本文選題：磁共振彈性成像　切入點(diǎn)：肝纖維化　出處：《吉林大學(xué)》2017年碩士論文　論文類型：學(xué)位論文

【摘要】：目的:采取Meta分析方式,探討磁共振彈性成像(MRE)對肝纖維化分級診斷優(yōu)勢及臨床適用性價值,為臨床肝纖維化治療提供有效依據(jù)。材料與方法:1、篩選文獻(xiàn):搜索2017年2月2日前國內(nèi)、外公開發(fā)表的磁共振彈性成像肝纖維化分期中文及英文文獻(xiàn)。納入數(shù)據(jù)庫包括萬方數(shù)據(jù)庫(WANFANG DATA),中國生物醫(yī)學(xué)文獻(xiàn)數(shù)據(jù)庫(CBM)、中國知網(wǎng)(CNKI)、維普資訊網(wǎng)(VIP)、EMBASE、PUBMED、Cochrane Library、Web of Science,并輔以手工查找相關(guān)文獻(xiàn),根據(jù)預(yù)先納入及排除標(biāo)準(zhǔn)進(jìn)行篩選。2、文獻(xiàn)質(zhì)量評價:依據(jù)診斷性Meta分析專用文獻(xiàn)質(zhì)量評價工具QUADAS-2(the Quality Assessment for Studies of Diagnostic Accuracy)進(jìn)行文獻(xiàn)質(zhì)量評價,并進(jìn)行文獻(xiàn)A、B等級分類。3、在納入文獻(xiàn)中提取所需數(shù)據(jù),并進(jìn)行數(shù)據(jù)處理。(1)提取文獻(xiàn)中所需數(shù)據(jù),制成表格。(2)對各組數(shù)據(jù)進(jìn)行異質(zhì)性檢驗(yàn),計(jì)算I2值判定異質(zhì)性水平。無明顯異質(zhì)性:I250%,異質(zhì)性較大:I2≥50%。根據(jù)異質(zhì)性檢測結(jié)果采用相應(yīng)模型進(jìn)行合并統(tǒng)計(jì)量。(3)應(yīng)用stata12.0軟件分別對F0 VS F1-F4病例組、F0-F1 VS F2-F4病例組、F0-F2 VS F3-F4病例組、F0-F3 VS F4病例組敏感度(Sensitivity,SEN)、特異度(Specificity,SPE)、陽性似然比(PosititiveLikelihood Ratio,+LR)、陰性似然比(Negative Likelihood Ratio,-LR)、診斷比值比(Diagnostic Odds Ratio,DOR)進(jìn)行合并及異質(zhì)性檢驗(yàn),繪制分層綜合受試者工作特征曲線(Hierarchical Summary Receiver Operating Characteristic,HROC),并計(jì)算曲線下面積(the Area Under the Receiver Operating Characteristic Curve,AUROC),對磁共振彈性成像肝纖維話分級診斷價值進(jìn)行評估。(4)通過計(jì)算本研究中各病例組的驗(yàn)后概率,并假設(shè)驗(yàn)前概率為25%、50%、75%,計(jì)算驗(yàn)后概率,通過Fagan Plot分析對MRE肝纖維化分級診斷臨床適用性進(jìn)行評估。(5)若納入文獻(xiàn)異質(zhì)性較大,通過閾值效應(yīng)及非閾值效應(yīng)對其異質(zhì)性來源進(jìn)行探討。結(jié)果:1、篩選文獻(xiàn):初步檢索出1332篇,最后納入22篇,其中英文21篇,中文1篇。2、文獻(xiàn)質(zhì)量評價:A級文獻(xiàn)4篇,B級文獻(xiàn)18篇。3、F0 VS F1-F4病例組、F0-F1 VS F2-F4病例組、F0-F2 VS F3-F4病例組、F0-F3 VS F4病例組敏感度、特異度異質(zhì)性均具有較大異質(zhì)性。4、(1)采用混合雙變量模型進(jìn)行合并統(tǒng)計(jì)量。F0 VS F1-F4病例組:SEN合并、SPE合并、+LR合并、-LR合并、DOR、AUROC值分別為88.8%(85.0-91.7)、95.9%(91.5-98.0)、21.435(10.215-44.979)、0.117(0.086-0.159),183.187(72.533-462.650)、0.96(0.94-0.98)。F0-F1 VS F2-F4病例組:SEN合并、SPE合并、+LR合并、-LR合并、DOR、AUROC分別為93.3%(89.2%-35.9%)、94.1%(90.2%-96.5%)、15.839(9.344-26.848)、0.072(0.044-0.117)、221.224(100.980-484.648)、0.98(0.96-0.99);F0-F2 VS F3-F4病例組:SEN合并、SPE合并、+LR合并、-LR合并、DOR、AUROC分別為92.9%(88.9-95.5)、94.6%(91.2-96.8)、17.348(10.496-28.671)、0.075(0.048-0.119)、230.434(111.482-476.317)、0.98(0.96-0.99)。F0-F3 VS F4病例組:SEN合并、SPE合并、+LR合并、-LR合并、DOR、AUROC分別為97.7%(93.0-99.3)、93.2%(90.3-95.2)、14.337(9.910-20.742)、0.025(0.008-0.075)、580.405(144.871-2325.307)、0.98(0.96-0.99);(2)Fagan分析結(jié)果:F0 VS F1-F4病例組:患病率為69.8%,陰性驗(yàn)后概率21%,陽性驗(yàn)后概率98%。假設(shè)驗(yàn)驗(yàn)前概率25%、50%、75%時,其陰性驗(yàn)后概率4%、10%、26%,陽性驗(yàn)后概率88%、96%、98%。F0-F1 VS F2-F4病例組:患病率58.5%,陰性驗(yàn)后概率9%,陽性驗(yàn)后概率96%,假設(shè)驗(yàn)驗(yàn)前概率25%、50%、75%時,其陰性驗(yàn)后概率2%、7%、18%,陽性驗(yàn)后概率84%、94%、98%。F0-F2 VS F3-F4病例組:患病率41.6%,其陰性驗(yàn)后概率1%,陽性驗(yàn)后概率84%。假設(shè)驗(yàn)驗(yàn)前概率25、50%、75%時,其陰性驗(yàn)后概率2%、7%、18%,陽性驗(yàn)后概率85%、95%、98%。F0-F3 VS F4組病例組:患病率26.3%,其陰性驗(yàn)后概率1%,陽性驗(yàn)后概率84%。假設(shè)驗(yàn)驗(yàn)前概率25%、50%、75%時,其陰性驗(yàn)后概率1%、2%、7%,陽性驗(yàn)后概率83%、93%、98%。(3)異質(zhì)性分析結(jié)果:本Meta分析異質(zhì)性主要來源于非閾值效應(yīng),不存在閾值效應(yīng)引起的異質(zhì)性。結(jié)論:磁共振彈性成像(MRE)作為一種新型的、無創(chuàng)的影像檢查手段,對肝纖維化各級均具有較高診斷價值,并具有較好的臨床適用性,可以為臨床精準(zhǔn)治療提供可靠參考。
