Colour Changes and Surface Roughness After Air-Polishing for Tobacco Stain Removal

Abstract

Introduction and aims

We aimed to investigate the efficacy of air-polishing in restoring the original tooth colour of standardised tobacco-stained tooth specimens.

Methods

Seventy-two specimens consisting of half dentine and half enamel were daily exposed to the smoke of five cigarettes in an automated smoking chamber. Four repetitions of a 14-day smoking cycle were performed. Specimens were cleaned after each cycle, either by air-polishing with erythritol or sodium bicarbonate powder, or with a rubber cup and pumice stone as control. Spectral photometric colour changes (ΔE) and profilometric surface roughness were measured before and after each cycle. All samples were stored for the entirety of the study period in artificial saliva to mimic oral conditions and to establish a pellicle layer on the samples.

Results

The tested cleaning procedures effectively removed tobacco staining from dentine and enamel samples, however, none of the treatments could restore the original colour.

Mean [SD] tooth colour of enamel specimens after four cycles was not statistically significantly different between erythritol (E = 77.6 [6.6]) and sodium bicarbonate air-polishing (E = 79.8 [7.0]) (P > .05). ΔE showed no significant differences between the air-polishing groups after four cycles (P > .05) but with the control group (P < .001). On dentine there was no statistically significant difference of ΔE between all three groups (P > .05). No significant differences in enamel and dentine roughness compared to baseline were observed after repeated air-polishing with erythritol (P > .05).

Conclusions

Erythritol air-polishing removed smoker's discolourations just as effectively as air-polishing with sodium bicarbonate and polishing with pumice, without altering the roughness of the tooth structures. Tobacco stains do not reoccur faster or stronger after the use of air-polishing devices than after polishing with rubber cup and paste.

Clinical relevance

Air-polishing with erythritol emerged as a safe and efficient option for removing extrinsic tooth discolourations with the least abrasive character. These findings support the clinical relevance of incorporating air-polishing devices, particularly with erythritol powder, in dental practice for managing tobacco-induced tooth discolourations.

Introduction

According to data compiled by Eurostat, 19.7 % of the European population are daily smokers.¹ In 2019, 5.9 % smoked ≥ 20 cigarettes on a daily basis, and 12.6 % smoked < 20 units.¹

Smokers` plaques on teeth are not only made up of nicotine, but a mixture of substances including sugar, tar, cocoa, various chemicals, and impurities found in cigarette smoke.² However, nicotine is one of the main causes of tooth discolouration. Nicotine deposits on dental surfaces may penetrate enamel and cause dental yellowing due to its tendency to turn from colourless to yellow when exposed to oxygen.³ Tobacco staining on dental surfaces is not simply a cosmetic issue. It is widely acknowledged that it is also a significant risk factor for the accumulation and alterability of dental plaque.4, 5, 6, 7, 8 Thus, the smokers’ teeth need to be cleaned regularly by a dental professional to remove stains, extrinsic discolourations and dental biofilm. Professional tooth cleaning may entail the employment of diverse mechanical equipment, including ultrasonic and sonic devices, along with hand instruments, polishing brushes with pastes, and air-polishing devices. The aim is to extract plaque, tartar, and discolourations in a precise and methodical manner which is often time consuming and frustrating as smoke deposits remain in narrow fissures and dimples when traditional instruments are used. In contrast, air-polishing is easy to use and requires less manual dexterity compared to traditional methods, and at the same time - when the right powders are selected - they are gentle on the hard substances.9, 10, 11, 12, 13

