Application of metabolomics in establishing primary nephrotic syndrome diagnosis model

Objective To establish diagnosis model and explore related metabolic pathways by analyzing the serum metabolic profile of patients with primary nephrotic syndrome (PNS) through metabolomics. Methods Thirty PNS patients hospitalized in Huai'an First People's Hospital between December 2010 and April 2012 were enrolled. High performance liquid chromatography-mass spectrometry (LC-MS) was employed to detect metabolites in the serum of 30 PNS patients and 30 healthy controls. Metabolic fingerprint profiling and multivariate pattern recognition analysis were combined to establish disease-specific metabolic diagnosis model, and metabolic pathway analysis was performed. Results PNS group and control group could be well separated by principal component analysis (PCA) model as well as partial least-squares discriminant analysis (PLS-DA) model with Q2 of 0.300. There was well interpretation in PLA-DA model (R2X=0.581,R2Y=0.452). Compared with healthy controls,PNS patients had decreased cholestane 3, 7, 12, 15 alcohol, acyl glycerine, phytosphingosine and tryptophan, and increased sphingomyelin, arginine and glutamic acid (all VIP＞1, P＜0.05). The metabolic disorders pathways of PNS patients included sphingolipid metabolism, arginine and proline metabolism, linoleic acid metabolism and pyrimidine metabolism (all impact＞0.10 and P＜0.05). Conclusions Metabolomics combined with multivariate pattern recognition analysis may be a new tool for diagnosis and monitoring of PNS.