SAT-121 Immediate Dysglycemia After Pancreatic Resection: Prevalence and Risk Factors

Introduction: Postoperative hyper- and hypoglycemia are associated with increased morbidity and mortality following many types of surgical procedures. Long-term development of diabetes mellitus (DM) following pancreatic resection is well documented; however, less is known regarding glycemic control in the immediate postoperative period for patients undergoing pancreatic resection. Methods: We conducted a retrospective study investigating rates of hyperglycemia (blood glucose >140 mg/dL) and hypoglycemia (blood glucose ≤ 70 mg/dL) and associated risk factors in the 10-day postoperative period in patients undergoing pancreatic resection. Patients that underwent pancreaticoduodenectomy (PD), distal pancreatectomy (DP), or complete pancreatectomy (CP) from 7/1/12-7/1/18 were included. Patients with a functional tumor, end-stage renal disease, liver failure, deceased within 14 days of surgery, and solid-organ transplant recipients were excluded. Results: After applying exclusion criteria, 227 of 261 patients were included. Median age was 61 years and 50.2% were male. Preoperative DM was present in 46/227 (20.3%). A total of 130 patients (57.3%) had a pancreatic primary indication for surgery, with other abdominal malignancies accounting for most of the non-pancreatic indications. Most underwent DP (59.5%) followed by PD (40.1%) and CP (0.6%). A total of 162 (71.4%) patients had at least one hyperglycemic episode with 83 (36.6%) having a blood glucose >180 mg/dl. The type of surgical procedure was not a risk factor for hyperglycemia, even when stratified for pancreatic vs non-pancreatic indication. Rates of hyperglycemia were higher in patients with pancreatic cancer (78.5%) compared to those with a non-malignant pancreatic lesion (44.0%), p <0.05. A preoperative diagnosis of DM (p <0.01) and use of TPN or enteral feeds (p <0.01) were risk factors for having any hyperglycemic episode and a hyperglycemic episode >180 mg/dL. Peri-operative steroids and BMI were not associated with hyperglycemia. A total of 46 patients (20.3%) had at least one hypoglycemic episode, including 20 of 105 patients who received insulin. Hypoglycemia risk was higher in patients with BMI < 25 kg/m(2) compared to BMI >= 25 kg/m(2) (p <0.01). Most hypoglycemic episodes happened within 3 days postoperatively (73.0%) and while the patient’s diet order was NPO or ice chips without nutritional supplements (63.5%). Renal function was not associated with risk of hypoglycemia. Conclusions: Hyper- and hypoglycemia are both common after pancreatic resection. Hyperglycemia is especially common in patients with pancreatic cancer, preoperative DM, and/or on supplemental nutrition. Hypoglycemia after pancreatic resection may be largely explained by nutritional status. Understanding risk factors for hyper- and hypoglycemia can assist in targeted methods to reduce rates of dysglycemia after pancreatic resection.