ABSTRACT
Background SARS-CoV-2 viral load and kinetics assessed in serial blood samples from hospitalised COVID-19 patients by RT-PCR are poorly understood. Methods We conducted an observational, prospective case series study in hospitalised COVID-19 patients. Clinical outcome data (Intensive Care Unit admission and mortality) were collected from all patients until discharge. Viremia was determined longitudinally during hospitalisation, in plasma and serum samples using two commercial and standardised RT-PCR techniques approved for use in diagnosis of SARS-CoV-2. Viral load (copies/mL and log10) was determined with quantitative TaqPath TM COVID-19 test. Results SARS-CoV-2 viremia was studied in 57 hospitalised COVID-19 patients. Persistent viremia (PV) was defined as two or more quantifiable viral loads detected in blood samples (plasma/serum) during hospitalisation. PV was detected in 16 (28%) patients. All of them, except for one who rapidly progressed to death, cleared viremia during hospitalisation. Poor clinical outcome occurred in 62.5% of patients with PV, while none of the negative patients or those with sporadic viremia presented this outcome (p<0.0001). Viral load was significantly higher in patients with PV than in those with Sporadic Viremia (p<0.05). Patients presented PV for a short period of time: median time from admission was 5 days (Range=2-12) and 4.5 days (Range=2-8) for plasma and serum samples, respectively. Similar results were obtained with all RT-PCR assays for both types of samples. Conclusions Detection of persistent SARS-CoV-2 viremia, by real time RT-PCR, expressed as viral load over time, could allow identifying hospitalised COVID-19 patients at risk of poor clinical outcome.