Back to 2011 Program

Background Hyperglycemia is a risk factor for perioperative morbidity and mortality. Recent studies evaluate the role of hyperglycemia on postoperative outcomes in patients undergoing cardiac, vascular, and colorectal surgery, yet none specifically evaluate patients undergoing pancreatic surgery. Objective The aim of this study was to examine the impact of postoperative hyperglycemia on patient outcomes after pancreatectomy. Methods We reviewed the medical records of all patients who underwent pancreaticoduodenectomy (PD) and distal pancreatectomy (DP) between January 1, 2008 and December 31, 2009. We gathered data on demographics, blood glucose during the first 72 postoperative hours, and perioperative complications for all preoperative non-diabetic patients. Three-day mean blood glucose measurements and episodes of hyper- (blood glucose greater than 200 mg/dL) and hypo- (blood glucose less than 70 mg/dL) glycemia were recorded. Continuous variables were compared using the Wilcoxon rank-sum test. Categorical variables were compared using chi-square or Fisher’s exact test. Results A total of 201 patients underwent either PD (118) or DP (84). The mean age of patients was 62.9 years with the majority being female (56%) and white (76%). The mean body mass index was 25.7 kg/m². Ninety-nine (49%) patients received an insulin infusion postoperatively. Eleven (6%) patients experienced an episode of hypoglycemia without serious sequelae. All patients had an average of 5.6 blood glucose measurements daily. During the first 72 postoperative hours, 81 (40%) patients had mean blood glucoses greater than 140 mg/dL. Patients who had three-day mean blood glucoses greater than 140 mg/dL experienced a significantly greater rate of overall complications (57% vs. 41%, p=0.03), including more frequent infectious complications (37% vs. 17%; p<0.01), major complications (31% vs. 10%, p<0.01), and pancreatic fistulae (17% vs. 8%, p=0.04) than patients who had three-day means less than or equal to 140 mg/dL. Patients with higher mean blood glucose measurements also had significantly longer median lengths of stay (8 days vs. 7 days, p<0.01) compared to patients with lower mean blood glucose measurements. During the first 24 postoperative hours, 72 (36%) patients had a blood glucose measurement greater than 200 mg/dL. Patients who had an episode of hyperglycemia experienced a significantly greater rate of pancreatic fistulae (19% vs. 7%, p=0.01) than patients who did not.Conclusion Among non-diabetic patients, even modest postoperative hyperglycemia following PD or DP is associated with increased morbidity and longer lengths of stay. Careful attention to adequate glycemic control during the early postoperative period may lead to better outcomes in patients undergoing pancreatectomy.