呋塞米晶型的近红外光谱法分析及其定量模型的建立  

Analysis of the Polymorphism of Furosemide by Near Infrared Spectroscopy and Establishment of the Quantitative Model

作　　者：黄蓉 叶晓霞[1] 陆丹[1] 毛丹卓[1] 杨永健[1] HUANG Rong;YE Xiaoxia;LU Dan;MAO Danzhuo;YANG Yongjian(Shanghai Institute for Food and Drug Control,Shanghai 201203)

机构地区：[1]上海市食品药品检验所,上海201203

出　　处：《中国医药工业杂志》2018年第10期1433-1439,共7页Chinese Journal of Pharmaceuticals

基　　金：国家重大科学仪器设备开发专项(2012YQ180132);药典会药品医疗器械审评审批制度改革专项课题(ZG2017-4-02)

摘　　要：采用近红外光谱法分析呋塞米的3种晶型,建立了3种晶型的近红外对照图谱,并采用OPUS数据分析系统建立了呋塞米晶型的定量模型。应用偏最小二乘法(PLS)建立回归模型,采用线性补偿差减法、直线差减法、矢量归一法、最小-最大归一法、多元散射校正法和1～2阶导数等不同数据预处理算法来减少系统干扰。所建立的定量模型对于晶型I、II、III的交叉验证均方差(RMSECV)分别为1.51、2.03、2.23;决定系数R^2分别为99.73、99.43、99.16,维数(rank)分别为9、6、8。呋塞米晶型Ⅰ、Ⅱ、Ⅲ的检测限分别为1.58%、1.35%、0.67%,定量限分别为4.74%、4.05%、2.01%。A near-infrared spectroscopy was used to analyze three crystal forms of furosemide, and the reference spectra were established. The quantitative model of furosemide crystal form was established by OPUS data analysis system. Partial least-squares （PLS） regression models were developed, and different data preprocessing algorithms such as linear compensation subtraction method, linear difference subtraction method, vector normalization method, minimummaximum normalization method, multiple scattering correction method and 1st to 2nd derivatives were applied to reduce the influence of systematic disturbances. The root mean square error of cross validation （RMSECV） of the quantitative model established for form Ⅰ, Ⅱ and Ⅲ was 1.51, 2.03 and 2.23, their determination coefficient R2 was 99.73, 99.43 and 99.16, and the rank was 9, 6 and 8, respectively. Their LODs were 1.58%, 1.35% and 0.67%, and their LOQs were 4.74%, 4.05% and 2.01%, respectively.

关 键 词：呋塞米 晶型 近红外光谱法 定量分析 

分 类 号：O657.33[理学—分析化学] TQ460.72[理学—化学]