Effect of whitening mouthrinses on color change of stained resin composites

Objective: The present study compared the effect of whitening mouthrinses (WM) on the color change of stained resin composites (RC). Material and Methods: Cylindrical specimens (6mm-diameter and 1mm-thickness) were prepared with the following RC (n=60/group): Filtek Z350XT (Z350methacrylate-based), Admira Fusion (ADormocer-based), TPH3 (TPHmethacrylate-based), and Beautifil II (BIIgiomer/methacrylate-based). The initial color was assessed with reflectance spectrophotometer using CIE L*a*b* system. The specimens were immersed in staining broth during 14 days, submitted to color evaluation (ΔE1) and randomly allocated in 4 subgroups (n=15), according to WM adopted: Listerine Whitening (LW-2% hydrogen peroxide), Plax Whitening (PW-1.5% hydrogen peroxide), Bromelain/papain (BP-experimental solution), and Deionized water (DW-negative control). The whitening cycle consisted of RC immersion in WM for 1 min and in artificial saliva for 30 min, simulating 12 weeks, and final color assessment was performed (ΔE2). Color change data were analysed by ANOVA and Tukey’s tests (α=5%). Results: After staining, TPH showed the lowest ΔE1 values and Z350 showed the highest color change (p=0.001). The whitening effect promoted by LW was significantly higher than color alteration obtained with PW (ΔE2), and BII showed the highest color change values (ΔE2) after whitening cycle. Conclusion: LW exhibited the greatest whitening potential on stained RC, mainly with the Giomer (Beautifill II) and the Ormocer-based (Admira Fusion) materials. Bromelain/papain solution showed no whitening effect on stained RC.


INTRODUCTION
Esthetic demands in daily clinical practice are increasing over the years. Perfect smile involving whiter teeth and restorations with optical properties similar to the tooth structure are often requested by patients. In this sense, resin composites are widely used for direct restorations in both anterior and posterior teeth, and tooth whitening techniques are commonly performed to minimize tooth discoloration [1,2].
Traditional modalities of tooth whitening are performed with professional supervision (at-home and in-office techniques). Nevertheless, over-the-counter products such as whitening mouthrinses, painting on gels, toothpastes, and whitening strips are often available to consumers. The whitening mouthrinses are usually composed by low-concentrated hydrogen peroxide solutions and their ability to result in tooth color alteration has been demonstrated in previous studies [3,4].
As an alternative to peroxide, vegetative enzymes such as bromelain and papain have been investigated as stain removal agents [5,6]. Bromelain is extracted from pineapple and other immature fruits [7]. This substance contains different proteases such as peroxidases, phosphatases, catalases and other substances with phytomedical purposes [7,8]. Papain is a proteolytic enzyme of cysteine peptidases extracted from papaya plant. The benefit of papain is related to hydrolysis of amide substratebased with application in industrial sectors, as food industry and biomedical approaches [9]. These enzymes have been previously investigated in relation to tooth whitening [4].
Although the improvements of chemical and mechanical properties of resin composites are expected to result in color stability of restorations, the differences in materials composition, such as type of organic matrix, size and type of particles influence its staining potential [10,11]. The challenges caused by the exposition to saliva, stains, acids and alcoholic food / beverages contribute to the materials discoloration [2,[11][12][13][14]. It has been reported that 50% of direct nanohybrid composite restorations presented surface staining after 2 years [15].
The effect of conventional whitening treatments on tooth-colored restorations has been previously shown [16,17], contributing to reduce staining and favoring the color recovery of resin composites [18,19]. Nevertheless, the potential of color alteration of the restorations resulting from whitening mouthrinses has not been properly described, although these products are easily available for consumers.
Therefore, this study aimed to compare the effect of whitening mouthrinses on the color change of stained resin composites. The null hypothesis tested was that there was no difference among the effect of the whitening mouthrinses on the color alteration of the stained resin composites tested.

Sample size calculation
A power of 80% and significance level of 5% was adopted to determine the sample size. With a standard deviation set at 2 [20] and a minimal difference to be detected of 2.7 (acceptable color difference threshold using CIE L*a*b* system) [21], 14 samples per group were required, considering 4 treatment groups. Thus, 15 samples per group were used. The sample size calculation was performed using an online software (http://estatistica.bauru. usp.br/calculoamostral/index.php).

