Background/aims: Pancreatic fistula is a common complication after pancreaticoduodenostomy. Prevention of a concomitant bile leak from hepaticojejunostomy is important because it could lead to more serious complications including intraperitoneal abscess, subsequent sepsis and massive hemorrhage by activating pancreatic fistula. This study was designed to determine perioperative risk factors of the hepaticojejunostomy leak for the purpose of decreasing this morbidity.
Methodology: Clinical records of 107 consecutive pancreaticoduodenal resections were reviewed. hepaticojejunostomy anastomoses were performed using absorbable sutures in an end-to-side, single-layer and interrupted fashion. A total of 8 presumed perioperative risk factors were analyzed. They included advanced age, low serum albumin, low serum total cholesterol, impaired glucose tolerance and placement of a biliary drainage catheter as preoperative factors, and dilated common hepatic duct and undone anastomotic leak test as intraoperative factors. In addition, transanastomotic stenting techniques including retrograde transhepatic bile drainage, T-tube and transjejunal drainage were compared with respect to hepaticojejunostomy leak rates.
Results: Hepaticojejunostomy leak was demonstrated in 9 patients (8%). Anastomotic leak testing only achieved statistical significance (p = 0.04). It is noteworthy that no hepaticojejunostomy leak developed among 28 patients who underwent this test through a retrograde transhepatic bile drainage catheter. In addition, the frequency of bile leaks (14/107) associated with the transanastomotic stenting techniques urged the necessity of appropriate intraperitoneal drain placement.
Conclusions: Careful anastomotic procedures with a subsequent anastomotic leak test most effectively prevent hepaticojejunostomy leak after pancreaticoduodenal resection.