Objective To access the effectiveness and safety of levofloxacin in controlling multidrug resistant tuberculosis (MDR-TB). Methods The electronic searches in databases of PubMed, EMbase, The Cochrane Library, CNKI, CBM and VIP, handsearches and other searches were conducted from the date of their establishment to April 2011 for collecting randomized controlled trials (RCTs) on levofloxacin treating MDR-TB. Two researchers independently screened the literature according to the inclusive and exclusive criteria, extracted the data, assessed the quality of the included studies by adopting the Jadad scale, and performed Meta-analysis by using RevMan 5.0 software. Results A total of 31 RCTs involving 2836 cases were included. The results of meta-analysis showed: a) Compared with the placebo group, levofloxacin could increase the sputum negative conversion rate after 3-month taking and at the end of the treatment period; b) Compared with the ofloxacin group, levofloxacin could increase the sputum negative conversion rate after 3 months and at the end of the treatment period; c) levofloxacin replacing either ethambutol or streptomycin could increase the sputum negative conversion rate after 3 months and at the end of the treatment period; d) Compared with the levofloxacin group, gatifloxacin and moxifloxacin could increase the sputum negative conversion rate after 3 months and at the end of the treatment period; e) There was no significant difference in the adverse reaction rate between each of the medication regimens (P=0.19). Conclusion Levofloxacin is more effective for MDR-TB than ofloxacin, ethambutol and streptomycin, but it is inferior to gatifloxacin or moxifloxacin. Its adverse reaction rate is equivalent to other medicines’.