双酚A暴露与儿童青少年肥胖关系的Meta分析

Meta analysis of the association of Bisphenol A with obesity in children and adolescents

目的采用Meta分析系统评价双酚A(Bisphenol A,BPA)暴露与儿童青少年肥胖的关系,为预防和控制儿童青少年肥胖的发生提供思路。方法全面检索Medline、Web of Science、中国生物医学文献数据库、中国知网数据库、维普数据库和万方数据库等中文和外文数据库,查找国内外关于BPA暴露与儿童青少年肥胖关系的中英文文献。使用Stata12.0进行Meta分析,选择标准化均数差(standardized mean difference,SMD)及其置信区间和随机效应模型合并效应值。Q检验和I2统计量用于评价研究的异质性,漏斗图、Egger和Begg检验评价发表偏倚。结果共纳入13项研究,3 932名研究对象,其中病例组1 046例,对照组2 886人。尿液中BPA水平与肥胖呈正相关(SMD=0.26,95%CI=0.01~0.51),而血清中BPA水平与肥胖不存在相关性(SMD=0.07,95%CI=-0.44~0.58)。各研究之间存在高度异质性(I^2=87. 3%,P <0.01)。亚组分析发现,美国(SMD=0.27,95%CI=0. 16~0. 39)和欧洲(SMD=3. 76,95%CI=2. 77~4. 74)地区的研究显示BPA暴露与肥胖正相关,而亚洲地区的研究显示肥胖儿童青少年与正常儿童青少年差异无统计学意义(SMD=0.10,95%CI=-0.08~0.27)。敏感性分析表明,Meta分析合并结果稳定,漏斗图、Egger检验和Begg检验均未发现发表偏倚(P值均>0.05)。结论 BPA暴露与儿童和青少年肥胖有关,BPA可能会促进儿童和青少年肥胖的发生。

Objective To evaluate the relationship between the bisphenol A( BPA) exposure and obesity in children and adolescents with Meta-analysis, and to provide a new idea for prevention and control of obesity. Methods Reviewer independently searched and screened potentially relevant studies that reported the association between BPA and obesity in children and adolescents through Medline, Web of Science, CBM, CNKI, WANFANG DATA and CQVIP to March 7, 2019. Chi-square based Q test and I^2 statistic were used to assess the heterogeneity. Forest plot was used to display results graphically. Potential publication bias was assessed by the Egger and Begg tests and funnel plot. Results In total, 13 studies with 3 932 participants met the eligibility criteria,13 articles were indentified involving 1 046 obesity children and 2 886 normal control children. Overall summary estimates indicated significant positive correlation between urinary BPA and obesity( SMD = 0.26,95%CI = 0.01-0.51), and a nonsignificant association between serum BPA and obesity( SMD = 0.07,95% CI =-0.44-0.58). Significant heterogeneity was observed in overall estimates( I^2= 87.3%, P<0.01). In additon, a significant positive association was estimated in American( SMD = 0.27, 95%CI = 0.16-0.39) and Europe( SMD = 3.76, 95 CI = 2.77-4.74). The association between BPA and obesity was not changed in each individual( sensitivity analysis by leaving one out approach). The funnel plots did not reveal severe asymmetry, and both Egger and Begg tests were non-significant( P>0.05);therefore, publication bias was assessed as "undetected". Conclusion The finding from our systematic review and Meta-analysis suggest that children and adolescents exposure to BPA may result in obesity.