Abstract
AbstractBackgroundPrevious metabarcoding studies based on 16S rRNA sequencing in patients with extrapulmonary sepsis have found early pulmonary dysbiosis associated with a poor prognosis. To further discern this association, here we aimed to better characterize the pulmonary bacterial communities in these patients by leveraging metagenomics and to evaluate if the presence of antibiotic resistance genes (ARGs) could explain the higher mortality of the patients.Material and methodsMetagenomic sequencing was performed using the Nextera XT Library Prep Kit and HiSeq 4000 (Illumina Inc.) on tracheal aspirate samples that were obtained within 24 h from diagnosis from patients with extrapulmonary sepsis admitted to the Intensive Care Unit (ICU). Analysis involved MetaSpades for contig assembly, Kraken2 and Metaphlan4 for taxonomic classification, and CARD and GTDB-tk for ARGs annotation and assignment to the bacterial species. The relationship between the presence of antibiotic resistance and ICU mortality was evaluated using the Wilcoxon test and logistic regression models adjusting for clinical and demographic variables.ResultsIn total, 127 different ARGs were detected circumscribed only to seven patients. The most common ARGs found were from the antibiotic groups of aminoglycosides and beta-lactams, both present in most of the patients. These ARGs were found, almost entirely linked toKlebsiella pneumoniae, Escherichia coli, andPseudomonas aeruginosa. The results also show a significant enrichment of ARGs among patients who died while admitted in the ICU (57%, 95% confidence interval [CI]: 18-90%) compared to surviving patients (20%, 95% CI: 7-40%) (p=0.022). Analyses adjusting for clinical and demographic variables did not alter this result.ConclusionMetagenomic sequencing has allowed an unprecedented characterization of the sepsis lung microbiome showing that antibiotic resistance is common among these patients. The results also suggest a relationship between the early accumulation of ARGs in the lung of patients with extrapulmonary sepsis who die while admitted in the ICU. Studies in independent samples will be needed to validate our findings.
Publisher
Cold Spring Harbor Laboratory