经颅多普勒超声对大脑中动脉狭窄诊断价值的Meta分析

Diagnostic value of transcranial doppler ultrasonography for middle cerebral artery stenosis： a Meta-analysis

目的探讨经颅多普勒超声（TCD）对大脑中动脉狭窄的诊断价值。方法计算机检索PubMed、EMBase及The Cochrane Library数据库和中国知网、万方及维普数据库自建库至2017年1月收录的TCD诊断大脑中动脉狭窄的相关文献，筛选后提取文献的一般信息和诊断参数信息。采用Revman5．3统计软件提供的QUADAS-2风险评估标准评价纳入文献的质量，采用Stata13．0软件对资料进行Meta分析，绘制综合受试者工作特征曲线（SROC），并计算曲线下面积（AUC）。结果最终纳入11篇文献，总样本量为2550例；异质性检验结果显示各文献的灵敏度、特异度、阴性似然比、诊断比值比（DOR）间均存在异质性。研究设计类型、TCD检验标准这2个因素可能为异质性的来源（P〈O．05）。Meta分析结果显示汇总灵敏度为0．88（95％CI：0．84-0．91）、特异度为0．93（95％CI：0．91-0．95）、阳性似然比为12．93（95％CI：9．90-16．90）、阴性似然比为0．13（95％CI：0．09-0．17）、诊断评分（SCORE）为4．63（95％CI：4．32-4．94），DOR为102．64（95％CI：75．44-139．65）；SROC的AUC为0．96（95％CI：0．94-0．98）；敏感性分析显示剔除文献后的各诊断参数的95％CI与原数据有较大部分重叠。结论TCD对大脑中动脉狭窄具有较高的诊断价值，可用于此疾病的诊断筛查。

Objective To evaluate the diagnostic value of transcranial doppler ultrasonography （TCD） for middle cerebral artery stenosis. Methods A comprehensive electronic search was performed in PubMed, EMBase, The Cochrane Library, CNKI, Wanfang Database and VIP Database to retrieve relevant studies on TCD in diagnosis of middle cerebral artery stenosis from initiation of the databases to January 2017. The general information and diagnostic parameters were documented from the retrieved studies, The quality of the prepared studies was evaluated by QUADAS-2 and the risk assessment maps brought by RevMan 5.3 statistical software. Pooled sensitivity, specificity, positive likelihood ratio （PLR）, negative likelihood ratio （NLR） and the heterogeneity of the included studies were analyzed using the Stata 13.0 software. Summary receiver operating characteristic curve （SROC） was drawn and the area under the curve （AUC） was calculated. The publication bias was assessed with funnel plotting. Results A total of 11 studies were included for this Meta analysis, involving 2 550 patients. Heterogeneity test showed that the heterogeneity was established between studies in sensitivity, specificity, NLR and diagnostic odds ratio （DOR）, resulting from disparity in study design and TCD test criteria （P〈0.05）. Meta analysis revealed a pooled sensitivity of 0.88 （95% CI： 0.84-0.91）, a specificity of 0.93 （95% CI： 0.91-0.95）, a PLR of 12.93 （95% CI： 9.90-16.90）, a NLR of 0.13 （95% CI： 0.09-0.17）, a diagnostic score （SCORE） of 4.63 （95% CI： 4.32-4.94）, a diagnostic odds ratio （DOR） of 102.64 （95% CI： 75.44-139.65）, and an AUC of SROC of 0.96 （95% CI： 0.94-0.98）. The sensitivity analysis after removing studies with sample capacity 〈100 showed that the confidence interval of the diagnostic parameters mostly overlapped with that of the original data. Conclusion As TCD shows good accuracy in diagnosing middle cerebral artery stenosis, it can be used in diagnostic screening of the disease.