Association of the hs-CRP-TyG index with coronary artery disease risk and angiographic severity: a retrospective comparative study with the TyG index

Figures

• 

  Figure 1. Restricted cubic spline plots showing the nonlinear association between the hs-CRP-TyG index and the risk of coronary artery disease. Solid lines represent the estimated odds ratios (ORs); shaded areas indicate 95% confidence intervals (CIs). The horizontal dashed line indicates OR = 1.0. The clinical interpretation of this threshold effect, where risk significantly increases beyond a value of 1.42, is discussed in detail in the Discussion section. Models 1–4 represent progressively adjusted regression models, ranging from the unadjusted model (Model 1) to the fully adjusted model including all covariates (Model 4).

• 

  Figure 2. Subgroup analysis of the association between hs-CRP-TyG index and risk of CAD. hs-CRP-TyG, hs-CRP–Triglyceride-Glucose; CAD, coronary artery disease; DM, diabetes mellitus; Echo_EF, echocardiographic ejection fraction; Fatty liver, presence of fatty liver disease; OR, odds ratio; CI, confidence interval. Note: All models were adjusted for age, sex, smoking status, alcohol consumption, hypertension, diabetes, lipid-lowering medication use, uric acid, lymphocyte ratio, fibrinogen, and HDL-C unless otherwise specified in subgroup stratification.

• 

  Figure 3. Calibration plots for predicted vs. observed probability of CAD across four models. This plot assesses the calibration accuracy of the four logistic regression models (Model 1 to Model 4) by comparing predicted probabilities (x-axis) against observed outcomes (y-axis). The diagonal dotted line represents perfect calibration (predicted = observed). Curves closer to the diagonal indicate better calibration. CAD, coronary artery disease; Models 1–4 represent progressively adjusted regression models, ranging from the unadjusted model (Model 1) to the fully adjusted model including all covariates (Model 4).

• 

  Figure 4. Decision curve analysis evaluating the clinical utility of the hs-CRP-TyG index for predicting CAD. The x-axis represents the threshold probability, defined as the minimum probability of disease at which a clinician would recommend further intervention. The y-axis represents the net benefit. The DCA curves for the hs-CRP-TyG models (Models 1–4) are plotted alongside the "Treat All" and "Treat None" strategies. The results show that using the hs-CRP-TyG index to guide decision-making yields a higher net benefit than the "Treat All" or "Treat None" strategies across a moderate range of threshold probabilities. This suggests that the index has potential clinical utility for risk stratification, aiding in the identification of high-risk patients who may benefit from further evaluation. hs-CRP-TyG, hs-CRP–Triglyceride-Glucose; CAD, coronary artery disease. Models 1–4 represent progressively adjusted regression models, ranging from the unadjusted model (Model 1) to the fully adjusted model including all covariates (Model 4).

• 

  Figure 5. ROC curves for predicting CAD using hs-CRP-TyG and TyG indices. (A) ROC curve of the hs-CRP-TyG in predicting CAD risk across four multivariable models (Model 1 to Model 4). (B) ROC curve of the TyG index alone in predicting CAD risk across four multivariable models (Model 1 to Model 4). The diagonal dashed line represents the reference line (AUC = 0.5, no discrimination). Model 1, unadjusted; Model 2, adjusted for age, sex, smoking, and alcohol; Model 3, further adjusted for hypertension, diabetes, and lipid-lowering medication; Model 4, fully adjusted, including uric acid, lymphocyte ratio, fibrinogen, and HDL-C. hs-CRP-TyG, hs-CRP–Triglyceride-Glucose; TyG, triglyceride-glucose index; CAD, coronary artery disease; ROC, receiver operating characteristic; AUC, area under the curve; FPR, false positive rate (1-Specificity); TPR, true positive rate (Sensitivity).