Specimen's preparation
The composition of the tested resin composites is described in Table I. Cylindrical specimens (6 mm diameter and 1 mm thickness) were prepared for each material (color A2) using a stainless steel matrix (n=60/material). The materials were inserted in a single increment and a mylar strip was placed on the resin surface to produce a smooth and standardized surface. They were light-cured (Radii Cal curing light, SDI, Victoria, Australia; 440-480nm; 1200 mW/cm 2 ) for 40 s according to manufacturer's instructions.
The specimens were attached to a metal holder and polished with sequential aluminum oxide abrasive papers (#1200, #2400, and #4000; FEPA-P, Struers, Ballerup, Denmark) mounted in a polishing machine (DP-10, Panambra, São Paulo, SP, Brazil) at 600 rpm for 60 s each under water cooling. The specimens were placed in an ultrasonic bath with deionized water for five minutes between each paper change and after polishing procedures to remove residues of abrasive grains.

Color assessment
The initial color assessment of each specimen was performed using a reflectance spectrophotometer with an integrating sphere (CM 2600d -Konica Minolta, Osaka, Japan), according to the CIE L*a*b* system [22]. The device was adjusted to use the D65 standard light source with 100% ultraviolet light (UV) and specular component included (SCI). The observer angle was set at 2º adjusted to a small reading area (SAV). The color of each sample was measured three times and averaged.
The coordinate values (L*, a*, b*) locates the color of an object in a three-dimensional (3D) color space. The L* axis represents the degree of lightness within a sample and ranges from 0 (black) to 100 (white); a* values correspond to color on the axis of red-green; and b* values correspond to yellow-blue axis.

Staining
The specimens were stored in individual containers with 2 ml of a staining broth solution (1.5 g of finely ground instant coffee; 1.5 g of finely ground instant black tea, 1.1 g of gastric mucin dissolved into 500 ml of water, 0.33 ml of FD&C Red 40, 0.33 ml of FD&C Yellow 5, 41.6 ml of red wine, 0.125g of methylparaben and 0.075 g of propylparaben) for 14 days (adapted from Wozniak et al., 1991) [23] under agitation. The pH of the solution was measured (pH=6.9), and it was renewed in the 7 th day of the experiment [24]. After staining, the specimens were washed with deionized water and a second color assessment was performed with the same specifications described previously. The stained specimens were stored in closed containers (100% relative humidity) preventing their dehydration during color measurement procedures.
The ΔL, Δa, and Δb values were calculated comparing the initial color and after staining assessments, and color alteration was determined (∆E 1 ) according to the following formula:
This cycle was repeated 84 times, which simulates 12 weeks (three months) of whitening mouthrinses use [3]. At the end of the cycle, the specimens were washed with deionized water and the final color assessment was performed (∆E 2 ), comparing the values obtained after whitening with the values obtained after staining.

Statistical analysis
Data were analyzed for normality by the Shapiro-Wilk test. Analysis of variance (ANOVA) parametric test followed by Tukey's test were used. One-way ANOVA was adopted for comparison among resin composites after staining. Aiming to compare the effect of the whitening mouthrinses on the different materials, two-way ANOVA was applied. The significance level used was set at 5%.

RESULTS
ΔL, Δa, Δb and ΔE 1 values after staining are expressed in Table III Table IV shows the results of the Tukey's test among the different stained materials and mouthrinses tested (Table IV).

