Testing the carbohydrate-insulin model: Short-term metabolic responses to consumption of meals with varying glycemic index in healthy adults.

Abstract

The carbohydrate-insulin model predicts that meals with varying glycemic indices will elicit distinct metabolic and hunger responses, including greater intake at subsequent meals following high-glycemic-index meals. To test this, a randomized trial (NCT05804942) was conducted in healthy adults using intervention meals with low, medium, and high glycemic indices and a constant macronutrient composition. After intake of the intervention meals, glucose and insulin followed the predicted pattern, but subjective hunger did not. At the group level, low glycemic index meals led to lower energy intake changes. At the individual level, energy intake changes were unrelated to body fatness or levels of glucose, β-hydroxybutyrate, free fatty acids, L-lactate, leptin, adrenaline, glucagon-like peptide-1, glucagon, and insulin-glucagon ratio. A weak negative association was observed between energy intake changes and insulin or insulin-glucagon ratio at 300 min, opposite to the model's prediction. These data provide little support for the carbohydrate-insulin model.