Are S-PRG Based Composites Able to Resist and Protect the Adjacent Enamel Against Erosive Wear?

This in-vitro study evaluated the resistance of S-PRG-based-composites against erosive wear and their protective effect on enamel adjacent to restoration. Bovine-enamel-blocks were randomized into 12 groups (n=10/group), according to the factors material and type of wear (erosion-e or erosion+abrasion-a): S-PRG-based-composite-Beautil II®(SPRGe/SPRGa); S-PRG-based bulk-ll-Beautil Bulk Restorative®(SPRGBFe/SPRGBFa); composite-Filtek Z350 XT®(RCe/RCa); bulk-ll-composite-Filtek Bulk Fill®(BFe and BFa); glass-ionomer cement-EQUIA Forte®(GICe/GICa); resin-modied glass-ionomer cement-Riva®(RMGICe/RMGICa). Standardized cavities were prepared in specimens and restored. Initial prole was performed on the material and on the adjacent enamel at distances of 100/200/300/600 and 700μm. Specimens were immersed in 0.5%citric-acid (2min-6x/day-during 5days), and abrasive challenge was performed using a toothbrushing-machine (1min-after erosive challenge). Final prole was obtained following initial. Data were analyzed by two-way ANOVA and Tukey-test (α<0.05). On erosion, the GICe and RMGICe groups presented greater loss of material compared to other groups; up to 300μm away from the restoration, GICe and SPRGBFe were able to promote less enamel loss than composite groups. For erosion+abrasion S-PRG-based groups showed intermediate material wear compared to GICs (higher wear) and composites (less wear); there was no difference of enamel wear adjacent to restorations among groups. It is concluded that S-PRG-based-composites are a good alternative for restorative treatment of erosive tooth wear.


Introduction
Erosive tooth wear (ETW) corresponds to a tooth tissue destroy due to chemical-mechanical process, in which extrinsic (dietary) and intrinsic (gastric) acids interacts with attrition and/or abrasion [1]. The prevalence of this alteration is high and seems to be increasing [2][3][4]. The ideal treatment for ETW is based on early diagnoses and implementation of non-operative management strategies, acting on the risk factors for its development over time [5,6]. When risk factors are not effectively controlled, the enamel, and eventually the dentine, is lost. According to the Radboud philosophy, even for patients who present severe tooth wear when there is no complaints, counselling and monitoring is the treatment of choice. Nevertheless, restorative treatment is recommended when there is loss of vertical dimension of occlusion, pain, and/or loss of esthetics. In these cases, minimally invasive and adhesive restorative strategies are indicated [7].
The longevity of restorative materials under erosive and abrasive challenges depends on durability of the material and durability of the interface between tooth substance and restoration [8,9,10]. In general, glass-ionomer cements (GIC) are more susceptible to wear than composites under chemical-mechanical challenges [11]. On the other hand, GICs are able to release uoride, which could enhance the acid resistance of the tooth tissue adjacent to restorations [12,13]. Therefore, there is no ideal material with potential bene ts as resistance and preventive effects.
Giomer's technology is based on Surface Pre-Reacted Glass-ionomer (S-PRG) llers that are synthesized by the reaction between uoro-boro-aluminosilicate glass and a polyacrylic acid solution [14]. This material has shown to release multiple ions including F − , Sr 2+ , Na + , BO 3 3− , Al 3+ , and SiO 3 2− [15,16]. The ions release promotes acid buffering action in the low pH lactic acid solution [17] and prevents demineralization around the material [18][19][20][21][22][23]. This ller is used in various dental materials including composite resins [24]. The buffering effect and the potential to prevent demineralization might be interesting characteristics of a material aiming to diminish the occurrence of enamel loss around restorations in patients with the presence of etiological factors for the development of erosive tooth wear. However, the evidence about the chemical mechanical resistance of this material or even its ability to prevent adjacent tooth tissue loss due to erosion is not clear yet. The factors under study were type of material (six levels) and wear condition (two levels). One hundred and twenty crowns of bovine teeth composed the sample, the specimens were randomly assigned into 12 groups (n = 10 per group): S-PRG-based composite resin-Beauti l II® (SPRGe/SPRGa);S-PRG-based bulkll composite resin-Beauti l Bulk Restorative® (SPRGBFe/SPRGBFa); composite resin-Filtek Z350 XT® (CRe/CRa); bulk-ll composite resin-Filtek Bulk Fill® (BFe and BFa); glass-ionomer cement-EQUIA Forte® (GICe/GICa); resin-modi ed glass-ionomer cement-Riva® (RMGICe/RMGICe). Group followed by "e" was subjected to erosion and by "a" to erosion + abrasion.
