Assessing the performance of LumiraDx™ SARS-CoV-2 Ag test in detecting Omicron lineages: 2022-2023 study

Abstract

Background: The introduction of rapid antigen tests revolutionized the approach to SARS-CoV-2 diagnosis, offering prompt and accurate results with high sensitivity and specificity. Although it is more cost- and time-saving than the gold standard, real-time polymerase chain reaction (RT-PCR), the efficacy in general population screening in both hospital- and community-based settings remains unknown. Moreover, rapid antigen testing is limited by qualitative results. This study aims to evaluate the diagnostic reliability of the LumiraDx™ rapid antigen test during the Omicron era and to investigate its quantitative (analogue-to-digital converter (ADC)) results in comparison with RT-PCR Ct values.

Methods: This prospective study included all adult patients with mild-to-moderate SARS-CoV-2 symptoms who were not hospitalised and did not require oxygen supplementation, consented to participate, and attended the Infectious and Tropical Diseases Unit of Padua University Hospital from July 14th, 2022 to January 3rd, 2023. The patients underwent two different tests simultaneously: a nasal LumiraDx™ swab and a real-time RT-PCR assay performed on a nasopharyngeal swab. Sampling was repeated several times for a subset of subjects.

Results: We enrolled 266 consecutive participants and collected 601 pairs of LumiraDx™ and RT-PCR samples. The most prevalent variant was BA.4/BA.5 Omicron (60.2 %). The sensitivity and specificity of LumiraDx™ test when compared to real-time RT-PCR results as the reference standard were 93.1 % and 79.75 %, respectively. No significant differences in diagnostic reliability were found based on the available characteristics, age, sex, symptom status, or COVID-19 variant, except for the days from symptom onset. According to the multilevel logistic regression analysis, the only independent variable significantly associated with test concordance was the Ct value (adjusted odds ratio (OR) = 0.56, p < 0.001). Significant differences in quantitative ADC values were found between false negative (FN) versus true negative (TN), and false positive (FP) and true positive (TP) tests.

Conclusions: This study showed that LumiraDx™ test is reliable for SARS-CoV-2 diagnosis in patients with mild-to-moderate SARS-CoV-2 symptoms. This finding confirms the efficacy of rapid antigen tests in monitoring vulnerable individuals during the current post-vaccination era. When compared with the RT-PCR, LumiraDx™ test effectively quantitatively distinguishes between FN and TN cases, as well as FP and true TP tests, despite inaccuracies in qualitative results.

Keywords: COVID-19; Coronaviruses; Diagnostic testing; Omicron; Quantitative test; RT-PCR; Rapid antigen testing; SARS-CoV-2; Sensitivity; Specificity; Viral epidemic.

Fig. 1

Bivariate kernel density estimation and marginal distributions of Ct values and days after symptoms among true positive (TP) and false negative (FN) cases. The black oblique line represents the estimated linear association (regression coefficient = 0.56 (95 % CI 0.45, 0.66), p-value <0.001, Spearman correlation = 0.47, p-value<0.001). The background heat-map represents the sensitivity variation with respect to both Ct values and days cumulative threshold. The darker the colour, the higher the level of sensitivity of the LumiraDx™ test compared to the standard real time RT-PCR.

Fig. 2

ADC (Analog to Digital Converter) quantitative value distribution. (A) Grouped boxplots comparing ADC distribution between TN, FN, FP and TP results. Statistical significance was assessed conducting pairwise Wilcoxon-signed rank test, p-values were adjusted according to Bonferroni method. (B) Bivariate kernel density estimation and marginal distributions of Ct and ADC values. The black oblique line represents the estimated linear association (regression coefficient = −0.0012, p-value <0.0001), the grey shade area represents the 95 % confidence bands. Note that 5000 ADC is the threshold value discriminating SARS-CoV-2 presence (dashed Grey line).