论文部分内容阅读
目的 探索基于前列腺影像报告和数据系统第二版(PI-RADS v2)联合前列腺癌相关生物指标建立预测前列腺高级别肿瘤的列线图模型.方法 回顾性分析2014年1月至2016年8月本院接受前列腺多参数磁共振成像检查的患者资料,根据PI-RADS v2标准对前列腺主要病灶进行评分,纳入患者年龄、PI-RADS v2、总前列腺特异抗原(tPSA)、游离前列腺特异抗原(fPSA)、前列腺体积、前列腺特异抗原密度(PSAD),游离/总前列腺抗原百分比(f/t)比值等相关指标进行多因素Logistic回归分析,病理采用超声引导穿刺活检或前列腺切除作为“金标准”.各指标在前列腺高级别肿瘤中的诊断价值采用受试者工作特征(ROC)曲线分析.筛选出的预测因子通过R软件建立nomogram模型,最后采用留一交叉验证评估模型判别能力.结果 共111例患者纳入研究,ROC曲线分析显示PSAD在诊断前列腺高级别肿瘤中曲线下面积(AUC)最大(AUC =0.84,95％ CI:0.77,0.90);多因素Logistic回归分析显示患者年龄(OR=1.10,95％ CI:1.01,1.20,P=0.023)、PI-RADS v2评分(OR=3.05,95％ CI:1.70,5.49,P=0.001)、前列腺体积(OR=0.96,95％ CI:0.93,0.99,P=0.020)为高级别肿瘤的独立预测因素,拟合ROC曲线AUC达0.92(95％ CI:0.87,0.97).留一交叉验证该模型对82％的病例进行了准确分类.结论 基于患者年龄、PI-RADS v2、前列腺体积建立的前列腺高级别肿瘤预测模型诊断准确性明显提高,值得推广运用.“,”Objective To explore a predictive nomogram for high-grade prostate cancer(HGPCa) based on Prostate Imaging Reporting and Data System version 2 (PI-RADS v2) and other traditional classical parameters.Methods Between January 2014 and August 2016,111 consecutive patients with multiparametric prostate MRI (mp-MRI) were scored according to PI-RADS v2.Patient age,tPSA、fPSA,f/t ratio,prostate volume,PSAD were also considered as predictive factors.Pathological examination from ultrasound-guided biopsy or prostatectomy was the gold standard.Their performance was evaluated as area under the curve (AUC) of the receiver operating characteristic.Multiple logistic regression analyses were used to select independent predictors for HGPCa.Leave-one-out cross validation was used to estimate the performance of the classifier.Results The highest classification accuracy was achieved by the PSAD (AUC 0.84,95％ CI:0.77,0.90).The Logistic regression model demonstrated that patient age (OR =1.10,95％ CI:1.01,1.20,P =0.023),PIRADS v2 (OR =3.05,95％ CI:1.70,5.49,P =0.001) and prostate volume (OR =0.96,95％ CI:0.93,0.99,P =0.020) were independent factors for HGPCa,which led to an AUC of 0.92 (95％ CI:0.87,0.97).Leave-one-out crossvalidation confirmed that the model has accurately classified 82％ cases.Conclusion A nomogram based on patient age,PI-RADS v2 and prostate volume were developed to predict HGPCa,and it achieved the higher test accuracy compared with other parameters alone.