陳能　呂燃　馬少云　全華山　賈曉軍　譚方　李敏龍

【摘 要】目的：系統(tǒng)評價單髁置換術(shù)治療自發(fā)性膝關(guān)節(jié)骨壞死的有效性和翻修率情況。方法：計算機(jī)檢索Web of Science、Pubmed、EMbase等英文數(shù)據(jù)庫，以及中國知網(wǎng)（CNKI）、萬方（WANFANG）、維普（VIP）、中國生物醫(yī)學(xué)文獻(xiàn)數(shù)據(jù)庫（CBM）等中文數(shù)據(jù)庫，檢索時間均為數(shù)據(jù)庫建立至2018年9月30日。檢索單髁置換術(shù)治療膝關(guān)節(jié)自發(fā)性骨壞死的單臂隨機(jī)對照試驗，經(jīng)文獻(xiàn)篩選、提取資料后，采用Stata 15.1軟件進(jìn)行療效Meta分析。結(jié)局指標(biāo)包括治療前后視覺模擬評分法（VAS）評分、美國特種外科醫(yī)院膝關(guān)節(jié)評分（HSS評分）、關(guān)節(jié)活動度（ROM）、股骨脛骨角（FTA）以及翻修率，采用隨機(jī)效應(yīng)模型分析。結(jié)果：初次檢索獲得文獻(xiàn)277篇，按照納入及排除標(biāo)準(zhǔn)，最終納入21篇文獻(xiàn)，共計703例，Meta分析結(jié)果顯示，9篇文獻(xiàn)報道了治療前后VAS評分，治療前VAS評分為7.04分（95%CI =?[6.50，7.58]，P < 0.001，I? = 93.5%）;治療后VAS評分為1.96分（95%CI = [1.39，2.54]，P < 0.001，I? = 95.2%）。10篇文獻(xiàn)報道了治療前后HSS評分，治療前HSS評分為62.76分（95%CI = [59.00，66.53]，P < 0.001，I? = 94.5%）;治療后HSS評分為89.68分（95%CI = [86.90，92.45]，P < 0.001，I?= 95.6%）。10篇文獻(xiàn)報道了治療前后ROM，治療前ROM為112.82°（95%CI = [105.52，120.12]，P < 0.001，I? = 98.8%）;治療后ROM為119.63°（95%CI = [114.05，125.21]，P < 0.001，I? = 97.2%）。11篇文獻(xiàn)報道了手術(shù)前后FTA，手術(shù)前FTA為179.44°（95%CI = [177.99，180.89]，P < 0.001，I? = 94.0%）;手術(shù)后FTA為177.36°（95%CI = [176.85，177.88]，P < 0.001，I? = 60.3%）。21篇文獻(xiàn)報道了翻修率，為0.02（95%CI = [0.01，0.05]，P = 0.01，I? = 46.63%）。結(jié)論：單髁置換術(shù)治療膝關(guān)節(jié)自發(fā)性骨壞死療效良好，翻修率低。

【關(guān)鍵詞】 膝關(guān)節(jié)自發(fā)性骨壞死;單髁置換術(shù);翻修率;Meta分析

【ABSTRACT】Objective：To systematically evaluate the effectiveness and revision rate of unicondylar replacement in the treatment of spontaneous knee osteonecrosis.Methods：The English databases of Web of Science，Pubmed，EMbase and Chinese databases such as CNKI，WANFANG，VIP and CBM were searched by computer.The searching time was from the establishment of above databases to September 30，2018.Retrieval was done for the single arm randomized controlled trials of unicondylar replacement for the treatment of spontaneous osteonecrosis of the knee joint.After literature screening and data extraction，meta-analysis of the efficacy was carried out with Stata 15.1 software.Outcome indicators included VAS before and after treatment，HSS，ROM，F(xiàn)TA and revision rate.Random effect model was used to analyze the outcome.Results：Two hundred and seventy-seven articles were retrieved for the first time，and 21 of them were included according to inclusion and exclusion criteria.A total of 703 patients were included.Meta analysis showed that nine articles reported VAS scores before and after treatment.The VAS score before treatment was 7.04（95% CI = [6.50，7.58]，P < 0.001，I? = 93.5%）.The VAS score after treatment was 1.96（95% CI = [1.39，2.54]，P < 0.001，I? = 95.2%.Ten articles reported HSS scores before and after treatment.HSS score before treatment was 62.76（95% CI = [59.00，66.53]，P < 0.001，I? = 94.5%），while HSS score after treatment was（95%CI = [86.90，92.45]，P < 0.001，I? = 95.6%）;Ten articles reported ROM before and after treatment.ROM before treatment was 112.82°（95%CI =?[105.52，120.12]，P < 0.001，I? = 98.8%），while ROM after treatment was 119.63°（95%CI = [114.05，125.21]，P < 0.001，I? = 97.2%）.Eleven articles reported FTA before and after surgery.FTA before surgery was?179.44°（95% CI = [177.99，180.89]，P < 0.001，I? = 94.0%），while FTA after surgery was 177.36°（95%CI =?[176.85，177.88]，P < 0.001，I? = 60.3%）.Twenty-one articles reported he renovation rate 0.02（95%CI = [0.01，0.05]，P = 0.01，I? = 46.63%）.Conclusion：Unicondylar arthroplasty is effective in the treatment of knee spontaneous osteonecrosis with low revision rate.It is expected to be further verified by high quality，large sample and long-term follow-up randomized controlled trials.