摘要：

目的 系统评价颅内破裂动脉瘤夹闭术和栓塞术后分流依赖性脑积水发生率。方法 以脑积水、夹闭、介入、栓塞、颅内动脉瘤、脑动脉瘤、蛛网膜下腔出血/蛛网膜下隙出血，以及hydrocephalus、shunt、clipping、coiling、surgical、endovascular、embolization、treatment、intracranial aneurysm、cerebral aneurysm、subarachnoid hemorrhage等中英文词组，计算机检索1990年1月-2015年9月美国国立医学图书馆生物医学信息检索系统、荷兰医学文摘、Cochrane 临床对照试验中心注册库、中国知网中国知识基础设施工程、万方数据库，辅助手工检索《中华神经外科杂志》、《中国现代神经疾病杂志》和《中国脑血管病杂志》等相关杂志，查阅关于颅内破裂动脉瘤夹闭术和栓塞术后分流依赖性脑积水发生率的临床研究。采用Jadad 量表和Newcastle-Ottawa 量表评价文献质量，RevMan 5.3 和Stata 13.1 统计软件进行Meta 分析。结果 共获得731 篇文献，经剔除重复和不符合纳入标准者，最终纳入18 项临床试验共计15 920 例颅内破裂动脉瘤患者，行夹闭术者10 038 例、行栓塞术5882 例。Meta 分析显示：两种治疗方式术后分流依赖性脑积水发生率差异无统计学意义（OR = 0.860，95%CI：0.720 ~ 1.030；P = 0.110）；分析结果的稳定性较差，但不存在发表偏倚（Egger 法：P = 0.795）。结论 颅内破裂动脉瘤夹闭术和栓塞术后分流依赖性脑积水发生率无显著差异，但尚待进一步的高质量临床研究加以证实。

关键词: 脑积水, 颅内动脉瘤, 栓塞, 治疗性, 显微外科手术, Meta分析

Abstract:

Objective  To systematically review the risk of shunt-dependent hydrocephalus after treatment of ruptured intracranial aneurysms by microsurgical clipping versus endovascular coiling.  Methods  A comprehensive literature search from January 1990 to September 2015 was conducted in electronic databases, such as PubMed, EMBASE/SCOPUS, Cochrane Central Register of Controlled Trials (CENTRAL), China National Knowledge Infrastructure (CNKI) and Wanfang data, for randomized controlled trials (RCTs) or observational studies. Key words were defined as hydrocephalus, shunt, clipping, coiling, surgical, endovascular, embolization, treatment, intracranial aneurysm, cerebral aneurysm, subarachnoid hemorrhage both in Chinese and English. Manual search was also conducted on several relevant journals, such as Chinese Journal of Neurosurgery, Chinese Journal of Contemporary Neurology and Neurosurgery, Chinese Journal of Cerebrovascular Diseases, and so on. Data were extracted and evaluated by two reviewers independently. Jadad Scale was used to assess the quality of RCTs. Newcastle-Ottawa Scale (NOS) was used to evaluate the quality of nonrandomized controlled trials. RevMan 5.3 and Stata 13.1 were used for Meta-analysis.  Results  Eighteen observational studies were finally included, involving 15 920 patients with ruptured intracranial aneurysms, among whom 10 038 patients underwent clipping and 5882 underwent coiling. The result of Meta-analysis indicated that there was no statistical significance on the incidence of shunt-dependent hydrocephalus between clipping and coiling (OR = 0.860, 95%CI: 0.720-1.030; P = 0.110). However, the findings showed low stability as the results of fixed effects model and random effects model were different. The result of Egger's test indicated no publication bias (P = 0.795).  Conclusions  The frequency of shunt-dependent hydrocephalus was not significantly different between clipping and coiling. However, the conclusion still needs to be further verified by high-quality studies.

Key words: