key: cord-0271826-8nec3b4b authors: Huang, Q.; Marzouk, T.; Cirligeanu, R.; Malmstrom, H.; Eliav, E.; Ren, Y. title: Ventilation rate assessment by carbon dioxide levels in dental treatment rooms date: 2021-02-08 journal: nan DOI: 10.1101/2021.02.04.21251153 sha: dc4d13d58f377c3e68e6af07fdee47e42be56241 doc_id: 271826 cord_uid: 8nec3b4b Objectives: The purpose of the present study was to monitor and evaluate CO2 levels in dental operatories using a consumer-grade CO2 sensor and determine the utility and accuracy of various methods using CO2 levels to assess ventilation rate in dental clinics. We aim to find a practical tool for dental practitioners to conveniently and accurately monitor CO2 levels and assess the ventilation rates in their office in order to devise a pragmatic and effective strategy for ventilation improvement in their work environment. Methods: Mechanical ventilation rate in air change per hour (ACHVENT) of 10 dental operatories was first measured with an air velocity sensor and air flow balancing hood. CO2 levels were measured in these rooms to analyze the effects of ventilation rate and number of persons in the room on CO2 accumulation. Ventilation rates were estimated using natural steady state CO2 levels during dental treatments and experimental CO2 concentration decays by dry ice or mixing baking soda and vinegar. We compared the differences and assessed the correlations between ACHVENT and ventilation rates estimated by steady states CO2 model with low (0.3 L/min, ACHSS30) or high (0.46 L/min, ACHSS46) CO2 generation rates, by CO2 decay constants using dry ice (ACHDI) or baking soda (ACHBV), and by time needed to remove 63% of excess CO2 generated by dry ice (ACHDI63%) or baking soda (ACHBV63%). Results: ACHVENT varied from 3.9 to 35.0 with a mean of 13.2 ({+/-}10.6) in the 10 dental operatories. CO2 accumulation occurred in rooms with low ventilation (ACHVENT [≤]6) and more persons (n>3) but not in those with higher ventilation and less persons. ACHSS30 and ACHSS46 correlated well with ACHVENT (r=0.83, p=0.003), but ACHSS30 was more accurate for rooms with low ACHVENT. Ventilation rates could be reliably estimated using CO2 released from dry ice or baking soda. ACHVENT was highly correlated with ACHDI (r=0.99), ACHBV(r=0.98), ACHDI63%(r=0.98), and ACHBV63% (r=0.98). There were no statistically significant differences between ACHVENT and ACHDI63% or ACHBV63%. Conclusions: Dental operatories with low ventilation rates and overcrowding facilitate CO2 accumulations. Ventilation rates could be reliably calculated by observing the changes in CO2 levels after a simple mixing of household baking soda and vinegar in dental settings. Time needed to remove 63% of excess CO2 generated by baking soda could be used to accurately assess the ventilation rates using a consumer-grade CO2 sensor and a basic calculator. Risks of disease transmission in healthcare settings during infectious disease pandemics have consistently challenged dental care professionals (DCPs) in their efforts to maintain a safe environment for their staffs and patients. DCPs have gained tremendous experiences in infection control from the ongoing HIV/AIDS pandemic by implementing universal or standard precautions against contact and droplet transmissions, but we are less confident in dealing with an infectious respiratory disease that may be transmitted through aerosol particles emitted by patients who have no overt symptoms. With mounting evidence that COVID-19 is transmissible through aerosols in an indoor environment [1, 2] , additional preventive measures beyond the standard care using personal protective equipment (PPE) are essential to minimize risks and alleviate anxieties experienced by staff and patients due to uncertainties associated with a novel infectious respiratory disease pandemic. Engineering controls through mechanical ventilation are important mechanisms to reduce the risks of airborne disease transmission in an indoor environment such as the dental offices. Though CDC recommends improving ventilation and air filtration in its guidance for dental settings during the COVID-19 pandemic [3] , few information is available on how to assess the ventilation condition and what measures to take to achieve more effective engineering control of disease transmission in dental offices. environments. CO2 is a byproduct of human metabolism and exists in high levels in exhaled air. Atmospheric CO2 level is at approximately 400 parts per million (ppm) in most outdoor environments, but CO2 in human exhaled air reaches on average 40,000 ppm in concentration [4] . CO2 levels is therefore often used as a proxy for indoor air quality as well as a risk marker for transmission of airborne diseases since it is inert, its indoor emission source (human) is known, and its measurement is inexpensive and accurate [5] ). Accumulation of CO2 may occur in indoor spaces with poor ventilation and overcrowding, and consequentially increase the risks of disease transmission, which is of particular significance under the current COVID-19 pandemic. High levels of CO2 in indoor environments have been associated with the transmission of infectious respiratory diseases such as tuberculosis, influenza and rhinovirus infections [6] [7] [8] . Indoor CO2 levels have therefore been widely used to model the risks of airborne infectious disease transmission [4, 9, 10] , including that of coronavirus infection transmission in dental offices [11] . CO2 levels is directly associated with ventilation rate in an occupied space. A clinical space with good ventilation should have a CO2 level close to that of outside air at approximately 400 ppm in most areas [4, 12] . Higher indoor CO2 levels indicate poor ventilation, accumulation