目的 系统评价口服中成药治疗卒中后抑郁的有效性和安全性。方法 检索中国期刊全文数据库（CNKI）、维普数据库（VIP）、万方数据库、PubMed、Cochrane Library，检索起止时间为建库至2020年12月。由2位研究员独立筛选文献、提取资料并按照Jadad量表对文献进行质量评价，采用Stata 16.0软件进行统计分析。结果 共纳入59个研究，总样本量6 074例，干预措施包括乌灵胶囊、舒肝解郁胶囊、解郁丸、养血清脑颗粒、益心舒胶囊、舒肝颗粒、甜梦口服液共7种中成药。网状Meta分析结果显示，在汉密尔顿抑郁量表（HAMD）评分方面，累积概率排序为：舒肝颗粒+常规治疗（90.3%）>舒肝解郁胶囊+常规治疗（74.9%）>甜梦口服液+常规治疗（67.4%）>益心舒胶囊+常规治疗（55.0%）>乌灵胶囊+常规治疗（47.7%）>解郁丸+常规治疗（38.3%）>养血清脑颗粒+常规治疗（25.7%）>常规治疗（0.6%）。在抑郁临床疗效方面，累积概率排序为：解郁丸+西医常规治疗（74.5%）>乌灵胶囊+西医常规治疗（65.9%）>养血清脑颗粒+西医常规治疗（56.1%）>舒肝解郁胶囊+西医常规治疗（53.4%）>西医常规治疗（0.1%）。在美国国立卫生研究院卒中量表（NIHSS）评分方面，累积概率排序为：舒肝颗粒+常规治疗（95.6%）>养血清脑颗粒+常规治疗（74.4%）>舒肝解郁胶囊+常规治疗（61.8%）>解郁丸+常规治疗（34.2%）>乌灵胶囊+常规治疗（32.5%）>常规治疗（1.5%）。在Barthel指数方面，养血清脑颗粒+常规治疗（73.6%）>舒肝颗粒+常规治疗（73.1%）>益心舒胶囊+常规治疗（65.1%）>舒肝解郁胶囊+常规治疗（61.7%）>乌灵胶囊+常规治疗（26.1%）>常规治疗（0.2%）。结论 对于卒中后抑郁患者，舒肝颗粒在HAMD量表和NIHSS评分方面疗效排序靠前，解郁丸和养血清脑颗粒分别在抑郁临床疗效以及提高Barthel指数方面效果最好。但受纳入研究质量所限，且存在发表偏倚，该结论仍需开展高质量研究进一步验证，并根据中医证型、病人情况进行选择。

Objective To evaluate the efficacy and safety of Chinese patent medicine in the treatment of post-stroke depression. Methods China National Knowledge Infrastructure (CNKI), VIP database (VIP), Wanfang Database, PubMed and Cochrane Library were all searched until December 2020. Two researchers independently screened the literature, extracted the data, and evaluated the quality of the literature according to the Jadad scale. Stata 16.0 was used for statistical analysis. Results A total of 57 studies were included, with a total sample size of 6 074 cases. The measures included 7 Chinese medicines, covering Wuling Capsule (WLC), Shugan Jieyu Capsule (SGJYC), Jieyu Pill (JYP), Yangxue Qingnao Granule (YXQNG), Yixinshu Capsule (YXSC), Shugan Granule (SGG), Tianmeng Oral Liquid (TMOL). On the part of HAMD, results of network meta analysis showed that the rank of cumulative probability was: SGG + CT (90.3%)>SGJYC + CT (74.9%)>TMOL + CT (67.4%)>YXSC + CT (55.0%)>WLC + CT (47.7%)>JYP + CT (38.3%)>YXQNG + CT (25.7%) >CT (0.6%). On the part of clinical effects in depression, results of network meta analysis showed that the rank of cumulative probability was: JYP + CT (74.5%)>WLC + CT (65.9%)>YXQNG + CT (56.1%)>SGJYC + CT (53.4%)>CT(0.1%). In terms of NIHSS, the rank of cumulative probability was: SGG + CT (95.6%)>YXQNG + CT (74.4%)>SGJYC + CT (61.8%)>JYP + CT (34.2%)>WLC + CT (32.5%)>CT (1.5%). In the aspect of Barthel index, the rank of cumulative probability was: YXQNG + CT (73.6%) >SGG + CT (73.1%)>YXSC + CT (65.1%)>SGJYC + CT (61.7%)>WLC + CT (26.1%)>CT (0.2%). Conclusion For patients with post-stroke depression, SGG had better effect in HAMD scale and NIHSS score, while JYP and YXQNG ranked first in improving clinical curative effects and Barthel index. However, limited by the quality of the included studies, this conclusion still needs to be further verified by highquality studies. Moreover, we should select Chinese patent medicines referring to TCM syndrome and patient condition.

R965.1;R971

北京市双一流高层次人才科研经费资助项目（1000041510053）