Remember me
The objective of this experimental proof of concept study was to evaluate the SARS-CoV-2 viral load and its progression in paired serial oronasopharyngeal swab and exhaled breath samples of COVID-19 patients. The viral load in EB samples was detected through a noninvasive and simple method using an air filter-based device, followed by routine qRT-PCR. The main purpose of this study was to resolve the question of whether the viral load of samples from the upper respiratory tract allows an accurate prediction of the viral load in EB, which may represent a more appropriate biological material for assessing the contagiousness of infected COVID-19 patients.
Methods Study designIn this prospective proof of concept study, we examined patients who initially tested positive for SARS-CoV-2 on admission to the hospital. In total, 187 specimens from 15 hospitalized patients diagnosed with COVID-19 were collected between July 18th and November 16th 2020. The screening consisted of collecting swabs of the upper respiratory tract (oronasopharyngeal swabs) and EB specimens and testing for the presence of SARS-CoV-2. In the statistical analysis, paired results were considered to compare the rate of the two sampling methods and to describe differences in the viral loads during the progression of COVID-19.
If the disease progression of patients was classified according to the NIH classification of severity of illness [] as asymptomatic or presymptomatic, mild and moderate patients were recruited to participate in the study. Furthermore, participants were informed and required to sign a written consent form. Patients who developed a critical condition during hospitalization that led to admission to the intensive care unit and those who were discharged from the hospital due to recovery were considered dropouts.The Ethics Committee of the Faculty of Medicine, Kiel University, Germany, approved this study (D527/20). Informed consent for COVID-19 research was waived by the data protection office of the Faculty of Medicine, Kiel University. The investigators were not blinded to allocation during experiments and outcome assessment.
Specimen collectionAfter initial positive COVID-19 diagnosis, a simultaneous paired collection of oronasopharyngeal and EB specimens was performed. The day after the initial diagnosis of COVID-19 was set as day 1, and specimen collection was repeated every 1–3 days during hospitalization, particularly on days 3, 5, 7, 10 and 14.
Patients were instructed to avoid eating, drinking, smoking, chewing gum or brushing teeth 30 minutes prior to sample collection. EB specimens were collected using a filter-based device (SensAbues®, Stockholm, Sweden) consisting of a mouthpiece, a polymeric electret filter enclosed in a plastic collection chamber, and an attached clear plastic bag. The mouthpiece is designed in such a way to avoid oral fluid contamination during sampling, allowing only microparticles to pass through and to be collected on the filter inside the device. A clear plastic bag indicates adequate individual use and a sufficient volume of exhaled breath passing the electret filter [20Characterization of exhaled breath particles collected by an electret filter technique., 21Clinical trial of a new technique for drugs of abuse testing: A new possible sampling technique., 22Detection of drugs of abuse in exhaled breath using a device for rapid collection: Comparison with plasma, urine and self-reporting in 47 drug users.].The patients were instructed to inhale through the nose and tidally exhale without 20 times through the mouthpiece onto the filter inside the collection device. A new device was used for each EB specimen collection. EB specimen collection was performed under the supervision of an investigator.
Next, specimens from the upper respiratory tract were collected. Oro-nasopharyngeal sampling was performed using sterile swabs (Nerbe Plus GmbH & Co. KG, Winsen/Luhe, Germany) following the standard recommended procedures []. Both swabs and the EB samples were stored at -80°C until extraction. Extraction of SARS-CoV-2 RNA and qRT-PCRPaired nasopharyngeal throat swabs and EB specimens were analysed simultaneously and under identical conditions. Viral RNA was extracted using the QIAamp viral RNA mini kit (QIAGEN GmbH, Hilden, Germany). Nasopharyngeal throat swabs were extracted in 0.5 mL buffer. Two hundred microliters of the extract was taken and further diluted (1:1).
