DCE-MRI对正常卵巢及囊实性卵巢肿瘤的临床价值研究

发布时间：2018-04-02 15:08

本文选题：动态增强磁共振成像　切入点：月经周期　出处：《天津医科大学》2014年硕士论文

【摘要】：目的评估正常卵巢动态增强磁共振成像(dynamic contrast-enhanced magnetic resonance imaging, DCE-MRI)检查的可行性,观察不同月经周期正常卵巢3.0TDCE-MRI参数及时间-信号曲线(time-intensity curve, TIC)特点。探讨DCE-MRI在不同性质囊实性卵巢肿瘤鉴别诊断方面的价值,分析其定量参数值的特点,以及DCE-MRI与扩散加权成像(Diffusion-weighted imaging, DWI)对囊实性卵巢肿瘤鉴别诊断效能的差异,为卵巢肿瘤定性诊断选择最优检查方法。对象与方法采用Siemens MAGNETOM Trio Tim3.0T超导型磁共振扫描仪,对33名处于不同月经周期的正常卵巢组织患者行盆腔MR平扫+DCE-MRI检查。同时对2012年11月-2014年1月间因盆腔肿块就诊的87例患者行盆腔MR、DWI (b值为1000s/mm2)及脂肪抑制DCE-MRI扫描,最终有48例患者入组。扫描完成后均在Siemens Syngo后处理工作站采用Tissue4D软件进行图像后处理,得到卵巢间质组织的TIC图像及Ktrans、Kep、Ve值。所有卵巢肿瘤患者均于检查后10天内获得病理结果,并根据病理结果分为卵巢良性肿瘤、交界性肿瘤及恶性肿瘤(1组、2组、3组)三组。采用独立样本t检验分别比较不同月经周期卵巢间质Ktrans、Kep及Ve值的差异。分别比较DWI、DCE-MRI及两者联合对卵巢肿瘤诊断效能的差异。分析TIC诊断卵巢肿瘤的灵敏度、特异度、阴性预测值、阳性预测值,比较不同组间有无差异。分别测量各组卵巢肿瘤实性成分的Ktrans、Kep、Ve值及ADC值,采用单因素方差分析比较各组间四个参数值的差异。采用ROC曲线比较Ktrans、Kep、Ve值及ADC值对不同性质卵巢肿瘤的鉴别诊断效能并确定最佳诊断阈值。结果 (1)不同月经周期正常卵巢间质成分TIC形态无明显差异(P0.05),均表现为I型曲线；(2)卵泡期及黄体期正常卵巢间质Ktrans值分别为0.213±0.129(min-1)、0.189±0.124(min-1),差异无统计学意义；Kep值于卵泡期及黄体期分别为0.237±0.103(min-1)、0.248±0.14差异无统计学意义；Ve值于卵泡期及黄体期分别为1.692±1.365、1.575±1.085,差异无统计学意义；(3)DWI对卵巢肿瘤鉴别诊断的灵敏度为68.18%,特异度为69.23%,阴性预测值为72%,阳性预测值为65.22%；DCE-MRI对卵巢肿瘤鉴别诊断的灵敏度为76.19%,特异度为74.07%,阴性预测值为80%,阳性预测值为69.57%；两者联合对卵巢肿瘤鉴别诊断的灵敏度为85.72%,特异度为100%,阴性预测值为86.96%,阳性预测值为100%,以上指标均高于单一检查方法；(4)卵巢良性肿瘤与恶性肿瘤之间实性成分的ADC值差异有统计学意义(P=0.000),卵巢交界性肿瘤与良性及恶性肿瘤之间实性成分的ADC值无统计学差异(P=0.052,P=0.051)；(5)卵巢良性肿瘤TIC以I型曲线为主,交界性及恶性肿瘤TIC均以Ⅲ型曲线为主,其中良性肿瘤与交界性和恶性肿瘤之间的TIC差异有统计学意义(P0.05),而交界性肿瘤与恶性肿瘤之间的TIC形态无显著差异(P0.05)；(6)不同性质卵巢肿瘤实性成分Ktrans值差异均具有统计学差异(P0.05),而Kep值及Ve值差异不明显(P0.05)；(7)ADC值、Ktrans、Kep及Ve值对良性和恶性肿瘤的最佳鉴别诊断阈值分别为1.174×10-3mm2/s、0.123min-1、0.262min-1、0.549,且均有较高的敏感度、阳性预测值及阴性预测值,但Kep、Ktrans值及ADC值的AUC无统计学差异(P0.05)；对交界性与恶性肿瘤的最佳鉴别诊断阈值分别为0.933×10-3mm2/s、0.212min-1、0.634,敏感度、特异度、阳性预测值较高,但阴性预测值较低,其中Kep值无意义；Ve值对良性与交界性肿瘤鉴别无意义；ADC值与Ktrans值对良性与交界性肿瘤的最佳鉴别诊断阈值分别为1.217×10-3mm2/s、0.108min-1,具有很高的敏感度和阴性预测值。结论利用3.0T DCE-MRI对正常卵巢间质检查具有可行性：正常卵巢间质的Ktrans值、Kep值及Ve值分别为卵巢肿瘤特点的研究提供了正常参考价值；应用DCE-MRI无创性鉴别囊实性卵巢肿瘤性质具有可行性,能够直观有效量化对不同性质囊实性卵巢肿瘤进行鉴别诊断,且Ktrans值对于卵巢肿瘤良恶性的鉴别具有最佳诊断效能。
[Abstract]:objective
Evaluation of normal ovarian dynamic enhanced magnetic resonance imaging (dynamic contrast-enhanced magnetic resonance imaging, DCE-MRI) to examine the feasibility of observing the signal curve of different menstrual cycle in normal ovarian 3.0TDCE-MRI parameters and time (time-intensity curve, TIC). To explore DCE-MRI in differential diagnosis of different cystic ovarian tumors and to analyze the characteristics of quantitative parameters the value of DCE-MRI and diffusion weighted imaging (Diffusion-weighted imaging, DWI) in the differential diagnosis of cystic ovarian cancer diagnostic efficacy differences, ovarian cancer diagnosis to select the optimal inspection method. Object and method
The Siemens MAGNETOM Trio Tim3.0T superconducting magnetic resonance scanner in 33 patients at different menstrual cycle in normal ovarian tissue underwent pelvic MR scan and +DCE-MRI examination. On November 2012 -2014 year in January for 87 cases of patients with pelvic pelvic masses were MR, DWI (b = 1000s/mm2) and fat suppression DCE-MRI scan, the final 48 patients were enrolled. After scanning were Siemens Syngo postprocessing workstation for postprocessing images using Tissue4D software, get the ovarian interstitial TIC image and Ktrans, Kep, Ve. All patients with ovarian tumor were obtained in 10 days after the examination of pathological results, and according to the pathological results were divided into benign ovarian the tumor, borderline tumors and malignant tumors (1 group, 2 group, 3 group). Three groups of independent samples t test was used to compare different menstrual cycle and ovarian stromal Ktrans, the difference of Kep and Ve value were compared. DWI, DCE-MRI And the combination for diagnosis of ovarian neoplasms. The efficiency difference sensitivity analysis, TIC diagnosis of ovarian tumor specificity, negative predictive value, positive predictive value, comparison between different groups have no difference. The ovarian tumor solid component were measured Ktrans, Kep, Ve and ADC values, the difference produced by single factor analysis of variance comparisons between groups of values of four parameters. The ROC curve of Ktrans, Kep, Ve and ADC values in differential diagnosis of ovarian tumor efficacy of different nature and determine the best diagnostic threshold.
