1

Introduction

Minimally invasive dentistry (MID) advocates the maximum preservation of intact and repairable dental hard tissues through minimising the unnecessary alteration of healthy tooth structure . Ideally, dental polishing techniques aim at removing surface stains efficiently and selectively without altering the underlying sound tooth surface. In air-abrasion, abrasive particles are emitted from a nozzle in an air stream and aimed at the tooth surface. These particles impact the hard tooth surface at high velocity, resulting in the transfer of kinetic energy and the resulting physical removal of adherent extrinsic surface stains/debris . Air-abrasion eliminates bone vibration and minimises a rise in tissue temperature and consequently, reduces the unpleasant characteristics associated with the use of conventional mechanical instruments . However, air-abrasion can result in alterations in an intact enamel surface due to its lack of clinical tactile feedback during use leading to operator over-use on the tooth surface concerned . Hence, air-abrasion operating parameters should be subjected to precise control, and the polishing powder should be minimally invasive, not damaging sound tissue whilst still efficient at surface stain removal at the same time .

Historically, different air-abrasion powders have been used in clinical practice including alumina, calcium carbonate, glycine and sodium bicarbonate. Bioactive glass 45S5 (BAG) powder has been introduced due to its unique properties such as antibacterial effects, remineralisation potential and selective removal of softer diseased/damaged tooth structure . A summary of the properties of a selection of clinical powders currently available can be found in Table 1 . To the authors’ knowledge, there are no previous published studies assessing the susceptibility of dental enamel to acid challenge following the air-abrasion procedures using BAG powder. Therefore, the aim of this study was to compare the effect of three different powders (sodium bicarbonate, alumina and BAG) on the susceptibility of sound dental enamel to subsequent acid challenge. The assessment was conducted in vitro using white light confocal profilometry, a “gold standard” method for assessing enamel surface loss , and multiphoton excitation fluorescence to examine surface topography. Two null hypotheses investigated in this study were (i) the use of air-abrasion has no effect on increasing the susceptibility of dental enamel to acid challenge when compared to a negative control group, and (ii) there is no difference in the level of mineral loss using different powders.

2

Materials and methods

Extracted, caries-free human molars were collected using ethics approval reviewed by the East Central London Research Ethics Committee (Reference 10/H0721/55), stored in refrigerated de-ionised water and used within a month from extraction. One buccal enamel slab from each tooth was sectioned using a diamond-wafering blade (XL 12205, Benetec Ltd., London, UK). Forty enamel slabs were included in this study after inspecting the integrity of the surface using a confocal tandem-scanning microscope (TSM) (Noran Instruments, Middleton, WI, USA), with an ×20 air objective in reflection scanning mode. The samples were included face down in acrylic resin using a hard-anodized aluminium and brass sample former (Syndicad Ingenieurbüro, München, Germany). The outer enamel layer was removed using a water-cooled rotating polishing machine (Meta-Serv 3000 Grinder-Polisher, Buehler, Lake Bluff, Illinois, USA) using a sequential polishing protocol; 180-grit silica carbide disk (Versocit, Struers A/S, Copenhagen, Denmark) for 5 s, 600-grit for 10 s, 1200-grit for 20 s, 2400-grit for 30 and 4000-grit for 45 s, followed by 3 min of ultrasonication to remove the smear layer at the enamel surface. This standardised polishing protocol permitted the removal of approximately 400 μm from the outer enamel layer.

Each samples’ surface topography was scanned prior to surface air-abrasion, after surface air-abrasion and after subsequent pH-cycling, using non-contact white light confocal profilometry (Xyris™ 4000 WL, TaiCaan™ Technologies Ltd., Southampton, UK) with a 10 μm step-over distance and 10 nm vertical resolution. The quantification flatness of the profilometry system used in this study was calibrated using the National Physical Laboratory optical flat. The maximum of the flatness error in the present system is 0.5 μm. Therefore, the baseline required flatness of the samples included in the present study was the step height value of less than 0.5 μm.

A standard scan area (3 mm × 2 mm) was selected on the enamel sample surface, including the targeted area in the centre surrounded by sound enamel acting as an internal sample reference level (control). The resulting topographic images were analysed using surface metrology software (Boddies v1.81, TaiCaan™ Technologies Ltd., Southampton, UK) by levelling the reference peripheral sound enamel areas to a “zero” plane. The step height of the lesion surface in relation to the sound enamel level was obtained by averaging five measurements taken within each sample. The differences in the enamel step height for each sample were calculated between pre-abrasion and post-abrasion, and between and post-abrasion and post-pH-cycling ( Fig. 1 ).

A diagram showing the experimental stages and the step height measurements steps.

An Aquacut™ clinical air-abrasion unit (Velopex, Harlesden, UK) was used to treat the enamel surface for 5 sec using the following operating parameters: air pressure, 60 psi; powder flow rate dial, 3 g/min; nozzle angle, 90°; nozzle-surface distance, 3 mm and the internal nozzle diameter, 600 μm . The samples were allocated into four experimental groups ( n = 10) according to the air-abrasion powder, alumina, sodium bicarbonate and BAG, with their compositions described in Table 2 . The negative control samples remained untreated. Following air-abrasion, the samples were rinsed thoroughly with de-ionised water and then submitted to pH-cycling. Before commencing the pH-cycling, the reference areas of sound enamel adjacent to the abraded area were coated with two layers of nail varnish for protection. The samples were submitted to a pH-cycling regimen for 10 days to mimic acid erosion as follows; demineralisation for 6 h each day in a buffer solution (40 ml/tooth) containing 75 mM acetic acid, 2.0 mM Ca(NO ₃ ) ₂ and 2.0 mM KH ₂ PO ₄ (pH 4.3). Remineralisation for 17 h overnight in remineralisation solution (20 ml/tooth) containing 20 m M Hepes, 130 mM KCl, 1.5 mM CaCl ₂ and 0.9 mM KH ₂ PO ₄ (pH 7). The pH-cycling was carried out at 37 °C .

The step height difference measurements were analysed statistically using SPSS statistical package (Version 20, SPSS Inc/IBM, Chicago, IL). Data were tested for normality using Histogram/Q–Q plots/Shapiro–Wilk tests. One-Way ANOVA with Tukey’s HSD post-hoc tests were conducted to calculate significant factors at p < 0.05.

Imaging of the specimens was performed using an in-house manufactured Two-Photon system. Excitation was provided by a tuneable (680–1080 nm) (140 fs) Ti:sapphire laser (Coherent, Chameleon Vision II) coupled to an inverted microscope. Samples were imaged using a ×40 (1.3NA) oil immersion objective (Nikon) and excited at 810 nm. The laser power at the sample was measured as 19 mW and the image format was 256 × 256 pixels with a pixel size of 321 nm.

Surface profile images of enamel topographies were generated using Image J software ( imagej.nih.gov/ij/ ) from the compilation of z -stack images (xyz plane) acquired sequentially in 0.5 μm increments beginning away from the sample surface and extending to 60 μm beneath the surface. Following air-abrasion and subsequent pH-cycling, three points, spaced at least 300 μm apart were examined on the both control and test areas of each specimen.