Research
Print page Print page
Switch language
The Capital Region of Denmark - a part of Copenhagen University Hospital
Published

Integrated model of insulin and glucose kinetics describing both hepatic glucose and pancreatic insulin regulation

Research output: Contribution to journalJournal articleResearchpeer-review

  1. Digital questionnaire platform in the Danish Blood Donor Study

    Research output: Contribution to journalJournal articleResearchpeer-review

  2. BCI using imaginary movements: the simulator

    Research output: Contribution to journalJournal articleResearchpeer-review

  1. Intestinal sensing and handling of dietary lipids in gastric bypass-operated patients and matched controls

    Research output: Contribution to journalJournal articleResearchpeer-review

  2. Augmented GLP-1 secretion as seen after gastric bypass may be obtained by delaying carbohydrate digestion

    Research output: Contribution to journalJournal articleResearchpeer-review

  3. GLP-2 and GIP exert separate effects on bone turnover: A randomized, placebo-controlled, crossover study in healthy young men

    Research output: Contribution to journalJournal articleResearchpeer-review

  4. Postprandial Nutrient Handling and Gastrointestinal Hormone Secretion After Roux-en-Y Gastric Bypass vs Sleeve Gastrectomy

    Research output: Contribution to journalJournal articleResearchpeer-review

View graph of relations

BACKGROUND AND OBJECTIVES: Modeling of glucose kinetics has to a large extent been based on models with plasma insulin as a known forcing function. Furthermore, population-based statistical methods for parameter estimation in these models have mainly addressed random inter-individual variations and not intra-individual variations in the parameters. Here we present an integrated whole-body model of glucose and insulin kinetics which extends the well-known two-compartment glucose minimal model. The population-based estimation technique allow for quantification of both random inter- and intra-individual variation in selected parameters using simultaneous data series on glucose and insulin.

METHODS: We extend the two-compartment glucose model into a whole-body model for both glucose and insulin using a simple model for the pancreas compartment which includes feedback of glucose on both insulin secretion and formation of insulin in pancreas. The model has 15 unknown parameters of which 8 have been selected for both intra- and inter-individual variations. The statistical technique for parameter estimation is based on first order conditional estimation.

RESULTS: The model has been evaluated on two datasets: Study group 1 includes 13 healthy subjects with 3-5 repeated IVGTT series of simultaneous plasma glucose and insulin measurements and Study group 2 includes 26 obese patients (3 subgroups: 10 type 2 diabetes (T2D), 7 impaired glucose tolerance (IGT) and 9 normal glucose tolerance (NGT)) with a single IVGTT series. In general the estimated population parameters compares well with reported values in similar studies. Overall the model fits the data series well and the random variation in the 8 selected parameters can account for both intra- and inter-individual variations in the data series. Simulation studies perform reasonable in response to either a slow glucose infusion or a staircase experiment with increasing glucose infusion. Furthermore, the parameters related to the pancreas compartment add useful interpretations in relation to discrimination between populations with varying degree of glucose intolerance.

CONCLUSIONS: We report a new and improved whole-body model of glucose and insulin kinetics which performs robustly under differing conditions and adds useful interpretations in relation to glucose intolerance.

Original languageEnglish
JournalComputer Methods and Programs in Biomedicine
Volume156
Pages (from-to)121-131
ISSN0169-2607
DOIs
Publication statusPublished - Mar 2018

    Research areas

  • Journal Article

ID: 52703624