目的 筛选经典名方血府逐瘀汤（Xuefu Zhuyu Decoction，XZD）的最优醇提工艺，使其能保留传统水提样品中的小分子物质，为透皮制剂开发提供实验依据。方法 以XZD中12种活性成分（腺苷、梓醇、苦杏仁苷、羟基红花黄色素A、芍药苷、阿魏酸、甘草苷、芸香柚皮苷、柚皮苷、橙皮苷、新橙皮苷、甘草酸）含量、出膏率及指纹图谱相似度作为关键质量属性（critical quality attributes，CQAs）；以乙醇体积分数、提取时间、溶剂倍数作为关键工艺参数（critical process parameters，CPPs），采用Box-Behnken设计-响应面法（Box-Behnken design-response surface methodology，BBD-RSM）开展试验。计算各试验样品指标的基准关联度，通过熵权法确定各CQAs的权重并计算综合评分；运用TOPSIS模型计算各XZD样品与其基准样品的相对接近度并排序，结合综合评分筛选最佳醇提工艺。此外，采用超高效液相色谱-四级杆-飞行时间质谱技术（UPLC-Q-TOF-MS/MS）对XZD基准样品与其最优醇提工艺样品的化学成分进行表征，进一步验证小分子物质的保留效果。结果 经熵权-TOPSIS模型分析，确定XZD最优醇提工艺为10倍量60%乙醇，提取3 h。3批验证样品的综合评分均值达0.501，RSD为0.75%，符合要求，证实该工艺稳定、方法适用。同时，UPLC-Q-TOF-MS/MS分析表明，XZD最优工艺样品与其基准样品的主要化学成分组成相似。结论 基于基准关联度与熵权-TOPSIS模型的综合筛选策略可行有效，所得XZD醇提工艺参数稳健可靠，能有效保留基准煎液的小分子物质，化学物质基础与基准煎液相似，可为XZD透皮制剂研发提供参考。

Objective To select the optimal alcohol extraction process for the classic prescription Xuefu Zhuyu Decoction (XZD, 血府逐瘀汤), enabling the retention of small-molecule substances in traditional aqueous extracts, thereby providing experimental basis for the development of transdermal formulations. Methods The contents of 12 active components (adenosine, catalpol, amygdalin, hydroxysafflor yellow A, paeoniflorin, ferulic acid, liquiritin, narirutin, naringin, hesperidin, neohesperidin, glycyrrhizic acid) in XZD, the extract yield, and the fingerprint similarity were used as critical quality attributes (CQAs). Ethanol concentration, extraction time, and solvent ratio were selected as critical process parameters (CPPs), and experiments were conducted using a Box-Behnken design-response surface methodology (BBD-RSM). The benchmark correlation degree of each experimental sample index was calculated, and the weights of each CQA were determined using the entropy weight method to compute a comprehensive score. The TOPSIS model was applied to calculate the relative closeness of each sample to the benchmark sample and rank them, with the optimal alcohol extraction process screened based on the comprehensive score. Additionally, ultra-performance liquid chromatography coupled with quadrupole time-of-flight tandem mass spectrometry (UPLC-Q-TOF-MS/MS) was used to description the chemical components of the benchmark sample and the optimal alcohol extraction process sample, further verifying retention effect of small-molecule components.. Results Analysis using the entropy weight-TOPSIS model determined the optimal alcohol extraction process for XZD as follows: ten times the amount of 60% ethanol, with each extraction lastin 3 h. The average comprehensive score of three validation samples reached 0.501, with RSD of 0.75%, which met the requirements and confirmed that the process was stable and the method was applicable. Meanwhile, UPLC-Q-TOF-MS/MS analysis showed that the main chemical composition of the optimal process sample were similar to those of the benchmark sample. Conclusion The comprehensive screening strategy based on the benchmark correlation degree and entropy weight-TOPSIS model is feasible and effective. The obtained XZD alcohol extraction process parameters are robust and reliable, and the small molecule substances of the benchmark decoction can be effectively retained. The chemical substance basis is similar to that of the benchmark decoction. This study can provide a reference for the research and development of XZD transdermal formulations.

R283.6

国家自然科学基金项目（82060721）；甘肃省自然科学基金项目（23JRRA1623）；甘肃省自然科学基金项目（23JRRA1622）