文献速递乳腺MR

扩散MRI鉴别乳腺病变诊断效能的Meta分析

A Meta-analysis of the Diagnostic Performance of Diffusion MRI for Breast Lesion Characterization.
PubMed 文献 · Radiology · 2019-04-23

该Meta分析比较了扩散加权成像(DWI)、扩散张量成像(DTI)和体素内不相干运动(IVIM)在鉴别乳腺良恶性病变中的诊断效能,纳入73项研究共6791个病灶(3930个恶性,2861个良性)。ADC的汇总敏感度、特异度和AUC分别为89%、82%和0.92;DTI的λ1分别为93%、90%和0.94;IVIM的D分别为88%、79%和0.90。三种技术诊断准确性相当,IVIM与ADC可比,DTI有潜力但研究数量有限。

要点

  • ADC鉴别乳腺良恶性病变的汇总敏感度89%、特异度82%、AUC 0.92。
  • DTI的λ1诊断效能最高,汇总敏感度93%、特异度90%、AUC 0.94,但研究数量有限。
  • IVIM的D汇总敏感度88%、特异度79%、AUC 0.90,与ADC可比。

核心结论

  • DWI、DTI和IVIM在鉴别乳腺良恶性病变方面具有相当的诊断准确性,敏感度和特异度均较高。
  • IVIM与ADC的诊断效能可比。
  • DTI可能具有应用前景,但迄今为止相关研究数量有限。

研究设计与样本

  • 研究类型为Meta分析,检索PubMed和Embase数据库2018年1月至3月发表的英文文献。
  • 共纳入73项符合纳入和排除标准的研究,包含6791个病灶(3930个恶性,2861个良性)。
  • 采用双变量模型获得各参数汇总敏感度、特异度和AUC,并评估发表偏倚和研究间异质性。
  • 亚组分析采用Student t检验或单因素方差分析评估MRI参数对诊断效能的影响。

诊断标准与关键数值

  • ADC的汇总敏感度89%、特异度82%、AUC 0.92。
  • DTI中诊断效能最高的参数为λ1,汇总敏感度93%、特异度90%、AUC 0.94。
  • IVIM中诊断效能最高的参数为D,汇总敏感度88%、特异度79%、AUC 0.90。
  • 除灌注分数(f)外,所有参数均存在显著异质性(P < .05)。
  • 评估ADC的研究存在发表偏倚。

局限性

  • 评估ADC的研究存在发表偏倚。
  • 除灌注分数(f)外,所有参数均存在显著异质性(P < .05)。
  • DTI可能具有应用前景,但迄今为止相关研究数量有限。

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

英文原文(摘要)· 点击展开

Background Various techniques are available to assess diffusion properties of breast lesions as a marker of malignancy at MRI. The diagnostic performance of these diffusion markers has not been comprehensively assessed. Purpose To compare by meta-analysis the diagnostic performance of parameters from diffusion-weighted imaging (DWI), diffusion-tensor imaging (DTI), and intravoxel incoherent motion (IVIM) in the differential diagnosis of malignant and benign breast lesions. Materials and Methods PubMed and Embase databases were searched from January to March 2018 for studies in English that assessed the diagnostic performance of DWI, DTI, and IVIM in the breast. Studies were reviewed according to eligibility and exclusion criteria. Publication bias and heterogeneity between studies were assessed. Pooled summary estimates for sensitivity, specificity, and area under the curve were obtained for each parameter by using a bivariate model. A subanalysis investigated the effect of MRI parameters on diagnostic performance by using a Student t test or a one-way analysis of variance. Results From 73 eligible studies, 6791 lesions (3930 malignant and 2861 benign) were included. Publication bias was evident for studies that evaluated apparent diffusion coefficient (ADC). Significant heterogeneity (P < .05) was present for all parameters except the perfusion fraction (f). The pooled sensitivity, specificity, and area under the curve for ADC was 89%, 82%, and 0.92, respectively. The highest performing parameter for DTI was the prime diffusion coefficient (λ1), and pooled sensitivity, specificity, and area under the curve was 93%, 90%, and 0.94, respectively. The highest performing parameter for IVIM was tissue diffusivity (D), and the pooled sensitivity, specificity, and area under the curve was 88%, 79%, and 0.90. Choice of MRI parameters had no significant effect on diagnostic performance. Conclusion Diffusion-weighted imaging, diffusion-tensor imaging, and intravoxel incoherent motion have comparable diagnostic accuracy with high sensitivity and specificity. Intravoxel incoherent motion is comparable to apparent diffusion coefficient. Diffusion-tensor imaging is potentially promising but to date the number of studies is limited. © RSNA, 2019 Online supplemental material is available for this article.

来源:Radiology,查看原文。版权归原作者所有。

引用

  1. Baxter GC, Graves MJ, Gilbert FJ, et al. A Meta-analysis of the Diagnostic Performance of Diffusion MRI for Breast Lesion Characterization. Radiology. 2019. PMID: 31012817. doi:10.1148/radiol.2019182510

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。