Risk of SARS-CoV-2 transmission by fomites: a clinical observational study in highly infectious COVID-19 patients

Study cohort

Hospitalized patients (age > 18 years) treated at St. Josef-Hospital Bochum, Germany, with confirmed SARS-CoV-2 infection with a high virus load (RT-PCR from combined nasopharyngeal and oropharyngeal swab (swab) with a cycle threshold (CT) ≤ 25 on admission) were included in this study. The initial viral load was determined using Allplex™ 2019-nCoV Assay (Seegene Inc., Seoul, Republic of Korea) targeting three SARS-CoV-2 specific genes (E gene, RdRP gene and N gene) with a sensitivity of 100 copies per run. Exclusion criteria were acute myocardial infarction, current need of ventilation support systems (e.g., high flow-or non-invasive ventilation), current treatment in an intensive care unit, evidence of drug or alcohol abuse, acute psychiatric disorders and any clinical or mental disorder that might deteriorate the patient’s condition during the standardized procedure of sampling (such as dysphagia), as per investigator’s judgement. Collected clinical data included medical history, current daily medication, laboratory results, blood gas analysis, and results of x-rays or computed tomography (to define the “clinical classification of COVID-19-infection” following the recommendation of the World Health Organization (WHO) [8] and Robert Koch Institute (RKI) [9]).

Study design

After written informed consent, two combined nasopharyngeal and oropharyngeal swabs were collected from each patient. Then, patients were asked to forcefully cough two-times on a pre-defined surface area containing nine standardized steel-carriers, each with a one-centimeter diameter (“cough”), using a specially designed tripod with a defined distance of 15 centimeters (Appendix Figure 1). In addition, patients were asked to moisten nine steel carriers with saliva for ten seconds within their mouth (“moisten”). After defined time points at room temperature (1 min, 5 min, 15 min, 30 min, 45 min, 90 min, 120 min and 240 min), the steel-carriers were placed in containers containing 2 mL cold Dulbecco’s modified Eagle’s medium (DMEM complete, supplemented with 10 % (v/v) fetal calf serum, 1 % non-essential amino acids, 100 IU/mL penicillin, 100 μg/mL streptomycin and 2 mM L-Glutamine) and transported on ice to the biosafety level three laboratory of the Ruhr-University Bochum. The study was conducted between November 2020 and April 2021.

Virus culture

VeroE6 cells were seeded at 3×10⁵ cells/well in a six well cell culture plate and incubated for at least four hours at 37 °C and 5 % CO₂. Hereafter, the medium was replaced with 1·8 mL of patient swabs, “cough” samples or “moisten” samples and 2·5 μg/mL amphotericin B was added. Over a maximum period of ten days, cells were monitored daily for the appearance of a cytopathic effect (CPE), indicating productive virus infection. Upon visible CPE, cells were harvested for RT-qPCR and the supernatant (SN) was collected for viral titration and RT-qPCR. Viral titers in the SN were quantified by endpoint-dilution and the 50 % tissue culture infective dose (TCID50/mL), calculated according to Spearman and Kärber [10].

Reverse Transcription Quantitative PCR (RT-qPCR)

SARS-CoV-2 RNA was isolated from the supernatant using AVL buffer and the QIAamp Viral RNA Kit (QIAGEN®, www.qiagen.com) according to the manufacturer’s instructions. RNA was directly subjected to one-step quantitative PCR (RT-qPCR) running a GoTaq Probe 1-Step RT-qPCR System (Promega®, www.promega.com). Total RNA was purified from VeroE6 cells using the RNeasy Mini Kit (QIAGEN®, www.qiagen.com). Subsequently, 500 ng of total RNA were reverse transcribed using the PrimeScript™ RT Master Mix (Takara®, www.takarabio.com) and subjected to two-step RT-qPCR running a GoTaq Probe 2-Step RT-qPCR System (Promega®, www.promega.com). RT-qPCR was performed as described previously [11] using a light cycler LC480 to quantify the M-Gene abundance.

Data analysis and sample size

Clinical patient parameters are expressed as mean ± SD or n (% of total). Results are expressed as means (± SEM). Clinical characteristics were screened for correlations using Spearman’s correlation coefficient. Statistical significance was defined as α=0·05. Statistical analysis was performed using GraphPad Prism version 8.0.0 for Windows (GraphPad Software, San Diego, California USA, www.graphpad.com). Sample size calculation was performed using G*Power Version 3.1.9.6 for windows [12].