Circular cavities were prepared on each specimen and then restored according to manufacturers' instructions (Table 1). During 5 days, erosion was simulated in vitro (Groups e) by specimens' immersion in 0.5% citric acid for 2 min, 6X/day. For groups subjected to erosion associated with abrasion (Groups a), each erosive challenge was followed by abrasion using a toothbrush machine with uoridated dentifrice slurry (3:1) for 60 s. Between challenges, the blocks were kept in arti cial saliva. The response variable was material and enamel loss measured by pro lometry.

Specimen preparation
Approximately 140 bovine teeth were used in the present study. First the roots were separated from their crowns using a cutting machine (National Factory of Nevoni Single-phase Motors / Series 16.223, Type: TG1 / 3, São Paulo, SP) and a dia ex-F diamond disc (Wilcos do Brasil, Indústria e Comércio Ltda., Petrópolis, RJ). Crowns were individually placed in a cylindrical silicone mold (inner radius of 2.8 cm) and embedded in acrylic resin (Jet Ltd, Campo Limpo Paulista, SP, Brazil). Then, the specimens were ground at with water-cooled silicon carbide discs (600, and 1200 grades of silicon carbide paper; Buehler Ltd, Lake Bluff, IL, USA) and polished with felt disc wet by 1 µm diamond spray (1 µm; Buehler Ltd., Lake Bluff, IL, USA). The enamel specimens were ultrasonicated (Ultrasonic Cleaner Mod USC 750, Unique Ind. And Com. Ltda, São Paulo, SP, Brazil) in deionized water for 2 min between the polishing steps.
One pro le per specimen was performed to ascertain their planning and selecting 120 specimens. The pro le was obtained with contact pro lometer (Mahr Perthometer, Göttingen, Germany), coupled to a computer with a MarSurf XCR 20 contour software. Then, a random distribution was made using the Microsoft Excel program, using the "RANDOM" function of the mathematical category (10 specimens per group, 12 groups).
Circular cavities were prepared at the center of the crown using #2096 cylindrical diamond bur (KG Sorensen, São Paulo, SP, Brazil), with a diameter of 1.4 mm. A custom-made automatic device was used to standardize the depth of preparation (1.5 mm). For the composite with and without S-PRG llers, 37% phosphoric acid and adhesive system were applied. Each material was handled according to the manufacturers' instructions ( Table 1). All restorative materials were inserted in a single increment and covered with polyester strip, followed by a glass slab kept under pressure to expel excess material from cavity. After 7 days of storage at 37º C in 100% relative humidity, the restorations were polished with watercooled silicon carbide discs as described before.

Initial Pro lometric Analysis
Enamel blocks were marked with a scalpel blade (Embramac, Itapira, SP, Brazil) at two opposite sites with a distance of 0.3 mm from the margin of the restoration, resulting in two reference areas with 1.0 mm (at the border) and a test area with 2.0 mm, containing the restoration (at the center). Subsequently, initial surface pro les were obtained from the specimens using a pro lometer (MarSurf GD 25, Göttingen, Germany) and contour software (MarSurf XCR 20). To standardize their position, specimens were xed to a special holder and their locations were recorded allowing their exact replacement after the erosiveabrasive challenges. To analyze restorative material loss, two surface pro les were obtained through scanning from the reference to the test area, at the center of the restoration with a distance of 100 µm between them. On the other hand, to analyze enamel loss adjacent to restoration, ve pro les were obtained through scanning from the reference to the test area, at 100-200 -300-600 and 700 µm distant from the restoration margin.
