Decoding the delirium proteome: linking clinical vulnerabilities to acute brain dysfunction
The Journals of Gerontology, Series A: Biological Sciences and Medical Sciences, 2026
Wang X., Liu K., Guo X., Li N., Li Z.
| Disease area | Application area | Sample type | Products |
|---|---|---|---|
Neurology | Pathophysiology | Plasma |  Olink Explore 3072/384 |
Abstract
Background
Delirium is an acute brain dysfunction syndrome with high morbidity and mortality, however its mechanisms remain poorly understood. Although multiple risk factors have been identified, directly targeting them is challenging. Proteins may provide new insights into this challenge.
Methods
Using data from the UK Biobank, we explored associations between 18 established delirium risk factors and incident delirium using Cox proportional hazards models. Plasma proteomic profiling was conducted to identify proteins associated with delirium. Identified proteins underwent enrichment and mediation analyses to evaluate their potential involvement in the associations between risk factors and delirium.
Results
Over a median 13.5-year follow-up, 701 participants developed delirium. Among 2915 plasma proteins, 78 were significantly associated with delirium after Bonferroni correction, including Neurofilament light polypeptide (NEFL), Glial fibrillary acidic protein, WAP four-disulfide core domain protein 2 (WFDC2), Urokinase plasminogen activator surface receptor, and Growth/differentiation factor 15 (GDF15). Enrichment analysis suggested that these proteins were predominantly involved in immune, inflammatory, and metabolic processes. Mediation analysis indicated that some proteins may statistically mediate the associations between clinical risk factors and incident delirium; for example, NEFL (11.7%) and EDA2R (10.4%) for age, PLAUR (47.2%) and Mesothelin (MSLN; 40.1%) for smoking, GDF15 (40.6%) and C7 (16.3%) for diabetes, WFDC2 (13.4%) and PLAUR (13.1%) for frailty. Overall, 24 proteins statistically mediated the associations between at least 9 risk factors and delirium, with GDF15, PLAUR, Amphiregulin, Hepatitis A virus cellular receptor, Occludin, and, showing relatively higher statistical estimates.
Conclusions
This large-scale study identifies plasma proteins that may help explain some of the observed associations between clinical risk factors and incident delirium, generating hypotheses about underlying mechanisms and potential avenues for further research.