对比弥散张量成像与弥散加权成像评估子宫内膜样腺癌病理分级的价值

Comparison on the value of diffusion tensor imaging and diffusion weighted imaging for predicting histopathological classification of endometrial adenocarcinoma

目的对比弥散张量成像(DTI)与弥散加权成像(DWI)评估子宫内膜样腺癌病理分级的效能。方法回顾性分析41例经病理证实的子宫内膜样腺癌患者,其中高分化(G1级)15例、中分化(G2级)14例、低分化(G3级)12例,术前均接受常规MR平扫及DTI、DWI,比较各级子宫内膜样腺癌的表观弥散系数(ADC)、平均弥散系数(MD)及各向异性分数(FA)值,分析各参数对病理组织学分级的诊断效能。结果 G1、G2、G3级子宫内膜样腺癌之间ADC值及MD值差异均有统计学意义(P均<0.001),FA值差异无统计学意义(P>0.05)。子宫内膜样腺癌ADC值(r=-0.589,P<0.001)、MD值(r=-0.724,P<0.001)均与其组织学分级呈负相关。将G3级归为高危组、G1和G2级归为低危组,高危组ADC值、MD值均低于低危组(P均<0.01)。受试者工作特征(ROC)曲线结果显示,ADC、MD值鉴别诊断高、低危子宫内膜样腺癌的曲线下面积(AUC)分别为0.792和0.868(P均<0.01)。结论 DTI和DWI均可用于评估子宫内膜样腺癌病理分级,且MD值效能较ADC值更高。

Objective To compare the efficacy of diffusion tensor imaging(DTI) and diffusion weighted imaging(DWI) for predicting histopathologic classification of endometrioid adenocarcinoma. Methods Data of 41 patients with pathology confirmed endometrioid adenocarcinoma were retrospectively analyzed. Among them, 15 cases were well differentiated(G1 grade), 14 were moderately differentiated(G2 grade) and 12 were poorly differentiated(G3 grade). All patients underwent routine MR scan, DTI and DWI before surgical operation. The apparent diffusion coefficient(ADC), the mean diffusion coefficient(MD) and fractional anisotropy(FA) were compared among different grades of endometrioid adenocarcinoma, and the diagnostic efficiency of each parameter for histopathological grading was evaluated. Results There were significant differences of ADC value and MD value among G1, G2 and G3 endometrioid adenocarcinoma(all P<0.001), but no statistically significant difference of FA value was found(P>0.05). The ADC value(r=-0.589, P<0.001) and MD value(r=-0.724, P<0.001) were both negatively correlated with histological grade. G3 was classified as high-risk group, while G1 and G2 were classified as low-risk group. The ADC and MD values of high-risk group were lower than those of low risk group(both P<0.01). Receiver operating characteristic(ROC) curves showed that the area under the curve(AUC) of ADC and MD values in differential diagnosis of high-risk and low-risk endometrioid adenocarcinoma was 0.792 and 0.868(both P<0.01), respectively. Conclusion Both DTI and DWI could be used to predict the histological classification of endometrial adenocarcinoma, and MD value was more effective than ADC value.