This in vitro experiment was conducted to simulate a clinical environment of the outpatient dentistry service at our institutional hospital. The treatment rooms were separated by walls without windows from ceiling to floor, the back side of the dental unit was open to provide passage for staff, and mechanical ventilation was provided once every 30 min through the supply and exhaust vents in the ceiling. All experimenters wore safety glasses and were not allowed to stand facing the laser light source.

A dental manikin (Simple Manikin III; Nissin Dental Products, Kyoto, Japan) and jaw model (Hard Gingiva Jaw Model, Nissin Dental Products) were mounted in a dental treatment unit (Signo G50; Morita, Saitama, Japan) in the horizontal position. An AT (TwinPower Turbine P PAR-EX-O-DI, Morita; 450,000 rpm), a contra-angle EM (TorqTech CA-DC-O, Morita; 40,000 rpm), and a 1:5 speed-up contra-angle EM (×5EM; TorqTech CA-5IF-O, Morita; 200,000 rpm) were used to simulate cutting procedures on an artificial tooth on the maxillary left central incisor (Fig. 1). The water injection volume per minute for the AT and EM was measured three times in advance using a measuring cylinder, and the average water injection volume was determined to be 50.5 mL/min and 43.0 mL/min, respectively. A flexible arm stand was used to secure the handpieces, such that the bur was parallel to the tooth axis. To verify the effectiveness of the HVEs, an intraoral vacuum (IOV) accompanying the dental treatment unit and an extraoral vacuum (EOV) (Free arm Arteo-T, Tokyo Giken Inc., Japan) was used. The IOV was grasped with a flexible arm stand or hand such that it was positioned on the labial side of the artificial tooth. This placement was selected to ensure that the direction of spray dispersal matched that of water injection from the instruments, as determined in preliminary experiments. Moreover, based on these preliminary experiment findings, dental sprays were rarely observed outside the manikin’s mouth during procedures involving the molars. As a result, both qualitative and quantitative evaluations were considered difficult, and the molars were excluded from the experiments. The EOV was placed 10 cm from the mouth, as recommended by the manufacturer. The operating conditions of the HVEs were set as follows: (1) no HVE, (2) EOV only, (3) IOV only, and (4) EOV + IOV.

The dental manikin and jaw model settings in a dental treatment unit. High-speed handpieces (air turbine, contra-angle electric micromotor, and 1:5 speed-up contra-angle electric micromotor) were operated to simulate cutting procedures

To evaluate dental spray dispersion, a laser light sheet (PIV Laser G2000, Katokoken Co., Kanagawa, Japan; 532 nm wavelength, 2 W output power, continuous wave) was used for visualization. A high-speed camera (k5, Katokoken Co.) was used to capture images (Fig. 2). The frame rate was 1000 frames per second. The imaging angle of view was 294 × 220 mm, and its resolution was 640 × 480 pixels (Fig. 3). The dynamics and dispersal range of the visualized dental sprays were qualitatively evaluated from the chin of the manikin to the position of the surgeon. Note that the actual size of one pixel that can be captured at this angle of view is 458.68 μm, which is larger than the WHO-defined boundary between droplets and droplet nuclei (5 μm); therefore, this experiment is only an observation of droplets. For quantitative analysis, five still images were randomly captured from the video taken in each operating condition. The images of the experimenter and manikin were blacked out in the monochrome still images, and the ratio of the white droplets occupying the screen was calculated and averaged. We used an image analysis software (ImageJ version 1.53 t; National Institutes of Health, Bethesda, MD) to binarize all image data with the same luminance threshold.

Equipment settings for dental spray visualization with laser light sheet

Images taken by the high-speed camera (calibration data). Each image was taken with an angle of view of 640 × 480 pixels, and the actual size of one pixel was 0.45868 mm

To evaluate the pollution of environmental surfaces in clinical dental settings, we observed three-dimensional spray dispersal with water-sensitive papers (WSPs; Syngenta Water-Sensitive Paper, 52 × 76 mm; Spraying Systems Co., Glendale Heights, IL). WSPs were fixed at equal intervals in the 12, 3, 6, and 9 o’clock positions around the manikin’s head using plastic boards and disposable chopsticks. For vertical measurements, a fishing line was hung from the ceiling, to which a WSP was attached. Three 30-s mock treatments were performed under conditions wherein the WSPs were placed at distances of 10, 20, and 30 cm from the manikin’s head (Fig. 4). We used ImageJ to binarize all image data with the same luminance threshold, and the area discolored by water adhesion was measured and averaged over three times.

Experimental setup. A Setting up an experiment to observe the pollution of environmental surfaces using water-sensitive paper. B Scanned image of water-sensitive paper (12 o’clock direction, 10-cm distance, 1:5 speed-up contra-angle electric micromotor). The area where water adheres to the surface turns blue

Comparisons between each operating condition were made using one-way analysis of variance, followed by post-hoc group comparisons using Tukey’s honestly significant difference test. All statistical analyses were performed using JMP®13 (SAS Institute Inc., Cary, NC, USA), and the statistical significance level was set at α = 0.05.