CT影像组学预测肾透明细胞癌病理分级:Meta分析

CT radiomics for predicting pathological grade of renal clear cell carcinoma:Meta-analysis

目的采用Meta分析观察CT影像组学预测肾透明细胞癌(ccRCC)病理分级的价值。方法检索建库至2021年1月PubMed、WebofScience、EMbase及中国知网、中国生物医学文献服务系统和万方医学网CT影像组学预测ccRCC病理分级相关文献,并进行筛选、质量评价及资料提取;以Stata16.0软件行Meta分析。结果纳入16篇文献、2489例患者共2495个ccRCC病灶。CT影像组学预测ccRCC病理分级无明显阈值效益(r=0.12,P<0.01)而具有较高异质性(I^(2)≥50%),其合并敏感度0.85[95%CI(0.80,0.89)]、合并特异度0.86[95%CI(0.81,0.90)],阳性似然比6.00[95%CI(4.30,8.30)]、阴性似然比0.18[95%CI(0.13,0.24)]、诊断比值比34.00[95%CI(20.00,57.00)],曲线下面积0.92。结论CT影像组学预测ccRCC病理分级效能较佳。

Objective To observe the value of CT radiomics for predicting pathological grade of clear cell renal cell carcinoma(ccRCC)with meta-analysis.Methods Literature concerning predicting pathological grade of ccRCC based on CT radiomics in the PubMed,Web of Science,EMbase,CNKI,SinoMed and Wanfang Med Online were searched from the time of establishment to January 2021.Literature screening,quality evaluation and data extraction were performed.Stata 16.0 was used for meta-analysis.Results A total of 16 articles were enrolled,including 2489 patients with 2495 ccRCC lesions.There was no significant threshold benefit in predicting the pathological grade of ccRCC based on CT radiomics(r=0.12,P<0.01)but was high heterogeneous(I^(2)≥50%).The combined sensitivity was 0.85(95%CI[0.80,0.89]),combined specificity was 0.86(95%CI[0.81,0.90]),positive likelihood ratio was 6.00(95%CI[4.30,8.30]),negative likelihood ratio was 0.18(95%CI[0.13,0.24]),the diagnostic odds ratio was 34.00(95%CI[20.00,57.00])and area under the curve was 0.92.Conclusion CT radiomics was effective for predicting pathological grade of ccRCC.