Objective To compare anastomotic fistula of modified triple-layer duct-to-mucosa pancreaticojejunostomy and end-to-end invagination pancreaticojejunostomy following pancreaticoduodenectomy. Methods The clinical data of 147 patients underwent pancreaticoduodenectomy from January 2015 to June 2017 in the West China Hospital of Sichuan University were retrospectively analyzed. The modified triple-layer duct-to-mucosa pancreaticojejunostomy were used in 101 cases (MTL group) and end-to-end invagination pancreaticojejunostomy were used in 46 cases (IPJ group). The differences of intraoperative and postoperative statuses were compared between the two groups. Results The baseline data of these two groups had no significant differences (P>0.05). Except for the average time of the pancreaticoenterostomy of the MTL group was significantly longer than that of the IPJ group (P<0.05), the intraoperative blood loss, the first postoperative exhaust time, postoperative hospitalization time, reoperation rate, death rate, and rates of complications such as the pancreatic fistula, biliary fistula, anastomotic bleeding, gastric emptying disorder, and intraperitoneal infection had no significant differences between these two groups (P>0.05). Conclusions Both modified triple-layer duct-to-mucosa pancreaticojejunostomy and end-to-end invagination pancreaticojejunostomy following pancreaticoduodenectomy are safe and effective. An individualized selection should be adopted according to specific situation of patient.