1 Melbourne Brain Centre at The Royal Melbourne Hospital, University of Melbourne, Australia.
2 Department of Neuroscience, Eastern Health, Australia.
Int J Stroke. 2019 Jun;14(4):390-399. doi: 10.1177/1747493018806167. Epub 2018 Oct 10.
Glial fibrillary acidic protein (GFAP) has shown promise in several studies for its ability to diagnose intracerebral hemorrhage (ICH). We evaluated the diagnostic accuracy of blood GFAP level to differentiate (ICH) from acute ischemic stroke (AIS) and stroke mimics, both overall, and in the first three hours after symptom onset.
We searched multiple databases, without language restriction, from inception until December 2017. Hierarchical summary receiver operating characteristic (HSROC) modeling was used to meta-analyze results. We conducted subgroup analyses restricted to blood samples collected within 0-60, 60-120, and 120-180 min time groups after symptom onset, to evaluate diagnostic accuracy in the early pre-hospital phase. Between and within study heterogeneity was explored using meta-regression.
The search identified 199 potentially relevant citations from which 11 studies involving 1297 participants (350 ICH, 947 AIS, or mimic) were included. The pooled sensitivity, specificity, and area under the HSROC curve were 0.756 (95% CI 0.630-0.849), 0.945 (95% CI 0.858-0.980), and 0.904 (95% CI 0.878-0.931), respectively. Differences in assays used, but not the other covariates, partially explained between-study heterogeneity ( = 0.034). The summary estimates for the 0-60, 60-120, and 120-180 min subgroups were comparable to the primary analysis and there was no statistically significant difference in diagnostic accuracy between subgroups.
GFAP is a promising diagnostic biomarker for ICH diagnosis in the early pre-hospital phase. Test accuracy is affected by assay subtype, but there are still unexplained sources of heterogeneity. High quality, international multi-center trials are warranted to develop and validate a point-of-care GFAP assay for the rapid triage and evaluation of acute stroke in the pre-hospital setting.
胶质纤维酸性蛋白（GFAP）在多项研究中显示出诊断脑出血（ICH）的潜力。我们评估了血液 GFAP 水平对区分（ICH）与急性缺血性中风（AIS）和中风模拟的诊断准确性，包括总体情况和症状发作后前 3 小时。
我们在没有语言限制的情况下，从成立到 2017 年 12 月，搜索了多个数据库。使用分层汇总受试者工作特征（HSROC）模型进行荟萃分析。我们进行了亚组分析，仅限于症状发作后 0-60、60-120 和 120-180 分钟时间组采集的血液样本，以评估早期院前阶段的诊断准确性。使用荟萃回归探索了组间和组内异质性。
搜索确定了 199 篇潜在相关引文，其中包括 11 项研究，涉及 1297 名参与者（350 例 ICH、947 例 AIS 或模拟）。汇总的敏感性、特异性和 HSROC 曲线下面积分别为 0.756（95% CI 0.630-0.849）、0.945（95% CI 0.858-0.980）和 0.904（95% CI 0.878-0.931）。使用的检测方法存在差异，但其他协变量不存在差异，部分解释了组间异质性（ = 0.034）。0-60、60-120 和 120-180 分钟亚组的汇总估计值与主要分析相当，并且亚组间的诊断准确性没有统计学差异。
GFAP 是早期院前阶段 ICH 诊断的有前途的诊断生物标志物。测试准确性受检测亚型的影响，但仍存在未解释的异质性来源。需要高质量、国际多中心试验来开发和验证一种即时 GFAP 检测，以便在院前环境中快速分诊和评估急性中风。
