Abstract
BACKGROUND: The diagnostic value of metagenomic next-generation sequencing (mNGS) in central nervous system (CNS) infectious diseases after empirical treatment has not been reported. AIM: To investigate the diagnostic value of mNGS of cerebrospinal fluid (CSF) in the empirically treated CNS infectious diseases. METHODS: A total of 262 CSF samples from patients with suspected CNS infections were collected between August 2020 and December 2021. Both mNGS and conventional methods were used for testing. The conventional methods included microbial culture, smear, polymerase chain reaction, etc. RESULTS: Among 262 suspected cases, 183 cases (69.84%) were diagnosed as CNS infection, including 86 cases of virus infection (47.00%), 70 cases of bacterial infection (38.25%) and 27 cases of fungal infection (14.76%). The sensitivity and specificity of mNGS were 65.6% (95%CI: 58.2%-72.3%) and 89.6% (95%CI: 79.1%-95.3%), respectively. The PPV of mNGS was 94.5% (95%CI: 88.6%-97.6%), and the NPV was 48.8% (95%CI: 39.7%-57.9%). The pathogen detective sensitivity and accuracy of mNGS were higher than those of conventional methods (Sensitivity: 65.6% vs 37.2%; P < 0.001; Accuracy: 72.0% vs 50%, P < 0.001). The results showed that compared with conventional methods, mNGS technology was a more sensitive method for the diagnosis of CNS infection after empirical treatment. CONCLUSION: mNGS can be a better method applied in the diagnosis of CNS infection after empirical treatment.