Efficacy of Stannous Ions on Enamel Demineralization under Normal and Hyposalivatory Conditions

Tooth erosion is recognized as a multifactorial condition in which salivary flow rate and saliva components might impact the erosion initiation and progression and can be considered as predisposing factors [Järvinen et al., 1991; Meurman and ten Cate, 1996; Piangpach et al., 2009; Ramsay et al., 2015]. It seems to be biologically plausible that a reduced saliva flow rate might be associated with an increased risk for erosion [Jensdottir et al., 2013; Lieshout and Bots, 2014], since several protective functions of saliva, such as the neutralizing potential or the potential of pellicle formation might depend on salivary flow rate [Tenovuo, 1997; Hara and Zero, 2014]. However, the information in the literature on the relevance of the saliva flow rate in this context is contradictory. Whilst some studies found a relation to salivary flow and buffering capacity [O'Sullivan and Curzon, 2000], other studies reported no relation between erosion and any basic saliva parameter [Järvinen et al., 1991; Meurman and ten Cate, 1996; Johansson et al., 2002; Schlueter et al., 2012]. Even if the impact of changes in saliva flow rate on erosion initiation or progression is still under debate, it could be that changes in saliva flow impact the efficacy of oral health care products by changing interactions between tooth surface, protein coverage of tooth surface (pellicle) and active agents.

To maintain oral health, commonly, sodium fluoride (NaF)-containing toothpastes are recommended; however, their effects to prevent and treat tooth erosion are limited [Vieira et al., 2005; Huysmans et al., 2011; West et al., 2015; Hove et al., 2015]. Other fluorides turned out to be more effective. The best results can be achieved by the application of polyvalent metal cations, such as stannous ions containing fluoride preparations [Ganss et al., 2011, 2012; Hove et al., 2014; Algarni et al., 2015]. A further increase in efficacy can be achieved if novel technologies, such as the addition of biopolymers like chitosan to F/Sn preparations, are used [Lussi and Carvalho, 2015] both under in vitro [Ganss et al., 2012; Carvalho and Lussi, 2014; Pini et al., 2016] and under in situ conditions [Schlueter et al., 2013, 2014]. Nevertheless, even if a good effect of the mentioned preparations has been shown under in situ conditions with healthy volunteers showing a normal saliva flow rate, nothing is known about the impact of saliva flow rate reduction (hyposalivatory conditions) on the efficacy of the erosion-inhibiting agents stannous ions, fluoride and chitosan.

Therefore, the aim of this in situ pilot study was to assess the protective effect of different experimental toothpastes containing F/Sn in comparison to experimental NaF or placebo toothpastes against initial enamel erosion in participants with normal and low stimulated salivary flow rates. Moreover, the interaction between a salivary parameter (flow rate) and active ingredients of toothpastes were statistically evaluated.

Materials and Methods

Experimental Design

The study was planned and conducted as a single-center, crossover, randomized and double-blinded in situ study with 2 additional in vitro procedures (see also flow chart, Fig. 1). After preparation specimens underwent an in vitro erosive demineralization; next, the participants wore specimens in situ. Within this period, the specimens were treated intraorally with the test toothpastes and were exposed to saliva. After this phase a second in vitro phase was performed. Analyses of specimens were made at baseline (BL), after the first in vitro demineralization phase (D1), after the in situ phase and after the second in vitro demineralization phase (D2).

Open multimedia modal

Preparation of Specimens

A total of 320 enamel specimens (4 × 4 mm) from bovine incisors and with surface microhardness between 310 and 360 kg/mm² were used for quantitative analysis. The bovine incisors were previously stored at least for 1 month in 0.1% thymol solution [Ingram et al., 1997; Amaechi et al., 1998], and specimens were obtained using a precision saw (Isomet 1000; Buehler, IL, USA) and a diamond disk (Buehler). The enamel surface was planed and flattened using silicon carbide papers (until #4000), felts, and diamond paste (1 µm) (all from Buehler) in a polishing machine, under water cooling (Arotec, São Paulo, SP, Brazil). Between each polishing step, the blocks were sonicated to remove debris in distilled water (15 min). The specimens were randomly assigned to the 2 experimental groups (normal and low salivary flow rate), and were further divided into 4 subgroups (toothpastes) after the first in vitro phase (D1). Additional specimens were included to be qualitatively analyzed by scanning electron microscopy (SEM). All specimens were stored in 100% humidity until use.

