Objective To evaluate the effectiveness and safety of early enteral nutrition (EN) versus total parenteral nutrition (TPN) after pancreaticoduodenectomy (PD). Methods Such databases as MEDLINE, EMbase, The Cochrane Library, CBM, VIP, CNKI were electronically searched to collect the randomized controlled trials (RCTs) about EN versus TPN after PD published from 2000 to March 2010. The quality of the included trials was assessed according to the inclusive and exclusive criteria, and the data were extracted and analyzed by using RevMan 5.0 software. Results A total of 4 RCTs involving 322 PD patients were included. The meta-analysis showed that the EN (the treatment group) was superior to the TPN (the control group) in the average postoperative hospital stay (MD= –2.34, 95%CI –3.91 to –0.77, Plt;0.05), the total incidence rate of complication (RR=0.75, 95%CI 0.57 to 0.99, P=0.04), the recovery time of enterocinesia (MD= –29.87, 95%CI –33.01 to –26.73, Plt;0.05) and the nutrition costs (MD= –30.51, 95%CI –35.78 to –25.24, Plt;0.05); there were no differences in mortality (RR=0.23, 95%CI 0.03 to 2.03, P=0.19), pancreatic leakage (RR=0.78, 95%CI 0.45 to 1.35, P=0.38), infectious complications (RR=0.71, 95%CI 0.43 to 1.18, P=0.19), non-infectious complications (RR=0.78, 95%CI 0.5 1 to 1.20, P=0.26) and postoperative serum albumin level (MD= –0.79, 95%CI –2.84 to 1.27, P=0.45). Conclusion Compared with total parenteral nutrition, the enteral nutrition used earlier after pancreatoduodenectomy shows significant advantages. But more reasonably-designed and double blind RCTs with large scale are expected to provide high quality proof.