ePrints

Simple hyperinflammation scores predict mortality in hospitalized patients with COVID-19 and offer a personalized medicine approach to dexamethasone intervention

Lookup NU author(s): Trevor Liddle, Professor Matthew CollinORCiD

Downloads

• Published version [.pdf]

Licence

This work is licensed under a Creative Commons Attribution 4.0 International License (CC BY 4.0).

Abstract

© 2025 The Authors.Background Dexamethasone is recommended for use in all patients with COVID-19 requiring supplemental oxygen, however, only some patients develop hyperinflammation (COV-HI) potentially influencing their response to corticosteroids. This study tested the ability of criteria defining COV-HI to predict response to dexamethasone. Methods A retrospective, multicentre, observational cohort study of 1313-patients with PCR-confirmed COVID-19 during first and second waves of community-acquired infection including 212 patients who received dexamethasone monotherapy. Demographic data, laboratory tests and clinical status were recorded from admission until death or discharge, with minimum 28-days follow-up. Patients were stratified at admission as COV-HI-YES/COV-HI-NO based on three published COV-HI definitions. Results Patients with COV-HI shared a biological phenotype of hypoalbuminemia/anemia, and elevated D -dimer/lactate dehydrogenase/alanine transaminase/respiratory rates. Combining these features predicted 28-day mortality and stratified COV-HI-YES from COV-HI-NO more effectively compared to individual markers/demographic features alone. In COV-HI-YES patients, dexamethasone treatment halved mortality-risk (relative risk = 0.50) compared to untreated patients. However, in COV-HI-NO patients mortality-risk was 3.03x higher (CI = 1.3-7.0) in treated versus untreated patients during a 28-day admission period. Conclusions We present a framework for a new machine-learning based scoring system for COV-HI combining clinical assessment with laboratory markers for prediction of mortality and targeting glucocorticoids in hospitalized COVID-19 patients.

Publication metadata

Author(s): Oppong AE, Coelewij L, Hutchinson M, Carpenter B, Robinson GA, Liddle T, Hawkins E, Cox MF, Ciurtin C, Venkatachalam S, Collin M, Tattersall RS, Ardern-Jones M, Duncombe AS, Jury EC, Manson JJ

Publication type: Article

Publication status: Published

Journal: International Journal of Infectious Diseases

Year: 2025

Volume: 161

Print publication date: 05/11/2025

Online publication date: 13/10/2025

Acceptance date: 09/10/2025

Date deposited: 27/11/2025

ISSN (print): 1201-9712

ISSN (electronic): 1878-3511

Publisher: Elsevier

URL: https://doi.org/10.1016/j.ijid.2025.108119

DOI: 10.1016/j.ijid.2025.108119

PubMed id: 41093003

Altmetrics

Funding

Share