Maltez de Almeida João Ricardo, Gomes André Boechat, Barros Thomas Pitangueira, Fahel Paulo Eduardo, de Seixas Rocha Mário
1 Department of Diagnostic Imaging, Clínica de Assistência à Mulher, Grupo CAM, Avenida ACM, 237, Ed. Prof. Carlos Aristides Maltez, Itaigara, Salvador, Bahia, Brazil 41825-000.
2 Department of Biomedicine, Bahiana School of Medicine and Public Health, Campus Brotas, Salvador, Bahia, Brazil.
AJR Am J Roentgenol. 2015 Jul;205(1):222-31. doi: 10.2214/AJR.14.13834.
The purposes of this study were to investigate whether dynamic contrast-enhanced MRI is adequate for subcategorization of suspicious lesions (BI-RADS category 4) and to evaluate whether use of DWI improves diagnostic performance.
The study group was composed of 103 suspicious lesions found in 83 subjects. Patient ages and lesion sizes were compiled, and two radiologists reanalyzed the images; subcategorized the findings as BI-RADS 4A, 4B, or 4C; and calculated apparent diffusion coefficient (ADC) values. The stratified variables were tested by univariate analysis and inserted in two multivariate predictive models, which were used to generate ROC curves and compare AUCs. Positive predictive values (PPVs) for each subcategory and ADC level were calculated, and interobserver agreement was tested.
Forty-four (42.7%) suspicious findings proved malignant. Except for age (p = 0.08), all stratified predictor variables were significant in univariate analyses (p < 0.01). Logistic regression models did not differ substantially after comparison of the ROC curves (p = 0.09), but the one including ADC values was slightly better: AUC of 0.89 (95% CI, 0.82-0.95) against AUC of 0.85 (95% CI, 0.78-0.93). PPV increased progressively in each BI-RADS 4 subcategory (4A, 0.15; 4B, 0.37; 4C, 0.84). ADC values of 1.10 × 10(-3) mm(2)/s or less had the second highest PPV (0.77). Interobserver agreement was substantial at a kappa value of 0.80 (95% CI, 0.70-0.90; p < 0.01).
Risk stratification of suspicious lesions (BI-RADS category 4) can be satisfactorily performed with DCE-MRI and slightly improved when DWI is introduced.
本研究旨在调查动态对比增强磁共振成像（DCE-MRI）是否足以对可疑病变（乳腺影像报告和数据系统[BI-RADS]4类）进行亚分类，并评估使用扩散加权成像（DWI）是否能提高诊断性能。
研究组由83名受试者中发现的103个可疑病变组成。收集患者年龄和病变大小，两名放射科医生重新分析图像；将结果分类为BI-RADS 4A、4B或4C；并计算表观扩散系数（ADC）值。通过单因素分析对分层变量进行检验，并将其纳入两个多变量预测模型，用于生成ROC曲线并比较曲线下面积（AUC）。计算每个亚类和ADC水平的阳性预测值（PPV），并检验观察者间的一致性。
44个（42.7%）可疑发现被证实为恶性。除年龄外（p = 0.08），所有分层预测变量在单因素分析中均具有显著性（p < 0.01）。比较ROC曲线后，逻辑回归模型无显著差异（p = 0.09），但包含ADC值的模型略好：AUC为0.89（95%可信区间，0.82 - 0.95），而另一个模型的AUC为0.85（95%可信区间，0.78 - 0.93）。每个BI-RADS 4亚类的PPV逐渐增加（4A，0.15；4B，0.37；4C，0.84）。ADC值为1.10×10⁻³mm²/s或更低时，PPV第二高（0.77）。观察者间一致性较高，kappa值为0.80（95%可信区间，0.70 - 0.90；p < 0.01）。
使用DCE-MRI可以令人满意地对可疑病变（BI-RADS 4类）进行风险分层，引入DWI后可略有改善。
