Cerebrospinal fluid lipidome in central nervous system infections: a study of diagnostic accuracy
In brief
A cerebrospinal fluid lipid marker scored 0.98 for bacterial meningitis detection
In the discovery cohort, elevated Hex2Cer [d] distinguished bacterial meningitis from other diagnoses with an area under the curve of 0.98, with similar results in internal and independent Danish validation cohorts. It also performed best among patients with intermediate cerebrospinal fluid white-cell counts. The findings support further testing, but do not yet establish how the marker would perform in routine care.
- Journal
- EBioMedicine (Q1)
- Published
- 17 November 2026
- Study design
- Prospective / inception cohort
- Evidence level
- Level 2, Moderate (CEBM 2b)
- Authors
- Steven L Staal, Sabine E Olie, Jan-Bert van Klinken, Michel van Weeghel, Frédéric M Vaz, Jacob Bodilsen, et al.
- PMID
- 43044421
- DOI
- 10.1016/j.ebiom.2026.106558
Why clinicians should know about it
- Picked for Biochemistry (medical) (top studies of the week, 22 November 2026): Human CSF lipid biomarkers validated for infection diagnosis
Abstract
BACKGROUND: Diagnosing central nervous system (CNS) infections can be challenging when no pathogen is detected, emphasising the need for new diagnostics. METHODS: We assessed the diagnostic accuracy of lipids in cerebrospinal fluid (CSF) for diagnosing CNS infections. Episodes were derived from four prospective community-acquired CNS infection cohorts. Three Dutch cohorts were used for the discovery and internal validation. An independent Danish cohort was used for external validation. All studies included adults with confirmed or suspected CNS infections who underwent a diagnostic lumbar puncture. Lipidomics, the index test, was conducted on CSF using liquid chromatography high resolution mass spectrometry. Microbiologically confirmed diagnoses were the reference standard. FINDINGS: In total, 229 episodes were included in the discovery cohort. A CNS infection occurred in 81/229 (35%) episodes and consisted of bacterial meningitis in 39 (17%), viral meningoencephalitis in 31 (14%) and other CNS infections in 11 (5%). We analysed 626 lipid species across 33 classes and identified a distinct profile for bacterial meningitis. Elevated Hex2Cer [d] showed the highest diagnostic accuracy for bacterial meningitis: area under the curve (AUC) = 0·98 (95% CI 0·97-1·00) when compared to all other diagnoses and AUC = 0·98 (95% CI 0·96-1·00) when compared to viral meningoencephalitis. Among episodes with CSF leucocytes between 5 and 1000 cells/mm3, Hex2Cer [d] remained the best predictor. These findings were supported by the internal and external validation cohorts. INTERPRETATION: Lipid alterations in CSF, particularly Hex2Cer [d], have high diagnostic accuracy for bacterial meningitis, highlighting their potential as a diagnostic biomarker. FUNDING: European Research Council; Netherlands Organisation for Health Research and Development; Danish Independent Research Foundation; Lundbeck Foundation.
Abstract as published, via PubMed.
For healthcare professionals. The summary is generated by AI from the published abstract, and the evidence level is assigned automatically from the study design on the Oxford CEBM hierarchy. Neither is medical advice. Read the full paper before changing practice.