TyHGB combined with diagonal earlobe crease and traditional risk factors for constructing a diagnostic model of coronary heart disease

TyHGB is a novel insulin resistance (IR)-related indicator, and its association with coronary heart disease (CHD) remains unclear. Additionally, studies have shown a close correlation between the diagonal earlobe crease (DELC) and CHD, yet it has not been fully applied in clinical practice to date. Therefore, this study constructed and validated a diagnostic model for CHD by combining TyHGB and DELC. A total of 1664 patients suspected of CHD who underwent coronary angiography (CAG) in the Department of Cardiology, Chengde Central Hospital from September 2021 to April 2025 were recruited for this study. Participants were categorized into a CHD group (n = 1,316) and a non-CHD group (n = 348). Univariate and multivariate logistic regression analyses were employed to assess the associations of TyHGB, DELC, and other relevant factors with CHD. Restricted cubic spline (RCS) analysis was employed to assess the dose-response correlation between TyHGB and the risk of CHD, and TyHGB levels were compared across quartiles. A nomogram incorporating TyHGB and DELC was developed, and its diagnostic performance was assessed using decision curve analysis (DCA) and receiver operating characteristic (ROC) curves. Age, sex, hypertension, diabetes, CR, Lp(a), TyHGB, and DELC were identified as independent risk factors for CHD through multivariate logistic regression analysis (P < 0.05). RCS curve analysis demonstrated a linear relationship between TyHGB and CHD risk (P for nonlinear = 0.477). Further quartile grouping of TyHGB revealed a progressive increase in CHD risk with rising TyHGB levels. Compared to the first quartile group, patients in the fourth quartile group exhibited a 2.087-fold higher risk of CHD (95% CI: 1.374–3.170). Optimal predictive variables were selected in the training set using Lasso regression and multivariate logistic regression. ROC curves were utilized to validate diagnostic performance, with the training set yielding an area under the curve (AUC) of 0.706 (95% CI: 0.669–0.743) and the validation set an AUC of 0.716 (95% CI: 0.657–0.775). Both TyHGB and DELC have been identified as independent risk factors for CHD, with a linear relationship observed between TyHGB levels and CHD risk. A diagnostic model for CHD, developed by integrating TyHGB, DELC, and traditional risk factors, demonstrates strong diagnostic efficacy.