Differences in detected viral loads guide use of SARS-CoV-2 antigen-detection assays towards symptomatic college students and children.

ABSTRACT

Limitations in timely testing for SARS-CoV-2 drive the need for new approaches in suspected COVID-19 disease. We queried whether viral load (VL) in the upper airways at presentation could improve the management and diagnosis of patients. This study was conducted in a 9 hospital system in Allegheny County, Pennsylvania between March 1-August 31 2020. Viral load was determined by PCR assays for patients presenting to the Emergency Departments (ED), community pediatrics practices and college health service. We found that for the ED patients, VL did not vary substantially between those admitted and not. VL was relatively equivalent across ages, except for the under 25 age groups that tended to present with higher loads. To determine if rapid antigen testing (RAT) could aid diagnosis in certain populations, we compared BD Veritor and Quidel Sofia to SOC PCR-based tests. The antigen assay provided a disease-detection sensitivity of >90% in a selection of 32 positive students and was modeled to have an 80% sensitivity in all positive students. In the outpatient pediatric population, the antigen assay detected 70% of PCR-positives. Extrapolating these findings to viral loads in older hospitalized patients, a minority would be detected by RAT (40%). Higher loads did correlate with death, though the prognostic value was marginal (ROC AUC of only 0.66). VL did not distinguish between those needing mechanical ventilation and routine inpatients. We conclude that VL in upper airways, while not prognostic for disease management, may aid in selecting proper testing methodologies for certain patient populations.