Abstract： Objective To evaluate the effects of high PEEP versus low PEEP upon mortality in patients with acute respiratory distress syndrome(ARDS).Methods The databases Medline,Embase,Web of Science,CENTRAL and CBM(Chinese Bio-Medical Literature & Retrieval System) were retrieved by using the key words ARDS,ALI,acute hypoxemic respiratory failure,acute respiratory distress syndrome,adult respiratory distress syndrome,respiratory failure,systemic inflammatory response syndrome,non-cardiogenic pulmonary edema,ventilation,PEEP,and positive end expiratory pressure,so as to search the material about the randomized controlled trials comparing the mortality of ARDS or ALI treated by high PEEP and low PEE P.Study reports were included if they:(1) enrolled over 10 patients;(2) enrolled adult human ARDS or ALI subiects treated with mechanical ventilation;(3) included high PEEP and low PEEP with significant differences;and (4) included hospital mortality.Results Five documents about randomized controlled clinical trials,including 2447 patients.met the enrollment criteria.The hospital mortalities of high PEEP and low PEEP groups were 34.2% (415/1215) and 37.7% (464/1232) respectively (RR=0.91,95% CI:0.82-1.01,P=0.07).Sensitive analysis confirmed the nonexistence of differences in hospital mortality between these two groups (RR=0.94,95% CI:0.84-1.05.P=0.25).Neither funnel plot nor rank correlation test regarding hospital mortality indicated the existence of publication bias(χ2=8.26.P=0.09).Conclusion The use of high or low PEEP in oxygenation function does not affect hospital mortality in ARDS patients.