Evaluation of the remineralizing capacity of silver diamine fluoride on demineralized dentin under pH-cycling conditions

Abstract Objective (1) to determine the effects of the silver diamine fluoride (SDF) and sodium fluoride (NaF) in demineralized dentin exposed to an acid challenge by pH-cycling, (2) to evaluate the remineralizing capacity of SDF/NaF products based on the physicochemical and mechanical properties of the treated dentin surfaces. Methodology In total, 57 human molars were evaluated in different stages of the experimental period: sound dentin – negative control (Stage 1), demineralized dentin – positive control (Stage 2), and dentin treated with SDF/NaF products + pH-c (Stage 3). Several commercial products were used for the SDF treatment: Saforide, RivaStar, and Cariestop. The mineral composition and crystalline and morphological characteristics of the dentin samples from each experimental stage were evaluated by infrared spectroscopy (ATR-FTIR), X-ray diffraction, and electron microscopy (SEM-EDX) analytical techniques. Moreover, the mechanical response of the samples was analyzed by means of the three-point bending test. Statistics were estimated for ATR-FTIR variables by Wilcoxon test, while the mechanical data analyses were performed using Kruskal-Wallis and Mann Whitney U tests. Results Regarding the chemical composition, we observed a higher mineral/organic content in the SDF/NaF treated dentin + pH-c groups (Stage 3) than in the positive control groups (Saforide p=0.03; Cariestop p=0.008; RivaStar p=0.013; NaF p=0.04). The XRD results showed that the crystallite size of hydroxyapatite increased in the SDF/NaF treated dentin + pH-c groups (between +63% in RivaStar to +108% in Saforide), regarding the positive control. SEM images showed that after application of the SDF/NaF products a crystalline precipitate formed on the dentin surface and partially filled the dentin tubules. The flexural strength (MPa) values were higher in the dentin treated with SDF/NaF + pH-c (Stage 3) compared to the positive control groups (Saforide p=0.002; Cariestop p=0.04; RivaStar p=0.04; NaF p=0.02). Conclusions The application of SDF/NaF affected the physicochemical and mechanical properties of demineralized dentin. According to the results, the use of SFD/NaF had a remineralizing effect on the dentin surface even under acid challenge.


