期刊文献+

FIGO ⅠB~ⅡA期宫颈癌原发灶MRI动态对比增强的定量参数预测盆腔淋巴结转移 被引量:9

Predictive Value of Quantitative Parameters on Pelvic Lymph Node Metastasis by Dynamic Contrast-Enhanced Magnetic Resonance Imaging of FIGO ⅠB ~ⅡA Cervical Cancer
原文传递
导出
摘要 目的探讨国际妇产科联盟(FIGO)ⅠB~ⅡA期宫颈癌原发灶3.0 T MRI动态对比增强(DCE-MRI)定量参数对盆腔淋巴结转移的预测价值。方法回顾性分析经病理证实并进行DCE-MRI的FIGOⅠB~ⅡA宫颈癌63例患者资料,利用血流动力学双室Tofts模型,获取原发灶容量转移常数(Ktrans)、回流速率常数(Kep)、血管外细胞外间隙容积分数(Ve)值,并记录所有患者MRI上肿瘤最大径、年龄、病理组织学类型、有无宫旁浸润(PI)、临床分期。将患者分为盆腔淋巴结转移(LNM)阳性和LNM阴性两组,采用独立样本t检验对MRI上肿瘤最大径、DCE-MRI定量参数(Ktrans、Kep、Ve)、年龄等参量进行比较,采用卡方检验对病理组织学类型、PI、临床分期等指标进行比较,对具有统计学差异的参量或指标行多因素Logistic回归分析和受试者工作特征曲线(ROC)分析。结果 63例患者中盆腔LNM阳性者20例,盆腔LNM阴性者43例。病理确诊PI者19例,无PI者44例。鳞癌55例,腺癌8例。患者平均年龄(51.33±10.99)岁(33~74岁);MRI上肿瘤最大径平均值(3.90±1.45)cm(1.31~6.99 cm)。原发灶Ktrans平均值(0.25±0.07)min-1(0.07~0.43 min-1);Kep平均值(0.48±0.17)min-1(0.08~0.96 min-1);Ve平均值0.57±0.12(0.31~0.86)。两组间原发灶Ktrans(P=0.021)、MRI上肿瘤最大径(P<0.001)、有无PI(P<0.001)、临床分期(P=0.007)有统计学差异,年龄(P=0.879)、Kep(P=0.914)、Ve(P=0.103)、病理组织学类型(P=0.211)无统计学差异;肿瘤最大径(P=0.002)是盆腔LNM阳性的独立风险因子;Ktrans值与肿瘤最大径联合应用对盆腔LNM阳性的预测效能最优,ROC曲线下面积为0.936,特异度97.7%,敏感度80.0%,准确率77.7%。结论 FIGOⅠB~ⅡA期宫颈癌原发灶Ktrans值与盆腔LNM相关,其联合MRI上肿瘤最大径,可在术前更好地预测盆腔LNM的可能性。 Objective To investigate the predictive value of tumorous quantitative parameters on pelvic lymph node metastasis(PLNM)by dynamic contrast-enhanced magnetic resonance imaging(DCE-MRI)of patients with FIGOⅠB~ⅡA cervical cancer(CC).Methods 63 cases of FIGOⅠB~ⅡA CC were retrospectively analyzed,which had been confirmed by pathology and went on to undergo DCE-MRI.The quantitative parameters(Ktrans,Kep,Ve)were obtained from enhancement tumors through Extended Tofts model;and other parameters and indicators,including the tumorous maximum diameter(TMD)of MRI,age,histological type,parametrial infiltration(PI),clinical stage,were recorded.Then,63 patients were divided into group of FIGOⅠB~ⅡA CC with PLNM(positive PLNM group)and group without PLNM(negative PLNM group).By means of independent sample t-test,differences of the various parameters,including age,Ktrans,Kep,Ve,and TMD,were compared between positive PLNM group and negative PLNM group.By means of Chi-square test analysis,differences of histological type,PI,clinical stage were compared between the two groups.And then using statistically different parameters or indicators,multivariate Logistic regression analysis and receiver operating characteristic curve(ROC)was performed for PLMN.Results Among the 63 patients,20 were positive on PLMN and 43 were negative on PLNM,PI was pathologically confirmed in 19 cases,55 cases were squamous cell carcinomas and 8 cases were adenocarcinomas.The average age of all patients was(51.33±10.99)years old(33~74 years old),the TMD of MRI was(3.90±1.45)cm(1.31~6.99 cm)on MRI.The average value of Ktrans was(0.25±0.07)min-1(0.07-0.43 min-1),that of Kep was(0.48±0.17)min-1(0.08-0.96 min-1),and that of Ve was 0.57±0.12 0.31~0.86).There were statistical differences on Ktrans(P=0.021),TMD(P<0.001),PI(P<0.001)and clinical stage(ⅠB vs.ⅡA,P=0.007)between the two groups;but not statistical differences on age(P=0.879),Kep(P=0.914),Ve(P=0.103),histological type(P=0.211).TMD of MRI(P=0.002)was independent risk factor for PLNM;the Ktrans combined with TMD possessed the best predictive efficacy for PLNM with an area under the ROC curve of 0.936,a specificity of 97.7%,a sensitivity of 80.0%and an accuracy of 77.7%.Conclusion The Ktrans value of the primary tumor in FIGOⅠB~ⅡA stage CC was correlated with PLNM.The possibility of PLNM should have a better prediction before operation if the Ktrans had combined with TMD of MRI.
作者 张禹 邓雪飞 张雪健 张茜 莫子 骆祥伟 朱友志 ZHANG Yu;DENG Xuefei;ZHANG Xuejian(Department of Medical Imaging,The 901st Hospital of Joint Logistics Support Force of PLA,Hefei 230031,P.R.China)
出处 《临床放射学杂志》 CSCD 北大核心 2019年第8期1449-1454,共6页 Journal of Clinical Radiology
关键词 宫颈肿瘤 淋巴转移 磁共振成像 定量分析 Cervical neoplasm Lymphatic metastasis Magnetic resonance imaging Quantitative analysis
  • 相关文献

参考文献15

二级参考文献215

共引文献223

同被引文献85

引证文献9

二级引证文献12

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部