Dual-Biofluid Metabolomic Profiling by MALDI-MS Enables Noninvasive Diagnosis and Prognosis of Central Nervous System Infections

Central nervous system (CNS) infections represent a common but serious complication following neurosurgical procedures. However, the metabolic alterations induced by infection remain incompletely understood, and noninvasive monitoring strategies remain urgently needed. In this study, we conducted a systematic metabolomic analysis of two biofluids─cerebrospinal fluid (CSF) and urine─obtained from patients undergoing neurosurgical procedures, using matrix-assisted laser desorption/ionization mass spectrometry (MALDI-MS). Metabolomic analysis of CSF revealed significant metabolic reprogramming associated with CNS infections, primarily characterized by the dysregulation of energy metabolism, amino acid metabolism, lipid and membrane-related pathways. Interestingly, similar pathway perturbations were also observed in the urinary metabolome. We further observed a significant comprehensive correlation between the urinary and CSF metabolomes. Then, machine learning models were successfully constructed based on urinary metabolomic features for noninvasive disease diagnosis and prognostic evaluation. Both the diagnostic and prognostic models achieved satisfactory classification performance, with area under the receiver operating characteristic curve values greater than 0.9. This work presents a novel, rapid, and noninvasive strategy for the detection and clinical management of CNS infections, which may facilitate timely and accurate clinical intervention and ultimately improve patient outcomes.