Effect of finishing protocols and staining solutions on color stability of dental resin composites

Abstract Objective The objective of this study was to evaluate the effect of finishing protocol and exposure to staining solutions on color stability of dental resin composites. Materials and Methods A nanofill and microhybrid composite, three finishing protocols (mylar, Soflex disc, and white polishing stone) and four staining solutions (tea, red wine, khat extract—two concentrations, control—distilled water) were evaluated. A digital spectrophotometer was used for color change (ΔE) measurements employing the CIE‐Lab‐color system. Paired/independent‐sample t test and two‐way analysis of variance (ANOVA) followed by Tukey's honestly significant difference posthoc test were used for inferential statistics at α = .05. Results Soflex finish was associated with least staining and comparable color stability for the two materials in tea and red wine. In Khat 2, microhybrid composite had statistically significant better color stability than nanofill for Soflex finish (14 days t = 3.270, p = .011). For microhybrid composite, mylar resulted in highest mean ΔE, whereas Soflex recorded the least in all staining solutions. For nanofill composite, white stone resulted in highest mean ΔE, whereas Soflex demonstrated the least mean ΔE in all staining solutions, except red wine where mylar demonstrated the least mean ΔE. For mylar finish, nanofill demonstrated statistically significant better color stability than microhybrid in both red wine (14 days t = 4.902, p = .001) and Khat 1 (14 days t = 3.252, p = .012). For stone finish, microhybrid demonstrated statistically significant better color stability than nanofill in all staining solutions (14 days t ≥ 4.785, p ≤ .001). Two‐way ANOVA showed a statistically significant difference in mean ΔE between and within specimens (F = 42.658, p < .001). All staining solutions caused clinically unacceptable discoloration for mylar and white stone finish. For Soflex finish, red wine produced clinically unacceptable color difference beyond 48 h. Conclusion There was a difference in color stability of resin composites depending on filler type, further influenced by finishing protocol. Soflex disc finish results in better color stability than mylar and white stone in both microhybrid and nanofill composites. Clinical significance Esthetic dental restorations such as resin composites are routine in contemporary restorative practice. Color stability of composites may be influenced by surface finish, dependent on the filler type, and consumption of chromogenic substances such as khat. To prolong their service, selection of suitable finishing protocols is an important consideration.

