MON-184 Hypoglycemia During Mixed Meal Tolerance Testing and Early Glucose Dysregulation in Cystic Fibrosis

Post-prandial and oral glucose tolerance test (OGTT)-related hypoglycemia is common in cystic fibrosis (CF). The mechanisms underlying this “reactive” hypoglycemia are unclear. We hypothesized that late hypoglycemia would be associated with higher peak plasma glucose, impaired early-phase insulin secretion compared to PI-NGT, and augmentation of late insulin secretion. To understand these relationships, we performed standardized mixed-meal tolerance tests (MMTT) with timed blood sampling for determination of plasma glucose, insulin and C-peptide in non-diabetic adolescents and young adults with pancreatic insufficient CF (PI-CF). Glucose tolerance was defined by OGTT as: normal (NGT; 1-hour glucose <155 and 2-hour glucose <140mg/dL, n=13), or abnormal (AGT; 1-hour glucose ≥155 or 2-hour glucose ≥140 mg/dL and <200 mg/dL, n=21). Hypoglycemia was defined as plasma glucose <70mg/dL during MMTT and used to assign subjects to exposure groups for analysis. Insulin secretory rates (ISRs) were derived by parametric deconvolution of C-peptide kinetics using a two-compartment model. Glucose and ISR incremental areas under curve (Glc-AUC; ISR-AUC) were calculated. Peak plasma glucose and ISR-AUC(interval) were compared between hypoglycemia groups [hypo(+) vs hypo(-)] using Student’s t-test and Mann Whitney U Tests, respectively. Linear regression was performed to explore whether ISR AUC was different in hypo(+) vs hypo(-) after adjustment for Glc-AUC. 34 subjects with PI-CF were evaluated. Participants were 50% female, aged 25±8.7 years, with FEV-1 % predicted 81±19, HbA1c 5.5±0.4%, and BMI z-score 0.17±1. Hypoglycemia occurred in 9 subjects (2 NGT, 7 AGT) at 163±30 min following the start of the meal. No group differences in gender, age, lung function, HbA1c, or BMI z-score were found. Peak plasma glucose was higher in hypo(+) vs hypo(-) (215±21 vs 168±33 mg/dL, p=0.0004). Moreover, early-phase ISR-AUC(15-45 min) was lower in hypo(+) vs hypo(-) PI-NGT (609±190 vs 1313±850 µU/mL, p=0.03), including the 2 NGT-hypo(+): 430mg/dL and 548mg/dL, respectively. ISR-AUC(120-180 min) was not different in hypo(+) than hypo(-) PI-AGT (2528 ±1723 vs 2219±1072 µU/mL, p=0.85); however, glc-AUC(120-180 min) was lower in hypo(+) than hypo(-) PI-AGT (-785±1288mg·min/dL vs 750±1115 mg·min/dL, p=0.015). After adjusting for glc-AUC over this interval, hypo(+) tended to have higher late insulin secretion than hypo(-) PI-AGT (p=0.089). Meal-related hypoglycemia in CF is common and not only seen following the glucose load of an OGTT. Hypoglycemia is associated with early glucose dysregulation (higher peak plasma glucose), loss of early-phase insulin secretion (lower ISR AUC(15-45)( min)), and possible late compensatory hyperinsulinemia (higher ISR AUC(120-180 min) relative to glucose). Whether hypoglycemia may predict progression to CFRD in individuals with PI-CF requires further study.