PGE_1治疗对肝衰竭患者生存及并发症发生率影响的荟萃分析

A meta-analysis of PGE_1 Therapy for Survival and Incidence of Complications in Patients with Liver Failure

目的评价PGE1治疗对肝衰竭患者生存及并发症发生率的影响。方法检索MEDLINE、EMBASE.com、Cochrane Controlled Trial Register、中国生物医学文献数据库,收集PGE1治疗肝衰竭的随机对照研究,筛选、评价文献并提取数据,计算PGE1治疗和安慰剂或空白对照治疗的患者在治疗结束后发生死亡结局及并发症的优势比,并对结果做敏感性分析和偏倚分析。结果Meta分析结果提示PGE1组死亡优势比对照组低29%,OR=0.30,95%可信区间0.23-0.37。PGE1对降低肝衰竭患者肝肾综合征的发生率有统计学意义,而对肝性脑病、出血、感染等并发症的发生均无显著影响。所有结果通过敏感性分析未见结果逆转,偏倚分析提示无显著发表偏倚存在。结论PGE1能降低肝衰竭患者的病死率及肝肾综合征的发生率。

Objective To evaluate the effectiveness of PGE1 therapy for survival and incidence of complications in patients with liver failure. Methods MEDLINE, EMBASE. corn, Cochrane Controlled Register and CBMdisc were searched. Randomized controlled trials of PGE1 therapy for liver failure were collected and reviewed. The quality of articles was evaluated and data was extracted from it. The odds ratio for death and complications among the patients with PGE1 therapy as well as the blank controls was calculated at the end of therapy. The results of meta - analysis were assessed according to sensitivity and bias. RevMan4.2 was applied to process the data. Results The results of metaanalysis showed that PGE1 reduced the casefatality ratio of liver failure [ OR =0.30, 95% CI（0.23 - 0.37）] and the risk difference was decreased 29% in PGE1 patients compared with the blank controls. And there was statistical significance in PGE1 reducing the incidence of hepatorenal syndrome, but occurance of the complieations such as hepatic encephalopathy, bleeding, infection etc. These results were not reversed in the analysis of sensitivity, and the analysis of bias showed little bias of them. Conclusions PGE1 can reduce the case - fatality ratio of liver failure and the incidence of hepatorenal syndrome. The accuracy of the results are low due to the poor quality of the included trials. Thus, the effect of PGE1 on liver failure needs to be further evaluated through randomized controlled trials of high quality to be carried out in multicenter and large samples.