Anatomy-specific pancreatic stump management to reduce the risk of pancreatic fistula after pancreatic head resection.

BACKGROUND

The anatomical status of the pancreatic remnant after a pancreatic head resection varies greatly among patients. The aim of the present study was to improve management of the pancreatic remnant for reducing pancreatic fistula after pancreatic head resection.

METHODS

Ninety-five consecutive patients who underwent an end-to-side, duct-to-mucosa pancreaticojejunostomy after pancreatic head resection were included in the study. To approximate the pancreatic stump to the jejunum, the transfixing and interrupted suture techniques were used in 51 and 44 patients, respectively. We modified the interrupted suture technique according to the anatomical status of the pancreatic remnant, i.e., the shape of the pancreatic stump and the location of the pancreatic duct.

RESULTS

There was no operative mortality in this study. Overall, 14 patients (15%) developed a clinically relevant pancreatic fistula. Certain anatomical features, including a small pancreatic duct, a soft, nonfibrotic pancreatic gland, and a pancreatic duct adjacent to the posterior cut edge, were significantly associated with pancreatic fistula. The fistula rate in the interrupted suture group was 7%, lower than that (22%) in the transfixing suture group (P = 0.036), and it was not influenced by pancreatic anatomy. Multivariate analysis identified a nonfibrotic pancreas (versus fibrotic pancreas; odds ratio [OR] 12.58, 95% CI 1.2-23.9; P = 0.001), a soft pancreas (versus hard pancreas; OR 4.67, CI 1.2-51.1; P = 0.006), and the transfixing suture technique (versus interrupted suture technique; OR 9.91, CI 1.7-57.5; P = 0.003) as significant predictors of clinically relevant pancreatic fistula.

CONCLUSIONS

Pancreatic anastomosis modified according to the pancreatic anatomy is effective in reducing the risk of pancreatic fistula formation with end-to-side, duct-to-mucosa pancreaticojejunostomy after pancreatic head resection.