目的 建立白脉软膏的指纹图谱并测定其主要成分的含量，结合多元统计分析对其2个产地14个批次质量进行科学全面地评价，为其质量控制提供依据。方法 采用HPLC法对白脉软膏中的主要成分甘草苷、甘草酸铵、甘松新酮、姜黄素进行含量测定，建立了白脉软膏的指纹图谱，通过指纹图谱相似度分析、聚类分析、主成分分析（PCA）和因子分析对2个产地的不同批次白脉软膏质量进行全面评价。结果 甘草苷、甘草酸铵、甘松新酮、姜黄素的含量测定方法学考察符合要求，质量分数分别为0.051～0.200mg/g、0.136～0.622 mg/g、0.030～0.345 mg/g、0.001～0.069 mg/g。建立的白脉软膏指纹图谱共标定17个共有峰，经对照品指认共鉴定出甘草苷、甘草酸铵、甘松新酮3个色谱峰，14批样品的相似度均大于0.975。聚类分析中2个产地14批白脉软膏可以分为4类，其中西藏林芝市生产的批次S1～S11聚为一类，甘肃省兰州市生产的S12聚为一类，S13聚为一类，S14聚为一类。PCA和因子分析结果显示甘肃省兰州市生产的3个批次S12～S14主成分综合得分均高于西藏林芝市生产的11个批次S1～S11，推测是因为产地生产条件或药材来源变化导致了质量的差异，与聚类分析结果一致。结论 首次建立白脉软膏含量测定、指纹图谱结合多元统计分析的质量控制方法科学可靠，不但可以用于白脉软膏的质量控制，还可以对其不同产地批次质量进行全面评价，为白脉软膏质量标准的提高和中药批次间的质量评价提供借鉴。

Objective To establish a fingerprint method of Baimai Ointment (BO) and determine the content of its main components. The BO of 14 batches from two production areas was scientifically and comprehensively evaluated based on multivariate statistical analysis, which provided the basis for the quality control. Methods HPLC method was used to determine the content of liquiritin, ammonium glycyrrhizinate, nardosinone and curcumin in BO, and the fingerprint of BO was established. The fingerprint similarity analysis, cluster analysis, principal component analysis (PCA) and factor analysis were performed to comprehensively evaluate the different batches of BO in two producing areas. Results The methodological determination of liquiritin, ammonium glycyrrhizinate, nardosinone and curcumin met the requirements, and the content was 0.051-0.200 mg/g, 0.136-0.622 mg/g, 0.030-0.345 mg/g, 0.001-0.069 mg/g, respectively. The established fingerprints of BO were calibrated with 17 common peaks. Three chromatographic peaks of liquiritin, ammonium glycyrrhizinate and nardosinone were identified by reference. The similarity of 14 batches of sample was greater than 0.975. In the cluster analysis, 14 batches of BO from two producing areas can be divided into four categories, among which batches S1-S11 produced by Linzhi City of Tibet were grouped into one category. And S12 produced in Lanzhou City of Gansu Province was clustered into one class, and S13 was clustered into one class, S14 was grouped into one class. The results of PCA and factor analysis showed that the comprehensive scores of the three batches of S12-S14 produced in Lanzhou City of Gansu Province were higher than the 11 batches of S1-S11 produced by Linzhi City of Tibet, presumably because of the changes in production conditions or sources of medicinal materials. The result was consistent with cluster analysis. Conclusion This study is the first to establish a scientific and reliable quality control method of Tibetan medicine BO based on multi-component determination, fingerprint and multivariate statistical analysis. It can be used not only for the quality control of Baimai Ointment, but also for the comprehensive evaluation of batch quality consistency. It provides reference for the improvement of the quality standard of BO and the quality evaluation among Chinese medicine batches.

R286.02

中央高校基本科研业务费专项（2017-JYB-JS-011）