目的 通过系统评价和试验序贯分析（TSA）综合评价石杉碱甲治疗阿尔茨海默病（AD）的有效性和安全性。方法 计算机检索PubMed、Embase、Cochrane Library、Web of Science、中国期刊全文数据库（CNKI）、万方数据库（Wanfang data）、维普数据库（VIP）、中国生物医学文献数据库（CBM），纳入比较石杉碱甲治疗AD的随机对照试验和队列研究，检索时限均从建库至2024年4月1日。由2位评价员独立筛选和评价，应用RevMan 5.4和Stata 17.0软件进行系统评价，TSA 0.9.5.10 beta软件进行TSA。结果 共纳入25项研究，包括1 813例患者。Meta分析结果显示：（1）与对照组相比，试验组有效率更高[RR＝1.46，95% CI（1.29，1.65)，P＜0.000 01]，不同药物、剂量、疗程、痴呆程度、研究类型的亚组分析结果一致；（2）试验组MMSE量表评分、MQ量表评分、HDS-R量表评分更高[MD＝3.99，95% CI（2.32，5.67)，P＜0.000 01]、[MD＝5.09，95% CI（2.71，7.47)，P＜0.000 1]、[MD＝2.59，95% CI（1.62，3.55)，P＜0.000 01]；（3）试验组不良反应高于对照组，差异有统计学意义[RR＝1.62，95% CI（1.21，2.19)，P＝0.001]；（4）敏感性分析提示结果较为稳健；（5）有效率和不良反应的Egger’s检验提示存在发表偏倚；（6）TSA进一步表明石杉碱甲治疗AD有效率和不良反应结论的可靠性。结论 石杉碱甲能改善AD患者的认知功能，存在的不良反应多为轻度不良反应，临床可根据患者总体情况综合考虑。

Objective To analyze the efficacy and safety of huperzine A for the treatment of Alzheimer's disease (AD) by systematic evaluation and trial sequencing.Methods Computerized searches of PubMed, Embase, Cochrane Library, Web of Science, CNKI, Wanfang, VIP, and Chinese Biomedical Literature Database (CBM) were performed to include randomized controlled comparing huperzine A treatment of AD trials and cohort studies, and the search timeframe was from the establishment of the database to April 1st, 2024 for all of them. Two evaluators independently screened and evaluated the studies, applying RevMan 5.4 and Stata 17.0 software for systematic evaluation, and TSA 0.9.5.10 beta software for TSA. Results A total of 25 papers were included, including 1 813 patients. The results of the Meta-analysis showed that:(1) compared with the control group, the effectiveness rate of the huperzine A group was higher [RR=1.46 with a 95% CI (1.29, 1.65), P < 0.000 01], with consistent results for subgroup analyses of different drugs, dosages, duration of treatment, degree of dementia, and type of study; (2) The MMSE scale scores, MQ scale scores, and HDS-R scale scores were higher in the sarcosine A group [MD = 3.99, 95% CI (2.32, 5.67), P < 0.000 01], [MD = 5.09, 95%CI (2.71, 7.47), P < 0.000 01], [MD = 2.59, 95%CI (1.62, 3.55), P < 0.000 01]; (3) The adverse reactions in the huperzine A group were higher than those in the control group; (4) Sensitivity analysis suggested that the results were more robust; (5) Egger's test for the effective rate and adverse reactions suggested that there was publication bias; And (6) TSA further affirmed the reliability of the conclusions on the effective rate and adverse reactions in the treatment of AD with huperzine A. Conclusion Huperzine A can improve cognitive function in patients with AD, and the common adverse reactions are mostly mild. The clinical selection may vary based on the overall condition of the patients.

R971

国家自然科学基金资助项目（82160933）;广西壮族自治区自然科学基金项目（2022GXNSFAA035577）;广西壮族自治区中医药管理局科技项目（GXZYZ20210072）;广西中医药适宜技术与开发与推广项目（GZSY22-20）