Coronavirus Disease 2019 Patients in Earlier Stages Exhaled Millions of Severe Acute Respiratory Syndrome Coronavirus 2 Per Hour

Abstract

Coronavirus disease 2019 (COVID-19) patients exhaled millions of severe acute respiratory syndrome coronavirus 2 RNA copies per hour, which plays an important role in COVID-19 transmission. Exhaled breath had a higher positive rate (26.9%, n = 52) than surface (5.4%, n = 242) and air (3.8%, n = 26) samples.

COVID-19, SARS-CoV-2, exhaled breath, airborne transmission

The coronavirus disease 2019 (COVID-19) pandemic has left a major mark on human history. Global efforts to stop the spread are accelerating. However, scientific information on the major routes of COVID-19 transmission is required. Analysis of environmental samples provides clues [1–4]. Notably, severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has been detected in air [2–4], on ventilation fans [1], and on hospital floors [1, 4]. Surface swabs from keyboards, cell phones, and patients’ hands have also tested positive [1]. Other studies have shown that aerosolized SARS-CoV-2 not only survives on various surfaces for sustained periods of time [5] but also remains viable in the air for up to 3 hours [6]. Despite these rapid developments, the key COVID-19 transmission routes remain debated [7], and evidence is extremely sparse on how SARS-CoV-2 is emitted into the air. Recently, scientists called for recognition of airborne transmission of COVID-19 [8], and the World Health Organization (WHO) made a change to their guidelines accordingly, that is, not excluding airborne transmission in crowed and closed settings. Here, we investigated the breath emission of SARS-CoV-2 from 49 COVID-19 patients recruited in Beijing, China, in addition to its environmental detection.

METHODS

We recruited 57 patients with COVID-19 and 4 patients without COVID-19 from hospital A and hospital B and 15 healthy individuals in Beijing (Supplementary Table 1). Exhaled breath condensate (EBC) samples were collected from 20 COVID-19 patients imported from Canada, France, Iran, Italy, Japan, Spain, Thailand, the United Kingdom, and the United States and from 29 local cases in Beijing (Supplementary Table 2). Supplementary Figures 1 and 2 show the intensive care unit (ICU) and general ward floor settings of hospital A, respectively. EBC samples were collected using a BioScreen device developed by Peking University (Supplementary Figure 3). A total of 52 EBC samples were collected from 49 COVID-19 patients (multiple samples were taken for certain patients) (Supplementary Table 2). EBC samples were also collected from 15 healthy controls. Twenty-six air samples were taken using 2 impingers (Supplementary Figure 4, Supplementary Table 3) as described in the Supplementary Materials. A total of 242 surface swabs (10 cm² or 25 cm²) from quarantine hotels and hospitals or from personal items of COVID-19 patients were obtained using wet cotton swabs (Supplementary Table 4). All samples collected were analyzed using reverse-transcription polymerase chain reaction (RT-PCR) (Roche 96 fluorescence qPCR instrument, Roche Molecular Systems, Inc, Pleasanton, CA) for SARS-CoV-2, targeting both ORF1ab and N genes using a detection kit (Jiangsu Bioperfectus Technologies, Nanjing, China). The quantitative viral loads in all samples was estimated using the RNA amplification equation; experimental and calculation details are provided in the Supplementary Materials. The ethics requirement involving human subjects, including the noninvasive collection of EBC samples, was waived due to the urgency of the infectious disease outbreak. The Center for Disease Control and Prevention of Chaoyang District of Beijing Ethics Committee approved the study.

RESULTS

The overall SARS-CoV-2 positive rate for EBC samples was 26.9% (n = 52), while surface swabs and air samples had low positive rates of (5.4%, n = 242) and (3.8%, n = 26), respectively (Table 1). Cycle threshold (Ct) values (35.54 ± 3.14) were obtained for each positive EBC sample (Table 1). The Ct values for EBC samples varied greatly among patients, with lower values generally detected during earlier disease stages (Supplementary Table 2). The breath emission rate was estimated to be from 1.03 × 10⁵ to 2.25 × 10⁷ viruses per hour (n = 14; Table 1). The detection kit had different amplification efficiencies for ORF1ab and N target genes in EBC samples (Supplementary Table 2). Although EBC samples from 2 patients (A and B) were shown to contain SARS-CoV-2 (Supplementary Table 2), surface swabs from their cell phones, hands, and toilet surfaces were negative for the virus (Supplementary Table 4). In the ward of patient C, the virus was present on the surface of an air ventilation duct entrance that was located below the patient’s bed (see Supplementary Video 1). In addition to causing air contamination, the exhaled SARS-CoV-2 could be partially responsible for the contamination that was observed on the surfaces.