Introduction
Dental caries is a major health problem in most countries, where the disease affects 60 to 90% of children and most adults. 1 Although the prevalence of dental caries decreased over the past decade, it is still one of the most common diseases worldwide. 2 In fact, dental caries majorly impacts the global clinical and economic burden, thus, an improved approach for its prevention and therapy is essential. 3 Dental caries is a multifactorial, dynamic, and chronic disease characterized by demineralization and cavitation of susceptible dental hard tissues. 4 The process involved in the caries development is biofilm mediated and sugar driven, resulting in the phasic demineralization and remineralization of tooth structures. 5 At the inorganic level, the demineralization process during dental caries starts with the acidic dissolution of hydroxyapatite crystals, leading to the formation of cavities.
Dental caries in enamel and dentin can be remineralized and arrested at an early stage of the disease by using chemical products at clinical level. The minimally invasive approach to caries treatment should respect dental tissues and traumatize less the patients, avoiding surgery as long as possible and focusing on the preservation of natural tooth structure. 5,6 Preventive and non-restorative treatments, such as the use of fluoride agents, are widely employed in the treatment of dental caries. 7,8 For the application of fluorides, established protocols must be followed, with each fluoride in its own recommended concentration, frequency of use, and dosage schedule. 8 Moreover, the use of fluoride products, such as toothpastes and mouthwashes, is becoming a common complement in oral care.
Currently, the use of oral rinses containing 0.2% NaF may prevent caries. 9,10 On the other hand, silver diamine fluoride (SDF) is a colorless, odorless, and alkaline topical biomaterial. 7 The commercial market has different concentrations of SDF (i.e., 12%, 30%, and 38%) available. 8 However, the most common SDF concentration is 38%, which contains a high concentration of fluoride (equivalent to a relative concentration of 5%; 44.800 ppm), silver ions (25%; 253.900 ppm), 8% ammonia, and 62% water. 11,12 Since the effectiveness of the SDF with concentration of 38%, the product is used for management of dentin hypersensitivity, prevention of caries, pits, and fissures in adults and children, and arrest of root caries. 13 Moreover, previous studies showed that SDF can prevent and arrest the formation of new caries lesions in young children as well as in older adults. 14,15 Systematic reviews also concluded that SDF treatment achieves a 70 to 85% success rate in arresting pediatric caries. 16 The uncomplicated SDF application protocol has minimal potential risk and is painless, allowing to treat apprehensive young children, disadvantaged population, and elderly. 7 The mechanism by which SDF prevents and remineralizes dental caries is only partially understood. 17 The preferred assumption is that the fluoride ions in the SDF help to remineralize the carious dentin, while the silver ions provide antibacterial activity and reduce collagen degradation by inhibiting dentin collagenases. 18 Additionally, the high concentration of fluoride ions reacts with the dentin to form calcium fluoride, as silver reacts with chlorine or phosphate compounds of the dentin, leading to the formation of silver salts. 19 This reaction was previously described in the immediate effects of SDF on demineralized dentin, resulting in the precipitation of different crystalline phases and increased mineralization on the dentin surface after the products application. 20 Moreover, the use of SDF inhibits the proteolytic activity of the matrix metalloproteinases (MMPs) and prevents dentin collagen degradation from bacterial collagenase action. 13,17,21 Although previous in vitro studies demonstrated the properties of SDF compounds in the treatment of caries, the effects of these fluoride products on demineralized dentin under acid challenge (i.e., pH-c process) were scarcely analyzed. 22,23 The pH-cycling (pH-c) procedure, standardized by Marquezan, et al. 24 (2009), is one of the best methods to simulate affected dentin caries layers.
The procedure is easy to reproduce and cost-effective, showing high experimental control and sensitivity to variable treatment behavior. 25,26 In this way, pH-c is appropriate to evaluate the demineralization effects on in vitro caries simulation and the response of dental materials (e.g., SDF products) for caries treatment in clinical practice. Therefore, the present study aimed (1) to determine the effects of the SDF/NaF in demineralized dentin exposed to an acid challenge by pH-c and, (2) to evaluate the remineralizing capacity of these products based on the physicochemical and mechanical properties of the treated dentin surfaces.
For this purpose, we employed several analytical techniques to characterize the morphological, compositional, and crystalline properties of the dentin mineral surfaces at different experimental stages.
Furthermore, we also determined the mechanical strength of the treated dentin to evaluate its possible clinical performance. The hypotheses were: (1) the use of SDF/NaF would not affect the physicochemical properties of the demineralized dentin after the pH-c and (2) the use of SDF/NaF would not affect the morphological structural of the demineralized dentin after the pH-c.

Specimen preparation and experimental design
The study was approved by the local Ethics Committee on Human and Animal Research (Reference number #1020-2020). In total, 57 human permanent molar teeth (non-carious or morphological defects) were stored in 0.1% thymol solution at 4°C before preparation.
Tooth roots were separated from the crowns 2 mm below the cement-enamel junction using an Isomet 11/1180 low-speed saw (Buehler, Lake Bluff, IL, USA) with a 456CA diamond disk (Struers, Copenhagen, Denmark). Teeth slices (9 mm diameter x 1 mm thickness) were obtained by transversal sections of the mid-coronal for each tooth. Subsequently, four dentin beams (6x1x1 mm 3 ) were obtained from the teeth slices, resulting in 228 beam-shaped specimens. The enamel was removed to result in a dentin beam. The samples were rinsed and cleaned by ultrasonication for 30 min to remove the remaining debris. Then, the specimens were carefully examined under an optical microscope to confirm the absence of microcracks and dental imperfections for further experiments. Finally, the samples corresponding to the negative control (Stage 1) were stored at room temperature and 90% relative humidity.
The carious dentin lesions were created using a pH-c procedure modified by Marquezan, et al. 24 (2009).
All the specimens were immersed, for 8 hours, in 1 ml of a demineralizing solution containing 2.2 mM CaCl 2 , 2.2 mM NaH 2 PO 4 , and 50 mM acetic acid adjusted to a pH=4.8. Thereafter, samples were introduced, for 16 hours, in 1 ml of a remineralizing solution containing Subsequently, the demineralized dentin specimens were randomly divided into four experimental groups, according to the cariostatic remineralization products applied (commercial SDF): Saforide, Cariestop, RivaStar, and NaF treatments. Table 1 shows the product characteristics and application procedures.
After the treatment with the different products, the samples were resubjected to pH-c for 14 days to evaluate the remineralizing effect of the products under an acid challenge (i.e., in vitro caries process). In this study, the samples were analyzed during the different stages of the experimental period: sound dentinnegative control (Stage 1), demineralized dentinpositive control (Stage 2), and SDF/NaF treated dentin + pH-c (Stage 3), using several analytical techniques.  Sodium fluoride and deionized water 9 Apply with a brush for 3 min. Rinse  The second derivative method was used to resolve the peak estimations within each spectrum region. The degree of smoothing was adjusted to 10% (Savitzky-Golay algorithm) and a mixed Gaussian-Lorentzian function was used to adapt the profile of the peaks.
The following parameters were calculated to describe dentin compositional properties: (1) the relative amount of mineral to the organic matrix (PO 4 /amide I) as the ratio of the main phosphate (PO 4 stretch: 1030 cm -1 ) to the amide I area ratio (type I collagen: 1640 cm -1 ); (2) Crystallinity Index (CI) is the ratio between phosphate sub-bands areas at 1030 cm -1 (high crystalline apatite phosphates) to 1020 cm -1 (poorly crystalline apatite phosphates). 27