For the extraction of viral RNA from the filter of the EB collection device, the manufacturer's instructions were modified. The electret air filter was first wetted by frequently adding 1 mL of buffer every 5 minutes to a total volume of 3 mL. Then, the collection device was gently agitated and vortexed for 2 min, and an additional 0.5 mL of buffer was pipetted onto the filter. To elute the solvent from the prewetted filter, the EB collection device was placed into plastic test tubes, and 400 µL of the extracted EB samples were taken for RNA isolation. After concentration, the RNA suspension was eluted in 50 μL of buffer. Ten microliters of the sample eluate was added to 15 µL master mix for each PCR. qRT-PCR was performed on a BioRAD CFX96 Real-Time Thermal Cycler with Maestro Software (Hercules, California) using the ampliCube Coronavirus SARS-CoV-2 kit (Mikrogen, Neuried, Germany) that targets the E and Orf1a genes of the virus. Thermal cycling conditions were set at 50°C for 8 min and 95°C for 3 min, followed by 45 cycles of 95°C for 10 s and 60°C for 45 s.
SARS-CoV-2 RNA detection was determined by amplification of both genes targeted (E, ORF1a). A cut-off cycle threshold (Ct) value of 40 was set as negative. The assay included an internal control (IC) for monitoring nucleic acid extraction efficacy and qRT-PCR inhibition in each reaction. A negative control (NC) was included and processed parallel to the clinical sample. The qRT-PCR analysis was replicated once for each positive sample.
SARS-CoV-2 quantificationCt values of the targeted E gene were converted to log10 SARS-CoV-2 RNA copies/µL using calibration curves based on an in vitro transcribed (IVT)-quantified coronavirus 2019 E gene control (European Virus Archive GLOBAL, Charité University Berlin). The IVT stock solution at a concentration of 106 copies/10 µL was serially diluted to 105,104, 103, 102, 101 and 10° copies/10 µL, and 10 µL of each was added to the master mix. qRT-PCR of the calibration was performed under the conditions described above.
Statistical AnalysesThe results are presented as the means or medians, SDs and interquartile ranges (IQRs), unless stated otherwise. Statistical details for each analysis are described in each figure legend or in the respective part of the text. Student's t-test was used to assess group differences for continuous numerical variables, and a one-tailed p-value was calculated. Additionally, a Welch t-test correction was applied because of unequal sample distribution variance. P-values were considered to be significant at p <0.05. No data points were excluded. Statistical analyses were carried out using R 2020 version 1.3.1093 and Origin 2020 version 9.70.
ResultsA comparison of oronasopharyngeal swabs and EB samples is required to analyse the correlation between the respective viral loads. Here, we tested 187 specimens of 15 hospitalized patients with a confirmed SARS-CoV-2 infection. Of these, 87 were of the upper respiratory tract (oronasopharyngeal swabs), and 100 EBs were collected with a filter-based device (Supplementary Table 1). The included COVID-19 patients had a mean age of 56 years (IQR 33-79, range 27-85).
SARS-CoV-2 RNA detection was determined with qRT-PCR. All 87 swabs from the upper respiratory tract tested positive. Moreover, 70 of the 100 EB specimens tested positive for SARS-CoV-2, whereas the remaining 30 were classified as negative due to Ct values > 40. The negative tested EB samples mostly occurred 7 days after clinical onset, i.e., 16 out of 30 (53.3%) EB samples taken on days 7-14 tested negative, whereas only 14 of 70 (20.0%) EB samples collected on days 1-5 showed a negative result (Supplementary Table 2).
Figure 1 provides an overview of SARS-CoV-2 samples from the upper respiratory tract as well as exhaled breath of all patients enrolled in this study. Individual COVID-19 progression is presented by means of Ct values of the targeted E and Orf1a genes of both specimen types. As shown, samples of the upper respiratory tract exhibited lower Ct values than the simultaneously collected EB samples. The mean difference in Ct values was 10.64. Furthermore, neither targeted E nor Orf1a genes showed a significant difference in oronasopharyngeal swabs (CI 99%, t(171) = -0.184, p = 0.855) nor in EB samples (CI 99%, t(137) = -0.447, p = 0.656) (Supplementary Figure 1). The mean differences between the Ct values of replicates of swabs and EB samples were shown to be 1.26 and 0.88, respectively.Figure 1Dynamics of SARS-CoV-2 in the upper respiratory tract and EB samples of 15 infected hospitalized patients. Progression of Ct values of the targeted E gene and Orf1a gene in serial swabs (n = 87) compared to EB samples (n = 70) of each individual during the 14 days after COVID-19 diagnosis
The Ct values were used for further calculation of the respective viral RNA load of each sample. Viral loads of both sampling methods were calculated using a calibration curve (correlation coefficient R2 = 0.9984, Supplementary Figure 2) with a quantification range shown to be between Ct values of 17.48 (106 copies/10 µL) and 38.56 (1 copy/10 µL).