Legal and ethical considerations

The study was conducted according to the revised principles of the Declaration of Helsinki and was approved by the ethics committee of the Ruhr-University Bochum (registration number 20-7065) in November 2020. All patients gave written informed consent.

Sequencing and strain assignment

RNA of the initial swaps was isolated using the NucleoSpin RNA kit (Macherey & Nagel) followed by a reverse transcription utilizing the SuperScript IV together with Oligo dT and random hexamer primer (Thermo Fisher) according to the manufacturers’ instructions. Subsequently, the cDNA was subjected to deep sequencing. Sequencing libraries were prepared from 4·5μl cDNA using NEBNext® ARTIC SARS-CoV-2 Library Prep Kit for Illumina sequencing platforms (New England BioLabs® Inc., neb.com, catalog #E7650). Concentration and size of the cDNA amplicons and libraries were assessed using Qubit fluorometer and Tapestation (High Sensitivity D1000 ScreenTape), respectively. High throughput paired end sequencing was performed using Illumina MiSeq sequencer and MiSeq Reagent Kit v2 (500-cycles) following the manufacturer’s recommendations. Raw reads were quality checked, trimmed and mapped to the SARS-CoV-2 reference sequence (NCBI Reference Sequence: NC_045512) using QIAGEN CLC Genomics Workbench 21.0.5. After removing duplicates, partially full length consensus sequences were extracted and samples were assigned to respective lineages using the pangolin tool [13] (Table 2). In addition, variants in spike domains were identified and annotated using Geneious prime 2021.2.2 (https://www.geneious.com).

Study cohort

A total of 15 patients (33.3 % female) between 39 and 89 years (mean age 70·5 (± 12·5) years) were recruited. Baseline parameters including laboratory findings on admission are presented in Table 1. All study patients had risk factors for a severe course of COVID-19 according to the criteria of the WHO [8] and RKI [14]. Three (20 %) patients had up to two risk factors, while most patients had multiple risk factors for a severe COVID-19 illness (3-4 risk factors: 7 (46·7 %), 5-6 risk factors: 2 (13·3 %), >6 risk factors: 1 (6·7 %)). Body temperature on admission was 37·0 (± -0.9) °C. On the study day, 9 (60 %) patients were categorized with a mild COVID-19 disease according to STACOB criteria (adaptation following the WHO Therapeutics and COVID-19:living guideline [8]). Nasal oxygen support was required by seven (46·7 %) patients. Mean peripheral oxygen saturation on admission was 93·4 % (SEM ± 7·4 %). None of the included patients were vaccinated against SARS-CoV-2.

On the day of the study, most patients had only mild symptoms (n= 9/15, 60 %) and were categorized to mild COVID-19 disease. Follow-up revealed a clinical worsening in 10 patients and 3 died (2 patients died of COVID-19, 1 patient died with COVID-19). Consistent with literature [15], a high level of lactate dehydrogenase (r=0·53, p=0·044), leucocytes (r=0·69, p=0·0056), but also C-reactive protein (r=0·54, p=0·035) on admission was significantly correlated with a more severe COVID-19 infection (Appendix Figure 2). No significant correlation was found for COVID-19 severeness and other described severeness predictors including age (r=0·44, p=0·11), alanine aminotransferase (r=-0·31, p=0·26), aspartate aminotransferase (r=0·07, p=0·81), procalcitonin (r=0·53, p=0·47), or D-dimers (r=0·45, p=0·11).

SARS-CoV-2 viral load and sequencing

The patients’ viral load before sample acquisition as determined with the combined nasopharyngeal and oropharyngeal swabs are displayed in Table 2. Mean CT-values in RT-qPCR analyses were E-Gene 15.3 (SEM ± 2·7), S-Gene 18·1 (SEM ± 4·9) RdRP-Gene 17·2 (SEM ± 3·7) and N-Gene 19·4 (SEM ± 5·5). Viral variants included supposed wildtype (n=10, 66·7 %), variant of concern (VoC) Alpha (n=4, 26·7 %) and VoC Beta (n=1, 6·7 %). Deep sequencing using the pangolin tool [13] confirmed lineage assignment for 14/15 (93·3 %) samples for nearly full-length genomes. Variant patterns identified by detailed investigation of spike domains underlined lineage assignment according to RKI variant reports [16] (Figure 1, Table 2).