of exhaled air, and increase in the fraction of "rebreathed air" or "shared air" in the indoor environment, which is proven to be a risk factor for transmissions of respiratory infectious diseases [4, 7, 8, 10] . CO2 levels have been used to estimate ventilation rates in dental offices through mathematic modeling [13] [14] [15] . Godwin and colleagues reported that the ventilation rate was 1.12 air change per hour (ACH) in a typical dental clinic in the US [13] , and Helmis and colleagues found that it was on average 5 ACH in a dental school clinic in Greece where the doors and windows were opened for cross ventilation [15] . Both studies used natural build-up of CO2 levels in the dental clinic to estimate the ventilation rate through mathematic models, but none of them actually verified the ventilation estimates using conventional methodologies such as a highprecision airflow sensor [16] . It is not known if these estimates were accurate as both methods require mathematical modeling based on several assumptions related to CO2 generation, buildup and dispersion over a relatively lengthy period of time, which may result in erroneous estimates if any of the assumed conditions are not met [5] . A simpler and more reliable method is needed if CO2 level is to be used by dental practitioners to assess the ventilation rate of their treatment rooms. The purpose of the present study was two-folds: 1) to monitor and evaluate CO2 level and its associated factors in dental operatories using a consumer-grade CO2 sensor, and 2) to determine the utility and accuracy of various methods that use CO2 levels to assess ventilation rate in dental clinics. Our aim was to find a practical tool that will enable all dental practitioners to conveniently and accurately monitor CO2 levels and assess the ventilation rates in their treatment rooms in order to devise a pragmatic and effective strategy for ventilation improvement in their work environment. Study settings: We conducted the CO2 concentration and ventilation rate assessments in 10 treatment rooms ranging from 667 to 1221 cubic feet (ft 3 ) in sizes. Mechanical ventilations of the treatment rooms are provided by three separate air handlers that drew 30%, 50% and 50% outside air to the ventilation systems. The outside air percentage were later increased to 60% for all three air handlers as part of our institutional response to COVID-19. Determining room airflow and mechanical ventilation rates: The mechanical ventilation of the 10 selected dental treatment rooms was measured with an air velocity sensor integrated in an air All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Instruments Inc., Scottsdale, AZ) as described elsewhere [17] . Briefly, the volumetric airflow rates of the dental treatment rooms were measured in cubic feet per minute (CFM, or ft 3 /min) at both the air supply inlets and air exhaust returns using the air flow balancing hood. The mechanical ventilation rates of each space in number of ACH was calculated based on supply and exhaust airflow rates for each room. The larger value between ACH from supply air and ACH from return air was used as the room's mechanical ventilation rate (ACHVENT) [18] . Assessing CO2 levels during dental treatment procedures: We measured CO2 levels in two dental treatment rooms when dental procedures were performed. The two rooms represented two extremes in ventilation rates, with one at the lowest at 3.9 ACH, and the other at the highest at 35 ACH. The number of persons in the rooms were recorded in real time when a person was entering and leaving the room. One of the two rooms were used for dental implant therapy and may often have more than one graduate students observing the procedures in addition to the treating dentist, the dental assistant and the patient in the room. 24-hour Continuous monitoring CO2 in dental treatment rooms: To further explore the dynamics of CO2 levels in dental treatment rooms throughout the work day and assess of accuracies of the steady state models of CO2 for ventilation assessments, we continuously measured the CO2 levels in the 10 selected dental treatment rooms for 24 hours and recorded the procedures performed and number of persons in the room. Assessing ventilate rate by natural CO2 level modeling in dental treatment rooms: We used the steady state model described by Batterman [5] to calculate the ACH of the treatment rooms and compared the outcomes with ACHVENT determined by mechanical ventilation. The steady state air change rate (ACHSS) is calculated as follows [5] : The CO2 generation rate GP is affected by many factors and may vary by human activity, physical size, sex and race [19, 20] . GP values of 0.46 L/min and 0.30 L/min have been used in previous studies to represent CO2 generation by moderately active adults [5] , [13] ). As the CO2 generation rates and activity levels by dental care providers and their patients are unknown and may not be constant, we decided to use both values to calculate the steady state air change All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted February 8, 2021. ; https://doi.org/10.1101/2021.02.04.21251153 doi: medRxiv preprint rate ACHSS and assess the correlations between ACHSS and ACHVENT at two GP levels (0.46 and 0.30 L/min, respectively). Assessing ventilation rates by CO2 decays using dry ice: Ventilation rates were also measured in the 10 dental treatment rooms using the CO2 decay method [5] . Outside air CO2 level was first measured for 5 minutes near the air intake of the ventilation system outside the building before each experiment. To raise the peak CO2 levels inside the dental treatment rooms to approximately 2000 ppm, 250g of dry ice were placed in a water bath and left in the room for two