In vitro Erosive Challenge

The erosive agent was a citric acid solution (1%, pH 3.5, citric acid monohydrate >99.5% PA; Merck, Darmstadt, Germany) freshly prepared before each treatment. The specimens were fixed in a holder and immersed in 40 mL of the citric acid for 20 min at room temperature [Hara et al., 2009] and under agitation (100 rpm). After this treatment (demineralization 1, D1), the specimens were gently rinsed with tap water for 2 min [Hara et al., 2009]. This erosive treatment had been previously determined in a pilot study to cause a reduction of enamel surface microhardness by around 30%. After the in situ phase (in situ), a second demineralization (D2) was performed in the same way as D1.

In situ Phase

Before in situ procedures in each experimental phase, stimulated saliva was collected from each participant, resulting in a total of 4 collections per participants during participation in the study. They were instructed to chew unflavored parafilm (Parafilm M, Pechiney Plastic Packaging Company, Chicago, IL, USA) and dispense the saliva in a falcon tube (15 mL) until the collection of 5 mL of saliva; the time until collection was measured. As a longer exposure to the air could impair the analysis of the salivary parameters, the falcon tubes used for the patients were given to them inside a cup with ice and the patients were instructed to keep the tubes closed and to open them just at the moment to dispense saliva. Right after the collection, the tubes were put closed in a box with ice, and pH and buffering capacity of stimulated saliva were analyzed. The collection of saliva was performed always at the same time at least with 1 h of fasting and after oral hygiene.

At each in situ phase, the participants wore custom-made acrylic palatal appliances. Four specimens were mounted somewhat below the appliance surface in order to avoid any contact between the specimens' surface and the tongue, however, not too deep in order to allow sufficient brushing. After the collection of saliva, participants were instructed to wear their palatal appliance for 5 min to allow the specimens to equilibrate with oral fluids. Afterwards, the participants received one soft-bristle toothbrush (Curaprox-Curaden Swiss GmbH, Kriens, Switzerland) loaded with the assigned toothpaste (1.5 g per phase and participant). The participants were instructed to brush the buccal surface of their teeth for 25 s without contact to the specimens, to create a mixture of toothpaste/saliva (slurry). After this, they remained with the appliances in position for 2 min to enable the contact of the specimens with the slurry. Then, the appliance was removed, and 1 researcher (N.I.P.P.) brushed the specimen's surfaces with the same toothbrush used by the participant for 20 s (1 stroke/s) [Stenhagen et al., 2014]. The brushing procedure was always performed by the same researcher. The researcher was previously calibrated, in particular regarding load, brushing velocity, and duration. Velocity and duration were trained by using a stop watch, load was trained by using the respective load on an electronic scale. A new toothbrush was used per participant and experimental phase. After brushing, the participants reinserted the appliance, gently rinsed the mouth with 15 mL of water for 10 s and spit out. Appliances were worn intraorally for 4 h [Hara et al., 2009]. After this period, the specimens were removed and stored at 100% humidity in individual vials at 37°C until surface microhardness measurement. The same procedures were repeated in the subsequent phases, changing the toothpaste used according to the crossover design and randomization list.

During wearing appliances participants were instructed not to eat, they were only allowed to drink water. Before starting the study and between each phase, a lead-in period of 2 days and washout period of 5 days were planned. During the washout, the participants were instructed to use standard nonfluoridated toothpaste.

Toothpastes

Four experimental toothpastes were tested, with the same composition except for the active agent. All experimental toothpastes were formulated using the same basic composition (Drogal Pharmaceutics, Piracicaba, SP, Brazil): glycerin, silica, carboxymethyl cellulose, water, methyl paraben, saccharine, titanium dioxide, sodium lauryl sulfate, mint oil. Toothpastes were always freshly prepared immediately before each experimental phase by adding the respective amount of active agents to the basic formula: placebo, no fluoride, no stannous ions (pH 6.9); sodium fluoride (NaF: 1,450 ppm F^-, pH 7.0); stannous fluoride and sodium fluoride (SnF/NaF; SnF₂: 1,100 ppm F^-, 1,090 ppm Sn²; NaF: 350 ppm F^-, pH 4.3) and the combination of sodium fluoride, stannous chloride and chitosan (NaF/Sn/Ch, NaF: 1,450 ppm F^-, SnCl₂: 3,500 ppm Sn²⁺, 0.5% Ch, pH 5.0). All compounds were purchased from Sigma-Aldrich (Saint Louis, MO, USA).

