Cleaning Efficacy of Toothpastes With Varying Abrasiveness—An In Vitro Investigation on Bovine Dentin

ABSTRACT

Objective

To determine how differences in abrasiveness (RDA) influence cleaning capabilities of toothpastes.

Methods

For this in vitro trial, 60 bovine dentin samples were prepared and divided into six groups (G1–G6; n = 10). Groups G1–G5 were arranged in order from low to high toothpaste abrasiveness (G1: RDA: 12, G2: RDA: 29, G3: RDA: 43, G4: RDA: 71, and G5: RDA: 143). Samples in group G6 were used as a negative control and later brushed with pure artificial saliva. In preparation for testing, the dentin samples were stained with a black tea mixture. The respective toothpastes were mixed with artificial saliva in a 1:2 ratio to produce a toothpaste slurry. The samples were brushed for a total of 25 min (with measurement intervals at 2‐, 5‐, 10‐, and 25‐min) using a manual toothbrush in an automated brushing machine. The cleaning efficacy of the toothpastes was determined using photographs connected to a planimeter computer program and statistically evaluated.

Results

Analysis of the median cleaning efficacy values showed that the toothpastes with a higher abrasiveness achieved a greater cleaning effect at each time point. All toothpastes showed a statistically significant difference (p < 0.05) in their cleaning efficacy (taking into account the median, interquartile range (IQR), minimum and maximum values) after 25 min of brushing.

Conclusions

With increasing RDA value, the cleaning efficiency of the toothpastes tested rises. However, a doubling of the RDA value does not necessarily elicit a twofold increase in cleaning effect.

1. Introduction

Toothbrushing using toothpaste is one of the primary methods used for dental care at home and has become a standard daily practice [1]. While therapeutic effects such as prevention of caries and reduction of hypersensitivity are of great clinical importance, consumers seem primarily motivated to use toothpastes to combat dental staining and achieve whiter smiles [1, 2]. Amongst the factors determining dental appearance, tooth colour seems to be one of the primary concern of patients [3]. Numerous studies have illuminated a dissatisfaction concerning tooth colour across various adult populations [4].

Toothpastes targeting stain removal, stain prevention, and tooth brightening typically contain ingredients with mechanical, chemical, and optical modes of action [5]. Abrasive particles are amongst the key functional mechanical components found in toothpastes to improve the removal of superficial stains. Toothpastes contain different types, concentrations, sizes, and hardness of abrasive substances in the form of carbonates, phosphates, and silicates [5] and more recently diamond powder [6] to achieve the desired effects.

While the main aim is to remove food debris, plaque, and surface stains, toothbrushing is typically also accompanied by a certain amount of abrasion of the natural tooth substances. To assess this loss of tooth structure caused by foreign bodies (in the case of tooth brushing, caused by the cumulated effect of the toothbrush and toothpaste), relative dentine abrasivity (RDA) and relative enamel abrasivity (REA) values of a toothpaste are measured. A study by Tawakoli, Sener, and Attin [7] showed that there is a significant positive association between the RDA value of a toothpaste and its cleaning efficacy. In an aim to provide consumers with a toothpaste, which is highly efficient in removing extrinsic staining, some contemporary toothpastes boast very high levels of abrasiveness. While recommendations on toothpaste RDA vary, a toothpaste with an RDA value up to 250 is considered safe and effective to be used on a daily basis by the American Dental Association [8].

To date, there are no known values to quantify whether and by how much the cleaning effect improves with increasing RDA values. It remains unclear, what level of abrasiveness is sufficient to achieve the desired cleaning effects without damaging the tooth structure. By analysing and comparing the cleaning efficacy of five commercially available toothpastes with a large RDA value range, this study aimed to determine the influence of toothpaste RDA on cleaning efficacy.

2. Materials & Methods

A schematic overview of the study procedures is presented in Figure 1.

FIGURE 1.

Schematic overview of study design.

