Rapid diagnosis of bacterial co-infection and antimicrobial resistance in patients with SARS-Cov-2 infection

Introduction: To date there has been limited research into bacterial co-infections in COVID-19. Culture data suggests less than 10% of COVID-19 patients isolate a bacterial co-pathogen. Molecular diagnostics can rapidly identify respiratory pathogens and antibiotic resistance genes. In this observational study we used the Biofire® FilmArray® and nanopore sequencing to rapidly diagnose co-infection and resistance in patients with suspected COVID-19. Methods: Respiratory samples were collected from patients with suspected COVID-19 at Ninewells Hospital (Dundee, UK). Samples were tested using the BioFire® FilmArray® Pneumonia Plus Panel according to manufacturers’ instructions. Also, samples underwent host DNA depletion and nanopore sequencing. Results: A total of 127 suspected COVID-19 samples were analysed using the Biofire® FilmArray® and 51.2% (65/127) of samples were positive for a bacterial pathogen. A total of 81 samples were positive for COVID-19 (73 sputum, 8 BAL/ETA) and 39/81 (48.1%) were positive for a bacterial pathogen. The most frequent organisms in COVID-19 positive and negative patients were H. influenzae (24.7% v 20%), S. aureus (24.7% v 16%, p=0.2) and S. pneumoniae (2% v 8%). Differences between COVID-19 positive and negative groups were not statistically significant. Using the nanopore metagenomic workflow, common respiratory pathogens and clinically relevant antibiotic resistance genes were detected, with both concordant and discordant data noted with respect to targeted PCR. Conclusion: Common respiratory pathogens and antibiotic resistance genes were detected in patients with confirmed COVID-19 using both the Biofire® FilmArray® and nanopore sequencing.