基于PI-RADS v2建立预测前列腺高级别肿瘤的列线图模型

发布时间：2018-03-01 09:41

  本文关键词： 多参数 磁共振成像 PI-RADS 前列腺癌 列线图　出处：《临床放射学杂志》2017年12期 　论文类型：期刊论文

【摘要】：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(t PSA)、游离前列腺特异抗原(f PSA)、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。
[Abstract]:Objective to establish a linear model of prostate cancer prediction based on prostate imaging report and data system (PI-RADS v2) combined with prostate cancer related biomarkers. Methods A retrospective analysis was performed from January 2014 to August 2016. The data of patients undergoing multiparameter magnetic resonance imaging of prostate in our hospital, According to the PI-RADS v2 criteria, the main prostate lesions were graded, and the patients were included in the age of PI-RADS v2, the total prostate specific antigen (TPCA), the free prostate specific antigen (PSA), the volume of the prostate, the volume of the prostate. The density of prostate specific antigen (PSAD) and the ratio of free to total prostatic antigen (f / t) were analyzed by multivariate Logistic regression analysis. Ultrasound-guided biopsy or prostatectomy was used as the "golden standard". The diagnostic value of each index in high grade prostate tumor was analyzed by using the operating characteristics of the subjects. The predicted factors were established by R software to establish the nomogram model. Results A total of 111 patients were included in the study. Results the maximum value of PSAD in diagnosing prostatic high grade tumors was 0.84% CI: 0.770.90%. Multivariate Logistic regression analysis showed that the patients suffered from the disease. The age of the patients was 1.1095 CI1: 1.01C 1.20 P0. 023a PI-RADS v2 score: 3.05% 95 CI: 1.705.49% P0.001, prostate volume OR0.9695 CI0.99P0.020) as an independent predictor of high grade neoplasms. The fitting ROC curve AUC reached 0.92% 0.87% 0.97%. The model was used to classify 82% cases accurately. Conclusion based on the patient's age and PI-RADS v2, the diagnostic accuracy of the prostatic high-grade tumor prediction model established by prostate volume is obviously improved, which is worth popularizing.
【作者单位】： 成都大学附属医院放射科;成都大学附属医院中心实验室;
【基金】：2015年成都市卫计委医学科研课题(编号:2015080) 2017年四川省卫计委科研课题(编号:17PJ430)
【分类号】：R445.2;R737.25
，

本文编号：1551388