目的 采用星点设计-效应面法优化丹皮酚脂微球（Pae-LM）的处方工艺并考察其体外释药行为。方法 以平均粒径和离心稳定性常数（K_e）为评价指标，分别对油相以及复合油比例、磷脂（PC）和硬脂酸用量、乳化剂种类、稳定剂种类及用量、PC与CH（胆固醇）的质量比、高速剪切温度及时间、均质压力及时间进行处方工艺筛选。采用星点设计效应面法考察丹皮酚投药量、高压均质压力对Pae-LM制剂性质的影响，用二项式模型及多元线性回归模型拟合建立指标与因素之间的数学关系，根据评价指标的最佳数学模型描绘效应面，经效应面法分析最佳处方。按照优化所得最佳处方制备Pae-LM，并考察体外释药特性。结果 最优处方制得的Pae-LM外观基本圆整，平均粒径为（149.27±0.57）nm，Zeta电位为（-36.01±3.09）mV，包封率为（98.24±0.32）%，载药量为（11.94±0.04）%。K_e与2个考察因素之间存在可信定量关系，且二项式模型较多元线性模型置信度高。丹皮酚原料药12、24、36 h的累积释放量分别为71.84%、85.21%、95.07%，而Pae-LM在12、24、36 h时累积释放量仅为57.21%、59.66%、63.91%，药物释放符合Ritger-peppas模型。结论 星点设计-效应面法适用于Pae-LM的处方优化。优化的Pae-LM粒径适宜，包封率高，可为丹皮酚心血管递送系统的开发提供参考。

Objective To optimize the formulation of paeonol lipid microspheres (Pae-LM) through central composite design- response surface method and determine its in vitro release characteristics. Methods Using the mean particle size and centrifugal stability constant (K_e) as evaluation indexes, the oil phase type and the ratio of composite oil, the amount of phospholipid and stearic acid, the type of emulsifier, the type and amount of stabilizer, the quality of PC and CH, the high-speed shear temperature and time, the homogenization pressure and time was screened in prescription process. Effects of dosage of paeonol and high pressure homogenizing pressure on the properties of Pae-LM preparation were investigated by central composite design-response surface method. The binomial model and multivariate linear regression model were used to establish the mathematical relationship between the indexes and the factors. According to the best mathematical model of evaluation index, the response surface was depicted and the best prescription was analyzed by the response surface method. According to the optimized formulation Pae-LM, the in vitro drug release characteristics were investigated. Results The best prescription of Pae-LM was basically round, with mean particle size of (149.32 ±0.57) nm, Zeta potential of (−36.01 ±3.09) mV, encapsulation rate of (98.24 ±0.32)% and drug-loading rate of (11.94 ±0.04)%. There was a credible quantitative relationship between K_e and the two factors, and the binomialmodel was more reliable than the multivariate linear model. The cumulative release of paeonol drug substance at 12, 24 and 36 h were 71.84%, 85.21% and 95.07%, while the cumulative release of Pae-LM was only 57.21%, 59.66%, and 63.91% at 12, 24 and 36 h, respectively. The drug release was in accordance with the Ritger-peppas model. Conclusion Central composite design-response surface method can be applied to optimize prescription of lipid emulsion microspheres. The optimized particle size of Pae-LM was suitable with a higher encapsulation rate, which can provide a reference for the development of paeonol cardiovascular delivery system.

R283.6

贵阳市科技计划项目（筑科合同[2017]30-25号）；贵州省科技创新团队项目（黔科合人才团队[2015]4025号）；贵州省高层次创新型人才百层次人才项目（贵州省科技厅黔科合人才[2015]4029号）；贵州医科大学药学国际科技合作基地（黔科合平台人才[2017]5802）