Then, the previously demarcated reference areas were protected with cosmetic nail varnish (Colorama Maybelline -Ultra Dura, Cobra Cosméticos Ltda, São Paulo, SP, Brazil) to maintain its integrity during erosive and/or abrasive challenges.
In vitro erosive and abrasive challenges All the specimens were subjected to six erosive cycling daily for 5 days, by their immersion in 30 ml of 0.5% citric acid pH 2.5 for 2 minutes under agitation at a speed of 50 rpm and at a controlled temperature of 25 ° C. After erosion, half of the specimens (erosion groups) were rinsed with deionized water for 5 s and kept immersed in arti cial saliva (0.33g KH 2 PO 4 , 0.34g Na 2 HPO 4 , 1.27g KCl, 0.16g NaSCN, 0.58g NaCl, 0.17 g CaCl 2 , 0.16 g NH 4 Cl, 0.2 g urea, 0.03 g glucose, 0.002 g ascorbic acid, pH 7 (KLIMEK et al., 1982 modi ed without mucin) [25], for 2 hours, until the next cycling. At the end of each day of cycling, the specimens were also immersed in arti cial saliva, overnight (14 hours), under a temperature of 37°C.
For the groups in which erosion was associated with abrasion, toothbrush abrasion was performed after each erosive challenge, (6x / day for 5 days). Extra-soft toothbrushes (Colgate Twister®, Colgate Palmolive Industrial LTDA, S.B. Campo, SP, Brazil) were personalized for each specimen and xed parallel to the dental surface on a brushing machine (Dental Biopdi, São Carlos, SP, Brazil). The dentifrice slurry was prepared daily by diluting uoride dentifrice (Colgate triple action®, Colgate-Palmolive Industrial LTDA, SB Campo, SP, Brazil) in distilled water in the proportion 1: 3 (weight-volume ratio, according to ISO 14569 -1), and was always agitated before use. The slurry was automatic dropped on each specimen (≈ 3 ml). Each abrasive cycling consisted of brushing the specimens for 60 s with 100 reciprocal linear motion (back and forth) and force of 250 g, at temperature of 37.5°C. After abrasion, specimens were rinsed with deionized water for 5 s and kept immersed in arti cial saliva similarly to erosion groups.

Pro lometric Analysis
After the in vitro erosive and abrasive cycling, the cosmetic nail varnish was removed from the reference areas and the pro lometric analysis was performed at the same sites of the initial measurements. Baseline and nal pro les were perfectly matched, since the enamel specimens could be precisely repositioned in the pro lometer wells. The material and enamel loss were quantitatively determined using a speci c software (MarSurf XCR 20) by calculating the vertical difference (average depth of the surface) between baseline and nal surface pro les. The material loss corresponded to the average value of the two pro les made at the center of the material (in micrometers). On the other hand, the enamel loss was analyzed in each evaluated distance from the restoration margin.

Statistical analysis
The assumptions of equality of variances and normal distribution of errors were satis ed. Two-way ANOVA and Tukey test were applied to analyze materials loss as well as the enamel loss in each distance in relation to the restoration. The signi cance level adopted was 5% and the software used was Sigma Plot for Windows (version 11.0, Erkrath, Germany). Table 2 shows the results for material loss. A statistic difference was found for type of material (α = 0.0001), for condition (α = 0.0001) and their interaction was also signi cant (α = 0.0001). When considering the erosion condition, the composite groups (CRe, BFe) and the S-PRG-based groups (SPRGe, SPRGBFe) presented statistic similar material loss, which was less than that of the glass-ionomer cement groups (GICe, RMGICe). On the erosion + abrasion condition, the composite groups (CRa, BFa) presented less material loss, followed by S-PRG-based composite (SPRGa, SPRGBFa) and then by the glassionomer groups (GICa, RMGICa), with statistic difference among them. Considering each material, both glass-ionomers and both S-PRG-based composites showed higher material loss when erosion was associated with abrasion in comparison to erosion alone. The composite groups presented statistic similar material loss when subjected to erosion and erosion + abrasion conditions. Tables 3, 4, 5, 6 and 7 show the results for enamel loss at 100, 200, 300, 600 and 700 µm of distance from restoration margin, respectively. There was statistic difference for type of material, for condition and signi cant interaction on each distance. For erosion associated with abrasion condition, there was no statistic difference among materials in relation to enamel wear in all distances. When considering each material singly, the same behavior between conditions on studied distances were observed. All materials resulted in higher enamel loss when erosion was associated with abrasion compared to erosion alone, except CR (Z350 resin), which did not show any difference between the conditions (ERO and ERO + ABR). When considering only erosion, materials behavior was different on the studied enamel distances from the restoration margin.

