Analysis of color stability and degree of conversion of different types of resin composites

Abstract Resin composites containing surface pre-reacted glass (S-PRG) have been introduced to reduce demineralization and improve remineralization of the tooth structure. However, water diffusion within the material is necessary for its action, which can impair its overall physicomechanical properties over time, including color stability. This study aimed to evaluate the color stability and related degree of conversion (DC) of four resin composites. Discs (6 x 4 mm, n = 5/group) of microhybrid (MH), nanofilled (NF), nanohybrid (NH), and S-PRG-based nanohybrid (S-PRG-NH) composites with two opacities (A2/A2E and A2O/A2D) were prepared. Color (CIELab and CIEDE2000) was evaluated with a spectrophotometer after aging in grape juice (2 x 10 min/10mL/7days). The DC was analyzed by using Fourier transform infrared spectroscopy before and after light-curing. Data were statistically analyzed by using two-way analysis of variance and post-hoc least significant difference tests (p<0.05). In the color stability analysis, the interaction between filler type and opacity was significant (CIELab, p = 0.0015; CIEDE2000, p = 0.0026). NH presented the highest color stability, which did not differ from that of MH. The greatest color alteration was observed for S-PRG-NH. S-PRG fillers also influenced DC (p < 0.05). The nanohybrid resin composite presented favorable overall performance, which is likely related to its more stable organic content. Notwithstanding the benefits of using S-PRG-based nanohybrid resins, mostly in aesthetic procedures, professionals should consider the susceptibility of such resins to color alteration, probably due to the water-based bioactive mechanism of action.


Introduction
Resin composites are often indicated to replace lost tooth structures because of their ability to mimic the shape and color of natural teeth, resulting in a nearly imperceptible restoration. 1 In addition, physical and chemical properties allow resin-based restoration to endure daily oral challenges and preserve its integrity, luster, and color stability over the years. 2However, there is no single ideal material for all clinical Declaration of Interests: The authors certify that they have no commercial or associative interest that represents a conflict of interest in connection with the manuscript.
][5] As established in the literature, the composition of materials, 6 filler size, and their distribution modulate the interaction of a resin composite with light 7 by having a direct impact on the material roughness. 3Nanosized composites have shown reduced surface roughness when compared to their macro and microsized counterparts. 8This is important because increased surface roughness may lead to increased staining, 6 particularly in patients with frequent intake of staining beverages, such as coffee, 9 wine, 10 and grape juice, 11 whose low pH can also harm the integrity of the surface. 12][15] The optical properties of resin composites have been a significant factor, as their interaction with light plays an important role in mimicking tooth structures.Therefore, their opacity and/or translucency has enhanced layering techniques, enabling a higher level of reproducibility of natural dental aspects.The level of translucency of these materials is also mainly regulated by their composition and configuration of inorganic fillers. 16Most resin composites on the market are available in different opacities generally referred to as dentin (opaque) and enamel (translucent) resins.The variation in composition, as well as the optical interaction with light and pigments, can influence the color stability of these materials.
Additionally, another major concern with resin composite restorations is the frequent occurrence of dental caries around restorations.8][19][20] These fillers consist of a pre-reacted fluorosilicate filler with a polyacrylic acid-treated surface layer.They combine core particles containing glass ionomer in a resin matrix. 14,18,209][20][21][22] In addition, because of the release of boron and fluoride, they have antibacterial properties, thereby reducing the risk of carious lesions. 18,20,21omposites containing S-PRG fillers have great structural strength and good physical, chemical, and mechanical properties, 18,20,23 and a beyond-acceptable degree of conversion (DC).However, owing to the presence of S-PRG fillers, a certain amount of water should be diffused within the material to allow ion release into the environment, 19,24,25 which could influence the optical properties of this resin composite over time, including color stability. 24,26][33] Considering that S-PRG fillers partially depend on the aqueous environment, it is still unclear how their presence can interfere in color appearance. 3,34,35iven that unsatisfactory esthetic appearance has been considered the main reason for the replacement of anterior restorations 36 and the inclusion of ion-releasing fillers requires some level of water diffusion within the material, 37 the optical properties of different composite resins should be further investigated.Therefore, this study aimed to evaluate the color stability and related degree of conversion (DC) of four resin composites (microhybrid, nanofilled, nanohybrid, and S-PRGbased nanohybrid) in two different opacities (A2/ A2E and A2O/A2D).The first null hypothesis stated that no difference in color stability would be detected among the types of resin composites evaluated, regarding their type and opacity.The second null hypothesis stated that there was no difference in DC among the tested resins.

