Diagnostic Performance and Clinical Utility of Metagenomic Next-Generation Sequencing in Suspected Central Nervous System Infections: A Prospective Comparative Study

Cao Xiaoguang^1*xin jia hu²jian chong huang^3*min shao^4*MENG HUADONG^5*

• ¹Anhui Provincial Hospital, Hefei, China
• ²Huashan Hospital, Fudan University, Shanghai, Shanghai Municipality, China
• ³Suzhou Municipal Hospital, Suzhou, Jiangsu Province, China
• ⁴First Affiliated Hospital of Anhui Medical University, Hefei, Anhui Province, China
• ⁵Hefei No.1 People's Hospital, Hefei, Anhui Province, China

Objective: To assess the diagnostic performance and clinical utility of metagenomic next-generation sequencing (mNGS) in patients with suspected central nervous system (CNS) infections. Methods:A prospective study was conducted from December 2019 to January 2024, enrolling 110 patients with suspected CNS infections. Cerebrospinal fluid (CSF) samples were subjected to mNGS, conventional biochemistry, and culture. Clinical features and outcomes were compared between patients confirmed with CNS infections and those without. Results: Of the enrolled patients, 69 were diagnosed with CNS infections. mNGS identified pathogens in 62 cases (77.11%), including 54 clinically confirmed true positives (49.09%), significantly surpassing traditional CSF culture (6.36%). mNGS reported results within 24 hours, considerably shorter than the 72~120 hours required for culture. Compared to the non-infection group, patients with CNS infections had significantly higher ICU admission(ICUA) rates, prolonged hospital stays, increased healthcare costs, and elevated rates of antibiotic adjustment and mNGS positivity (P<0.05). CSF turbidity, cell count, and protein levels were significantly elevated, while glucose and chloride levels were reduced. Logistic regression identified mNGS, CSF protein, and glucose levels as independent predictors of CNS infection. Receiver operating characteristic (ROC) analysis demonstrated superior diagnostic accuracy for continuous CSF variables over binary ones, with mNGS showing robust performance [area under the curve (AUC) = 0.794]. Conclusion: mNGS offers rapid and accurate pathogen detection, outperforming conventional methods in sensitivity and turnaround time, and provides valuable guidance for individualized antimicrobial treatment in CNS infections. Key word：CNS infections，mNGS， CSF analysis，diagnostic accuracy；pathogen detection.