ABSTRACT
BACKGROUND: Pancreatic fistula (PF) is one of the most common complications after pancreaticoduodenectomy. There have been no large prospective randomized trials evaluating PF rates comparing invagination versus duct to mucosa pancreaticojejunostomy. We tested the hypothesis that a duct to mucosa pancreaticojejunostomy would reduce the PF rate. STUDY DESIGN: Between August 2006 and May 2008, 197 patients at two institutions underwent pancreaticoduodenectomy by a total of 8 experienced pancreatic surgeons as part of this prospective randomized trial (clinical trial no. NCT00359320). All patients were stratified by pancreatic texture and randomized to either an invagination or a duct to mucosa pancreaticojejunal anastomosis. Recorded variables included pancreatic duct diameter, operative time, blood loss, complications, and pathology. Primary end point was PF rate, as defined by the International Study Group on Pancreatic Fistula. Secondary end points included PF grade, postoperative length of hospital stay, other morbidities, and mortality. RESULTS: Rate of PF for the entire cohort was 17.8%. There were 23 fistulas (24%) in the duct to mucosa cohort and 12 fistulas (12%) in the invagination cohort (p < 0.05). The greatest risk factor for a PF was pancreas texture: PF developed in only 8 patients (8%) with hard glands, and in 27 patients (27%) with a soft gland. There were two perioperative deaths (both in the duct to mucosa group), with the proximate causes of death being PF, followed by bleeding and sepsis. CONCLUSIONS: This dual-institution prospective randomized trial reveals considerably fewer fistulas with invagination compared with duct to mucosa pancreaticojejunostomy after pancreaticoduodenectomy. Results confirm increased PF rates in soft as compared with hard glands. Additional studies are needed to define the optimal technique of pancreatic reconstruction after pancreaticoduodenectomy.