• Download figure
• Open in new tab

Figure 1: Viral Isolates analyzed in this study.

Spike domains (nt) of 14 samples (black dots) and the SARS-CoV-2 reference genome NC_045512 (open circle) were assembled using Clustal Omega at EMBL-EBI [28]. Phylogenetic analysis was conducted with “One click” at Phylogeny.fr [29]and tree was visualized with MEGA X [30]. Scale bar indicates the number of changes per site in maximum likelihood inference (HKY85 substitution model); numbers at branches represent bootstrap values (1000 repetitions; cut-off ≥ 70%).

Two patient samples were excluded from the study due to bacterial/fungal contamination within the cultures (Appendix Table 1). Viral RNA could be detected from all (n=13/13) combined nasopharyngeal and oropharyngeal swabs (inoculum) and viral RNA could be successfully detected within the inoculated cell cultures (Figure 2, Appendix Figure 3). After inoculation with the “swabs”, viral loads in the cells ranged from 2·23×10¹ to 2·03×10⁹ RNA copies/50 ng and in the supernatant from not detectable to 6·58×10⁷ RNA copies/mL (Appendix Table 1). Infectious virus (“Infectivity”), determined as TCID₅₀/mL, could be recovered from the nasal-oropharyngeal swabs from n=6/13 (46·2 %) patients (Figure 2; P1, P2, P3, P5, P8, P10). Of note, despite inclusion criteria defining a high viral load (CT ≤ 25, Table 2), some patients displayed lower viral loads at the time-point of the study, with none of the swab samples resulting in productive virus infection in cell culture (Appendix Figure 3; P.11, P12, P13, P14).

• Download figure
• Open in new tab

Figure 2: Quantification of viral loads and infectivity of patient swabs, “moisten” samples and “cough” samples that could be successfully recovered in cell culture.

VeroE6 cells were inoculated with the patient material and monitored on a daily basis. Upon the emergence of cytopathic effects, the supernatant was collected to determine viral loads by RT-qPCR (RNA copies/mL; indicated by ¹) and viral titers by an endpoint-dilution assay (TCID₅₀/mL, indicated by ³). Additionally, RNA was isolated from the cells and subjected to RT-qPCR to determine viral loads (RNA copies/50 ng total RNA, indicated by ²). For each patient (P1-P10), three panels were designed. The top small panel includes exclusively data regarding the patient swabs, while the larger middle panel shows the data for the “moisten” samples and the lower panel the data collected from the “cough” samples. For “moisten” and “cough” samples viral loads and infectivity at nine different time points were determined. The color indicates the amount of virus being detectable in each sample, with light grey being the lower limit of detection to dark blue resulting in 10¹⁰ RNA copies/mL, RNA copies/50 ng or TCID₅₀/mL. The visible CPE was rated two dimensionally, with light grey being “no visible CPE” and dark green being “visible CPE”.

Evaluation of SARS-CoV-2 transmission risk

Steel-carriers contaminated via intensive moistening with saliva (Figure 2, “moisten”) resulted in a visible CPE and detectable viral RNA within the cells (at least one time point positive) in n=6/13 (46·2 %) cases (P1, P2, P3, P5, P9, P10). Despite the absence of a visible CPE in patient 6 and patient 10 (Figure 2, time points 30 and 45 min), viral RNA was detected in the cells and supernatants, and infectious virus was quantified, indicating that harvesting might have been too early for the appearance of CPE. Infectious virus was recovered from n=5/13 (38·5 %) contaminated steel carriers (Figure 2; “SN infectivity” P1, P3, P5, P6, P10) with viral titers ranging from 5·59×10¹ to 8·68×10⁵ TCID₅₀/mL. In some patients, infectious virus could be recovered from the steel-carriers for up to 240 min after incubation at room temperature (Figure 2; P1, P3, P5), underlining the environmental stability of SARS-CoV-2 over several hours. For other samples (Figure 2; P2, P6, P10), infectious virus could be recovered at early time points only (1 min, 15 min, 30 min, 45 min).

After contamination with coughing, viral RNA could be weakly detected within five of the cellular samples (Figure 2; “cough” P1, P7, P8, P10, Appendix Figure 3; P11). However, none of the via coughing contaminated surfaces resulted in productive cellular infection as determined by the appearance of CPE and quantification of infectivity (Figure 2, Appendix Figure 3).