DISCUSSION
The color stability of resin composites represents an important condition in esthetic restorations, and it depends on several factors [26,27]. The central factors are the composition and proportion of the organic matrix and inorganic content in the material, since they are related to the hydrophilic properties and consequently, staining potential of resin composites [28,29]. Then, the lower the volume of inorganic fillers, the higher the water sorption extent [10,11].
TEGDMA is a hydrophilic monomer contributing to water sorption in resin composites. When this monomer is combined with Bis-GMA, a higher water uptake is detected [23,28]. Since Filtek Z350XT and Beautifil II are composed by Bis-GMA and TEGDMA, this may explain why these materials presented the highest color change after staining (ΔE 1 = 8.23 and 7.52, respectively).
The susceptibility to staining may be also explained by the filler's characteristics. Although Filtek Z350XT has 78.5% filler content, this composite presents an aggregated zirconia/silica clusters and these structures may promote gaps among them, favoring staining [11,24]. On the other hand, TPH3 showed the lowest color alteration values (ΔE 1 = 3.07), despite of its lower filler content (60% wt). This material presents barium fillers, which has shown to restrict water/ pigments absorption on material surface due to the high reactivity of barium with water [28].
It must be highlighted that, despite the differences among the materials, all resin composites tested presented ΔE 1 values above 2.7 (acceptability threshold) [21], indicating t h a t t h e s e r e s t o r a t i o n s w o u l d r e q u e s t replacement, repolishing, or whitening for color adjustments [30].
In the present study, the specimens were artificially stained aiming to standardize resin composite discoloration in order to observe the whitening effect of mouthrinses. Then, the specimens were submitted to whitening cycle with the different products. The whitening efficacy was observed only for the LW and PW products. This whitening capacity has is attributed to the hydrogen peroxide component [4]. Thus, the null hypothesis was rejected.
Hydrogen peroxide is able to reduce colored molecules, breaking down the complex pigments structure. As a result of the oxidizing process, chromophores exhibit a decreasing in light absorption, resulting in the perception of a whiter substrate [31,32]. The LW group showed overall higher whitening efficacy in comparison to PW, which might be attributed to its higher concentration of hydrogen peroxide. Despite the low concentration of peroxide in these products, they were able to change the color of the stained composites, due to the longer period used with mouthrinses compared to professionally supervised whitening treatments.
Bromelain/papain and deionized water (control group) presented similar results, indicating no whitening effect provided by bromelain/papain-based experimental solution.
Previous in-vitro studies demonstrated that the incorporation of these enzymes in toothpastes resulted on tooth stain removal [5,6,33]. The mechanism of action of vegetative enzymes is related to their proteolytic activity in disrupting the protein portion of the acquired pellicle and, consequently, removing the dye pigments that are attached to the proteins [33]. Nevertheless, bromelain/papain efficacy is controversial, since no significant tooth color alteration was observed in a previous study testing these actives in solutions [4], as observed in our study with stained composites. Thus, its whitening efficacy may be a result of the association of the enzymatic action with the abrasive action of the toothpastes.
Some resin composites were more susceptible to the whitening effect of the mouthrinses, decreasing ΔE 2 values. Beautifil II presented the most evident whitening potential after mouthrinses application compared to the other composites tested. This resin is classified as a Giomer, an ion releasing material containing surface prereacted fluoro-boro-alumino-silicate glass fillers (S-PRG). Despite its high filler content (83.3%), hydrogen peroxide was able to effectively act on giomer material organic matrix [16]. In order to contribute to the ion releasing process, Beautifil II organic content presents hydrophilic monomers, contributing to diffusion of hydrogen peroxide into the organic matrix [34]. TPH3 showed the lower differences in ΔE 2 values after the whitening cycle. Possibly, this condition is related to the limited color change after staining.
The pure Ormocer-based resin composite (Organically Modified Ceramic), Admira X-tra, exhibited intermediate values after staining (Table III). This material is composed by a great amount of filler content (84%), which contributes to a lower water sorption, and consequently is less prone to staining, as observed in a previous study [24]. Additionally, Ormocer has shown higher values of microhardness and degree of conversion, probably due to numerous polymerizable units of ormocer molecule. This increases the number of chemical bonds between the methacrylate groups and contributes for maintenance of its superficial integrity after water storage [24,35,36].
Although there were different responses in whitening outcomes depending on the material, all resins exhibited some degree of color change when subjected to peroxide-based products. These mouthrinses resulted in whitening effect superior to the color change acceptability threshold (2.7) for AD and BII materials. Thus, such products might be considered as an alternative to maintenance of restoration color over time, mainly in individuals frequently exposed to colored food and beverages, since these colorants tend to stain restorations more intensely than the tooth [37]. Additionally, the whitening mouthrinses might contribute to potentially reducing restoration premature replacement due to esthetic concerns. Nevertheless, caution should be taken to extrapolate these results to clinical situation, since the in vitro nature of the study presume the exacerbation of the staining conditions, as well as the absence of salivary flow and clearance, diluting the mouthrinse. Additionally, the period of whitening treatment with rinses is long and this might affect patient's compliance.

CONCLUSIONS
The 2% hydrogen peroxide-based mouthrinse (Listerine Whitening) presented the greatest whitening potential on stained composites, mainly on the Giomer-based (Beautifill II) and the ormocer (Admira Fusion) materials. Bromelain/ papain solution showed no whitening effect on stained RC.