目的应用Meta分析法综合评价X线胸片在肺癌筛查中的价值。方法在PubMed、Cochrane、ISIWebofKnowledge等数据库检索1996年至2011年期间的相关文献资料;根据纳入标准筛选符合条件的文献;对文献进行信息统计、科学性评估及质量分级;采用随机效应模型进行数据分析;通过似然比、汇总灵敏度和特异度及汇总受试者工作特征曲线(SROC曲线)等统计指标综合评价X线胸片在肺癌筛查中的价值。结果共纳入9篇文献,其中英文6篇,中文3篇,由8个独立研究构成,研究对象94490例。异质性检验提示无阈值效应,但存在其他原因导致的异质性,按随机效应模型进行Meta分析得出X线胸片在肺癌筛查中价值的汇总阳性似然比为9.13,汇总阴性似然比为0.72,汇总灵敏度为0.250(0.219~0.282),汇总特异度为0.932(0.930~0.933),SROC曲线下面积为0.8536。结论X线胸片对肺癌具有一定的诊断价值,但其敏感度为25%,不适用于筛查。
Objective To evaluate the value of chest X ray in screening for lung cancer by Meta analysis. Methods The relevant documentation from PubMed, Cochrane, ISI Web of Knowledge and other databases from 1996 to 2011 was searched. The eligible literatures according to the inclusive criteria were screened. The literatures were gathered as statistical information to assess the quality of scientific and classified. The random effect model was used to analyze data. The value of ehest X ray in screening for lung cancer was comprehensively evaluated by the likelihood ratio, the pooled sensitivity, specificity and summary receiver operating characteristic (SROC) curve. Results Nine literatures were colleeted, including six in English and three in Chinese, with 8 independent researches and a total of 94 490 cases were brought into the study. Heterogeneity was found except for threshold effect. A Meta-analysis was performed using the random effect model, and the value of the positive likelihood ratio was 9.13, negative likelihood ratio was 0.72, the weighted sensitivity was 0. 250 (0. 219-0. 282), the weighted specificity was 0. 932 (0. 930-0. 933), and SROC area under curve was 0. 853 6 by using chest X-ray in screening lung cancer. Conclusions Chest X ray has certain diagnosis value for lung cancer, but it is not applied to screen for lung cancer because the sensitivity is only 25%.