基于CT影像的虚拟膀胱镜技术对膀胱肿瘤诊断价值的Meta分析

Accuracy of virtual cystoscopy based on CT images for differentiating bladder cancer lesions and normal tissues: a meta-analysis

目的:评价基于CT影像三维重建的虚拟膀胱镜技术对于膀胱肿瘤病灶及正常组织的鉴别诊断价值。方法:检索PubMed、Embase、Cochrane、Medline数据库中关于CT影像、虚拟膀胱镜及膀胱癌的英文文献,由2名研究人员独立检索文献并提取数据,有异议时交由第三位研究人员进行评定。然后根据纳入排除标准筛选纳入研究的全文,最终根据QUADAS量表评价纳入文献的质量,采用Stata14.0软件和R语言(v4.1.1)进行数据处理和Meta分析,计算合并后的灵敏度及特异度,绘制受试者工作特征曲线并计算曲线下面积。结果:共纳入12篇文献,合计786个病灶。Meta分析显示,合并效应量后的灵敏度为0.96(95%CI:0.92~0.98),合并后的特异度为0.92(95%CI:0.82~0.97),合并后的基于CT影像的虚拟膀胱镜对于膀胱肿瘤病灶的诊断效能曲线下面积为0.98(95%CI:0.96~0.99)。结论:基于CT影像三维重建的虚拟膀胱镜技术对于膀胱肿瘤病灶及正常组织或良性病变的诊断有较高的应用价值。

Objective:To evaluate the value and accuracy of virtual cystoscopy based on three-dimensional reconstruction of CT images in the differential diagnosis of bladder tumor lesions and normal tissues.Methods:The English literatures in PubMed,EMBASE,Cochrane and MEDLINE databases were searched by the following keywords:CT images,virtual cystoscopy and bladder cancer.Two researchers independently searched the literatures and extracted the data.If there was any objection,they were handed over to the third researcher for evaluation.Following this,included studies were screened according to the inclusion and exclusion criteria,and their quality was assigned by the QUADAS scale.Finally,Stata14.0 software and R language(v4.1.1)were used for data processing and meta-analysis to calculate the combined sensitivity and specificity,draw the receiver operating characteristic(ROC)curve,and calculate the area under the curve(AUC).Results:A total of 12 literatures were included,with a total of 786 lesions.Meta analysis showed that the sensitivity of the combined effect was 0.96(95%CI:0.92-0.98),the specificity of the combined effect was 0.92(95%CI:0.82-0.97),and the AUC of the combined virtual cystoscope based on CT image for bladder tumor lesions was 0.98(95%CI:0.96-0.99).Conclusion:Virtual cystoscopy based on three-dimensional reconstruction of CT images has high application value in the diagnosis of bladder tumor lesions and normal or benign lesions.