目的 对比黄柏配方颗粒汤剂（DGD）与标准汤剂（SD）、传统汤剂（TD）的差异，对市售配方颗粒（DG）的质量进行评价，并建立SD、TD的相关标准及DG的评价方法。方法 采用HPLC建立指纹图谱，从化学成分种类、代表性指标成分含量、指纹图谱相似度、共有峰峰面积总和、主成分分析（PCA）5个方面对DGD（A～E 5个厂家各3批）与SD（10批）、TD（10批）共7类35个供试品进行全面对比研究；对DGD进行临床建议当量校正。结果 ①SD和TD中的21个共有峰均在DGD指纹图谱中得到保留。②DGD中A厂木兰花碱含量比SD低34.3%（P<0.05）；C厂木兰花碱含量比SD低35.6%（P<0.01），比TD低37.0%（P<0.05）；D厂盐酸黄柏碱含量比SD低22.0%（P<0.05），比TD低27.5%（P<0.05）；D厂盐酸小檗碱的含量比SD低20.8%（P<0.05），比TD低23.8%（P<0.05）；其他厂家各成分与SD、TD无显著性差异（P>0.05）。③各DGD与SD平均图谱相似度均大于0.992 6，与TD平均图谱相似度均大于0.991 2，成分相似度较高。④采用归一化法以SD的21个共有峰的总峰面积值为1个单位，7类供试品比值分别为0.90、1.03、0.69、0.77、0.73、1.00、1.06。⑤PCA显示B厂与SD、TD距离接近，差异小。以SD的21个共有峰信息为标准，采用峰面积加和法对DG的临床建议当量进行校正，建议A、C、D、E厂家由1 g相当于原饮片12 g分别降低至10.7、8.3、9.2、8.8 g，B厂家无需校正，仍为1 g相当于原饮片10 g。结论 真实世界中多数黄柏DGD与SD、TD间存在成分含量上的差异，并无成分种类和成分种类间比例的明显差异；这些整体基本一致的差异可通过对临床建议当量进行校正加以调整，从而促进临床合理用药。

Objective Comparing the differences between dispensing granule decoction (DGD), standard decoction (SD) and traditional decoction (TD) of Phellodendri Chinensis Cortex (PCC) prescription to evaluate the quality of commercially available dispensing granule (DG), and establish the relevant standards for SD, TD and evaluative methods for DG. Methods Fingerprint was established by HPLC. A comprehensive comparative study was conducted on 35 samples of DGD (three batches from each of the five A-E manufacturers), SD (10 batches) and TD (10 batches) in seven categories from five aspects of chemical composition type, representative index component content, fingerprint similarity, total peak area sum and principal component analysis (PCA); Clinically recommended equivalent corrections were performed for DGD. Results ① Twenty-one common peaks in SD and TD were preserved in the DGD fingerprint. ② The content of magnoflorine in manufacturer A of DGD was 34.3% lower than that of SD (P<0.05); The content of magnoflorine in manufacturer C was 35.6% lower than SD (P<0.01), and 37.0% lower than TD (P<0.05); The content of phellodendrine hydrochloride in D manufacturer was 22.0% lower than SD (P<0.05), and 27.5% lower than TD (P<0.05), The content of berberine hydrochloride in D manufacturer was 20.8% lower than SD (P<0.05), and 23.8% lower than TD (P<0.05). There were no significant differences between the other manufacturers' components. ③ The average similarity of each DGD and SD was greater than 0.992 6, and the average similarity of each DGD and TD was greater than 0.991 2, with high component similarity. ④ Using the normalization method, the total peak area of the 21 common peaks of SD was 1 unit, and the ratios of the seven types of samples were 0.90, 1.03, 0.69, 0.77, 0.73, 1.00, and 1.06. ⑤ PCA showed that the distance between the B manufacturer and SD and TD was close, and the difference was small. Using the 21 common peak information of SD as the standard, the peak area plus method was used to correct the clinical recommended equivalent of DG. It was recommended that manufacturers A, C, D, and E could be reduced from 1 g equivalent to 12 g of the original decoction pieces to 10.7, 8.3, 9.2, and 8.8 g, respectively. B manufacturer was not needed to be corrected, and still 1 g was equivalent to 10 g of the original decoction pieces. Conclusion There are differences in the content of components between DGD, SD, and TD in the real world. There is no significant difference in the proportion of components and components. These overall basically consistent differences can be adjusted by correcting the clinical recommended equivalent, thus promoting clinical rational drug use.

国家自然科学基金面上项目（81773892）；河南省首批自然科学基金面上项目（162300410187）；河南省高等学校重点科研项目（16A360021）；河南省中医药科学研究专项课题（2016ZY2055）；河南省中医管理局国家中医临床研究基地科研专项（2018JDZX039）；河南省中医药拔尖人才培养项目资助（2019ZYBJ07）