Broad testing for respiratory viruses among persons under investigation (PUI) for SARS-CoV-2 is performed inconsistently, limiting our understanding of alternative infections and co-infections in these patients. Here, we used unbiased metagenomic next-generation sequencing (mNGS) to assess the frequencies of 1) alternative viral infections in SARS-CoV-2 RT-PCR negative PUIs and 2) viral co-infections in SARS-CoV-2 RT-PCR positive PUIs.
Methods: A convenience sample set was selected from PUIs who were tested for SARS-CoV-2 in the Emory Healthcare system during the first 2 months of the pandemic from 02/26-04/23/20. Laboratory results were extracted by chart review; Flu/RSV and multiplex respiratory pathogen PCRs had been performed at the discretion of treating physicians. Excess nasopharyngeal swab samples were retrieved within 72 hours of collection and underwent RNA extraction and SARS-CoV-2 testing by triplex RT-PCR. mNGS was performed by DNAse treatment, random primer cDNA synthesis, Nextera XT tagmentation, and high-depth Illumina sequencing. Reads underwent taxonomic classification by KrakenUniq, as implemented in viral-ngs.
Results: 53 PUIs were included, 30 negative and 23 positive for SARS-CoV-2 by RT-PCR. Among SARS-CoV-2 negative PUIs, 28 (93%) underwent clinical testing for alternative infections, and 8 (29%) tested positive for another respiratory virus. In all cases, mNGS identified the same virus (Table 1). In another 3 PUIs, mNGS identified two viruses that were not tested for and one that was missed by routine testing. No SARS-CoV-2 was detected by mNGS among RT-PCR negative PUIs. Among SARS-CoV-2 RT-PCR positive PUIs, 18 (69%) underwent clinical testing for co-infections, and none were detected. mNGS did not identify any viral co-infections but did detect SARS-CoV-2 in all 23 PUIs.Table 1: Molecular and Metagenomic Testing of Persons Under Investigation.png)
Conclusion: Unbiased mNGS offers the powerful opportunity to streamline testing for PUIs by assessing for SARS-CoV-2 and alternative infections simultaneously; this technique can also be used to identify co-infections, but none were observed in our study population. Interestingly, many PUIs had no infection identified on routine testing or mNGS, which may reflect inadequate sampling, rapid virus clearance, or a non-viral process.
Ahmed Babiker, MBBS
Assistant Professor
Emory University School of Medicine
Atlanta, GA
Disclosure: I do not have any relevant financial / non-financial relationships with any proprietary interests.
Heath L. Bradley, MSc
Senior Research Specialist
Emory University
Atlanta, Georgia
Disclosure: I do not have any relevant financial / non-financial relationships with any proprietary interests.
Victoria D. Stittleburg, BSc
Research Specialist
Emory University
Atlanta, Georgia
Disclosure: I do not have any relevant financial / non-financial relationships with any proprietary interests.
Autum Key, MSc
Research Specialist
Emory University
Atlanta, Georgia
Disclosure: I do not have any relevant financial / non-financial relationships with any proprietary interests.
Jesse Waggoner, MD
Assistant Professor
Emory University
Atlanta, Georgia
Disclosure: I do not have any relevant financial / non-financial relationships with any proprietary interests.
Anne Piantadosi, MD, PHD
Assistant Professor
Emory University
Atlanta, Georgia
Disclosure: I do not have any relevant financial / non-financial relationships with any proprietary interests.