2.1. Bovine Dentine Sample Preparation

For this in vitro study, 60 dentine samples of bovine origin were required. The sample size selection was based on previous similar studies, where sample size ranged from 60 [9] to 80 [10] dentin samples. The teeth were obtained from the lower jaws of cattle, which were previously sacrificed for human consumption. As they are considered slaughter waste, the use of these teeth is not covered by the Animal Welfare Act and thus, no approval from the Ethics Committee was required. The cleaned, anterior, mandibular root dentin samples were selected to be as flat as possible and of uniform colour and size. The dentin samples were ground to a maximum thickness of 6 mm and where necessary, samples were shortened to a maximum length of 15 mm using the Planopol (Struers Inc., Cleveland, United States of America). The minimum length specified was 11 mm to ensure the sample could be embedded in the containers used (self‐made by the laboratory of the Clinic of Conservative and Preventive Dentistry, Centre for Dental Medicine, University of Zürich) and to permit visualisation of the differences between brushed and non‐brushed surfaces.

Following this, the dentine samples were polished. To facilitate this process, the samples were adhesively attached to a scanning electron microscope carrier. The polishing process was conducted in two steps using water cooling and a Tektronix TM 503 device (Tektronix, Beaverton, United States) to ensure the loading pressure remained at a constant 0.4–0.6 N. First, the surfaces were polished for 2 min with the rough, light‐blue Sof‐Lex disk (Sof‐Lex PopOn Disk No. 1982SF, 3 M, Saint Paul, MN, USA) and then for additional 2 min using the light‐yellow Sof‐Lex disk (Sof‐Lex PopOn Disk No. 2382SF, 3 M). A new polishing disc was used for each dentine sample and mounted in a handpiece (KaVo Dental GmbH, Biberach an der Riss, Germany), ensuring a constant speed of 1500 rpm.

After the polishing procedure, the dentine samples were stored in a moist environment and handled solely with the use of gloves to prevent contamination of their standardised surfaces, which could have compromised the uniform staining. The root canals, which were open on both ends of the sample, were plugged with a white plasticine (Pelikan Plastilin white Art.60139, Pelikan Group, Berlin, Germany) and sealed with a layer of black nail polish (Cosnova, Sulzbach, Germany) to prevent intrinsic discoloration when staining the outer surfaces of the samples.

2.2. Sample Staining

To perform a uniform extrinsic discolouration of the dentine, the samples were stained using a black tea solution. To prepare the solution, 390 mL of deionised water was brought to a boil in a beaker (covered with aluminium foil to prevent excessive evaporation). One bag of Lipton Yellow Label Tea Quality No.1 (Unilever, London, UK) and Marks & Spencer Extra Strong Teabags, rich, bright & malty, Strong 3 (Marks & Spencer, Lancing, UK) were left to seep for 10 min. Following this, the teabags were removed from the water, 400 mL of deionised water was added, and the solution was left to cool to room temperature. Finally, the tea was titrated to a pH value of exactly 4.0 using 0.1 M citric acid.

For the staining, the dentin specimens were individually attached (polished surface pointing upwards) to the base of 40 mL polypropylene tubes using putty. 10 mL of the black tea solution was pipetted into each tube, which was then sealed shut with the lid and stored for 17 h, under constant agitation, in an incubator at 40°C (Binder, Tuttlingen, Germany). After completion of the soaking procedure, the samples were removed from the tubes using plastic tweezers and then stored in moist petri‐dishes, ready for use over the next 2 days. A total of 18 dentine samples were stained each round. From these, the 12 most uniformly stained dentin samples were selected for further use. Sixty bovine dentine samples were finally divided into six groups (G1–G6) of 10 samples each. Groups G1–G5 contained samples to be brushed by toothpastes in order from low to high abrasiveness (RDA range 12–143), while samples in group G6 were used as the negative control and later brushed with artificial saliva.

2.3. Embedding of Dentine Sample

Dentine samples were embedded in specially fabricated containers (self‐made in the laboratory of the Clinic of Conservative and Preventive Dentistry, Centre for Dental Medicine, University of Zurich) using a light‐body impression material (President Coltene, Altstätten, Switzerland). The samples were set in a central position, 3 mm below the container's borders. In addition, in order to provide a flat and even surface for the brush heads during the test and to simulate adjacent teeth, two small glass cylinders and one flat glass platelet were embedded on either side of the dentin sample. A photograph of this embedding set‐up can be found in Figure 2.

FIGURE 2.

Photograph of dentine‐sample embedding. (A) bovine dentine sample, (B) engraved demarcation line, (C) ruler (was put in place before taking photographs of the samples), (D) glass cylinders, (E) flat glass platelet, and (F) custom‐made container.

