Postoperative acute pancreatitis as a major determinant of postoperative delayed gastric emptying after pancreaticoduodenectomy

Abstract

The aim of this study was to prospectively analyze the possible association of delayed gastric emptying and postoperative pancreatic complications after pancreaticoduodenectomy. Although hospital mortality after pancreaticoduodenectomy is minimal, morbidity is still high; delayed gastric emptying is one of the most frequent complications. Thirty-nine consecutive patients undergoing pancreaticoduodenectomy were included in this study: 14 females and 25 males (median age 65 years; range, 7-82). Delayed gastric emptying was defined as the need for a nasogastric tube or recurrent vomiting that prevented normal feeding on the 10th postoperative day. Blood analysis was performed on postoperative days 4, 6, and 10; Gastrografin examination on day 6; CT scan on days 2 and 5; and drain amylases were measured on day 5. Pancreatitis was defined as pancreatitis changes in CT scan interpreted by an experienced radiologist without knowing other data. Pancreatic fistula was defined according to the recent international recommendations. We had no mortality. Twelve patients (31%) developed delayed gastric emptying. Surgical (9/12 vs. 5/27; P = 0.001) but not medical complications occurred more often in the delayed gastric emptying group. Of the single complications, postoperative CT-detected pancreatitis (6/12 vs. 4/27; P = 0.03) and postoperative pancreatic fistula (5/12 vs. 1/27; P = 0.0007) were significantly associated with delayed gastric emptying compared with the patients without delayed gastric emptying. This pancreatitis was already detected in CT scan on day 2 in most patients (6/10, 60%). In delayed gastric emptying patients, the only parameters in blood analysis that differed significantly from patients without this complication were serum amylase activity (mean +/- SEM, 715 +/- 205 vs. 152 +/- 70 IU/L; P = 0.02), blood leukocyte count (16 +/- 2 vs. 9 +/- 0.6 x 10(9)/L; P = 0.007) and serum C-reactive protein (CRP) concentration (144 +/- 28 vs. 51 +/- 14 mg/L, P = 0.01). Postoperative pancreatic (subclinical) fistula was also associated with postoperative pancreatitis (6/10 vs. 0/29; P = 0.003). Preoperative coronary artery disease (OR = 16; 95% CI, 1.0-241; P = 0.05) and soft pancreatic texture at operation (OR = 9; 95% CI, 1.4-52; P = 0.02) were significant risk factors for the development of postoperative pancreatitis. The diagnosis of delayed gastric emptying after pancreaticoduodenectomy often follows postoperative pancreatitis. Delayed gastric emptying is also associated with postoperative pancreatic fistula, for which this pancreatitis seems to be a risk factor. Preoperative coronary artery disease and soft texture of the pancreas are significant risk factors for postoperative CT-detected pancreatitis.