Abstract： Objective To observe the effect of two different ways of subglottic secretion drainage on the incidence of ventilator-associated pneumonia.Methods A total of 149 patients receiving mechanical ventilation from October 2007 to February 2010 in our hospital were enrolled in our study.According to the random number table,they were divided into 3 groups:a control group (47),a intermittent drainage group (54),and a continuous drainage group (48).There was no significant difference in APACHE Ⅱ score before intubation in the patients of the 3 groups.Conventional treatment and care was performed in the control group,while subglottic rinse was performed every 4 h in the intermittent drainage group and continuous aspiration in the continuous drainage group.The incidence and mortality of ventilator-associated pneumonia were compared among groups within 30 days after intubation.The tolerance and complications were observed.Results The incidence rate of early onset ventilator-associated pneumonia in the intermittent drainage group(13.0％) and the continuous drainage group(14.6％) was significantly lower than that in the control group (40.4％),and the difference was statistically significant (x2 =9.914,7.978 ; P =0.002,0.005),but there was no difference between the former 2 groups (x2 =0.056 ; P =0.812).There was no significant difference in the incidence of late-onset ventilator-associated pneumonia or mortality among groups (P ＞ 0.05).Good tolerance of the 2 methods was observed.Conclusion Subglottic secretion drainage by both the intermittent and the continuous methods can efficiently prevent early onset ventilator-associated pneumonia.