MRI和超声对产前胎盘植入诊断价值的Meta分析

Meta-analysis of the value of MRI and ultrasonography in the diagnosis of prenatal placental implantation

目的 通过Meta分析对MRI和超声对产前胎盘植入的诊断价值进行系统评价.方法 检索PubMed、Web of Science、EMBASE、Cochrane图书馆、中国生物医学文献数据库、中国知网、维普、万方数据库中建库至2018年7月8日公开发表的有关MRI和超声对产前胎盘植入诊断研究的中、英文文献.对纳入文献进行数据提取和质量评估后,采用Stata15.1统计软件进行随机效应模型的Meta分析.计算所纳入研究的合并敏感度、合并特异度、合并阳性似然比、合并阴性似然比、合并诊断比值比以及相对应的95%可信区间(CI),绘制森林图和合并受试者操作特征(SROC)曲线,并计算SROC曲线下面积(AUC).采用I2评估研究之间的统计学异质性.结果 共纳入30项研究,总样本病例数2266例.MRI对产前胎盘植入诊断的合并敏感度、合并特异度、合并阳性似然比、合并阴性似然比、合并诊断比值比、AUC分别为0.83(95%CI:0.79~0.87)、0.90(95%CI:0.84~0.94)、8.6(95%CI:5.2~14.2)、0.18(95%CI:0.14~0.24)、47(95%CI:24~91)、0.91(0.88~0.93).US的上述相应的诊断指标值分别为0.81(95%CI:0.76~0.85)、0.88(95%CI:0.82~0.92)、6.6(95%CI:4.5~9.8)、0.22(95%CI:0.17~0.28)、31(95%CI:18~52)、0.90(0.87~0.92).MRI和US诊断的敏感度和特异度均有较高异质性(I2均>50%,P均<0.001).Deek漏斗图提示存在发表偏倚(MRI,P=0.021;US,P=0.049).结论 MRI对产前胎盘植入的诊断价值略高于超声,是其临床评估的一种有效且无创的检查方法.

Objective To evaluate the diagnostic value of MRI and Ultrasonography(US)in prenatal placenta implantation using meta-analysis.Methods The PubMed,Web of Science,EMBASE,Cochrane library,Chinese Biomedicine Literature Database,China National Knowledge Infrastructure,Weipu Chinese Science and Technology Periodicals Database,and Wanfang Digitalized Journal Data base were searched for the publications in Chinese and English on the use of MRI and US for diagnosing prenatal placenta implantation for the period up to July 82018.After data extraction and quality assessment of the included literatures,statistical software Stata 15.1 was used for a random-effects Meta-analysis.The combined sensitivity,combined specificity,combined positive likelihood ratio,combined negative likelihood ratio,combined diagnostic odds ratio and the corresponding 95%confidence interval(CI)were calculated.Forest plots and combined reciever operating characteristics(SROC)curves were plotted and the area under the SROC curve(AUC)was calculated.I2 was used to assess statistical heterogeneity between studies.Results A total of 30 studies were included,with a total sample number of 2,266 cases.The combined sensitivity of MRI for the diagnosis of prenatal placental implantation was 0.83(95%CI:0.79-0.87),combined specificity was 0.90(95%CI:0.84-0.94),combined positive likelihood ratio(PLR)was 8.6(95%CI:5.2-14.2),combined negative likelihood ratio(NLR)was 0.18(95%CI:0.14-0.24),and combined diagnostic odds ratio(DOR)was 47(95%CI:24-91),AUC was 0.91(0.88-0.93).The combined US sensitivity for the diagnosis of prenatal placental implantation was 0.81(95%CI:0.76-0.85),combined specificity was 0.88(95%CI:0.82-0.92),combined PLR was 6.6(95%CI:4.5-9.8),combined NLR was 0.22(95%CI:0.17-0.28),and combined DOR was 31(95%CI:18-52),AUC was 0.90(0.87-0.92).The sensitivity and specificity of MRI and US diagnosis were significantly heterogeneous(all I2>50%,P<0.001).Deek funnel plots suggested a potentially publication bias(MRI bias P=0.021,US bias P=0.049).Conclusion MRI is slightly more valuable than US in the diagnosis of prenatal placenta implantation,and is an effective and non-invasive method for clinical evaluation.