Pre-operative dysglycemia is associated with decreased survival in patients with pancreatic neuroendocrine neoplasms

Background: Diabetes mellitus is associated with increased risk of pancreatic cancer and impaired postresection survival. For pancreatic neuroendocrine neoplasms, no evidence is available for a similar effect of diabetes mellitus. The aim of this study was to evaluate the glycemic profile in patients with pancreatic neuroendocrine neoplasms and to assess the potential impact of glycemic control on the pathology and long-term outcomes in patients undergoing resection of pancreatic neuroendocrine neoplasms. Methods: Pancreatic resections from 2001 to 2017 for pancreatic neuroendocrine neoplasms were analyzed from prospective databases. Blood glucose and HbA1c levels were collected from preoperative tests. Preoperative dysglycemia was defined as a blood glucose >= 140 mg% and/or HbA1c >= 6.5%. Uni- and multivariate analyses were performed according to the presence of perioperative dysglycemia. Survival analyses were performed by Kaplan-Meier curves and Cox-proportional hazards method. Results: Four hundred and seventeen patients were analyzed. Medical history was positive for diabetes mellitus in 88 (21.1%) patients. Blood glucose evaluation identified 30 additional patients without a prior diagnosis of diabetes mellitus who had preoperative dysglycemia. No differences regarding pathologic characteristics or outcomes were detected between diabetics and non-diabetics. Conversely, patients with dysglycemia had greater rates of metastasis (16.8% vs 27.4%; P = .027) as well as vascular, perineural, and lympho-vascular involvement than those with normal blood glucose (89.2% vs 57.4%; P < .001, 90.0% vs 65.1%; P = .046, and 89.3% vs 61.3; P = .006, respectively). Preoperative dysglycemia was associated with impaired overall survival (hazard ratio = 1.57 [1.01-2.46]) and recurrence-free survival (hazard ratio = 1.78 [1.01-3.12]). By multivariate analysis, preoperative dysglycemia was independently associated with recurrence-free survival (hazard ratio 2.32 [1.29-4.17]), together with lymph-node involvement (hazard ratio = 2.01 [1.14-3.57]) and metastatic disease (hazard ratio = 5.10 [2.73-9.55]). Conclusion: Preoperative dysglycemia, but not diabetes mellitus per se, is associated with advanced disease and impaired long-term outcomes in patients undergoing resection for a pancreatic neuroendocrine neoplasm. For those patients, closer surveillance and strict glycemic control are warranted. (C) 2019 Elsevier Inc. All rights reserved.