X-ray Diffraction (XRD) analyses
A convenience sample of three dentin beams was randomly selected and two-dimensional X-ray   ; phosphate band) after the pH-c process (positive control and treated dentin groups). Furthermore, the positive control and SDF treated groups (i.e., Saforide, Cariestop, and RivaStar) showed a higher relative amide II (1550 cm -1 ), as well as the broadening and decreased intensity at 1400 cm -1 of amide III (1240 cm -1 ), in the SDF treated groups regarding the negative control. On the other hand, the specimens treated with NaF product showed similar spectra as the positive control. Table 2 shows the compositional parameters related to the main inorganic/organic components, associated to the 1030/1640 absorption area (i.e., PO 4 /amide I ratio) and the CI. We observed a significant decrease

X-ray Diffraction (XRD)
We studied the identification of the crystalline phases precipitated during remineralizing treatments and the crystallinity properties of the dentin mineral by means of XRD analyses. Figure 3 shows the integrated 2Theta patterns (one-dimensional scanning) for each of the treatments and the control groups. The diffraction patterns show highly crystalline phases formed due to the remineralizing treatments on the dentin surfaces, corresponding to silver chloride   and silver iodide (iodoargyrite) for RivaStar and NaF crystalline phase. Table 3       3) (i.e., Ca>23.50% and P>8.67%, respectively), showing higher Ca/P ratios for the SDF treatment groups (above 2.63). Furthermore, all groups treated with SDF products also presented silver (Ag), although the products' distribution was heterogeneous on the dentin surfaces. Finally, iodine (I) was detected in RivaStar, as well as Na in the NaF treatment group.

SEM-EDX analyses
Three-point bending test Table 4

Discussion
The present study investigated the remineralizing capacity of commercial SDF/NaF products on demineralized dentin exposed to a cariogenic challenge. Our results indicate that the application of these products increased the mineral/organic content and the crystallite size of the hydroxyapatite crystals in the SDF/NaF treated dentin. We associated these compositional and microstructural alterations, as well as the formation of different crystalline phases on the dentin surfaces, with a partial improvement in the mechanical properties of the remineralized dentin. These products promote many effects in the dentin properties that resisted an experimental acid challenge (induced by pH-c) and, thus, they preserve the remineralizing capacity of the SDF/NaF and prevent demineralization of the dentin structure. In this way, we rejected the two hypotheses since the results showed differences between control (sound and demineralized dentin) and SDF/NaF treatment groups considering the different experimental stages.

