OBJECTIVES: To evaluate the efficacy and safety of ceftriaxone and levofloxacin in treating community-acquired pneumonia and provide references for clinical application. METHODS: This was a systematic review. Data were collected from literature (randomized controlled trails) published from January 2003 to January 2014 through searching databases both at home and abroad, such as CNKI, WanFang Data, VIP, PubMed, Science Direct, Springer, Ovid, Wiley-Blackwell and The Cochrane Library(Issue1,2014). Both qualitative analysis and quantitative analysis were conducted. Quantitative analysis (meta-analysis) was performed by RevMan 5.2.  RESULTS: Nine studies were included, involving 2233 patients. Of these, six studies (553 patients) met meta-analysis criteria. The results of meta-analysis were clinical effective rates (RR=0.90, 95%CI, P=0.002), and adverse events rates (RR=0.70, 95%CI, P=0.22). The result of qualitative description as follows: one prospective, randomized study showed success rate were 89% in the ceftriaxone group and 96% in the levofloxacin group; one multicenter retrospective study showed the mortality of the two groups were 3.1% and 2.0%, the length of hospital stay(LOS) were 5.5±3.5 days and 4.8±2.9 days respectively, with no significant difference found between groups; the other retrospective study showed ceftriaxone could shorten LOS since it had a trend toward earlier switch to oral therapy.  CONCLUSIONS: In general, current evidence shows that the efficacy for the treatment of community-acquired pneumonia of levofloxacin is superior to ceftriaxone, there were no significant difference in the incidence of adverse reaction.