目的 探讨尿液代谢组学应用于诊断慢性乙肝患者抑郁症状的潜在可行性。方法 采用核磁共振检测75例慢性乙肝合并抑郁患者和64例慢性乙肝无抑郁患者尿液中的代谢物质，利用正交偏最小二乘聚类法(OPLSDA)分析检测到的尿液代谢物质。结果 共检测出31种代谢物质。OPLS-DA分析结果显示，这些代谢物质的组合能准确区分两组样本(准确率约88%)，其中12种代谢物质起主要区分作用。结论 尿液代谢组学具有诊断慢性乙肝患者抑郁症状的潜在可行性，值得科研人员研究和临床推广。

Objective To explore the potential feasibility of using urine metabolomics for diagnosing the major depression (MD) of patients with chronic hepatitis B. Methods Metabolites in the urine of 75 chronic hepatitis B patients with MD and 64 chronic hepatitis B patients without MD were detected by the nuclear magnetic resonance spectroscopy and analyzed by the orthogonal partial least-squares discriminant analysis (OPLS-DA). Results A total of 31 metabolites were identified. The results of OPLS-DA analysis showed that samples of two groups could be correctly separated by combinations of these metabolites (the accurate rate was about 88%). Twelve metabolites played a major role in separation. Conclusion Urine metabolomics has the potential feasibility for diagnosing the MD of patients with chronic hepatitis B and deserves further study and clinical application.