In a dental operatory (with the door closed), we conducted a benchtop exercise to compare the spatter pattern obtained while performing a simulated tooth preparation procedure. We covered the air inlet vent to the operatory so that no airflow currents were present, as they could have affected the spatter pattern. We inserted a typodont manikin head (KaVo Dental, Charlotte, N.C.) into the headrest position of a dental chair. We constructed a 4 × 3-foot wooden platform to surround the manikin head as it was reclined into a usual position for operative dentistry so that the maxillary dental occlusal plane was perpendicular to the floor. We then inserted a typodont (D95SDP-200 32 Teeth Soft Gingivae Type, Kilgore International, Cold-water, Mich.) into the maxillary and mandibular positions of the manikin head. Royal blue fadeless bulletin board paper was trimmed to fit and placed on the wooden platform.
We placed two laboratory stands next to the manikin at the 2-o'clock and 12-o'clock positions. Each stand held a three-pronged clamp in which we placed a dental handpiece or the HVE and maintained it in a fixed position. We used vinyl polysiloxane (VP Mix regular set mint [102–8752], Henry Schein, Melville, N.Y.) to adapt the clamps closely to the handpiece and HVE. We oriented the dental handpiece and HVE in such a way as to simulate the position of a right-handed dentist during preparation of the occlusal surface of three posterior teeth (nos. 18, 19 and 20) (Figure 1). We placed a carbide 330 bur in the dental handpiece and oriented it into a small occlusal preparation in the tooth to act as an index for reproducibility.

Overview of the experimental design setup.

To simulate the volumetric size of the oral cavity, we placed the vinyl polysiloxane putty in areas of the typodont in which water may flow. In a clinical situation, the Isolite system and a dental dam adapt closely to the oral soft tissues so that essentially no spaces exist through which water spray could flow into the oropharynx (Figure 2). In the typodont, various spaces exist that may skew the spatter pattern or volume of spray. Placement of the vinyl polysiloxane putty allowed for a closer approximation to in vivo conditions.

Placement of the Isolite system (Isolite Systems, Santa Barbara, Calif.) in the mouth, showing close adaptation to the oral soft tissues. Image of the Isolite system reproduced with permission of Isolite Systems, Santa Barbara, Calif.

We oriented the orifice of the HVE to be parallel to and 1 centimeter from the buccal surface of the experimental tooth during the control and dental dam trials (Figure 3). We used water and air spray to approximate the aerosol plume produced during operative dental procedures, and we used one high-speed handpiece (KaVo INTRAmatic LUX 3 25LHA, KaVo Dental). The handpiece was operated at the maximum torque and rotation speed of 200,000 revolutions per minute for 10 seconds. We set the water flow through the handpiece at 25 milliliters/minute19 (link), 20 (link) and set the air pressure to achieve a normal aerosol plume. We measured the rate of suction of the HVE and the Isolite device by inserting each into a 2-liter graduated cylinder filled with 2,000 mL of water. The HVE cleared all water in the cylinder in 14 seconds, equating to a rate of 142.9 mL/second. The Isolite device cleared all water in the cylinder in 35 seconds, equating to a rate of 57.1 mL/second. We conducted all 72 trials (as described later) in one session and did not adjust any settings.

Experimental setup for the dental dam with a high-volume evacuator. Note the proximity of the high-volume evacuator to the buccal surface of the tooth.

The control consisted of a simulated preparation on teeth nos. 18, 19 and 20, with a bite block in place and an HVE positioned adjacent to the operative site as described earlier. The first experimental condition consisted of a simulated tooth preparation with use of a bite block, a dental dam and the HVE. The second experimental condition consisted of a simulated tooth preparation and use of the Isolite system set at maximum strength. The Isolite system is designed to provide simultaneous isolation of the maxillary and mandibular quadrants with use of a mouthpiece that has flexible flanges. The system also provides illumination and is used to aspirate oral fluids. The dental dam trials involved the use of a standard 6-inch non-latex dental dam punched with three holes to isolate teeth nos. 18, 19 and 20 for each trial.
We added a 0.1 percent fluorescein dye solution (Sigma-Aldrich, St. Louis) to the dental unit water supply. During the simulated tooth preparation procedure, the water spray aerosolized and scattered away from the test tooth; the resulting spatter settled outside of the typodont mouth and onto the paper-covered platform. We removed the bulletin board paper after each trial and allowed it to dry thoroughly. We then numbered each sheet of paper randomly so that scoring would be masked.
We constructed a 5–square centimeter overlay grid with the use of framer's string fixed tightly at 5-cm intervals. We placed the grid on each sheet of paper. We held a light-emitting diode dental curing light (Demi, Kerr, Orange, Calif.) (emitting blue light with a spectral range of between 425 and 500 nanometers) 8 cm from the surface of the paper and used it to fluoresce the spatter droplets that had collected. When viewed through amber-colored protective glasses, the fluorescence was visualized easily (Figure 4). If the operator (M.C.H., J.M.L.) found even one spot of fluorescence within a 5-cm2 grid, he scored the sample as being contaminated. The operator then counted the number of squares with contamination to determine the amount of spatter produced in each trial.

Fluorescent spatter with overlay grid. The image is from one of the control trials for tooth no. 18.

To achieve a power of 0.80 (effect size = 0.20; P < .05), 24 trials in each group were necessary. Therefore, we conducted eight trials for each of the three teeth tested in each group (that is, the control and two experimental groups). This resulted in a total of 72 trials for the experiment. Two graders (M.C.H., J.M.L.) each scored 36 trials. To analyze the data, we conducted a two-way analysis of variance (ANOVA) with the use of statistical software (PASW statistics 18.0.0, IBM, Armonk, N.Y.). The results of the ANOVA indicated both significant main effects and interaction effects. We subsequently conducted a post hoc Tukey honestly significant difference test for the main effects, and we used a Bonferroni correction to evaluate pairwise comparisons of the interaction effects.