动态增强MRI联合集合序列对乳腺良恶性病变的诊断价值

Diagnostic Value of Dynamic Contrast Enhanced MRI Combined with Compilation Sequence in Differential Diagnosis between Benign and Malignant Breast Lesions

目的探讨动态增强磁共振成像(DCE-MRI)和磁共振成像集合(MAGiC)序列对乳腺良恶性病变的鉴别诊断价值。方法回顾性分析经病理证实的86例乳腺病变患者,共87个病灶,其中,良性病变32个,恶性病变55个。患者均行常规乳腺MRI平扫和动态增强扫描,并于增强前和注射对比剂后分别采集MAGiC序列图像。分析病灶的形态、边界、强化方式、早期强化率(EER)和时间-信号强度曲线(TIC),根据Fischer评分标准及乳腺影像报告与数据系统(BI-RADS)2013版进行评分和分类。使用GE ADW 4.7后处理工作站对MAGiC序列原始图像进行处理,测量病灶增强前后的T1、T2和质子密度(PD)值,分别用T_(1pre)、T_(2pre)、PD_(pre)和T1post、T2post、PDpost表示,计算各值变化量和变化率{以T1值为例,△T1=T_(1pre)-T_(1post),△T_(1%)=[(T_(1pre)-T_(1post))/T_(1pre)]×100%}。分类变量采用卡方检验或R×C列表检验,连续变量采用Mann-Whitney U检验。采用上述具有统计学意义的指标建立单独应用DCE-MRI、MAGiC序列及联合应用DCE-MRI与MAGiC序列的Logistic回归模型,并采用受试者工作特征(ROC)曲线评价其对良恶性病变的鉴别诊断价值。结果良恶性组间病灶的形态、边界、强化方式和TIC类型差异具有统计学意义(P<0.05);除T_(2post)和PD_(pre)外,其他定量参数在两组间差异均有统计学意义(P<0.05);DCE-MRI联合MAGiC序列的Logistic回归模型具有最高的曲线下面积(AUC),其准确率、敏感度和特异度分别为90.8%、94.5%和84.4%。结论MAGiC序列作为一种辅助检查方法能提高DCE-MRI检查对乳腺良恶性病变的鉴别诊断效能。

Objective To evaluate the differential diagnostic value of dynamic contrast enhanced MRI(DCE-MRI)combined with magnetic resonance image compilation(MAGiC)sequence in benign and malignant breast lesions.Methods A retrospective study performed in 86 patients with pathologically confirmed malignant tumors(n=55)and benign lesions(n=32).All patients underwent the routine MR plain scan sequences and dynamic contrast enhancement scan sequences,and axial compilation sequences for T1,T2 and PD mapping were added before and after enhancement.The MRI features of the lesions including shape,margin,pattern of enhancement,early enhancement rate(EER)and the type of time intensity curve(TIC)were retrospectively scored and graded according to Breast Imaging Reporting and Data System(BI-RADS 2013 ed).The original MAGiC image was processed by GE ADW 4.7 post-processing workstation.The quantitative values were measured before and after the administration of the contrast medium(recored as T_(1pre),T2 pre,PDpre and T1 post,T2 post,PDpost respectively),and the parameter variation values and its change ratio were calculated as follows:take T1 value for example,△T1=T_(1pre)-T1 post,△T1%=[(T_(1pre)-T1 post)/T_(1pre)]×100%.The classified variables were compared using the chi-square or the R×C list test.The continuous variables of two groups were compared using Mann-Whitney U test.Using the above indexes with statistically differences,the logistic regression model was established to constructed the newly combined parameters and evaluate the diagnostic performance of DCE-MRI,MAGiC sequence and DCE-MRI plus MAGiC sequence.Receiver operating characteristic curve was used to analyze and compare the ability of these models in differentiation of benign and malignant breast lesions.Results There were significant differences of shape,margin,pattern of enhancement and the type of TIC between the two groups(P<0.05).Except T2 post and PDpre,other quantitative parameters have significant statistical differences between benign and malignant breast lesions(P<0.05).The Logistic regression model of combined DCE-MRI+MAGiC sequence had the highest AUC(0.953),theaccuracy,specificity and sensitivity was 90.8%,94.5%and 84.4%,respectively.Conclusion MAGiC sequence can significantly improve the diagnostic efficacy of DCE-MRI in the differential diagnosis of benign and malignant breast lesions.