Methodology Experimental design
This in vitro study analyzed two factors: a) resin composites classified into four levels according to the size and type of filler particles: microhybrid (MH), nanofilled (NF), nanohybrid (NH), and nanohybrid with S-PRG (S-PRG-NH); and b) opacity at two levels: translucent/enamel (T/E) and dentin/opaque (D/O).A2 color was set for all specimens.For color assessments, two response variables were used: color alteration measured by a spectrophotometer and calculated based on the CIELab (ΔE * ab ) and CIEDE2000 (ΔE 00 ) equations after aging in grape juice.For the DC, the absorbance of the materials before and after light-curing by Fourier transform infrared spectroscopy (FTIR) was the response variable.The experimental unit consisted of a specimen of resin composite.
The technical specifications of each material are displayed in Table 1.

Sample size calculation
In the color assessments, the effect size for the CIELab and CIEDE2000 equations was estimated to be 0.779 and 0.741, respectively, based on the findings from a pilot study.Consequently, the total sample size was estimated to be n = 4/group and n = 5/group.However, considering that the same specimens were used to calculate the color alteration using both equations, the final sample size used was n = 5/group.
For the DC analysis, the effect size was estimated to be 0.791; resulting in a total sample size of n = 4/ group.However, to account for potential losses, a sample size of n=6/group was selected to be used in this study.

Specimen preparation
Forty discs (6 mm in diameter and 2 mm in thickness) were prepared and randomized into eight groups (n = 5/group) by a blinded trained operator.The material was inserted into a Teflon mold in increments of 2 mm, covered with a polyester strip, and pressed against glass plates to prevent the formation of air bubbles and to remove excess material.The upper glass plate was removed, and the resin composite was light-cured using a light-curing unit (Radii-Cal LED, 1,000mW/cm 2 ; SDI, Bayswater, Victoria, Australia).The intensity was monitored with a radiometer (Demetron; Kerr, Middleton, WI, USA) for each of the five specimens.The bottom surfaces of the discs were marked with a scalpel blade, stored in deionized water for 24 hours, and protected from light at 37°C.
The top surface was polished with sequential aluminum oxide discs (Sof-lex, Pop-on, 3M ESPE, St. Paul, USA): medium (10 s), fine (10 s), and extra fine grit sizes (10 s).After polishing, the specimens were placed in an ultrasonic bath for 5 minutes for elimination of debris.
The tests were performed in accordance with the ISO guidelines 4049. 38

Cycling protocol
This study followed the protocol proposed by Svizero et al. 11 The artificial aging process was performed with two daily immersion cycles of 10 minutes in 10 mL of undiluted grape juice (Suco de Uva Integral UniSabor, Indústria do Sucos 4 Léguas, Caxias do Sul, RS, Brazil) for 7 days.The grape juice had a pH of 3.47 and was kept at room temperature (23 ± 2ºC).After each immersion, the specimens were washed and stored in deionized water at 37ºC.The juice and water were replaced in each cycle.After 7 days, the specimens were subjected to an ultrasonic bath with deionized water for 5 minutes, dried with absorbent paper, and their final color was measured. 11

