Comparison of SARS-CoV-2 Detection by Rapid Antigen and by Three Commercial RT-qPCR Tests: A Study from Martin University Hospital in Slovakia.

The global pandemic of coronavirus disease 2019 (COVID-19) caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is having a tremendous impact on the global economy, health care systems and the lives of almost all people in the world. The Central European country of Slovakia reached one of the highest daily mortality rates per 100,000 inhabitants in the first 3 months of 2021, despite implementing strong prophylactic measures, lockdowns and repeated nationwide antigen testing. The present study reports a comparison of the performance of the Standard Q COVID-19 antigen test (SD Biosensor) with three commercial RT-qPCR kits (vDetect COVID-19-MultiplexDX, gb SARS-CoV-2 Multiplex-GENERI BIOTECH Ltd. and Genvinset COVID-19 [E]-BDR Diagnostics) in the detection of infected individuals among employees of the Martin University Hospital in Slovakia. Health care providers, such as doctors and nurses, are classified as "critical infrastructure", and there is no doubt about the huge impact that incorrect results could have on patients. Out of 1231 samples, 14 were evaluated as positive for SARS-CoV-2 antigen presence, and all of them were confirmed by RT-qPCR kit 1 and kit 2. As another 26 samples had a signal in the E gene, these 40 samples were re-isolated and subsequently re-analysed using the three kits, which detected the virus in 22, 23 and 12 cases, respectively. The results point to a divergence not only between antigen and RT-qPCR tests, but also within the "gold standard" RT-qPCR testing. Performance analysis of the diagnostic antigen test showed the positive predictive value (PPV) to be 100% and negative predictive value (NPV) to be 98.10%, indicating that 1.90% of individuals with a negative result were, in fact, positive. If these data are extrapolated to the national level, where the mean daily number of antigen tests was 250,000 in April 2021, it points to over 4700 people per day being misinterpreted and posing a risk of virus shedding. While mean Ct values of the samples that were both antigen and RT-qPCR positive were about 20 (kit 1: 20.47 and 20.16 for Sarbeco E and RdRP, kit 2: 19.37 and 19.99 for Sarbeco E and RdRP and kit 3: 17.47 for ORF1b/RdRP), mean Ct values of the samples that were antigen-negative but RT-qPCR-positive were about 30 (kit 1: 30.67 and 30.00 for Sarbeco E and RdRP, kit 2: 29.86 and 31.01 for Sarbeco E and RdRP and kit 3: 27.47 for ORF1b/RdRP). It confirms the advantage of antigen test in detecting the most infectious individuals with a higher viral load. However, the reporting of Ct values is still a matter of ongoing debates and should not be conducted without normalisation to standardised controls of known concentration.

Eleven routine clinical features predict COVID-19 severity uncovered by machine learning of longitudinal measurements.

Severity prediction of COVID-19 remains one of the major clinical challenges for the ongoing pandemic. Here, we have recruited a 144 COVID-19 patient cohort, resulting in a data matrix containing 3,065 readings for 124 types of measurements over 52 days. A machine learning model was established to predict the disease progression based on the cohort consisting of training, validation, and internal test sets. A panel of eleven routine clinical factors constructed a classifier for COVID-19 severity prediction, achieving accuracy of over 98% in the discovery set. Validation of the model in an independent cohort containing 25 patients achieved accuracy of 80%. The overall sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were 0.70, 0.99, 0.93, and 0.93, respectively. Our model captured predictive dynamics of lactate dehydrogenase (LDH) and creatine kinase (CK) while their levels were in the normal range. This model is accessible at https://www.guomics.com/covidAI/ for research purpose.