ADC值联合第二版前列腺影像报告和数据系统对有临床意义前列腺癌的诊断效能分析

Analyze of diagnostic efficiency of ADC value combined with prostate imaging reporting and data system version 2 for clinically significant prostate cancer

目的评估ADC值或ADC分类联合第二版前列腺影像报告和数据系统(PI-RADS v2)对有临床意义前列腺癌(CSPCa)的诊断效能,并与单独PI-RADS v2评分、ADC值或ADC分类进行比较。方法回顾性分析310例疑似前列腺癌[前列腺特异抗原(PSA)异常]患者的临床资料,先行3.0T多参数磁共振成像(mpMRI)检查,在mpMRI检查后4周内行经直肠超声(TRUS)引导下前列腺穿刺;对可疑病灶行PI-RADS v2评分和ADC值测定。以穿刺病理结果为金标准,用多变量二元Logistic回归分析对预测指标建立联合预测模型,使用受试者工作特征(ROC)曲线分析联合预测模型和单独观察指标的诊断效能,用Delong检验比较曲线下面积(AUC)。结果纳入可疑病灶共375个,病理证实CSPCa为140个。Logistic回归分析结果显示,PI-RADS v2评分、ADC值和ADC分类均为CSPCa的独立预测指标(P<0.001)。PI-RADS v2评分+ADC值、PI-RADS v2评分+ADC分类诊断CSPCa的AUC值均为0.92,两者比较,差异无统计学意义(P=0.93),且均高于单独PI-RADS v2评分的0.88(P<0.001);PI-RADS v2评分+ADC分类诊断外周带(PZ)CSPCa和移行带(TZ)CSPCa的AUC高于PI-RADS v2评分(P<0.05)。PI-RADS v2评分为4、5分时,CSPCa的阳性率分别为25.8%、83.7%,分类后阳性率上升为47.3%、90.5%。ROC结果显示,ADC分类在PI-RADS v2评分为4分时的预测效能最佳。结论ADC值或ADC分类能明显提高PI-RADS v2评分对CSPCa的预测效能,在PI-RADS v2评分为4分时的诊断效能最佳。

Objective To evaluate the diagnostic efficacy of ADC value or ADC categories combined with prostate imaging reporting and data system version 2(PI-RADS v2)for clinically significant prostate cancer(CSPCa),and to compare it with PI-RADS v2 score,ADC value or ADC categories alone.Methods The clinical data of 310 patients with suspected prostate cancer[prostate specific antigen(PSA)abnormality]were analyzed retrospectively.The 3.0T multiparameter magnetic resonance imaging(mpMRI)was performed first,and transrectal ultrasound(TRUS)guided prostate puncture was performed within 4 weeks after mpMRI;PI-RADS v2 score and ADC value of suspicious lesions were measured.Taking the pathological results of puncture as the gold standard,a joint prediction model was established for the prediction indexes by multivariable binary Logistic regression analysis,the diagnostic efficacy of the joint prediction model and individual observation indexes were analyzed by receiver operating characteristic(ROC)curve,and the area under the curve(AUC)was compared by Delong test.Results A total of 375 suspicious lesions were included,of which 140 were as confirmed CSPCa by pathology.Logistic regression analysis result showed that the PI-RADS v2 score,ADC value and ADC categories were independent predictors of CSPCa(P<0.001).The AUC value of PI-RADS v2 score+ADC value and PI-RADS v2 score+ADC categories in the diagnosis CSPCa was 0.92,there was no significant difference between them(P=0.93),and they were higher than 0.88 of PI-RADS v2 score alone(P<0.001);the AUC of PI-RADS v2 score+ADC categories in the diagnosis of peripheral zone(PZ)CSPCa and transitional zone(TZ)CSPCa was higher than that of PI-RADS v2 score(P<0.05).When PI-RADS v2 score was 4 and 5 points,the positive rates of CSPCa were 25.8%and 83.7%respectively,after classification,the positive rates increased to 47.3%and 90.5%.ROC results showed that ADC categories had the best diagnostic efficiency when PI-RADS v2 score was 4 points.Conclusion ADC value or ADC categories can significantly improve the diagnostic efficiency of PI-RADS v2 score for CSPCa,and the diagnostic efficiency is the best when PI-RADS v2 score is 4 points.