Abstract
PURPOSE: Cerebrospinal fluid (CSF) metagenomic next-generation sequencing (mNGS) has the potential to identify the majority of pathogens in a single test. Accurate pathogen identification is vital for central nervous system infection (CNSi). However, there are few related studies investigating in a municipal hospital. PATIENTS AND METHODS: A total of 52 suspected CNSi patients were retrospectively recruited in Xinxiang central hospital between July 2019 and April 2023. The diagnostic performance of CSF mNGS, conventional microbiological tests (CMT), and the combination of CSF mNGS and CMT were evaluated by comparing to the final diagnosis. RESULTS: Among 52 suspected CNSi patients, 35 were diagnosed as CNSi. In comparison to the final diagnosis, the area under curves (AUC) for CSF CMT, CSF mNGS, and the combination of CMT and mNGS for the diagnosis of CNSi were 0.56 (95% CI 0.4-0.72), 0.74 (95% CI 0.61-0.84), and 0.76 (95% CI 0.63-0.88), respectively. The sensitivities were 11.43% (95% CI 4.54%-25.95%), 48.57% (95% CI 32.99%-64.43%), and 51.43% (95% CI 35.57%-67.01%), respectively. The accuracy was 40.38 (95% CI 27.01%-54.90%), 65.38% (95% CI 50.91%-78.03%), and 67.31% (95% CI 52.89%-79.67%), respectively. Furthermore, based on CSF mNGS results, seven patients confirmed initial treatment, two escalated, and one de-escalated. Additionally, we identified the optimal cutoff values as 1.75 U/L for CSF adenosine deaminase (ADA), 75.44 U/L for CSF protein, and 185 mmH(2)O for CSF pressure, when these values were exceeded, CSF mNGS tended to yield positive results. CONCLUSION: CSF mNGS showed superior diagnostic performance in CNSi and hence could serve as a complementary tool to CMT and conjunctively guide the precision therapy. Additionally, the values for CSF ADA, protein and pressure could assist in predicting mNGS positive result. With technical improvements for mNGS sample processing to increase throughput and reduce costs, clinicians may use mNGS more widely in municipal hospital laboratories.