key: cord-0752838-y6y54pfh authors: Thommes, Lis; Burkert, Francesco Robert; Öttl, Karla-Wanda; Goldin, David; Loacker, Lorin; Lanser, Lukas; Griesmacher, Andrea; Theurl, Igor; Weiss, Günter; Bellmann-Weiler, Rosa title: Comparative evaluation of four SARS-CoV-2 antigen tests in hospitalized patients date: 2021-02-17 journal: Int J Infect Dis DOI: 10.1016/j.ijid.2021.02.052 sha: 2760fb6e09167f38f6d0938cdaad7e2be75ad9ef doc_id: 752838 cord_uid: y6y54pfh Objectives Rapid identification of infected subjects is a corner stone for control of a pandemic like the current one with the SARS-CoV-2. Easy to handle antigen tests can provide timely results which is of specific importance in a primary care setting. However, concerns exist regarding their sensitivity, which let us evaluate four commercially available tests in patients hospitalized for COVID-19. Methods We analyzed in parallel nasopharyngeal/oropharyngeal swabs from 87 consecutive patients admitted to our department with moderate to severe COVID-19 by means of quantitative RT-PCR (Cobas, Roche) and up to four antigen tests from different distributors. Antigen test results were linked to Ct (cycle threshold) values as a marker for infectivity of patients. Results We found that two out of four antigen tests correctly identified subjects with high viral loads (Ct ≤25) and three out of four tests detected more than 80% of subjects with a Ct ≤30 which is considered as a threshold for infectivity. However, one test investigated had a poor clinical performance. When investigating subjects with Ct values >30 we found that in up to 45% of the cases the antigen test was still positive. Conclusion Most antigen tests had a sufficient sensitivity to identify symptomatic subjects infected with SARS-CoV-2 and with transmissible infection. On the other hand, antigen testing may not be suitable to identify loss of infectivity in COVID-19 subjects during follow-up. Newly introduced antigen tests need to be validated in a clinical or primary care setting to define their clinical usefulness. A central component for the control of a pandemic, like the current one caused by the SARS-CoV-2 with no specific available therapy, is the rapid identification and containment of infected individuals (Weissleder et al., 2020) . Easily accessible and relatively inexpensive antigen tests offer an easy to use diagnostic tool to quickly identify such patients. Antigen tests are useful in primary and emergency care settings because they do not require laboratory equipment. They have been shown to aid in the specific diagnosis of respiratory infections such as influenza specifically during the cold season when many respiratory viruses with similar clinical symptoms are circulating (Moriyama et al., 2020) . However, a major drawback for the use of antigen tests is their lack of sensitivity which is sometimes much lower within a clinical setting than described in the user's manual (Lanser et al., 2020) as shown also for influenza bed-side tests during the 2009 pandemic (Bellmann-Weiler et al., 2011) . Several antigen tests for rapid detection of SARS-CoV-2 have been introduced into clinical practice for diagnosis of symptomatic patients. However, information on their sensitivity in symptomatic patients in a real-life clinical setting are scarce (Diao et al., 2020) . Therefore, we performed a comparative analysis of four commercially available SARS-CoV-2 antigen tests in symptomatic COVID-19 patients who needed hospitalization and compared their analytical outcomes with the results and cycle threshold (Ct) values obtained with a commercially available RT-PCR test. We investigated consecutive COVID-19 patients admitted to our inpatient ward at the Depart- When studying patients with a Ct value <25, which reflects the population with the highest viral loads and thus the highest infectivity, we found that two tests were able to detect all affected patients, whereas the remaining tests had sensitivities of 83.3% and 60%, respectively (Table 1) . We then examined the clinical performance of the four antigen tests including all J o u r n a l P r e -p r o o f patients with a Ct value <30 (Diao et al., 2020) . Thereby, we observed a sensitivity of the tests between 45.2% and 88.9%. This indicated large variances in the clinical usefulness of those tests to identify infected individuals with a high probability. Thus, the positive predictive values (PPV) of the antigen tests to detect patients with transmissible infections (Ct values ≤30) were between 52.5 (CI 45.2% -59.6%) and 73.8 (CI 62.3% to 82.8%) ( Table 1) . Three out of four tests had a sensitivity above 79%. However, one out of the four antigen tests showed correct positive results in only 60% of patients with Ct values ≤30. To determine whether the antigen tests would remain positive despite non-infectivity, we swabbed patients after prolonged hospitalization during follow up with Ct values above 30, as proposed by the RKI (Laferl et al., 2020) . We found that in subjects with Ct values >30, 9.3% to 46.0% of samples showed positive results in the antigen tests. In this single-center study, we present a clinical evaluation and comparison of four commercially available COVID-19 antigen tests, using RT-PCR as a reference. A strength of our study is that the analyses were performed at a single center with experienced