Modeling the glucose regulatory system in extreme preterm infants

Area of Science:

• Neonatal physiology
• Mathematical modeling
• Critical care medicine

Background:

• Premature infants often experience disturbed blood glucose homeostasis due to immature regulatory systems and physiological stress.
• Hypoglycemia and hyperglycemia are common in very low birth weight infants and more mature neonates.
• A model of the neonatal glucose regulatory system is needed for improved glycemic control strategies.

Purpose of the Study:

• To adapt an existing metabolic system model from adult critical care for neonatal physiology.
• To identify time-varying insulin sensitivity and glucose uptake profiles in neonates using integral-based fitting methods.
• To assess the predictive accuracy of the adapted model for clinical decision support.

Main Methods:

• Adapted a metabolic system model from adult critical care to neonatal physiology.
• Employed integral-based fitting to determine time-varying insulin sensitivity and non-insulin mediated glucose uptake.
• Evaluated model predictive ability using clinical data (1091 glucose measurements, 3567 patient hours, N=25) by comparing simulated vs. actual glucose values at 1-, 2-, and 4-hour intervals.

Main Results:

• The model achieved a median absolute percentage error of 2.4% for glucose value fitting.
• Median absolute prediction errors were 5.2% at 1 hour, 9.4% at 2 hours, and 13.6% at 4 hours.
• Low fitting and prediction errors demonstrate the model's accuracy in capturing neonatal glucose dynamics.

Conclusions:

• The developed model accurately captures and predicts neonatal metabolic dynamics for effective glycemic control decision support.
• The adaptation of adult critical care models to neonates is feasible and supported by literature data.
• Similar mathematical models can describe glucose metabolism dynamics in both premature neonates and critical care adults, indicated by low errors.