To investigate the aerosol generation by a noninvasive real-time observation device and assess the conditions relating to aerosolization during intraocular pressure (IOP) measurements using a commercial noncontact tonometer (NCT).
Prospective experimental and healthy eye studies.
In an initial experimental study, we devised a model mannequin eye to investigate how air puff pressure and IOP of the eye affected aerosol generation. In the human study including 20 healthy volunteer control subjects, the number of tear aerosol particles generated at 20 and 40 mm Hg air puff pressures with and without eye drop was investigated. The recorded aerosol visualization video was analyzed and the number of aerosol particles generated in 5 seconds after IOP measurement was measured.
The experimental and human studies confirmed the aerosol generation during NCT measurements. In the experimental study, when the air puff pressures were set at 20 and 40 mm Hg, a lower IOP (5 mm Hg) generated significantly more aerosols than a higher IOP (25 mm Hg) (20 mm Hg, P = .0159; 40 mm Hg, P = .0079). There was also a significant positive correlation between the air puff pressure and the number of aerosol particles in both high- and low-IOP eyes ( P < .001). At an air puff pressure of 40 mm Hg, the amount of aerosol generated was significantly higher with eye drop than without eye drop ( P = .047).
NCT generates significant aerosolization from the tear film, the amount of which is determined by the IOP and the air puff pressure and the presence of eye drop use before the measurements.
I n general, conjunctivitis is transmitted by direct contact with an infected person or from an area touched by an infected person. The presence of various viruses in ocular infections has been confirmed in the tear fluid. Currently, infections caused by SARS-CoV-2 have been reported worldwide, and COVID-19 is known to cause conjunctivitis symptoms. As with other viral conjunctivitis, detection of viral RNA from 0 to 24.0% of conjunctiva and tear fluid samples has been reported in COVID-19 patients. , Based on this evidence and the presence of multiple potential receptors for SARS-CoV-2, a recent review report suggests that the ocular surface may be a potential route of SARS-CoV-2 infection.
COVID-19 can be caused by droplet infection through coughing or sneezing and through direct contact with the virus. The possibility of SARS-CoV-2 virus transmission via microaerosols has been raised previously, , but it is still not known if the transmission actually occurs. However, in the medical field, it is necessary to consider possible scenarios of transmission and prevent airborne transmission of the virus. Airborne transmission of the virus can occur when certain medical procedures, called aerosol-generating procedures, produce aerosols. In particular, droplets with a particle size of 5 µm or less are called microaerosols. It has been reported that experimental microaerosols can stay in the air for a long time and that SARS-CoV-2 viral RNA can survive in the air for a long time (World Health Organization, Transmission of SARS-CoV-2: implications for infection prevention precautions, www.who.int/news-room/commentaries/detail/transmission-of-sars-cov-2-implications-for-infection-prevention-precautions ). ,
Intraocular pressure (IOP) measurement is a routine examination in ophthalmology and is frequently used not only for glaucoma screening and determining the effect of its treatment but also as a postoperative examination after intraocular surgery. There are several methods for measuring IOP. Among them, the noncontact tonometer (NCT) is considered to be more effective in preventing conjunctivitis through tear fluid than the Goldman applanation tonometer, which measures IOP through direct contact.
In the past, similar concerns about the possibility of infection from tear aerosols have been raised as disturbing when the world faced the HIV infections concomitant with an initial report that showed the generation of tear aerosols with preliminary methodology during NCT. In that previous report, color fluorescence photography was taken during IOP measurement with NCT after instilling fluorescein solution into the tear film. A recent review reported the implementation of environmental control measures, to reduce respiratory droplet transmission of COVID-19, installation of protective shields on slitlamps, frequent disinfection of equipment, and provision of eye protection to the staff. This report also suggested that IOP measurement by NCT should better be avoided from the viewpoint of COVID-19 prevention.
It is important to prevent not only COVID-19 but also future infection transmissions through the tear fluid. The purpose of this study was to investigate the aerosol generation during NCT measurements and describe different conditions of IOP and air puff pressure.