The typical progression of COVID-19 evaluated through samples of the upper respiratory tract begins with a steady increase in viral load until reaching a shedding peak, which is followed by continuously decreasing viral loads. Patients 3 and 4 merely showed decreasing viral loads due to inclusion in the study after being symptomatic 3-5 days before being tested positive. The EB samples show a different progression in comparison to the swab samples, with viral loads marginally increasing or remaining nearly constant during the 14 days after diagnosis of COVID-19 (Figure 2). This results in a higher variance of the viral load of oronasopharyngeal swabs compared to the EB samples.Figure 2Comparison of viral load progression of swab and exhaled breath samples from six selected COVID-19-patients over 14 days. *Extractable log10 SARS-CoV-2 RNA copies per oronasopharyngeal swab and per 20 times exhaling through EB collection device.
Figure 3 shows the viral load per sampling device, i.e., the extractable number of SARS-CoV-2 RNA copies of an oronasopharyngeal swab and of the EB collection device (after exhaling 20 times). By comparing the number of SARS-CoV-2 copies, the viral load of the oronasopharyngeal swabs (n = 87) was significantly higher (CI 99%, t(86) = 3.526, p6 (IQR = 1.63 × 104 - 4.17 × 106, range 1.65 × 102 - 1.40 × 108), whereas the mean viral load of EB samples (on day 1-10 after diagnosis) was 2.47 × 103 (IQR = 2.17 × 102-1.85 × 103, range 7.19 × 101-2.94 × 104). Estimating that exhaling 20 times is approximately equivalent to 10 L of exhaled breath [], a viral load up to 2.94 × 103 can be found in one litre of exhaled breath.Figure 3SARS-CoV-2 viral load in oronasopharyngeal swabs and EB samples. *Extractable number (mean [SD]) of log10 SARS-CoV-2 RNA copies per oronasopharyngeal swab and per EB collection device. Samples were collected on day 1 (swabs n = 28, EB n = 26), day 3 (swabs n = 20, EB n = 17), day 5 (swab n = 16, EB n = 13), day 7 (swab n = 12, EB n = 6) and day 10 (swabs n = 8, EB n = 7) after COVID-19 diagnosis.
All patients have shed SARS-CoV-2 in exhaled breath in at least one occasion. Nevertheless, the distinctive viral loads emitted by patients show a high heterogeneity, e.g. 9.83 × 101 to 2.94 × 104 virus copies per 20 times exhaling on day 1 after the first diagnosis.
Finally, the respective viral loads of 70 simultaneously collected paired samples of oronasopharyngeal swabs and EB were not found to correlate (correlation coefficient R2Figure 4.Figure 4Correlation of SARS-CoV-2 viral load of paired oronasopharyngeal swabs and EB samples. The viral load of samples (n = 70) of both specimen types simultaneously collected on respective days after COVID-19 diagnosis is shown by means of log10 SARS-CoV-2 RNA copies (R2 < 0.01).
