超声造影对子宫内膜病变的诊断价值

Diagnostic values of contrast-enhanced ultrasonography in the diagnosis of endometrial lesions

目的探讨不同类型子宫内膜病变超声造影(CEUS)图像特征及定量分析结果,以及CEUS诊断子宫内膜病变的价值。方法选择2016年6月1日至2018年1月31日,于四川省中西医结合医院妇科诊断为疑似子宫内膜病变的62例年龄为27～66岁女性受试者为研究对象。对其进行经阴道CEUS检查,并通过手术或诊刮术获取子宫内膜组织标本进行活组织病理学检查。根据活组织病理学检查结果,将其分为恶性组(n=10,子宫内膜癌患者),良性组(n=41,子宫内膜单纯性增生或子宫内膜息肉患者),对照组(n=11,无子宫内膜病变者)。分析62例受试者的CEUS图像特征。CEUS采用与子宫内膜病变区相同深度的正常子宫肌层为参照,通过肘静脉注射造影剂后,运用时间-强度曲线(TIC)测定造影剂开始增强时间、达峰时间、峰值强度、梯度[(造影剂峰值强度-造影剂基线强度)/造影剂强度上升时间],以及TIC曲线下面积(TIC-AUC)。3组受试者造影剂开始增强时间比较,采用单因素方差分析;峰值强度、达峰时间、梯度及TIC-AUC比较,采用Kruskal-Wallis H检验。绘制子宫内膜的峰值强度、TIC-AUC预测子宫内膜癌发生的受试者工作特征(ROC)曲线,计算ROC曲线下面积(ROC-AUC),根据约登指数最大原则,确定这2个指标预测子宫内膜癌发生的最佳临界值,并计算其敏感度和特异度。本研究遵循的程序符合2013年修订的《世界医学协会赫尔辛基宣言》要求。本研究征得所有受试者知情同意,并在进行经阴道CEUS检查前,与受试者本人签署临床试验知情同意书。结果经肘静脉注射造影剂后,经阴道CEUS图像特征为:(1)恶性组子宫内膜造影剂增强早期,呈快速均匀或非均匀性高增强,根据子宫内膜癌组织病理学类型不同,表现为子宫内膜开始增强时间明显早于、稍早于子宫肌层者分别为7例(子宫内膜腺癌)与3例(子宫恶性中胚叶混合瘤、子宫内膜透明细胞癌),峰值强度明显高于、稍高于子宫肌层者分别为8例(子宫内膜腺癌、透明细胞癌)与2例(子宫恶性中胚叶混合瘤),消退时间均晚于子宫肌层。(2)良性组41例受试者中,9例为子宫内膜单纯性增生,其子宫内膜造影剂开始增强时间晚于子宫肌层,清退时间早于子宫肌层,子宫内膜早期增强模式主要为稍低增强;32例为子宫内膜息肉,其子宫内膜造影剂开始增强时间稍晚于子宫肌层,消退时间早于子宫肌层,子宫内膜早期增强模式呈多样化,主要表现为结节状稍高增强。(3)恶性组、良性组、对照组峰值强度分别为25.1dB(14.6～26.1dB)、10.7dB(7.0～15.5dB)、10.1dB(7.2～15.1dB),TIC-AUC分别为1 880.7(734.9～2 084.2)、418.0(240.0～718.7)、349.8(238.3～563.6)。3组峰值强度与TIC-AUC分别比较,差异均有统计学意义(H=6.732、6.732,P=0.035、0.011)。3组达峰时间、梯度、开始增强时间分别整体比较,差异均无统计学意义(P>0.05)。(4)TIC指标中,峰值强度、TIC-AUC预测子宫内膜癌发生的ROC-AUC分别为0.92(95%CI:0.78～0.97,P=0.028)和0.96(95%CI:0.84～0.99,P=0.016)。根据约登指数最大原则,峰值强度、TIC-AUC预测子宫内膜癌发生的最佳临界值分别为16.608dB、767.45,此时其预测子宫内膜癌发生的敏感度分别为81.2%、85.1%,特异度分别为85.8%、87.2%。结论在经阴道CEUS时,子宫内膜病变CEUS图像具有一定特征性,对于子宫内膜恶性病变的诊断具有临床价值。