personnel and standardized procedures for sampling and processing of nasopharyngeal/oropharyngeal swabs, which were immediately analyzed in parallel by RT-PCR and antigen tests. This provides information on the true value of antigen detection kits as a point of care method for identifying infected COVID-19 patients and estimates the clinical utility of such tests to diagnose symptomatic patients in a primary care setting. A limitation of our study is that we cannot provide information on the specificity of the tests as we only investigated hospitalized patients with already confirmed COVID-19. A further limitation of our study is the usage of Ct-values to define levels of infectivity. Ctvalues can vary considerably, either because of inconsistent sampling methods, or based on differences between RT-PCR methods, which currently lack standardized reference materials. To reliably determine infectivity based on Ct-values, further validation and international standardization of the available quantitative RT-PCR assays is required. Our study indicates that some antigen tests have a very good sensitivity to identify infected patients with COVID-19 like symptoms needing hospitalization specifically those with higher viral loads and thus higher infectivity, which has already been shown for other antigen tests than those used by us (Lambert-Niclot et al., 2020) . In our opinion, this makes them a suitable tool in primary care situations to rapidly detect SARS-CoV-2 infected individuals among symptomatic subjects. Nevertheless, one out of four CE-certified antigen tests showed very poor clinical performance. However, we cannot provide information whether or not those tests perform similarly well in subjects who are less severely affected by the infection. Moreover, our results do not allow speculation on the usage of antigen tests in strategies aimed at screening of asymptomatic patients in the incubation phase of the infection, although some governments have implemented such strategies. Currently, the WHO does not recommend rapid immunoassays for unselected screening for COVID-19 but rather for testing of individuals with a high pre-test probability (e.g. local outbreak situation in semi-closed communities or groups for early detection and isolation, health workers during outbreaks etc.) -especially when an analysis by PCR is not immediately available (World Health, 2020). To have a definitive answer to these important public health issues, further evaluations of these tests with parallel PCR analyses have to be carried out during screening events which will also provide more information regarding the specificity of the respective tests. Numerous studies evaluating POCT antigen tests have recently emerged in the literature. Strömer et al. compared the SARS-CoV-2 NADAL COVID-19® Ag POCT with RT-PCR, and also found that the POCT was more likely to be positive with lower Ct values (Strömer et al., 2020) . Our results regarding the Panbio TM COVID-19 Ag Rapid test by Abbott are concordant with the findings of Albert et al. (Albert et al., 2020) ; due to the absence of reported J o u r n a l P r e -p r o o f Ct values, no direct comparison is possible between our study and that by Olearo et al, nonetheless in accordance with our finding they also detected an increase in sensitivity with higher viral loads, (Olearo et al., 2020) . Our comparative evaluation of the antigen tests was extended to patients being considered as non-infectious according to the recommendations of the RKI (Laferl et al., 2020) . These investigations showed persistence of positivity in a considerable number of subjects even with Ct values above 30 and lack of COVID-19 specific symptoms. This suggests that antigen tests may not serve as a suitable tool to determine either persisting infectivity or end of infectivity in subjects suffering from COVID-19. There was no funding source for this study. The study was approved by the ethics committee of the Innsbruck Medical University (ID of ethical vote: 1167/2020) and conformed to the principles of the Declaration of Helsinki. Field evaluation of a rapid antigen test (Panbio™ COVID-19 Ag Rapid Test Device) for COVID-19 diagnosis in primary healthcare centres Accuracy of bedside antigen tests in the diagnosis of new influenza A/H1N1v infection Accuracy of a nucleocapsid protein antigen rapid test in the diagnosis of SARS-CoV-2 infection. Clinical microbiology and infection : the official publication of the European Society of An approach to lifting self-isolation for health care workers with prolonged shedding of SARS-CoV-2 RNA Evaluation of a Rapid Diagnostic Assay for Detection of SARS-CoV-2 Antigen in Nasopharyngeal Swabs Evaluating the clinical utility and sensitivity of SARS-CoV-2 antigen testing in relation to RT-PCR Ct values Seasonality of Respiratory Viral Infections Handling and accuracy of four rapid antigen tests for the diagnosis of SARS-CoV-2 compared to RT-qPCR Performance of a Point-of-Care Test for the Rapid Detection of SARS-CoV-2 COVID-19 diagnostics in context Antigen-detection in the diagnosis of SARS-CoV-2 infection using rapid immunoassays: interim guidance The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. We thank Prof. Dr. Germar Pinggera and Prof. Dr. Franz Allerberger for generously providing antigen tests.J o u r n a l P r e -p r o o f