多参数MRI对前列腺癌诊断价值的Meta分析

Multi-parameter Magnetic Resonance Imaging in the Diagnosis of Prostate Cancer: A Meta-analysis

目的通过Meta分析探讨多参数磁共振成像(mp-MRI)对前列腺癌(PCa)的诊断价值。资料与方法利用计算机检索2013年1月1日至2017年9月31日Pub Med、Ebsco、Embase、Ovid、中国生物医学文献数据库、维普、万方和中国知网等数据库收录的mp-MRI诊断PCa的相关文献。由2位研究人员按照纳入与排除标准独立筛选文献并提取资料。采用QUADAS-2工具评价纳入研究的方法学质量,计算合并敏感度(Sen)、特异度(Spe)、阳性似然比(+LR)、阴性似然比(-LR)以及诊断比值比(DOR),绘制汇总受试者工作特性曲线并计算曲线下面积(AUC)。结果最终纳入20篇文献,共3334个病灶。Meta分析结果显示,mp-MRI诊断PCa的Sen合并=0.89(95%CI 0.87～0.92)、Spe合并=0.76(95%CI 0.67～0.84)、+LR合并=3.8(95%CI 2.7～5.4)、-LR合并=0.14(95%CI 0.11～0.17)、DOR合并=27(95%CI 17～44),AUC合并=0.92(95%CI 0.89～0.94)。结论 mpMRI对PCa具有较高的诊断价值,可作为诊断PCa的一种重要方法。

Purpose Meta-analysis of diagnostic value of prostate cancer（PCa） based on multi-parameter magnetic resonance imaging（mp-MRI）. Materials and Methods Computer-based search of Pub Med, Ebsco, Embase, Ovid, CBM, VPCS, Wanfang, and CNKI databases for related literatures on PCa diagnosed via mp-MRI from January 1, 2013 to September 31, 2017. Two researchers independently screened out literatures, extracted data based on inclusion and exclusion criteria, applied QUADAS-2 to evaluate the quality of methodologies included for the study, calculated combined sensitivity（Sen）, specificity（Spe）, positive likelihood ratio（＋LR）, negative likelihood ratio（-LR）, and diagnostic odds ratio（DOR）, drew Summerized receiver operating characteristic（SROC） curve and calculated the area under the curve（AUC）. Results Eventually 20 articles of literatures were included, totally 3334 lesions. Meta-analysis results showed that Senmerge of PCa diagnosed via mp-MRI was 0.89（95% CI 0.87-0.92）, Spemerge 0.76（95% CI 0.67-0.84）, ＋LRmerge 3.8（95% CI 2.7-5.4）,-LRmerge=0.14（95% CI 0.11-0.17）, DORmerge=27（95% CI 17-44）, AUCmerge=0.92（95% CI 0.89-0.94）. Conclusion mp-MRI has a high diagnostic value for PCa and can be used as an important method for PCa diagnosis.