Study population and setting
This study included 610 hospitalized COVID-19 patients clinically diagnosed by chest computed tomography (CT) indicative of viral pneumonia between February 2 – February 17, 2020 in Hankou Hospital, a hospital designated for treating COVID-19 patients in Wuhan, China. RT-PCR on pharyngeal swabs was done to confirm infection. All patients had pharyngeal swab specimens RT-PCR prior to admission, and patients with initially negative, “dubious”, or weakly positive samples had follow-up samples retaken and tested 1-2 days later. Initially positive patients had the test repeated after clinical symptoms improved.
Summary of Main Findings
Of 610 patients hospitalized with viral pneumonia, only 168 (27.5%) had a positive RT-PCR result on pharyngeal swab specimen prior to admission, and only 241 (39.5%) had at least one positive RT-PCR result at any time point. 17 patients with confirmed COVID-19 (positive RT-PCR) had subsequent negative RT-PCR test results, followed again by positive RT-PCR results several days later after treatment and improved symptoms.
The major strength is that a large number of hospitalized individuals with clinical diagnosed COVID-19 were followed over time with repeat RT-PCR testing.
The major limitation is the lack of detailed information on the clinical course of individual patients, particularly related to timing of exposures and symptom onset which would be helpful in discerning when in a clinical time-course individuals may be more or less likely to be testing negative when truly infected. Serological confirmation would also be beneficial to confirm COVID-19 infection beyond just CT-confirmed viral pneumonia.
Longitudinal follow up of clinically diagnosed patients with serial RT-PCR test results demonstrate testing sensitivity and stability issues.