Postoperative morbidity and mortality of pancreaticoduodenectomy for periampullary cancer.

OBJECTIVE

To find out factors that may influence the mortality after Whipple's operation, whether duct to mucosa anastomosis is better than classic "dunking" pancreaticojejunostomy and whether age over 70 is a contraindication for this procedure.

DESIGN

Prospectively randomised study.

SETTING

Teaching hospital, Taiwan, R.O.C.

PATIENTS: 93 patients with periampullary cancer undergoing Whipple's operation were randomly divided into two groups. Forty-six with periampullary cancer underwent invaginating pancreaticojejunostomy, and 47 patients underwent duct to mucosa anastomosis for reconstruction.

MAIN OUTCOME MEASURES: Mortality and morbidity were compared between two groups.

RESULTS

The over all mortality was 8% (7/93). An albumin concentration of less than 30 g/L before operation and operative blood loss influenced the surgical mortality both in the univariate and multivariate analysis. Age over 70 years was not a factor. Patients with duct to mucosa anastomoses had a leak rate of 4% (2/47), morbidity of 21% (10/47) and mortality of 6% (3/47). Patients with an invaginated pancreaticojejunostomy had a leak rate of 15% (7/46), morbidity of 33% (15/46), and mortality of 9% (4/46). The need for total parenteral nutrition in the invaginated group (33%) was statistically greater than in the other group (11%) (p = 0.01).

CONCLUSIONS

The morbidity and mortality of pancreaticoduodenectomy for periampullary cancer although slightly greater for patients over the age of 70 are acceptable. The factors that may influence the mortality are an albumin concentration of less then 30 g/L and the amount of blood lost during operation. The duct to mucosa anastomosis is a safe procedure, which has a lower leak rate and less need for total parenteral nutrition than pancreaticojejunostomy.