Summary
Background
Detection of pooled samples for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) may improve laboratory testing capacity. This study evaluated the impact of pooling and retesting individual swabs on the overall detection rate and reduction in retesting frequency.
Methods
One hundred respiratory swab samples were tested individually and in groups of three or five samples using Cepheid’s Xpert® Xpress SARS-CoV-2 Test Kit. The optimal number of samples per pool was calculated using ‘A Shiny App for Pooled Testing’.
- Settings
This study was conducted prospectively at King Fahd Hospital of the University (KFHU), which is a 550-bed secondary healthcare institution in Al-Khobar, Saudi Arabia, receiving referrals from severely ill suspected COVID-19 cartridge patients from all surrounding regions. as well as outpatient cases.
- Samples and Materials
The following parameters were used: category 2 × 2 for the contingency table; 90.0% power; type I error of 0.05; κ2 (Cohen’s kappa [κ] coefficients for hypothesis testing) was set to 0.9 since the expected sensitivity of the assay is greater than 90%; and the value of κ1 was set to the values of 0.0, 0.3, 0.5 and 0.7. Considering these parameters, the estimated sample size ranged from 8 to 96 samples. One hundred and one samples were included in the study.
Combined nasopharyngeal (NP)/oropharyngeal (OP) swab samples were sent by the emergency department (ED) or employee health clinic to the hospital’s diagnostic microbiology laboratory for SARS-CoV-2 testing by RT-PCR. Swab samples were transported in viral preservative medium (VPM) (Guangzhou Improve Medical Instruments, Guangzhou, China) refrigerated, and processed within 24 h of collection.
The Xpert® Xpress SARS-CoV-2 Test Kit from Cepheid (California, USA) was used to test all clinical specimens according to the manufacturer’s instructions. It is a cartridge-based system for the detection of the target sequence of the E and N2 gene of arbovirus and SARS-CoV-2, respectively. Residual sample volumes were stored at 8 °C after initial testing and then used in our pooling study within 24 h of collection.
- Inclusion and exclusion criteria
We included clinical specimens obtained from the COVID-19 low-risk group (score less than 5) based on the COVID-19 clinical visual classification score. Samples were enrolled in the study in the order they were received in the laboratory to simulate a real-life scenario.
In addition, we included healthcare workers after known and definite but unprotected exposure and patients evaluated as routinely screened on admission and before any surgical or radiological procedure. We excluded repeat samples obtained from laboratory-confirmed COVID-19 patients from the study.
Results
Twenty-five groups were generated from 101 samples. Of the 13 pools containing five samples each, three pools gave true positive results. While of the 12 pools containing three samples each, five pools gave true positive results. Four samples gave a combined false negative result. The overall sensitivity and specificity of the assay across groups were 66.6% and 100%, respectively. The cycle threshold was lowered in most groups compared to individual sample tests.
Conclusion
The overall pooled test had a notable impact on laboratory resources. However, caution is warranted when selecting cases for pooled testing, as reduced sensitivity can have a significant impact and increase the risk of exposure to infection.
Keywords: SARS-CoV-2, COVID-19, pooling, Saudi Arabia, RT-PCR