目的 系统回顾并量化非酒精性脂肪性肝病（NAFLD）与缺血性脑卒中发病之间的关系。方法 截至 2020年 1月 31日，系统检索了PubMed、Web of Science、中国学术期刊网络出版总库、万方数据期刊论文资源、Cochrane Library、中国生物医 学文献数据库和Embase，使用STATA 15.0进行统计分析，合并纳入研究的OR及95%CI，同时根据地区、诊断方法、血清转氨酶 水平进行亚组分析。结果 共纳入符合标准的9篇文章4206例患者。在8项研究中，NAFLD与缺血性脑卒中之间存在显著关 联（OR=2.53，95%CI=1.83~3.49）。在所有比较中，各研究之间没有发现显著的异质性。在敏感性分析中排除任何一项研究后， 这些结果基本保持不变。按地区进行亚组分析结果显示，在欧美地区，NAFLD患者发生缺血性脑卒中的综合风险为OR=2.17 （95%CI=1.03~4.58）；在亚洲地区，风险为 OR=3.23(95%CI=1.83~5.7)；在非洲，风险为 OR=2.28（95%CI=1.44~3.61）。按脂肪肝 诊断的不同方法进行亚组分析表明，诊断方法可能是影响NAFLD与缺血性脑卒中风险关系的混杂因素（腹部超声诊断，OR= 2.60，95%CI=1.87~3.61；肝脏活检诊断，OR=1.08，95%CI=0.18~6.52）。按血清转氨酶水平进行亚组分析，结果表明谷丙转氨酶 （ALT）水平与 NAFLD 患者发生缺血性脑卒中的风险增加有显著相关性（ALT 的平均差值为 1.14 U/L, 95%CI=0.68~1.61）。谷 草转氨酶（AST）水平与 NAFLD 患者发生缺血性脑卒中风险的增加有显著的相关性（AST 平均水平差异为 0.57 U/L，95%CI= 0.25~0.80）；谷氨酰转移酶（GGT）水平与 NAFLD 患者发生缺血性脑卒中风险的增加有显著相关性（GGT 平均水平差异为 1.08 U/L，95%CI=0.38~1.78）。结论 NAFLD可能会增加缺血性脑卒中的发生风险。

Objectives Several studies have emerged indicating that ischemic stroke is associated with nonalcoholic fatty liver disease (NAFLD), and we aimed to systematically review and quantify the association between ischemic stroke and NAFLD. Methods PubMed, the Web of Science, China Academic Journals Full-text Database, Wanfang Data, the Cochrane Library, Chinese biomedical literature and Embase were searched (until January 2020) for studies examining the relationship of NAFLD and ischemic stroke. The odds ratio (OR) value and 95% confidence interval (CI) of the included studies were pooled. At the same time, the subgroup analysis was conducted according to region, diagnostic method and serum transaminase level. Results Nine articles met inclusion criteria, with a total of 4, 206 individuals. A significant association was found between NAFLD and ischemic stroke in eight studies (OR = 2.53, 95%CI=1.83-3.49). No significant heterogeneity was detected among studies in all comparisons. These results remained essentially unchanged after excluding any of the studies in the sensitivity analysis. Subgroup analysis by continent showed that, in Europe and America, the combined risk of NAFLD patients occurred ischemic stroke was 2.17 (95%CI=1.03-4.58); in Asia, the risk was 3.23 (95%CI=1.83-5.7); in Africa, the risk was 2.28 (95%CI=1.44-3.61). Subgroup analysis by different methods of diagnosing fatty liver showed that diagnostic methods may be a confounding factor affecting the relationship between NAFLD and ischemic stroke risk (abdominal ultrasound diagnosis, OR=2.60, 95%CI=1.87-3.61; liver biopsy diagnosis, OR=1.08, 95%CI=0.18-6.52). Subgroup analysis by serum transaminase levels showed that the levels of ALT were significantly associated with an increased risk of ischemic stroke in NAFLD patients (the mean difference of ALT was 1.14 U/L, 95%CI=0.68-1.61). The levels of AST were significantly associated with an increased risk of ischemic stroke in NAFLD patients (the mean difference of AST was 0.57 U/L, 95%CI=0.25-0.80). The level of GGT was significantly associated with an increased risk of ischemic stroke in NAFLD patients (the mean difference of GGT was 1.08 U/L, 95%CI=0.38-1.78). Conclusions This meta⁃analysis indicated that nonalcoholic fatty liver disease can increase the risk incident of ischemic stroke.