An in vitro screening assay for dental stain cleaning

The natural colour of permanent teeth is largely determined by dentine and modified by the thickness and translucency of the overlying enamel [1]. The appearance of the teeth, particularly whiteness, is aesthetically important to individuals and tooth discolouration is a common dental patient complaint. Personal dissatisfaction with the appearance of the dentition has been reported to range from 17.9 to 52.6% [2‐6] and the causes of tooth discolouration are multifactorial and are classified as extrinsic, intrinsic and internalised discolouration [7]. Intrinsic and internalised discolouration arise generally during tooth development or during disease and the more extensive localisation deep within the dental tissues constrains its reversal. Extrinsic staining of the tooth can arise from a variety of sources, such as smoking, red wine consumption, and the intake of cationic compounds, such as chlorhexidine or stannous salts [7‐10]. Stain removal can be challenging as staining compounds can be bound to dental plaque and the acquired pellicle, as well as directly to the enamel surface [11]. Thus, oral hygiene strategies need to address both the removal of dental plaque and the extrinsic stain bound in these different ways.

While a variety of stain removal and tooth whitening procedures are used professionally in the clinic [12‐15], they are relatively costly and labour intensive [16]. Furthermore, there is considerable demand for ‘over-the-counter’ (OTC) tooth-whitening products that can be integrated into a normal oral hygiene regime, and whitening toothpastes are commonly used for this [17]. In general, tooth-whitening toothpastes function by abrasive removal of extrinsic stain associated with the dental plaque, the acquired pellicle and the enamel surface together with the chemical cleaning action of constituents, such as sodium tripolyphosphate (STP), sodium pyrophosphate, sodium hexametaphosphate, hydrogen peroxide, as well as by the activity of the enzymes papain and bromelain [4].

Although randomised controlled trials (RCTs) can provide a high degree of evidence for cleaning efficacy, development of a robust laboratory model to evaluate the mechanisms of action and screen efficacy of whitening products or bleaching agents would be of significant benefit [18]. In vitro models allow rapid screening of a range of potential products; the most effective can subsequently be clinically assessed in RCTs. Various approaches have been used to investigate extrinsic stain removal by oral care products using a wide range of substrates such as polymethylmethacrylate, hydroxyapatite and enamel (human or bovine). Model stains, including tea, coffee, gastric mucin, soy broth, bacteria, chlorhexidine, saliva, tea with orange II Rhodamine, blood, cola, wine, tobacco and ferric-tannate, have all been used [19‐24]. One of the most commonly applied in vitro approaches uses cut, polished and acid etched bovine enamel specimens stained with a solution containing a combination of coffee/tea/gastric mucin/Sarcina lutea turtox [19]. The stained specimens are then repeatedly brushed for a period of time using an automated approach with toothpaste slurries. Furthermore, a recent model has utilised a ferric-tannate coating deposited onto highly polished sintered hydroxyapatite discs and was reported to mimic the daily control of pellicle growth, maturation and staining [24].

A significant limitation of current protocols is the weak bonding between the stain and substrate constraining discrimination between different cleaning products and regimes. A further issue is the need for a staining time of several days necessary to achieve sufficient stain build-up for high-throughput analysis and discrimination of cleaning regimes. There is also a need to better model stain interactions with natural enamel to provide good clinical relevance. Subsequently, the aim of this study was to establish a relatively rapid and reproducible staining protocol with appropriate bonding to enamel, which would allow high-throughput assessment of the cleaning ability, and discrimination between oral care products, following brushing.

The in vitro stain methodology developed in this study provides a relatively inexpensive, easily performed, rapid and reproducible staining protocol which allows examination of stain bound to a natural enamel substrate. Of the many laboratory methods used to evaluate the cleaning potential of toothpaste and toothpaste ingredients, that developed by Stookey et al. [19] (PCR methodology) is commonly used by the oral care industry. However it requires a time consuming stain deposition protocol with opportunities for reproducibility issues when used at different test sites. The data presented here demonstrate good agreement between our stain methodology and the PCR methodology in terms of brightness changes after brushing using the same batches of toothpastes. The current study has also demonstrated the retention of the stain after toothbrushing in the presence of either standard toothpastes or whitening toothpastes as well as the most common chemically active compound in whitening toothpastes, STP. Furthermore it demonstrates the ability of the methodology to discriminate between these various oral hygiene products.

Model stains have been previously described [19‐23] for simulation of the extrinsic staining of the teeth. However, one of the main disadvantages of these staining approaches has been that the stain properties could not be directly compared due to the differences in materials used, i.e., different brand/supplier of reagents used in different studies. However this ferric-tannate stain now provides a reliable standardised model for stain removal purposes in which the stain properties can be controlled due to the chemically defined nature of the stain deposits.

