摘要
目的对比T_(1)mapping成像与扩散峰度成像(DKI)的定量参数在宫颈癌组织学特征评估中的应用价值。方法收集2018年11月至2020年6月经郑州大学第一附属医院病理证实的58例宫颈癌患者和30例宫颈正常的其他疾病患者的资料,将宫颈癌患者按照病理分型分为鳞癌(46例)组和腺癌组(12例),按照病理分化程度分为中高分化组(42例)和低分化组(16例)。患者术前接受常规盆腔磁共振成像(MRI)平扫、DKI序列扫描检查,采用B1修正变量翻转角序列分别于增强前及增强后5 min采集T_(1)mapping图像,测定肿瘤实质区和正常宫颈表观扩散系数(ADC)值、平均扩散峰度(MK)值、平均扩散系数(MD)值以及增强前弛豫时间(T_(1pre))和增强后弛豫时间(T_(1post))。采用独立样本t检验比较组间ADC值、MK值、MD值、T_(1pre)和T_(1post)差异,有统计学意义的参数绘制受试者工作特征(ROC)曲线,以曲线下面积(AUC)表示参数的诊断性能,采用Spearman相关分析各定量参数值与宫颈癌分型、分化的相关性。结果(1)肿瘤实质区和正常宫颈的ADC值、MK值、MD值、T_(1pre)和T_(1post)间差异均有统计学意义(P<0.05);鳞癌和腺癌的MK值、T_(1pre)间差异均有统计学意义(P<0.05);中高分化组和低分化组的ADC值、MK值、MD值、T_(1pre)和T_(1post)间差异均有统计学意义(P<0.05)。(2)在鳞癌和腺癌的鉴别中,T_(1pre)具有较好的鉴别诊断能力(AUC=0.764,P<0.001),其次为MK值(AUC=0.685,P=0.037);在低分化和中高分化宫颈癌的鉴别中,MK值具有最高鉴别诊断能力(AUC=0.812,P<0.001),其次是T_(1post)(AUC=0.797,P<0.001),但T_(1post)的敏感度最高(88.10%)。(3)T_(1pre)与宫颈癌分型呈正相关(r=0.338,P<0.05);在分化程度方面,T_(1post)(r=-0.460,P<0.001)和MK值(r=-0.483,P<0.001)与宫颈癌分化呈负相关,ADC值(r=0.455,P<0.001)、MD值(r=0.377,P=0.004)和T_(1pre)(r=0.273,P=0.038)与宫颈癌分化程度呈正相关。结论在宫颈癌组织学特征评估中,T_(1)mapping成像与DKI具有相似诊断效能,但在临床应用中T_(1)mapping成像技术更简单、快速、全面,因此更推荐其作为宫颈癌的首选辅助检查。
Objective To compare the application value of quantitative parameters of T 1 mapping and diffusion kurtosis imaging(DKI)in histological characteristics assessment of cervical cancer.Methods The data of 58 patients with pathologically proven cervical cancer and 30 patients with normal cervix were collected in the First Affiliated Hospital of Zhengzhou University from November 2018 to June 2020.Cervical cancer patients were divided into squamous cell carcinoma(46 cases)and adenocarcinoma(12 cases).According to the degree of pathological differentiation,patients were divided into middle-high differentiation group(42 cases)and low differentiation group(16 cases).Preoperative routine pelvic magnetic resonance imaging(MRI)plain scan,DKI sequence scan.Using B1 correction variable-flip-angle sequence,the T 1 mapping images were collected 5 minutes before and after enhancement.The apparent diffusion coefficient(ADC),mean diffusion kurtosis(MK),mean diffusion coefficient(MD),pre-enhancement relaxation time(T_(1pre))and post-enhancement relaxation time(T_(1post))were measured in the parenchymal area of the tumor and normal cervix.The independent sample t test was used to compare the difference of ADC,MK,MD,T_(1pre) and T_(1post) values among these groups.Statistically significant parameters were used to draw receiver operating characteristic(ROC)curves.Spearman correlation analysis was used to analyze the correlation between quantitative parameter values and cervical cancer pathological characteristics.Results The ADC,MK,MD,T_(1pre) and T_(1post) values showed significant differences between cervical cancer and normal tissue(P<0.05).There were significant differences in the MK and T_(1pre) values between squamous cell carcinomas and adenocarcinomas(P<0.05).The ADC,MK,MD,T_(1pre) and T_(1post) values showed significant differences between middle-high differentiation group and low differentiation group(P<0.05).T_(1pre) had better differential diagnosis ability in the squamous cell carcinoma and adenocarcinoma(AUC=0.764,P=0.001),MK values were followed(AUC=0.685,P=0.037).T_(1pre) had better differential diagnosis ability in middle-high differentiation group and low differentiation group(AUC=0.812,P<0.001),T_(1post) values were followed(AUC=0.797,P<0.001),but the sensitivity of T_(1post) was the highest(88.10%).The T_(1pre) value showed a positive association with cervical cancer classification(r=0.338,P<0.05).In terms of differentiation,T_(1post)(r=-0.460,P=0.001)and MK values(r=-0.483,P=0.001)were negatively correlated with cervical cancer differentiation,and ADC(r=0.455,P<0.001),MD(r=0.377,P=0.004),T_(1pre)(r=0.273,P=0.038)values were positively correlated with cervical cancer differentiation.Conclusion In the pathological features assessment of cervical cancer,both T 1 mapping and DKI shows similar diagnostic performance,but T 1 mapping is simpler,faster and more comprehensive in clinical application that it is more recommended as the first choice of magnetic resonance imaging for cervical cancer.
作者
杨梦
李淑健
刘洁
张斐斐
张勇
程敬亮
YANG Meng;LI Shujian;LIU Jie;ZHANG Feifei;ZHANG Yong;CHENG Jingliang(Department of Magnetic Resonance,the First Affiliated Hospital of Zhengzhou University,Zhengzhou 450052,China)
出处
《河南医学研究》
CAS
2021年第6期978-982,共5页
Henan Medical Research
作者简介
通信作者:程敬亮,E-mail:cjr.chjl@vip.163.com。
