基于大样本研究的双能量CT诊断痛风准确性的Meta分析

A Meta-analysis of the diagnostic accuracy of dual-energy CT for gout based on larger sample studies

目的本研究通过Meta分析评估基于大样本研究的双能量计算机断层扫描(dual-energy CT,DECT)对痛风的诊断效能。方法在Web of Science、Embase、PubMed、Cochrane、中国知网(CNKI)、万方和中国生物医学文献数据库(CBM)中收集DECT诊断痛风的临床试验文献,对其进行了筛选、汇总和分析。使用统计学软件对纳入文献进行质量以及偏倚风险评估,并且合并敏感性和特异性、阳性似然比、阴性似然比和诊断比值比(DOR)以及各自95%置信区间(95%CI);绘制综合受试者操作特性曲线(SROC),以获得Cochran Q指数和曲线下面积(AUC);使用敏感性分析及Meta回归探索潜在的异质性来源以及Deek’S漏斗图评估是否存在发表偏倚。结果共纳入15篇文献,汇总合并灵敏度为0.87(0.86~0.89)、特异度为0.90(0.87~0.92)、阳性似然比为8.68(5.38~13.98)、阴性似然比为0.12(0.09~0.17)、AUC为0.96(0.94~0.97),诊断比值比为78.24(41.08~149.04)。结论DECT对痛风有着较高的诊断效能,可以作为临床无创筛查方法之一,有着较好的临床应用前景。

Objective To evaluate the diagnostic efficacy of dual-energy CT(DECT)for gout based on large sample studies through Meta-analysis.Methods Clinical trial literature on DECT diagnosis of gout was collected from Web of Science,Embase,PubMed,Cochrane,China National Knowledge Infrastructure(CNKI),Wanfang,and China Biomedical Literature Database(CBM).The included literature was screened,summarized,and analyzed.Statistical software was used to assess the quality and bias risk of the included literature,and the sensitivity,specificity,positive likelihood ratio,negative likelihood ratio,diagnostic odds ratio(DOR),and 95%confidence intervals(95%CI)were combined.The summary receiver operating characteristic curve(SROC)was plotted to obtain the Cochran Q index and the area under the curve(AUC);Sensitivity analysis and Meta-regression were used to explore potential sources of heterogeneity,and Deek’s funnel plot was used to evaluate the presence of publication bias.Results A total of 15 literatures were included.The combined sensitivity was 0.87(0.86~0.89),specificity was 0.90(0.87~0.92),positive likelihood ratio was 8.68(5.38~13.98),negative likelihood ratio was 0.12(0.089~0.17),AUC was 0.96(0.94~0.97),and DOR was 78.24(41.08~149.04).Conclusion DECT demonstrates high diagnostic efficacy for gout and can be considered as a non-invasive screening method with promising clinical applications.