中文摘要：

目的 采用荟萃分析的方法评价股骨近端防旋髓内钉（PFNA）与动力髋螺钉（DHS）治疗老年股骨粗隆间骨折的效果.方法 检索PubMed、EMbase、Cochrane Controlled Center Register ofControlled Trials、中国生物医学文献数据库、中国知网、万方数据库数据库的随机对照试验,检索时间从建库到2013年5月.采用Revman5.1与GRADEprofile 3.6软件进行分析和证据等级评定.结果 共纳入17个随机对照试验,1 344例患者.结果显示：与DHS相比,PFNA在内固定失败率（OR=0.34,95％ CI=[0.14,0.81],P=0.02）、骨折的平均愈合时间（WMD=-15.11,95％CI=[-24.36,-5.85],P=0.001）和Harris评分优良率（OR =1.97,95％CI=[1.15,3.37],P=0.02）及Harris评分分数（WMD=6.88,95％ CI=[0.96,12.80],P=0.02）、术后髋内翻的发生率（OR=0.40,95％ CI=[0.18,0.92],P=0.03）等方面更有优势;而在术后1年死亡率、术后股骨干骨折、骨折的不愈合和延迟愈合、术后股骨头坏死等方面,两者差异无统计学意义.GRADE证据质量为“极低级”到“低级”.结论 现有研究显示PFNA在老年股骨粗隆间骨折治疗效果上优于DHS,但仍需开展大样本、多中心、高质量RCT,并根据骨折分型进行亚组分析.

英文摘要：

Objective To assess the efficacies of proximal femoral nail antirotation （PFNA） versus dynamic hip screw （DHS） for intertrochanteric fracture in elders. Methods The databases of PubMed, Embase, CENTRAL, CBM, CNKI and WANFANG were searched for the relevant randomized controlled trials （RCTs） up to May 18, 2013. After quality evaluation and data extraction by two authors independently, meta-analyses were performed with the RevMarrS. 1 software. And the levels of evidence were evaluated by the GRADEprofile 3.6 software. Results A total of 17 studies （n = 1 344） were included and there were 656 patients in PFNA group and 688 in DHS respectively. Meta-analyses showed that, as compared with DHS, PFNA could significantly decrease fixation failure rate （odds ratio （OR） = 0. 26, 95% confidence interval （CI） = [ 0. 12, 0. 57 ], P = O. 0008 ）, reduce average fracture healing time （weighted mean difference （WMD） = - 15. 11, 95% CI = [ - 24. 36, - 5.85 ~, P = O. 001 ） , improve excellent and good rate for Harris score （ OR =0. 29, 95% CI = [0. 15, 0. 56] , P =0. 0002）, yield a higher Harris score （ WMD = 6. 88, 95 % CI = [ 0. 96, 12. 80 ], P = 0. 02） and reduce the rate of coxa vara （ OR = 0. 40, 95 % CI = [ 0. 18,0. 92 ], P = 0. 03 ） . However, there was no statistical significance in 1-year mortality, postoperative nonunion and delayed union, postoperative femoral head necrosis, postoperative femoral fractures or femoral head necrosis. The importance of outcomes was ＂critical＂. And the level of evidences based on the GRADE approach was from ＂very low＂ to ＂low＂. Conclusion PFNA is superior to DHS. Due to the limitations of the included studies, more large-sample and high-quality RCTs are required. And subgroup analysis based on fracture types should be performed.