Powder-blasting devices have a long history in dental medicine.^11,12,14,15 While in the beginning corundum or silicate crystals were used for cavity preparation, sodium bicarbonate was the first powder employed for supragingival cleaning of enamel.^11,16 Bicarbonate powders` impressive cleaning efficiency was tempered by significant abrasion of dentine of exposed root surface, therefore limiting their potential use especially in periodontitis patients with significant gingival recession.^11,12,17 Furthermore, postoperative sensitivity of the gums following epithelial abrasion was reported.18, 19, 20, 21, 22 In the early 2000s, minimally abrasive powders (MAPs) and improved devices were developed to efficiently remove biofilms without damaging oral surfaces.^9,10,23 The most common MAPs are based on the amino-acid glycine or on erythritol, a sugar alcohol.¹³ Recent clinical investigations in dental prophylaxis showed an increased biofilm-removing capacity of air-polishing using erythritol powder compared to rubber cup polishing with significantly shorter treatment times, and an increased efficacy in the short-term resolution of gingivitis.^13,24,25 Also, in long-term periodontal maintenance MAPs have been proofed to be safe with comparable clinical outcomes as conventional mechanical debridement.²⁶

Reduced treatment times while efficiently removing plaque and stains appear to be particularly attractive for professional tooth cleaning of heavy smokers. We wonder whether the frequent and repeated use of air-polishing can restore the original tooth colour in smokers and whether the regrowth of tobacco stains is affected by the selected procedure. Therefore, the aim of the present study was to compare the colour changes and surface roughness of standardised enamel and dentine samples, which were repeatedly exposed to cigarette smoke and cleaned in vitro, either with erythritol air-polishing, sodium bicarbonate air-polishing or rubber cup and pumice stone paste. The null hypothesis was that there would be no difference in colour change between the control procedures and air-polishing with erythritol after repeated use.

Material and methods

Ethical approval

The Ethics committee of the Medical University of Innsbruck, Austria, approved the study (EK 1027/2023). Patients who donated their extracted teeth for scientific research signed a consent form before tooth preservation. The study was conducted per the Helsinki Declaration of 1964 and its later amendments.

Samples

A sample size of 24 specimens per group was calculated based on the results of our pilot study.²⁷ Ten extracted molars and 14 incisors of the permanent dentition of non-smokers with a baseline colour-range of A1 to A2 (VITA classical shade guide; VITA Zahnfabrick; Bad Säckingen, Germany) were used for this study from the Biobank of extracted teeth, Medical University of Innsbruck, Austria. All teeth were sound and extracted due to medical reasons. Before preservation, teeth were cleaned by rinsing with 3 %-hydrogen peroxide for 15 minutes and stored in a 1 % thymol solution (University pharmacy Innsbruck, Innsbruck, Austria). The teeth were separated in an apical-coronal direction into particles of the same size as the diameter of the spectrophotometer by cutting with a diamond-coated rotating separation-disc (Komet, Lemgo, Germany) at 10,000 rpm. Finally, 72 specimens consisting of half dentine and half enamel were embedded in an auto-polymerizate resin (Kulzer; Vienna, Austria) and then perfectly fitted into the jig of the profilometer for exact and reproducible positioning and measurement.

Colour measurement

Colour measurements were performed with the spectrophotometer VITA Easyshade V (VITA Zahnfabrick; Bad Säckingen, Germany), a cordless portable, battery-operated, contact-type spectrophotometer with a measuring range from 400-700 nm.^28,29 The tooth colour is calculated using the CIELAB Formula with the L*a*b-colour space, where luminosity “L” indicates the beam of light, “a” indicates the position on the red-green-axis and “b” indicates the position on the yellow-blue-axis.³⁰ These three coordinates define one point in the coordinate system, which corresponds to a specific colour. The total variable is calculated with the formula $\mathrm{E}=\sqrt{L^2+a^2+b^2}$, and the difference from baseline to every point of measurement ∆$\mathrm{E}=\sqrt{\Delta L^2+\Delta a^2+\Delta b^2}$. The CIELAB formula, which was used for this study uses ΔE to calculate the shade difference between two colours.³⁰ In contrast to the clinical shade determination, whereby the colour is visually determined by the examiner, the shade determination with the spectrophotometer is more reliable and exhibits greater reproducibility.^29,31,32 The higher the difference the greater deviates the colour, and hence the difference is more perceptible to the human eye. There are two relevant thresholds in dentistry for assessing colour: the perceptibility threshold (PT) and the acceptability threshold (AT).³³ While the 50: 50 PT refers to a colour difference were only 50 % of the observers detect a colour difference, the 50: 50 AT refers to the situation, where only 50 % of the observers consider that the colour of the restoration needs to be corrected or newly fabricated.³⁴ This means that a perfect colour match is a colour difference below PT, while an acceptable colour match in dentistry is a colour difference at or below AT.³³ Based on clinical data previously investigating colour matching with a spectrophotometer, the AT was set at ΔE = 2.7 and the PT at ΔE = 1.2 by Riseco et al. 2021.^31,34,35 These values were also used in the present study. Additionally, for a more comprehensive understanding of our results, ΔE was transmitted to the VITA classical shade guide with the same VITA Easyshade V that was used for our trials.

