Multivariate analysis based on chromatographic fingerprinting for the evaluation of batch-to-batch reproducibility in traditional Chinese medicinal production

Abstract

In traditional Chinese medicinal production, a high degree of batch-to-batch reproducibility is required to achieve successful batches that meet the quality standard of traditional Chinese medicinal preparations (TCMPs). In this paper, multivariate statistical process monitoring (MSPM) based on chromatographic fingerprinting has been introduced to evaluate the batch-to-batch reproducibility of TCMPs. A commonly used TCMP product, Sheng-Mai-Yin (SMY) oral liquid, is selected as a typical example to demonstrate the feasibility of this approach. A number of batches are prepared through normal operation procedures to simulate the common-cause variations induced by preparation days, herb materials, and operators. Another number of abnormal batches are designed to simulate different types of deviations, including the changes of herb material quality attributes and abnormal operation conditions. The high-performance liquid chromatographic fingerprinting is used to construct the integrative pattern of SMY products. A principle component analysis model is established of the fingerprint data of the normal batches. Multivariate (Hotelling T2 and DModX) control charts are successfully applied to monitor the abnormal batches. As an effective multivariate analysis method in this study, MSPM based on chromatographic fingerprinting promises great applications in batch monitoring and evaluation for the production of TCMPs.