Prevalence of dysglycemia and associated risk factors in patients with pancreatic benign and low-grade malignant tumors before pancreatic surgery: A prospective observational study

Abstract

BackgroundPancreatic benign and low-grade malignant tumors (PBLMT) have experienced a rapid increase in incidence rates worldwide. Few studies have focused on the glucose metabolism status of patients with PBLMT before pancreatic surgery.MethodsFrom August 2017 to June 2018, 70 patients with PBLMT were prospectively screened for abnormalities in glucose metabolism by an oral glucose tolerance test (OGTT) before pancreatic surgery. Patients were classified as having normal glucose tolerance (NGT), prediabetes mellitus (pre-DM), or new-onset DM (NOD) according to the American Diabetes Association (ADA) criteria. Glucose metabolism indices were calculated based on the OGTT parameters. Tumor volume and remnant pancreatic volume (RPV) were measured by computed tomography.ResultsForty-nine of 70 patients with PBLMT developed dysglycemia (pre-DM and NOD). RPV was smaller in the pre-DM (57.44 ± 18.20 cm3 vs. 70.48 ± 14.08 cm3, P = 0.001) and NOD groups (37.38 ± 20.40 cm3 vs. 70.48 ± 14.08 cm3, P < 0.001) than in the NGT group. The homeostasis model assessment of β-cell function (HOMA2-β), insulinogenic index (IGI), and insulin secretion/insulin resistance index (ISSI-2) were worse in the pre-DM and NOD groups compared with NGT group (all P < 0.05). After univariate and multivariate analyses, age over 60 years (P = 0.049, OR = 5.76, 95% CI: 1.01-32.92) and RPV less than 49.36 cm3 (P = 0.024, OR = 8.59, 95% CI: 1.34-55.22) were recognized as independent risk factors for dysglycemia. The analysis of all patients revealed inverse correlations between RPV and both in age (r = -0.28, P = 0.019) and tumor volume (r = -0.28, P = 0.032). Positive correlations were found between RPV and both IGI (r = 0.29, P = 0.019) and ISSI-2 (r = 0.39, P = 0.0011).ConclusionIn patients with PBLMT, 70% had dysglycemia before surgery. Old age and a reduction in RPV were independent risk factors for developing dysglycemia before pancreatic surgery. The decisions to treat PBLMT with resection should hinge more on the risk of dysglycemia as well as potential malignancy.