银黄颗粒HPLC指纹图谱与模式识别分析

Analysis of HPLC fingerprint and pattern recognition of Yinhuang Granules

目的:建立主要色谱峰化学成分明确的银黄颗粒HPLC指纹图谱,通过化学模式识别的方法全面评价其内在质量。方法:采用HPLC法,确定了100个不同厂家银黄颗粒指纹图谱的20个共有峰,指认了其中14个共有峰的化学成分,采用指纹图谱相似度评价,结合主成分分析和聚类分析对不同生产厂家的质量进行评价。结果:100批银黄颗粒样品图谱与对照图谱比较,相似度均>0.974之间,主成分分析与聚类分析均可将不同生产厂家的银黄颗粒很好地分类,且分类结果一致。结论:所建立的HPLC指纹图谱分析方法全面、准确、稳定,结合化学模式识别研究可有助于银黄颗粒的整体质量控制,同时为其质量评价提供一种有效手段。

Objective: To establish the fingerprint with clear chemical composition of Yinhuang Granules by HPLC, and to assess the quality of Yinhuang Granules by chemical pattern recognition. Methods: HPLC method was used to establish Yinhuang Granules standard fingerprint, consisting of 20 common peaks, and the chemical constituents of 14 common peaks were identified. The fingerprint was further analyzed by chemometric methods including similarity analysis, principal component analysis(PCA), and hierarchical clustering analysis(HCA). Results: The similarity of 100 batches of Yinhuang Granules was greater 0.974. The samples were clearly classified into three groups by using PCA, and four groups by using HCA, and the results of classification were consistent. Conclusion: The developed method of HPLC fingerprint is accurate, comprehensive and reliable, and the combination of chemical pattern recognition can contribute to the overall quality control of Yinhuang Granules, which can be applied as an effective method for the quality evaluation of Yinhuang Granules.