目的 建立清热消炎宁胶囊（Qingre Xiaoyanning Capsules，QXC）UPLC指纹图谱及多指标成分定量分析方法，结合化学模式识别技术对多批次制剂进行质量评价。方法 通过优化样品前处理及色谱检测方法，建立合适UPLC指纹图谱和含量测定条件。色谱柱为UPLC HSS T3柱（50 mm×2.1 mm，1.8 μm）；乙腈-0.1%甲酸水溶液为流动相梯度洗脱；检测波长280 nm；柱温40℃；体积流量0.5 mL/min。采用层次聚类分析（hierarchical cluster analysis，HCA）、主成分分析（principal component analysis，PCA）、正交偏最小二乘-判别分析（orthogonal partial least squares-discriminant analysis，OPLS-DA）和聚类热图分析对20批清热消炎宁胶囊进行质量评价。结果 清热消炎宁胶囊UPLC指纹图谱及含量测定方法学考察结果均符合测定要求，获得21个共有峰，运用对照品比对方式指认了7个色谱峰；20批样品的相似度均大于0.954，样品间一致性及稳定性良好；由HCA可知，20批样品可大致聚成2类；PCA从21个共有峰中提取了4个主成分，通过OPLS-DA筛选了迷迭香酸、绿原酸、隐绿原酸、丹参素钠、新绿原酸等8个影响样品质量差异性较大的化合物；7个定量成分线性关系均良好（r≥0.999 8），平均加样回收率98.41%～101.18%，RSD均不大于1.81%；聚类热图分析结果表明，20批样品可聚为2类。结论 建立的同一色谱条件下的清热消炎宁胶囊UPLC指纹图谱结合多指标成分定量方法专属性强、简便、准确，可为其整体质量控制和品质评价提供参考依据。

Objective To establish the UPLC fingerprint and multi-component quantitative analysis of Qingre Xiaoyanning Capsules (清热消炎宁胶囊, QXC), at the same time, chemical pattern recognition technology was used to evaluate the quality of multiple batches of QXC. Methods The sample pretreatment conditions and chromatographic analysis conditions of QXC were optimized, and the optimal UPLC fingerprint and multi-component quantitative analysis method were established. Column:UPLC HSS T3 column (50 mm×2.1 mm, 1.8 μm); mobile phase acetonitrile-water (containing 0.1% formic acid) with gradient elution; detection wavelengths 280 nm; column temperature 40℃; flow rate 0.5 mL/min. Cluster analysis (CA), principal components analysis (PCA), orthogonal partial least squares-discriminant analysis (OPLS-DA) and cluster heatmap analysis were applied for evaluating the quality of 20 batches of QXC. Results Methodological investigation of UPLC fingerprint and content determination were well verified and meeting the analysis requirements. A total of 21 common peaks were obtained by full peak matching, and seven of them were identified by comparing with the retention time of mixed reference substance. The similarity of 20 batches of samples was greater than 0.954, which showed good consistency and stability between the samples. Twenty samples could be classified into two clusters; Four principal components from 21 common peaks were extracted by PCA. Eight quality differential compounds were presented in the fingerprint by OPLS-DA, including rosmarinic acid, chlorogenic acid, 4-dicaffeoylquinic acid, salvianic acid A sodium, neochlorogenic acid and so on. The resolution and linear relationship of seven components in quantitative analysis were good. The average recovery rates were 98.41%-101.18% with RSD ≤ 1.81%. Results of cluster heatmap analysis showed that 20 batches of QXC also could be divided into two categories. Conculsion In this study, the qualitative analysis of UPLC fingerprint and quantitative analysis of multiple index components based on the same chromatographic analysis conditions is specific, simple and accurate, which can provide a reference for the quality control and quality evaluation of QXC.

R283.6

东莞市社会发展科技项目（20211800900442）