Fatality and adverse event rates of prostaglandin E in treating acute liver failure: a meta-analysis
Objective To evaluate the fatality and adverse event rates of prostaglandin E (PGE) versus placebo in treating acute liver failure (ALF). Methods A search was performed using Cochrane Library, MEDLINE, EMBASE, VIP, CNKI, CBM, and other electronic databases to select randomized controlled trials (RCT...
Saved in:
Published in | Linchuang gandanbing zazhi Vol. 29; no. 9; pp. 681 - 684 |
---|---|
Main Authors | , , |
Format | Journal Article |
Language | English |
Published |
01.09.2013
|
Online Access | Get full text |
Cover
Loading…
Summary: | Objective To evaluate the fatality and adverse event rates of prostaglandin E (PGE) versus placebo in treating acute liver failure (ALF). Methods A search was performed using Cochrane Library, MEDLINE, EMBASE, VIP, CNKI, CBM, and other electronic databases to select randomized controlled trials (RCTs) for comparing PGE and placebo in treating ALP published up to April 2012. There were no limits to language and publication. Data extraction and quality evaluation were performed independently by two researchers for the RCTs meeting inclusion criteria; a meta-analysis was performed with RevMan 5.1 software, and the GRADE system was used to grade the quality of evidence and strength of recommendation. Results Two RCTs involving 59 cases met the inclusion criteria. Compared with the placebo, PGE could not reduce the fatality rate (RR = 0.99; 95% CI:0.62-1.57; P = 0.96), as demonstrated by the meta-analysis. Both interventions caused slightly adverse events, but no incidence rate was reported. Based on the GRADE system, the quality of evidence was low (2C), and the strength of recommendation was weak. Conclusion PGE cannot reduce the fatality of ALF, according to the systematic review. Large-scale, high-quality basic and clinical researches should be performed to confirm the conclusion because the systematic review is secondary study, the literature included has low quality of evidence, and there may be bias in review. |
---|---|
Bibliography: | ObjectType-Article-1 SourceType-Scholarly Journals-1 content type line 23 ObjectType-Feature-2 |
ISSN: | 1001-5256 |
DOI: | 10.3969/j.issn.1001-5256.2013.09.012 |