IntroductionThis study aimed to investigate the correlation between various plasma metabolites and the likelihood of developing diabetic nephropathy (DN) and construct a diagnostic model for DN in Chinese patients with type 2 diabetes mellitus (T2DM). MethodsA cross-sectional investigation was conducted in a hospital setting. Based on medical data, a total of 743 patients from a tertiary hospital were selected and categorized into two groups: the diabetic nephropathy group (DN group) and the non-diabetic nephropathy group (non-DN group). Plasma levels of metabolites, including amino acids and acylcarnitines, were determined using a laser counter measurement system (LC-MS). Subsequently, partial least-squares regression was used to assess the significance of these metabolites. Receiver operating characteristic (ROC) curves were generated for factors that ranked highest in terms of relevance. Model performance was assessed using the curve (AUC). ResultsOf the 743 patients with T2DM admitted to the hospital, 145 had DN. Compared with the non-DN group, the DN group exhibited elevated systolic blood pressure (P = 0.001), high-density lipoprotein cholesterol (P = 0.01), and low-density lipoprotein cholesterol (P = 0.042). Additionally, the DN group had a higher prevalence of stroke patients (P < 0.001) and diabetic retinopathy patients (P < 0.001). Finally, a risk model that included citrulline, leucine, tyrosine, valine, propionylcarnitine (C3), and palmitoylcarnitine (C16) was developed. This model achieved an AUC of 0.709, with a 95% confidence interval (CI) ranging from 0.626 to 0.793. ConclusionsA diagnostic model consisting of six plasma metabolites to assess the risk of DN in Chinese patients with T2DM may provide clues for future research.