Colorimetric assessment by CIELab and CIEDE2000 systems
Color was assessed on the top surface of the specimens using a spectrophotometer (Easy Shade Advance Vita; Vita Zahnfabrik, Bad Säckingen, Baden-Württemberg, Germany) on a flat matte white standardized acrylic background under standardized lighting.Three measurements for each specimen were performed by a single blinded operator, and their average was calculated.Color readings were conducted at two different time points: after polishing (at 24 hours; Δ baseline ) and after aging in grape juice (at 7 days; Δ final ).The spectrophotometer was calibrated for each specimen.
In CIELab (Equation 1) and CIEDE2000 (Equation 2), color systems used are based on three main parameters: L* refers to luminosity (L* = 0 = black; L* = 100 = white), a* indicates the chroma on the red-green axis (a* > 0 = red and a* < 0 = green), and b* the chroma on the yellow-blue axis (b* > 0 = yellow; b* < 0 = blue).CIEDE2000 sought to enhance the blue and gray color performance with a more specific equation, where ΔL', AC' and AH' represent luminosity, chroma, and hue, respectively.ΔR = RT (ΔC' x ΔH') refers to the interaction between chroma and hue in the blue region.SL, SC, and SH are weighting functions that adjust the total color difference in L*, a*, and b* coordinates.KL, KC, and KH are parametric factors that serve as correction terms for experimental conditions. 39

Degree of conversion
The DC analysis was measured using FTIR (IRPrestige-21, Shimadzu,Tokyo, Japan) associated with an attenuated total reflectance (ATR) device. 40orty-eight discs were prepared (n = 6/group).For the initial reading of the unpolymerized material, the resin composite was inserted into the same Teflon mold (6 mm × 2 mm) on the ATR crystal.
The readings were conducted in the absorption mode within the spectral range of 4000 to 650 cm -1 and included 32 scans at a resolution of 4 cm -1 .DC was calculated based on changes in the intensity of aliphatic (1636 cm -1 /1638 cm -1 ) and aromatic bonds (1608 cm -1 for NF and NH, and 1715 cm -1 for MH and S-PRG-NH), according to Equation 3.

Statistical analysis
Dat a were entered i nto Microsof t Excel spreadsheets (Excel 2016; Microsoft, Redmond, USA) and analyzed for normal distribution and homogeneity using Sigma Plot software (Systat Software, Inc., San Jose, USA).
For color alteration, the data were subjected to two-way analysis of variance (ANOVA) and posthoc least significant difference (LSD) tests.Twoway ANOVA and post-hoc LSD tests were applied for DC analysis.The significance level was set at 5% for all tests.

Colorimetric assessment by CIELab and CIEDE2000 systems
Both homogeneity (p = 0.164 for CIELab and p = 0.135 for CIEDE2000) and normality (p = 0.370 for CIELab and p = 0.656 for CIEDE2000) of the data were tested.Statistically significant differences were found for the type of fillers (CIELab and CIEDE2000, p = 0.0001) and an interaction between the type of filler and opacity (CIELab, p = 0.0015; CIEDE2000, p =0 .0026)was observed.No significant difference was found for the opacity factor (CIELab, p = 0.348; CIEDE2000, p = 0.0645).The mean ΔE values for the CIELab and CIEDE2000 systems are presented in Table 2. Table 3 shows the mean value of each analyzed parameter in a descriptive way.
Based on the CIELab equation, among the T/E resin composites, NH presented the lowest ΔE* ab , which did not differ from that of MH.The greatest ΔE* ab was observed for S-PRG-NH, which did not differ from NF; however, both differed from MH and NH.
For the D/O resin composites, all groups were different.The lowest ΔE* ab was also noted for NH, which differed from MH, NF, and S-PRG-NH.The highest ΔE* ab was observed for S-PRG-NH, which differed from all others.
Nonetheless, despite these results, all ΔE* ab values of both T/E and D/O composites were classified as clinically notable (> 3.46), except for NH, whose ΔE* ab was the only one classified as clinically perceptible (> 1.8).
Regarding the CIEDE2000 equation, favorable outcomes were also detected for T/E composites in the case of NH and MH, which did not differ between  According to the established thresholds, all ΔE 00 values of T/E were classified as clinically unacceptable (> 2.25), except for NH, whose ΔE* ab values were classified as clinically perceptible (> 1.3).For D/O opacity, the materials were classified as clinically unacceptable (> 2.25), except for MH and NH materials, which were within the acceptable range (< 1.3).

