糖皮质激素对儿童紫癜性肾炎预防作用的Meta分析

Effect of glucocorticoids in preventing Henoch-Schonlein purpura nephritis in children:a meta-analysis

目的评价糖皮质激素对儿童紫癜性肾炎（Henoch-Sch？nlein purprua,HSPN）的预防作用。方法检索 Cochrane Library、Embase、PubMed、Medline、Web of Science、Google Scholar、Springer、中国期刊全文数据库（CNKI）、中文科技期刊全文数据库库（VIP）、万方数据库等,收集随机对照试验文献,检索文献时间截止到2013年12月31日。对文献进行资料提取和文献质量评价,应用RevMan 5.2软件进行 Meta分析。结果6篇随机对照试验进入 Meta分析。其中多中心前瞻性随机对照试验研究（Ⅰa级）5篇,单中心前瞻性随机对照试验研究（Ⅰb 级）1篇。5篇文献描述了具体随机化方法,采用了充分的分配隐藏和盲法,1篇文献未提及具体随机化方法和分配隐藏,未采用盲法。纳入6篇文献均报道了失访和退出情况,5篇为低度偏倚风险,1篇为高度偏倚风险。Meta分析结果显示,诊断过敏性紫癜（Henoch-Sch？nlein purpura,HSP）后糖皮质激素预防组 HSPN 发生率为16.9%（89/527）,对照组 HSPN发生率为15.9%（84/527）,合并相对危险度 RR=1.07[95%可信区间（95%CI）：0.83~1.39],P=0.61,差异无统计学意义,进行敏感性分析,结果无改变。结论本Meta分析结果尚不支持糖皮质激素能预防儿童 HSPN 的发生,并有待于大规模、多中心、前瞻性随机对照试验进一步证实。

Objective To evaluate the effect of glucocorticoids （GC）in preventing Henoch-Sch？nlein purprua nephritis （HSPN）in children.Methods Cochrane Library,Embase,PubMed, Medline,Web of Science,Google Scholar,Springer,National Knowledge Infrastructure （CNKI）of China,Chongqing VIP Database and Wanfang Database were searched for randomized controlled trials （RCTs）by December 31,2013.Literature was evaluated and the Meta analysis was performed by Rev-Man5 .2 software.Results Six RCTs were included in this study,including 5 articles of multi-centers RCT （Ⅰa）,and 1 article of single center RCT （Ⅰb）.Five articles described the specific randomized method and adopted adequate allocation concealment and blinding,and 1 article did not.All of them reported loss of follow up and withdrawal.Five articles had lower risk of bias,while 1 article had high risk of bias.527 cases were treated with glucocorticoids and 89 of them showed HSPN.In control group,84 of 527 cases showed HSPN.No significant difference was observed between two groups （RR=1.07,95% CI：0.83-1.39,P=0.61）.For sensitivity analysis,the same result was found as above.Conclusions Early glucocorticoids do not prevent Henoch-Sch nlein purprua nephritis in chil-dren.However,prospective RCTs from large scale and multi-centers should be adopted to further confirm the conclusion.