EXPERIMENTAL STUDY USING MODEL EYE
As shown in Fig. 1 , a silicone rubber was fixed with 2 magnets and attached to a mannequin head to create an eye model. The radius of curvature of the silicone rubber part was 8.45 mm, and the thickness was adjusted to create 2 models: one with IOP measured at 5 mm Hg by NCT and the other with IOP measured at 25 mm Hg. A hydrogel soft contact lens (Base curve, 8.6 mm; Power, +5.0 diopter (D); diameter, 14.2 mm; Medalist 1-day [Bausch & Lomb]) was placed on the hemispherical silicone artificial cornea made of a certain thickness after adjusting the size to 10 mm diameter with a circular trepan.
This study was approved by the Institutional review board of Tsurumi University School of Dental Medicine (No. 1825, September 29, 2021) and was conducted in accordance with the principles of the Declaration of Helsinki. Twenty right eyes of 20 healthy volunteers (17 male, 3 female; mean age: 45.6 years) without any history of ocular and systemic diseases, or ocular surgery were investigated in this study. Subjects using contact lenses were requested to stop wearing contact lens 1 day before the IOP measurements. None of the subjects had a history of COVID-related symptoms, body temperature of >36.5°C, history of close contact with a polymerase chain reaction–positive patient or suspected COVID patient, or history of travel in the last 2 weeks. Written informed consent was obtained from all subjects after explanation of possible adverse events during the experiments.
Before the aerosol measurement experiments, tear volume, tear film breakup time, fluorescein staining, and lissamine green staining scores were measured. The Ocular Surface Disease Index (OSDI) questionnaire was also used to investigate the subjective symptoms of dry eye.
TEAR VOLUME MEASUREMENTS
Tear retention volume was measured using strip meniscometry as previously described. In brief, the strip meniscometry tube was placed in contact with the inferior temporal tear meniscus for 5 seconds, and the length of the wetted area was measured.
TEAR FILM BREAKUP TIME
Two microliters of 1% fluorescein solution was instilled into the conjunctival sac, and the tear film breakup time was observed by slitlamp microscopy. The time from blinking to the appearance of a dark break spot on the tear film was measured 3 times, and the average value was used as the tear film breakup time.
OCULAR SURFACE VITAL STAINING SCORE
After measuring the tear film breakup time, the fluorescein staining score was measured. The scoring method was based on the previously reported method of scoring temporal bulbar conjunctiva, nasal bulbar conjunctiva, and cornea with a maximum score of 3 points for each location (a total score of 9 points). The fluorescein staining score was measured after measuring the tear film breakup time. The same method was used to measure the lissamine green staining score using 2% lissamine green staining solution.
OCULAR SURFACE DISEASE INDEX QUESTIONNAIRE
The Japanese version of the OSDI questionnaire was administered to all subjects before examination. The OSDI score was calculated by multiplying the total score by 25 and dividing it by the number of questions that were answered.
MICROAEROSOL VISUALIZATION PROTOCOL IN THE MODEL EYE
All aerosol measurements were performed in a specially designed clean room. The room is a 2 × 5-m enclosed room with an air suction system using a high-efficiency particulate air (HEPA) filter on the entire ceiling, which can be activated to remove dust and other fine particles from the air in a short period of time to enable aerosolization experiments. The current HEPA filter clean room is the sole experimental platform for aerosol visualization experiments in Japan. This special room was air-conditioned to maintain a temperature of 25°C and humidity of 50%. The eye model was attached to the model head and placed at the measurement position of the NCT. A new soft contact lens was applied to the model eye for each measurement, and 30 µL of ophthalmic artificial eye drop solution (Soft Santear; Santen) was applied after absorbing excess water on the surface with a surgical sponge.
The aerosols generated around the eye were visualized by irradiating a light source using a YAG laser (Parallel Eye H; Shin Nippon Air Technologies) designed for aerosol visualization from underneath the model eye, and aerosol images were captured with an ultrasensitive camera (Eye Scope; Shin Nippon Air Technologies). Before each measurement, the measurement room was closed and suspended particles were forcibly eliminated with the filtration device before conducting the experiment. Three consecutive IOP measurements were performed under the same IOP and the same air puff pressure conditions. At first, aerosol images were taken preliminarily around the entire circumference of the tonometer, but because the generation of aerosols was scarcely observed above the tonometer, only the temporal sides were photographed. The visualized images were recorded on the hard disk of a laptop computer, and the aerosol particles were quantified as described later in this section.