Roughness measurements

Surface roughness was evaluated using a profilometer (Talysurf; Taylor Hobson AMETEK; Warrenville, IL, USA). In brief, a tactile sensor fumbles a defined length of 2 mm (Cut off: 0.25 mm, measurement-velocity: 0.50 mm / s) and creates a profile of the surface. The profiles' roughness parameters were calculated as follows: Rt represents the total roughness and is determined by measuring the highest peak and deepest valley in µm. Rz is the average roughness, while Ra is the arithmetic average roughness and is measured by calculating the average deviation of the surface roughness of a profile from its mean line (in µm). To calculate the average roughness (Rz), the profile was separated into eight sampling lengths, and the peak and the valley of each sampling length were averaged. This study used Rz instead of Ra, which is more commonly used for standardised samples of dental materials with a plain surface, to calculate the average roughness.^36,37 Ra could be biased due to the complex anatomical surface of the natural human enamel and lead to a falsified average roughness. The analysis also included the total roughness (Rt).

Automated smoking chamber

An automated smoking chamber with an in vitro environment mimicking typical human smoking conditions was developed to generate standardised tooth discolourations.

The smoking chamber's schematic design is drawn from the study by Bernhard et al., in which the effects of cigarette smoke on cultured cells were investigated.³⁸ The design was adapted to meet the dental requirements of the present study. The chamber itself is a 3D-printed cube with internal dimensions of 100 × 100 × 100 mm³. In both, the inlet and outlet, there are two hermetically sealed pneumatic hose connectors (Camozzi; Hall, Austria). The cigarettes are stuck into one of the hose connectors and combusted by a liquid pump that creates negative pressure in the smoking chamber. The liquid pump performs with 35 ml / s water taken from a 2-liter glass bottle (UNILAB, Innsbruck, Austria) for 30 seconds, ensuring that the chamber is flooded homogenously with cigarette smoke. One cigarette produces enough smoke to fill the chamber with a dense concentration of smoke. As the cigarette is finished, the lid is opened to exhaust the residual smoke and the next cigarette will be changed by hand, following the same protocol as mentioned before. This verifies that each sample has been exposed to the same amount of cigarette smoke.

Experimental procedures

Duration of smoking cycles, numbers of smoked cigarettes, and optimal storage for reproducible staining were defined in preliminary studies.²⁷ Finally, the experimental protocol involved four repetitions of a 14-day smoking cycle and professional tooth cleaning after each cycle. Every day the samples were exposed to cigarette smoke of five cigarettes to produce extrinsic tooth discolourations. All samples were stored in artificial saliva (Pickering laboratories; Mountain View, CA, USA) for 48 hours prior to the first smoking cycle and for the entirety of the study period, with the exception of the exposure time in the smoking chamber and during the measurement and cleaning process. This was done to mimic oral conditions and to establish a pellicle layer on the samples.

Every two weeks, the first test group was cleaned with air-polishing using erythritol powder, the second test group with air-polishing using sodium bicarbonate, and the control group with rubber cup and pumice stone. In both air-polishing groups, samples were cleaned ten seconds with the airflow device Prophylaxis master and the Airflow MAX handpiece (both by E.M.S.; Nyon, Switzerland) as recommended with the settings of water 100 % and powder 100 % and in a circumferential movement.

