基于PSAD及mp-MRI的联合模型对临床显著性前列腺癌诊断价值的研究  

Study on the diagnostic value of combined models based on PSAD and mp-MRI in clinically significant prostate cancer

作　　者：辛建英 张文轩 韩雨 巩平 徐爱霞 王锡臻 XIN Jianying;ZHANG Wenxuan;HAN Yu;GONG Ping;XU Aixia;WANG Xizhen(Medical Imaging Center,Affiliated Hospital of Shandong Second Medical University,Weifang 261031,China)

机构地区：[1]山东第二医科大学附属医院影像中心,潍坊261031

出　　处：《磁共振成像》2025年第2期72-76,141,共6页Chinese Journal of Magnetic Resonance Imaging

基　　金：潍坊市卫生健康委员会科研项目(编号:WFWSJK-2023-281);潍坊医学院附属医院医学研究培育基金项目(编号:2022wyfyzzjj01)。

摘　　要：目的探讨前列腺特异性抗原密度(prostate specific antigen density,PSAD)联合多参数磁共振成像(multi-parameter magnetic resonance imaging,mp-MRI)在临床显著性前列腺癌(clinically significant prostate cancer,csPCa)诊断中的价值。材料与方法回顾性分析105例疑诊为PCa并取得病理结果患者的临床及影像资料。术前均行mp-MRI检查、前列腺特异性抗原(prostate specific antigen,PSA)检测。根据病理及Gleason评分,将患者分为csPCa和非csPCa两组,比较两组间PSAD和mp-MRI的表观扩散系数(apparent diffusion coefficient,ADC)值、开始强化时间(T0)、短暂增强时间、流入速率(wash in rate,WIR)等参数的差异,采用二元logistic回归构建联合诊断模型,运用受试者工作特征(receiver operating characteristic,ROC)曲线评估各参数及构建模型诊断csPCa的效能。结果csPCa组的ADC、T0低于非csPCa组,WIR、PSAD、短暂增强时间高于非csPCa,差异均有统计学意义(P<0.05)。PSAD、ADC、WIR单独诊断csPCa的ROC曲线下面积(area under the curve,AUC)及95%置信区间分别为0.902(0.829~0.952)、0.890(0.814~0.942)、0.812(0.724~0.882),具有较高的诊断效能,临床诊断界值分别为0.47 ng/(mL·cm^(3))、0.82×10-3 mm^(2)/s、50.33 s^(-1),敏感度分别为79.1%、67.4%、100.0%,特异度分别为100.0%、100.0%、54.8%。PSAD、ADC、WIR任意两参数(WIR+PSAD、ADC+PSAD、WIR+ADC)及多参数(WIR+PSAD+ADC)联合诊断csPCa的AUC、敏感度、特异度分别为:0.929(0.862~0.970)、83.7%、96.8%;0.940(0.877~0.977)、90.7%、91.9%;0.935(0.870~0.974)、79.1%、95.2%;0.955(0.896~0.986)、90.7%、91.9%。ROC曲线对比分析显示WIR+PSAD+ADC、ADC+PSAD、WIR+ADC联合模型与ADC、WIR单参数诊断csPCa的AUC差异均具有统计学意义(P<0.05);WIR+PSAD联合模型与WIR单独诊断csPCa的AUC差异具有统计学意义(P<0.05)。结论PSAD联合mp-MRI对csPCa具有较高的诊断价值,其主要指标构建的诊断模型可用于预测csPCa。Objective:To evaluate the value of prostate specific antigen density(PSAD)combined with multi-parameter magnetic resonance imaging(mp-MRI)used in diagnosing clinically significant prostate cancer(csPCa).Materials and Methods:Retrospective analysis of clinical and imaging data of 105 patients with suspected PCa and pathological findings were selected,prostate specific antigen(PSA)detection and mp-MRI were performed before surgery.Based on pathological results and Gleason score,patients were divided into csPCa and non-csPCa groups.Parameters of mp-MRI including apparent diffusion coefficient(ADC),Begin time of enhancement(T0),Brevity of enhancement,wash in rate(WIR)and PSAD were compared between the two groups,combined diagnostic models were constructed by binary logistic regression,and receiver operating characteristic(ROC)curves were used to evaluate the diagnostic efficacy of each parameter and model for csPCa.Results:The ADC and T0 in csPCa group were lower than those in non-csPCa group,while the PSAD,WIR and Brevity of enhancement were opposite,and all differences reach statistical significance(P<0.05).The area under the curve(AUC)for the diagnosis of csPCa in PSAD,ADC and WIR were 0.902(0.829 to 0.952),0.890(0.814 to 0.942)and 0.812(0.724 to 0.882)respectively with higher diagnostic efficacy,the clinical diagnostic boundaries were 0.47 ng/(mL·cm^(3)),0.82×10-3 mm^(2)/s,50.33 s^(-1),the sensitivities were 79.1%,67.4%,100.0%,and the specificities were 100.0%,100.0%and 54.8%,respectively.The AUC,sensitivity and specificity of any two and multi-parameter combined diagnosis of csPCa by WIR,PSAD and ADC:WIR+PSAD 0.929(0.862 to 0.970),83.7%,96.8%;ADC+PSAD 0.940(0.877 to 0.977),90.7%,91.9%;WIR+ADC 0.935(0.870 to 0.974),79.1%,95.2%;WIR+PSAD+ADC 0.955(0.896 to 0.986),90.7%,91.9%,respectively.ROC curve contrast analysis revealed significant differences in AUC between WIR+PSAD+ADC,ADC+PSAD,WIR+ADC combined diagnostic models and ADC,WIR single parameter diagnosis of csPCa(P<0.05);the AUC of WIR+PSAD model was different

关 键 词：临床显著性前列腺癌 前列腺特异性抗原密度 多参数磁共振成像 扩散加权成像 动态对比增强 诊断效能 

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]