动态对比增强MRI定量分析对外周带前列腺癌神经侵犯的预测价值  

The predictive value of dynamic contrast-enhanced MRI quantitative analysis for perineural invasion in peripheral prostate cancer

作　　者：蔡二朋 汤凯 胡晓峰 张虎 朱先锋 王颜 CAI Erpeng;TANG Kai;HU Xiaofeng;ZHANG Hu;ZHU Xianfeng;WANG Yan(Department of Radiology,the Second People’s Hospital of Wuhu,Wuhu,Anhui Province 241000,China;Department of Pathology,the Second People’s Hospital of Wuhu,Wuhu,Anhui Province 241000,China;Department of Radiology,the First Affiliated Hospital of USTC,Hefei 230001,China)

机构地区：[1]芜湖市第二人民医院影像科,安徽芜湖241000 [2]芜湖市第二人民医院病理科,安徽芜湖241000 [3]中国科学技术大学附属第一医院影像科,安徽合肥230001

出　　处：《实用放射学杂志》2024年第10期1649-1652,1657,共5页Journal of Practical Radiology

基　　金：中国红十字基金会医学赋能领航菁英科研项目(XM_LHJY2022_05_16)。

摘　　要：目的探讨动态对比增强磁共振成像(DCE-MRI)定量参数速率常数(Kep)、转运常数(K^(trans))预测外周带前列腺癌(PCa)神经侵犯(PNI)的价值。方法回顾性分析行根治性前列腺切除术(RP)的45例外周带PCa患者临床及术前MRI资料,根据病理结果分为PNI组(n=27)和无PNI组(n=18)。比较2组间年龄、总前列腺特异性抗原(tPSA)、K^(trans)值、Kep值、表观扩散系数(ADC)值、前列腺体积、病灶最大径、前列腺特异性抗原密度(PSAD)的差异。多因素logistic回归分析得出PNI的独立预测因子并构建联合预测模型;DeLong检验比较联合预测模型和各独立预测因子曲线下面积(AUC)间的差异。结果单因素分析显示tPSA、K^(trans)值、ADC值、病灶最大径、PSAD在2组间差异有统计学意义(P<0.01)。多因素分析显示K^(trans)值和病灶最大径是PNI的独立预测因子,AUC分别为0.854、0.874(P<0.01)。联合预测模型诊断PNI的AUC为0.955(P<0.001)。DeLong检验结果表明联合预测模型诊断PNI的AUC优于K^(trans)、病灶最大径(P<0.05)。结论K^(trans)值可用于预测PNI,K^(trans)值联合病灶最大径预测PNI的价值更大,较传统方法有额外收益,为临床治疗方式的选择提供更多参考依据。Objective To investigate the value of quantitative parameters(Kep and K^(trans))of dynamic contrast-enhanced magnetic resonance imaging(DCE-MRI),in predicting perineural invasion(PNI)in peripheral prostate cancer(PCa).Methods The clinical and preoperative MRI data of 45 patients with peripheral PCa who underwent radical prostatectomy(RP)were analyzed retrospectively.According to the pathological results,the patients were divided into PNI group(n=27)and non-PNI group(n=18).Various parameters,including age,total prostate specific antigen(tPSA),K^(trans)value,Kep value,apparent diffusion coefficient(ADC)value,prostate volume,maximum lesion diameter,and prostate-specific antigen density(PSAD)were compared between the two groups.Multivariate logistic regression analysis was used to identify independent predictors of PNI,and a joint prediction model was established.The DeLong test was used to compare differences in the area under the curve(AUC)between the joint prediction model and each independent predictor.Results The univariate analysis identified statistically significant differences in the tPSA,K^(trans)value,ADC value,maximum lesion diameter,and PSAD between the two groups(P<0.01).The multivariate analysis showed that the K^(trans)value and the maximum lesion diameter were independent predictors of PNI,with AUC of 0.854 and 0.874,respectively(P<0.01).The AUC of the joint prediction model for PNI diagnosis was 0.955(P<0.001).The DeLong test showed that the AUC of the joint prediction model for PNI diagnosis was better than that of the K^(trans)and the maximum lesion diameter(P<0.05).Conclusion The K^(trans)value can be used to predict PNI.Furthermore,the combination of the K^(trans)value and the maximum lesion diameter is more effective for predicting PNI than traditional methods.This provides more reference basis for the selection of clinical treatment methods.

关 键 词：动态对比增强磁共振成像 定量参数 前列腺癌 神经侵犯 

分 类 号：R445.2[医药卫生—影像医学与核医学] R737.25[医药卫生—诊断学]