基于裂区的D-最优实验设计法优化丹参川芎嗪注射液水提工艺

Optimization of water extraction process of Salviae Miltiorrhizae and Ligustrazine Hydrochloride Injection via split-plot based D-optimal experimental design

目的在质量源于设计(quality by design,QbD)理念的指导下优化丹参川芎嗪注射液(Salviae Miltiorrhizae and Ligustrazine Hydrochloride Injection,SMLHI)的水提工艺。方法首先通过筛选实验设计确定加热温度、保温时间、药材规格和溶剂倍量为丹参水提工艺的关键工艺参数。随后为了提高实验效率,采用基于裂区的D-最优实验设计优化SMLHI的水提工艺。为了尽可能多的提取实验数据潜在信息,采用了响应曲面法和人工神经网络法2种方法建立关键工艺参数和关键质量属性之间的回归模型。最后,选择较优的模型进一步采用基于满意度函数的多指标优化算法综合考察提取液各关键质量属性,确定水提工艺的最佳操作条件。结果基于裂区的D-最优实验设计可以有效提高实验效率,同时发现人工神经网络模型较响应曲面法模型具有更好的拟合能力和预测能力。以人工神经网络模型为基础进行多指标优化,最终确定SMLHI水提工艺的最佳操作条件为加热温度150℃,保温时间84 min,药材规格11 cm,溶剂用量为10倍。结论利用基于裂区的D-最优实验设计法优化了SMLHI水提工艺,为传统的中药制药工艺研究过程中面临的工艺参数改动受限、更改成本高昂的情况提供合理的实验设计方案,提高研究效率,降低研究成本,对不同试验规模下中药制药工艺的过程研究具有较大参考价值,为中药的工艺二次开发提供了可供参考的新方法。

Objective Quality by design(QbD)concept was applied to guide the efficient optimization of the water extraction process for Salviae Miltiorrhizae and Ligustrazine Hydrochloride Injection(SMLHI).Methods Firstly,Plackett-Burman experimental design method was used to screen critical process parameters(CPPs).Secondly,in order to improve the study efficiency,split-plot based D-optimal methodology was applied to implement experimental design.To obtain more process information,response surface method(RSM)and artificial neural network(ANN)were used to establish the regression model between CPPs and critical quality attributes(CQAs)for the extraction.Finally,after comparison of the two models,the optimum operating conditions were determined with the multi-index optimization algorithm by a better regression model.Results The study efficiency was improved by the split-plot method to conduct the experimental arrangement.Then,the ANN method showed better modeling performance than the RSM in this study,and then it was taken as the final optimum model for the multi-index optimization.The optimal operating conditions of the extraction process for SMLHI were follows:The heating temperature was 150℃,extraction time was 84 min,specification of medicinal material was 11 cm,and the ratio of material to water was 10.Conclusion This study optimized the water extraction process for SMLHI by introducing the split-plot design into the process optimization of traditional Chinese medicine.Split-plot design methodology can provide reasonable experimental design solutions to improve research efficiency and reduce research costs for pharmaceutical process research in traditional Chinese medicine,which faces limited changes in process parameters and high costs.And it has a greater reference value for the process study of Chinese medicine pharmaceutical process under different test scale,and provides a new method for the secondary development of Chinese medicine pharmaceutical process for reference.