Colour and surface roughness measurements were carried out both pre- and post-treatment. The whole experiment was independently repeated three times (Figure 1).

Figure 1.

Flowchart of the experimental procedure. After storing the samples for 48h in sterile saliva, the samples were separated in three groups: Cleaning with MAPD and erythritol, MAPD and sodium bicarbonate and cleaning with rubber cup and pumice stone as control group. Baseline colour and surface roughness were measured and afterwards the samples were exposed to cigarette smoke in the smoking chamber for 14 days. After 14 days the samples were cleaned with the appropriate treatment and colour and surface roughness were measured before and after the cleaning. The exposure in the smoking chamber for 14 days and the cleaning process were repeated four times.

Statistical analysis

The statistical analysis was performed by using IBM SPSS Statistics V.29.0.0.0 (IBM Armonk; NY, USA). Mean and standard deviation were used for descriptive analysis, if not stated otherwise. Significance tests were calculated by a one-way ANOVA analysis and t-test for pairwise comparisons. If not normally distributed, non-parametric tests (Kruskal-Wallis) were performed. Significance level was set at α = .05 and the power was set at 80 %. It should be noted that the data pertaining to the colour of enamel and dentine were found to be normally distributed. The data related to the roughness of enamel samples also exhibited a normal distribution, whereas the data related to dentine samples did not.

Results

Changes of enamel colour

Baseline spectrophotometric analysis of enamel revealed a mean colour E [SD] of 87.4 [7.14] for erythritol, 84.59 [6.44] for sodium bicarbonate and 83.05 [6.1] for control with no statistically significant differences, corresponding to a baseline colour-range of A1 to A2 (VITA classical shade guide; VITA Zahnfabrick; Bad Säckingen, Germany).

During the first cycle of 14 days of in vitro smoking, the samples of all three groups darkened statistically significantly compared to baseline with a mean ΔE [SD] of 17.3[6.84] (P < .001 for all post-smoking comparisons with E baseline). Subsequently, all three cleaning procedures led to a statistically significant and consistent lightening of the tooth colour after polishing with a mean ΔE [SD] of 14.08 [4.49] (P = .008) for air-polishing with erythritol, 18.81 [7.05] (P = .079) for sodium bicarbonate, and 22.90 [6.00] (P < .001) for rubber cup and pumice. There were no statistically significant differences between the three cleaning procedures (P = .800), however, the original enamel colour was not achieved in any group (Table 1; Figure 2A).

Table 1.

Colour changes after each cycle of smoking and polishing compared to baseline.

Enamel specimens	Colour difference ΔE to baseline; mean [SD]
	1st cycle	2nd cycle	3rd cycle	4th cycle
Air-polishing erythritol (n = 24)	8.61 [5.00]	16.11 [18.17]	11.75 [5.86]	8.51 [4.41]*
Air-polishing sodium-bicarbonate (n = 24)	9.35 [6.95]	12.98 [4.53]	9.06 [6.00]	8.04 [3.68]†
Rubber cup and pumice (n = 24)	8.19 [6.22]	14.81 [13.51]	12.43 [7.72]	13.08 [5.05]

Dentine specimens	Colour difference ΔE to baseline; mean [SD]
	1st cycle	2nd cycle	3rd cycle	4th cycle
Air-polishing erythritol (n = 24)	11.36 [5.80]	21.92 [18.72]*	16.11 [6.66]	17.74 [6.00]
Air-polishing sodium-bicarbonate (n = 24)	8.41 [5.55]	11.8 [3.71]	14.19 [4.40]	15.36 [4.74]
Rubber cup and pumice (n = 24)	8.09 [3.82]	18.02 [15.74]	13.68 [5.23]	14.29 [6.45]

^⁎P < .05.

^†P < .001.

The tooth colour E with the L*a*b-colour space was measured with the spectrophotometer VITA Easyshade V (VITA; Bad Säckingen, Germany) baseline, and after each cycle of 14-days in vitro smoking and cleaning. The colour difference ΔE is given as mean [SD] and was calculated between baseline and the end of each cycle. The higher ΔE, the greater is the colour difference after each cycle to baseline. None of the polishing procedures could restore the baseline colour at any time, neither on enamel nor on dentine specimens. Statistically significant differences between each test and the control group are marked with * or **.

