Detection of Pathogens and Antimicrobial Resistance Genes in Ventilator-Associated Pneumonia by Metagenomic Next-Generation Sequencing Approach

BACKGROUND: The early identification of pathogens and their antibiotic resistance are essential for the management and treatment of patients affected by ventilator-associated pneumonia (VAP). However, microbiological culture may be time-consuming and has a limited culturability of many potential pathogens. In this study, we developed a rapid nanopore-based metagenomic next-generation sequencing (mNGS) diagnostic assay for detection of VAP pathogens and antimicrobial resistance genes (ARGs). PATIENTS AND METHODS: Endotracheal aspirate (ETA) samples from 63 patients with suspected VAP were collected between November 2021 and July 2022. Receiver operating characteristic (ROC) curves were established to compare the pathogen identification performance of the target pathogen reads, reads percent of microbes (RPM) and relative abundance (RA). The evaluation of the accuracy of mNGS was performed comparing with the gold standard and the composite standard, respectively. Then, the ARGs were analyzed by mNGS. RESULTS: ROC curves showed that RA has the highest diagnostic value and the corresponding threshold was 9.93%. The sensitivity and specificity of mNGS test were 91.3% and 78.3%, respectively, based on the gold standard, while the sensitivity and specificity of mNGS test were 97.4% and 100%, respectively, based on the composite standard. A total of 13 patients were virus-positive based on mNGS results, while the coinfection rate increased from 27% to 46% compared to the rate obtained based on clinical findings. The mNGS test also performed well at predicting antimicrobial resistance phenotypes. Patients with a late-onset VAP had a significantly greater proportion of ARGs in their respiratory microbiome compared to those with early-onset VAP (P = 0.041). Moreover, the median turnaround time of mNGS was 4.43 h, while routine culture was 72.00 h. CONCLUSION: In this study, we developed a workflow that can accurately detect VAP pathogens and enable prediction of antimicrobial resistance phenotypes within 5 h of sample receipt by mNGS.