胶质细胞原纤维酸性蛋白诊断脑缺血性疾病的Meta分析

通过Meta分析探讨胶质细胞原纤维酸性蛋白（GFAP）在鉴别脑出血（ICH）及缺血性脑卒中（IS）的诊断价值。方法检索Pubmed、Web of Science、Embase、万方中文科技期刊数据库、中国知网数据库,筛选有关GFAP对ICH和IS 诊断的研究文献,检索时间均由建库至2016年12月31日。提取数据后计算合并敏感性、特异性、诊断比值比（DOR）、阳性似然比（PLR）、阳性似然比（NLR）及对应的95%CI;异质性采用I2检验分析,绘制汇总受试者工作特征曲线,并计算曲线下面积（AUC）。采用Stata 12.0 软件绘制漏斗图,评价纳入文献的发表性偏倚。采用Metadisc 1.4 软件对数据进行分析。结果纳入9 篇文献,合计1057 例。Meta 分析结果显示,GFAP诊断汇总敏感性为74%（68%～80%）,汇总特异性为91%（88%～93%）,汇总PLR及NLR分别为11.63（4.75～28.50）和0.30（0.17～0.51）,DOR的95%CI：（21.51,106.40）,AUC 为0.9109。结论GFAP 对于鉴别ICH 和IS 有较高的诊断价值,值得临床推广应用。

Value of serum GFAP in differential diagnosis of intracerebral hemorrhage and ischemic stroke: a systematic review and Meta-analysis

To investigate the value of glial fibrillary acidic protein (GFAP) in the differential diagnosis of intracerebral hemorrhage (ICH) and ischemic stroke (IS). Methods Pubmed, Web of Science, Em-base, Chinese Wanfan databases and CNKI were retrieved. Studies about GFAP for diagnosis of cerebral hem-orrhage and ischemic stroke were included. The retrieval time was from the date of establishment of the databases until December 2016. Software Metadisc 1.4 was used to analyze data, and calculate SE, specific SP and DOR, likelihood ratio (LR), negative LR (NLR) and 95% CI. Heterogeneity was analyzed by I2 test. The summary receiver-operating characteristic curve was drawn, and the AUC was calculated. The Stata 12.0 was used to test the publication bias. Results As a result, 9 articles involving 1,057 cases were included. Meta-analysis results showed that the total diagnostic sensitivity of GFAP was 74% (68%-80%), the total specificity was 91% (88%-93%), the total positive likelihood ratio and negative likelihood ratio were 11.63 (4.75-28.50), and 0.30 (0.17-0.51), 95% confidence interval of DOR was (21.51, 106.40), and AUC was 0.9109. Conclusions GFAP is a useful biomarker for identification of IS and ICH, and thus worthy of clinical popu-larization and application.