DiscussionThe evaluation and analysis of exhaled breath is important to extend existing knowledge, providing further proof for SARS-CoV-2 quantification using this biological matrix. Although EB sampling seems challenging, it is a promising biological matrix to explore, especially as it is a noninvasive and more comfortable sampling method than oronasopharyngeal swab sampling and can be easily repeated. The main purpose of this pilot study was to detect SARS-CoV-2 RNA in exhaled breath and to compare the viral load of simultaneously collected serial swabs and EB samples taken at different time points during hospitalization of COVID-19 patients. Further details of the medical condition of the patients were not included in the evaluation. This study is comparable to other studies of similar design aiming to find viral RNA in different biological materials, e.g., sputum, urine, feces and plasma or serum [[5]Detection of SARS-CoV-2 in Different Types of Clinical Specimens.,[6]Virological assessment of hospitalized patients with COVID-2019.,[8]Viral load dynamics and disease severity in patients infected with SARS-CoV-2 in Zhejiang province, China, January-March 2020: Retrospective cohort study.]. However, it is still elusive whether these biological matrices correlate with the infectiousness of patients diagnosed with the virus [[24]Transmissibility and transmission of respiratory viruses.].EB sampling is not common for diagnosing SARS-CoV-2 infection. Although recent studies have reported the detection of the virus in exhaled breath condensate (EBC) samples [16Use of exhaled breath condensate (EBC) in the diagnosis of SARS-COV-2 (COVID-19)., 17COVID-19 patients in earlier stages exhaled millions of SARS-CoV-2 per hour., 18Breath-, air- and surface-borne SARS-CoV-2 in hospitals.], no documented time-dependent progression of SARS-CoV-2 viral loads in EB has yet been described. To fill this gap, we collected serial samples during a period of 14 days after COVID-19 diagnosis to analyse the progression described by means of Ct values and viral loads. To ensure a comparison obtaining valid data, the commonly used sampling method of oronasopharyngeal swabs was performed in parallel to analyse the viral loads of the two different specimen types collected from each patient at the same time.EB testing was performed using a simple device containing an electret air filter, which is commonly utilized to detect nonvolatile exogenous and endogenous molecules, including drugs, metabolites, proteins and other biomarkers [20Characterization of exhaled breath particles collected by an electret filter technique., 21Clinical trial of a new technique for drugs of abuse testing: A new possible sampling technique., 22Detection of drugs of abuse in exhaled breath using a device for rapid collection: Comparison with plasma, urine and self-reporting in 47 drug users.,[25]Lovén Björkman S et al.Peanuts in the air - clinical and experimental studies.]. The results of our study clearly show that viral RNA is detectable as well. The utilized method provides reproducible and reliable results since the mean difference between Ct values of replicates of EB samples was shown to be 0.88. This mean difference was even lower than that evaluated for swab samples (1.26).Analogous to other specimen types, such as sputum, urine, feces or serum[5,6,8], both the targeted genes (E and Orf1a) of SARS-CoV-2 can be detected in respiratory droplets of EB, showing no significant difference (CI 99%, t(137) = -0.447, p = 0.656). Ryan et al. recently reported that other genes, such as the N and S genes, can also be found in EBC samples [[16]Use of exhaled breath condensate (EBC) in the diagnosis of SARS-COV-2 (COVID-19).], indicating that a sufficient amount of SARS-CoV-2 RNA in EB is perceptible.Although EB samples show a lower viral load than swab samples, SARS-CoV-2 was detected in EB samples even 12 days after COVID-19 diagnosis. Furthermore, the progression of viral loads described in this study remains nearly constant over a time of 14 days. The viral loads slightly decreased in the exhaled breath of COVID-19 patients during 7-10 days after diagnosis. However, in contrast, the progression of the viral loads of swab samples described here differs from that. The viral load of swab samples slowly increases, leading to a shedding peak and subsequently decreases [[6]Virological assessment of hospitalized patients with COVID-2019.,[26]Transmission heterogeneities, kinetics, and controllability of SARS-CoV-2.], confirming the results of studies that present a similar progression of COVID-19.While interpreting the data, it should be considered that the days after diagnosis do not necessarily equate with the days after onset, as some patients were not tested immediately due to delayed hospital submission. Hence, 7 days after a positive test may represent the end of the infection of some patients, whereas others might exhibit the shedding peak because of early detection of SARS-CoV-2 infection (Figure 1).EB samples showed a significantly lower viral load (CI 99%, t(86)=3.526, p3 in EB samples are comparable to the results of Ma et al. evaluating EBC [[17]COVID-19 patients in earlier stages exhaled millions of SARS-CoV-2 per hour.]