Hassaninejad-Darzi Seyed Karim, Samadi-Maybodi Abdolraouf, Nikou Seyed Mohsen
Department of Chemistry, Faculty of Science, Babol University of Technology, Babol, Iran.
Analytical Division, Faculty of Chemistry, University of Mazandaran, Babolsar, Iran.
Iran J Pharm Res. 2016 Summer;15(3):379-391.
Resolution of binary mixtures of theophylline (THEO), montelukast (MKST) and loratadine (LORA) with minimum sample pre-treatment and without analyte separation has been successfully achieved by multivariate spectrophotometric calibration, together with partial least-squares (PLS-1), principal component regression (PCR) and hybrid linear analysis (HLA). Data of analysis were obtained from UV-Vis spectra of three compounds. The method of central composite design was used in the ranges of 2-14 and 3-11 mg L for calibration and validation sets, respectively. The models refinement procedure and their validation were performed by cross-validation. The minimum root mean square error of prediction (RMSEP) was 0.173 mg L for THEO with PCR, 0.187 mg L for MKST with PLS1 and 0.251 mg L for LORA with HLA techniques. The limit of detection was obtained 0.03, 0.05 and 0.05 mg L by PCR model for THEO, MKST and LORA, respectively. The procedure was successfully applied for simultaneous determination of the above compounds in pharmaceutical tablets and human plasma. Notwithstanding the spectral overlapping among three drugs, as well as the intrinsic variability of the latter in unknown samples, the recoveries are excellent.
通过多元分光光度校准，结合偏最小二乘法（PLS-1）、主成分回归（PCR）和混合线性分析（HLA），在最小化样品预处理且不进行分析物分离的情况下，成功实现了茶碱（THEO）、孟鲁司特（MKST）和氯雷他定（LORA）二元混合物的分辨。分析数据来自三种化合物的紫外可见光谱。分别在2 - 14和3 - 11 mg/L范围内使用中心复合设计方法构建校准集和验证集。通过交叉验证进行模型优化及其验证。采用PCR技术测定THEO时预测的最小均方根误差（RMSEP）为0.173 mg/L，采用PLS1技术测定MKST时为0.187 mg/L，采用HLA技术测定LORA时为0.251 mg/L。通过PCR模型分别获得THEO、MKST和LORA的检测限为0.03、0.05和0.05 mg/L。该方法成功应用于同时测定药片中以及人血浆中的上述化合物。尽管三种药物之间存在光谱重叠，且未知样品中药物存在固有变异性，但回收率良好。
