热毒宁注射液金银花和青蒿(金青)醇沉过程中多指标的近红外快速检测

Multi-indicators rapid determination in alcohol precipitation of Lonicerae Japonicae Flos and Artemisiae Annuae Herba in Reduning Injection by near-infrared spectroscopy

目的应用近红外光谱(NIR)透射法对热毒宁注射液大生产中金银花和青蒿(金青,LA)醇沉过程绿原酸(chlorogenic acid,CA)质量浓度和固含量进行快速检测,提高醇沉过程质量控制水平。方法制备金青醇沉过程样品,以HPLC法和烘干法为对照分析方法,分别测定CA质量浓度和固含量,运用偏最小二乘(PLS)法分别建立NIR光谱与CA质量浓度和固含量之间的定量校正模型,并对醇沉过程的未知批次中的CA质量浓度和固含量进行预测。结果所建立的CA质量浓度和固含量模型的相关系数(R)均达到0.980,采用独立的验证集对模型进行验证,验证集预测误差均方根(RMSEP)分别为0.541和0.218,相对预测偏差(RSEP)均小于8%。结论研究结果表明NIR分析技术可实现热毒宁注射液醇沉过程中CA质量浓度和固含量2个关键质控指标的快速检测,结果准确可靠。

Objective Near-infrared（NIR） spectroscopy was used to rapidly determine the concentration of chlorogenic acid（CA） and soluble solid content in alcohol precipitation of Lonicerae Japonicae Flos and Artemisiae Annuae Herba（LA） in Reduning Injection, so as to increase the level of quality control of alcohol precipitation process. Methods The LA samples of alcohol precipitation were collected. The PLS algorithm was used to establish NIR models between spectra and concentration of CA and total solid content, and the concentration of CA and soluble solid content in unknown batch were predicted. Results The correlation coefficients（R） of NIR models were above 0.980 in calibration sets. The R of verification model was above 0.980, the root mean square error of prediction（RMSEP） for CA and soluble solid content were 0.541 and 0.218, respectively, and the relative standard errors（RSEP） for CA and soluble solid content were no more than 8%. Conclusion The results demonstrate that NIR could determine the concentration of CA and soluble solid content in alcohol precipitation of Reduning Injection rapidly and accurately.