Objective To evaluate the efficacy and safety of colistin in the treatment of severe infections. Methods PubMed, ISI Web of Knowledge and Wanfang databases were searched. The initial literatures and references listed in the literature were manually searched. Controlled studies were analyzed using RevMan 5. 0 software.Results Eleven studies were enrolled, including five prospective studies and six retrospective studies. Pooled analysis showed that, compared with other therapies, treatment with colistin in severe infections did not improve 28 or 30-day mortality, clinical symptoms, or bacteria clearance,however, increased the risk of kidney damage. Subgroup analysis showed that colistin did not improve symptoms, mortality ( which was even higher in the patients with drug resistant bacteria infection) , or kidney damage in drug resistant bacteria infections and ventilator associated pneumonia ( VAP) compared with the other antibiotic group. Conclusions Colistin is not superior to the other antibiotics in severe infections.However, there are some shortcomings in our meta-analysis due to limited high-quality RCTs, thus welldesigned RCTs are still needed before final conclusion is made.