Clinical significance: Esthetic dental restorations such as resin composites are routine in contemporary restorative practice. Color stability of composites may be influenced by surface finish, dependent on the filler type, and consumption of chromogenic substances such as khat. To prolong their service, selection of suitable finishing protocols is an important consideration.
composite materials, esthetic dentistry, restorative dentistry, surface chemistry/properties 1 | INTRODUCTION Dental resin composites are ceramic-reinforced polymer systems applied in direct and indirect tooth-colored restorations (Osiro et al., 2016). Since their introduction in the 1960s, they have become increasingly popular for both anterior and posterior restorations (Demarco et al., 2017;Sarkis, 2012). This is due to growing esthetic demands from patients and a paradigm shift in philosophies of operative dentistry that have resulted in transition from GV Blacks' principles of "extension for prevention" to current concepts of minimal intervention dentistry (Osiro et al., 2019).
Direct composite restorations have several advantages as follows: they are less technically demanding, cheaper than indirect restorations, do not cause wear to opposing dentition, and can be easily repaired (Kohli & Bhatia, 2015). Moreover, bonding mediated by dental adhesive systems favor conservative preparations and lightactivated versions enable operator-controlled working time (Ritter et al., 2017). However, their disadvantages include polymerization shrinkage stress resulting in microleakage, color instability, incomplete polymerization, and limited curing depth (Ritter et al., 2017). Nonetheless, current advances have attempted to address the inhomogeneous polymerization and limited depth of cure by employing strategies such as nanofiller technology, silorane instead of methacrylate resins and photoinitiators other than camphoroquinone (Gonçalves, 2018). Resin composites have an average clinical longevity of 6-10 years (Henry, 2009).
The failure behavior of composites differs in the anterior and posterior regions. Posterior composites fail primarily due to secondary caries and fracture (Opdam et al., 2014), whereas anterior composites commonly fail due to esthetic factors such as color alterations (Berber et al., 2013;Catelan et al., 2013;Fernando Demarco et al., 2015). Discoloration of resin composites may be intrinsic or extrinsic (Singh, 2014). Intrinsic discolouration is permanent (Malekipour et al., 2012) and is related to the materials' composition, that is, the matrix, filler type and amount, photoinitiator system, and percentage of remaining double carbon bonds as expounded by the degree of conversion (Catelan et al., 2013). On the other hand, extrinsic discoloration may be due to accumulation of plaque biofilm and related staining, low degree of polymerization, exposure to environmental factors including heat, water, food colorants, and ambient and ultraviolet light. Notably, color change is often a result of a combination of these factors. Finishing and polishing protocols of composites are also an important determinant of extrinsic discoloration (Ashok & Jayalakshmi, 2017).
Finishing removes residual surface imperfections following contouring by cutting or grinding, whereas polishing provides lustre on a material surface. This is essential for maintenance of oral health, function, and esthetics (Kumari et al., 2015). Finishing and polishing minimize extrinsic discoloration of restorations by preventing accumulation of plaque biofilm and staining agents (Samra et al., 2008;Schmitt et al., 2011). Typically, the technique employs a stepwise approach with methodical gradual use of finer instruments including diamond and carbide-finishing burs, abrasive-impregnated rigid points, impregnated rubber cups and points, aluminum oxide-coated abrasive discs, abrasive strips, and polishing pastes (Kumari et al., 2015). The aim is first to contour the restoration using diamond burs, carbide burs, or coarse abrasive-coated discs. This is followed by finishing with either fine or extra-fine diamond burs, carbide burs, white aluminum oxide stones, white Arkansas stones, or medium and fine abrasive coated discs. The final step is polishing to achieve an enamel-like lustre using fine and extra-fine polishing paste (aluminum oxide or diamond), abrasive coated discs, silicon carbide-impregnated brushes or diamond-impregnated rubber polishing discs, cups, or points. It is recommended that all the finishing and polishing instruments within a selected system are used in the proper sequence (Patel et al., 2016). Early studies on finishing of resin-based restorations showed that very smooth surfaces could be obtained when restorations were allowed to set in contact with mylar strips (Heath & Wilson, 1976). However, composites polymerized in contact with a mylar strip leave a resin-rich surface layer, the oxygen inhibition layer, which is easily abraded in the oral environment exposing rough inorganic fillers, tends to absorb more water and is more prone to staining (Schmitt et al., 2011;Uçtaşli et al., 2007). There is a tendency to consider the smooth surface achieved with the mylar strip a final finish (Patel et al., 2016), but it is associated with more color change than the Soflex disc finish (Kumari et al., 2015;Yew et al., 2013).
Several staining agents and their effects on color stability of resin composites have been evaluated, including red wine, tea, and khat.
Red wine has been associated with greater color change comparatively (Catelan et al., 2013;Kisumbi, 1998). The effect of tea may be worse when sugar is added (Guler et al., 2005) or chlorhexidine rinse is used concomitantly (Omata et al., 2006). Khat is considered a drug of abuse with less addictive potential than alcohol and tobacco (Aden et al., 2006), and its use has been linked to a range of health issues affecting the gastrointestinal, cardiovascular, and reproductive systems (Al-Motarreb et al., 2002). Oral conditions associated with khat include periodontal disease, dry mouth, dental caries, oral mucosal lesions, temporomandibular joint disorders, and oral cancer (Astatkie et al., 2015). Staining of teeth from khat chewing is caused by tannins together with the presence of khat-favoring chromogenic bacteria (Aden et al., 2006;Al-Maweri et al., 2018).
The influence of the filler type on finishing is an important variable when studying the effect of staining solutions and finishing protocols on the color stability of resin composites (Schmitt et al., 2011). Further, information is scarce on the staining effects of khat, yet it is consumed in large quantities by certain communities in Africa, Asia, and Europe (Aden et al., 2006;Al-Maweri et al., 2018;Al-Motarreb et al., 2002;Astatkie et al., 2015).
Therefore, the aim of this study was to evaluate the effect of three finishing protocols and exposure to four staining solutions on the color stability of dental resin composites, testing the hypothesis that there is no difference in the color stability of microhybrid and nanofill composites.