2.4. Preparation of Artificial Saliva Solution

Artificial saliva was used in this study for the preparation of toothpaste slurries (G1–G5) and as a negative control during the brushing procedures (G6). To prepare 1 L of artificial saliva, a volumetric flask (Witeg Labortechnik Gmbh, Wertheim, Germany) of 1000 mL was filled with 500 mL of deionised water and placed on a magnetic stirrer (IKA Werke, Staufen im Breisgau, Germany). The artificial saliva was produced according to the instructions of Klimek et al. [11], by the addition of all the listed chemicals and components, with the exception of mucin. Prior to the addition of di‐sodium hydrogen phosphate, the volumetric flask was filled to approximately 900 mL with deionised water to prevent the precipitation of the phosphate. After all the solid components had dissolved, the pH of the artificial saliva solution was adjusted to 6.4 through titration with 1 M/L HCl (780 pH Meter, Metrohm Schweiz, Zofingen, Switzerland).

2.5. Preparation of Toothpaste Slurries

Five toothpastes with a wide range of RDA values (RDA 12–RDA 143) were selected for testing. These were Candida White Diamond (RDA 12 ± 2; Migros France SAS, Archamps, France), Candida Parodin (RDA 29 ± 2; Migros France SAS), Signal Micro Granuli (RDA 43 ± 4; Unilever RA, Rotterdam, Netherlands), Signal Anti Caries (RDA 71 ± 6; Unilever RA), and Signal White System (RDA 143 ± 6; Unilever RA). The RDA values were obtained in a previous study [6]. An overview of the detailed chemical composition of the toothpastes used is presented in Table 1.

TABLE 1.

Overview of detailed chemical composition of the toothpastes used. RDA values and Abrasive data according to Hamza et al. [6].

Tested toothpastes (Manufacturer)	RDA value	Composition	Abrasive
Candida white diamond (Migros France SAS, Archamps, France)	12	Aqua, hydrogenated starch hydrolysate, potassium citrate, hydrated silica, sodium lauryl sulfate, xantham gum, aroma, sodium acrylates/C10‐30 alkyl acrylate crosspolymer, sodium fluoride, sodium saccharin, zinc chloride, diamond powder, methylparaben, allantoin, limonene, linalool, benzyl alcohol, CI 77891 (https://www.codecheck.info/p/kosmetik/mund‐zahnpflege/zahnpasta/candida‐white‐diamond.html)	SiO2 · nH2O diamond powder
Candida parodin (Migros France SAS, Archamps, France)	29	Aqua, dicalcium phosphate dihydrate, hydrogenated starch hydrolysate, sorbitol, hydroxyapatite, hydrated silica, sodium lauryl sulfate, aroma, panthenol, curcuma xanthorrhiza root extract, caprylic triglyceride, cellulose gum, sodium monofluorophosphate, aroma, panthenol, paraffinum liquidum, trisodium phosphate, allantoin, zinc chloride, sodium saccharin, sodium hydroxide, bromochlorophene, aluminium lactate, aluminium hydroxide, bisabolol, limonene, cinnamal, CI 73360. 1300 ppm fluorid (https://www.codecheck.info/p/kosmetik/mund‐zahnpflege/zahnpasta/candida‐parodin‐plus‐zahnpasta.html)	CaHPO4 · 2H2O Ca5(PO4)3OH SiO2 · nH2O
Signal micro‐granuli (Unilever RA, Rotterdam, Netherlands)	43	Aqua, soorbitol, hydrated silica, peg‐32, sodium lauryl sulfate, aroma, zinc citrate, cellulose gum, sodium fluoride, sodium saccharin, CI 77891, limonene, CI 74260, CI 42090 (https://wiop.unilever.ch/brands/signal/signal‐zahncreme‐micro‐granuli‐959‐8500575‐7/)	SiO2 · nH2O
Signal anti‐caries (Unilever RA, Rotterdam, Netherlands)	71	Aqua, hydrogenated starch hydrolysate, hydrated silica, PEG‐32, sodium lauryl sulfate, aroma, cellulose gum, benzyl alcohol, CI 77891, sodium fluoride, sodium saccharin, limonene, CI 73360 (https://wiop.unilever.ch/brands/signal/signal‐zahncreme‐anti‐caries‐945‐67648875‐300002075341/)	SiO2 · nH2O
Signal white system (Unilever RA, Rotterdam, Netherlands)	143	calcium carbonate, aqua, hydrogenated starch hydrolysate, hydrated silica, sodium lauryl sulfate, sodium monofluorophosphate, aroma, benzyl alcohol, trisodium phosphatee, cellulose gum, sodium saccharin, CI 74160, glycerin, sodium laureth sulfate (https://wiop.unilever.ch/brands/signal/signal‐zahnpasta‐white‐system‐6594‐67657353‐300002103450/)	CaCO3 SiO2 · nH2O

