Association of multiple cholesterol-related indices with diabetic nephropathy: A cross-sectional analysis of the NHANES database

Objective To investigate the independent associations, dose-response relationships, and heterogeneity across different populations between eight cholesterol-related indices (WHR, LHR, NHR, MHR, PHR, AIP, CHG, and RCII) and diabetic nephropathy (DN), utilizing data from the National Health and Nutrition Examination Survey (NHANES). Methods This cross-sectional analysis included 9421 participants from the NHANES 2015–2020 cycles, including 631 DN patients. Multivariable logistic regression was used to assess associations between each index and DN. Restricted cubic spline (RCS) models were employed to explore nonlinear relationships, and effect modification was assessed across subgroups defined by gender, BMI, glycemic control, and other characteristics. Results After adjusting for multiple confounders, WHR, LHR, NHR, AIP, CHG, and RCII were all significantly positively associated with DN risk (all P < 0.001), with RCS analyses indicating nonlinear “threshold effects” for most indices. Subgroup analyses revealed significant heterogeneity in associations: WHR and LHR showed stronger associations in females; AIP and CHG remained robust across BMI and glycemic control subgroups; RCII showed stronger associations among non-hypertensive and non-smoking individuals. Conclusion In this cross-sectional study, six cholesterol-inflammation composite indices were independently associated with DN, predominantly exhibiting non-linear dose-response relationships. The variation in association strength across sex, metabolic status, and lifestyle factors suggests a potential pattern of “pathway heterogeneity” in DN risk. These indices warrant further investigation in future studies to explore their value as early clues for risk identification.