MRI两种表观弥散系数值对子宫内膜良恶性病变的诊断价值

Diagnostic value of two apparent diffusion coefficient measurement methods for benign and malignant endometrial lesions

目的探讨MRI两种表观弥散系数(ADC)值对子宫内膜良恶性病变的诊断价值及其一致性。方法回顾性分析79例子宫内膜病变的MRI资料,以病理结果为金标准,分别测量并比较良恶性病变MRI整体ADC值(ADCw)与局部ADC值(ADCr)的差异,并以受试者工作特征(ROC)曲线评估两种方法的诊断效能,以Bland-Altman法评估两者的一致性。结果子宫内膜良性病变的ADCw、ADCr值均高于恶性病变(均P<0.05);ADCw值取1.046×10^(-3)mm^2/s、ADCr值取1.053×10^(-3)mm^2/s作为鉴别子宫内膜良恶性病变的诊断阈值时,其ROC曲线下面积分别为0.947和0.911,ADCw ROC曲线下面积大于ADCr的曲线下面积(P<0.05);Bland-Altman分析显示:两者差值尚有7.9%(7/79)在95%一致性范围外。结论 ADCw和ADCr值均有助于子宫内膜良恶性病变的定性诊断,但两种方法的一致性较差,尚不能相互替代,ADCw的诊断效能更优。

Objective To investigate the diagnostic value and consistency of two apparent diffusion coefficient(ADC)measurement methods for benign and malignant endometrial lesions.Methods The MRI data of 79 cases of endometrial lesions were analyzed retrospectively.The pathological result was taken as golden standard,and the MRI ADC values by Whole Measurement Method(ADCw)and Regional Measurement Method(ADCr)were determined respectively and compared between benign and malignant lesions.Receiver operating characteristic(ROC)curve was used to assess the diagnostic efficiencies of the two methods,and Bland-Altman method to assess the consistency of the two methods.Results The ADCw and ADCr values of benign endometrial lesions were higher than those of malignant lesions(all P<0.05);with ADCw value of 1.046×10-3 mm 2/s and ADCr value of 1.053×10-3 mm 2/s as the thresholds for differential diagnosis of benign and malignant endometrial lesions,the areas under ROC curve were 0.947 and 0.911,respectively,and the area under ROC curve of ADCw was greater than that of ADCr(P<0.05).Bland-Altman analysis showed that the cases with the difference between ADCw and ADCr beyond the 95%consistency range accounted for 7.9%(7/79).Conclusion ADCw and ADCr values are conducive to qualitative diagnosis of benign and malignant endometrial lesions,having a poor consistency of the two methods,and can not replace each other yet.ADCw achieves better diagnostic efficiency.