In the previously reported PCR methodology [19], acid etching of polished enamel surfaces has been used to expedite stain accumulation and adherence, although the reproducibility of the etching process can be difficult to accurately reproduce. In the present study, stain could be most readily removed from the diamond polished bovine enamel surfaces, followed by the 400-grit ground enamel bovine surfaces while the stain on the 280-grit ground bovine enamel surfaces was retained to the greatest extent. The surface roughness of enamel specimens can be reproducibly achieved with grinding/polishing protocols contributing to the reproducibility of the presently reported methodology. The surface roughness of the tooth is also of clinical importance, particularly in respect to bacterial retention. A threshold surface roughness for bacterial retention of 0.2 μm has been reported [26]. Indeed, the mean surface roughness values for 400-grit ground bovine enamel were approximately 0.15 μm, which is below the threshold roughness of 0.2 μm and the stain was more readily removed from the 400-grit ground surfaces than from the 280-grit ground bovine enamel surfaces, the latter of which had a mean surface roughness of approximately 0.4 μm. The polished bovine enamel surfaces had mean surface roughness values of approximately 0.02 μm, which is ten-fold less than the threshold roughness for bacterial retention and the stain on these polished surfaces was removed much more readily than that from the 400-grit ground and 280-grit ground bovine enamel surfaces.

The acquired enamel pellicle (AEP) which occurs in vivo results from rapid binding of salivary constituents after saliva contacts a newly exposed enamel surface and reportedly contributes to enamel protection [11]. It is generally understood that the AEP reduces friction between teeth and between teeth and the oral mucosa. It has also been reported that the pellicle enables the initial attachment of bacteria to the tooth surface, which is the first step in plaque formation [11]. In the present study, four acidic polymers were used to mimic an AEP layer on the 280-grit ground bovine enamel surfaces, by either direct application of a thin layer of the acidic polymer prior to application of the ferric-tannate stain or by partly replacing tannic acid with acidic polymers for the stain formation. Appreciable differences in stain removal efficacy were found between the four acidic polymers with a rank order of stain retention of Gantrez > Carbopol > Carrageenan > Xanthan. Gantrez on both occasions showed a very similar pattern of stain removal, indicating its potential to mimic an AEP layer prior to the stain deposits.

STP, a sodium salt of triphosphoric acid with surfactant and chelating properties, has been used in whitening toothpastes for stain removal. In vitro studies with crystalline hydroxyapatite (HA) powder showed that STP was effective in removing existing stain and impeding stain formation through inhibition of the adsorption of salivary proteins or tea stain and the desorption of existing protein and stain from HA surfaces [16, 27]. In vivo trials have also reported significant extrinsic stain removal efficacy for a STP containing dentifrice versus the control [28] as well as a reduction in staining by a chewing gum which contained STP [29]. In the present study, the effects of STP on in vitro stain removal efficacy were examined both in combination with abrasive particles and alone using 280-grit ground finish bovine surfaces. Approximately 35% of the stain was removed when brushing with 10.0% (w/w) STP alone after 5000 strokes, while a control of water resulted in minimal stain removal (less than 1%). The addition of 5.0 (w/w)% or 10.0 (w/w)% STP to a slurry of 15.0 (w/w)% Zeodent 113 abrasive improved stain removal compared with a 15.0 w/w% Zeodent 113 abrasive slurry alone for the same number of brush strokes. These data also highlight the efficacy of STP in contributing to stain removal during oral hygiene measures. However, no differences were observed between the STP concentration of 5.0 (w/w)% and 10.0 (w/w)%. Interestingly, a lack of influence of STP concentration on stain desorption from HA has also been reported elsewhere [27]. Compared with the 10.0 (w/w)% STP solution alone (without abrasive particles), the slurries with 15.0 w/w% Zeodent 113 abrasive particles (+/− STP) showed significantly higher stain removal efficacy, demonstrating that the abrasive was mainly responsible for the removal of extrinsic stain. Similar results have been reported by others when testing different types of whitening toothpaste products [30].

The present ferric-tannate/bovine enamel (280-grit ground finish) methodology provides an easily performed and reproducible model, which can be readily established in laboratories. The chemically defined nature of the stain contributes to reproducibility of the methodology as does the enamel sample preparation protocol. The rapid stain deposition (about 2 h for 10 cycles) facilitates the use of the methodology for rapid screening of large numbers of oral care products to target those products most suitable for testing under clinical conditions. Importantly, it allows good discrimination between various oral care products, including those used for tooth whitening purposes.

With the growth in the desire for whiter teeth by consumers and patients, toothpaste manufacturers aim to maximise the cleanability of enamel extrinsic staining whilst minimising possible damage (wear) to dental hard tissues. Therefore, it is important to investigate the wear and roughness in parallel to stain removal when using the present methodology.