Metabolomics analysis of diabetic kidney disease for discovering early diagnostic biomarkers: A systematic review and meta-analysis of prospective studies

Abstract

Background Classical biomarkers of diabetic kidney disease (DKD), including serum creatinine and albuminuria, cannot detect the disease in early stages, leading to worsened complications. This study aimed to introduce a panel of early diagnostic biomarkers for DKD through meta-analysis of longitudinal metabolomics studies. Methods A systematic search was conducted across PubMed, Web of Science, and Scopus up to May 12, 2025. Only studies were included in which urine or blood samples were collected from individuals with diabetes and the participants were followed over time. The outcomes were defined as DKD incidence, albuminuria progression, rapid estimated glomerular filtration rate decline, end-stage renal disease, or all-cause mortality. Relative ratio (95 confidence intervals (CI)) or correlation coefficients (95 CI) of baseline metabolites with these outcomes were extracted from the included studies. Results The analysis included 39 studies covering 52 populations, with a total sample size of 31,012 individuals. The meta-analysis incorporated 170 blood and 12 urine metabolites, of which 65 and 11 showed significant associations with the outcomes, respectively. Enrichment analyses of the differential metabolites highlighted the reprogramming of amino acid, lipid, and energy metabolism. Conclusion This meta-analysis introduces metabolic biomarkers strongly associated with DKD incidence or progression. Furthermore, this study underscores the rewiring of metabolic pathways related to energy homeostasis as an adaptation to the prolonged insults of diabetic milieu. The limitations of this study are, the variation in multivariable adjustment methods used across the included studies and the lack of established decision thresholds for the proposed biomarkers. Copyright © 2025. Published by Elsevier Inc.