Figure 2.

Colour measurements of tooth enamel (A) and dentine (B). The spectrophotometer VITA Easyshade V (VITA GmbH.; Bad Säckingen, Germany) measures the enamel colour E with the L*a*b-colour space. The mean values of E with the SD range are shown for the three groups erythritol air-polishing, sodium-bicarbonate air-polishing, and control cleaning with rubber cup and pumice stone. The E-value decreased during the first two smoking cycles nearly by the same amount and subsequent cleaning by any procedure led to statistically significant brightening. Thereafter, tooth colour in each group reached a plateau, which was significantly darker than the initial tooth colour. This was observed for enamel (A) and dentine (B) surfaces.

After the second cycle of smoking and polishing ΔE compared to baseline was statistically significantly greater than after the first cycle for all groups (P = .030), revealing a consistent darkening of the specimens (Table 1). All procedures showed a similar cleansing efficacy with no statistically significant differences between the groups for ΔE after polishing (erythritol ΔE 12.5 [7.09]; sodium bicarbonate ΔE 16.9 [9.56]; control ΔE 17.8 [7.95]; P = .505).

Thereafter, no further darkening compared to baseline was observed after the third and fourth cycle of smoking and polishing with no statistically significant differences of ΔE compared to the first two cycles (Table 1). While there was no statistically significant difference in ΔE between erythritol and sodium bicarbonate air-polishing after four cycles of smoking and cleaning compared to baseline (ΔE = 8.51 [4.41] and 8.04 [3.68], respectively; (P = .928), cleaning with rubber cup and pumice reached a statistically significant greater deviation of the colour of enamel with 13.08 [5.05] compared to erythritol (P = .002) and sodium bicarbonate (P < .001) (Table 1). The colour difference after four cycles of smoking and cleaning between air-polishing and rubber cup and pumice (ΔE = 4.57 for erythritol and ΔE = 5.04 for sodium bicarbonate respectively) was above the threshold of acceptability (AT = 2.7), while the difference between the air-polishing powders was below the threshold of percebtability (PT = 1.2) (ΔE = .47). None of the polishing procedures could restore the baseline colour with statistically significantly lower E-values at any time after the first smoking cycle (Figure 2).

Changes of dentine colour

The same trend was also observed for dentine surfaces (Table 1). Baseline spectrophotometric analysis revealed a statistically significantly lighter mean colour for the specimens for erythritol with E [SD] = 92.06 [5.03], compared to E = 88.18 [5.4] for sodium bicarbonate and E = 84.39 [6.1] for control (P < .001).

During the first cycle of 14 days of in vitro smoking, the samples of all three groups darkened statistically significantly compared to baseline with a mean ΔE [SD] of 27.00 [7.06] (P < .001). Subsequently, all three cleaning procedures led to a statistically significant lightening of the tooth colour with a mean ΔE [SD] after polishing of 14.10 [6.70] for air-polishing with erythritol, 22.1 [7.68] for sodium bicarbonate, and 23.8 [6.56] for rubber cup and pumice (Figure 2B). There were no statistically significant differences in lightening between the three cleaning procedures (P = .059), however, the original dentine colour was not achieved in any group with a noticeable difference of ΔE compared to the baseline (Table 1; Figure 2B. After the fourth cycle, the difference in colour change was ΔE [SD] 17.74 [6.00] for erythritol, 15.36 [4.74] for sodium bicarbonate and 14.29 [6.45] for rubber cup and pumice, with no statistically significant difference between the groups in final colours E (P = .114). The specimens exhibited a comparable level of colour stability after the second cycle to that observed in the enamel samples. However, a more pronounced darkening was observed during the smoking cycles with overall ΔE [SD] being statistically significantly greater in comparison to enamel (P < .001).

