Link between SARS-CoV-2 emissions and airborne concentrations: closing the gap in understanding

TL;DR

This study combines experimental measurements and a novel theoretical model to quantify airborne SARS-CoV-2 concentrations, providing evidence that speaking significantly increases airborne viral load, thus clarifying transmission pathways.

Contribution

It introduces a new predictive approach to estimate airborne SARS-CoV-2 concentrations and compares it with experimental data to improve understanding of airborne transmission.

Findings

Airborne SARS-CoV-2 is detectable during speaking but not during breathing.

The theoretical model's estimates closely match experimental measurements within uncertainty ranges.

Speaking increases airborne viral concentration by an order of magnitude compared to breathing.

Abstract

The question of how SARS-CoV-2 is transmitted remains surprisingly controversial today, especially with reference to airborne transmission. In fact, despite a large body of scientific evidence, health and regulatory authorities still require direct proof of this mode of transmission. To close this gap, we measured the saliva viral load of SARS-CoV-2 of an infected subject located in a hospital room, as well as the airborne SARS-CoV-2 concentration in the room resulting from the person breathing and speaking. As the next step, we simulated the same scenarios to estimate the concentration of RNA copies in the air through a novel predictive theoretical approach. Finally, we conducted a comparative analysis (i.e. a metrological compatibility analysis) of the differences between the experimental and theoretical results by estimating the uncertainties of these two approaches. Our results showed that for an infected subject's saliva load ranging between 2.4x106 and 5.5x106 RNA copies mL-1, the corresponding airborne SARS-CoV-2 concentration was not detectable when the person was breathing, but was 16.1 (with an uncertainty of 32.8) RNA copies m-3 when speaking. The application of the novel predictive estimation approach provided average concentrations of 3.2 (uncertainty range of 0.2-8.3) and 18.5 (uncertainty range of 4.5-43.0) RNA copies m-3 for breathing and speaking scenarios, respectively, thus confirming that for the breathing scenario, the airborne RNA concentration would be undetectable, being below the minimum detection threshold of the experimental apparatus (< 2 RNA copies m-3).