摘要
背景与目的:基于多参数磁共振成像(magnetic resonance imaging,MRI)的膀胱影像报告和数据系统(Vesical Imaging-Reporting and Data System,VI-RADS)在术前评估膀胱癌肌层浸润方面的价值越来越得到认可,但当以3分为诊断阈值时仍存在较高的假阳性率。肿瘤大小在肿瘤浸润评估方面具有一定的辅助诊断价值。因此本研究旨在探究VI-RADS联合肿瘤大小用于评估膀胱癌肌层浸润方面的诊断效能。方法:回顾性收集2019年11月—2022年2月复旦大学附属肿瘤医院收治的119例经手术后病理学检查证实为膀胱癌患者(共159个病灶)的术前膀胱多参数MRI及临床资料。由两名放射科医师独立对每个病灶进行VI-RADS评分以及肿瘤基底接触长度(tumor contact length,TCL)的测量,对于评分或大小存在差异的病灶由两名医师讨论后达成一致结论。采用受试者工作特征(receiver operating characteristic,ROC)曲线分析VIRADS、TCL及两者联合模型的肌层浸润诊断效能,在计算出相应的曲线下面积(area under curve,AUC)、灵敏度、特异度、阳性预测值(positive predictive value,PPV)、阴性预测值(negative predictive value,NPV)和诊断准确度后进行比较。结果:术后病理学检查证实,非肌层浸润性膀胱癌(non-muscle-invasive bladder cancer,NMIBC)和肌层浸润性膀胱癌(muscle-invasive bladder cancer,MIBC)病灶分别为75和84个。MIBC组平均TCL(6.15~6.23 cm)与NMIBC组平均TCL(2.26~2.35 cm)存在显著差异,差异有统计学意义(P<0.05)。VI-RADS联合TCL在预测膀胱癌肌层浸润方面的特异度、PPV及诊断准确度均显著高于单独应用以3分为诊断阈值的VI-RADS(P<0.05),但灵敏度和NPV差异无统计学意义(P>0.05)。TCL(AUC=0.89)、VI-RADS(AUC=0.90)及VI-RADS联合TCL(AUC=0.91)的AUC差异无统计学意义(P>0.05)。结论:VI-RADS联合TCL能够在一定程度上降低VI-RADS 3分病灶在评价膀胱癌肌层浸润方面的假阳性率,有利于避免过度治疗。
Background and purpose:The value of Vesical Imaging-Reporting and Data System(VI-RADS)based on multiparametric magnetic resonance imaging(MRI)in the preoperative assessment of bladder cancer muscle-invasive is increasingly recognized.However,there is still a high number of false positives when the diagnostic cut-off value is 3 points.Tumor size has certain auxiliary diagnostic value in the assessment of tumor infiltration.Therefore,this study mainly explored the diagnostic performance of VI-RADS combined with tumor size in assessing bladder cancer muscle-invasive.Methods:The preoperative bladder multiparametric MRI and clinical data of 119 patients with bladder cancer confirmed by surgery and pathology(a total of 159 lesions)who were treated in Fudan University Shanghai Cancer Center from November 2019 to February 2022 were retrospectively collected.VI-RADS score and tumor contact length(TCL)measurements were performed independently for each lesion by two radiologists.Lesions with differences in score or size were given consistent results following discussion by two physicians.The receiver operating characteristic(ROC)curve was used to analyze the diagnostic performance of VI-RADS,TCL and their combined models for muscle invasion,and the corresponding area under curve(AUC),sensitivity,specificity,positive predictive value(PPV),negative predictive value(NPV)and diagnostic accuracy were compared.Results:Postoperative pathology confirmed that there were 75 and 84 lesions of non-muscle-invasive bladder cancer(NMIBC)and muscle-invasive bladder cancer(MIBC),respectively.The mean TCL of MIBC group(6.15-6.23 cm)was significantly different from that of NMIBC group(2.26-2.35 cm),and the difference was statistically significant(P<0.05).The specificity,PPV and diagnostic accuracy of VI-RADS combined with TCL in predicting bladder cancer muscle-invasive were significantly higher than those of VI-RADS with a diagnostic threshold of 3 points alone(P<0.05),whereas there was no statistically significant difference in the sensitivity and NPV(P>0.05).There was no significant difference in AUC between TCL(AUC=0.89),VI-RADS(AUC=0.90)and VI-RADS combined with TCL(AUC=0.91)(P>0.05).Conclusion:VI-RADS combined with TCL can reduce the false positive rate of VI-RADS 3-point lesions in the evaluation of bladder cancer muscle-invasive to a certain extent,which is beneficial for avoiding overtreatment.
作者
陈如串
刘伟
周冰妮
刘晓航
周良平
CHEN Ruchuan;LIU Wei;ZHOU Bingni;LIU Xiaohang;ZHOU Liangping(Shanghai Institute of Medical Imaging,Shanghai 200032,China;Department of Radiology,Fudan University Shanghai Cancer Center,Department of Oncology,Shanghai Medical College,Fudan University,Shanghai 200032,China)
出处
《中国癌症杂志》
CAS
CSCD
北大核心
2023年第3期260-266,共7页
China Oncology
关键词
膀胱影像报告和数据系统
肿瘤基底接触长度
肌层浸润性膀胱癌
磁共振成像
Vesical Imaging-Reporting and Data System
Tumor contact length
Muscle-invasive bladder cancer
Magnetic resonance imaging