Model-based analysis of postprandial glycemic response dynamics for different types of food

Abstract

•Knowledge of postprandial glycemic response (PPGR) dynamics is important in nutrition research and diabetes management.•Large variability in PPGRs to different types of food is inadequately predicted by existing glycemic measures.•PPGR dynamics are quantified more accurately by extending the AUC with kinetic properties (peak height, width, rise time).•The physiology-based dynamic model (PBDM) integrates physiological knowledge with postprandial glucose and insulin data.•The PBDM describes PPGR dynamics more precise than the kinetic properties and accounts for sparse and heterogeneous data.•The PBDM can be adjusted to simulate impaired glucose tolerance and insulin resistance. Background & aimsKnowledge of postprandial glycemic response (PPGR) dynamics is important in nutrition management and diabetes research, care and (self)management. In daily life, food intake is the most important factor influencing the occurrence of hyperglycemia. However, the large variability in PPGR dynamics to different types of food is inadequately predicted by existing glycemic measures. The objective of this study was therefore to quantitatively describe PPGR dynamics using a systems approach.MethodsPostprandial glucose and insulin data were collected from literature for many different food products and mixed meals. The predictive value of existing measures, such as the Glycemic Index, was evaluated. A physiology-based dynamic model was used to reconstruct the full postprandial response profiles of both glucose and insulin simultaneously.ResultsWe collected a large range of postprandial glucose and insulin dynamics for 53 common food products and mixed meals. Currently available glycemic measures were found to be inadequate to describe the heterogeneity in postprandial dynamics. By estimating model parameters from glucose and insulin data, the physiology-based dynamic model accurately describes the measured data whilst adhering to physiological constraints.ConclusionsThe physiology-based dynamic model provides a systematic framework to analyze postprandial glucose and insulin profiles. By changing parameter values the model can be adjusted to simulate impaired glucose tolerance and insulin resistance. Knowledge of postprandial glycemic response (PPGR) dynamics is important in nutrition management and diabetes research, care and (self)management. In daily life, food intake is the most important factor influencing the occurrence of hyperglycemia. However, the large variability in PPGR dynamics to different types of food is inadequately predicted by existing glycemic measures. The objective of this study was therefore to quantitatively describe PPGR dynamics using a systems approach. Postprandial glucose and insulin data were collected from literature for many different food products and mixed meals. The predictive value of existing measures, such as the Glycemic Index, was evaluated. A physiology-based dynamic model was used to reconstruct the full postprandial response profiles of both glucose and insulin simultaneously. We collected a large range of postprandial glucose and insulin dynamics for 53 common food products and mixed meals. Currently available glycemic measures were found to be inadequate to describe the heterogeneity in postprandial dynamics. By estimating model parameters from glucose and insulin data, the physiology-based dynamic model accurately describes the measured data whilst adhering to physiological constraints. The physiology-based dynamic model provides a systematic framework to analyze postprandial glucose and insulin profiles. By changing parameter values the model can be adjusted to simulate impaired glucose tolerance and insulin resistance.