他汀类药物治疗ALI／ARDS的系统回顾和Meta分析：基于国际数据库的结果

Statin in the treatment of ALI/ARDS： a systematic review and Meta-analysis based on international databases

目的系统评价他汀类药物对急性肺损伤／急性呼吸窘迫综合征（ALI／ARDS）患者预后的影响。方法检索美国国立医学图书馆PubMed／Medline数据库、荷兰医学文摘Embase数据库、科学网（Web of Science）和Cochrane图书馆数据库，自建库至2015年11月17日发布的关于他汀类药物用于ALI或ARDS患者的随机对照临床试验（RCT）和队列研究。对照组给予常规治疗，试验组在对照组基础上应用他汀类药物。检索词为急性肺损伤、ALI、急性呼吸窘迫综合征、ARDS、他汀、辛伐他汀、罗素他汀；无发表日期和语言限制；主要结局为院内病死率，其他指标包括脱机时间、非重症加强治疗病房（ICU）住院时间、ICU住院时间和ICU病死率。采用RevMan5．2和STATA13．0软件进行系统评价和Meta分析，绘制漏斗图分析纳入文献发表偏倚情况。结果共纳入5项研究、1636例患者，试验组739例，对照组897例；3项RCT，2项队列研究。Meta分析结果显示，试验组和对照组院内病死率差异无统计学意义[相对危险度（RR）=0．96，95％可信区间（95％CI）=0．79。1．15，P=0．63]；无论是基于RCT和队列研究的亚组分析，还是不同他汀类药物的亚组分析，均提示试验组和对照组院内病死率差异无统计学意义（均P〉O．05）。试验组与对照组脱机时间[均数差（MD）=1．41，95％CI=-0．32—3．13，P=0．11]、非ICU住院时间（MD=-0．23，95％CI=-1．61—1．15，P=0．75）、ICU住院时间（MD=-1．03，95％CI=-6．55—4．50，P=0．72）和ICU病死率（RR=0．88，95％CI=0．68—1．14，P=0．33）比较差异均无统计学意义。漏斗图显示各研究间院内病死率不存在发表偏倚。结论他汀类药物不能降低ALI／ARDS患者院内和ICU病死率，也不能改善脱机时间及非ICU和ICU住院时间。

Objective To confirm the effects of statin therapy on mortality of patients with acute lung injury/ acute respiratory distress syndrome （ALI/ARDS）. Methods PubMed/Medline, Embase, Web of Science and Cochrane Central Register of Controlled Trials were searched for articles using the terms ＂acute lung injury＂, ＂ALl＂, ＂acute respiratory distress syndrome＂, ＂ARDS＂, ＂statin＂, ＂simvastatin＂ and ＂rosuvastatin＂ updated to November 17, 2015. Randomized controlled trial （RCT） or observational cohort studies investigating the effects of statin therapy on mortality in patients with ALI or ARDS were all identified, without date or language restriction. The control group was given conventional treatment, while the experimental group was treated with statins additionally. The primary outcome was in-hospital mortality. Meanwhile, ventilator-free day, intensive care unit （ICU）-free day, ICU length of stay （LOS） and ICU mortality were also analyzed. RevMan 5.2 and STATA 13 software were used for systematic review and Meta analysis, and funnel plot was used to analyze the publication bias. Results A total of five trials including three randomized controlled trials and two observational studies were included. Among 1 636 patients enrolled in the study, there were 739 patients in experimental group, and 897 in control group. It was shown by Meta analysis that there was no significant difference in in-hospital mortality between experimental group and control group [relative risk （RR） = 0.96, 95% confidence interval （95%CI） = 0.79-1.15, P = 0.63]. The subgroup analysis based on RCT and cohort study, or the subgroup analysis of different statins showed that there was no significant difference in in-hospital mortality between the experimental group and the control group （both P 〉 0.05）. There were no significant differences in ventilator-free days [mean difference （MD） = 1.41, 95%CI = -0.32-3.13, P = 0.11], ICU-free days （MD = -0.23, 95%CI = -1.61-1.15, P = 0.75）, ICU length of stay （MD = -1.03, 95%CI = -6.55-4.50, P = 0.72）, or ICU mortality （RR = 0.88, 95%CI = 0.68-1.14, P = 0.33） between the experimental group and the control group. It was shown by funnel plot that there was no publication bias in in-hospital mortality. Conclusion The systematic review and meta-analysis suggests that statin may not be associated with a significant reduction in mortality, ventilator-free day, ICU-free day and ICU length of stay in patients with ALI/ARDS.