Studies On Quality Control Of Compound E-jiao Oral Liquid Using Near-infrared Spectroscopy

All key aspects in the pharmaceutical manufacturing process exert potential influence on the quality of end products. To ensure good quality, all these links, from raw materials and in-process intermediates to end products, should thus be put under rigorous monitoring and control. Compound E-jiao oral liquid (CEOL) is a wildly used oral TCM preparation in China. It has both botanical and animal medicines in its formulation; its manufacturing process encompasses typical unit operations such as extraction and precipitation. Given the status quo, near-infrared (NIR) spectroscopy in conjunction with multivariate data analysis (MVDA) was employed herein for quality control of the whole manufacturing process of CEOL, starting from one of the raw materials E-jiao (Donkey-hide gelatin) to in-process intermediates and eventually the end products. The main contents and achievements of this thesis are summarized as follows:1. Rapid quality evaluation methods for E-jiao were established based on NIR spectroscopy. For manufacturer and storage time discrimination, different identification methods were developed respectively. For the former, fingerprints with Hotelling T2, DModX and similarity match values (SMVs) as criteria were proposed; while for the latter, a PLS-DA model was established. These NIR-based methods comprehensively reflect information pertain to the quality of E-jiao and can be useful in quick quality tests of the raw materials before they enter the production line.2. Rapid quantification methods for the in-process extract solutions were established based on NIR spectroscopy. Simultaneous quantification of four quality indicators (viz. total flavonoids, total saponins, total saccharides and soluble solid contents) in the extract solutions was developed. A timely analysis of the extract solutions can help capture information regarding the completeness of the extraction as well as quality fluctuation between batches, upon which the extracting end-point determination is based.3. In-line monitoring of the precipitation after the extract solutions and the melted E-jiao were mixed was investigated combining NIR spectroscopy and multivariate statistical process control (MSPC). Control charts including principal component scores, Hotelling T2 and DModX were jointly employed for detection of abnormalities. This approach offers an opportunity to achieve timely fault diagnosis and to create a knowledge base concerning the various abnormalities that might occur during the process.4. The feasibility of analyzing the end products through unopened glass vials was explored. Short wavelength NIR region is endowed with this particular feature that materials such as glass or quartz are virtually transparent to NIR radiation. In light of this, a non-invasive method for the rapid analysis of CEOL was developed. In addition, NIR-based fingerprints of CEOL were established to comprehensively reflect the quality of the end products. With these fast and non-invasive methods, a larger portion of products could be sampled and tested.