Results

All volunteers completed the study, and none of them reported adverse effects due to the toothpastes or procedures applied. Salivary flow rate of each participant as well as pH and buffering capacity values are given in Table 1. There was a significant difference between both groups for flow rate but not for the other parameters.

Open multimedia modal

Figure 2 shows the SMH values for the different groups and time points tested. No significant differences were found between groups regarding initial SMH (BL). These values were used as the basis for the calculation of the parameters under evaluation: %SMH, %RHL, and %RER.

Open multimedia modal

Table 2 presents the calculated values after D1 (%SMH_initial), after in situ phase (%SMH_total) as well as %RHL and %RER.

Open multimedia modal

Regarding normal flow rate, best performance was found for SnF/NaF for %SMH_total, %RHL, and %RER, followed by NaF/Sn/Ch, NaF and placebo, with statistical difference between each of them (p < 0.05). In case of low flow, SnF/NaF and NaF/Sn/Ch presented the best result for all parameters, without significant differences between both Sn preparations. NaF and placebo were comparable with no statistically significant difference between them. Comparing the flow rate conditions (normal and low), differences for %SMH_total and for %RHL after application of NaF and for %RER in case of all preparations were found (p = 0.0125), with a better performance under normal salivary flow conditions. The resistance was distinctly lower in case of low flow rate; however, the Sn-containing preparations were able to compensate this saliva effect.

Factorial ANOVA showed an impact of saliva flow on study outcome for all calculated values (%SMH_initial, %RHL, %RER, all p < 0.001; %SMH_total, p < 0.01); an impact of toothpaste type was found for %SMH_total, %RHL, %RER (all p < 0.001). An effect of interaction between salivary flow rate and toothpaste on study outcome was shown for %SMH_total (p < 0.01) and for %RHL (p < 0.05), but not for %SMH_initial and %RER.

Representative SEM images are shown in Figure 3. Very similar patterns were observed for placebo and NaF, irrespective of salivary flow. The etching pattern is more evident for these groups in case of low salivary flow. In general, it is possible to recognize some etching pattern with visibility of the orientation and location of the prisms in all groups. However, for SnF/NaF and NaF/Sn/Ch these characteristics were clearly less evident, mainly under normal salivary flow.

Open multimedia modal

Discussion

The present study aimed to assess the effect of toothpastes containing different active agents on SMH in enamel after a single erosive challenge in a short-term in situ application and ex vivo abrasion and erosion model. Most studies investigated the in situ effect in persons with a normal saliva flow rate [Magalhães et al., 2008; Hara et al., 2009; Wiegand et al., 2010; Schlueter et al., 2013; Hove et al., 2014; Stenhagen et al., 2014], showing the effect under more or less healthy conditions. The novelty of the present study is, however, to investigate the toothpastes' effect in persons with an extremely low flow rate (xerostomia). In persons, who were treated with ionizing radiation in the head-neck area, the salivary glands, which are in the field of radiation, are often injured and therefore the saliva flow rate is severely impaired [Jendsottir et al., 2013; Lieshout and Bots, 2014]. As resting saliva is often in this group virtually not present anymore [Andrews and Griffiths, 2001; Jhan et al., 2008], it was decided to classify according only to the stimulated saliva in order to homogenize this group. As the flow rate was so low in the present study, it was not possible to collect saliva in the appropriate time. Therefore, the saliva was collected according to volume, and the needed time was measured. This increased the blinding of the analysis since an allocation to one group by volume was excluded.

The recruitment of persons with low flow rate is difficult, and motivating them for attendance in a study is often complicated. Therefore, it was planned to use a very short-term in situ model (4 h) [Hara et al., 2009]. The specimens were pre-eroded in order to shorten the in situ period, since it is well known that antierosive agents react much more quickly with erosively altered surfaces than with sound or polished ones [Attin et al., 2000]. In addition, with this model, the very first interaction between active agent, abrasives and acid could be investigated. Even if an intraoral brushing model was to be preferred, the brushing procedure was performed extraorally in order to increase standardization regarding duration, force, and velocity, which is according to current recommendations [Wiegand and Attin, 2011]. For further standardization, only 1 calibrated examiner performed the abrasive procedures, who was extensively trained. Laboratory-prepared toothpastes were chosen to exclude the influence of other ingredients, which can be different between commercially available products.

