Rapid diagnosis of SARS-CoV-2 pneumonia on lower respiratory tract specimens

• Main Authors: Vanessa De Pace, Patrizia Caligiuri, Valentina Ricucci, Nicola Nigro, Barbara Galano, Valeria Visconti, Giorgio Da Rin, Bianca Bruzzone
• Format: Article
• Language: English
• Published: BMC 2021-09-01
• Series: BMC Infectious Diseases
• Subjects: Rapid PCR; Bronchoalveolar lavage (BAL); Bronchoaspirates (BAS); SARS-CoV-2
• Online Access: https://doi.org/10.1186/s12879-021-06591-w

Abstract Background The ongoing SARS-CoV-2 pandemic requires the availability of accurate and rapid diagnostic tests, especially in such clinical settings as emergency and intensive care units. The objective of this study was to evaluate the diagnostic performance of the Vivalytic SARS-CoV-2 rapid PCR kit in lower respiratory tract (LRT) specimens. Methods Consecutive LRT specimens (bronchoalveolar lavage and bronchoaspirates) were collected from Intensive Care Units of San Martino Hospital (Genoa, Italy) between November 2020 and January 2021. All samples underwent RT-PCR testing by means of the Allplex™ SARS-CoV-2 assay (Seegene Inc., South Korea). On the basis of RT-PCR results, specimens were categorized as negative, positive with high viral load [cycle threshold (Ct) ≤ 30] and positive with low viral load (Ct of 31–35). A 1:1:1 ratio was used to achieve a sample size of 75. All specimens were subsequently tested by means of the Vivalytic SARS-CoV-2 rapid PCR assay (Bosch Healthcare Solutions GmbH, Germany). The diagnostic performance of this assay was assessed against RT-PCR through the calculation of accuracy, Cohen’s κ, sensitivity, specificity and expected positive (PPV) and negative (NPV) predictive values. Results The overall diagnostic accuracy of the Vivalytic SARS-CoV-2 was 97.3% (95% CI: 90.9–99.3%), with an excellent Cohen’s κ of 0.94 (95% CI: 0.72–1). Sensitivity and specificity were 96% (95% CI: 86.5–98.9%) and 100% (95% CI: 86.7–100%), respectively. In samples with high viral loads, sensitivity was 100% (Table 1). The distributions of E gene Ct values were similar (Wilcoxon’s test: p = 0.070), with medians of 35 (IQR: 25–36) and 35 (IQR: 25–35) on Vivalytic and RT-PCR, respectively (Fig. 1). NPV and PPV was 92.6% and 100%, respectively. Table 1 Demographic characteristics and data sample type of the study cases (N = 75) Male, N (%) 56 (74.6%) Age (yr), Median (IQR) 65 (31–81) BAS, N (%) 43 (57.3%)  Negative 30.2%  Positive—High viral load [Ct ≤ 30] 27.9%  Positive—Low viral load [Ct 31–35] 41.9% BAL, N (%) 32 (42.7%)  Negative 37.5%  Positive—High viral load [Ct ≤ 30] 40.6%  Positive—Low viral load [Ct 31–35] 21.9% Data were expressed as proportions for categorical variables. Specimens were categorized into negative, positive with high viral load [cycle threshold (Ct) ≤ 30] and positive with low viral load (Ct of 31–35). BAS bronchoaspirates, BAL bronchoalveolar lavage, Ct cycle threshold Fig. 1 Distribution of E gene cycle threshold values of the rapid PCR and RT-PCR Conclusions Vivalytic SARS-CoV-2 can be used effectively on LRT specimens following sample liquefaction. It is a feasible and highly accurate molecular procedure, especially in samples with high viral loads. This assay yields results in about 40 min, and may therefore accelerate clinical decision-making in urgent/emergency situations.