Roughness (Rz)

There were no statistically significant differences in Rz between the groups baseline (P > .05) (Table 2). After four cycles of smoking and air-polishing there were no statistically significant differences to baseline in Rz, neither for erythritol nor for sodium bicarbonate powders, and neither for enamel nor for dentine (Table 2). In detail, for the erythritol group, Rz [SD] values at baseline were 4.83 [2.76] compared to 4.77 [2.21] after the fourth cycle (P > .999), and for sodium bicarbonate 5.69 [3.33] and 4.76 [1.20], respectively (P = .804). In contrast, the positive control group that was treated with pumice stone achieved a statistically significant surface roughness reduction with baseline Rz [SD] 4.37 [1.42] and final Rz [SD] = 2.54 [0.97] (P = .007).

Table 2.

Roughness measurements before and after four cycles of smoking and cleaning.

	Enamel			Dentine
	Baseline	Final	P-value	Baseline	Final	P-value
Erythritol powder (n = 24)	4.83 [2.76]	4.77 [2.21]	> .999	4.62 [1.69]	5.19 [2.61]	.855
Sodium bicarbonate (n = 24)	5.69 [3.33]	4.76 [1.20]	> .999	4.26 [1.46]	5.18 [1.82]	.350
Control (n = 24)	4.37 [1.42]*	2.54 [.97]*	.007	5.15 [2.03]*	3.27 [1.38]*	.022

The profilometric surface roughness was evaluated by creating a raw profile of the surface and calculating the average roughness Rz in μm. P-values refer to the comparison of finale Rz with baseline values. Surface roughness was not statistically significantly different after four cycles of smoking and cleaning for air-polishing with erythritol or sodium-bicarbonate, neither for dentine nor enamel. Statistically significantly lower Rz-values after cleaning with rubber cup and pumice reveal smoother surfaces probably due to abrasion (see also Figure 3). Statistically significant differences between the groups are marked with * < .05 or **< .001.

Also, for dentine samples, there was no statistically significant difference in surface roughness for air-polishing, with baseline Rz [SD] = 4.62 [1.69] and final Rz [SD] = 5.19 [2.61] for erythritol polishing (P = .855) and for sodium bicarbonate 4.26 [1.46] and 5.18 [1.82], respectively (P = .350). The control group again showed a statistically significant surface roughness reduction with a decrease in Rz [SD] from 4.76 [3.06] baseline to 3.27 [1.38] after four cycles (P = .022).

Discussion

Removing tooth stains caused by tobacco smoke is both time-consuming and challenging, particularly when dealing with surface irregularities, fissures and small indentations on the teeth. Based on the authors' experience, eliminating tobacco stains is a significant motivation for many smoking patients to seek regularly for professional dental cleaning. However, it is crucial to consider the potential abrasiveness of powder devices when repeatedly used on enamel or exposed root surfaces during periodontal maintenance therapy.¹³ In this study, we aimed to investigate the colour changes of natural tooth surfaces caused by moderate smoking and evaluate the efficiency of different cleaning methods. We designed an in vitro smoking chamber to create standardised tobacco stains and evaluate the cleaning efficiency, in terms of colour change, between erythritol and sodium bicarbonate air-polishing and cleaning with rubber cup and pumice.

The null hypothesis, which stated that there is no difference in colour change between the control procedures and air-polishing with erythritol after repeated use, was not rejected. Consequently, air-polishing with erythritol proved to be as effective in removing tobacco stains as air-polishing with sodium bicarbonate and cleaning with a rubber cup and pumice. However, none of the cleaning methods were able to restore the original baseline tooth colour at any point after the first smoking cycle, likely due to nicotine deposits penetrating the enamel and causing dental yellowing. This yellowing occurs because nicotine tends to turn from colourless to yellow when exposed to oxygen.³ In order to restore the original tooth colour, by eliminating intrinsic discolourations, dental bleaching or the use of microabrasive pastes should be considered in vivo.³⁹ It has been demonstrated that in office or at home bleaching is an effective method for brightening teeth, even in the case of discoloured teeth, provided smoking cessation.40, 41, 42 From a clinical point of view, the enamel colour in our study changed on the VITA classical shade guide from A1/A2 at baseline to A3 / A3.5 after four smoking cycles. For dentine specimens the colour changed from A1/A2 baseline to C2 / C3 on the VITA Classical Shade Guide. The colour differences between the test and control groups were below the perception threshold (PT), meaning that neither patients nor dental professionals would recognize a colour difference between the cleaning procedures.

