定量DCE-MRI对乳腺非肿块型强化病变良恶性的预测价值

Differentiation between benign and malignant non-mass enhancement lesions using volumetric quantitative dynamic contrast-enhanced MR imaging

目的:探讨定量动态对比增强MRI(qDCE-MRI)对乳腺非肿块型强化病变(NME)良恶性的预测价值。方法:回顾性收集83例行乳腺常规MR增强及DCE-MRI检查、术后病理资料完整的乳腺NME病变患者,对病变感兴趣区(ROI)进行三维分割,获得qDCE-MRI定量及半定量参数。定量参数包括容量转移常数(Ktrans)、速率常数(Kep)、血管外细胞外间隙容积比(Ve)、血管(血浆)间隙容积分数(Vp);半定量参数包括达峰时间(TTP)、最大浓度(Max Con)、浓度-时间曲线下面积(AUC)、最大斜率(Max Slope)。采用独立样本t检验或Mann-Whitney U检验分析NME良恶性病变间的统计学差异;采用单因素回归分析各参数的独立诊断效能,并用多因素回归分析最终诊断方程;绘制ROC曲线评估qDCE-MRI参数对NME病变良恶性的鉴别诊断效能。结果:NME病变良、恶性组的Ktrans值分别为(0.0563±0.042)和(0.1092±0.0550)min-1,两者差异具有统计学意义(P<0.001);Kep值分别为(0.4334±0.2516)和(0.6863±0.3039)min-1,两者差异具有统计学意义(P=0.001);Vp值分别为(0.0046±0.0063)和(0.0106±0.0092),两者差异具有统计学意义(P=0.012);TTP值分别为(4.4499±0.7441)和(3.7880±0.7571)min,两者差异具有统计学意义(P=0.001);Max Slope值分别为(0.2021±0.1084)和(0.3067±0.1178),两者差异具有统计学意义(P<0.001)。单因素回归分析得出Ktrans、Kep、Vp、TTP、Max Slope对于NME良恶性均具有独立诊断效能,其中以Ktrans的诊断效能最高,AUC为0.790(95%CI:0.675~0.906,P<0.05);多因素回归分析后,Ktrans及TTP进入最终方程,最终方程对于NME良恶性鉴别的AUC为0.831(95%CI:0.727~0.935,P<0.05)。结论:qDCE-MRI对乳腺NME病变良恶性具有一定的鉴别诊断价值。

Objective:The purpose of this study was to determine the diagnostic performance of volumetric quantitative dynamic contrast-enhanced MRI(qDCE-MRI)in the differentiation between malignant and benign non-mass enhancement(NME)lesions.Methods:The clinicopathological data of 82 NME patients were retrospectively analyzed.All patients underwent MRI enhanced and DCE-MRI scans before treatment.3 D-ROI was drawn on the NME lesions.Quantitative pharmacokinetic parameters Ktrans,Kep,Ve,Vp semi-quantitative parameters TTP,Max Con,Max Slope,AUC were obtained by using a two-compartment extended Tofts model.Comparisons between groups were respectively analyzed using the t test or Mann-Whitney U test.Logistic regression analysis was used to determine predictors of malignancy,followed by receiver operating characteristics(ROC)analysis to evaluate the diagnostic performance.Results:Ktrans values were(0.0563±0.042)min-1 and(0.1092±0.0550)min-1 in malignant and benign NME groups,respectively,with statistically significant difference(P<0.001).Kep values were(0.4334±0.2516)min-1 and(0.6863±0.3039)min-1,respectively,with statistically significant difference(P=0.001).Vp values were(0.0046±0.0063)and(0.0106±0.0092),respectively,with statistically significant difference(P=0.012).TTP values were(4.4499±0.7441)min and(3.7880±0.7571)min,respectively,with statistically significant difference(P=0.001).The Max Slope values were(0.2021±0.1084)and(0.3067±0.1178),respectively,with statistically significant difference(P<0.001).Ktrans provided the highest accuracy of 0.790(95%CI:0.675~0.906,P<0.05).Univariate regression analysis showed that Ktrans,Kep,Vp,TTP and Max Slope were independent predictors and Ktrans has the highest diagnostic performance.Multivariate logistic regression analyses showed that Ktrans and TTP were independent predictors of malignancy and the area under ROC curve was 0.831(95%CI:0.717~0.935,P<0.05).Conclusion:qDCE-MRI has a certain value in differential diagnosis of breast non-mass enhancement lesions.