Dietary inflammatory potential and dietary quality in relation to advanced cardiovascular-kidney-metabolic syndrome and mortality risk: traditional and machine learning-based analysis

Abstract

Background: Cardiovascular-kidney-metabolic (CKM) syndrome, recently defined by the American Heart Association, represents a complex interplay of cardiovascular, renal, and metabolic disorders linked by shared inflammatory pathways. This study investigates Dietary Inflammatory Index (DII) and Healthy Eating Index (HEI) influence CKM progression and mortality, while exploring the mediating role of inflammatory markers and identifying key dietary predictors.

Methods: We analyzed 19,742 National Health and Nutrition Examination Survey participants (1999–2020) with complete dietary and clinical data. DII and HEI were calculated from 24-hour dietary recalls. Advanced CKM syndrome (stages 3–4) and all-cause mortality were primary outcomes. We employed (1) multivariable logistic/cox regression to assess DII/HEI-outcomes associations; (2) mediation analysis for inflammatory markers; (3) machine learning with SHAP analysis to identify critical dietary components.

Results: Higher DII was associated with advanced CKM syndrome (OR = 1.046, 95%CI: 1.001–1.092) and mortality (HR = 1.096, 95%CI: 1.046–1.149), while HEI showed protective effects (OR = 0.993, 95%CI: 0.988–0.999; HR = 0.993, 95%CI: 0.987–0.999). Cluster analysis revealed the pro-inflammatory low-quality diet (Pattern D) had 35% higher CKM risk (OR = 1.351) and 41% greater mortality risk (HR = 1.411) versus anti-inflammatory high-quality diet. Neutrophils mediated 26.50% (DII) and 24.48% (HEI) of advanced CKM syndrome, 11.16% (DII) and 14.72% (HEI) of mortality risk. Machine learning identified vitamin B6, vitamin D, niacin, monounsaturated fatty acids (MUFA), polyunsaturated fatty acids (PUFA) emerged as critical predictors for CKM staging and mortality risk.

Conclusion: Higher DII scores are associated with increased risks of advanced CKM syndrome and all-cause mortality, while higher HEI scores are associated with reduced risks. Neutrophils were identified as a key inflammatory mediator in these associations. Vitamin B6, vitamin D, niacin, MUFA, PUFA emerged as key dietary factors, suggesting precision nutrition strategies for CKM management.

Graphic abstract:

Supplementary Information: The online version contains supplementary material available at 10.1186/s12937-025-01268-5.

Keywords: Cardiovascular-kidney-metabolic syndrome; Dietary inflammatory index; Healthy eating index; Mortality; Risk stratification.

Fig. 1

Cluster Analysis of Dietary Patterns Based on DII and HEI. Note: Pattern A: Anti-inflammatory High-Quality Diet; Pattern B: Balanced Mid-Quality Diet; Pattern C: Pro-inflammatory Mid-Quality Diet; Pattern D: Pro-inflammatory Low-Quality Diet

Fig. 2

Mediation Effects of Inflammatory Markers on the Associations of DII, HEI with Advanced CKM Syndrome and All-Cause Mortality. Note: Arrows indicate the direction of effects, and mediation is represented by the proportion of the total effect explained by the mediator. ACME represents the average causal mediation effects, and ADE represents the average direct effects

Fig. 3

ROC Curve and SHAP Feature Importance Analysis of Machine Learning Models in Predicting Advanced CKM Syndrome and All-Cause Mortality. Note: (A) ROC curves for predicting advanced CKM syndrome. B ROC curves for predicting all-cause mortality in advanced CKM syndrome. C SHAP summary plot for predicting advanced CKM syndrome. D SHAP summary plot for predicting all-cause mortality in advanced CKM syndrome. ROC, receiver operating characteristic; SHAP, shapley additive exPlanations; AUC, area under the curve; LightGBM, light gradient boosting machine; XGBoost, eXtreme gradient boosting; GBM, gradient boosting machine; SVM, support vector machine; NeuralNet, neural network; MUFA, monounsaturated fatty acids; PUFA, polyunsaturated fatty acids.