MICROAEROSOL VISUALIZATION PROTOCOL IN THE HUMAN STUDY
Fifteen minutes after the tear film and ocular surface examinations were completed, the aerosol measurement experiment was conducted. The measurements were performed in the same measurement room as in the experimental study. Three LED light sources (Parallel Eye D; Shin Nippon Air Technologies) were placed behind the subject as shown in Fig. 1 , and aerosol generation was observed by measuring the scattered light at a location just beside the right eye using an ultrasensitive camera (Eye Scope; Shin Nippon Air Technologies). The other experimental conditions were the same as for the experiment using the model eye.
First, the subject’s IOP was measured in the conventional IOP measurement mode. Then, the air puff pressure was fixed at the IOP measurement mode of 20 mm Hg, and the IOP was measured 3 times consecutively. Next, the air puff pressure was changed to the IOP measurement mode of 40 mm Hg, and the IOP was measured 3 times consecutively. Next, 30 µL of artificial tear drop (Soft Santear) was applied, and after 1 minute, IOP was measured 3 times in the IOP measurement mode of 20 mm Hg. In the next step, 30 µL of artificial tear drop was again applied, and after 1 minute, IOP was measured 3 times in the IOP measurement mode of 40 mm Hg.
The aerosol formation was visualized during all IOP measurements and saved as a movie on the hard disk of a personal computer.
For all IOP measurements, the TONOREF III (Nidek) was used. IOP values were measured in the conventional measurement mode, or in the modes that can measure IOP from 20 to 60 mm Hg when measuring at a constant air puff pressure.
MEASUREMENT OF THE NUMBER OF AEROSOL PARTICLES IMMEDIATELY AFTER IOP MEASUREMENT
A 5-second movie from the air puff was cut out from the original movie file and saved in audiovisual interleave format.
This was followed by quantification analysis using ImageJ (v1.47, NIH).
To prevent the background from interfering with the quantification of the aerosol particles, the background objects were subtracted from all images.
The area where the background was continuously devoid of noise for 5 seconds (mainly from the center to the right side of the video) was specified by looking at the video.
We set the luminance value to be the threshold for identifying the particles from the video. After preliminary investigations, we considered more than 4 pixels to be particles and less than 4 pixels to be noise.
The results of counting the number of clusters of bright spots above the threshold value were saved in a file. The total number of particles generated in 5 seconds was calculated by counting the number of particles that increased from the previous frame as one count in Excel (Microsoft) file.
In the experimental study, the Wilcoxon signed rank test was used for comparison of aerosol numbers between patients with and without eye drop at the same air puff pressure. The correlation between air puff pressure and amount of aerosol particles in the experimental study was determined by Spearman rank correlation. In the human study, the Wilcoxon signed rank test was used for comparison of tear aerosol numbers between different air puff pressures at the same IOP. In both the experimental study and the human study, a P value less than 5% was considered to be clinically significant.
EXPERIMENTAL STUDY USING MODEL EYE
Supplemental Video S1 shows a typical case of aerosol generation immediately after NCT measurement, and Fig. 4 was created from its video. The lower the IOP and the higher the air puff pressure, the more aerosol generation was observed. Almost no aerosol generation was observed in the upper part of the eye.
Fig. 5 A shows a comparison of the mean amount of aerosol particles generated in the model eye at low (5 mm Hg) and high (25 mm Hg) IOPs. When air puff pressure was 20 mm Hg, significantly more aerosol particles were generated at low IOP (142.0±64.1 particles/field) than at high IOP (37.6±29.9 particles/field) ( P = .0159). Similarly, at NCT air puff pressure of 40 mm Hg, significantly more aerosol particles were generated at low IOP (390.2±64.9 particles/field) than at high IOP (165.2±54.3 particles/field) ( P = .0079).
Fig. 5 B shows the relationship between NCT air puff pressure and number of aerosol particles when the IOP of model eye was 5 mm Hg. Similarly, Fig. 5 B shows the relationship between NCT air puff pressure and number of aerosol particles when the IOP of model eye was 5 mm Hg, and there was a significant positive correlation between the amount of aerosol particles and air puff pressure ( P < .0001, R 2 = 0.625). There was also a significant positive correlation between the amount of aerosol particles and air puff pressure ( P < .0001, R 2 = 0.514).
The demographic information of the study participants and the results of tear fluid, ocular surface staining scores, and OSDI are shown in Table 1 .