Insulin resistance and its association with coronary heart disease : insights from bioinformatics analysis

Xinyue Lu¹Minghuan Liu¹Yang Song¹Shiying Lin¹Xiaowan Zhou¹Hang Zeng¹Aina Li²Huangyuan Li¹Xiaoxu Xie¹Shaowei Lin¹Siying Wu^1*

• ¹Fujian Medical University, Fuzhou, China
• ²The First Affiliated Hospital of Fujian Medical University, Fuzhou, China

Background: Evidence on the relationship between no-insulin-based insulin resistance (IR) surrogates and the prevalence of coronary heart disease (CHD) in remains limited. Here, we assess the associations between multiple non–insulin-based IR surrogates and CHD, delineating subgroups at heightened susceptibility across demographic and cardiometabolic strata. We further explore plausible mechanistic pathways and perform mediation analyses to elucidate the pathophysiological links between IR and CHD. Methods: This study analyzed 7,419 participants from Fujian Medical University hospitals (2018–2024). Multivariate logistic regression and restricted cubic splines (RCS) were performed to evaluate the relationship between the atherogenic index of plasma (AIP index), triglyceride glucose index (TyG index), the metabolic score for insulin resistance (METS-IR), and triglyceride(TG) /high density lipoprotein cholesterol(HDL-C) ratio (TG/ HDL-C ratio) and the prevalence of CHD. Subgroup analysis and interaction analysis were employed to identify effect modifiers associated with the IR-CHD relationship, whilst bioinformatics analysis and mediation analysis were utilised to identify and quantify potential mediating effects within this association. Results :In primary analyses, surrogate markers of IR were positively associated with the prevalence of CHD, with odd ratios ranging between 1.039 and 1.612. Moreover, physical inactivity was the predominant effect modifier (interaction P-value < 0.05), with a possible additional influence of younger age, so that sedentary adults <60 years with elevated IR surrogates appeared particularly vulnerable. Mechanistically, bioinformatic enrichment analyses highlighted three core pathways, including AGE–RAGE signaling in diabetic complications, FoxO signaling, and adipocytokine signaling, providing context for the observed patterns. Finally, mediation analyses indicated that fasting plasma glucose mediated 8.15%–55.33% of the associations, including AIP index, TyG index, METS-IR, and ln(TG/HDL-C). Conclusion:In this large hospital-based cross-sectional sample, non–insulin-based IR surrogates were positively associated with the prevalence of CHD and may help identify groups with a higher likelihood of having CHD. Glycemic pathways and physical inactivity emerged as important correlates of these associations and deserve further evaluation in longitudinal studies.