儿童腺病毒感染与脑炎相关性的Meta分析

Correlation between Adenoviral Infection and Encephalitis in Children: A Meta-Analysis

目的系统评价儿童腺病毒感染与脑炎之间的相关性。方法计算机检索CHKD、CNKI、WanFangData、CMCC、PubMed、Ovid、EBSCO和ScienceNew，查找儿童腺病毒感染与脑炎相关性的研究，检索时限为1978～2013年。按照纳入与排除标准进行文献筛选和提取资料，然后采用Comprehensive MetaAn出sis2．0软件进行Meta分析。结果最终纳人18个研究。Meta分析结果显示：脑炎患儿腺病毒的感染率为P2=10．8％[95％CI（6．5％～15．2％），P〈0．00001]，人腺病毒感染儿童中脑炎的发病率为P2=31．4％[95％CI（14．1％～48．6％），P〈0．00001o2个研究结果分别显示，人腺病毒与儿童脑炎之间的相关性为OR1=2．22[95％CI（0．32，14．79），P〉0．05]和OR2=1．12[95％CI（0．14，9．10），P〉0．05]，其差异均无统计学意义。与儿童脑炎相关的人腺病毒以B组中的3型和7型为主，而C组中的1型、2型和D组中的8型次之。结论当前证据表明人腺病毒感染与儿童脑炎不存在相关性，上述结论尚需开展更多设计合理的研究，以明确二者的相关性。

Objective To systematically review the correlation between adenoviral infection and encephalitis in chil- dren. Methods Studies on the correlation between adenoviral infection and encephalitis among children were retrieved in the databases of CHKD, CNKI, WanFang Data, CMCC, PubMed, Ovid, EBSCO, and Science New during 1978-2013. According to our inclusion and exclusion criteria, literature was screened. After data extraction, meta-analysis was conducted using Comprehensive Meta Analysis 2.0. Results A total of 18 studies were enrolled. The results of meta-analysis indicated that： the prevalence of infection of human adenovirus in children with encephalitis was p~=10.8% （95%CI 6.5% to 15.2%, P〈0.000 01）. The incidence of encephalitis in children with adenoviral infection was p2=31.4% （95%CI 14.1% to 48.6%, P〈0.000 01）. The results of two studies showed that, the correlation coefficients between adenoviral infection and encephalitis were OR1=2.22 （95%CI 0.32 to 14.79, P〉0.05）, and OR2=1.12 （95%CI 0.14 to 9.10, P〉0.05）, respectively, both with no signifi- cant difference. In addition, the subgenus B adenoviruses, including type 3 and type 7 were more associated with encephalitis, followed by type I and type 2 in subgenus C and type 8 in subgenus D. Conclusion Current evidence shows that no correlation exists between adenoviral infection and encephalitis in children. However, the above conclusion needs to be verified by conducting more rationally-designed studies so as to clarify the correlation.