Based on COVID-19 case reports, environmental surfaces and air specimens were collected by Guangzhou CDC (GZCDC) from Feb 6 to Apr 10, 2020. The environmental surfaces specimens of COVID-19 cases sampled in home, hotel, public area, restaurant, marketplace, car and pet, which was associated with COVID-19 cases’ life trajectory before hospitalization. Air specimens of COVID-19 cases were also sampled in their room (home or hotel). All specimens were collected before disinfection.

Epidemiological investigations were performed for COVID-19 cases including basic demographic (age, sex, imported cases or not), date of symptoms onset, date of diagnosis, date of environmental specimens collection, clinical severity, symptoms at onset (fever, dry cough, expectoration, fatigue, myalgia and diarrhea), radiological examinations (chest computed tomography [CT]), laboratory testing (white blood cell count, lymphocyte, lymphocyte percentage, neutrophilic granulocyte and neutrophilic granulocyte percentage) and hospital admission.

The environmental specimens of SARS-CoV-2 were collected by qualified technicians who had received biosafety training (who had passed the training) and were equipped with corresponding laboratory skills. Personal protective equipment (PPE) were required for sampling personnel including N95 masks or masks with higher filtration efficiency, goggles, protective clothing, double-layer latex gloves and waterproof boot covers [13]. Sterile premoistened swabs were used for sampling, and then were put into a tube containing 2 to 3 mL virus preservation solution (or isotonic saline solution, tissue culture solution or phosphate buffer) with swabs’ tail discarded and cap tightened. According to Technical Guidelines for COVID-19 Laboratory Testing in China [13], three kinds of positions for environmental sampling were recommended: high frequency contact sites, surface of daily necessaries and toilet related specimens. Two to four specimens should be collected for each category. We further classified the three categories into 9 sub-categories: (1) toilet (sink, faucet, shower, toilet bowl, floor drain); (2) anteroom (chair, table, TV remote, water dispenser, cup, bottle, TV bench, handrail); (3) kitchen (cutting board, bowl, chopsticks); (4) bedroom (bed, sheet pillow, bedside table, telephone, computer, mouse, air conditioner, fan); (5) door knobs and switch buttons in room; (6) air in room; (7) outside room (elevator, elevator button, stair armrest); (8) car (steering wheel, seat, handlebar, air); (9) pet (mouth, nose, anus). Door knobs and switch buttons in room were not classified into specific room locations due to the specimens were collected with mixed form, where were commonly used by COVID-19 cases. And specimens in car (steering wheel, seat and handlebar) were also mixed form.

Air specimens were sampled using an MD8 microbiological sampler (Sartorius, Germany) and sterile gelatin filters (3 μm pores and 80 mm diameter, Sartorius, Germany) [7,14]. Air sampling was set with a volume of 1000 L at a speed of 50 L/minute, and it took about 20 minutes to complete the sampling process. The filters were dissolved aseptically in 30 mL virus preservation solution (or isotonic saline solution, tissue culture solution, or phosphate buffer).

The environmental surfaces and air specimens for nucleic acid detection were stored at -70°C or below (specimens may be temporarily stored in -20°C refrigerators in the absence of -70°C storage condition) and were tested as soon as possible. Laboratory confirmation of environmental surfaces and air specimens was performed by qualified staff, and results were identified through open reading frame 1ab (ORF1ab) and nucleocapsid protein (N) of SARA-CoV-2 by RT-PCR testing in accordance with the protocol established by China CDC [13]. Details on laboratory processes are provided in Appendix (appendix 1 p1-2).

The asymptomatic persons infected with SARA-CoV-2 (asymptomatic cases in short) are referring to those who have no relevant clinical manifestations including clinically detectable signs or self-perceived symptoms such as fever, cough or sore throat, but who have tested positive for SARA-CoV-2 in respiratory specimens or other specimens [15]. Clinical classification of symptomatic cases includes 4 categories [16]: mild, moderate, severe, and critical (appendix 2 p3). In this study, the COVID-19 cases included asymptomatic, mild, moderate, severe, and critical cases. Fever was defined as an axillary temperature of 37.3°C or higher. Public area included elevator, elevator buttons and stair armrests. An imported case is a person who was infected with SARA-CoV-2 in a foreign country and diagnosed in Guangzhou.

