Endogenous glucose production from infancy to adulthood: A non-linear regression model

Objective To construct a regression model for endogenous glucose production (EGP) as a function of age, and compare this with glucose supplementation using commonly used dextrose-based saline solutions at fluid maintenance rate in children.

Design A model was constructed based on EGP data, as quantified by [6,6-2H2] glucose dilution after fasting overnight during normoglycaemia, in 40 healthy subjects aged 2.5-54.3 years old. The data were analysed using non-linear regression modelling with a 1-phase exponential decay curve fit. This model was compared to the amount of glucose provided with 2.5% or 5% dextrose-based saline solutions infused at fluid maintenance rate.

Results Non-linear regression analysis of the EGP data yielded the following regression model: EGP (mg/kg/min) = 6.50 × 2.72^{-0.145 × age (y)}+1.93. Glucose supplementation at fluid maintenance rate with a 5% dextrose-based saline solution ranged from 46% at age 1 year to 55% at age 18 years of the glucose required to preclude the need for EGP. With a 2.5% dextrose-based solution, these percentages are 23% at age 1 year to 27% at age 18 years.

Conclusions we present an accurate non-linear regression model for EGP as a function of age. With standard dextrose-based saline solutions infused at fluid maintenance rate, only approximately 50% or less of EGP is provided. With prolonged infusion of these solutions, the deficit between exogenous glucose supplementation and EGP may induce a catabolic state and may ultimately lead to hypoglycaemia, especially in younger children.