammatory mediators, thus causing decreased cell viability [21]. Even though, CP is known to exhibit lower cytotoxicity due to its gradual release of HP, it might be plausible that higher a concentration like in our study (45% CP vs 32% HP) may have significantly more impact on cell viability than the type of the bleaching agent [22]. However, this remains speculation as it would be necessary to compare the effect of both bleaching agents with the same concentrations on cell viability which has not been performed in this study within the In-Office bleaching group.

Interestingly, there was no statistically significant differences regarding quantitative evaluation compared to the At-Home bleaching products which mostly ranged from 10% to 20% CP. According to previous studies, higher viable cells would have been expected in bleaching agents with lower concentrations which could not be confirmed in our study [4,20]. However, within the At-Home bleaching group, products with CP concentrations above 10% caused fewer viable cells than the control which is supported through other studies [4]. Nevertheless, Viva Style 16 with the same concentration and agent as Perfect Bleach 16 did not cause this effect. Possibly, different levels of cytotoxicity using the same concentration of bleaching agent may be due to the additives in the products though this is not closer analyzed in this study [20,23]. Apart from cell death, other cellular changes can also be caused by HP, such as damage to biological membranes and DNA [4,24]. Agents that include stabilizers, thickeners, dyes, preservatives or the viscosity of Viva Style 16 might be formulated differently from Perfect Bleach 16 since the former did not result in less cell viability in contrast to the latter. A possible explanation might be, that Perfect Bleach 16 causes cellular changes leading to irreparable damage, while the other might cause altered cell processes not leading up to cell death [25]. Possibly, cellular changes as observed in our qualitative evaluation may be hinting towards those mechanisms. However, since both products merely caused slight cellular changes in our observed results, it might be possible to suggest that repair mechanisms of the cell are able to reverse potential former damages. Based on the fact that studies have shown that soft tissue injuries and symptoms are reversible and full recovery occurs [4,26,27]. On the other hand, it might be possible that more severe cellular alterations occur after a longer time and thus extended observation periods may be necessary to see cytotoxic effects. A study has observed cytotoxic effects of CP bleaching products only after 48h and especially 72h, while there were no significant effects after 24h [23]. However, the potential explanations for the effects of bleaching agents were not tested in this study and thus cannot be verified.

Similarly to the viable cell count, Opalescence 15 and 20 with concentrations over 10% CP exhibited the highest reactivity index which can be categorized as cytotoxic. However, the At-Home bleaching products with 16% could not be described as cytotoxic according to the reactivity grading. Further both products, Perfect Bleach 16 and Viva Style 16, are from different manufacturers compared to the Opalescence products which may contribute to this theory. Studies found that bleaching agents with the same concentration of CP also resulted in different levels of cytotoxicity and it has been suggested that unknown agents within the bleaching products might influence the toxicity [20,23]. This study supports the possible theory that apart from the concentration, other factors like additives may have an influence on the cytotoxicity [23].

For future research and detailed understanding of the underlying mechanisms of cytotoxicity of bleaching products, further testing methods are necessary that examine the molecular pathways triggered within the cells. Further, comparing different observation periods may be beneficial for determining if products exert cytotoxicity over a short or long period of time and if potential effects change with time, giving more ground for assessing possible cellular recovery. Considering different effects of products with the same bleaching agent at the same concentration, it would be of interest to evaluate the cytotoxicity of other components besides the active agent.

Apart from that, it is interesting to note that there was no significant difference regarding cytotoxicity between Easydam and Opaldam. The former contains Bis-GMA and UDMA as monomers while the latter only contains UDMA as a monomer. While all monomers are known to exhibit cytotoxicity, existing research has shown that Bis-GMA has a higher cytotoxic potential than UDMA [28]. However, our results could not confirm this, possibly because of a high conversion rate after curing and thus minimal monomer release into the eluate [29]. For future studies, the monomer release into the medium would be advised to prove possible hypotheses.

It must also be stated that in vitro models do not accurately represent the in vivo situation and therefore cannot replace human studies. Factors like saliva flow, microbial environment and individual gum tissue differences are not reflected in an in vitro model. Further, the three-dimensional complex tissue is not accurately represented by an isolated monoculture [21]. Nevertheless, they offer valuable indicative information for the effects of materials on tissue [30].

5. Conclusions

Our study has found all In-Office bleaching products to exert cytotoxic effects on fibroblasts. Out of the tested At-Home bleaching products, only three with concentrations of 15%, 16% and 20% CP showed effects of cytotoxicity and reduced cell viability. Meanwhile, the In-Offices did not result in statistical significantly fewer viable cells compared to the At-Home bleaching products. Through observation, Opalescence products with concentrations over 10% CP caused significant cellular alterations while other products with similarly high concentrations did not or only partly cause cytotoxic effects. Gingiva barriers did not exert any cytotoxic alterations. Therefore, it seems that using them as protective measurements during in-Office bleaching procedures can be recommended. For At-Home bleaching, selecting products with CP concentrations up to 10% might be beneficial to decrease the risk of soft tissue sensitivity.