In both groups results for placebo were as expected for D1, with a reduction of hardness by approximately 30% [Attin et al., 1997; Carvalho and Lussi, 2014]. Treatment with placebo inclusive brushing led to a slight increase in microhardness (%RHL ranged between 78 and 85% in the placebo groups), which is plausible since the brushing procedure removed parts of the outer, partially demineralized enamel. As no remineralization occurs during a 4-h in situ period [Ganss et al., 2007; Lussi et al., 2014], the recovery of hardness in the placebo group was mainly to be attributed to the partial removal of the outer demineralized structures. However, this outer layer was not completely removed by brushing [Kuroiwa et al., 1994] indicated by the still reduced surface hardness confirming results of the literature [Carvalho and Lussi, 2014].

The findings of the erosion resistance calculations are of particular interest, turning the relevance of saliva during the development of erosion out. Interestingly, no differences in pH and buffering capacity were found between both groups, which is in concordance to the literature. Buffering capacity can be reduced during radiotherapy; however, this parameter seems to normalize after a period of 3 months after therapy [Marangoni-Lopes et al., 2016]. Other parameters than pH and buffering capacity might be of more relevance in the present study. The saliva forms the acquired pellicle on the surface, which has substantial thickness after the 4-h in situ phase under normal flow conditions [Hannig et al., 2003], showing a considerable erosion protecting effect [Hara et al., 2006], relevant for the second demineralization (D2). In the low flow rate group, the protective effect seems to be impaired. Qualitative and quantitative effects could play a role. Either the pellicle is thinner due to the minor amount of saliva as well as the minor presence of pellicle-forming proteins in the oral cavity, or the pellicle is differently structured, since the saliva composition changes under certain pathological conditions [Paszynska et al., 2015; Marangoni-Lopes et al., 2016]. Both options could lead to a less protective pellicle. The SEM pictures show indeed differences between the groups, perhaps due to different pellicle effects in each group (Fig. 3).

The results of the active agents' effect in the normal flow rate group are in good concordance to the literature. The sodium fluoride preparation showed a relative increase in efficacy in comparison to placebo by approximately 25% (%SMH_total), an effect dimension which was previously reported [Schlueter et al., 2009a, b; Huysmans et al., 2011; West et al., 2015]. In contrast, this effect could not be verified under hyposalivatory conditions, under which the effect of NaF was reduced to a dimension similar to placebo for all evaluated parameters. The effect of NaF is apparently modulated by the amount of saliva [Scaramucci et al., 2013]. The efficacy of sodium fluoride-containing preparations is based on the formation of CaF₂-like particles on the surface. These precipitates are rapidly dissolved under erosive in vitro conditions; however, they seem to be more stable under in situ conditions due to the impact of saliva and pellicle [Ganss et al., 2007]. Maybe, the effect of the preparation tends more towards in vitro than in situ conditions if the saliva flow is reduced.

The stannous ion-containing preparations showed better efficacy regarding postprocedure in situ relative %SMH_total, %RHL and %RER under both low and normal flow conditions, confirming protective effects found in previous in situ studies [Hove et al., 2008; Huysmans et al., 2011; Hove et al., 2014; Schlueter et al., 2014; Hove et al., 2015] and in vivo studies [Young et al., 2006] including participants with a normal saliva flow rate. Stannous ions strongly react with hydroxyapatite, forming stable coatings composed of different salts already after a single application, or being incorporated into the enamel after multiple applications creating a more resistant substrate [Babcock et al., 1978; Schlueter et al., 2009a]. In the presence of saliva, the stannous ions interact not only with mineralized tissue, but also with proteins of the acquired pellicle [Huysmans et al., 2011; Rakhmatullina et al., 2013; Bellamy et al., 2014], like mucins and albumins [Algarni et al., 2015]. Such an interaction can lead to an increase in pellicle thickness [Algarni et al., 2015] resulting in a stronger barrier against acids [Buzalaf et al., 2008; Hara et al., 2014]. These findings support the above hypothesized relevance of pellicle thickness. Interestingly, the stannous ions seem to compensate, at least in part, the reduced protective effect of saliva under hyposalivatory conditions. Whether it is due to an increased acid resistance of the dental hard tissue by the interaction of stannous ions with the dental hard tissue, the increase in pellicle thickness or a combination of both cannot be answered with the present study. However, the SEM pictures (Fig. 3) showed an enamel surface with less evidence of an etching pattern for the Sn-based toothpastes, strengthening the effect of this ion on acid solubility of enamel and, when comparing salivary conditions, the suspected pellicle effect.

