Serum hepcidin as a host-derived biomarker of disease severity and risk stratification in tuberculosis
In brief
Hepcidin levels were higher in severe tuberculosis but did not predict death
In a prospective cohort of 140 adults with tuberculosis, median hepcidin was about 202 ng/mL in severe disease versus 146 ng/mL in non-severe disease, but it was not linked to 3-month mortality. Creatine kinase-MB and a combined creatine kinase-MB and adenosine deaminase measure better separated survivors from non-survivors; these findings need validation in larger, multicentre studies.
- Journal
- Infection (Q1)
- Published
- 15 October 2026
- Study design
- Prospective / inception cohort
- Evidence level
- Level 2, Moderate (CEBM 2b)
- Authors
- Utkarsh Kaushik, Durga Shankar Meena, Sojit Tomo, Gopal Krishana Bohra, Deepak Kumar, Sarika Kombade
- PMID
- 42878786
- DOI
- 10.1007/s15010-026-03020-5
Why clinicians should know about it
- Picked for Epidemiology (top studies of the week, 18 October 2026): Prospective cohort linking hepcidin levels to TB severity
- Picked for Biochemistry (medical) (top studies of the week, 18 October 2026): Serum hepcidin as biomarker of TB disease severity
Abstract
BACKGROUND: Hepcidin is a key regulator of systemic iron metabolism. Hepcidin may reflect the inflammatory activity and disease burden in tuberculosis. However, prospective evidence on the prognostic value of serum hepcidin and other host-biomarkers in tuberculosis is limited. We evaluated the association of hepcidin with disease severity and outcomes. METHODS: This prospective cohort study enrolled consecutive adults (age ≥ 18 years) with newly diagnosed tuberculosis. Baseline serum hepcidin, adenosine deaminase (ADA), and creatine kinase-MB (CK-MB) levels were measured. Disease severity was assessed using a predefined tuberculosis severity score. The prognostic performance of biomarkers was evaluated using receiver operating characteristic analysis and multivariable logistic regression. RESULTS: Among 140 patients (median age, 48 years; 67.1% male), 25% had pulmonary tuberculosis, 25% extrapulmonary tuberculosis, and 50% disseminated disease; HIV co-infection was present in 16.4%. The 3-month all-cause mortality rate was 22.1%. Serum hepcidin levels were significantly higher in severe than non-severe tuberculosis (201.5 vs. 145.9 ng/mL; p = 0.017), but were not associated with 3-month mortality. Conversely, CK-MB was significantly elevated among non-survivors (24.5 vs. 13.8 U/L; p < 0.001). CK-MB showed good predictive performance for mortality (AUC 0.795), while ADA demonstrated moderate discrimination (AUC 0.647). The ADA/CK-MB combination showed the highest discriminative performance for mortality prediction (AUC 0.82). Triple-marker panel (Hepcidin + ADA+CK-MB) provided the best discrimination for severe disease (AUC 0.717). CONCLUSIONS: Serum hepcidin was associated with disease severity in tuberculosis but was not associated with 3-month mortality. Combined biomarker models showed promising prognostic performance, warranting further validation in larger multicentre studies.
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.