Abstract
BACKGROUND: SARS-CoV-2 seems mainly transmissible via respiratory droplets. We compared the time-dependent SARS-CoV-2 viral load in serial pharyngeal swab with exhaled breath (EB) samples of hospitalized COVID-19 patients. METHODS: In this prospective proof of concept study, we examined hospitalized patients who initially tested positive for SARS-CoV-2. Paired oronasopharyngeal swab and EB specimens were taken at different days of hospitalization. EB collection was performed through a simple, noninvasive method using an electret air filter-based device. SARS-CoV-2 RNA detection was determined with real-time quantitative reverse transcription polymerase chain reaction. RESULTS: Of 187 serial samples from 15 hospitalized patients, 87/87 oronasopharyngeal swabs and 70/100 EB specimens tested positive. Comparing the number of SARS-CoV-2 copies, the viral load of the oronasopharyngeal swabs was significantly higher (CI 99%, P<<0,001) than for EB samples. The mean viral load per swab was 7.97 × 10(6) (1.65 × 10(2)-1.4 × 10(8)), whereas EB samples showed 2.47 × 10(3) (7.19 × 10(1)-2.94 × 10(4)) copies per 20 times exhaling. Viral loads of paired oronasopharyngeal swab and EB samples showed no correlation. CONCLUSIONS: Assessing the infectiousness of COVID-19 patients merely through pharyngeal swabs might not be accurate. Exhaled breath could represent a more suitable matrix for evaluating infectiousness and might allow screening for superspreader individuals and widespread variants such as Delta.