Experimental Groups Mpa
Negative  One of the main limitations in the application of in vitro models in Cariology is their limited capability to reproduce some interindividual factors of the intraoral environment (e.g., plaque pH values, saliva, and microbial composition). However, these models have important advantages regarding the high level of scientific control and the consequent lower intrinsic variability and, thus, the smaller sample size required.
In vitro simulation of dental caries using a standardized model, such as the pH-c, is an accepted method in dental research due to its ease of technique and reproducibility. 26,28,29 The pH-cycling method is widely used for many studies on dentin caries, with slight modifications depending on the study objectives. 25,26 For example, pH-c models were employed to detect the dose and pH responses of fluoride dentifrices and their association with other anticaries treatments. 30 Previous studies have also described that pH-c models can simulate mineral alterations at the compositional, microstructural and morphological levels, which resemble the characteristics of natural caries. 25 In our study, samples were treated with different remineralizing products in previous demineralized dentin and subjected to a subsequent pH-c protocol to evaluate their potential remineralization capacity under acidic conditions (simulating caries formation). In the SEM images, we observed the regular morphology with a rough surface in the sound dentin (Stage 1) ( Figure 4A-B). 47 After pH-c, we detected a higher aperture of dentinal tubules and a smoother surface, which coincided with the characteristics of demineralized dentin (Stage 2) ( Figure 4C-D). 24 The application of the SDF products (i.e., Saforide, Cariestop, and RivaStar) resulted in the formation of different crystalline phases, heterogeneously distributed on the dentin surface and partially occluding the dentinal tubules ( Figure 4E-N).
Particularly in RivaStar group, the precipitation of silver particles increased occlusion of the dentin tubules.
These observations have been previously described in SDF treatment applications associated with areas of greatest dentin demineralization. 13 The EDX analysis Moreover, an increase in the Ca/P ratio after the application of SDF products suggested that occurred remineralization of the previously demineralized dentin. Similar remineralization processes using SDF applications have been reported, 23,50 SDF/KI fluoride concentrations inhibited demineralization and/or enhanced remineralization of artificial dentin lesions. These overall alterations in the morphological, compositional, and structural properties of the dentin due to remineralizing products can lead to changes in its mechanical response.
One criterion for assessing dental tissue remineralization is the evaluation of its mechanical properties. 45 The demineralization in a carious process reduces the mineral/organic content of the dental tissues and, consequently, the mechanical properties of the tooth may be impaired. 26,51 The inorganic phase provides strength, whereas the organic phase is responsible for the toughness of dentin structure. 26 Our results demonstrated that the experimental groups treated with SDF/NaF products (Stage 3) improved the mechanical properties of the demineralized dentin, even after the caries simulation by pH-c. Furthermore, a higher concentration of SDF generated a greater inhibitory effect on MMPs, 17 which may contribute to avoid collagen degradation, reinforcing the mechanical properties. Moreover, the improvement in the mechanical properties could be also attributed to the alteration in the mineral/organic content and its crystalline characteristics, confirmed with ATR-FTIR and XRD analyses. On the other hand, RivaStar group showed the lowest increase in the flexural strength values, suggesting that the combination of SDF and potassium iodide (KI) may affect properties of SDF. In fact, the presence of KI decreases the amount of silver compounds on the dentin surface, which negatively affects its mechanical response. 13,50 These mechanical properties may critically influence the success of the future restoration using these remineralizing products.
The application of SDF products is a simple, noninvasive, and affordable treatment for dental caries.
Moreover, SDF is also a potential agent in the nonrestorative control of dental caries. 14,52 The present study reveals how the alteration of the physicochemical properties of dentin treated with SDF/NaF improves the mechanical properties of the dentin surfaces under pH-cycling conditions, simulating a dental caries process. Thus, the remineralization capacity of these treatments, due to the potential reaction of their chemical compounds with the hydroxyapatite of the dentin mineral, 8 appeared to show a mechanical effect on demineralized dentin that favors the stability of the treatment under acidic conditions (i.e., cariogenic challenge). Based on these results, we observed that the Saforide reacted better than Cariestop and RivaStar, both presenting a similar fluoride high concentration between the SDF products. On the other hand, we used the NaF product at a very low fluoride concentration compared to the other fluoride sources.
Therefore, the effect of 0.2% sodium fluoride solution (~850 ppm fluoride) should be directly contrasted with the SDF products (~40.000 ppm fluoride).
In the present study, even at a very low fluoride concentration, this product significantly changed the hydroxyapatite crystalline structure. These findings are corroborated by other studies, which concluded that the use of 0.2% NaF can decrease caries. 9,10 These differences provide important information for clinicians on the use of the appropriate material for each patient.
However, this experimental study was insufficient to determine the impact of the in vivo antimicrobial activity attributed to these products. In future studies, it would be interesting to determine the long-term evaluation of these products in caries processes (i.e., progression under pH cycling conditions), as well as the aesthetic issue related to the application of these products.

Conclusions
The use of silver diamine fluoride and 0.2% sodium fluoride solution has a remineralizing effect on the dentin surface even under acid attack.
Despite the fluoride concentration, the application of these products causes the precipitation of different crystalline salts, which is associated with an increase in the mineral/organic content and an alteration of the crystalline characteristics of dentin, thus, ultimately increasing the mechanical properties of demineralized dentin. Within the silver diamine fluoride products, the crystalline properties (higher CI and larger crystallite size) in Saforide treatment seem to be related to a higher mechanical response after acid attack. Due to the limitations of this study, further research in settings containing all oral factors to evaluate its clinical application is essential.