Abstract: Background　Both total hip arthroplasty（THA） and hemiarthroplasty（HA） are common treatment methods for fresh femoral neck fractures in the elderly.However，although there are many retrospective studies or non-randomized controlled trials comparing the efficacy and safety of the two，relevant systematic reviews are rare.Objective　To systematically evaluate the efficacy and safety of THA versus HA for fresh femoral neck fractures in the elderly.Methods　From January to March 2017，we searched Cochrane Library，Medline，EMBase，CNKI，VIP and Wanfang Data Knowledge Service Platform and printed journals for randomized controlled trials（RCTs） evaluating the efficacy and safety of THA and HA for fresh femoral neck fractures in the elderly，and reviewed the references of the relevant RCTs published in the printed journals.Data were extracted from eligible RCTs and quality assessment was conducted.Results　Twelve RCTs 〔quality grade：A（3），B（6），C（3）〕 were included involving 1 310 patients.Meta-analysis showed that compared with those receiving HA，patients receiving THA had much higher Harris score at 1-year follow-up〔MD=3.93，95%CI（2.09，5.77），P<0.000 1〕.However，there were no significant differences between the two groups in Harris score when the follow-up interval was 2 years，3-4 years and more than 4 years（P>0.05）．The revision rate was similar in both groups at 1-year，2-year and 3-4-year follow-up（P>0.05），but patients with THA showed much lower revision rate when the follow-up interval was longer than 4 years〔RR=0.27，95%CI（0.14，0.54），P=0.000 2〕.The wear rate of the acetabular component showed no significant differences between two groups at 1-year，and 2-year follow-up（P>0.05），but it was much lower in patients with THA when the follow-up interval was 3-4 years〔RR=0.05，95%CI（0.01，0.22）〕 and longer than 4 years〔RR=0.05，95%CI（0.01，0.31）〕（P<0.05）．The fatality rate was similar in both groups at 3 different follow-up periods（1 year，2-4 years，> 4 years）（P<0.05）．The dislocation rate was similar in both groups at 1-year，2-4-year follow-up（P>0.05），but it was significantly increased in THA patients when the follow-up interval was longer than 4 years 〔RR=2.33，95%CI（1.27，4.29），P=0.006〕.Two groups showed no significant differences in the deep infection rate at 3 different follow-up periods（1 year，2-4 years，> 4 years），incidence of postoperative pain，and incidence of postoperative complications（P>0.05）．Conclusion　The differences in Harris score，revision rate，wear rate of acetabular component，and dislocation rate between patients with THA and those with HA changed according to the time of follow-up，while the fatality rate and deep infection rate did not.

Key words: Femoral neck fractures；Arthroplasty, replacement, hip；Treatment outcome；Systematic reviews

摘要： 背景　全髋关节置换（THA）与半髋关节置换（HA）是老年新鲜股骨颈骨折常用的治疗手段，但当前对两者疗效和安全性的比较多为回顾性或非随机对照试验，尚缺乏系统评价。目的　对THA与HA治疗老年新鲜股骨颈骨折的疗效和安全性进行系统评价。方法　2017年1—3月，计算机检索Cochrane图书馆、Medline、EMBase、中国知网、维普网、万方数据知识服务平台关于THA与HA治疗老年新鲜股骨颈骨折疗效和安全性的随机对照试验，同时手工检索相关期刊及追溯参考文献。提取纳入文献相关信息，并进行质量评价。结果　本研究纳入12篇文献（文献质量A级3篇，B级6篇，C级3篇），共1 310例患者。系统评价显示，THA患者随访1年时Harris评分高于HA〔MD =3.93，95%CI（2.09，5.77），P<0.000 1〕，随访2、3~4年及4年以上时Harris评分与HA无差异（P>0.05）。THA患者随访1、2年及3~4年翻修率与HA无差异（P>0.05），随访4年以上时翻修率低于HA〔RR=0.27，95%CI（0.14，0.54），P=0.000 2〕。THA患者随访1、2年髋臼磨损率与HA无差异（P>0.05），随访3~4年〔RR=0.05，95%CI（0.01，0.22）〕及4年以上〔RR=0.05，95%CI（0.01，0.31）〕时髋臼磨损率低于HA（P<0.05）。THA患者随访1、2~4年及4年以上时病死率与HA无差异（P>0.05）。THA患者随访1、2~4年脱位率与HA无差异（P>0.05），随访4年以上时脱位率高于HA〔RR=2.33，95%CI（1.27，4.29），P=0.006〕。THA患者随访1、2~4年及4年以上时深部感染率，以及术后疼痛发生率、并发症发生率与HA无差异（P>0.05）。结论　THA与HA治疗老年新鲜股骨颈骨折的Harris评分、翻修率、髋臼磨损率、脱位率差异随着随访时间的延长而不同，而病死率、深部感染率则不受随访时间的影响。

关键词: 股骨颈骨折, 关节成形术, 置换, 髋, 治疗结果, 系统评价