目的 采用HPLC-UV-MS法建立丹红注射液多元指纹图谱，并对其中9种主要药效成分同时定量分析。方法 采用Atiantis T3色谱柱（250 mm×4.6 mm，5 μm），流动相为0.05%甲酸水溶液-50%乙腈水溶液，梯度洗脱；质谱检测采用负离子选择离子（SIM）模式。结果 该多元指纹图谱较全面地体现了制剂中2味处方药材丹参与红花的化学信息，11批制剂指纹图谱相似度均在0.988以上。5-羟甲基糠醛、丹参素钠、原儿茶醛、香豆酸、丹酚酸D、迷迭香酸、丹酚酸B、丹酚酸A、山柰酚-3-O-芸香糖苷9种成分在各自质量浓度范围内呈良好线性关系，r≥0.999 0；9种化合物的平均加样回收率分别为（99.0±1.4）%、（102.0±1.7）%、（99.3±1.6）%、（97.6±1.6）%、（100.0±1.8）%、（97.9±1.6）%、（100.5±4.4）%、（100.6±2.0）%、（106.0±4.7）%（n＝9）；方法重复性的RSD均小于1.45%；测定的11批制剂中9种成分总量在2.61～3.06 mg/mL。结论 该方法简单、准确、重复性好，多元指纹图谱结合定量测定能更全面地反映丹红注射液的质量，可用于制剂的质量控制。

Objective To establish the multiple fingerprints of Danhong Injection (DI) using HPLC-UV-MS method and to simultaneously determine nine kinds of the medicinal components in DI. Methods Separation was performed on Atiantis T3 analytical column (250 mm × 4.6 mm, 5 μm) with the mobile phase of 0.05% formic acid-50% acetonitrile by gradient elution, and negative-ion SIM mode was selected for mass spectrometric detection. Results The multiple fingerprints reflected the chemical information of Salvia miltiorrhiza and safflower in DI. The similarity of the fingerprints was higher than 0.988 in all 11 batches of DI. 5-Hydroxymethyl furfural, sodium danshensu, protocatechuic aldehyde, coumaric acid, salvianolic acid D, rosmarinic acid, salvianolic acid B, salvianolic acid A, and kaempferol-3-O-rutinoside showed the good linearity in their respective ranges of concentration, r ≥ 0.999 0. The average recovery of low-, mid-, and high-dose medicinal components (n = 9) was (99.0 ± 1.4)%, (102.0 ± 1.7)%, (99.3 ± 1.6)%, (97.6 ± 1.6)%, (100.0 ± 1.8)%, (97.9 ± 1.6)%, (100.5 ± 4.4)%, (100.6 ± 2.0)%, and (106.0 ± 4.7)%, respectively. The relative standard deviation for the method reproducibility was lower than 1.45%. The total contents of nine medicinal components in the 11 batches of DI were 2.61—3.06 mg/mL. Conclusion This method is simple, accurate, and with good reproducibility, and the multiple fingerprints combined with the quantitative analysis could reflect the quality of DI better, which could be used to control the quality of DI.

国家自然科学基金资助项目（81273992）