. Here, we show the measured mean viral load of 2.47 × 103 (range 7.19 × 101-2.94 × 104), representing the amount of SARS-CoV-2 obtained by breathing 20 times into the device. Thus, up to approximately 2.94 × 103 can be found in 1 L of exhaled breath. Considering the typical respiratory rate for a healthy adult at rest of 18 breaths per minute [[27]Barrett KE Barman SM Boitano S Brooks HL. Ganong's review of medical physiology.], an infected patient would shed approximately 3.89 × 103 - 1.59 × 106 per hour simply via regular breathing. This amount of viral load could possibly remain in the air at least for several minutes [[14]Turbulent Gas Clouds and Respiratory Pathogen Emissions: Potential Implications for Reducing Transmission of COVID-19.]. While analysing the difference in the viral load of EB and swab samples, the distinctive sampling method as well as the biological material should be considered. Cells of the oropharyngeal or nasopharyngeal mucosa containing the viral RNA were mechanically abraded using a swab, which subsequently led to an artificially generated higher viral load in swabs. In comparison, the collection of respiratory droplets in exhaled breath is noninvasive and consequently does not undergo such mechanical stress.Ljungkvist et al. and Beck at al. describe the collection efficiency as well as the recovery from the filter to be more than 90% [[28]Two techniques to sample non-volatiles in breath-exemplified by methadone.,[29]Demonstration that methadone is being present in the exhaled breath aerosol fraction.]. Moreover, the collection efficiency seems to be 99% in the particle diameter range of 0.5–20 μm. As respiratory droplets are typically 5-10 µm in diameter, it can be assumed that the collection efficiency of the system lies between 90 and 99%. Hence, the sampled viral load measured from EB samples almost completely translate to the actual viral emission. Although the mean viral load in EB samples was low, it represents the minimal viral load that would have been exhaled into the environment by the patient just while tidal breathing. Consequently, a non-infected close contact would be exposed at least to this viral load in a non-ventilated room. Hence, talking, singing or even coughing and sneezing might cause higher aerosol emissions [[14]Turbulent Gas Clouds and Respiratory Pathogen Emissions: Potential Implications for Reducing Transmission of COVID-19.,[30]Aerosol emission and superemission during human speech increase with voice loudness.]. In contrast, patients swallow the nasopharyngeal and oropharyngeal mucus with cells containing viral RNA. Therefore, there is a high possibility that even though the viral load of swab samples is significantly higher, the viral amount found in swabs is not fully shed into the environment, as is the case when exhaling.Furthermore, each patient enrolled in this study exhaled at least once SARS-COV-2 RNA. Nevertheless, the emitted amount of virus copies is extremely heterogeneous as some patients shed in the very beginning of the illness only 92 virus copies per 20 times exhaling whereas others may exhale 3 orders of magnitude higher (2.94 × 104). The infection risk among population could possibly increase the higher the amount of EB viral shedding get. Therefore, exhaled breath testing may allow distinguishing between non-, low- and super-spreader for the aerosol route of transmission.
It is commonly discussed that COVID-19 patients are most infectious in the first week of the infection [[6]Virological assessment of hospitalized patients with COVID-2019.,[26]Transmission heterogeneities, kinetics, and controllability of SARS-CoV-2.]. Here, we show that patients might exhale a significant amount of the virus in the later stage of the infection. However, all EB samples collected on day 14 after COVID-19 diagnosis tested negative. Thus, even though swab samples show a positive result of SARS-CoV-2 with Ct values of 28-29 on day 14, the paired EB sample could be SARS-CoV-2 negative. Nevertheless, this is not always the case, as similar Ct values of swab samples of another patient (Ct 27.88, viral load of 1.23 × 104) showed a positive paired EB sample with a Ct value of 28.53, resulting in a viral load of 2.94 × 104 per 20 exhalations. These results clearly indicate that even after 2 weeks of infection, high viral loads can be found in swabs of the upper respiratory tract, but the patient might not shed the virus via exhalation and most likely might no longer be infectious. Nevertheless, it should be considered that this study does not inform about the distinctive infectiousness of the sampled individuals since no viral culture was performed.Lastly, no correlation between the viral load of swab samples and EB samples was found (Figure 4). In contrast, Pan et al. showed that sputum and swab samples seem to correlate [[31]Viral load of SARS-CoV-2 in clinical samples.]. This is most likely due to both specimen types being collected from the upper respiratory tract, whereas exhaled breath mainly originates from the lower respiratory tract.As mentioned earlier, SARS-CoV-2 is mainly transmitted via respiratory droplets [11Airborne transmission of SARS-CoV-2: The world should face the reality., 12Aerodynamic analysis of SARS-CoV-2 in two Wuhan hospitals., 13Aerosol and Surface Stability of SARS-CoV-2 as Compared with SARS-CoV-1., 14Turbulent Gas Clouds and Respiratory Pathogen Emissions: Potential Implications for Reducing Transmission of COVID-19.,[24]Transmissibility and transmission of respiratory viruses.].Hence, assessing the infectiousness merely through swabs of the upper respiratory tract might not be suitable in all cases.
Comments (0)