Macronutrient Regulation of Ghrelin and Peptide YY in Pediatric Obesity and Prader-Willi Syndrome

Abstract

Background: The roles of macronutrients and GH in the regulation of food intake in pediatric obesity and Prader-Willi Syndrome (PWS) are poorly understood.

Objective: We compared effects of high-carbohydrate (HC) and high-fat (HF) meals and GH therapy on ghrelin, insulin, peptide YY (PYY), and insulin sensitivity in children with PWS and body mass index (BMI) -matched obese controls (OCs).

Methods: In a randomized, crossover study, 14 PWS (median, 11.35 y; BMI z score [BMI-z], 2.15) and 14 OCs (median, 11.97 y; BMI-z, 2.35) received isocaloric breakfast meals (HC or HF) on separate days. Blood samples were drawn at baseline and every 30 minutes for 4 hours. Mixed linear models were adjusted for age, sex, and BMI-z.

Results: Relative to OCs, children with PWS had lower fasting insulin and higher fasting ghrelin and ghrelin/PYY. Ghrelin levels were higher in PWS across all postprandial time points (P < .0001). Carbohydrate was more potent than fat in suppressing ghrelin levels in PWS (P = .028); HC and HF were equipotent in OCs but less potent than in PWS (P = .011). The increase in PYY following HF was attenuated in PWS (P = .037); thus, postprandial ghrelin/PYY remained higher throughout. A lesser increase in insulin and lesser decrease in ghrelin were observed in GH-treated PWS patients than in untreated patients; PYY responses were comparable.

Conclusion: Children with PWS have fasting and postprandial hyperghrelinemia and an attenuated PYY response to fat, yielding a high ghrelin/PYY ratio. GH therapy in PWS is associated with increased insulin sensitivity and lesser postprandial suppression of ghrelin. The ratio Ghrelin/PYY may be a novel marker of orexigenic drive.

Trial registration: ClinicalTrials.gov NCT02464514.

Figures

Figure 1.. Glucose, insulin, and insulin/glucose ratio responses to isocaloric high-fat and high-carbohydrate breakfast meals.

Predicted values and SEs were based on linear mixed models. See the Results for statistically significant differences among the experimental groups. CARB, high-carbohydrate meal; FAT, high-fat meal.

Figure 2.. Ghrelin and PYY responses to isocaloric high-fat and high-carbohydrate breakfast meals.

Predicted values and SEs of both observed responses and percent changes relative to baseline (time = 0) of each group at each time point were based on linear mixed models. See the Results for statistically significant differences among the experimental groups. CARB, high-carbohydrate meal; FAT, high-fat meal.

Figure 3.. Ghrelin/PYY ratio responses to isocaloric high-fat and high-carbohydrate breakfast meals.

Predicted values and SEs of both observed responses and percent change relative to baseline (time = 0) of each group at each time point were based on linear mixed models. See the Results for statistically significant differences among the experimental groups. CARB, high-carbohydrate meal; FAT, high-fat meal.

Figure 4.. Glucose, insulin, and insulin/glucose ratio responses to isocaloric high-fat and high-carbohydrate breakfast meals in GH-treated and untreated PWS.

Predicted values and SEs of percent change relative to baseline (time = 0) of each group at each time point were based on linear mixed models. See the Results for statistically significant differences among the experimental groups. CARB, high-carbohydrate meal; FAT, high-fat meal; +GH, treated with GH.

Figure 5.. Ghrelin, PYY, and ghrelin/PYY ratio responses to isocaloric high-fat and high-carbohydrate breakfast meals in GH-treated and untreated PWS.

Predicted values and SEs of percentage change relative to baseline (time = 0) of each group at each time point were based on linear mixed models. See the Results for statistically significant differences among the experimental groups. CARB, high-carbohydrate meal; FAT, high-fat meal; +GH, treated with GH.