Objective To explore the characteristics and quantitative analysis results of contrast-enhanced ultrasonography （CEUS） images of different types of endometrial lesions, and analyze the values of CEUS in the diagnosis of endometrial lesions. Methods From June 1, 2016 to January 31, 2018, a total of 62 subjects with suspected endometrial lesions in the Department of Gynecology, Sichuan Integrative Medicine Hospital were enrolled into this study, aged 27-66 years. All subjects underwent transvaginal CEUS, as well as surgery or surgical curettage to obtain tissues for histo pathology. According to the endometrial histopathological results, they were divided into malignant group （ n =10, endometrial cancer patients）, benign group （ n =41, simple endometrial hyperplasia or endometrial polyps）, and control group （ n =11, without endometrial lesions）. The characteristics of the CEUS images were analyzed. The time-intensity curve （TIC） was used to determine the initial enhancement time, peak time, peak intensity and gradient of contrast agent in endometrium, and area under TIC curve （TIC-AUC） as the normal myometrium at the same depth as the endometrial lesion was used as a reference. The initial enhancement time of contrast agent among three groups was compared by one-way ANOVA analysis. The peak intensity, peak time, gradient and TIC-AUC among three groups were compared by Kruskal-Wallis H test. The receiver operator characteristic （ROC） curves of peak intensity and TIC-AUC for predicting the incidence of endometrial cancer were built respectively, and the area under ROC curve （ROC-AUC） was calculated. The optimal critical values of those two indicators for predicting the incidence of endometrial cancer were obtained when Youden index reaching the maximum value. And their sensitivities and specificities were calculated respectively. This study met the requirements of the World Medical Association Declaration of Helsinki revised in 2013. Clinical trials informed consent was obtained from each subject before undergoing transvaginal CEUS. Results After the injection of contrast agent through cubital vein, the characteristics of CEUS images were as follows. ①In the malignant group, the transvaginal CEUS results showed a rapid uniform or non-uniform enhancement in the early stage of endometrium. According to the different pathological types of endometrial cancer, the initial enhancement time of endometrium was significantly earlier than that of the myometrium （7 cases of adenocarcinoma endometrium） or slightly earlier than that of the myometrium （3 cases of malignant mesodermal mixed tumor or clear cell carcinoma）, and the peak intensity was significantly higher than that of the myometrium （8 cases of adenocarcinoma endometrium or clear cell carcinoma） or slightly higher than that of the myometrium （2 cases of malignant mesodermal mixed tumor）, and the regression times of endometrium all were later than that of the myometrium. ②Among the 41 subjects in the benign group, 9 cases were simple endometrial hyperplasia whose transvaginal CEUS results showed that the initial enhancement time of endometrium were later than that of the myometrium, and the regression time of endometrium was earlier than that of the myometrium; the early enhancement patterns of endometrium were mainly slightly enhanced. And 32 cases were endometrial polyps whose transvaginal CEUS results showed that the initial enhancement time of endometrium was slightly later than that of the myometrium, and the regression time of endometrium was earlier than that of the myometrium; the early enhancement patterns of endometrium were diverse, mainly characterized by a slightly higher nodular enhancement. ③The peak intensities of malignant group, benign group and control group were 25.1 dB （14.6-26.1 dB）, 10.7 dB （7.0-15.5 dB）, and 10.1 dB （7.2-15.1 dB）, respectively, and TIC-AUC of malignant group, benign group and control group were 1 880.7 （734.9-2 084.2）, 418.0 （240.0-718.7）, and 349.8 （238.3-563.6）, respectively. There were statistical differences among three groups in the peak intensity and TIC-AUC （ H =6.732, 6.732; P =0.035, 0.011）. But there were no statistical differences among three groups in the peak time, gradient and initial enhancement time of endometrium （ P 〉0.05）. ④Among the indicators of TIC, the ROC-AUC of peak intensity and TIC-AUC for predicting the incidence of endometrial cancer were 0.92 （95% CI ： 0.78-0.97, P =0.028） and 0.96 （95% CI ： 0.84-0.99, P =0.016）, respectively. The optimal cut-off values of peak intensity and TIC-AUC for predicting the incidence of endometrial cancer were 16.608 dB and 767.45 respectively, and the sensitivities were 81.2% and 85.1% respectively, and the specificities were 85.8% and 87.2% respectively. Conclusions The images of endometrial lesions in transvaginal CEUS have certain characteristics. And transvaginal CEUS has certain diagnostic values in the diagnosis of endometrial malignant lesions.