https://doi.org/10.4081/ecj.2026.15444
Metabolomic profiling using 1H NMR spectroscopy in emergency department patients with sepsis: association with SOFA severity scores and in-hospital mortality. A prospective observational study
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.
Published: 21 September 2026
Metabolomics offers mechanistic insights into the pathophysiology of sepsis and may complement conventional severity scoring. This prospective study evaluated the serum metabolomic profile of patients with sepsis presenting to the Emergency Department (ED) and its relationship with Sequential Organ Failure Assessment (SOFA) scores and in-hospital mortality. One hundred and fifty-five adults (≥18 years) meeting Sepsis-3 criteria were enrolled consecutively from a tertiary care ED between July 2023 and July 2024. Serum samples were collected at two standardised time points: admission (T1, 0 h) and 72 hours (T2). Samples were analysed using 1H NMR spectroscopy on an 800 MHz Bruker Avance III spectrometer. SOFA scores were recorded at T1 and T2. The primary outcome was in-hospital mortality. Associations between metabolites, SOFA scores, and mortality were examined using univariable non-parametric tests, Spearman correlation, and a parsimonious multivariable logistic regression model. All metabolite p-values are exploratory given the absence of formal multiplicity correction. Of 155 patients, 98 had sepsis and 57 had septic shock; 38 (24.5%) died in hospital. Baseline characteristics did not differ significantly between survivors and non-survivors. At T2, non-survivors had significantly lower levels of methanol (10.6 vs 18.1 µM, p=0.02), mannose (52.6 vs 72.1 µM, p=0.044), isoleucine (p=0.015), leucine (p=0.027), and valine (p=0.02), and higher pyruvate (56.2 vs 27.4 µM, p=0.011). No significant metabolite differences were observed at T1. SOFA score was associated with in-hospital mortality (OR 1.36, 95% CI 1.06–1.74, p=0.017; AUC 0.70, 95% CI 0.55–0.84). Distinct metabolic alterations at 72 hours — but not at ED admission — particularly in branched-chain amino acids, mannose, methanol, and pyruvate, were associated with in-hospital mortality in sepsis patients. These findings suggest that early metabolic trajectories may complement ongoing risk stratification after initial ED resuscitation, but require validation in larger, multicentre cohorts before clinical translation.
Downloads
1. Singer M, Deutschman CS, Seymour CW, et al. The Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3). JAMA 2016;315:801–10. DOI: https://doi.org/10.1001/jama.2016.0287
2. Ghatak T, Sajjad SA, Das A, et al. The diagnostic accuracy of serum procalcitonin for sepsis in adult emergency department patients: a systematic review and meta-analysis. CHEST Critical Care 2025;3:100179. DOI: https://doi.org/10.1016/j.chstcc.2025.100179
3. Shankar-Hari M, Phillips GS, Levy ML, et al. Developing a new definition and assessing new clinical criteria for septic shock. JAMA 2016;315:775–87. DOI: https://doi.org/10.1001/jama.2016.0289
4. Langley RJ, Tsalik EL, van Velkinburgh JC, et al. An integrated clinico-metabolomic model improves prediction of death in sepsis. Sci Transl Med 2013;5:195ra95. DOI: https://doi.org/10.1126/scitranslmed.3005893
5. Langley-Evans SC, Gardner DS, Jackson AA, et al. Metabolic profiling identifies key biochemical pathways altered in human sepsis. Crit Care Med 2015;43:2534–42.
6. Seymour CW, Yende S, Scott MJ, et al. Metabolomics in pneumonia and sepsis: an analysis of the GenIMS cohort study. Intensive Care Med 2013;39:1423–34. DOI: https://doi.org/10.1007/s00134-013-2935-7
7. Rogers AJ, McGeachie M, Baron RM, et al. Plasma metabolites in early sepsis identify distinct clusters of patients with differential outcomes. Crit Care Explor 2021;3:e0480. DOI: https://doi.org/10.1097/CCE.0000000000000478
8. Mosevoll KA, Gundersen T, Skrede S, et al. Metabolic profiling of Sepsis-3 patients compared to SIRS and Sepsis-2 patients. Biomolecules 2023;13:223. DOI: https://doi.org/10.3390/biom13020223
9. Pandey N, Kumar S, Singh R, et al. Metabolomics in sepsis: a comprehensive review of clinical and experimental studies. Clin Exp Emerg Med 2025;12:4–15. DOI: https://doi.org/10.15441/ceem.24.211
10. Evans L, Rhodes A, Alhazzani W, et al. Surviving Sepsis Campaign: international guidelines for management of sepsis and septic shock 2021. Intensive Care Med 2021;47:1181–247. DOI: https://doi.org/10.1007/s00134-021-06506-y
11. Chen Q, Liang X, Wu T, et al. Integrative analysis of metabolomics and proteomics reveals amino acid metabolism disorder in sepsis. J Transl Med 2022;20:123. DOI: https://doi.org/10.1186/s12967-022-03548-8
12. Peng J, Qiu C, Zhang J, Xiao X. Serum metabolite profiling reveals metabolic characteristics of sepsis patients using LC/MS-based metabolic profiles. BMC Med Genomics 2023;16:224. DOI: https://doi.org/10.1186/s12920-023-01666-w
13. Feng S, Cui N, Zhao W, et al. Prognostic biomarkers for sepsis mortality based on the literature and LC-MS-based metabolomics. Am J Transl Res 2023;15:5757–68.
Ethics Approval
CRediT authorship contribution
Ratender Kumar Singh, Tanmoy Ghatak, and Puja Gautam contributed to study conception and design, ethical approval, and patient evaluation, screening, and sample collection. Puja Gautam additionally compiled the clinical and demographic data for all patients from the institutional digital records. Gurvinder Singh and Pragati Gupta prepared samples for NMR analysis and acquired, processed, and analyzed the NMR spectral data. Dinesh Kumar performed the multivariate statistical analysis. Prabhakar Mishra performed the univariate statistical analysis. All authors reviewed and approved the final manuscript.
Supporting Agencies
Data Availability Statement
All data generated or analyzed during this study are included in this published article.
How to Cite

This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
PAGEPress has chosen to apply the Creative Commons Attribution NonCommercial 4.0 International License (CC BY-NC 4.0) to all manuscripts to be published.