minutes. A small oscillating fan was used to keep the CO2 well mixed in the room. CO2 level was then measured at one-minute interval using an Aranet4 CO2 sensor (range 0-9999ppm, accuracy ±50ppm, SAF Tehnika, Riga, Latvia) for up two hours. The ventilation rate by CO2 clearance using dry ice (ACHDI) were determined as described by Batterman [5] using the CO2 concentration decays : where Δt = period between measurements; C0 and C1 = CO2 levels measured at the beginning and the end of the decay period (ppm), and CR = CO2 level in outdoor air (ppm). The consumer-grade Aranet4 CO2 sensor used in the current study was purchased at amazon.com in the US. It was recommended by the Federation of European Heating, Ventilation and Air Conditioning Associations (REHVA) for monitoring CO2 levels in schools during the COVID-19 pandemic [21] , and was found to be comparable to a research-grade LI-COR CO2 sensor in accuracy and suitable for the time-response assessment in this study [22] . Assessing ventilation rates by CO2 decays using baking soda: Considering that dental practitioners in private practices may not have ready access to dry ice, we developed a method to rapidly generate CO2 in dental treatment rooms using household baking soda (Arm& Hammer Pure baking soda, Church & Dwight Co., Inc., Ewing, NJ, USA) and vinegar (Heinz all-natural distilled white vinegar, Pittsburgh, PA, USA). Mixing baking soda (NaHCO3) with vinegar containing 5% acetic acid (CH3COOH) will generate CO2 as follows: NaHCO3 + CH3COOH → CH3COONa + H2O + CO2 (#3) We tested peak CO2 values in the 10 treatment rooms using a weight (g):volume (ml) ratio of 1:15 based on the molar masses of the reagents. We aimed at a peak level range of 1500 to 2000 ppm in rooms with various mechanical ventilation rates and with doors closed. We determined that adding about 125g of baking soda (approximately 3/5 cup measure) to 1893 ml (a 64-oz bottle) of vinegar containing 5% acetic acid will elevate the CO2 level in a typical dental All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted February 8, 2021. ; https://doi.org/10.1101/2021.02.04.21251153 doi: medRxiv preprint treatment room (10 x 11 ft in area, 8 ft in ceiling height, or 880 ft 3 in volume) with a moderate ventilation rate (ACHVENT = 4) to above 1500 ppm. For rooms that are significantly larger or having very high ventilation rates, one full cup measure (about 8 oz or 227g) of baking soda may be used with 3785 ml (a one-gallon jar) of vinegar containing 5% acetic acid. Before the baking soda experiment, outdoor CO2 level was first measured for 5 minutes near the air intake of the ventilation system outside the building. To generate CO2 inside the dental treatment rooms, we first poured vinegar into a large container and added the baking soda powder. The mixture was vigorously shaken or stirred with a large spatula for two minutes and then removed from the room. A small oscillating fan was used to keep the CO2 well mixed in the room with doors closed. CO2 level was then measured at one-minute interval using the CO2 sensor as described above. The length of the CO2 measurement period is dependent on the ventilation rates of the room. For rooms with excellent ventilation, CO2 level decays rapidly and a measurement period of 30 minutes may be adequate. For rooms with very poor ventilation, CO2 measurements may need to continue for several hours or overnight to allow a sufficient level of CO2 clearance for calculating the ventilation rate using the baking soda and vinegar method (ACHBV) based on equation #2 described above. Estimating ventilation rate by time to 63% removal of excess CO2: Based on the commonly used formula for the rate of purging airborne contaminants, one complete air change will replace 63% of the airborne contaminants in the room with outdoor air [22] [23] [24] . Ventilation rate can therefore be simply calculated using the time needed to reach 63% reduction of excess CO2 from its peak level: where t1 = initial timepoint in minutes with indoor CO2 at peak level, and t2 = timepoint in minutes when excess CO2 is reduced by 63%. Indoor CO2 at peak level (CS) is the sum of outdoor CO2 (CR) and excess CO2 (CE) generated by dry ice or baking soda. As CO2 measurement starts at peak level in this experiment, t1 is therefore always 0. Time needed to remove 63% CE, or t2, is the timepoint when indoor CO2 level is at C63%E = CS -63% CE, where CE = CS -CR. Statistical analysis: To understand factors associated with CO2 levels in dental treatment rooms, we performed multiple regression analysis using CO2 levels as dependent variable and number of persons in the room, ventilation rate, room sizem and outdoor CO2 level as independent variables. We analyzed the dynamics of CO2 levels during dental treatment procedures using descriptive analysis and compared the differences in steady state CO2 levels between rooms All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted February 8, 2021. ; https://doi.org/10.1101/2021.02.04.21251153 doi: medRxiv preprint with poor and good ventilations using t-tests. ACHVENT was compared with ACH calculated with different methods based on CO2 levels to assess the correlation (Pearson's r) between the two methods of ventilation rate assessments. As the primary goal of the present study was to determine the validity of pragmatic methodologies that could be used by dental practitioners who do not have access to sophisticated instruments to assess the ventilation rate of their clinic spaces, we focused the analysis on CO2 generation using baking soda and vinegar and ventilation rate calculation using time needed to achieve 63% reduction of peak CO2 levels (ACHBV63%). Mechanical ventilation rate of the dental treatment rooms: The