动态增强MRI在卵巢良恶性肿瘤鉴别中的价值

Application of dynamic contrast-enhanced MRI in differentiating malignant from benign ovarian tumors

目的:探讨动态增强MRI(dynamic contrast-enhanced MRI,DCE-MRI)对卵巢良恶性肿瘤的诊断和鉴别诊断价值。方法:回顾性分析经手术和病理证实、含有实质成分的86例卵巢良恶性肿瘤(良性15个、恶性71个)的DCE-MRI资料。评价标准包括时间-强度曲线(time-intensity curve,TIC)类型(分为Ⅰ、Ⅱ、Ⅲ型)和3项半定量参数:增强幅度(enhancement amplitude,EA)、最大斜率(maximal slope,MS)及达峰值一半时间(time of half rising,THR),比较良恶性肿瘤的差异。结果:恶性肿瘤中,TICⅢ型59例(83%)、Ⅱ型12例(17%),未见Ⅰ型;良性肿瘤中,TICⅠ型5例(33%)、Ⅱ型10例(67%),未见Ⅲ型;TIC类型在良恶性肿瘤中差异有统计学意义(P<0.001)。恶性肿瘤的平均EA及MS值均大于良性肿瘤[(267.4±86.2)vs.(220.2±90.5),(11.0±6.3)vs.(6.1±4.7),P值分别为0.05和<0.001],而THR值则低于良性肿瘤[(32.8±7.6)s vs.(55.5±15.4)s,P<0.001)]。THR值是诊断效能最高的指标,受试者工作特性曲线的曲线下面积为89%。当THR<45 s时,诊断灵敏度、特异度、准确率、阳性和阴性预测值分别为94%、80%、92%、96%和75%。结论:DCE-MRI的TIC类型及其3个半定量参数有助于卵巢良恶性肿瘤的鉴别。

Objective: To investigate the value of dynamic contrast-enhanced MRI(DCE-MRI) in the differential diagnosis of ovarian malignant and benign tumors. Methods: DCE-MRI data of 86 ovarian tumors with solid compositions(benign 15, malignant 71) confirmed by surgery and pathology were retrospectively analyzed. The patterns(Ⅰ, Ⅱ and Ⅲ) of time-intensity curve(TIC) and three semi-quantitative parameters including enhancement amplitude(EA), maximal slope(MS) and time of half rising(THR) were evaluated and compared between ovarian benign and malignant tumors. Results: Among malignant ovarian tumors, 59(83%) were type Ⅲ, 12(17%) were type Ⅱ, and no type Ⅰ. Among benign ovarian tumors, 5(33%) were typeⅠ, 10(67%) were type II, and no type Ⅲ. There was significant difference in TIC pattern between benign and malignant tumors(P<0.001). The mean values of EA and MS in malignant ovarian tumors were significantly higher than those in benign tumors [(267.4±86.2) vs.(220.2±90.5),(11.0±6.3) vs.(6.1±4.7); P=0.05 and P<0.001, respectively]. Whereas the mean value of THR in malignant tumors was significantly lower than that in benign tumors [(32.8±7.6) s vs.(55.5±15.4) s; P<0.001]. THR was the best diagnostic indicator among three semiquantitative parameters, and the area under the receiver operating characteristic(ROC) curve was 89%. When the cutoff value of THR was 45 s, the sensitivity, specificity, accuracy, positive and negative predictive values were 94%, 80%, 92%, 96% and 75%, respectively. Conclusion: TIC patterns and semi-quantitative parameters of DCE-MRI are helpful in distinguishing malignant from benign ovarian tumors.