Degree of conversion
The homogeneity (p = 0.002) and normality (p = 0.109) of the data were assessed.Statistically significant differences were found between the evaluated materials, opacities, and their interaction (p < 0.0001).The DC values are listed in Table 4 and representative spectra of FTIR analyses before and after light-curing are presented in Figure.
In the case of T/E specimens, NF and NH presented higher DC compared to MH and S-PRG-NH (p < 0.05).Among O/D specimens, S-PRG-NH and MH showed a lower DC than did other materials with the same opacity (p < 0.05).Significant differences between the opacities were observed only for NF.For this material, the translucent version showed a higher DC than the opaque version (p < 0.05).

Discussion
Discoloration of restorative materials occurs for different reasons, such as DC, titratable acidity, food colorant absorption and penetration, 1,24,32 and the size and amount of inorganic filler. 40The tooth substrate and surface properties of the materials can also be included, as well as the type of polishing performed. 3,6,9n the present study, for both CIELab and CIEDE2000  equations, S-PRG-NH showed the highest values for color alteration, followed by NF, MH, and NH, which presented the lowest ΔE values.Regarding opacities, NF and S-PRG-NH showed difference in terms of color stability for CIELab, which was noted only for NF in the case of CIEDE2000.Therefore, the first null hypothesis was rejected.
In clinical practice, layering of different colors and opacities of resin composites is recommended in order to achieve excellent aesthetic outcomes that closely resemble natural teeth.The overall discoloration of this material can result from extrinsic factors (e.g., pigments from food and beverages) or intrinsic factors, such as the composition of the organic matrix, size and amount of the inorganic matrix, photoinitiators, and DC. 24,40The opacity and translucency of these materials can also be influenced by their composition, which is mainly determined by the difference in refractive index between the organic and inorganic matrix and by the amount and size of the filler. 40This difference between the opacities could influence their response to discoloration; however, most materials showed no difference in color change between the two opacities (T/E and D/O) in this study, except for NF (CIELab and CIEDE2000) and S-PRG-NH (CIEDE2000).
This alteration was evident for all evaluated parameters, particularly in the case of NF-T/E compared to NF-D/O, which explains the difference of ΔE between their opacities.Although all evaluated resins showed a decrease in luminosity (ΔL), this alteration was more evident in S-PRG-NH, which was the most affected parameter.The presence of S-PRG particles in resin composites based on this technology leads to a more whitish appearance, affecting the perception of luminosity of these materials.The greater opacity of D/O group also contributed to the fact that the decrease in ΔL values became more evident for this material.In addition, both resins showed an increase in hue (ΔH) values, in contrast to MH and NH, whose values decreased.In general, the yellow-blue axis (Δb) was affected, in line with the bluish coloration of the substance used to age the specimens.
The mechanism of action of S-PRG fillers depends on water diffusion, 37 which requires higher amounts of hydrophilic monomers and higher water sorption as compared to materials not subjected to this technology. 20,24,25Previous studies have shown a positive correlation between water sorption and staining, 24,25 which could explain the higher staining observed for S-PRG-based composites.Nonetheless, as water is absorbed, expansion and plasticization of the resin and hydrolysis of the silane bonds occur, which can generate microcracks and the release of residual monomers. 11This could also favor staining and compromise the aesthetic appearance of the restoration. 28,32Moreover, as silane bonds are degraded, the filler particles of the resin can start to detach from the organic matrix, contributing to the increase in their staining potential due to increased roughness and pigment retention. 28,32,37tudies have shown that the solubility of BisGMAbased composites increases when associated with triethylene glycol dimethacrylate (TEGDMA) and decreases with urethane dimethacrylate (UDMA). 24,29hese findings could explain the findings of the present study, in which greater staining was seen for BisGMA and TEGDMA-based composites (S-PRG-NH and NF), indicated mainly by a greater change in the ΔL* parameter, 29 corroborating the findings of this study.The high water sorption promoted by the association of BisGMA and TEGDMA, combined with the mechanism of S-PRG fillers, could explain the staining of the materials shown in this study.The manufacturers did not specify the dimethacrylates in the tested NH and MH resins, except for the UDMA diluent used in MH.The presence of UDMA in the organic matrix could explain the higher resistance of MH to staining, but the lack of this information precludes a more robust interpretation of the results.
In this case, limitations can also be extended to other properties, and the balance of the main properties needs to be considered for the appropriate recommendation of each material. 4he color stability of resin composites has also been strongly associated with their DC. 31In this study, S-PRG-NH and MH presented the lowest DC, while NH and NF materials showed the best percentages of monomer-to-polymer conversion.A higher opacity can also influence the passage of light through the material, thus decreasing DC and leading to greater susceptibility to discoloration. 31he impact of this property was only observed for NF, which showed lower values for its D/O version compared to T/E.Therefore, the second null hypothesis was also rejected.Incomplete light-curing generates a greater amount of residual monomers, which are easily degraded, resulting in greater susceptibility to pigmentation. 33ifferences in the organic matrix composition of methacrylate and the size, type, and volume of particles can affect the depth of light curing and scattering, and consequently, the DC. 30 Hence, the volume and size of fillers in the composition of the S-PRG-based resin composite used in this study could have influenced the DC results.Ilie and Fleming 34 compared different materials, including one with S-PRG fillers similar to this study, and the poor performance was attributed to the presence of the filler. 35This technology shows larger particles when compared with other technologies, and it is associated with more filler percentage, hindering light penetration into deep layers, thus decreasing the DC. 34S-PRG-NH presented a percentage of inorganic fillers of 83.3wt% against 73.0wt% of MH and 78.5wt% of NF.The percentage of fillers in the NH resin was not provided by the manufacturer.
Additionally, it is known that TEGDMA increases the DC, 35 which could have contributed to the higher values observed for NF.The similar performance of S-PRG-NH, which also contains TEGDMA monomer, can be limited by these fillers, as mentioned above.Nevertheless, the presence of TEGDMA can impair the mechanical properties 35 because the monomers used in the formulation of these composites are strongly related to the staining potential and DC of the material.Future studies addressing the physicochemical properties of these resins should be conducted.
Therefore, based on the analyses in the present study, MH and NH could be used as an excellent aesthetic treatment alternative.In contrast, S-PRG-NH should be indicated in specific situations.Gordan et al. evidenced optimal performance of posterior restorations using Beautifill II. 19In particular, in Class II situations, this 13-year follow-up proved its clinical effectiveness.In a recent study, Toz-Akalin et al. conducted a 2-year follow-up and demonstrated how this material containing low-shrinkage organic monomers and reduced amount of S-PRG can yield more interesting results under service conditions. 5herefore, as reported in other studies, clinical circumstances, such as daily brushing 10 and polishing protocols, 12 may also play important roles in the longevity of restorative materials.Future studies should investigate these parameters, given that repolishing can re-establish the original color of the restorative material and achieve clinically acceptable levels of aesthetic longevity. 12n equilibrium between the in vitro and in vivo performance of the materials promotes robust interpretation, aiding professionals to choose materials and techniques in different circumstances and assuring more precise benefits to the patients.