As shown in Table 1, during the first two smoking cycles the colour of enamel and dentine specimens decreased by approximately the same value in all three groups. During the third and the fourth cycle neither the darkening due to in vitro smoking nor the brightening after tooth cleaning were that embossed compared to the first two cycles. The more cycles the samples were exposed to cigarette smoke, the more the ΔE value equalized. For dentine, cleaning with rubber cup and pumice achieved a statistically significant greater brightening compared to erythritol, probably due to a more aggressive substance removal, which is undesirable in vivo. Although the sodium bicarbonate demonstrated a comparable brightening effect to that of the control group, with no statistically significant difference, it is not recommended for clinical use due to the aforementioned limitations.^11,17 In general, it is important to consider the limitations of the conventional procedure and the potential impact on the final outcome. Powders can reach fissures and pits more homogenously than a rubber cup and pumice, which probably results in levelling and greater loss of substance of uneven surfaces. A study of Kruse et al. investigated the effects of hand instruments in comparison to air polishing. Despite an increase in surface roughness for both groups, air polishing resulted in a tissue loss of ≤ 50% compared to hand instruments. Furthermore, polishing with a rubber cup and polishing paste was unable to reduce the increased roughness caused by hand instruments.⁴³

No significant differences in enamel and dentine roughness Rz were observed after repeated air-polishing compared to baseline (P > .05) (Table 2). The roughness of dentine treated with erythritol retained almost equal with Rz [SD] = 4.62 [1.69] baseline and 5.19 [2.61] after four cleaning cycles (P = .855). But also air-polishing with sodium bicarbonate resulted in a non-significant difference of Rz [SD] with 4.26 [1.46] baseline and 5.18 [1.82], respectively (P = .35). This is irritating, as other studies have shown an increase in surface roughness due to air-polishing with sodium bicarbonate.^9,23 The missing difference could be due to a type II error, as sample size calculation was done for colour changes, or due to the applied variable Rz as roughness profiles indeed show higher peaks and lower valleys for sodium bicarbonate than for erythritol (Figure 3). This results in a similar average roughness Rz, but from a clinical perspective, the surface is rougher than the erythritol samples. Indeed, most previous studies used Ra (average arithmetic roughness).^36,37 As we were using natural tooth samples with uneven natural surfaces Rz was undoubtedly the more reliable option in this case.

Figure 3.

Representative roughness profiles [µm] of dentine after four cycles of smoking and cleaning with erythritol (A), sodium bicarbonate (B) or control (C) . The x-axis of the measured profile equals 2 mm, the amplitude (y-axis) of the graphic is ± 4 µm. It clearly illustrates the differences, with the roughness of the sodium bicarbonate group (B) having higher peaks and lower valleys than the erythritol group (A). The smooth roughness profile of the control groups (C) results due to the wear of the pumice stone.

The enamel roughness after cleaning with rubber cup and pumice decreased statistically significantly from 4.37 [1.42] to 2.54 [.97] (P = .007). Although the surface becomes smoother, the reason for this decrease is likely due to the undesirable abrasive nature of pumice stone. Nimri et al. elucidated the variability of abrasiveness properties in pumice stone, attributed to various factors such as particle size, concentration, and surface pressure.⁴⁴ Thus, we raise a critical inquiry regarding the potential enamel and dentine wear caused by professional tooth cleaning with pumice stone. In the present study, we used pumice stone as an effective control procedure in removing discolourations. Concerning the aspects of the unpredictable wear and use of pumice stone, the MAPD should be preferred in general and especially erythritol powders as the sodium bicarbonate showed high abrasiveness in previous studies.^9,11 It is noteworthy that all samples darkened by a similar factor after each cleaning procedure, refuting statements that teeth might darken faster after air-polishing than after traditional cleaning methods. Furthermore, this was also shown by Okumus et al. who demonstrated that external discolouration was even significantly greater following polishing with a rubber cup than using air-polishing devices.⁴⁵

