Here we aim to build up a metagenomics-centered surveillance on the infectious microbiome showing in the fever of unknown origin (FUO) patients. We collected venous blood, bronchoalveolar lavage fluid, cerebrospinal fluid, tissue block, sputum, bone marrow biopsy, and purulent liquid samples from 123 patients. Metagenomic sequencing (mNGS) for both DNA and RNA sequences was performed to profile the total pathogenic microbiome in the samples. A large pool of infectious or conditional infectious bacteria was found, belonging to Enterobacteriaceae, Staphylococcaceae (10.55%), Burkholderiaceae (10.05%), and Comamonadaceae (4.25%). The major virus families detected from mNGS analysis include Adenoviridae, Anelloviridae, Peribunyaviridae, Flaviviridae, and Herpesviridae, showing up in 34.96%, 47.37%, 30.89%, 5.69%, 3.25%, and 1.63% of patients, respectively. Using the Ward clustering method, two clusters of patients were organized: high-variety group and low-variety group. The patients in the high-variety group demonstrated higher levels of immune cells and inflammatory indicators such as lactate dehydrogenase, aspartate aminotransferase, and alanine aminotransferase. The patients in the low-variety group showed higher levels of inflammatory lipids such as 13,14-dihy-15-keto PGE2 (fold > 10, P = 0.021); tetra-PGDM (fold = 5.29, P = 0.037); and 20-HETE (fold > 10, P = 0.02). The mNGS surveillance system demonstrated remarkable potential in preventing infectious diseases using mNGS data.
Keywords: bacterial infection; diagnosis; disease processes; metagenomics; microbiome.
© The Author(s) 2023. Published by Oxford University Press on behalf of Applied Microbiology International.