Chitosan shows notable antierosive as well as antierosive/antiabrasive effects under normal flow conditions if used in combination with stannous and fluoride ions in a toothpaste [Ganss et al., 2012; Schlueter et al., 2014; Carvalho and Lussi, 2014] or mouthrinse [Pini et al., 2016]. Chitosan is a cationic biopolymer able to bind with proteins [Van der Mei et al., 2007], fluoride [Keegan et al., 2012], and other ions from saliva and enamel surface [Lussi & Carvalho, 2015], and abrasives from toothpastes [Lundin et al., 2008]. Furthermore, chitosan is able to increase tin retention on tooth surfaces [Hove et al., 2008] and to reduce friction effects between the dental hard tissue and the abrasive agent by forming layers on the surface [Raviv et al., 2003], explaining the better efficacy of preparations with a chitosan additive. However, in contrast to the literature, the NaF/Sn/Ch formulation was less effective than the SnF/NaF formulation in the present study. Most likely the biopolymer needs several applications to form the protective layer or to increase tin retention for developing the antierosive effect in case of a normal flow pellicle, which was not the case in the single application model of the present study. Interestingly, under low flow conditions, both stannous ion-containing toothpastes showed the same efficacy (%RER), with an in tendency better effect of the chitosan-containing preparation. The chitosan seems to compensate in parts the reduced protective effect of saliva in low flow persons.

Besides all advantages of the study, one has to consider also the shortcomings of the study. The design only investigated the very initial stages of erosion, which is only a small part of the disease. Therefore, further studies with a longer observation period and a multiple application of the preparations should be performed in order to confirm the protective effect of the stannous or even stannous and chitosan-containing preparations in case of hyposalivation. Another shortcoming is the method for quantification. Surface hardness is only appropriate in case of very initial erosion, which should be simulated in the present study. However, it cannot be fully excluded that a surface loss was provoked. Furthermore, it is well stated that the dental surface hardness loss by an erosive attack cannot be recovered. Maybe, a small rehardening by antierosive agents in the presence of saliva can be achieved. However, it could also be that the measurement was performed in precipitates formed on the surface by pellicle, chitosan or minerals. To compensate at least in part this shortcoming, the interpretation of results was mainly focused on the %RER values, since these values seem to be more independent of potentially formed precipitates. A third shortcoming is the sample size, which is quite small, not allowing drawing final conclusions from the results, which have to be confirmed with a larger study.

From the findings of this pilot study, it can be concluded that salivary flow is important to modulate the efficacy of antierosive preparations probably influencing their action mechanism, which deserves more focus in future research. The NaF toothpaste showed a limited effect against enamel erosion in case of normal and even no efficacy in case of low salivary flow. Sn-based toothpastes provided greater antierosive protection, regardless of the salivary flow, being considered the most effective agent to be used against enamel erosion in patients with xerostomia, even if the efficacy is worse than under a normal flow rate. Although final conclusions cannot be drawn from the results of the present pilot study, they indicate that stannous ions are quite effective, and it seems to be worth paying attention to further research in this area to develop strategies for the care of patients suffering from hyposalivation due to different reasons. Chitosan may also be a promising additive, since it showed a slight increase in antierosive action of stannous ion-containing preparations even after a single application. As it was found for stannous ions, showing a better effect after multiple application, it could be that after long-term use the biopolymer is able to compensate the reduced effect of a low flow rate to a greater dimension, which should be elucidated in further studies.

Acknowledgments

The authors would like to thank the Department of Oral Diagnosis (Semiology and Oral Pathology Sections) for their support with the patients who received head and neck radiotherapy. Besides, the authors are grateful to Drogal Pharmaceutics (Piracicaba, SP, Brazil) for their support in formulating the toothpastes used, to Mr. Waldomiro Vieira Filho (Laboratory of Biochemistry of Piracicaba Dental School) and Mrs. Birgit Meier (University of Giessen) for their technical assistance. Financial support was provided by Coordination of Training of Higher Education Graduate/Capes, Brazil.

Ethics Statement

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki Declaration and its later amendments, or comparable ethical standards. Informed consent was obtained from all individual participants included in the study.

Disclosure Statement

The authors declare that they have no conflict of interest.