Abstract
BACKGROUND: Diabetic kidney disease (DKD) is a leading cause of kidney failure closely linked to lifestyle factors, but the mechanisms have not been systematically investigated. AIM: This study aimed to assess the long-term metabolic effects of lifestyle behaviors on DKD. DESIGN AND METHODS: This study aimed to examine links between lifestyle, metabolic biomarkers, and DKD incidence and mortality in a population with diabetes. This study analyzed data from 18 287 participants, evaluating five lifestyle factors (diet, sleep duration, physical activity, smoking and alcohol intake) alongside 251 metabolic biomarkers. Cox proportional hazards models and Mendelian randomization (MR) assessed associations. Mediation analysis was conducted on biomarkers linked to both lifestyle and DKD. Additionally, genome-wide association study (GWAS) and gene enrichment analysis were conducted on mediating biomarkers to explore biological mechanisms. RESULTS: Among 18 287 participants with diabetes, 3247 developed DKD over a median follow-up of 14.6 years. Lipids and amino acids were associated with DKD and mediated the effects of lifestyle factors. Mediating biomarkers, including triglycerides to total lipids in HDL percentage and glycoprotein acetyls, demonstrated both observational and causal associations with DKD. The mediation effects differed between various levels of blood glucose control. Pathway enrichment analysis identified both shared and distinct biological pathways. CONCLUSIONS: This comprehensive study underscores the importance of metabolomics in delineating the mechanisms by which lifestyle behaviors influence DKD, paving the way for targeted interventions.