| Materials and sample size
Two dental resin composites were evaluated as follows: a nanofill, Filtek Z350 (3M ESPE, St Paul, MN, USA), and a microhybrid, Vit-lescence (Ultradent, Inc., South Jordan, UT, USA). The materials were subjected to three finishing and polishing protocols, namely Mylar strip (Maquira Industries, Brazil), Soflex polishing discs (3M ESPE Dental Products, Neuss, Germany), and White polishing stone (Prima Dental, Gloucester, UK) in four staining solutions, tea, red wine, and khat extract (diluted to two different concentrations), whereas distilled water was the control. Color stability was determined as specified in ISO 7496:2000 for color stability of dental materials, with five test specimens for each finishing protocol and staining solution for both materials. Thus, for each material, there were three finishing protocols subjected to four staining solutions and one control. Each group contained five samples, resulting in a total of 150 samples as summarized in Table 1.

| Specimen preparation
Specimen discs were prepared (n = 150, 75 Vit-l-escence and 75 Filtek Z350 XT) in perspex moulds 8 mm diameter × 2 mm depth. A single layer of composite was packed into a mould and light cured on both sides for 40s (Blueluxcer M385, Monitex Industrial Co. Ltd, Taiwan), including a segment of dental floss to suspend the specimens in the staining solutions and fasten a label. Before curing, the irradiance intensity of the unit was checked using a handheld analog radiometer (Apoza, New Taipei City, Taiwan) and determined to be 1000 mW/cm 2 .
2.1.2 | Finishing/polishing protocols Mylar finish (M) was achieved by curing with Mylar strips on both sides of the specimen while the other protocols had a Mylar finish on one side and the other side exposed. The exposed side was polished using either a Soflex disc or white stone. Coarse, medium, fine, and superfine grit sequence of Soflex discs (D) held on a slow-speed hand piece were used sequentially for 30 s each. After each polishing, the specimens were rinsed and air-dried for 10 s before the next step.
White polishing stones (S) were used for 2 min on a slow-speed hand piece with minimal pressure. After treatment, all the specimens were placed in distilled water at room temperature for 24 h, to allow for rehydration and completion of polymerization (Afzali et al., 2015;Mundim et al., 2010), after which baseline color measurements were recorded.

| Staining protocols
Red wine (Robertson Winery, South Africa) (R), alcohol content of 7.5% and pH 3.4 was used as is. Black tea (Ketepa, Kenya) (T), pH 5.5, was brewed by placing a tea bag in 250 ml of boiling water for 4 min with no sugar added and then allowed to cool. The khat extraction protocol was previously described (Dimba et al., 2003) and is a modification of the methanolic extraction protocol (Lee, 1995), excluding the alkaloid purification to minimize acid or basic residues in the extract. Briefly, 870 g of fresh khat shots were chopped into small pieces and covered in methanol, the mixture agitated in a sonicator at room temperature, and filtered through an 11 µm filter to produce

| Color measurements
A digital spectrophotometer (Vita Easyshade, Vita Zahnfabrik) based on the CIE-Lab-color (Commission International de l' Eclairage L*a*b color) system was used for all color measurements against a white background. Color measurements were taken after 6 h and 1, 2, 4, 7, 10, and 14 days. The mean value for the five specimens was calculated. Paired-sample and independent-sample t test was used to determine the mean color difference at baseline and at the end of staining between the two materials, whereas two-way analysis of variance (ANOVA) was used to determine the mean color difference within or between groups of finishing protocols and staining solutions followed by Tukey's honestly significant difference posthoc test at 95% confidence level.

| RESULTS
At the end of the staining period all specimens, except (D) finished micro-hybrid specimens (Vit-l-escence) in the K1 and K2 staining solutions, demonstrated a clinically unacceptable color difference (ΔE).
Total ΔE above 2.6 units is considered clinically perceptible, whereas ΔE of 5.5 is considered clinically unacceptable, represented in the figures by the black and white arrows, respectively.   Filtek and Vit-l-escence samples in red wine (A, E), tea (B, F), Khat 1 (C, G), and Khat 2 (D, H). Table 3 shows the results of the independentsample t tests for the same. Table 4 shows that ANOVA test elicited a statistically significant difference in mean color (ΔE) between and within the specimens (F = 42.658, p < .001). For

