Association between atherogenicity indices and prediabetes: a 5-year retrospective cohort study in a general Chinese physical examination population

Abstract

Background and objective: Atherogenicity indices have emerged as promising markers for cardiometabolic disorders, yet their relationship with prediabetes risk remains unclear. This study aimed to comprehensively evaluate the associations between six atherogenicity indices and prediabetes risk in a Chinese population, and explore the predictive value of these atherosclerotic parameters for prediabetes.

Methods: This retrospective cohort study included 97,151 participants from 32 healthcare centers across China, with a median follow-up of 2.99 (2.13, 3.95) years. Six atherogenicity indices were calculated: Castelli's Risk Index-I (CRI-I), Castelli's Risk Index-II (CRI-II), Atherogenic Index of Plasma (AIP), Atherogenic Index (AI), Lipoprotein Combine Index (LCI), and Cholesterol Index (CHOLINDEX). To address the natural relationships between the atherogenicity indices and risk of prediabetes, we applied Cox proportional hazards regression with cubic spline functions and smooth curve fitting, using a recursive algorithm to calculate inflection points. Machine learning approach (XGBoost and Boruta methods) to address the high collinearity among indices and assess their relative importance, combined with time-dependent ROC analysis to evaluate the predictive performance at 3-, 4-, and 5-year follow-up.

Results: During follow-up, 11,199 participants developed prediabetes (incidence rate: 3.71 per 100 person-years). Significant nonlinear associations were observed between all atherogenicity indices and prediabetes risk. Through Z-score standardization of atherogenicity indices and comprehensive Cox proportional hazards regression and advanced machine learning techniques, we identified AIP as the most significant predictor of prediabetes [HR = 1.057 (95% CI 1.035-1.080, P < 0.0001)], with LCI emerging as a secondary important marker [HR = 1.020 (95% CI 1.002-1.038, P = 0.0267)]. Our innovative XGBoost and Boruta analysis uniquely validated these findings, providing robust evidence of AIP and LCI's critical role in prediabetes risk assessment. Time-dependent ROC analysis further validated these findings, with LCI and AIP demonstrating comparable discrimination, with overlapping AUC ranges of 0.5952-0.6082. Notably, the combined indices model achieved enhanced predictive performance (AUC: 0.6753) compared to individual indices, suggesting the potential benefit of using multiple atherogenicity indices for prediabetes risk prediction.

Conclusion: This study identifies statistically significant associations between atherogenicity indices and prediabetes risk, highlighting their nonlinear relationships and combined effects. While the predictive performance of these indices is modest (AUC 0.55-0.68), these findings may contribute to improved risk stratification when incorporated into comprehensive assessment strategies.

Keywords: Atherogenicity indices; Cohort study; Machine learning; Nonlinear relationship; Prediabetes; Risk prediction.

Fig. 1

Study flow diagram of participant selection 211,833 participants were assessed for eligibility in the original study. We further excluded 114682participants. The final analysis included 97,151 subjects in the present study

Fig. 2

Correlation matrix analysis of atherogenicity indices. Correlation matrix showing relationships between the six atherogenicity indices (CRI-I, CRI-II, AIP, AI, LCI, and CHOLINDEX) in the study cohort. The diagonal panels display histogram distributions for each variable. Lower triangular panels show scatter plots with fitted curves (red lines) illustrating bivariate relationships. Upper triangular panels present Pearson correlation coefficients, with asterisks (***) indicating statistical significance at p < 0.001. The correlation matrix reveals strong positive associations between most indices, with correlation coefficients ranging from 0.44 to 1.00

Fig. 3

Non-linear associations between atherogenicity indices and risk of prediabetes. Restricted cubic spline models showing the relationship between various atherogenicity indices and log-transformed relative risk (Log RR) for adverse outcomes. The solid red line represents the estimated Log RR, and the green shaded areas indicate the 95% confidence intervals. The reference values were set at the median of each index. A AI, B AIP, C CHOLINDEX, D CRI-I, E CRI-II, and F LCI. P values for non-linearity were <0.001 for all indices

Fig. 4

Relative importance of different atherogenicity indices in predicting incident prediabetes. A. Relative importance of atherogenicity indices in predicting prediabetes risk, determined by XGBoost algorithm. The x-axis represents the normalized relative importance, highlighting the predictive potential of each index. B. Boruta feature importance analysis of atherogenicity indices, showing the statistical significance and predictive power of different markers. The y-axis represents the importance score, with higher values indicating greater predictive relevance for prediabetes risk

Fig. 5

Temporal comparison of predictive performance for atherogenicity indices. This figure shows the area under the receiver operating characteristic curve values for various atherogenicity indices assessed over different time periods (3-year, 4-year, and 5-year follow-up). The x-axis represents different lipid parameters including CRI-I, CRI-II, AIP, AI, LCI, CHOLINDEX, and Combined Indicators. The y-axis shows AUC values ranging from 0.500 to 0.700. The Combined Indicators demonstrated superior predictive performance compared to individual indices