Objective To assess the effectiveness and safety of trazodone versus alprazolam on adults’ generalized anxiety disorder (GAD). Methods Such databases as PubMed (1980 to May 2012), CBM (1990 to May 2012), VIP (1989 to May 2012), CNKI (1990 to May 2012) and WanFang Data (1990 to May 2012) were searched to collect the randomized controlled trials (RCTs) about trazodone vs. alprazolam for adults’ GAD. According to the inclusion and exclusion criteria, two reivewers screened literature, extracted data and assessed the quality of the included studies, then meta-analysis was conducted using RevMan 5.0 software. Results A total of 5 RCTs involving 403 patients were included. The results of meta-analysis showed that: a) After four-week treatment, there were no significant differences between the two groups in the HAMA scores (RR=1.04, 95%CI 0.95 to 1.13, P=0.38) and cure rate (RR=1.05, 95%CI 0.75 to 1.48, P=0.76); and b) The somnolence rate of the trazodone group was lower than that of the alprazolam group (RR=0.42, 95%CI 0.25 to 0.72, P=0.001). But there were no significant differences between the two groups in dizziness (RR=0.52, 95%CI 0.27 to 1.01, P=0.05), fatigue (RR=0.10, 95%CI 0.01 to 1.41, P=0.09), and poor appetite (RR=2.82, 95%CI 0.28 to 28.23, P=0.38). Conclusion There is no significant difference between razodone and alprazolam in the effectiveness when treating GAD, but razodone has lower side effects while alprazolam tends easily to cause somnolence. For the quantity limitation and low methodological quality of the included studies, this conclusion still needs to be further proved by more high quality RCTs.