Journal Article

Serum copper modulates cognitive function in diabetic patients via an HDL-C-mediated pathway: identification of a 25 µg/dL exploratory threshold

Zhou, Jianlong, Jiang, Yayi, Yue, Rensong, Shi, Wenxiang, Li, Yadi

2025

Overview

Background Diabetes-associated cognitive dysfunction (DACD), a prevalent complication of diabetes with learning, memory, and executive function impairments, lacks targeted therapeutic options. While trace elements, oxidative stress, and inflammation are linked to DACD, the role of serum copper and its interaction with inflammatory/oxidative biomarkers in cognitive regulation remains unclear in diabetic populations. Methods This cross-sectional study analyzed data from the National Health and Nutrition Examination Survey (NHANES) 2011–2014, including 1,149 participants (861 non-diabetic, 288 diabetic). Cognitive function was assessed via the Animal Fluency Test (AFT) and Digit Symbol Substitution Test (DSST). Serum copper levels were measured, alongside inflammatory indices: neutrophil-to-lymphocyte ratio, monocyte-to-lymphocyte ratio, neutrophil-to-monocyte-lymphocyte ratio, systemic immune-inflammation index and systemic inflammation response index; and oxidative stress markers: γ-glutamyl transferase, uric acid, and high-density lipoprotein cholesterol (HDL-C). Associations were analyzed using multivariable linear regression, causal mediation analysis, restricted cubic spline models and sex/age subgroup stratification. Results Diabetic participants had lower DSST scores than non-diabetic individuals ( P  < 0.001). In diabetic participants, serum copper was negatively associated with AFT scores (β = − 0.132, P  = 0.034) and positively correlated with HDL-C (β = 0.559, P  = 2.11e-06). HDL-C was the sole factor that statistically mediated the association between serum copper and DSST scores (average causal mediation effect = 0.095, 95% CI: 0.046–0.153, P  < 0.001). A non-linear relationship emerged: HDL-C remained stable at serum copper < 20 µg/dL but increased significantly when copper exceeded 25 µg/dL ( P  < 0.001). Stratified analyses revealed threshold heterogeneity (all P  < 0.05): males had a lower serum copper threshold (24.5 µg/dL, 95% CI: 21.8–27.2) than females (26.1 µg/dL, 95% CI: 23.4–28.8), and adults ≥ 65 years had a higher threshold (27.3 µg/dL, 95% CI: 24.5–30.1) than those < 65 years (23.8 µg/dL, 95% CI: 21.1–26.5). Conclusions This study identifies a diabetes-specific statistical association between serum copper, HDL, and cognitive function in DACD. The 25 µg/dL copper threshold (exploratory inflection point) marks where HDL-C-mediated effects become prominent, while sex- and age-specific threshold differences highlight population heterogeneity. This threshold offers a reference for trace element-lipid interaction research but requires validation in independent cohorts before potential use in DACD risk stratification.

• Publisher: BioMed Central,BioMed Central Ltd,Springer Nature B.V,BMC
• Subject: Alzheimer's disease / Analysis / Biomarkers / Care and treatment / Cholesterol / Cognition & reasoning / Cognition disorders / Cognitive ability / Copper / Dementia / Development and progression / Diabetes / Diabetes mellitus / Diabetes-associated cognitive dysfunction / Diabetics / Disease / Endocrinology / Executive function / Glucose / Health aspects / High density lipoprotein / High density lipoprotein cholesterol / Inflammation / Leukocytes (neutrophilic) / Lipids / Lymphocytes / Mediation / Medicine / Medicine & Public Health / Metabolic Diseases / Monocytes / Negotiation, mediation and arbitration / Neutrophils / Older people / Oxidative stress / Population / Population studies / Serum copper / Standard deviation / Surveys / Threshold effect / Trace elements / Trace elements (nutrients) / Type 2 diabetes / Uric acid