Table 1.

Detection of Severe Acute Respiratory Syndrome Coronavirus 2 and Its Positive Rates from 52 Exhaled Breath Condensate Samples Collected from 49 Coronavirus Disease 2019 Patients, 26 Air Samples, and 242 Surface Swabs

	Exhaled Breath Condensate (n = 52)	Air Sample (n = 26)	Surface Swab (n = 242)
Sample SARS-CoV-2 positive rate	14/52 (26.9%)	1/26 (3.8%)	13/242 (5.4%)
Cycle threshold rangea (N or ORF1a/b)	35.54 ± 3.14	38.40	36.38 ± 1.92
Estimated SARS-CoV-2 emission rate/level	(1.03 × 105, 2.25 × 107) viruses/h	6.07 × 103 viruses/m3	(7.10 × 101, 1.72 × 103) viruses/cm2

The SARS-CV-2 emission rate or concentration level in air or on surfaces was estimated based on an assumed amplification efficiency of 75% and a reverse-transcription polymerase chain reaction detection limit of 100 copies/µL¹⁰. Lower and upper bounds of virus emission rates or levels corresponded to upper and lower bounds of cycle threshold (Ct) values.

Abbreviation: SARS-CoV-2, severe acute respiratory syndrome coronavirus 2.

^aLower Ct values were used for those of N or ORF1a/b genes for presentation and viral estimation.

From 26 air samples collected, including those using a robot (Supplementary Video 2), 1 sample (air-1) from an unventilated quarantine hotel toilet room was positive (estimated to be 6.07 × 10³ viruses/m³; Table 1, Supplementary Figure 5, Supplementary Table 4). Surface swab samples from a pillow case (swab-1) and hands (swab-2) of patient D who used the toilet room were shown to contain SARS-CoV-2, but no virus was detected in this patient’s EBC sample, which was collected on a different date (Supplementary Figure 5). Additionally, SARS-CoV-2 was detected on an air ventilation duct entrance surface as described above (the duct acted like an air sampler) (Supplementary Figure 5). These air sample data, despite the low positive rate, still show that the air in the hospitals that housed the COVID-19 patients was contaminated with SARS-CoV-2.

Of 242 surface swab samples, 13 were positive for SARS-CoV-2 (Table 1, Supplementary Figure 6, Supplementary Table 4). Among the 5 categories of surfaces, the toilet pit had the highest SARS-CoV-2 positive rate (16.7%, n = 12), followed by the hospital floor (12.5%, n = 16), the other surfaces (7.4%, n = 27), the surfaces touched by patients (4.0%, n = 149), and the surfaces touched by medical staff (2.6%, n = 38). Ct values (36.38 ± 1.92) were obtained for each positive surface swab sample (Table 1, Supplementary Table 4). The surface-borne viral level was estimated to be from 7.10 × 10¹ to 1.72 × 10³ viruses/cm² (Table1). For a toilet pit swab (swab-3), the EBC-1 of its associated patient E also tested positive. For the surfaces touched by patients (149 samples), we detected 6 positives from the hands of patient D, a pillow case of patient D, mobile phones of patients F and G, and computer keyboards of patients G and H (Supplementary Figure 6, Supplementary Table 4). Surprisingly, only 2 of 22 surface swabs from the mobile phones of COVID-19 patients tested positive (Supplementary Figure 6, Supplementary Table 4). None of the 26 surface swabs collected from handles of various objects appeared positive for the virus (Supplementary Table 4). These observations seem to not support the widely held belief that direct transmission by contact with surfaces plays a major role in COVID-19 spread.