Conclusions
Despite the limitations of this study, it can be concluded that the nanohybrid resin composite presented favorable overall performance, which is likely related to its more stable organic content.However, the use of S-PRG-based nanohybrid resins in aesthetic procedures should take into account their greater susceptibility to color alteration, probably due to the water-based bioactive mechanism of action.The opacity of the material may also affect the color stability and the DC of some materials.

Figure .
Figure.Representative FTIR spectra of a resin composite before and after light-curing.

Table 1 .
Classification of resin composite materials based on type of particles, commercial name, composition, opacity, and light-curing time.

Table 2 .
Means and standard deviations of ΔE values of resin composites analyzed by the CIELab and CIEDE2000 equations.

Table 3 .
Descriptive means of color coordinate values for the analyzed resin composites..The more perceptible changes were attributed to S-PRG-NH, which differed from those of the other materials.NF presented intermediate values, which differed from all groups.Among the D/O composites for CIEDE2000, a performance similar to that of CIELab was verified.NH presented the lowest ΔE 00 , which differed from those of the other materials.MH and NF presented intermediate values of ΔE 00 , differing from each other and from both NH and S-PRG-NH.The S-PRG-NH material showed the most perceptible results and differed from all materials.

Table 4 .
Degree of conversion (%) and standard deviation of the tested groups.
N=6/group.T/ E: translucent/ enamel; D/O: dentin/ opaque.Lowercase letters indicate statistical difference between type of materials and opacities.