Wang Bingqing, Xu Fei, Zhang Minheng
Department of Neurology, Taiyuan Central Hospital, Taiyuan, Shanxi, China.
Department of Gerontology, The First People's Hospital of Jinzhong, Jinzhong, Shanxi, China.
Front Med (Lausanne). 2025 Jun 5;12:1522028. doi: 10.3389/fmed.2025.1522028. eCollection 2025.
Emerging evidence suggests insulin resistance may contribute to neurodegeneration, yet its role in non-diabetic populations remains unclear. This study explores the relationship between estimated glucose disposal rate (eGDR), a measure of insulin sensitivity, and incident cognitive dysfunction in non-diabetic adults.
Our longitudinal analysis utilized data from 5,178 CHARLS participants (age ≥ 45 years). Insulin sensitivity was quantified using eGDR, calculated from waist circumference, hypertension status, and hemoglobin A1c levels. Participants were stratified by eGDR quartiles for comparative analysis. We employed multivariable Cox models, survival curves, restricted cubic splines, and sensitivity testing to evaluate associations with cognitive outcomes.
Over an 8.7-year follow-up, cognitive dysfunction developed in 36.9% of participants. Analyses revealed significant metabolic-cognitive associations, with each standard deviation increase in eGDR linked to a 15.8% reduction in risk (adjusted hazard ratio [HR] = 0.792, 95% confidence interval [CI]: 0.793-0.881). Restricted cubic spline analysis identified non-linear threshold effects, with risk accelerating below certain eGDR levels ( < 0.05). Kaplan-Meier survival analysis demonstrated significant differences in cognitive impairment incidence across eGDR quartiles ( = 0.003). Additionally, both eGDR and metabolic score for insulin resistance (METS-IR) showed comparable predictive value for cognitive impairment risk, outperforming other metabolic indices, including the atherogenic index of plasma (AIP), and the triglyceride glucose index (TyG).
These findings position eGDR as a promising biomarker for cognitive risk stratification in non-diabetic adults. However, further multi-database studies should validate these associations and explore the underlying mechanisms.
新出现的证据表明胰岛素抵抗可能导致神经退行性变，但其在非糖尿病人群中的作用仍不明确。本研究探讨了估计葡萄糖处置率（eGDR，一种胰岛素敏感性指标）与非糖尿病成年人新发认知功能障碍之间的关系。
我们的纵向分析利用了来自5178名中国健康与养老追踪调查（CHARLS）参与者（年龄≥45岁）的数据。胰岛素敏感性通过eGDR进行量化，该指标由腰围、高血压状态和糖化血红蛋白水平计算得出。参与者按eGDR四分位数分层进行比较分析。我们采用多变量Cox模型、生存曲线、受限立方样条和敏感性测试来评估与认知结局的关联。
在8.7年的随访中，36.9%的参与者出现了认知功能障碍。分析显示存在显著的代谢 - 认知关联，eGDR每增加一个标准差，风险降低15.8%（调整后风险比[HR]=0.792，95%置信区间[CI]：0.793 - 0.881）。受限立方样条分析确定了非线性阈值效应，在特定eGDR水平以下（<0.05）风险加速上升。Kaplan - Meier生存分析表明，eGDR四分位数之间的认知障碍发生率存在显著差异（P = 0.003）。此外，eGDR和胰岛素抵抗代谢评分（METS - IR）对认知障碍风险显示出相当的预测价值，优于其他代谢指标，包括血浆致动脉粥样硬化指数（AIP）和甘油三酯葡萄糖指数（TyG）。
这些发现表明eGDR是用于非糖尿病成年人认知风险分层的一个有前景的生物标志物。然而，进一步的多数据库研究应验证这些关联并探索潜在机制。