| DISCUSSION
Resin composite restorations are exposed to diverse stains and chemo-mechanical effects in the oral environment culminating in discolouration as demonstrated by the nanofill (Filtek) and microhybrid (Vit-l-escence) composite materials evaluated in this study.
Both materials depicted unacceptable discoloration with a mean ΔE > 5.5 units after 14 days, except Soflex disc-finished microhybrid specimens in khat. Based on the results, the null hypothesis was rejected. There was a significant difference in the color stability of nanofill and microhybrid dental resin composite restorative materials with different surface finishing treatments in different staining solutions.
In this study, the clinically perceptible ΔE was considered to be 2.6 units and the clinically unacceptable ΔE was 5.5 units, employing the previously determined color matching tolerances of dentists (Douglas et al., 2007). Distilled water was the control and did not produce a clinically perceptible ΔE in any of the specimens, which is consistent with observations from several other studies (Ertaş et al., 2006;Falkensammer et al., 2013;Guler et al., 2005), indicating that adsorption of water did not cause significant color change. The highest mean ΔE was observed in the Mylar-finished microhybrid composite in the diluted khat (K2), whereas the lowest was observed in the same solution and sample with a Soflex disc finish. The SDs for color difference were generally high for mylar and low for stone-and disc-finished specimens. This may be due to unpredictability of uniformity on the surface layers after mylar finish in comparison with the other two finishing protocols.
Overall, the disc finish resulted in the best color stability in both the microhybrid and the nanofill composites in all staining solutions over the duration of the study. It was only in red wine that the mylar finish resulted in less ΔE than the disc finish for the nanofill. Generally, the stone finish in tea and khat resulted in highest ΔE in the nanofill, whereas the mylar finish in khat resulted in the highest ΔE in the microhybrid. This is consistent with findings from other studies, which also reported better outcome with Soflex disc than mylar finish (Kumari et al., 2015;Schmitt et al., 2011).
Although surface roughness of the samples was not analyzed after finishing, the mylar-finished specimens were expected to produce the least surface roughness. A smooth finish can improve esthetics and longevity and the smooth finish achieved with the mylar strip is regarded as the gold standard for composites (Catelan et al., 2013). This assumes that the superficial layer contains more polymer matrix than fillers (Kumari et al., 2015). Although this layer is advantageous for a smooth surface, it becomes a disadvantage when its effect on ΔE is considered, due to its high affinity for water and stains, and may explain the high ΔE observed in the (M) finished samples and particularly the microhybrid composites. However, this layer is removed during the disc and stone finishes (Schmitt et al., 2011). Moreover, the disc was expected to produce a smoother finish compared to the stone, as was observed in both materials. The disc system incorporates a multistep polishing technique utilizing sequential discs with decreasing abrasiveness, which results in a smoother final finish compared to single-step materials such as stone (Schmitt et al., 2011;Turkun & Turkun, 2004). These findings are similar to those reported by several other investigators who have also associated mylar finish with poor color stability (Berber et al., 2013;Patel et al., 2004;Yew et al., 2013).
As the stone finish eliminates the resin-rich superficial layer, it was expected to result in better color stability than mylar. This was the case with the microhybrid but not the nanofill resin composite. As stated by the manufacturer, Filtek contains 63.3%, whereas Vit-l-escence contains  two materials. Red wine is known to cause significant ΔE in composites (Ertaş et al., 2006;Falkensammer et al., 2013;Kisumbi, 1998;Silva et al., 2014) and it has been proposed that alcohol causes softening of the resin matrix of the composite facilitating staining (Topcu et al., 2009). Several studies have also reported that tea can cause significant discoloration of dental composites due to adsorption of colorants in the form of tannins (Ghahramanloo, 2008;Malekipour et al., 2012;Omata et al., 2006

| CONCLUSION AND CLINICAL SIGNIFICANCE
The findings of this study may not be directly applied to the clinical situation because of the continuous staining procedure, absence of the effects of the complex oral environment, and habitual oral hygiene effects; nonetheless, they provide an indication of the effect of these staining solutions on the color stability of dental resin composites. Within the limitations of this study, we concluded that, overall, Soflex discs finish resulted in better color stability compared to White polishing stone and Mylar finish for both the nanofill and the microhybrid composites; therefore, clinicians may accomplish best results if they complete these restorations using the Soflex discs.
All staining solutions caused clinically unacceptable discolouration of all specimens finished with mylar and white stone. For Soflex finish, red wine produced clinically unacceptable color difference beyond 48 h; therefore, the effect of the staining solution was found to be dependent on the polishing protocol, which ideally should be determined by the type of filler system. Patients should be advised to consume potentially staining foods or beverages in moderation.
T A B L E 3 Independent-sample t tests of mean ΔE of M, Soflex D, and S finished V or F specimens in R, T, K1, and K2 at 14 days