OBJECTIVE To evaluate by meta-analysis the effects of dexmedetomidine versus midazolam on postoperative delirium in patients that received postoperative mechanical ventilation. METHODS The electronic databases of PubMed, Web of Science, EMbase, CNKI, CBM, Cochrane library and WanFang were searched by two reviewers. All the clinical studies related to dexmedetomidine versus midazolam on postoperative delirium were screened and collected in this meta-analysis. The combined postoperative delirium risk between dexmedetomidine and midazolam groups was pooled by random effect model. The publication bias was assessed by Begg's funnel plot and Egger's line regression test. RESULTS A total of six studies including 386 subjects (202 in the dexmedetomidine group and 184 in the midazolam group) were finally included in this meta-analysis. All six studies reported adequate sequence generation. Three studies used blindness methods and 2 publications were free of selective reporting. However, only 1 publication reported allocation concealment. Because of significant heterogeneity across the studies (I2=61.7%, p<0.05), the data were pooled by random effect model. Pooled data showed the postoperative delirium risk in the dexmedetomidine group was significantly lower than that of the midazolam group (RR=0.20 (095%CI:0.09~0.47, p<0.05)).The Begg's funnel plot showed obvious asymmetry at the bottom and Egger's line regression test also indicated significant publication bias (t=-6.51, p<0.05). CONCLUSION Compared with midazolam, patients that received dexmedetomidine for postoperative mechanical ventilation sedation had less risk of developing delirium.