目的:系统评价应用阿仑膦酸3~4年预防脆性骨折的结果。方法:系统搜索MED-LINE、EMBASE、CENTRAL和中国生物医学文献数据库及相关研究的引文。时间范围建库至2020年7月16日。使用Cochrane偏倚风险评估工具进行质量评估。根据异质性大小决定使用固定或随机效应模型。对高风险研究进行敏感性分析。按照PRISMA 2009规范进行报告。结果:共纳入4个研究8篇报告。2个研究2篇报告结果为高风险，其余2个研究6篇报告为低风险。结果显示使用阿仑膦酸3~4年可有效预防新发椎体骨折（n RR=0.54，95%n CI：0.44~0.66，n RD=-0.03）、临床骨折（n RR=0.82，95%n CI：0.73~0.92，n RD=-0.03）、非椎体骨折（n RR=0.84，95%n CI: 0.75~0.95，n RD=-0.02）、临床椎体骨折（n RR=0.51，95%n CI：0.34~0.76，n RD=-0.01）和髋部骨折（n RR=0.56，95%n CI：0.37~0.87，n RD=-0.01）；但未见能有效预防腕部骨折（n RR=0.85，95%n CI：0.67~1.09）、严重不良事件（n RR=0.95，95% n CI：0.80~1.14）和上消化道不良事件（n RR=1.02，95%n CI：0.96~1.07）。对纳入高风险报告的6个结果进行敏感性分析：临床骨折、非椎体骨折、椎体骨折、髋部骨折、腕部骨折、严重不良事件的n HR分别为0.81、0.85、0.49、0.62、0.94、0.94。n 结论:口服阿仑膦酸3~4年可有效预防社区绝经后低骨量或骨质疏松女性脆性骨折，其中对椎体骨折的预防效果优于非椎体骨折。“,”Objective:To evaluate systematically the outcomes of prevention of fragility fracture with alendronate use for 3 to 4 years.Methods:We searched CENTRAL, MEDLINE, EMBASE and CBM for relevant randomized controlled trials published before 16 July, 2020. The quality of included studies was evaluated according to the Cochrane tool for assessing risk of bias. Fix- or random-effects were taken for meta-analysis depending on the magnitude of heterogeneity. Sensitivity analysis was used for high risk studies to assess the robustness of results. The results were reported according to The PRISMA 2009 Check-list.Results:A total of 8 reports from 4 studies were included. Two reports from 2 studies were rated as high-risk while the other 6 reports from the other 2 studies as low-risk. The meta-analyses showed that use of alendronic acid for 3 to 4 years effectively prevented new vertebral fracture (n RR=0.54, 95%n CI: 0.44 to 0.66, n RD=-0.03), clinical fracture (n RR=0.82, 95%n CI: 0.73 to 0.92, n RD=-0.03), non-vertebral fracture (n RR=0.84, 95%n CI: 0.75 to 0.95, n RD=-0.02), clinical vertebral fracture (n RR=0.51, 95% n CI: 0.34 to 0.76, n RD=-0.01) and hip fracture (n RR=0.56, 95%n CI: 0.37 to 0.87, n RD=-0.01), but did not prevent wrist fracture (n RR=0.85, 95%n CI: 0.67 to 1.09), serious adverse event (n RR=0.95, 95%n CI: 0.80 to 1.14) or upper gastrointestinal adverse event (n RR=1.02, 95%n CI: 0.96 to 1.07). By the sensitivity analysis of the 6 results from the high-risk reports, the HRs for clinical fracture, non-vertebral fracture, vertebral fracture, hip fracture, wrist fracture, and serious adverse event were, respectively, 0.81, 0.85, 0.49, 0.62, 0.94 and 0.94.n Conclusions:Alendronate use for 3 to 4 years can effectively prevent fragility fractures in postmenopausal women with low bone mass or osteoporosis, leading to better prevention effect on vertebral fracture than on non-vertebral fracture.