Comparison of SARS-COV-2 nasal antigen test to nasopharyngeal RT-PCR in mildly symptomatic patients

Abstract

IntroductionCOVID 19 has been vastly spreading since December 2019 and the medical teams worldwide are doing their best to limit its spread. In the absence of a vaccine the best way to fight it is by detecting infected cases early and isolate them to prevent its spread. Therefore, a readily available, rapid, and cost-effective test with high specificity and sensitivity for early detection of COVID 19 is required. In this study, we are testing the diagnostic performance of a rapid antigen detection test in mildly symptomatic cases. (RADT).MethodsThe study included 4183 male patients who were mildly symptomatic. A nasal sample for the rapid antigen test and a nasopharyngeal sample was taken from each patient. Statistical analysis was conducted to calculate the sensitivity, specificity, positive predictive value, negative predictive value and kappa coefficient of agreement.ResultsThe prevalence of COVID 19 in the study population was 17.5% (733/4183). The calculated sensitivity and specificity were 82.1% and 99.1% respectively. Kappa’s coefficient of agreement between the rapid antigen test and RT-PCR was 0.859 (p < 0.001). A stratified analysis was performed and it showed that the sensitivity of the test improved significantly with lowering the cutoff Ct value to 24.ConclusionThe results of the diagnostic assessment of nasal swabs in the RADT used in our study are promising regarding the potential benefit of using them as a screening tool in mildly symptomatic patients. The diagnostic ability was especially high in cases with high viral load. The rapid antigen test is intended to be used alongside RT-PCR and not replace it. RADT can be of benefit in reducing the use of PCR.

Diagnostic performance of a SARS-CoV-2 rapid antigen test in a large, Norwegian cohort

“…In our study, the overall specificity of the antigen test was found to be very high (99.9 %). This is consistent with previous reports, which all showed specificity numbers close to 100 % ( [6] , [7] , [8] , [9] , [10] , [11] , [12] ). False positive results are thus rare.…”

“…In the samples with ct values below 30, we found the RAT’s sensitivity to be 83.8 %. This is in line with previous studies which also found an association between high viral load (ct values < 30 or viral load > 10 6 copies/ml) and increased sensitivity, with reports ranging from 80.0 % to 98.0 % ( [6] , [7] , [8] , [9] , [10] , [11] , [12] ). This indicates that the majority of infectious cases can be correctly identified with the RAT.…”

“…Our data shows that for SARS-COV-2 infected patients with symptom duration less than five days, the sensitivity of the RAT was about 80 %, and when restricting the analysis to individuals with an expected contagious viral load (ct value < 30), the sensitivity was 87.6 %. This is in line with other studies of the Panbio RAT, where sensitivity ranged from 77.2 % to 95.8 % when only patients with symptom duration of less than one week were considered ( [6] , [7] , [8] , [9] , [10] , [11] , [12] ).…”

“…The prevalence in Oslo during the study period was low to medium with a prevalence rate in all samples from Aker test station at 5.7 % (unpublished data), which made it particularly important to conduct an independent evaluation. The overall sensitivity reported by the aforementioned studies varied between 60.5 % and 90.5 % ( [6] , [7] , [8] , [9] , [10] , [11] , [12] ). Studies comprising mainly symptomatic persons described higher sensitivity, all above 80 % ( 6 , 8 , 9 , 12 ), whereas the populations in the two studies reporting the lowest sensitivities (60.5 % and 71.4 %) both contained large fractions of asymptomatic individuals ( 7 , 10 ).…”

“…Recent large prospective studies on the Panbio RAT have included from 913 to 4138 participants and were performed in settings with relatively high SARS-CoV-2 prevalence, ranging from 9.6 % to 40.1 % ( [6] , [7] , [8] , [9] , [10] , [11] , [12] ). The prevalence in Oslo during the study period was low to medium with a prevalence rate in all samples from Aker test station at 5.7 % (unpublished data), which made it particularly important to conduct an independent evaluation.…”