Abstract
BACKGROUND AND AIM: The antioxidant diet and lifestyle are widely believed to prevent and even treat various diseases; however, their applicability to cardiovascular-kidney-metabolic (CKM) syndrome remains unknown. In this study, the correlation between the oxidative balance score (OBS) and CKM syndrome was investigated, along with the underlying mechanisms involved. METHODS AND RESULTS: The study data were obtained from participants with complete OBS and diagnostic information on CKM syndrome in the NHANES from 1999 to 2018. The correlation between OBS and the incidence of CKM syndrome was investigated by weighted multivariable logistic regression analysis. Mediation analyses were conducted to explore the mediating role of inflammatory markers in the relationship between these two variables. A restricted cubic spline graph and threshold effects were constructed to identify nonlinear associations. OBS was applied to eight machine learning algorithms to develop a predictive model and assess its performance. A total of 30113 participants were included in our study. OBS was negatively correlated with CKM syndrome regardless of covariate adjustment. Compared with the lowest quartile of OBS, the highest quartile resulted in a 33% decrease in the incidence of CKM syndrome. A nonlinear association was identified using a restricted cubic spline, with 30 being the threshold point. Mediation analysis indicated that high-sensitivity C-reactive protein (hsCRP) and systemic immune inflammation (SII) partially influenced the relationship between these two parameters. Among the eight machine learning algorithms, XGBoost presented the highest area under the curve (AUC), demonstrating superior predictive performance and clinical efficacy. CONCLUSIONS: The prevalence of CKM syndrome was negatively correlated with an increase in OBS, which may be partially related to its anti-inflammatory effect.