The toothpastes were mixed with the artificial saliva in a 1:2 ratio to simulate the dilution of toothpaste as it occurs in vivo and create the toothpaste slurries for test groups one to five. A total of 90 g of slurry was prepared for each toothpaste (30 g of toothpaste and 60 g of artificial saliva). It was mixed for 5 min on the Ultraturrax T25 (IKA Werke) before it was filled into centrifuge tubes (Greiner, Kremsmünster, Austria) and placed in a refrigerator for storage until further use.

2.6. Brushing

The toothbrush selected for use in this study was the Paro M43 manual toothbrush (Esro, Kilchberg, Switzerland) with medium bristle hardness. Prior to testing, the brush heads were separated from the handle and glued to metal holders (self‐made in the laboratory of the Clinic of Conservative and Preventive Dentistry, Centre for Dental Medicine, University of Zurich). These were then attached to an automated brushing machine (self‐made, laboratory of the Clinic of Conservative and Preventive Dentistry). This device allows for a simulation of the forward and backward scrubbing motion, characteristic of tooth brushing, with six brush heads sweeping over six dentine samples simultaneously. Following each brushing cycle, the used brush heads were removed from the metal holders and replaced with new brush heads to be used in the next brushing cycle.

The embedded dentine samples were clamped into the brushing machine and positioned in such a way that the brush heads would sweep over the centre of the dentin specimens during their brushing motion but not touch the container borders. The boundaries between the surfaces brushed by the brush heads and the adjacent unbrushed surface areas were demarked by engraving the specimens using a pointed metal rod. This would allow easy and accurate repositioning of the dentin samples to ensure that samples would be repeatedly brushed on the same surface areas.

The contact pressure of the brush head on the dentine samples was regulated by a screw on the metal holders and set at a standard 2.5 N. This pressure was then systematically checked using a spring balance (Pesola 600 g, PESOLA Präzisionswaagen, Schindellegi, Switzerland) following each brushing cycle and re‐adjusted where necessary. The speed of the brushing motion was set at 60 cycles per min, where one cycle is defined as a forward and backward scrubbing movement.

Prior to their use in each experimental brushing cycle run, the slurry tubes were removed from the cold storage and placed in a rotator (Antylia scientific, Vernon Hills, IL, USA) to be remixed. A plastic pipette was then used to add 1 g of the group‐specific slurry or the artificial saliva to the individual samples, respectively. After 2‐, 5‐, 10‐, and 25‐min brushing intervals, the slurry was washed off, the dentine samples were photographed and 1 g of the same slurry was applied again.

2.7. Photography

To visually document the dentine‐sample surfaces at baseline, prior to brushing and after 2‐, 5‐, 10‐, and 25‐min brushing intervals, photographs were taken using a Canon EOS 2000D (Canon Europe, London, UK) under standardised ambient conditions. To ensure comparability between images, the shutters of the window in the room in which the specimens were photographed were closed and only the ceiling light was switched on. The camera's highest possible resolution was used, the aperture was set to 1/4, the focus was set to 5.6, and the grain was set to ISO 100. For each photograph, a ca. 11 mm portion of a ruler was placed next to the embedded dentin sample (as seen in Figure 2.) in order to have a visually measurable scale demarking 10 mm which was required for the evaluation.

2.8. Planimetry

In order to determine the cleaning efficacy, that is, to determine the total surface area of the sample which was cleaned, following each time cleaning interval, the initial photographs were displayed on a computer screen and the specimen outlines were traced onto transparent A4 writing films (Folex, Schwyz, Switzerland) overlaid on the computer screen, using foil pens (STABILO International, Heroldsberg, Germany). In addition, the 10 mm scale was marked as a reference and the engraved borders of the brushed surface were marked to create an outline‐template. Each initial writing film template for a sample was then copied four times. The photographs of the samples taken at the four time points (2‐, 5‐, 10‐, and 25‐min of brushing) were overlaid with the copied transparent writing films. Following this, the areas of the sample, which were deemed clean (visual evaluation), were traced onto the films.

