Effects of starvation and short-term refeeding on gastric emptying and postprandial blood glucose regulation in adolescent girls with anorexia nervosa.

Postprandial glucose is reduced in malnourished patients with anorexia nervosa (AN), but the mechanisms and duration for this remain unclear. We examined blood glucose, gastric emptying, and glucoregulatory hormone changes in malnourished patients with AN and during 2 wk of acute refeeding compared with healthy controls (HCs). Twenty-two female adolescents with AN and 17 age-matched female HCs were assessed after a 4-h fast. Patients were commenced on a refeeding protocol of 2,400 kcal/day. Gastric emptying (C-octanoate breath test), glucose absorption (3-O-methylglucose), blood glucose, plasma glucagon-like peptide-1 (GLP-1), glucose-dependent insulinotropic polypeptide (GIP), insulin, C-peptide, and glucagon responses to a mixed-nutrient test meal were measured on admission and 1 and 2 wk after refeeding. HCs were assessed once. On admission, patients had slower gastric emptying, lower postprandial glucose and insulin, and higher glucagon and GLP-1 than HCs ( P < 0.05). In patients with AN, the rise in glucose (0-30 min) correlated with gastric emptying ( P < 0.05). With refeeding, postprandial glucose and 3-O-methylglucose were higher, gastric emptying faster, and baseline insulin and C-peptide less ( P < 0.05), compared with admission. After 2 wk of refeeding, postprandial glucose remained lower, and glucagon and GLP-1 higher, in patients with AN than HCs ( P < 0.05) without differences in gastric emptying, baseline glucagon, or postprandial insulin. Delayed gastric emptying may underlie reduced postprandial glucose in starved patients with AN; however, postprandial glucose and glucoregulatory hormone changes persist after 2 wk of refeeding despite improved gastric emptying. Future research should explore whether reduced postprandial glucose in AN is related to medical risk by examining associated symptoms alongside continuous glucose monitoring during refeeding.