Objective To systematically review the effectiveness and safety of alanyl-glutamine dipeptide for severe acute pancreatitis (SAP). Methods Such databases as MEDLINE, EMbase, CENTRAL, VIP, WanFang Data, CBM and CNKI were electronically searched from inception to October, 2012 for randomized controlled trials on alanyl-glutamine dipeptide for SAP. Two reviewers independently screened literature according to the inclusion and exclusion criteria, extracted data, and assessed methodological quality. Then, meta-analysis was performed using RevMan 5.2. Results Five trials were included involving a total of 227 patients. The results of meta-analysis showed that: compared with the control group, the alanyl-glutamine dipeptide group had the lower incidence of SAP complications (RR=0.41, 95%CI 0.20 to 0.82), the lower incidence of infected pancreatic necrosis (RR=0.12, 95%CI 0.02 to 0.89), less time for alleviating bellyache (MD= –0.90, 95%CI –1.72 to –0.08). There was a tendency in decreasing SAP mortality (RR=0.15, 95%CI 0.02 to 1.19) and lessening the recovery time of blood amylase (SMD=0.37, 95%CI –0.04 to 0.79). Conclusion Current evidence shows that, alanyl-glutamine dipeptide can lower the incidence of complications and the incidence of infected pancreatic necrosis, and shorten the time for alleviating bellyache in SAP patients. Due to the limited quality of the included studies, the above conclusion needs to be verified by more high quality studies.