临床征象对识别儿童社区获得性肺炎支原体肺炎价值的系统综述

摘要

目的 评估临床征象在识别儿童社区获得性肺炎支原体肺炎中的价值.方法 通过检索从建库至2015年8月Cochrane图书馆、PubMed、万方、CNKI及维普数据库,按照纳入和排除标准筛选文献,提取纳入文献中的相关信息,并根据QUADAS评价系统制定质量评价标准,对纳入的各项研究进行质量评价.从纳入的文献中分析社区获得性肺炎支原体肺炎与非肺炎支原体肺炎在咳嗽、发热、喘息等临床表现以及相关胸X线片表现的不同点.采用Cochrane协作网提供的RevMan5.3软件进行Meta分析,并按要求整理输入数据,完成敏感度、特异度的汇总分析及相应的95％的置信区间(CI)并绘制森林图.进而得出阴性似然比和阳性似然比、诊断比值比(DOR)、验前概率和验后概率及相应的95％ CI:并进行汇总分析.结果 共纳入文献1 1篇.汇总胸痛、湿啰音、头痛、胸腔积液、肺实变和肺气肿等临床征象的真阳性(TP)和假阳性(FP)的病例数:胸痛:TP 287例,FP 1 090例;湿啰音:TP l 906例,FP 6 886例;头痛:TP 590例,FP 2 051例;胸腔积液:TP 10例,FP 16例;肺实变:TP 75例,FP 83例;肺气肿:TP 433例,FP 116例;各临床征象对识别肺炎支原体肺炎的汇总敏感度、汇总特异度、DOR和95％ CI:分别为:胸痛:汇总敏感度0.12,95％ CI:0.10,0.13,汇总特异度0.89,95％ CI:0.88,0.90,DOR:1.05,95％ CI:0.92,1.21;湿啰音:汇总敏感度0.66,95％ CI:0.64,0.67,汇总特异度0.36,95％ CI:0.88,0.90,DOR:1.12,95％ CI:1.02,1.22;头痛:汇总敏感度0.23,95％ CI:0.21,0.25,汇总特异度0.80,95％ CI:0.79,0.80,DOR:1.16,95％ CI:1.05,1.29;胸腔积液:汇总敏感度0.04,95％ CI:0.02,0.08,汇总特异度0.98,95％ CI:0.96,0.99,DOR:1.28,95％ CI:0.56,2.89;肺实变:汇总敏感度0.32,95％ CI:0.26,0.39,汇总特异度0.87,95％ CI:0.84,0.90,DOR:1.88,95％ CI:1.23,2.90;肺气肿:汇总敏感度0.22,95％ CI:0.17,0.29,汇总特异度0.73,95％ CI:0.69,0.77,DOR:1.05,95％ CI:0.68,1.61.结论 临床征象对识别儿童社区获得性肺炎支原体肺炎的意义不大.虽然本研究数据分析显示胸痛、头痛、湿啰音等症状体征及胸腔积液、肺实变和肺气肿等胸X线表现可以提示肺炎支原体肺炎的可能性,但是其中任何一项临床征象的存在或缺失都不能作为肯定或否定肺炎支原体肺炎的识别依据.%Objective To evaluate the value of clinical signs in the identification of Mycoplasma pneumonia in children's community acquired pneumonia.Method We searched the Cochrane library,PubMed,CNKI,Wan Fang and VIP databases.According to the inclusion and exclusion criterias,we selected and extracted the related information in the literature.According to the QUADAS evaluation system,we established the quality evaluation standard to evaluate the quality of the included studies and analyzed the difference of the clinical manifestations between Mycoplasmae pneumoniae and non-Mycoplasma pneumoniae in children's community acquired pneumonia.We used the RevMan 5.3 software to do the meta-analysis and collected the data according to the requirements.We calculated the pooled sensitivities,specificities and 95％ CIs.Then we calculated the negative and positive likelihood ratio,the ratio of the diagnosis and the pre-/post-test probabilities with 95％ CIs.Result A total of 11 articles were included in the literature.In summary,the cascs of the clinical signs of true positive (TP) and false positive (FP) were as follows:chest pain:TP:287,FP:1090;rales:TP:1906,FP:6886;headache:TP:590,FP:2051;pleural effusion:TP:10,FP:16;consolidation:TP:75,FP:83;emphysema:TP:443,FP:116.The pooled sensitivity,the pooled specificity,the diagnostic ratio (DOR) and 95％ CI were:chest pain:pooled sensitivity:0.12,95％ CI:0.10-0.13,pooled specificity:0.89,95％ CI:0.88-0.90,DOR:1.05,95％ CI:0.92-1.21;rales:pooled sensitivity:0.66,95％ CI:0.64,0.67,pooled specificity:0.36,95％ CI:0.35,0.37,DOR:1.12,95％ CI:1.02,1.22;headache:pooled sensitivity:0.23,95％ CI:0.21-0.25,pooled specificity:0.80,95％ CI:0.79-0.80,DOR:1.16,95 ％ CI:1.05-1.29;pleural effusion:pooled sensitivity:0.04,95％ CI:0.02,0.08,pooled specificity:0.98,95％ CI:0.96,0.99,DOR:1.28,95％ CI:0.56,2.89;consolidation:pooled sensitivity:0.32,95％ CI:0.26,0.39,pooled specificity:0.87,95％ CI:0.84,0.90,DOR:1.88,95％ CI:1.23,2.90;emphysema:pooled sensitivity:0.22,95％ CI:0.17,0.29,pooled specificity:0.73,95％ CI:0.69,0.77,DOR:1.05,95％ CI:0.68,1.61.Conclusion The value of clinical symptoms and signs in the identification of mycoplasma pneumonia in children's community acquired pneumonia was not significant.Although the clinical symptoms/signs of chest pain,headache,rales and chest X-ray manifestations of pleural effusion,consolidation,emphysema could suggest Mycoplasma pneumoniae infection,the presence or absence of any clinical signs were not positive or negative indicators for the identification of Mycoplasma pneumoniae infections.