Serological testing reveals the hidden COVID-19 burden among healthcare workers experiencing a SARS-CoV-2 nosocomial outbreak.

We describe the results of testing healthcare workers from a tertiary care hospital in Japan, which had experienced a COVID-19 outbreak during the first peak of the pandemic, for SARS-CoV-2 specific antibody seroconversion. Using two chemiluminescent immunoassays and a confirmatory surrogate virus neutralization test, serological testing unveiled that a surprising 42.2% (27/64) of overlooked COVID-19 diagnoses had occurred when case detection had relied solely on SARS-CoV-2 nucleic acid amplification testing. This undetected portion of the COVID-19 iceberg beneath the surface may potentially have led to silent transmissions and triggered the spread. A questionnaire-based risk assessment was further indicative of exposures to specific aerosol-generating procedures, i.e. non-invasive ventilation, having had conveyed the highest transmission risks and served as the origin of outbreak. Our observations are supportive of a multi-tiered testing approach, including the use of serological diagnostics, in order to accomplish exhaustive case detection along the whole COVID-19 spectrum.


Introduction
When the COVID-19 pandemic landed in January 2020, Japan was no exception Participants completed a questionnaire which included demographic data, past medical 120 history, occupational exposure to aerosol-generating procedures performed on confirmed  The results of molecular or serological testing were described as frequencies and 130 percentages among the participants screened. To assess the differences among 131 demographic characteristics between NAT-confirmed and serologically confirmed 132 COVID-19 patients, the following demographic variables were compared by t-tests (for 133 the "Age" variable) or Fisher's exact test (for the other variables; "Male sex", "Pre-134 existing risk condition", "Severity" and "Signs and symptoms"). Magnitude of 135 All rights reserved. No reuse allowed without permission.
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. ("Symptomatic"), expressing no symptoms ("Asymptomatic") and complaining of 139 isolated smell impairments ("Hyposmia/anosmia only"). Spearman's correlation 140 coefficient was calculated for the various indices of serological response. For the 141 procedural exposure risk assessment, risk ratio (RR), and risk difference (RD), per 142 exposure were calculated as the ratio, or the absolute difference, between COVID-19 143 incidence among those exposed to the aerosol-generating procedures and the reference 144 ("Not exposed") group. The association between exposures to aerosol-generating 145 procedures and COVID-19 incidence was tested by Fisher's exact test. To evaluate the 146 extent of harm attributable to each procedure regarding the actual increase of COVID-19 147 cases, the attributable fraction among the exposed (AFe) and the attributable number of 148 events (AN) were calculated. AFe is the proportion of COVID-19 diagnoses in the 149 exposed group that is attributable to the occupational exposure and was calculated per 150 exposure as; AFe = (RR-1) / RR (6). AN is the absolute number of COVID-19 diagnoses 151 attributable to the occupational exposure and was calculated per exposure as; AN = AFe 152 × (number of COVID-19 diagnoses among the exposed). P-values less than 0.05 were 153 All rights reserved. No reuse allowed without permission.
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

Nucleic acid amplification test
All rights reserved. No reuse allowed without permission.
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint this version posted July 19, 2021. ;https://doi.org/10.1101https://doi.org/10. /2021   (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint this version posted July 19, 2021. ; https://doi.org/10.1101/2021.07.15.21260585 doi: medRxiv preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.