We conducted a comprehensive analysis to evaluate the treatment efficacy and safety of enteral nutrition (EN) and parenteral nutrition (PN) in severe acute pancreatitis (SAP) patients, and to provide a basis for their evidence based application in a clinical setting. We conducted a systematic online search of the PubMed, Web of Science, Wanfang, and China National Knowledge Infrastructure databases, from their inception to November 2017. Studies were subjected to further screening if they met the inclusion/exclusion criteria. Eleven studies were subjected to qualitative and quantitative synthesis; these included a total of 562 patients (281 for EN and 281 for PN). No significant heterogeneity across studies was found. The results indicated that EN can significantly decrease the mortality rate (relative risk [RR] = 0.43, 95% confidence interval [CI]: 0.23–0.78, P=0.006), and lowers the risk of infection and complications (RR = 0.53, 95% CI: 0.39–0.71, P=0.000) more so than does PN. The EN group had a similar risk of multiple organ failure (MOF) compared with the PN group (RR = 0.63, 95% CI: 0.39–1.02, P=0.059). The use of EN was also found to significantly reduce mean hospitalization time (mean difference = −2.93, 95% CI: −4.52–1.34, P=0.000). No publication bias was found. Our meta-analysis suggested that EN, but not PN, significantly reduced the risk of mortality, infection, and complications for patients with SAP. EN support also decreased the rate of MOF and surgical intervention. EN is recommended as an initial treatment option for patients with SAP.