Finally, it should be mentioned that the enamel microhardness can vary significantly from patient to patient, depending on various factors such as nutrition, the use of fluoride, or oral diseases.^46,47 The evidence about the abrasiveness of air-polishing devices or polishing pastes on enamel varies greatly since there are many generations of air polishing powders, polishing pastes and devices, and also different methods of surface roughness measurements and values. Earlier studies showed the high abrasiveness of air-polishing devices used with sodium bicarbonate.^14,15 Glycine was the first powder that could also be used subgingival on root surfaces with air-polishing devices.²³ Erythritol represents a further development, as it can be used just as gently on root surfaces and composite materials, but the efficiency of discolouration removal appears to be greater.^48,49 Several studies demonstrate the superiority of conventional methods involving a rubber cup and polishing paste for enamel cleaning.^50,51 On the other hand, other studies suggest significant benefits of air-polishing devices with regards to their cleaning efficiency, particularly on challenging surfaces, and minimal abrasiveness in contrast to polishing pastes.^{23,37,45,51,52}

In preliminary studies we investigated different study protocols for in vitro smoking, the sample storage conditions, the frequency and number of standardised cigarettes smoked per day, the duration of the smoking cycles, and the positioning of the samples in the chamber.²⁷ The present study protocol was chosen to mimic a “near-physiological” cigarette combustion process that affects the tooth surfaces as closely as possible to a moderate smoker.^53,54 The ability to create standardised tooth discolourations represents a key strength of the study, as it allows for a comprehensive and accurate assessment of the effectiveness of different polishing procedures. Paolone et al. described the importance of standardization in this field of research, in order to achieve replicable results.⁵⁵ The protocol for standardisation is comparable to our own. While numerous studies have the specimen exposed to 20 cigarettes, our smoking chamber has generated a uniform smoke dispersion, whereby five cigarettes create discolourations that are analogous to those observed in vivo. In a clinical setting, a multitude of variables, including tobacco brand, oral hygiene, and dietary habits, may influence the efficacy of the procedure. Furthermore, the colour determination using spectral photometry is a more reproducible method for measuring colour than either a visual examination by a dental professional or the patient's own view.³² A recent study observed significant differences between the Vita EasyShade when compared to the gold standard spectroradiometer which means that the absolute values of L*a*b* and E from this study cannot be transferred one-to-one to the clinic. However, as all samples were measured with the same device, the colour differences within the study should be reliable.⁵⁶ As a non-negligible limitating factor, the extent of dental hard tissue wear could not be measured in our study using profilometry alone. Thus, further investigations such as optical coherence tomography or quantitative light-induced fluorescence are necessary. Although these methods do not account for enamel hardness variations, they show promising results and should not be disregarded.^57,58 Further studies should additionally investigate bleaching procedures for treating tobacco discolouration.

Conclusion

Based on our data, air-polishing with erythritol can be recommended as a safe and efficient method for removing tobacco stains. It removed smoker's discolourations just as effectively as the more abrasive methods air-polishing with sodium bicarbonate and polishing with pumice, without altering the roughness of the tooth structures. However, the original tooth colour could not be restored with any of the investigated procedures.

Author contribution

LS and IKS contributed to the study conception and design and funding acquisition. Material preparation and data collection were supervised by LS. The analysis of the data was conducted by VW and LS. The first draft of the manuscript was written by LS and IKS and VW commented on previous versions of the manuscript. All authors read and approved the final manuscript.

Ethics approval and consent to participate

The Ethics committee of the Medical University of Innsbruck, Austria, approved the study (EK 1027/2023). Patients who donated their extracted teeth for scientific research signed a consent form before tooth preservation. The study was conducted per the Helsinki Declaration of 1964 and its later amendments.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.