Using a computer application, Planimeter SC (self‐developed in the laboratory of the Clinic of Conservative and Preventive dentistry), the total brushed area (total area within the engraved markings of the sample), and the cleaned areas were measured and expressed in mm². The cleaned area was subsequently quantified as a percentage of the total brushed area, thus giving the cleaning efficacy of the toothpaste in question at the respective measurement times. The planimetric analysis was conducted by one, non‐blinded individual.

2.9. Statistical Analysis

The data collected from planimetric evaluation was recorded using Microsoft Excel (Microsoft Corporation, Redmond, WA, USA). All the statistical analyses were performed using the statistical software R including the packages tidyverse and PMCMR.

The target variable “cleaning efficacy” was first evaluated using descriptive statistical parameters. The median values of the cleaning efficacy of each toothpaste at each measurement time point were calculated as a percentage and listed in a table. Based on these values at the different measuring times, a progress curve was plotted for each toothpaste and also for the negative control. Subsequently, the cleaning efficacy for the brushing time of 25 min was displayed graphically and in tabular form. In addition, all values for the cleaning efficacy after a brushing time of 25 min were presented in box plots.

For the conclusive statistical analysis, the Kruskal‐Wallis test was used to determine whether there were any significant differences between the tested toothpastes and the plain saliva after the total 25 min brushing time. In addition, a post hoc Conover‐Test was performed to compare the mean values pairwise. The resulting p‐values were corrected for multiple testing according to Holm. A p‐value of < 0.05 was interpreted as statistically significant.

3. Results

3.1. Cleaning Efficacy at all Measurement Times

The line‐graph in Figure 3 shows the progressive curves of the median values of the cleaning efficacy (in %) at each measurement time point for the five toothpastes and the negative control. The tabular listing of the numerical median values for each group at the various time measurements are presented in Table 2.

FIGURE 3.

Median values of the cleaning efficacy (in %) of all six groups at each measurement time point.

TABLE 2.

Mean, standard deviation, median, interquartile range, minimum and maximum values of the cleaning efficacy (in %) of each toothpaste and saliva after 2‐, 5‐, 10‐, and 25‐min brushing periods.

Toothpaste	Brushing time [min]	Mean	Standard deviation (SD)	Median cleaning efficacy [%]	Interquartile range (IQR)	Minimum (min)	Maximum (max)
Candida white diamond (RDA 12)	2	1.83	0.93	1.89	1.49	0.57	3.17
	5	4.54	2.89	4.10	2.61	1.11	9.93
	10	7.29	3.96	6.77	3.15	1.37	15.51
	25	11.92	5.54	10.79	4.96	4.32	22.93
Candida parodin (RDA 29)	2	4.92	2.40	4.07	3.02	2.40	9.22
	5	13.67	5.87	14.57	10.01	6.72	21.81
	10	28.62	9.11	29.41	9.63	13.65	41.61
	25	62.77	10.53	61.16	12.49	45.66	78.65
Signal micro granuli (RDA 43)	2	7.94	3.19	9.09	3.89	2.09	10.88
	5	20.13	7.60	19.13	10.49	10.06	33.62
	10	42.19	12.55	40.27	14.06	28.74	64.95
	25	78.29	9.32	79.91	6.77	54.28	87.37
Signal anti caries (RDA 71)	2	13.89	3.84	14.81	3.92	5.61	19.94
	5	42.57	10.19	43.67	14.11	20.58	53.49
	10	72.47	13.04	70.75	6.80	53.59	97.58
	25	97.83	4.46	100.00	1.13	86.71	100.00
Signal white system (RDA 143)	2	22.59	6.03	23.05	6.32	13.74	33.73
	5	67.22	12.58	68.91	12.58	47.51	86.32
	10	97.12	4.31	98.24	3.30	85.98	100.00
	25	100.00	0.00	100.00	0.00	100.00	100.00
Saliva	2	0.08	0.18	0.00	0.04	0.00	0.55
	5	0.16	0.22	0.09	0.21	0.00	0.67
	10	0.50	0.64	0.24	0.42	0.00	2.18
	25	0.74	0.75	0.50	0.57	0.16	2.69

The median values of cleaning efficacy of the investigated toothpastes and artificial saliva showed significant differences (p < 0.05, respectively) at each time point. The toothpaste with the RDA value of 143 ± 6 showed the highest and most rapidly achieved cleaning effect, followed by the toothpaste with the RDA value 71 ± 6, where the median value at the 25 min brushing time recorded the same cleaning effect. These were followed, in order, by the toothpastes with RDA values 43 ± 4, RDA 29 ± 2, and RDA 12 ± 2. It should be noted, that the curves of the toothpaste with RDA 43 and RDA 29 were practically identical, with the cleaning efficacy of the toothpaste with RDA 43 being only slightly better than that of the toothpaste with RDA 29 at each time point. The lowest cleaning effect was achieved by the negative control, artificial saliva.

