A simple algorithm based on initial Ct values predicts the duration to SARS-CoV-2 negativity and allows more efficient test-to-release and return-to-work schedules

Especially during global pandemics but also in the context of epidemic waves, the capacity for diagnostic qRT-PCRs rapidly becomes a limiting factor. Furthermore, excessive testing incurs high costs and can result in an overstrained work force in diagnostics departments. Obviously, people aim to shorten their isolation periods, hospitals need to discharge convalescent people, and re-employ staff members after infection. The aim of the study was to optimize retesting regimens for test-to-release from isolation and return-to-work applications. For this purpose, we investigated the association between Ct values at the first diagnosis of SARS-CoV-2 infection and the period until test negativity was reached, or at least until the Ct value exceeded 30, which is considered to indicate the transition to a non-infectious state. We included results from the testing of respiratory material samples for the detection of SARS-CoV-2 RNA, tested from 01 March 2020 to 31 January 2022. Lower initial Ct values were associated with longer periods of SARS-CoV-2 RNA positivity. Starting with Ct values of <20, 20-25, 25-30, 30-35, and >35, it took median intervals of 20 (interval: 14-25), 16 (interval: 10-21), 12 (interval: 7-16), 7 (interval: 5-14), and 5 (interval: 2-7) days, respectively, until the person tested negative. Accordingly, a Ct threshold of 30 was surpassed after 13 (interval: 8-19), 9 (interval: 6-14), 7 (interval: 6-11), 6 (interval: 4-10), and 3 (interval: 1-6) days, respectively, in individuals with aforementioned start Ct values. Furthermore, the time to negativity was longer for adults versus children, wild-type SARS-CoV-2 variant versus other variants of concern, and in patients who were treated in the intensive care units. Based on these data, we propose an adjusted retesting strategy according to the initial Ct value in order to optimize available PCR resources.