Materials and Methods
Approval from Research Ethics Committee
This study protocol was approved by the Kocaeli University Ethical Committee of Clinical Research, Non-Invasive Clinical Research Ethical Committee (KOU GOKAEK 2019/10.34) and was conducted following the Declaration of Helsinki of 1975, as revised in 2000. All participants signed an informed consent form for being included in the study. This clinical trial is registered at ClinicalTrials.gov as NCT04853745.
Sample selection and processing of teeth
The patients were recruited from individuals seeking periodontal and/or dental treatment at the Faculty of Dentistry, Kocaeli University, Kocaeli, Turkey. The teeth samples were collected from the participants who met the following criteria: being older than the age of 18, having no history of any infectious disease such as hepatitis and/or HIV (+), being willing to participate in the study, having a single-rooted tooth with a gingival recession on all surfaces and need to be extracted, the tooth must be free of caries, defects, and restorations.
All patients who met the criteria were informed about the study. Patients who gave written informed consent were included in the study.
Before the teeth extractions, root surfaces were marked all around with a bur at the gingival margin. Extractions were performed under local anesthesia with articaine hydrochloride and epinephrine 1:200,000 (Ultracain® D-S, Sanofi Aventis, Istanbul, Turkey). Following the extractions, tooth surfaces were cleaned with an ultrasonic scaler (Woodpecker® Medical Instrument Co. Ltd., Guilin, China) to remove deposits and soft tissue remnants and were inspected under 2.7X magnification (Orangedental®, Biberach, Germany). They were sectioned using a metal disc attached to the handpiece by gingival margin level. Extracted teeth were collected and were randomly divided into 3 groups as follows: sodium bicarbonate group,glycine group, and erythritol group. The samples were stored in 1% thymol.
Evaluation with Micro-CT scanning
Samples were scanned before and after polishing powder application with a high-resolution micro-CT device (Bruker SkyScan 1275, Kontich, Belgium). The scanning conditions were 100 kVp; 100-mA, 0.5-mm Al/Cu filter; 13.1-μm pixel size; and rotation at 0.2 steps. According to the manufacturer’s instructions, each sample was rotated 360°. The mean scanning time was around 1h. Also, to measure tissue mineral density (TMD), two different mineral concentrations of conical MD phantoms (rods) of 0.25 and 0.75 gHAp cm3 were placed next to the samples, and the scan was repeated at the same conditions at baseline and after the polishing procedure.
Tooth surfaces instrumentation
The same air-polishing device (AIR-FLOW® Master Piezon, EMS SA, Nyon, Switzerland) was used for all powder instrumentations. Three types of air-polishing powders were used: sodium bicarbonate-based particles (CLASSIC®, EMS SA, Nyon, Switzerland ), glycine-based particles (PERIO®, EMS SA, Nyon, Switzerland ), and erythritol-based particles (PLUS®, EMS SA, Nyon, Switzerland). All surfaces were numbered, and application was made to the mesial and distal surfaces of the root, buccal and lingual surfaces of the crown to avoid repeated instrumentations. After the device and samples were fixed, a metal plate with a 5 mm diameter hole was placed on the sample to limit the application area. Surfaces one and three were air-polished at medium power setting (9 LED power setting) and surfaces two and four were air-polished at maximum power setting (17 LED power setting). The distance between the handpiece and the tooth surface was kept constant at 5 mm, and the treatment angulation was adjusted to 600. In all applications, the application time was 5 seconds, and the water setting was medium (6 LED). The powder chambers of the device were filled to the maximum level in each application.
Micro-CT image analysis
The special software (NRecon ver. 1.7.4.2, Skyscan, Kontich, Belgium) was used for reconstructions. For the reconstruction parameters, ring artifact correction and smoothing were fixed at 7 and 3 in the order, and the beam artifact correction was set at 40%. 16-bit gray-value images were obtained. Reconstructed images were superimposed using another software (DataViewer ver. 1.5.6.2, Skyscan, Kontich, Belgium). The pre-application images and the incisal edges and buccal cusps of the teeth in this image were chosen as a landmark to align the pre-application and post-application images. The reference and target images were superimposed, and a different image was obtained from the different areas of the two images. This image represented the change in the area after the air-polishing. After reconstruction, the interpolated regions of interest (ROI) were drawn using the software (CTAn ver. 1.16.1.0, Skyscan, Aartselaar, Belgium ). Using these ROIs, defect depth, defect volume, and demineralization depth were measured. Thresholding was applied to each of the images. The threshold was set as follows: the lower limit was set between 20 and 255 (in grayscale values), and the upper limit was set at the top end of the brightness spectrum. A multilevel Otsu threshold was used.
To evaluate TMD, ROIs were manually drawn to include the enamel and exposed root surface which is cementum. Other regions were excluded from the ROI selection. The ROIs were subtracted from the original image by software (CTAn ver. 1.16.1.0, Skyscan, Aartselaar, Belgium ) to allow average gray value and density analysis, and TMD was calculated. To aid TMD calculations, grayscale values were converted into MD values (gHAp cm3) with a linear calibration curve based on grayscale values obtained from two mineral concentration conical phantoms of 0.25 and 0.75 gHAp cm3.
Statistical analysis
Power analysis was performed using software (G*Power Software version 3.1.9.2, Düsseldorf, Germany) to estimate the minimum required sample size. Sample counts were calculated at 12 for each group. Type I error and test power was set to 5% and 90% respectively. Considering the losses that may arise during sample preparation, the sample size was determined to be 48.
The data were analyzed using commercially available statistical analysis software (MedCalc Statistical Software version 12.7.7, MedCalc Software bvba, Ostend, Belgium).
Descriptive statistics that were used to analyze the results included the mean, standard deviation, median, minimum, and maximum. Kruskal Wallis test was used to compare differences in more than two independent variables. Mann-Whitney U test was performed to compare differences between two independent variables. Wilcoxon test was used to compare differences between two dependent variables. P -values less than 0.05 were considered statistically significant.