目的:通过系统分析方法评价持续低剂量肝素及低分子肝素抗凝治疗对急性呼吸窘迫综合征（ARDS）患者的疗效及安全性。方法:检索万方数据库、维普数据库、中国知网（CNKI）、中国生物医学文献服务系统（SinoMed）、美国国立医学图书馆数据库（PubMed）、荷兰医学文摘数据库（Embase）、Cochrane图书馆数据库等中英文数据库中1999年7月至2020年6月发表的有关肝素或低分子肝素治疗ARDS的临床随机对照试验（RCT）。对照组给予常规治疗，试验组在常规治疗基础上加用肝素5～10 U·kgn -1·hn -1或低分子肝素皮下注射2 500～5 000 U，每12 h 1次，疗程均为7 d；主要结局指标为28 d病死率、治疗7 d后动脉血氧饱和度（SaOn 2）及急性生理学与慢性健康状况评分Ⅱ（APACHEⅡ）、重症监护病房（ICU）住院时间及机械通气时间；次要结局指标为治疗7 d后C-反应蛋白（CRP）、肿瘤坏死因子-α（TNF-α）、白细胞介素-6（IL-6）、肺损伤评分、呼吸频率、活化部分凝血活酶时间（APTT）、凝血酶原时间（PT）等指标。由2名研究者收集数据，依据Cochrane 5.1手册评价文献质量，采用RevMan 5.3软件进行Meta分析；Meta分析结果的稳定性用敏感性分析方法检验，同时通过漏斗图分析SaOn 2的发表偏倚。n 结果:最终纳入14篇中文文献，共涉及894例患者，其中试验组454例，对照组440例。Meta分析结果显示，与对照组相比，试验组28 d病死率明显降低〔相对危险度（n RR）＝0.54，95%可信区间（95%n CI）为0.38～0.76，n P＝0.000 4〕，治疗7 d SaOn 2均显著升高〔肝素：均数差（n MD）＝48.27，95%n CI为29.95～66.59，n P<0.000 01；低分子肝素：n MD＝56.67，95%n CI为41.22～71.13，n P<0.000 01〕；与对照组相比，治疗7 d后试验组APACHEⅡ评分（n MD＝-4.28，95%n CI为-5.15～-3.42，n P<0.000 01）、肺损伤评分（n MD＝-1.19，95%n CI为-1.35～-1.03，n P<0.000 01）及呼吸频率（n MD＝-4.76，95%n CI为-6.26～-3.26，n P<0.000 01）均明显降低，试验组患者ICU住院时间（n MD＝-4.85，95%n CI为-6.94～-2.76，n P<0.000 01）、机械通气时间（n MD＝-2.93，95%n CI为-3.34～-2.52，n P<0.000 01）亦均明显低于对照组；治疗7 d后试验组IL-6（n MD＝-38.50，95%n CI为-59.01～-17.99，n P<0.000 01）、TNF-α（n MD＝-16.24，95%n CI为-31.11～-1.38，n P<0.000 01）及CRP（n MD＝-5.5，95%n CI为-6.47～-4.27，n P<0.000 01）亦显著降低。但两组治疗7 d后APTT（n MD＝-0.55，95%n CI为-1.61～0.51，n P＝0.27）及治疗7 d PT（n MD＝-0.41，95%n CI为-1.48～0.66，n P＝0.45）比较差异均无统计学意义。对异质性较高的指标通过排除任意一项研究再次进行Meta分析，其总体效应值未发生明显改变，提示结果比较稳健。对纳入文献数量超过10篇的指标进行漏斗图分析，结果显示文献存在发表偏倚，但偏倚较小。n 结论:持续性低剂量肝素和低分子肝素抗凝治疗可改善ARDS患者预后，降低病死率。“,”Objective:The effects of low-dose heparin and low molecular weight heparin on the efficacy and prognosis of patients with acute respiratory distress syndrome (ARDS) were systematically evaluated.Methods:Chinese and English databases such as Wanfang data, VIP database, China National Knowledge Infrastructure (CNKI), China Biomedical Literature Service system (SinoMed), American National Medical Library database (PubMed), Netherlands Medical Abstracts database (Embase) and Cochrane library database were searched for the randomized controlled trials (RCTs) of heparin or low molecular weight heparin in the treatment of ARDS. The search time is from July 1999 to June 2020. The control group was given routine treatment, while the experimental group was given heparin 5-10 U·kgn -1·hn -1, or low molecular heparin subcutaneous injected 2 500-5 000 U once every 12 hours for 7 days on the basis of the routine treatment. The main outcome were 28-day mortality, arterial oxygen saturation (SaOn 2), acute physiology and chronic health evaluation Ⅱ (APACHEⅡ), length of stay in intensive care unit (ICU) and mechanical ventilation time, and the secondary outcome indexes were C-reactive protein (CRP), tumor necrosis factor-α (TNF-α), interleukin-6 (IL-6), lung injury score, respiratory frequency, activated partial thromboplastin time (APTT) and prothrombin time (PT) after 7 days' treatment. The two researchers collected data and evaluated the quality of the literature according to the Cochrane 5.1. The stability of the Meta-analysis results was tested by sensitivity analysis, and the publication bias was included in the funnel chart analysis.n Results:Fourteen Chinese literatures were included, with a total of 894 cases, including 454 cases in the experimental group and 440 cases in the control group. The results of Meta-analysis showed that the 28-day fatality rate in the experimental group was significantly lower than that in the control group [relative risk (n RR) = 0.54, 95% confidence interval (95%n CI) was 0.38-0.76, n P = 0.000 4], and 7-day SaOn 2 significantly increased [heparin: mean difference (n MD) = 48.27, 95%n CI was 29.95-66.59, n P < 0.000 01]. Low molecular weight heparin: n MD = 56.67, 95%n CI was 41.22-71.13, n P < 0.000 01]. Compared with the control group, the APACHE Ⅱscore ( n MD = -4.28, 95%n CI was -5.15 to -3.42, n P < 0.000 01), the lung injury score ( n MD = -1.19, 95%n CI was -1.35 to -1.03, n P < 0.000 01) and the respiratory rate ( n MD = -4.76, 95%n CI was -6.26 to -3.26, n P < 0.000 01) in the experimental group were significantly lower than those in the control group after 7 days' treatment. The time of staying in ICU ( n MD = -4.85, 95%n CI was -6.94 to -2.76, n P < 0.000 01) and the time of mechanical ventilation ( n MD = -2.93, 95%n CI was -3.34 to -2.52, n P < 0.000 01) in the experimental group were lower than those in the control group. After 7 days' treatment, the levels of IL-6 ( n MD = -38.50, 95%n CI was -59.01 to -17.99, n P < 0.000 01), TNF-α ( n MD = -16.24, 95%n CI was -31.11 to -1.38, n P < 0.000 01) and CRP ( n MD = -5.5, 95%n CI was -6.47 to -4.27, n P < 0.000 01) were significantly lower in the experimental group than those in the control group. However, after 7 days' treatment, there were no significant differences in APTT ( n MD = -0.55, 95%n CI was -1.61 to 0.51, n P = 0.27) or PT (n MD = -0.41, 95%n CI was -1.48 to 0.66, n P = 0.45). For the indicators with high heterogeneity, Meta-analysis was carried out again by excluding any study, and the overall effect value did not change significantly, suggesting that the results were relatively robust. Funnel chart analysis was carried out on the indexes with more than 10 articles included in the literature. The results showed that the literature had publication bias, but the bias was small.n Conclusion:Continuous anticoagulation therapy with low dose heparin and low molecular weight heparin can improve the prognosis and reduce the mortality of patients with ARDS.