Abstract
BACKGROUND: There is mixed evidence for an association between cardiometabolic risk factors and chronic kidney disease risk (CKD). This study aimed to determine whether different latent classes of cardiometabolic conditions were associated with chronic kidney disease risk. METHOD: Data from 7,195 participants in the China Health and Retirement Longitudinal Study (CHARLS) were analyzed. Latent class analysis was performed using data on obesity, high-density lipoprotein cholesterol, triglyceride, hypertension, diabetes, arthritis or rheumatism, and systemic inflammatory conditions and heart disease. Confounder-adjusted multiple logistic regressions were conducted to estimate CKD incidence by cardiometabolic latent classes. Sensitivity analyses were performed across cross-sectional and longitudinal samples, as well as derivation and validation cohorts. RESULTS: Three cardiometabolic classes were identified: relatively healthy cardiometabolic (RHC) phenotype, metabolic syndrome (MetS) phenotype, and cardiovascular disease (CVD) phenotype, which accounted for 66.2%, 19.9%, and 13.8%, respectively. The incidence of CKD was 12.7% in the CVD group, 9.4% in the MetS group, and 5.9% in the RHC group. After adjusting for confounding factors, it was found that the metabolic syndrome type had a 54% increased risk of newly diagnosed CKD compared to the healthy heart type (OR = 1.54, 95% CI: 1.22-1.93), while the cardiovascular type increased by 104% (OR = 2.04, 95% CI: 1.61-2.57). Sensitivity analyses showed high consistency (> 90%) in class assignments, confirming model robustness. CONCLUSION: Different cardiometabolic phenotypes are associated with an increased risk of new-onset CKD. Gender and age are important factors influencing the strength of this association. Phenotypic classification may improve CKD risk stratification and guide early prevention efforts.