Discussion
Many restorative materials have been tested regarding their resistance to erosive tooth wear [11,13,19,22,26,27]. When restorative treatment is indicated for patients who present erosive tooth wear, the ideal management is the association with measures that eliminate the causes of erosive wear [5][6][7][8]. However, this approach is not always feasible, implying that the use of bioactive materials capable of protecting the adjacent tooth structure is highly desirable. The results of the present study showed that the S-PRGbased composite materials were able to promote less enamel loss located at 100 µm distant from restoration margin, when compared to resin composite, therefore, the second formulated null hypothesis was rejected. The S-PRG-based bulk-ll composite resin (SPRGBFe-Beauti l bulk ll) promoted a reduction of 26% of enamel wear, which was similar to conventional glass-ionomer (GICe-EQUIA), with a 27% reduction. The reference for calculating the enamel loss reduction was composite resin group (Z350), as this material was the less effective material to protect against erosive wear, and the amount of enamel loss of this group was considered as 100%. This reduction was statistically signi cant up to 200 µm for SPRGBFe and on all distances for GICe. Therefore the protective effect on adjacent enamel against erosion, promoted by S-PRG-based composite groups was notable. Although restricted to enamel very close to the material, it might not be ignored, since the frequent acid exposure may affect the margins of the adhesive restorations, favoring the ow of uids through the adhesive interface [28].
Previous in vitro and in situ studies did not nd difference on the prevention of enamel loss adjacent to different types of materials (amalgam, composite resin and glass-ionomer cement) by means of pro lometry and hardness [19,27]. The explanation for the contradictory results, compared to the present study can be the pro le measurement method, materials composition and the erosive protocol. On these studies, the pro le reached around 1.5 mm distance from the material [19], probably not showing their potential protective effect, which was shown to be higher in the margins, in our study. On the other hand, some studies also found less enamel loss adjacent to glass-ionomer cements [12,13]. Rolim et al. 2012, evaluated the percentage of mineral loss on the surface around restorations under the use of highly uoridated dentifrices and showed that teeth restored with conventional GIC provided an additional protection against enamel erosion regardless of dentifrice used [12]. Similarly, Alghilan et al. investigated the effect of erosion on restorative materials and on adjacent enamel, simulating different salivary ow rates, and found that uoride-containing materials promoted less loss of enamel surface under erosive challenges [13]. Although there is no agreement on the protective ability of the glass-ionomer cement regarding adjacent enamel under erosive challenges [19,29], the present study found the best effect for CGIe. However, the modi ed glass-ionomer (RMGIe) resulted in similar enamel wear compared to composite resin group (CRe-Z350) and it was expected a better performance. In contrast, another study found that resin-modi ed glass ionomer cement (Fuji II LC) was the only material able to protect the enamel adjacent to the restorations against the erosive and erosive-abrasive challenges [30]. This result reinforces the knowledge that signi cant variation can exist among materials within the same category, depending on factors such as the nature, size of the ller particles [11] and the presence of resin. Resin modi ed glass ionomer cement in general exhibit a short-term uoride release which is lower and takes more time as compared with the conventional cements [31], this characteristic might have decreased the ability to protect the adjacent enamel against erosive challenge.
