目的 基于质量源于设计理念，优化经典名方桃红四物汤（Taohong Siwu Decoction，TSD）的提取工艺。方法 以出膏率及芍药苷、阿魏酸、羟基红花黄色素A的提取率作为TSD提取工艺的关键质量属性（critical quality attributes，CQAs），筛选关键工艺参数（critical process parameters，CPPs）。通过Box-Behnken设计建立数学模型，考察CPPs与CQAs之间的交互作用。结果 方差分析结果显示，所建模型的方差均有显著性差异（P<0.05），且失拟值不显著，表明模型具有统计学意义。基于以上结果，建立提取工艺的设计空间，对空间进行优化，得到最佳提取工艺为提取次数2次，提取时间40~80 min，加水量11~15倍。所得制剂中间体出膏率为49.48%，芍药苷提取率为2.53 mg/g，阿魏酸提取率为0.30 mg/g，羟基红花黄色素A提取率为0.58 mg/g。结论TSD提取工艺稳定，所得制剂中间体与基准样品质量属性接近，可为后续颗粒剂的研究开发奠定基础。

Objective To optimize the extraction process of Taohong Siwu Decoction (TSD, 桃红四物汤) based on the concept of quality by design (QbD).Methods With the extraction amount of solid matter, the extraction percentages of paeoniflorin, ferulic acid and hydroxysafflor yellow A (HSYA) as critical quality attributes (CQAs), the potential critical process parameters (CPPs) was screened. The mathematical model was established by Box-Behnken experimental design to investigate the interaction between CPPs and CQAs. Results Analysis of variance showed that the variance of models was significant (P < 0.05), and the lack-of-fit value was not significant, indicating that the model was statistically significant. Based on the results, through the establishment and optimization of the design space, the optimum extraction process of TSD was obtained as follows:two extraction times, 40-80 min extraction, and 11-15 times of water adding. The extraction rate of the intermediate was 49.48%, and the extraction percentages of paeoniflorin, ferulic acid and HSYA was 2.53 mg/g, 0.30 mg/g and 0.58 mg/g. Conclusion The extraction process of TSD was stable, and the quality properties of the preparation intermediates were close to reference samples, which could provide guidance for the research of subsequent granules.

R283.6

中央高校基本科研业务费专项资金（2020-JYB-ZDGG-031）