3.2. Cleaning Efficacy after 25 min Brushing Time

A boxplot and table of values of the cleaning efficacy (in %) of each toothpaste and saliva after a 25 min brushing period is given in Figure 4 and can be found in Table 2 respectively.

FIGURE 4.

Median, IQR, minimum and maximum values of the cleaning efficacy (%) of the six groups determined after 25 min of brushing. There was a significant difference between all groups.

All six groups investigated showed a statistically significant difference in their cleaning efficacy at the 25 min brushing time point (p < 0.05). The toothpaste with the RDA value of 143 showed the highest percentage (median/IQR: 100.00/0.0) of cleaned area. This was followed by the toothpaste with RDA 71 (100.00/1.1), RDA 43 (79.9/6.8), and RDA 29 (61.2/12.5) groups. Of all the toothpastes, the toothpaste with the RDA value 12 had the lowest percentage (10.8/5.0) of cleaned surface area, which was nevertheless greater than the cleaned area of the negative control with artificial saliva (0.6/0.2).

4. Discussion

There are numerous factors that might influence cleaning efficacy that may be achieved by toothbrushing. These include the type of toothpaste used, the type of toothbrush used [10], the pressure, speed and angle at which the brushing occurs, the type of tooth discoloration targeted, and the surface condition of the tooth.

The selection of the toothpastes used in this investigation was based on the results of a study by Hamza et al. [6], in which 20 commercially available toothpastes were tested for their RDA and REA values and subsequently classified into five categories as follows. RDA‐1: very low abrasive (RDA < 20); RDA‐2: low abrasive (RDA 20–40); RDA‐3: moderately abrasive (RDA 40–60); RDA‐4: strongly abrasive (RDA 60–80); and RDA‐5: very strongly abrasive (RDA > 80) [6]. The same RDA grouping was also used in previous studies [7, 12]. For this study, one toothpaste was selected from each category, ensuring that the difference in RDA values between each higher or lower abrasive toothpaste was approximately equal. An exception was made for the very strongly abrasive toothpaste group, from which the toothpaste with the highest RDA value was selected.

In the present study, bovine dentin samples were used. This was done, as the use of human teeth proves difficult due to the low availability and oftentimes poor condition of extracted teeth, with carious lesions or fractures. Dentin is, by nature, more porous and softer than enamel and may be more challenging to clean than a smooth enamel surface. Therefore, dentin was the dental hard tissue chosen to investigate the cleaning efficacy of the various toothpastes. In addition, while in a typical in vivo situation, primarily enamel is exposed in the human mouth, exposed dentin due to gum recession or enamel erosion is common. Bovine dentine samples have proven to be a valid alternative to human specimens as shown by Wegehaupt, Widmer, and Attin [13] and provide further advantages due to their larger planar surface and absence of caries lesions and defects [14]. Nonetheless, individual teeth, including the dentin portion, may have slightly different physical properties, as a result of chemical composition or exposure to external or internal stimuli, which may have an influence on the cleaning effect achieved. In order to standardise the dentine surfaces as much as possible, the samples were pre‐treated to achieve a uniform surface, prior to testing, as described above. The staining of the samples was conducted according to the methods described by Tanner et al. [10] and Hamza et al. [9] using a black tea solution with a pH of 4. The soaking in an acidic pH results in more intense discoloration of the dentin as the surface is simultaneously eroded which allows for a deeper staining of the sample. An intensive staining was required in order to be able to evaluate the cleaning efficacy of the tested toothpastes under the desired challenging conditions.

