摘要
目的探讨基于多参数磁共振成像(mp-MRI)的膀胱影像报告和数据系统(VI-RADS)评分对肌层浸润性膀胱癌(MIBC)的诊断效能。方法系统检索2018年9月至2023年12月PubMed、Web of Science和Embase数据库中与采用VI-RADS诊断MIBC相关的文献。纳入标准:研究中基于mpMRI的VI-RADS评分用于确定MIBC。排除标准:研究对象<10例、研究人群重叠、未测试VI-RADS总分在确定MIBC时的诊断效能等。经文献质量评价后,使用RevMan 5.4和Stata 15.1软件计算纳入研究VI-RADS评分诊断效能的合并敏感性和特异性,绘制森林图和层次综合受试者操作特征(HSROC)曲线,计算曲线下面积(AUC)评估VI-RADS评分的诊断效能。使用Deeks漏斗图评价发表偏倚。采用I 2评估文献间的异质性,并运用回归分析和亚组分析探讨异质性的来源。结果本研究共纳入29项研究,共3577例患者。当分界点为VI-RADS=3分时,VI-RADS预测MIBC的汇总敏感性和特异性分别为93%(95%CI 0.90~0.95)和82%(95%CI 0.76~0.88);当分界点为VI-RADS=4分时,汇总敏感性和特异性分别为83%(95%CI 0.78~0.87)和93%(95%CI 0.90~0.95)。VI-RADS=3分和VI-RADS=4分的HSROC曲线的AUC分别为0.95和0.94。亚组分析结果显示,当分界点为VI-RADS=3分时,纳入患者例数、研究设计、MRI场强、放射科医生数量、手术方式、扩散加权成像(DWI)和动态对比增强(DCE)的图像采集平面均导致敏感性的异质性(P<0.05);除研究设计类型和DWI图像采集平面外,其余因素均为导致同一分界点特异性异质性的因素(P<0.05)。当分界点为VI-RADS=4分时,所有纳入的因素都是导致敏感性和特异性异质性的原因(P<0.05)。Meta回归分析结果显示,分界点VI-RADS=3分和VI-RADS=4分均是敏感性和特异性异质性的重要来源(P<0.05)。结论VI-RADS评分诊断MIBC的效能在分界点为VI-RADS=3分或VI-RADS=4分时均能获得较好的敏感性和特异性,且VI-RADS=3分的诊断效能优于VI-RADS=4分。
Objective To evaluate the diagnostic performance of the Vesical Imaging-Reporting and Data System(VI-RADS)based on multiparametric magnetic resonance imaging(mp-MRI)for muscle-invasive bladder cancer(MIBC).Methods A systematic search was conducted in PubMed,Web of Science,and Embase databases for studies published between September 2018 and December 2023 that investigated the use of VI-RADS for diagnosing MIBC.Inclusion criteria were studies utilizing mp-MRI-based VI-RADS scoring to determine MIBC.Exclusion criteria were studies with fewer than 10 patients,overlapping study populations,or those failing to assess the diagnostic performance of VI-RADS for MIBC.After quality assessment,RevMan 5.4 and Stata 15.1 were used to calculate pooled sensitivity and specificity,generate forest plots and summary receiver operating characteristic(SROC)curves,and determine the area under the curve(AUC).Publication bias was assessed using Deeks funnel plot.Heterogeneity was evaluated using the I 2 statistic,with meta-regression and subgroup analyses to explore its sources.Results Twenty-nine studies involving 3577 patients were included.At a VI-RADS cutoff of 3,the pooled sensitivity and specificity for MIBC diagnosis were 93%(95%CI 0.90-0.95)and 82%(95%CI 0.76-0.88),respectively.At a cutoff of 4,these values were 83%(95%CI 0.78-0.87)and 93%(95%CI 0.90-0.95).The hierarchical SROC(HSROC)AUCs were 0.95 and 0.94 for cutoffs of 3 and 4,respectively.Subgroup and meta-regression analyses revealed that at a cutoff of 3,patient sample size,study design,MRI field strength,number of radiologists,surgical approach,and DWI/DCE imaging planes contributed to sensitivity heterogeneity(P<0.05).All factors except study design and DWI plane were sources of specificity heterogeneity(P<0.05).At a cutoff of 4,all factors significantly influenced heterogeneity in both sensitivity and specificity(P<0.05).Meta-regression confirmed that both cutoffs(3 and 4)were significant sources of heterogeneity(P<0.05).Conclusions VI-RADS demonstrates excellent diagnostic performance for MIBC at both cutoffs(3 and 4),with VI-RADS≥3 showing superior sensitivity and VI-RADS≥4 offering higher specificity.The cutoff of 3 provides better overall diagnostic efficacy.
作者
余青洋
陈康康
赵烔蕾
毛卫浦
王泽军
彭新阳
赵子慧
彭新桂
陈明
吴剑平
Yu Qingyang;Chen Kangkang;Zhao Tonglei;Mao Weipu;Wang Zejun;Peng Xinyang;Zhao Zihui;Peng Xingui;Chen Ming;Wu Jianping(Department of Urology,Zhongda Hospital Southeast University,Nanjing 210009,China;Clinical Medicine,School of Medicine,Southeast University,Nanjing 210003,China;Department of Radiology,Zhongda Hospital Southeast University,Nanjing 210009,China;Department of Urology,Zhongda Hospital Lishui Branch,Southeast University,Nanjing Lishui People’s Hospital,Nanjing 211299,China;School of Medicine,Southeast University,Nanjing 210003,China)
出处
《中华泌尿外科杂志》
北大核心
2025年第6期430-438,共9页
Chinese Journal of Urology
基金
“十四五”江苏省医学重点学科(2023YXZDXK02)。
关键词
膀胱癌
多参数磁共振
膀胱成像报告和数据系统
肌肉侵犯
荟萃分析
Bladder cancer
Multi-parameter magnetic resonance imaging
Vesical imaging-reporting and data system
Muscle invasion
Meta-analysis