Categorical variables were presented with frequency (percentage) and continuous variables were summarized with median (interquartile range [IQR]). Figures were drawn using Office Excel (version 2019). Analyses were all performed with the SAS software (version 9.4 for Windows, SAS Institute, Inc., Cary, NC, USA).

This study was based on the data from the work for an ongoing public health response to COVID-19 by GZCDC as required by the National Health Commission of China, and hence individual informed consent was waived. The study was determined not to be human subjects research and therefore was considered exempt from ethical approval after consultation with the ethics committee of GZCDC. Analytical datasets were constructed in an anonymized manner, and all analysis of personally identifiable data took place onsite at the GZCDC.

Among 39 COVID-19 cases, 23 (23/39, 59.0%) were male, 11 (11/39, 28.2%) were imported and the median age was 36.0 years (IQR, 31.0 to 48.0 years). A total of 9 (9/39, 23.1%) asymptomatic cases and 30 (30/39, 76.9%) symptomatic cases with 10 (10/30, 33.3%) mild, 19 (19/30,63.4%) moderate and 1 (1/30, 3.3%) severe cases were identified, and all (39/39, 100.0%) were cured and discharged. Among them, the common symptoms at onset was fever (24/39, 61.5%), dry cough (21/39, 53.9%), expectoration (9/39, 23.1%), fatigue (6/39, 15.4%), myalgia (3/39, 7.7%), diarrhea (2/39, 5.1%), and 31 (31/37, 79.5%) with CT lung abnormalities (Table 1).

A total of 641 environmental surfaces and air specimens were collected among 39 COVID-19 cases, and 20 specimens (20/641, 3.1%) were tested positive on RT-PCR from 9 COVID-19 cases (9/39, 23.1%), with 5 (5/101, 5.0%) positive specimens from 3 asymptomatic cases, 5 (5/220, 2.3%) from 3 mild cases and 10 (10/374, 2.7%) from 3 moderate cases (Table 2). The COVID-19 cases without fever (8/16 [50.0%] vs. 1/23 [4.3%]), dry cough (7/18 [38.9%] vs 2/21 [9.5%]), expectoration (8/30 [26.6%] vs. 1/9 [11.1%]), fatigue (9/33 [27.2%] vs. 0/6 [0.0%]) and myalgia (20/36 [55.6%] vs. 0/3 [0.0%]), diarrhea (9/37 [24.3%] vs. 0/2 [0.0%]) symptoms were more likely to have positive specimens of SARS-CoV-2 (Table 2).

All the 20 positive specimens were collected within 3 days (≤3 days) from diagnosis to sampling (Fig 1), and 13 (13/259, 5.0%) positive environmental surfaces specimens were collected from home, 6 (6/113, 5.3%) from hotel and 1 (1/5, 20.0%) from car that had driven. While, specimens in restaurant that had eaten (0/30, 0.0%), marketplace that had visited (0/122, 0.0%), pet that had lived with (0/4, 0.0%) and public area that had stayed (0/108, 0.0%) were all tested negative on RT-PCR (Table 2).

A total of 136 environmental surfaces and air specimens were collected from 9 COVID-19 cases who with at least one positive environmental specimen. Among them, 20 specimens (20/136, 14.7%) of specimens were tested positive on RT-PCR. Among 20 positive environmental specimens, 10 (10/42, 23.8%) were found in toilet (5 on toilet bowl, 4 on sink/faucet/shower, 1 on floor drain), 4 (4/21, 19.0%) in anteroom (2 on water dispenser/cup/bottle, 1 on chair/table, 1 on TV remote), 1 (1/8, 12.5%) in kitchen (1 on dining-table), 1 (1/18, 5.6%) in bedroom (1 on bed/sheet pillow/bedside table), 1 (1/5, 20.0%) in car (1 on steering wheel/seat/handlebar) and 3 (3/20, 21.4%) on door knobs. Surfaces specimens collected from outside room (0/13, 0.0%), air specimens collected from room (0/10, 0.0%) and car (0/1, 0.0%) were tested negative on RT-PCR (Table 3 and Fig 2).

This study has several limitations. Firstly, the results of RT-PCR testing do not indicate the amount of viable virus, and viral culture was not done to demonstrate viability. Secondly, due to operational limitations during the outbreak, days of sampling interval was inconsistent. However, it also provided additional information about SARS-CoV-2 survival. Thirdly, the number of air specimens represented only a small fraction of total specimens, and air exchanges in room would have diluted the presence of SARS-CoV-2 in the air. Further studies are required to confirm these preliminary results.