Application of respiratory metagenomics for COVID-19 patients on the intensive care unit to inform appropriate initial antimicrobial treatment and rapid detection of nosocomial transmission

Charalampous, T.; Alcolea-Medina, A.; Snell, L. B.; Williams, T. G. S.; Batra, R.; Camporota, L.; Meadows, C. I. S.; Wyncoll, D.; Barrett, N. A.; Hemsley, C. J.; Bryan, L.; Newsholme, W.; Boyd, S. E.; Green, A.; Mahadeva, U.; Patel, A.; Cliff, P.; Page, A. J.; O'Grady, J.; Edgeworth, J. D.

Background. Clinical metagenomics (CMg) is being evaluated for translation from a research tool into routine diagnostic service, but its potential to significantly improve management of acutely unwell patients has not been demonstrated. The SARS-CoV-2 pandemic provides impetus to determine that benefit given increased risk of secondary infection and nosocomial transmission by multi-drug resistant (MDR) pathogens linked with expansion of critical care capacity. Methods. Prospective evaluation of CMg using nanopore sequencing was performed on 43 respiratory samples over 14 weeks from a cohort of 274 intubated patients across seven COVID-19 intensive care units. Results. Bacteria or fungi were cultured from 200 (73%) patients, with a predominance of Klebsiella spp. (31%) and C. striatum (7%) amongst other common respiratory pathogens. An 8 hour CMg workflow was 93% sensitive and 81% specific for bacterial identification compared to culture, and reported presence or absence of {beta}-lactam resistance genes carried by Enterobacterales that would modify initial guideline-recommended antibiotics in every case. CMg was also 100% concordant with quantitative PCR for detecting Aspergillus fumigatus (4 positive and 39 negative samples). Single nucleotide polymorphism (SNP)-typing using 24 hour sequence data identified an MDR-K. pneumoniae ST307 outbreak involving 4 patients and an MDR-C. striatum outbreak potentially involving 14 patients across three ICUs. Conclusion. CMg testing for ICU patients provides same-day pathogen detection and antibiotic resistance prediction that significantly improves initial treatment of nosocomial pneumonia and rapidly detects unsuspected outbreaks of MDR-pathogens.