volumetric sizes, airflow rates and mechanical ventilation rates of dental treatment rooms are presented in Table 1 . The 10 operatories are on average 882 ft 3 in volume (range 667 to 1221 ft 3 ). ACHVENT varied from 3.9 to 35.0 with a mean of 13.2 (±10.6). A majority (7/10) of the treatment rooms have greater supply than exhaust air flow rates with positive differentials between ACHs and ACHE at two or greater; and one room have significantly greater exhaust than supply airflow rates with a highly negative differential between ACHs and ACHE. A majority of the treatment rooms (7/10) have ACHVENT at or greater than 6, and 4 of the 10 rooms had ACHVENT greater than 10. CO2 levels during dental treatment procedures: As shown in Figure 1 , CO2 levels were significantly higher in the room with low ventilation rate (ACH = 3.9) and reached a peak of nearly 1600 ppm when 6 persons were in the room. The increased number of persons were related to teaching activities involving dental implant surgery where additional graduate students were allowed to observe the procedures. Comparing the two rooms with the same number of persons in the room for the same restorative procedures, CO2 levels reached 1100 ppm at the peak in the room with 3.9 ACH but stayed below 700 ppm in the room with 35 ACH (p<0.0001). CO2 accumulation appeared to be associated with crowding and low ventilation rate. Continuous CO2 monitoring in dental treatment rooms: We continuously monitored the CO2 levels for 24 hours in 10 treatment rooms with various ventilation rates in different departments. The working hours were from 8:00am to 5:00pm with lunch breaks from 12:30pm to 1:30pm for some and 1:00pm to 2:00pm for other departments. The dental procedures included prophylaxis, extractions, restoratives, endodontics, dental implant surgery, periodontal surgery and exams. Number of persons in the rooms varied from 2 to 6, with more people in the room during dental implant surgeries. The CO2 levels in early morning (5:00 -7:00am) were at an level of 421±10 ppm, similar to the outdoor levels (413±15 ppm) (Table 2, Figure 2 ). The steady All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted February 8, 2021. ; https://doi.org/10.1101/2021.02.04.21251153 doi: medRxiv preprint state CO2 level (CSS) during dental procedures, which was the mean concentrations of CO2 at the plateau level when the number of persons in the room stays unchanged for at least 5 min, ranged from 543 ppm to 1,374 ppm (786±207 ppm) in the 10 dental treatment rooms. Multiple regression analysis showed that steady state CO2 levels in dental treatment rooms were statistically significantly correlated to number of persons in the room (b=90.2, p=0.006), mechanical ventilation rate (b=11.0, p=0.001) and the volumetric size of the room (b=-0.50, p=0.049), but not to outside air CO2 levels (b=4.15, p=0.160). These findings confirmed that more people in smaller room with low ventilation rate facilitate CO2 accumulation in dental treatment rooms. As shown in Figure 2 , CO2 levels in rooms with ACH higher than 6 rarely reach 800 ppm. In rooms with ACH lower than 6, however, the CO2 levels were consistently greater than 800 ppm and approached 1,400 ppm in a room with ACH 3.9 when number of persons in the room was as high as 6 during clinical teaching activities related to dental implant procedures. Ventilation rates by steady state CO2 level modeling in dental treatment rooms: Based on steady state CO2 level (CSS) and outdoor CO2 level (CR) presented in Table 2 , ventilation rates with CO2 generation at 0.30 L/min (ACHSS30) and 0.46 L/Min (ACHSS46) in the clinical space were calculated using equation #1 for two dental treatment procedures in each room. As expected, ACHSS30 values (Mean 8.3, SD 3.7) were significantly lower than ACHSS46 (Mean 12.6, SD 5.7) (mean difference = -4.5, paired t=-6.97, p<0.0001). Both ACHSS30 and ACHSS46 were similarly well correlated with ACHVENT (r=0.83, p=0.003). ACHSS30 approximated closely to the mechanical ventilation rates in rooms with ACHVENT £ 6 (mean difference=-0.6, paired t=-1.24, p=0.304), but significantly underestimated those in rooms with ACHVENT > 6 (mean difference=-8.7, paired t=-2.59, p=0.049). The opposite is true for ACHSS46, it significantly overestimated the ventilation rates in rooms with ACHVENT £ 6 (mean difference=3.7, paired t=6.78, p=0.007), but approximated closer to those in rooms with ACHVENT > 6 (mean difference=-3.5, paired t=-1.14, p=0.307) ( Table 3) . As ventilation rate is likely below 6 ACH in most dental treatment rooms in private dental practices in small free-standing buildings in the US [13] , CO2 generation rate (GP) of 0.30 L/min is more appropriate for ventilation rate estimate using equation #1. We listed the corresponding steady state CO2 levels and ventilation rates in rooms with various sizes with 8-ft ceiling height using n = 3, GP = 0.30 L/min, and CR= 400 ppm (Supplemental Table 1 ). DCPs may use the steady state CO2 levels measured in their dental treatment rooms to roughly estimate the ventilation rate using this table. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Ventilation rates by CO2 decays using dry ice or baking soda: Results of ventilation estimates by CO2 decay using dry ice or baking soda and vinegar are shown in Table 3 . Both methods appeared to be reliable in assessing the ventilation rates of dental treatment rooms. The CO2 decay curves demonstrate that CO2 levels decreased faster over time in rooms with high ACHVENT (Figure 3 A, B) . ACHDI values ranged from 3.6 to 28.5 (11.2±8.4) and were highly correlated with the mechanical ventilation rates ACHVENT (r=0.99, p<0.0001) (Figure 3 C) . ACHDI was lower than ACHVENT (mean difference = 2.0, paired t=2.32, p=0.046). Similarly, ACHBV ranged from 4.7 to 27.0 (11.8±8.1) and also correlated highly with ACHVENT (r=0.98, p<0.0001) (Figure 3 D) . There was no statistically significant difference between ACHBV and ACHVENT (mean difference = 1.4, paired t=1.48, p=0.174), or between ACHBV and ACHDI (mean difference = 0.59, paired t=1.26, p=0.239). Ventilation rates by time to 63% removal of excess CO2 generated by dry ice or baking soda: Ventilation rates calculated by time needed to remove 63% of excess CO2 generated by dry ice of baking soda are presented in Table 3 . ACHDI63 values ranged from 4.1 to 30.8 (11.5±8.6) and were highly correlated with the mechanical ventilation rates ACHVENT (r=0.98, p<0.0001) (Figure 4 A). There was no statistically significant difference between ACHDI63 and ACHVENT (mean difference = 1.8, paired t=1.93, p=0.086). Similarly, ACHBV63 ranged from 4.6 to 27.3 (11.9±8.1) and correlated highly with ACHVENT (r=0.98, p<0.0001) (Figure 4 B) . There was no statistically significant difference between ACHBV63 and ACHVENT (mean difference = 1.3, paired t=1.34, p=0.213), or between ACHBV63 and ACHDI63 (mean difference = 0.50, paired t=0.76, p=0.467). A Microsoft Excel template is provided in Supplemental Table 2 that will allow DCPs to calculate the ventilation rate of their offices by inputting the values of peak CO2 level (CS), outdoor CO2 level (CR) and time (min) needed to reach 63% removal of excess CO2 generated by dry ice or baking soda. The findings of the present study indicate that dental operatories with low mechanical ventilation rates and overcrowding facilitate CO2 accumulations. Ventilation rates could be measured by assessing natural or experimental build-up of CO2 levels in dental treatment rooms using a consumer-grade CO2 sensor. We found that ventilation rates in ACH could be accurately assessed by observing the changes in CO2 levels after a simple mixing of household baking soda and vinegar in dental settings. Time needed to remove 63% of excess CO2 generated by All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted February 8, 2021. ; https://doi.org/10.1101/2021.02.04.21251153 doi: medRxiv preprint baking soda could be used to accurately calculate the ventilation rates with the help of a basic calculator. Our findings show that CO2 level may consistently stay above 800 ppm in rooms with ventilation rates below 6 ACH, especially when three or more persons (including the patient who was not wearing a mask) are in the room during dental treatments. We observed that CO2 level stayed above 1000 ppm and approached 1600 ppm when 3 to 6 persons were in a room with 3.9 ACH in clinical teaching scenarios involving dental implant treatment. High levels of CO2 indicate high concentrations of respiratory aerosols in the room. It is possible that these aerosols contain pathogens if the patient is not wearing a mask and is infected but asymptomatic or presymptomatic. Effective mitigation measures will be required in these rooms to improve the air quality even without the ongoing infectious disease pandemic. Overcrowding should be avoided in rooms with poor ventilation. In dental operatories with ventilation rates higher than 10 ACH, the CO2 levels stayed consistently below 700 ppm in most cases with 3 persons in the room. Our data demonstrated a clear dependency of CO2 levels on number of persons in the room and the mechanical ventilation rate. CO2 level is a proxy for indoor air quality as it represents the fraction of rebreathed air, or the proportion of inhaled air that was exhaled by others in the same indoor environment. Though numerous epidemiological studies indicate CO2 begins to have negative health effects at 700 ppm and respiratory symptoms may occur when indoor CO2 concentration is above 1000 ppm [25] , our main concern is the concurrent accumulation of respiratory aerosols that may contain infectious disease pathogens. Numerous studies have shown that exhaled air from infected patients contains respiratory disease pathogens, including rhinovirus, influenza virus and Mycobacterium tuberculosis [4, [26] [27] [28] . Patients with early stages of COVID-19 may release millions of SARS CoV-2 viral copies per hour in exhaled air [29] . In a recent study that modeled factors associated with the spread of respiratory infectious disease in dental offices, CO2 levels were found to play the most important role on the risk of infectious disease transmission [11] . CO2 levels at 774 ppm was considered as low risk but those at or above 1135 ppm may increase the risk of disease transmission in dental offices [11] . It is important to point out that the setting for the current study is a postdoctoral dental training institution affiliated with an academic medical center, which may differ significantly in ventilation conditions from private dental practices that have a solo or a few dental practitioners. Ventilation conditions in different dental settings are largely unknown as the ventilation design of dental offices is not regulated as other outpatient healthcare facilities that are required to have 6 to 15 ACH by ASHRAE and CDC [30, 31] . Godwin and colleagues reported that ventilation rate was All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted February 8, 2021. ; https://doi.org/10.1101/2021.02.04.21251153 doi: medRxiv preprint 1.12 ACH in dental operatories of a small dental clinic (2400 ft 2 ) [13] , which is significantly lower than the mean of 13 ACH in the present study but resembles more closely to the mean of 1.09 ACH in typical residential households in the US [32] . Accurate and reliable measurements of ventilation rate in various dental settings are important for risk assessment and for risk mitigation planning in an era of frequent