Dual control system for comprehensive evaluation of quality consistency of herbal preparations

Abstract

This study was undertaken to establish an integrated quality control method based on the fingerprint of different testing instruments and a four-dimensional antioxidant activity profile. Firstly, the comprehensive ratio quantitative fingerprint method was used to rapidly calculate the similarity of GC fingerprint, UV-vis quantized fingerprint, HPLC fingerprint and ATR-FT-IR quantized fingerprint. In addition, a comprehensive evaluation strategy was proposed using the variation coefficient weighting algorithm, incorporating four complementary inputs, to achieve highly accurate analysis. At the same time, multi-markers assay by monolinear method was used for the first time for the quantitative analysis of multiple components. Based on this, the fingerprint-efficacy relationship was investigated, and substances that might be biologically active and the wavelength range of their distribution were predicted using the online-HPLC-DPPH-FIA method in conjunction with a PLS model built from individual and combined data matrices. The results showed that all 20 batches of samples were within 1–3 grades, with good quality consistency. According to index E, the samples' overall antioxidant capacity was also highly correlated with the year of production. The four-dimensional antioxidant activity profile also led to the conclusion that variations in antioxidant characteristics were caused by differences in the concentration and volatility of particular bioactive chemicals. Finally, mid-level data fusion produced better classification results than a single technique, further demonstrating the synergistic effects that may occur when the four types of data were combined. This study demonstrated that the combination of GC, UV–vis, ATR-FT-IR and HPLC can be used for the consistency control of herbal preparations and can elucidate the potential in predicting the antioxidant capacity of herbal preparations.