Result
(1) no significant difference in different menstrual cycle normal ovarian stromal component (P0.05), the morphology of TIC showed I curve; (2) follicular phase and luteal phase of normal ovarian interstitial Ktrans = 0.213 + 0.129 + 0.124 (min-1), 0.189 (min-1), there was no significant difference in Kep value; in the follicular phase and luteal phase were 0.237 + 0.103 (min-1), no statistically significant difference of 0.248 + 0.14; the value of Ve in follicular phase and luteal phase were 1.692 + 1.365,1.575 + 1.085, the difference was not statistically significant; (3) the sensitivity of DWI for differential diagnosis of ovarian tumors was 68.18%, the specificity was 69.23%, negative the predictive value was 72%, the positive predictive value was 65.22%; the sensitivity of DCE-MRI for differential diagnosis of ovarian tumors was 76.19%, the specificity was 74.07%, the negative predictive value was 80%, the positive predictive value was 69.57%; the combination of differential diagnosis of ovarian tumor sensitivity was 85.72%, specificity was 100%, negative The predictive value was 86.96%, the positive predictive value was 100%, the above indicators were higher than the single test method; (4) between ovarian benign tumor and malignant tumor solid composition ADC value difference was statistically significant (P=0.000), borderline ovarian tumors between benign and malignant tumors and solid components showed no significant difference (ADC value P=0.052, P=0.051); (5) TIC in benign ovarian tumor I curve, borderline and malignant tumors TIC were mainly type III curve, which was statistically significant between benign and borderline and malignant tumors TIC (P0.05), and the difference between the junction of tumor and malignant tumor morphology had no significant TIC the difference (P0.05); (6) ovarian tumor solid composition Ktrans value difference was statistically significant (P0.05), while Kep and Ve values were not significantly different (P0.05); (7) the value of ADC, Ktrans, Kep and Ve was the best for the identification of benign and malignant tumors The diagnostic threshold was 1.174 * 10-3mm2/s, 0.123min-1,0.262min-1,0.549, and has high sensitivity, positive predictive value and negative predictive value, but no significant difference in Kep, Ktrans and ADC values of AUC (P0.05); the optimal threshold for differential diagnosis of borderline and malignant tumors were 0.933 * 10-3mm2/s, 0.212min-1,0.634, sensitivity the specificity, the positive predictive value is higher, but the negative predictive value is low, the Kep value is meaningless; Ve value of benign and borderline tumors without meaning; the ADC value of the best discrimination of benign and borderline tumor differential diagnosis threshold was 1.217 * 10-3mm2/s, and the Ktrans value of 0.108min-1, with a sensitivity and negative predictive the very high value.
conclusion
The normal ovarian interstitial examination has the feasibility of using 3.0T DCE-MRI: normal ovarian interstitial Ktrans value, Kep value and Ve value of the normal reference value of ovarian tumor characteristics respectively; application of DCE-MRI noninvasive identification of cystic ovarian tumors is feasible, can effective quantification of the different nature of cystic ovarian the differential diagnosis of tumor, and the value of Ktrans has the best diagnostic efficacy for differential diagnosis of benign and malignant ovarian tumors.

【学位授予单位】：天津医科大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：R737.31;R445.2

【参考文献】