目的·采用系统评价和meta分析的方法评价呼出气一氧化氮（fractional exhaled nitric oxide，FeNO）检测对儿童咳嗽变异性哮喘（cough variant asthma，CVA）的诊断价值。方法·系统检索7个数据库（PubMed、Embase、Cochrane Library、Web of Science、中国生物医学文献服务系统、中国知网、万方数据库）建库至2022年3月1日的队列研究和病例对照研究。由2位研究人员分别独立根据纳入和排除标准对文献进行筛选，提取纳入文献中真阳性、假阳性、真阴性、假阴性的患儿人数以及FeNO的截断值。采用诊断性试验准确性质量评价工具2（diagnostic accuracy studies tool version 2，QUADAS-2）对纳入文献的风险偏倚和临床适用性2个方面进行质量评价。运用Revman 5.4软件绘制质量评价图。采用Stata 17.0统计分析软件的midas、spearman命令进行meta分析。采用灵敏度对数和（1－特异度）对数的Spearman相关系数检验阈值效应。采用双变量箱式图和I²统计量评价研究间非阈值效应异质性。以双变量随机效应模型合并效应量。以Deek漏斗图评价发表偏倚。结果·共检索到1 099篇文献，经筛选最终纳入9篇。纳入的文献均有明确的专家共识或指南作为诊断标准。QUADAS-2评估结果显示纳入文献的质量较低。灵敏度和（1－特异度）的Spearman相关系数为0.05（P=0.90），表明无阈值效应引起的异质性。双变量箱式图及I²=0的结果表明，研究间无明显的非阈值效应异质性。合并效应量结果显示，在慢性咳嗽患儿中，FeNO检测诊断CVA的灵敏度为0.82（95%CI 0.78~0.85），特异度为0.93（95%CI 0.89~0.95），阳性似然比为11.30（95%CI 7.40~17.10），阴性似然比为0.19（95%CI 0.16~0.24），诊断优势比为58（95%CI 34~101），综合受试者操作特征曲线（summary receiver operating characteristic curve，SROC曲线）下面积为0.89（95%CI 0.86~0.92）。Deek漏斗图显示不存在发表偏倚。结论·FeNO检测对儿童CVA的诊断价值呈中等程度。但受限于较低质量的原始研究证据，此结果需谨慎解读。

Objective ·To evaluate the values of fractional exhaled nitric oxide (FeNO) for the diagnosis of cough variant asthma (CVA) in children with chronic cough by systematic review and meta-analysis. Methods ·PubMed, Embase, Cochrane Library, Web of Science, China Biology Medicine Database (SinoMed), China National Knowledge Infrastructure (CNKI), and Wanfang Database were systematically searched for cohort studies and case-control studies to March 1, 2022. Two researchers independently screened the studies according to the inclusion and exclusion criteria, and the number of true positive, false positive, true negative, and false negative patients and the cut-off value of FeNO were extracted from the studies. Diagnostic accuracy studies tool version 2 (QUADAS-2) was used to evaluate the quality of the included literature in terms of risk bias and clinical applicability. Revman 5.4 was used to draw the quality evaluation chart. A meta-analysis was performed by the midas and spearman commands of Stata 17.0. Spearman correlation coefficients of sensitivity logarithm and (1-specificity) logarithm were used to test the threshold effect. The substantial heterogeneity caused by non-threshold effect was tested by bivariate box plot and I² statistic. The effect size was pooled by bivariate random effect model. Publication bias was evaluated by Deek′s Funnel plot. Results ·One thousand and ninety-nine studies were retrieved in total and 9 were finally included. The diagnostic criteria of the included study were all from expert consensus or guidelines. The QUADAS-2 showed that the quality of the included study was low. The Spearman correlation coefficient of sensitivity and (1-specificity) was 0.05 (P=0.90), indicating that there was no heterogeneity caused by threshold effect. The results ofbivariate box plot and I²=0 showed that there was no heterogeneity caused by non-threshold effect. Combined effect showed that the sensitivity was 0.82 (95%CI 0.78－0.85), the specificity was 0.93 (95%CI 0.89?0.95), the positive likelihood ratio was 11.30 (95%CI 7.40?17.10), the negative likelihood ratio was 0.19 (95%CI 0.16?0.24), the diagnostic odds ratio was 58 (95%CI 34?101), and the area under the summary receiver operating characteristic (SROC) curve was 0.89 (95%CI 0.86?0.92). The result of Deek′s Funnel plot showed that there was no publication bias. Conclusion ·FeNO test has a moderate diagnostic value in prediciting CVA in children with chronic cough. However, the results of the study need to be interpreted with caution due to the low quality of the original evidence.