Association of dietary quality, biological aging, progression and mortality of cardiovascular-kidney-metabolic syndrome: insights from mediation and machine learning approaches

Abstract Background To investigate the association between the Dietary Inflammatory Index (DII), biological aging, and the staging and mortality of cardiovascular-kidney-metabolic (CKM) syndrome. Methods Data of 7,918 participants were derived from the National Health and Nutrition Examination Survey 2005–2018. Cross-sectional analyses using multivariable logistic regression were conducted to evaluate the relationship between DII and CKM staging. Cox proportional hazards models were employed to assess the impact of DII on mortality in CKM patients. Mediation analyses were performed to determine whether biological aging mediated DII-staging and DII-mortality association. Machine learning models were developed to classify CKM stages 3/4 and predict all-cause mortality, with SHapley Additive exPlanations (SHAP) used to interpret the contribution of DII components. Results Over a median follow-up of 9.3 years, 819 deaths were recorded. Higher DII were associated with an increased risk of advanced CKM stages [OR (95% CI): tertile 2, 1.39 (1.17, 1.65); tertile 3, 1.85 (1.56, 2.20)], and all-cause mortality [(HR (95% CI): tertile 2, 1.20 (1.01–1.43); tertile 3: 1.45 (1.21–1.73)]. The optimal risk stratification threshold for DII to predict all-cause mortality was 1.93. Mediation analyses revealed that biological aging accounted for 23% (95% CI: 18-28%) of the effect of DII on advanced CKM stages and 13% (95% CI: 8-22%) of the effect of DII on all-cause mortality. Furthermore, the Light Gradient Boosting Machine model showed strong performance in predicting advanced CKM staging (AUC: 0.896, 95% CI: 0.882–0.911), while Logistic regression performed better in predicting all-cause mortality (AUC: 0.857, 95% CI: 0.831–0.884). SHAP analysis revealed that intake of magnesium and n-3 fatty acid were associated with reduced risk of both advanced CKM stages and all-cause mortality. Conclusion DII, a marker of pro-inflammatory dietary patterns, was significantly linked to CKM syndrome progression and mortality, partly by influencing biological aging. This underscores the importance of diet quality in managing CKM staging and mortality risk.