Pancreaticojejunostomy versus alternative treatment of the pancreatic stump after pancreaticoduodenectomy: a comparative analysis of early postoperative outcome

The operative management of the pancreatic stump after pancreaticoduodenectomy has been shown to be an important factor influencing the postoperative development of pancreatic fistula. Thus far, there is no ideal technique for reconstruction, and end-to-end pancreaticojejunostomy (PJS) represents the preferable method. Comparative analysis of early postoperative outcome was done between two groups of patients who underwent either end-to-end PJS or pancreatic remnant ligation (PRL) after pancreaticoduodenectomy. Between January 1997 and December 2001, 39 consecutive patients underwent pancreaticoduodenectomy at the 1st Department of Surgery, University of Athens Medical School. All operations were performed or supervised by two senior surgeons, and all patients underwent a Whipple's procedure. After pancreaticoduodenectomy, 23 patients underwent end-to-end PJS (PJS group), whereas the remaining 16 patients underwent PRL without pancreatic reconstruction (PRL group). We compared the two groups in terms of patients' characteristics, clinical presentation, serum laboratory values on admission, operative details, and postoperative course. The morbidity and mortality rates were 15.4% and 5.1%, respectively, for the whole series. In the PJS group, the morbidity rate was 8.7%, the pancreatic fistula formation rate was 4.3%, and the mortality rate was 4.3%. In the PRL group, the morbidity rate was 25%, the pancreatic fistula formation rate was 12.5%, and the mortality rate was 6.25%. These differences were not statistically significant. There were two deaths in the whole series (one in each group); however, none of the deaths were related to pancreatic fistula formation. Hospital stay was similar in both groups. Both PJS and PRL are valid surgical options that correlate with acceptable postoperative incidence of pancreatic fistula formation, morbidity, and mortality rates. Although PRL avoids the construction of the most risky anastomosis, the results of this study show that early postoperative results after PRL are not superior to PJS; therefore, the method should not be considered as the treatment of choice for the pancreatic stump after a Whipple's procedure. Meticulous surgical technique, surgical experience, and close postoperative care are essential for a successful outcome after this major abdominal operation.