A prospective study of nanopore-targeted sequencing in the diagnosis of central nervous system infections

Central nervous system (CNS) infections are a leading cause of death in patients. Nanopore-targeted sequencing (NTS) has begun to be used for pathogenic microbial detection. This study aims to evaluate the ability of NTS in the detection of pathogens in cerebrospinal fluid (CSF) through a prospective study. Fifty CSF specimens collected from 50 patients with suspected CNS infections went through three methods including NTS, metagenomic next-generation sequencing (mNGS), and microbial culture in parallel. When there was an inconsistency between NTS results and the results of the mNGS, the 16S rDNA gene was amplified followed by Sanger sequencing to further verify pathogens detected by NTS. Among 50 CSF specimens, 76% were NTS-positive, which is lower than mNGS (94.0%), yet higher than microbial culture (16.0%). The overall validation rate, diagnostic accordance rate (DAR), sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of NTS were 86.7%, 50.0%, 71.0%, 15.8%, 57.9%, and 25.0%, respectively. In the CSF total nucleated cell (TNC) number <= 10 cells/mu L, DAR, specificity, and PPV were 20%, 11.1%, and 11.1%, whereas in that with CSF TNC number >10 cells/mu L, DAR, sensitivity, specificity, PPV, and NPV were 57.5%, 70.0%, 20.0%, 72.4%, and 18.2%, respectively. Although NTS has a higher microbial detection rate than microbial culture, it should combine CSF TNC result to evaluate the value of NTS for the diagnosis of CNS infections.