Evaluation of clinical and imaging biomarkers for the prediction of new onset diabetes following pancreatic resection.

PURPOSE: To identify clinical and imaging biomarkers that can predict the new onset of diabetes mellitus (NODM) within 1 year of pancreatic resection. METHODS: A retrospective chart review was conducted of 426 non-diabetic patients who underwent a pancreaticoduodenectomy or distal pancreatectomy at the University of Pittsburgh Medical Center between 2006 and 2016. Clinical characteristics and the patient's diabetic status at 1-year post resection were collected from the EMR. Imaging biomarkers included hepatic and pancreatic fat replacement, pancreatic calcifications, pancreatic duct diameter, pancreatic volume and body composition. Univariate and multivariable analyses were performed to demonstrate any predictive biomarkers of diabetes occurrence within 1 year of pancreatic resection. RESULTS: 135/426 (31.7%) patients developed NODM. The only significant clinical predictor was older age (OR 1.02, 95% CI 1.002-1.039, p = 0.032). Imaging characteristics found to be significant included hepatic steatosis (OR 1.777, 95% CI 1.094-2.886, p = 0.02), larger reduction in pancreas volume (OR 0.989, 95% CI 0.979-0.999, p = 0.027), and greater preoperative visceral fat (OR 1.004, 95% CI 1.001-1.006, p = 0.001). CONCLUSION: Age, presence of hepatic steatosis, change in pancreatic volume, and preoperative visceral fat are independent predictive biomarkers for NODM following pancreatic resection.

Long-term health after pancreatic surgery: the view from 9.5 years.

BACKGROUND: Discussing the impact of pancreatic surgery on long-term health is poorly understood, but necessary for informed consent. Given the increased number of pancreatic operations being performed annually, further investigation is necessary. METHODS: Patients surviving longer than 5 years after pancreatic surgery were surveyed for postoperative hospitalizations, operations, pain, nutrition and diabetes. Variables were analyzed according to patient and peri-operative variables, and validated using medical records. RESULTS: Eighty individuals completed the survey; median follow-up was 9.5 years (IQR:6.43,12.73). 47.5% underwent a pancreatoduodenectomy, and 25.0% a distal pancreatectomy; 40.0% had adenocarcinoma. 57.1% reported long-term weight loss, of which 65.9% was unintentional. While 1.3% took pancreatic enzymes before surgery, 38.8% utilized after. 12.5% had diabetes before, and 28.6% after surgery; 22 of 30 patients required insulin replacement therapy (73.3%). 41.3% reported hospitalizations, 17.5% required endoscopies and 28.8% additional operations after full recovery. Need for additional interventions were not related to pathology or post-operative complications, but were more common among patients undergoing a Whipple. CONCLUSION: More than half of patients will have a long-term medical complication attributable to pancreatectomy. In comparison to the literature, it may be inferred that consequences occur within the first few years after surgery, and do not compound over time.