ObjectiveTo systematically review the effectiveness and safety of thalidomide for ankylosing spondylitis (AS). MethodsDatabases including Ovid MEDLINE (1946 to 2014.2.1), EMbase (1947 to 2014.2.1), CENTRAL (Issue 1, 2014), CBM (1978 to 2014.2.1), CNKI (1994 to 2014.2.1), WanFang Data (1980 to 2014.2.1) and VIP (1989 to 2014.2.1) were searched for randomized controlled trials about the effectiveness and safety of thalidomide for AS. Two reviewers independently screened literature according to the inclusion and exclusion criteria, extracted data, and assessed methodological quality of included studies. Meta-analysis was then conducted using RevMan 5.2 software. ResultsSeven RCTs were included involving 544 patients. The results of meta-analysis showed that, compared with the blank group, thalidomide increased clinical remission, but it showed no obvious advantage in Bath Ankylosing Spondylitis Disease Activity Index (BASDAI), Bath Ankylosing Spondylitis Functional Index (BASFI), and secondary outcome index, with an increased total rate of withdrawal/drop-out. Compared with SSZ, thalidomide increased the rate of maintaining remission when it was used in the maintenance treatment after the patients attained ASAS20; and for other outcomes it was similar to SSZ. Compared with NSAIDs, thalidomide increased the rate of maintaining remission when it was used in the maintenance treatment after the patients attained ASAS20; it increased clinical remission; for secondary outcomes it was similar to NSAIDs; and it had a higher incidence of adverse reaction as well as an increased total rate of withdrawal/drop-out. ConclusionCompared with the blank group, thalidomide increases clinical remission, with an obviously-increased total rate of withdrawal/drop-out. Compared with SSZ, thalidomide increases the rate of maintaining remission when it is used in the maintenance treatment after patients attain ASAS20. Compared with NSAIDs, thalidomide increases the rate of maintaining remission when it is used in the maintenance treatment after patients attain ASAS20; it also increases clinical remission; but it has a higher incidence of adverse reaction as well as an increased total rate of withdrawal/drop-out. Due to limited quantity and quality of the included studies, the above conclusion needs to be verified by conducting more high quality studies.