目的 建立基于多指标测定及化学计量方法的黄精属药材质量评价体系，对不同物种黄精属药材（滇黄精P. kingianum、多花黄精P. cyrtonema、点花黄精P. punctatum、互卷黄精P. altenicirrhosum及玉竹P. odoratum）进行综合质量评价。方法 测定38批不同物种黄精属药材水分、总灰分、浸出物、黄精多糖、总黄酮及薯蓣皂苷元含量，采用单因素方差分析比较各指标在不同物种间的差异；基于熵权法确定各指标权重，并结合灰色关联度法构建黄精属药材质量评价模型。结果 不同物种黄精属药材5项指标差异显著，浸出物含量不存在显著差异。38批样品相对关联度介于0.435 7～0.601 4，相对关联度＞0.500的样品有8批，总体上多花黄精和滇黄精的质量评价最高。结论 基于多指标成分的单因素方差分析及构建的熵权-灰色关联度模型可用于黄精属不同物种药材的质量评价，该方法可较好地反映黄精属药材质量，提高黄精属药材质量评价的科学性。

Objective To establish a quality evaluation system for Polygonatum medicinal materials based on multi-index determination and chemometric methods, and to comprehensively evaluate the quality of Polygonatum medicinal materials from different species (P. kingianum, P. cyrtonema, P. punctatum, P. altenicirrhosum, and P. odoratum). Methods The contents of moisture, total ash, extract, Polygonatum polysaccharides, total flavonoids, and diosgenin were determined in 38 batches of Polygonatum medicinal materials from different species. One-way analysis of variance was used to compare the differences in each index among different species. The entropy weight method was applied to determine the weight of each index, and a quality evaluation model for Polygonatum medicinal materials was constructed in combination with grey relational analysis. Results Significant differences were observed in five indexes among different species of Polygonatum medicinal materials, while no significant difference was found in extract content. The relative correlation degrees of the 38 batches of samples ranged from 0.435 7 to 0.601 4. Eight batches showed a relative correlation degree greater than 0.500. Overall, P. cyrtonema and P. kingianum exhibited the highest quality evaluation results. Conclusion One-way analysis of variance based on multi-index components, together with the entropy weight-grey relational analysis model, can be used for quality evaluation of Polygonatum medicinal materials from different species. This method can effectively reflect the quality of Polygonatum medicinal materials and improve the scientific basis of their quality evaluation.

R282.6

云南省科技厅科技计划项目（202304BI090004）；国家自然科学基金资助项目（82360751）