infectious disease pandemics. Mechanical ventilation rate is assessed by quantifying the amount of outdoor air flowing into and out of an indoor space using highly sophisticated instruments operated by trained professionals [16] . Technical barriers may have contributed to the scarcity of information regarding ventilation in dental settings. Besides direct air flow measurements, ventilation rate could be estimated using CO2 as a tracer gas. CO2 in an indoor space could be built up to a significantly higher level than in outdoor air, either through natural generation by the occupants or through experimental release of the gas in the indoor spaces [5, 33, 34] . Analysis of the steady state CO2 levels or the rate of CO2 concentration decays, which is directly dependent on the outdoor air flow rate from the ventilation system, will allow an estimate of the ventilation rate of the indoor space. We found that modeling the steady state CO2 levels using equation #1 correlated reasonably well with the mechanical ventilation rate, but may either under-or overestimate the ventilation rate based on different assumptions of human CO2 generation rates. In comparison, the CO2 concentration decay method relied on actual CO2 levels measured at the beginning and the end of a decay period (equation #2) and provided more accurate assessments and better approximation to the mechanical ventilation rates. CO2 concentrations in dental operatories could be built up to a level of about 1500 to 2500 ppm in 2 minutes using either dry ice or baking soda and vinegar. CO2 decays could then be monitored using a CO2 sensor that logs data in one-minute intervals. There are many affordable consumer-grade CO2 sensors readily available and suitable for the purpose of observing CO2 level changes over a period of time, from several minutes to several hours depending on the mechanical ventilation rates. The CO2 sensor used in the present study was purchased online for $159 and appeared to be a reliable tool for monitoring indoor CO2 levels in dental settings. Although ventilation rate in ACH could be calculated by fitting a linear regression line over time into the natural log scale of time-varying concentrations of CO2 levels (equation #2), we found that a simplified method (equation #4) provided equally if not more accurate estimate of ventilation rate. As it is known that one complete air change will replace 63% of the airborne contaminants in the room with outdoor air [22] [23] [24] , ventilation rate could be easily calculated using the time needed to remove 63% of excess CO2 as a contaminant. For example, assuming All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted February 8, 2021. ; https://doi.org/10.1101/2021.02.04.21251153 doi: medRxiv preprint CO2 level in outdoor air is 400 ppm, and peak CO2 level is 1500 ppm after placing dry ice or baking soda inside the dental office for 2 minutes, excess CO2 inside the dental office will be 1500 -400 = 1100 ppm at peak. The CO2 level that represents 63% removal of excess CO2 is therefore 1500 -63% x1100 = 807 ppm. If it takes 15 min for the CO2 level to reach 807 ppm from the peak of 1500 ppm, ventilation rate in ACH will be 60/15 = 4; and if it takes 2 hours for the CO2 level to reach 807 ppm, ACH will be 60/120 = 0.5. This method will allow dental practitioners to accurately estimate the ventilation rate in ACH using a simple calculator. Ventilation rates measured by the CO2 decay methods were approximately 15% lower on average than those measured by air flow sensors at the air supply inlets or exhaust returns of the ventilation system. This discrepancy is expected as indoor CO2 could only be removed by fresh outdoor air brought in by the ventilation system or by outdoor air infiltrated through leaky doors and windows [35] . As the mechanically ventilated air was not composed of 100%, but 60% outdoor air at the time of this study, the CO2 removal rate was understandably lower than the total air flow rate. In practical sense, ventilation rates measured by CO2 decay methods are better indicators of outdoor air flow, which is more important than the recycled indoor air flow measured by airflow sensors in terms of air contaminant removal efficiencies. Our data showed that household baking soda (NaHCO3) and vinegar (5% acetic acid) could be used to generate CO2 in dental office to assess the ventilation rate by observing the CO2 concentration decays using a CO2 sensor and a basic calculator. This method will allow dental practitioners to reliably estimate the ventilation rate in their dental offices without expensive equipment and advanced technical skills. The test could be completed within 30 minutes in spaces with ventilation rate higher than 2 ACH but may take longer time if the ventilation is significantly below 1 ACH. We recommend to plan a two-hour observation time during off-hours with the building ventilation system operating in its normal setting. We consider that it is very important for every dental practitioner to be able to accurately assess the ventilation rate in their working environments. Epidemiological data showed that transmission of COVID-19 is almost exclusively an indoor phenomenon, with 99.97% of the transmissions occurring in an indoor environment [29] . Airborne transmission through respiratory aerosols is increasingly recognized as a major driver for the COVID-19 pandemic [36] [37] [38] . As essential healthcare providers, dental professionals work in the frontline during the pandemic and need to adopt measures to mitigate the risk of aerosol transmission in addition to droplet and contact precautions that have been the standard of infection control in dental offices [39] . All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. We