Noninvasive Ventilation In Adult Acute Lung Injury/Acute Respiratory Distress Syndrome: Meta-analysis

Objectives:To systematically assess of noninvasive ventilation (NIV) on the rate of endotracheal intubation and intensive care unit (ICU) mortality in patients with acute lung injury (ALI) and acute respiratory distress syndrome (ARDS)Methods:Collect the RCTs on NIV treating ALI/ARDS adult patients extensively,search for English database of Pubmed and CENTRAL(the Cochrane Central Register of Controlled Trials) with the keywords of"NIV", "bipap" "CPAP", "bilevel positive airway pressure",''continuous positive airway pressure", "noninvasive positive pressure ventilation", "noninvasive ventilation".The searching criteria is limited to clinical trial and randomized controlled trial.Search for Chinese database of CNKI,Wanfang,Weipu,the MeSH is "NIV", "CPAP", "ALI" "ARDS"ect,using the methods of combining MeSH Subject Headings researsh,free words search and Boolean calculation connective words together.The references of included articles was second searched. Search time limited 1995-2010. we manual search for Critical Care Medicine, Intensive Care Medicine, The American Journal of Respiratory and Critical Care Medicine, Chest,no language limited.Identify the conclusion standards and exclusion standards,and manage the articles by the software of Endnote X4.All of the included articles was cataloged by different therapeutic methods.and Systematie review the quality of them,with different aspects:random, hiding,blind, follow-up.Extract the relevant data about NIV and common oxygen therapy treating adult ALI/ARDS patients. Systematic review the included trials by using the software of Cochrane Revman 5. Two-Category variable applys RD to analyse statistic with random effect model.Every effect size was expressed by 95% CI.Analyse the heterogeneity in the studies adopting x2 and P.If there is no statistical heterogeneity of all studies(P>0.1, I2≤50%),we apply fixed effect model to mata-analyse.If there is statistical heterogeneity(P<0.1, I2≥50%),according to probalble heterogeneity factors,we apply subgroup analysis to eliminate the heterogeneity.The studies which have clinical heterogeneity and couldn't get combined analysed would be descriptive analysed.At last all of the results will bu displayed by forest graph.Results:The search was totally 2343 pertinent literatures,by screening, secondary and supplementary search,4 articles met the inclusion criteria.Review the methodology quality of each article by using jaded Score Method,4 high qualitative articles have reached 4-5. In total,151 ALI/ARDS patients were detected,75 of them were accepted NIV treating,76 of them were accepted common oxygen therapy.Meta-analyze the included articles by adopting the random effective model.The result displays that on the trachea cannula events the risk difference(RD) is-0.31,95% confidence interval (CI) is [-0.60,-0.02], P=0.03, there was statistical significance.It indicates that comared with common oxygen therapy NIV can reduce the rate of trachea cannula. On the ICU death events,meta-analyse displays the risk difference(RD) is-0.24,95% confidence interval (CI) is [-0.58,0.10], P=0.16, there was no statistical significance.It indicates that comared with common oxygen therapy NIV couldn't reduce the mortalityConclusions:The evidence of this article indicates that compared with common oxygen therapy, NIV can reduce the rate of trachea cannula in adult patients with early acute lung injury (ALI) and acute respiratory distress syndrome (ARDS),but couldn't reduce the death events happened in ICU.