Purpose: Delayed gastric emptying (DGE) is an important postoperative complication after pancreaticoduodenectomy (PD), and its incidence may be associated with the utilized surgical procedures. Compared with pancreaticojejunostomy (PJ) after PD, it may be speculated that pancreaticogastrostomy (PG) is a risk factor for DGE, because it needs an anastomosis of the remnant pancreas to the back wall of the stomach. This study aimed to compare PG and PJ with regard to the incidence of DGE after PD. Methods: We performed a prospective open-label randomized clinical trial (RCT) including patients undergoing elective pancreaticoduodenectomy, who were randomly assigned PG or PJ the day before surgery. The primary endpoint was incidence of DGE. Results: The study included 60 patients (30 PG, 30 PJ), of whom seven were deemed unresectable, one was enucleated, and one was switched from PJ to PG during surgery according to the surgeon’s decision. Thus, modified intention-to-treat analyses were performed in 27 PG patients and 26 PJ patients. DGE occurred in three patients in the PG group and six patients in the PJ group, which did not constitute a significant between-group difference (P = 0.42). In the PG group, two cases were ISGPS grade A DGE and one was grade C. In the PJ group, one case was grade A, two grade B, and three grade C. The two groups also did not significantly differ in the incidence of other morbidities or postoperative hospital stay. Conclusions: Post-PD DGE incidences were similar after PG and PJ.
All Science Journal Classification (ASJC) codes