中国结核分枝杆菌／艾滋病病毒双重感染率分析

Meta analysis on the co-infection between Mycobacterium tuberculosis and HIV/AIDS in China

目的了解国内文献报道的结核分枝杆菌（MTB）和艾滋病病毒（HIV）双重感染率。方法系统检索PubMed、中国期刊全文数据库（CNKI）、万方数据资源系统和维普中文科技期刊全文数据库中2000—2012年有关MTB／HIV双重感染率的研究文献，摘录有关信息，采用广义倒方差模型估计双重感染率，并对性别、样本来源和地区进行亚组分析，使用观察性研究质量评价准则STROBE声明评价文献质量，利用敏感性分析评价研究质量对结果的影响。结果共纳入39篇文献，总样本量为168286人，利用广义倒方差模型估计结核病患者中HIV双重感染率为0．86％（95％C1：0．80％～0．93％）。亚组分析显示，MTB／HIV双重感染率男性（1．53％）高于女性（0．54％），艾滋病病毒感染者和艾滋病患者（H／V／AIDS）高发地区（0．91％）高于低发地区（0．71％），差异均有统计学意义。HIV／AIDS中结核双重感染率为14．44％（95％CI：13．62％～15．30％）。亚组分析显示，MTB／HIV双重感染率医院来源（28．64％）高于社会人群（4．74％），西部（15．79％）高于中部（13．88％）及东部地区（6．47％），差异均有统计学意义。敏感性分析中在删除了符合条目数少的文献后，Meta分析的结果与原结果接近。结论中国MTB／HIV双重感染率较高。

Objective To understand and estimate the rate of co-infection between Mycobacterium tuberculosis （M. tuberculosis） mad the human immunodeficiency virus （HIV）. Methods Based on the related literature regarding Chinese M. tuberculosis and HIV co-infection being published between 2000 and 2012 in PubMed, China National Knowledge Infrastructure （CNKI）, Wan Fang Databases and the Chinese Science ＆ Technology Journal Database （VIP） , we extracted related information and applied the generic inverse variance model to estimate the following parameters as： co-infection rate, differences on gender, region and subgroups. We also used the STROBE Statement and observational quality evaluation standard to evaluate the quality of literature, and employed sensitivity analyses to evaluate the impact of research quality. Results 39 papers were included in this Meta-analysis, which contained a total sample size of 168 286. The co-infection rate of HIV in TB patients was 0.86% （95%CI： 0.80%-0.93%）, when using the generic inverse variance model. According to subgroups analyses, at α =0.05 level, men had a higher co-infection rate （1.53%） than women （0.54%） and areas with high infection rate of HIV/AIDS （0.91%） were more commonly seen than low infection rate areas of HIV/AIDS （0.71% ） and with significant differences. The co-infection rate of M. tuberculosis in patients infected with HIV was 14.44% （95% CI： 13.62%-15.30% ）. Data from subgroup analyses also showed that at the α =0.05 level, the co-infection rates from hospital samples （28.64%） were higher than those from the population samples＇ （4.74%）. The co-infection rate （15.79%） from the western region was higher than that in the Central and Eastern regions （13.88% and 6.47% respectively）, with significant differences. After excluding the papers that only met a few items of the STRODE Statement, results of the Meta-analysis were very close to the original results. Conclusion The co-infection rate of both M. tuberculosis and HIV was high in China. Therefore, it was crucial for people and related organizations to pay more attention to this problem of co-infection.