摘要
目的 探讨DCE-MRI联合高b值DWI对卵巢上皮性肿瘤良性、交界性、恶性的鉴别诊断价值.方法 选取病理证实为卵巢上皮性肿瘤患者 55 例,其中良性 23 例(良性组),交界性 10 例(交界性组),恶性 22 例(恶性组);术前行MRI平扫及增强、高b值DWI(b值=1200s/mm2)、DCE-MRI定量分析,并行统计学分析确定最佳最大阈值.结果 1)良性组和恶性组A1 及Ktrans 值、Kep 值及Ve 值各组间差异均有统计学意义(P<0.05);恶性组和交界性组肿瘤A1 及Kep值各组间差异均有统计学意义(P<0.05);2)ROC 曲线分析,良、恶性组 A1 值的诊断阈值为 1179.4 mm2/s,AUC 为0.945,Ktrans值具有最高的诊断效能,诊断阈值为 0.0823min-1,AUC为 0.913;交界性组与恶性组A1 值的诊断阈值为1147.8 mm2/s,AUC为 0.891,Kep值AUC为 0.732,诊断阈值为 0.273 min-1;3)高b值DWI联合DCE-MRI对卵巢良性组和恶性组肿瘤鉴别及恶性组和交界性组肿瘤鉴别诊断具有高诊断效能.结论 1)高b值及Ktrans值在卵巢上皮性肿瘤良性组和恶性诊断中具有高诊断效能;高b值及Kep值在卵巢上皮性恶性组和交界性肿瘤诊断中具有高诊断效能.2)高b值DWI、DCE-MRI联合筛查明显提高了卵巢上皮性肿瘤术前诊断的正确性.
Abstract
Objective To investigate the potential value of dynamic contrast enhanced-MR imaging(DCE-MRI)and high-b-value diffusion weighted imaging(DWI)for differentiating benign,borderline and malignant ovarian epithelial tumor and to evaluate whether the preoperative diagnostic performance can be improved when combining two techniques.Methods This ret-rospectively study enrolled 55 patients with ovarian epithelial tumor confirmed by pathology.All patients underwent conventional MR imaging,high-b value DWI and DCE-MRI.The apparent diffusion coefficient(ADC)values were obtained from solid of tu-mor,while,Ktrans,Kep and Ve values were obtained from the solid portion of tumor.Receiver operating characteristic(ROC)curve analysis was performed to determine optimal thresholds,sensitivity(Se),specificity(Sp),positive predictive value(PPV),negative predictive value(NPV)and accuracy(ACC)for ADC,Ktrans,Kep and Ve values in the differentiation of be-nign,borderline and malignant ovarian epithelial tumor.Results 1)There were significant differences in A1,Ktrans,Kep and Ve values of solid portion between benign and malignant.There was significant difference in Ktrans value of solid portion between benign and borderline.There were significant differences in A1,Kep values of solid portion between borderline and malignant.2)In ROC curve analysis,A1 values provided a valuable diagnostic performance in differentiating malignant from benign ovarian epithelial tumor:AUC = 0.945 of solid portion with optimal thresholds,Se,Sp,PPV and NPV of 1149.9 mm2/s,95.7％,95.5％,95.5％and 95.7％,correspondingly.Beside,the AUC of Ktrans values(AUC = 0.913)was higher than that of Kep or Ve val-ues,with optimal thresholds,Se,Sp,PPV and NPV of 0.0823 min-1,90.9％,82.6％,83.9％,and 94％.A1 values provided a valuable diagnostic performance in differentiating malignant from borderline ovarian epithelial tumor:AUC = 0.891 of solid por-tion with optimal thresholds,Se,Sp,PPV and NPV of 1149.9 mm2/s,80％,95.5％,94.7％and 82.7％.Beside,the AUC of Kep values(AUC = 0.732)was higher diagnostic efficiency with optimal thresholds,Se,Sp,PPV and NPV of 0.0273min-1,72.7％,80％,78.4％,and 74.6％.The combination of DCE-MRI and high-b-value DWI showed the best diagnostic performance in the dis-crimination of malignant vs benign and borderline vs malignant ovarian epithelial tumor,but bad diagnostic performance in the discrimination of borderline vs benign.Conclusion High-b-value DWI and Ktrans values derived from DCE are useful for differ-entiating malignant from benign ovarian epithelial tumors.High-b-value DWI and Kep values are useful for differentiating malig-nant from borderline tumor.But the both are bad diagnostic performance in the discrimination of the borderline vs benign.The combination of DCE-MRI with high-b-value DWI can improve the preoperative diagnostic performance in the discrimination of malignant,borderline and benign ovarian epithelial tumors.
维普
万方数据