recommend that mitigation measures be taken for dental treatment rooms that have a ventilation rate below 15 ACH, which is required for procedure rooms in outpatient healthcare facilities by CDC guidelines [30] . While in theory the most effective measure for air quality improvement in dental offices is to increase outdoor air flow rate through the mechanical ventilation system or through natural ventilation by opening doors and windows, such measure is severely limited by the weather or climate conditions and often impractical or impossible to realize in practice. An effective alternative is to improve air filtration using upgraded filters in the ventilation system and portable air cleaners (PAC) equipped with high efficiency particulate air (HEPA) filters. We showed that a PAC with a HEPA filter and a clean air delivery rate of 250 cubic feet per minute could add an equivalent of 17 ACH to an average sized dental operatory (10 ft L, 11ft W, 8 ft H) [17] . The PAC was especially effective in removing aerosols from rooms with low mechanical ventilation rate because of the lack of disturbance from the high air flow of the ventilation system [17] . For larger spaces where PAC is not practical or effective, upper room ultraviolet germicidal irradiation (UVGI) system may be considered as it has been shown to be effective in healthcare settings [40] [41] [42] . The upper room UVGI system requires professional installation and maintenance to ensure its safe operation. In summary, we found that dental operatories with low ventilation rates (ACHVENT £ 6) facilitate CO2 accumulations. Crowding inside the room contributed to elevated CO2 levels. Ventilation rates could be reliably calculated by observing the changes in CO2 levels after a simple mixing of household baking soda and vinegar in dental settings. Time needed to remove 63% of excess CO2 generated by baking soda could be used to accurately assess the ventilation rates using a consumer-grade CO2 sensor and a basic calculator. For rooms with ventilation rate below 15 ACH, we suggest the addition of a PAC with HEPA filter and proper clean air delivery rate to facilitate air quality improvement in dental treatment rooms. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted February 8, 2021. ; https://doi.org/10.1101/2021.02.04.21251153 doi: medRxiv preprint Figure 1 : CO2 levels during dental treatment procedures in operatories with low and high mechanical ventilation. A: Significant CO2 accumulation occurred in a room with low ventilation (ACH = 3.9) and multiple persons in the room during clinical teaching activities for dental implant surgery. E, F: CO2 level is associated with ventilation rate in rooms with the same number of persons. Peak CO2 level reached 1100 ppm in the room with 3.9 ACH (B) but stayed under 700 ppm in the rooms with 35 ACH (C). All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted February 8, 2021. ; https://doi.org/10.1101/2021.02.04.21251153 doi: medRxiv preprint Figure 2 : 24-hour continuous measurements of CO2 levels in 10 dental treatment rooms with various ventilation rates. CO2 accumulation occurred in rooms with lower ventilation rates (ACH £ 6). CO2 levels stayed under 800 ppm in rooms with higher ventilation rate and lower number of persons. CO2 level in nonworking hours is close to that of outdoor at 400 ppm in all the rooms. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. 16.2 DI: dry ice. BV: baking soda and vinegar. CR: outdoor CO2 level. CS: peak CO2 level after CO2 generation by dry ice or baking soda. CE: excess CO2 generated by dry ice or baking soda (CS -CR). C63%E: CO2 level after 63% excess CO2 is removed (CS-63%CE). t2: time (min) needed to reach C63%E. ACHT63%: ventilation rate in air change per air based on t2 All rights reserved. No reuse allowed without permission. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted February 8, 2021. ; https://doi.org/10.1101/2021.02.04.21251153 doi: medRxiv preprint Using steady state CO2 level during dental treatments to estimate ventilation rates: The following table could be used to roughly estimate the ventilation rate of the dental treatment rooms in 3 steps. First, measure the length and width of the room to get the area in square feet (ft 2 ). Second, determine the steady state CO2 level during a dental treatment procedure that lasts more than 10 minutes as follows: with the dentist, dental assistant and the patient together in the room and without any person entering or leaving the room, read the CO2 sensor readings 10 minutes into the procedure and record the next 5 readings at 1 min interval, add the 5 readings to get the sum and divide the sum by 5, the result is the steady state CO2 level. Third, match the steady state CO2 level to the closest number under the area column of your room size, the number in the ACH column in the same row is the ventilation rate estimate. For example, if the CO2 level reaches a steady state level of about 1062 ppm during a dental treatment that lasted longer than 5 minutes with 3 persons in a room that is 110 ft 2 (10-ft W x11ft L) in area, the ventilation rate is about 3 ACH. It will be about 6 ACH if the CO2 level stays at about 761 ppm (Supplemental Table 1 ). Table 1 : Steady state CO2 levels and ventilation rate in air change per hour (ACH)* Review and Extension of CO₂-Based Methods to Determine Ventilation Rates with Application to School Classrooms How to quantify the ventilation rate of an indoor space using an affordable CO2 monitor, Medium Aerobiology and its role in the transmission of infectious diseases Airborne infection. Theoretical limits of protection achievable by building ventilation CO 2 at peak 2800 ppm B1. CS = Peak CO2 level CS CO 2 outdoors 400 ppm B2. CR = Outdoor CO 2 level CR Excess CO 2 2400 ppm B3. CE = Excess CO2. CE = CS-CR After removal of 63% excess CO2 1288 ppm B4. C63% = CO2 level after remova Time needed to remove 63% excess CO2 15.0 min B5. Time to reach C63% (Value in B Ventilation rate 4.0 h -1 B6. Air change per hour, or ACH To calculate air change per hour, 3 values need to be entered into the above template: 1. CS: this is the CO2 level when the measurement starts after CO2 generation by mixing baking soda and vineg 2 Time needed to reach 63% removal of excess CO2. You need to read the value in B5 and check the CO2 sensor readings to get the time needed to reach 63% remo References How COVID-19 spreads SARS-CoV-2 and Potential Airborne Transmission Guidance for Dental Settings -Interim Infection Prevention and Control Guidance for Dental Settings During the COVID-19 Pandemic Modelling the risk of airborne infectious disease using exhaled air Review and Extension of CO₂-Based Methods to Determine Ventilation Rates with Application to School Classrooms Detection of airborne rhinovirus and its relation to outdoor air supply in office environments Shared air: a renewed focus on ventilation for the prevention of tuberculosis transmission Quantification of Shared Air: A Social and Environmental Determinant of Airborne Disease Transmission An estimation of airborne SARS-CoV-2 infection transmission risk in New York City nail salons Risk of indoor airborne infection transmission estimated from carbon dioxide concentration Modeling of the Transmission of Coronaviruses, Measles Virus, Influenza Virus, Mycobacterium tuberculosis, and Legionella pneumophila in Dental Clinics Ventilation for Acceptable Indoor Air Quality Indoor environment quality in dental clinics: potential concerns from particulate matter Emissions of total volatile organic compounds and indoor environment assessment in dental clinics in Indoor air quality in a dentistry clinic RA 2017) Measurement, Testing, Adjusting and Balancing of Building HVAC Systems., ASHRAE Effects of mechanical ventilation and portable air cleaner on aerosol removal from dental treatment rooms What does air change mean?, Isolation Rooms: Design, Assessment, and Upgrade. , Francis J Curry National Tuberculosis Center Carbon dioxide generation rates for building occupants CO2 generation rate in Chinese people REHVA -Federation of European Heating Ventilation and Air Conditioning Associations, How to operate HVAC and other building service systems to prevent the spread of the coronavirus disease (COVID-19) in workplaces How to quantify the ventilation rate of an indoor space using an affordable CO2 monitor Aerobiology and its role in the transmission of infectious diseases Airborne infection. Theoretical limits of protection achievable by building ventilation Effects of low-level inhalation exposure to carbon dioxide in indoor environments: A short review on human health and psychomotor performance An optimized method to detect influenza virus and human rhinovirus from exhaled breath and the airborne environment Viable influenza A virus in airborne particles expelled during coughs versus exhalations Influenza virus RNA recovered from droplets and droplet nuclei emitted by adults in an acute care setting Indoor transmission of SARS-CoV-2 Guidelines for environmental infection control in health-care facilities; recommendations of CDC and Healthcare Infection Control Practices Advisory Committee (HICPAC) New Ventilation Guidelines for Health-Care Facilities Residential ventilation and energy characteristics Air Change Rate Measurements using Tracer Gas Carbon Dioxide from Dry Ice Assessing general ventilation effectiveness in the laboratory Simple and Cheap Air Change Rate Measurement Using CO2 Concentration Decays It Is Time to Address Airborne Transmission of Coronavirus Disease 2019 (COVID-19) A systematic review of possible airborne transmission of the COVID-19 virus (SARS-CoV-2) in the indoor air environment Identifying airborne transmission as the dominant route for the spread of COVID-19 Standard and transmission-based precautions: an update for dentistry Impact of Environmental Factors on Efficacy of Upper-Room Air Ultraviolet Germicidal Irradiation for Inactivating Airborne Mycobacteria Efficacy of ultraviolet germicidal irradiation of upper-room air in inactivating airborne bacterial spores and mycobacteria in full-scale studies The characterization of upper-room ultraviolet germicidal irradiation in inactivating airborne microorganisms The authors declare no conflict of interests. This study is supported in part by the Eastman (ft 2 ) ACH 100 110 120 130 140 150 160 170 180 190 200 1 2784 2567 2386 2234 2103 1989 1890 1802 1724 1655 1592 2 1592 1484 1393 1317 1251 1195 1145 1101 1062 1027 996 3 1195 1122 1062 1011 968 930 897 867 841 818 797 4 996 942 897 858 826 797 772 751 731 714 698 5 877 833 797 767 741 718 698 680 665 651 638 6 797 761 731 706 684 665 648 634 621 609 599 7 741 710 684 662 643 627 613 600 589 579 570 8 698 671 648 629 613 599 586 575 566 557 549 9 665 641 621 604 589 577 566 556 547 539 532 10 638 617 599 583 570 559 549 540 532 525 519 11 617 597 581 567 555 544 535 527 520 514 508 12 599 581 566 553 542 532 524 517 510 505 499 13 583 567 553 541 531 522 515 508 502 497 492 14 570 555 542 531 522 514 506 500 495 490 485 15 559 544 532 522 514 506 499 493 488 484 479 Supplemental Table 2 is a dynamic template that will allow you to enter 3 values to get the ventilation rate in air change per hour (ACH) for your treatment rooms: 1, the peak CO2 level (CS), 2, the outdoor CO2 level, and 3. Time needed to reach 63% removal of excess CO2. Table 2 : Ventilation rate estimate using time needed to remove 63% excess CO2 generated by baking soda and vinegar