DISCUSSION

For the first time, we report that SARS-CoV-2 is released directly into the air via breathing by COVID-19 patients. The detection limit for SARS-CoV-2 by RT-PCR was reported to be approximately 100 RNA copies/µL [9]. Using the equation described in the Supplementary Materials, the observed Ct values showed that SARS-CoV-2 levels in exhaled breath could reach 10⁵–10⁷ copies/m³ if an average breathing rate of 12 L/min is assumed. The SARS-CoV-2 breath emission rate is affected by many factors, such as disease stage, patient activity, and possibly age. We found that the SARS-CoV-2 breath emission rate into the air was the highest, up to 10⁵ viruses/min, during the earlier stages of COVID-19. This finding is in line with that from a previous report that showed that the highest SARS-CoV-2 load in throat swabs was observed at the time of symptom onset [10]. Another significant discovery from this work is that SARS-CoV-2 emission does not continue at the same rate but rather is a sporadic event. For example, 2 EBC samples (EBC-1, EBC-2) collected from patient E but on different dates and using the same method returned different test results (Supplementary Table 2).

SARS-CoV-2 has previously been detected in fine particles in hospital air [4]. A peak of fluorescence biological particle at around 1 µm was also detected in exhaled breath from healthy individuals [11]. The SARS-CoV-2–negative air samples (Supplementary Figure 5) may be due to low SARS-CoV-2 emissions, virus inactivation by disinfectants, and rapid dilution or removal of SARS-CoV-2 by fresh air flow (2.5 m³/min for general hospital wards, Supplementary Video 1; 12 air exchanges per hour for ICU rooms). The presence of SARS-CoV-2 in the toilet room air might be due to the exhaled virus or virus aerosolization from the toilet. The spread of COVID-19 by asymptomatic patients has also been documented [12]. The asymptomatic disease carriers do not generally cough or sneeze to generate respiratory droplets; thus, the observed transmission of the disease has been difficult to explain by respiratory droplet transmission but is rather logical for a fine aerosol route.

The dominant SARS-CoV-2 transmission routes need to be interrupted in order to effectively stop the ongoing COVID-19 pandemic. Large respiratory droplets and direct contact transmissions are presently cited as major transmission routes for COVID-19 by WHO. In contrast, we show that the surfaces of mobile phones (n = 22) and various handles (n = 35) frequently used by COVID-19 patients presented very low probabilities of SARS-CoV-2 presence (9.0% and 0%, respectively). Airborne transmission of SARS-CoV-2 has already played an important role in documented real-life COVID-19 spread in semi-enclosed environments [7, 13]; for example, cluster infection incidents in a choir in Washington State [14] and in a restaurant in Guangzhou, China [15]. Though we did not study infectivity or transmission probability and other virus-releasing activities such as talking and singing, our study demonstrates that exhaled breath emission plays an important role in SARS-CoV-2 emission into the air, which could have contributed greatly to the observed airborne cluster infections and the ongoing pandemic. Accordingly, measures such as enhanced ventilation and the use of face masks are essential to minimize the risk of infection by airborne SARS-CoV-2.

Supplementary Data

Supplementary materials are available at Clinical Infectious Diseases online. Consisting of data provided by the authors to benefit the reader, the posted materials are not copyedited and are the sole responsibility of the authors, so questions or comments should be addressed to the corresponding author.

Notes

Author contributions. M. Y. and J. M. contributed to the study design. J. M., M. Y., X. Q., Z. Z., H. W., L. S., and J. G. contributed to sample collection and experiments. J. M., X. Q., and M. Y. contributed to patients’ recruitment and clinical management. M. Y., J. M., X. L., H. C., L. Z., L. M., S. A. G., P. B., and R. C. F. contributed to data analysis, data interpretation, figure preparation, and literature search. M. Y. wrote the manuscript draft, and all authors revised the manuscript. All authors reviewed and approved the final version of the report.

Acknowledgments. The authors thank the local Center for Disease Prevention and Control medical stuff for their assistance in recruiting patients.

Financial support. This research was supported by the National Natural Science Foundation of China (NSFC; 22040101) and the NSFC Distinguished Young Scholars Fund (21725701 to M. Y.).

Potential conflicts of interest. The authors: No reported conflicts of interest. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest.

Author notes

J. M. and M. Y. contributed equally to this work.