The study aimed to mimic in vivo conditions as accurately as possible and chose to replicate validated experimental procedures as used in numerous previous studies [6, 7, 9, 12, 15]. This includes details such as the contact force [6, 7, 12, 15], the type of brushing head [6, 7, 9], the brushing speed [6, 7, 9, 12, 15] and the measuring times [12, 15] per cycle, the production of the slurries [9] and the staining method of the dentine samples [7, 9, 12, 15]. While these methods are legitimate and valid, it must be recognised that an accurate simulation of the in vivo environment using a single in vitro model is extremely challenging. This is largely due to the fact that further parameters, such as intraoral temperature and pH variations, fluctuations in salivary flow, the formation of the pellicle on the tooth surface and individual brushing behaviour variations or brush selection, the type of tooth discoloration tackled and the surface condition of the tooth, which may have an influence of the outcome of the cleaning efficacy, cannot be controlled or replicated with sufficient fidelity in an in vitro model.

In the present study, the dentine samples were brushed and photographed after four time‐intervals. This is in contrast to the studies by Tawakoli, Sener, and Attin [7] and Hamza et al. [9], where the cleaning effect was assessed only once, after 25‐min [7] and 26‐min [9] brushing duration. A more frequent assessment of the cleaning efficacy was deemed more precise and would allow for easier comparison and as such, this study recorded the cleaning effect at brushing time intervals as seen in two previous studies by Imfeld et al. [12, 15], namely after 2‐, 5‐, 10‐, and 25‐min. It is however important to note that solely the results observed after 2‐ and 5‐min brushing durations may be translated to the clinical situation, as brushing is not usually conducted for 10‐ or 25‐min intervals. Nevertheless, in the present study, the statistical evaluation was only performed for the measurement time point of 25 min brushing duration in order to be able to better compare the results with those of the RDA measurement (which was performed after 25 min of brushing) from the study by Hamza et al. [6] Furthermore, the order of the toothpastes based on their cleaning efficacy (from best to worse, Figure 2) was not changed between 2‐ and 25‐min brushing time.

Photographs were taken at baseline, prior to any brushing, as well as after each measurement time point. These were then first visually evaluated and subsequently the total cleaned area was quantified as a percentage of the brushed area, using the planimeter software. The method described above, which relies on a subjective, visual evaluation, has been used in several previous studies [7, 9, 12, 15]. To achieve a more objective evaluation, one might consider using an alternative optical method to assess colour change or stain removal, involving spectrophotometry, as has been used in previous studies [1, 16, 17].

This study's primary purpose was to compare the cleaning effect of toothpastes with varying RDA values and therefore did not investigate abrasiveness, as this data was already made available for the tested toothpastes through the study of Hamza et al. [6] As described in a study by Tawakoli, Sener, and Attin [7], high cleaning efficacy is often accompanied by strong abrasion and high tooth surface roughness. An interesting extension of the investigation could be to examine and measure the dentine surface roughness or substance loss after the use of each toothpaste, therefore providing insight into which toothpastes offer the best compromise between the cleaning effect, abrasiveness, and resulting surface roughness. This would allow for a more precise specification of the clinical indication of the individual toothpastes.

All toothpastes tested, as well as the negative control, showed a statistically significant difference in cleaning efficacy.

The toothpaste with the RDA value 143 showed the highest cleaning efficacy at each time point. After only 5 min of brushing time, more than 2/3rds of the surface had been cleaned. After 10 min of brushing time, almost 100% of the surface was clean, as shown by the curve in Figure 3, which flattened out considerably after this time point. In a study by Tawakoli, Sener, and Attin [7], this toothpaste already stood out as the toothpaste with the best cleaning efficacy. While the toothpaste has proven to be very efficient and effective in stain removal, this could come at the expense of sound tooth substance, which experiences strong abrasion due to the toothpaste's high RDA value of 143 ± 6.

The toothpaste with the RDA value 71 demonstrated the second‐best cleaning efficacy of the tested toothpastes. The results demonstrate, that although this toothpaste has an RDA value that is only half that of the toothpaste showing the highest cleaning efficacy (71 ± 6 vs. 143 ± 6, respectively), it cleans very effectively with a cleaning efficiency that is not half that of the toothpaste with the larger RDA. Similar observations were already documented in a previous study by Tawakoli, Sener, and Attin [7] It should however be noted, that while both toothpastes achieve 100% cleaning by 25 min of brushing time, the toothpaste with an RDA of 143 achieves almost complete cleaning at an earlier point in time (98% at 10 min). Possible reasons for the observed results may be due to differences in density, size or shape of the abrasives or further ingredients found in the individual toothpastes. This would require a more detailed investigation in a further study.