[Abstract]:Objective: to take Meta analysis of magnetic resonance elastography (MRE) on liver fibrosis diagnostic superiority and clinical applicable value, provide an effective basis for clinical treatment of hepatic fibrosis. Materials and methods: 1. Screening Literature: search domestic before February 2, 2017, magnetic resonance elastography for liver fibrosis staging and published Chinese English included in the literature. The database includes Wanfang database (WANFANG DATA), Chinese biomedical literature database (CBM), China HowNet (CNKI), VIP information network (VIP), EMBASE, PUBMED, Cochrane Library, Web of Science, supplemented by manual search of relevant literature, according to the inclusion and exclusion criteria for screening.2, the quality of literature the evaluation of diagnostic Meta: Based on the analysis of special literature quality assessment tool QUADAS-2 (the Quality Assessment for Studies of Diagnostic Accuracy) for quality evaluation of literature and literature, A, B The classification of.3, in the literature for extracting the required data, and data processing. (1) data, extract the required documents made form. (2) to test for heterogeneity of the statistical data, the calculation of I2 value to determine the heterogeneity level. No significant heterogeneity: I250%, heterogeneity: I2 = 50%. according to the heterogeneity test results using the corresponding model combined with statistics. (3) the application of stata12.0 software F0 VS F1-F4 respectively in the case group, F0-F1 VS F2-F4 F0-F2 VS case group, F3-F4 group, F0-F3 VS F4 group (Sensitivity, SEN) sensitivity, specificity (Specificity, SPE), positive likelihood ratio (PosititiveLikelihood Ratio, +LR), negative likelihood ratio (Negative Likelihood Ratio, -LR), diagnostic odds ratio (Diagnostic Odds Ratio, DOR) were combined and heterogeneity test, drawing layered integrated receiver operating characteristic curve (Hierarchical Summary Receiver Operating Characteris Tic, HROC), and calculate the area under the curve (the Area Under the Receiver Operating Characteristic Curve, AUROC), to evaluate the value of magnetic resonance imaging of liver fibrosis elastic grading diagnosis. (4) the posterior probability was calculated in the study group, and assuming that the prior probability was 25%, 50%, 75% and calculate the posterior probability, through Fagan Plot analysis to evaluate the MRE grading of liver fibrosis diagnosis and clinical applicability. (5) if the literature heterogeneity, the heterogeneity of the sources are discussed through the threshold effect and non effect. Results: 1 screening Literature: preliminary selected 1332 articles, 22 articles were included at last among them, English 21, Chinese 1.2, quality evaluation of literature: a literature 4, B grade 18 references.3, F0 VS F1-F4 F0-F1 VS F2-F4 case group, the case group, F0-F2 VS F3-F4 F0-F3 VS F4 case group, the case group sensitivity, specificity and heterogeneity has great heterogeneity .4, (1) the mixed bivariate model with statistic.F0 VS F1-F4 in the case group: SEN merger, SPE merger, +LR merger, merged with -LR, DOR, AUROC values were 88.8% (85.0-91.7), 95.9% (91.5-98.0), 21.435 (10.215-44.979), 0.117 (0.086-0.159), 183.187 (72.533-462.650), 0.96 (0.94-0.98).F0-F1 VS F2-F4 case group: SEN SPE +LR merger, merger, merger, merged with -LR, DOR, AUROC were 93.3% (89.2%-35.9%), 94.1% (90.2%-96.5%), 15.839 (9.344-26.848), 0.072 (0.044-0.117), 221.224 (100.980-484.648), 0.98 (0.96-0.99); F0-F2 VS F3-F4 case group: SEN merger with SPE, +LR and -LR, with DOR, AUROC, respectively, 92.9% (88.9-95.5), 94.6% (91.2-96.8), 17.348 (10.496-28.671), 0.075 (0.048-0.119), 230.434 (111.482-476.317), 0.98 (0.96-0.99).F0-F3 VS F4 case group: SEN merger, SPE merger, +LR merger, -LR merger, DOR, AUROC were 97.7% (93.0-99.3), 93 .2%(90.3-95.2),14.337(9.910-20.742),0.025(0.008-0.075),580.405(144.871-2325.307),0.98(0.96-0.99);(2)Fagan鍒嗘瀽緇撴灉:F0 VS F1-F4鐥呬緥緇，

本文編號：1559527