The protective effect found for S-PRG-based composite groups (SPRGe and SPRGBFe) was similar to glass-ionomer cement groups (GICe and RMGIe). We hypothesize that uoro-alumina-boro silicate glass ller (S-PRG) can release Silicon, Strontium, Aluminum, Boron, Sodium, and Fluoride ions, neutralizing the erosive acids and reducing enamel demineralization [29]. Nedeljkovic et al. 2016 found that Beauti l II was capable of increasing the pH of the solutions up to neutral (6 to 7) and attributed this ability to the S-PRG llers [32]. Strontium presents a synergistic effect when applied in association with Fluoride, with an advantageous of replacing hydroxyl and calcium ions in the apatite structure [33], this results in a more acid-resistant strontium and uoride-modi ed apatite that may be less soluble under acid exposure. However, in the study of Viana et al. 2020 -PRG-based composite (Beauti l II) was not able to protect adjacent enamel against erosion [30].
When abrasion was conducted after erosion, no protective effect was observed for the studied materials.
Probably the ions released by the S-PRG-based composite groups and the uoride by the glass-ionomer cement groups did not increase enamel mechanical resistance su ciently to reharden the softened enamel and reduce its loss due to abrasion. Even highly concentrated polyvalent metal uorides present limitation on the protective effect because the mechanical impact overcomes their chemical bene cial effect [34].
Regarding the material loss due to acid attack, the composite groups showed the lowest wear. This result is in accordance with previous studies [11,35] and can be explained mainly by the low acid-degradation of the composite matrix organic content [11,36]. In the composite groups, we did not nd statistic difference between wear conditions, since material wear of erosion alone was similar to erosion + abrasion. The mechanical resistance of the composite matrix in addition to bond stability between the ller and the matrix increases the abrasion resistance of the composite-based restorative materials [36].
Glass-ionomer cements showed the highest loss, which is in line with the literature [11,34]. Previous studies explain these results by the acid ability to dissolve the siliceous hydrogel layer, resulting in peripheral matrix dissolution and exposure the glass particles [26, 37,38]. Since the dissolution of this matrix causes a softening of the material, it is easily removed by toothbrush abrasion as we can see on the present results (Table 1) and in the study conducted by Yu et al. [26]. It was expected that conventional Glass-ionomer cements would present signi cant higher loss due to erosion than resinmodi ed ones; however, in the present study it was observed the opposite.
The S-PRG-based composite groups showed similar wear compared to the composite groups which is in line with previous study [30]. The rst formulated null hypothesis was rejected. This result can be attributed to the presence of bis-GMA and TEGDMA matrix, which is resistant to acid [24] and to high ller content. Contrary to the present study, Kooi et al. 2012, demonstrated that the Giomer's (resin composite with S-PRG llers) hardness and roughness are more affected than composite resins by citric acid, due to the uoroborosilicate glass llers greater susceptibility to degradation by weak acids than the zirconiasilicate ller of the conventional composite. [24]. For erosion + abrasion, S-PRG-based composite resins showed intermediate material wear compared to GIC (higher wear) and composite resins (less wear). Composites with nano llers particles exhibit homogeneous and less prominent particles on the surface, which are less susceptible to removal by mechanical forces [24]. Probably uorosilicate glass llers are more super cial and prominent to promote ions release, but this characteristic might also facilitate their removal.
Given the limitations of this in vitro study, S-PRG-based composite resins have showed to be a potential alternative for the restorative treatment of patient with erosive tooth wear, due to their higher resistance to erosive and/or abrasive wear than glass-ionomer cements, and their effective protection of enamel near to the restoration.

Conclusion
When the restoration was subjected solely to erosion challenge, S-PRG-based composite resins in similar performance to conventional composites, was less susceptible when compared to glass-ionomer cement.
In addition, they S-PRG -based composite resins promoted a reduction in enamel wear surround the margin of restoration, which was also observed when conventional glass-ionomer was used. None of the studied materials were able to protect enamel adjacent to restoration when erosion was associated with abrasion and this challenge resulted in intermediate loss of S-PRG -based composites compared to glassionomer cement, which showed the highest wear while composite resins were less prone to wear.