目的 通过检测白前Cynanchi Stauntonii Rhizoma et Radix中主要药效成分含量及醇溶性浸出物、总灰分和酸不溶性灰分，建立用于白前质量等级预测的化学计量学及Logistic回归分析方法。方法 对12省36批白前样品进行回流提取，以熊果酸为内参物，采用一测多评（quantitative analysis of multi-components by single-marker，QAMS）法检测提取物中白薇苷A、白薇苷B、白前苷B、熊果酸、胡萝卜苷、β-谷甾醇含量，并与外标法测得结果进行比较，同时对醇溶性浸出物、总灰分和酸不溶性灰分进行检测；采用化学识别模式、因子分析法及Logistic回归分析建立白前质量优劣评价模型，对其质量差异性进行综合评价。结果 熊果酸为内参物时，相对校正因子耐用性良好，相对保留时间值法可用于色谱峰定位，外标法与QAMS法所得含量结果无明显差异；白薇苷A、白薇苷B、白前苷B、熊果酸、胡萝卜苷、β-谷甾醇6个成分线性范围分别为0.37～9.25、1.56～39.00、1.85～46.25、0.29～7.25、0.58～14.50和1.15～28.75µg/mL，平均加样回收率为97.66%～100.12%，RSD为0.69%～1.51%；主成分分析（principal component analysis，PCA）和正交偏最小二乘判别分析法（orthogonal partial least squares-discriminant analysis，OPLS-DA）能明确区分不同产地的白前药材，提取了2个主成分，3个质量差异因子；因子分析法结果显示36批白前的综合得分在−1.225～0.966，其中S25综合得分最高。Logistic回归模型结果与因子分析法分析结果一致。结论 PCA、OPLS-DA、因子分析和Logistic回归模型可以用于评价不同产地白前的质量差异，为白前质量控制提供参考。

Objective To establish a degree prediction method of Baiqian (Cynanchi Stauntonii Rhizoma et Radix) base on chemometrics and Logistic regression analysis method by detecting the contents of main medicinal components, alcohol-soluble extracts, total ash and acid-insoluble ash. Methods Reflux extraction was performed on 36 batches of Cynanchi Stauntonii Rhizoma et Radix from 12 provinces, and the contents of cynatratoside A, cynantratoside B, vincetoxicoside B, ursolic acid, eleutheroside A and β-sitosterol in the extracts were determined by quantitative analysis of multi-components by single-marker (QAMS) method with ursolic acid as internal reference substance, and comparison with the external standard method, at the same time, alcohol-soluble extract, total ash and acid-insoluble ash were detected. The chemical identification model, factor analysis and Logistic regression model were used to establish a quality evaluation model for Cynanchi Stauntonii Rhizoma et Radix, and the quality differences were comprehensively evaluated. Results When ursolic acid was used as the internal reference, the relative correction factor had good durability. The relative retention time method could be used for chromatographic peak location. There was no significant difference between the content results obtained by the external standard method and the QAMS method. The linear ranges of six components were 0.37—9.25, 1.56—39.00, 1.85—46.25, 0.29—7.25, 0.58—14.50 and 1.15—28.75 μg/mL, respectively. The average recoveries were 97.66%—100.12% with RSDs of 0.69%—1.51%. Principal component analysis (PCA) and orthogonal partial least squares discriminant analysis (OPLS-DA) could clearly distinguish Cynanchi Stauntonii Rhizoma et Radix from different producing areas. Two principal components and four quality differential factors were extracted. The results of factor analysis showed that the comprehensive scores of 36 batches of Cynanchi Stauntonii Rhizoma et Radix were −1.225—0.966, and the comprehensive score of S25 was the highest. The results of Logistic regression model were consistent with the results of factor analysis. Conclusion PCA, OPLS-DA, factor analysis and Logistic regression model can be used to evaluate the quality differences of Cynanchi Stauntonii Rhizoma et Radix from different producing areas, so as to provide reference for the quality control of Cynanchi Stauntonii Rhizoma et Radix.

R286.2

山西省中医药管理局科研课题（2023ZYYB052）