Analysis of the median values of cleaning efficacy for the toothpastes with RDA value 43 and RDA 29 reveals that both toothpastes have very similar curve progressions at all measuring time‐points and feature a more linear relationship. The toothpaste with RDA value 43 did, however, show a better cleaning efficacy overall. This may most likely be attributed to the difference in RDA value. However, as mentioned above, further investigation as to the true causative feature is necessary to confirm this. The study by Tawakoli, Sener, and Attin [7] also observed a slight difference between the toothpastes in terms of cleaning efficacy, however the observed difference was not as large as in the present study. This may be attributed to the subjective nature of the evaluation method used, which relies on individual visual assessment.

The least satisfactory cleaning result amongst the tested toothpastes was achieved by the toothpaste with the RDA value of 12 (10.8% after 25 min). This toothpaste has a very low RDA value, as determined in previous studies and thus the poor cleaning effect might have been anticipated. It contains diamond powder, which has recently been added to toothpastes as an abrasive [6, 9, 18, 19]. Several studies [6, 18, 19] have been conducted on the abrasive properties of diamond powder, as a single abrasive or as an additional abrasive in toothpastes. These studies showed that toothpastes with diamond powder behave significantly more abrasively on enamel than on dentine and that they therefore have a lower RDA value, but significantly higher REA values, relative to conventional toothpastes [6, 18, 19]. Thus, taking this into account, it may be deduced that this toothpaste has a poor cleaning effect on dentine, while achieving a higher cleaning effect on enamel, despite its low RDA value. Further study would be required to confirm this.

Artificial saliva, as a negative control, was by far the worst performer and achieved practically no cleaning efficacy. This confirms the important role of toothpastes in achieving a cleaning effect.

There is some dispute in the question, whether RDA values are positively correlated with the cleaning effect. The observations in this and two other studies [7, 15] show that there is a positive relation between the RDA value and the cleaning efficacy. Further studies however call this assumption into question with results that show that the abrasiveness of toothpastes does not directly correlate with their cleaning efficacy [1, 9, 20]. A possible explanation for these differences may lie in the variability of abrasive substances found in the toothpastes investigated. Studies showing a positive correlation between the RDA value and the cleaning efficacy [7, 15] mainly studied toothpastes containing conventional abrasives such as silicates, calcium carbonates, or phosphates. In contrast, for example, the study by Hamza et al. [9], investigated different types of toothpastes, with varying abrasiveness and different and in some cases, novel abrasive substances. A more thorough investigation into the relationship between different abrasive substances and observed cleaning effect could provide an interesting opportunity for further study.

It remains unclear whether the behaviour of RDA value and cleaning efficacy may be applied to all types of abrasive substances contained in toothpastes and whether similar cleaning efficacy trends may be translated to other dental hard tissues such as enamel. One should keep in mind that in healthy dental patients, tooth brushing mainly takes place on the enamel and not on the dentine. Enamel differs greatly from dentine in its chemical and physical properties. Previous studies [6, 18, 19] have obtained different results regarding the transferability of the abrasiveness of toothpastes on dentine to enamel. This was primarily dependent on the abrasive substances contained in the toothpastes (conventional and/or diamond powder) and the condition of the tooth structure (healthy or eroded) and resulted in some toothpastes showing a positive correlation in the transferability of abrasiveness of dentin to enamel while others showed no correlation [6, 18, 19]. Thus, as no definite association may be found between RDA and REA values, transferability of results from one dental hard tissue to another remains limited.

5. Conclusions

The results of this in vitro trial confirm that with increasing RDA value, the cleaning efficiency of the toothpastes tested rises. However, a doubling of RDA value does not necessarily elicit a twofold increase in cleaning efficacy.

6. Clinical Relevance

6.1. Scientific Rationale for Study

The following study aimed to determine how differences in abrasiveness (RDA) affect the cleaning efficacy of toothpaste as, to date, there are no known values to quantify whether and by how much the cleaning effect improves with increasing RDA value.

6.2. Principal Findings and Practical Implications

An increase in the cleaning efficiency was observed with increasing RDA value of the toothpastes tested. It is still unknown whether these findings apply to other dental hard tissues such as enamel and whether the observed relationship applies for all types of abrasive substances contained in toothpastes.

Conflicts of Interest

The authors declare no conflicts of interest.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.