Repeatability of insulin sensitivity and glucose effectiveness from the minimal model. Implications for study design

Area of Science:

• Metabolic research
• Endocrinology
• Physiological modeling

Background:

• Insulin sensitivity is crucial for metabolic health.
• The minimal model is a common tool for assessing insulin sensitivity.
• Repeatability of measurements is essential for reliable clinical and research applications.

Purpose of the Study:

• To evaluate the interday repeatability of insulin sensitivity (SI) measurements derived from the minimal model.
• To compare the repeatability of SI measurements using a full 30-sample dataset versus a reduced 12-sample dataset from frequently sampled intravenous glucose tolerance tests (FSIGTs).

Main Methods:

• 11 healthy men underwent three FSIGTs over 12 days with controlled diet and activity.
• Minimal model analysis was performed using both 30-sample and 12-sample datasets.
• Interday coefficients of variation (CV) were calculated for insulin sensitivity (SI), glucose effectiveness (SG), and insulin secretion parameters.

Main Results:

• Minimal model SI measurements showed an average interday CV of 20.2% with the full dataset.
• Glucose effectiveness (SG) and insulin secretion parameters exhibited similar or slightly higher variability.
• SI calculated from the reduced 12-sample dataset was not significantly different in mean value but showed a trend towards higher variability (27.7% CV).

Conclusions:

• The minimal model provides repeatable measurements of insulin sensitivity (SI) in normal men.
• While a reduced 12-sample FSIGT dataset yields similar mean SI, it may reduce measurement precision.
• Careful consideration of data set size is warranted when applying the minimal model for population studies.