A comparison of simple allometric and maturation models for the prediction of morphine clearance in pediatrics

Abstract

The objective of this study is to predict morphine clearance in children (preterm neonates to 10-month-old infants) by maturation models that include age and weight and to compare the predictive performance of morphine maturation models with simple allometric models. Age, weight, and morphine clearance data were obtained from the literature. A maturation model (n=60) for morphine was developed using data from preterm and term neonates to 5-year-old children. The allometric models were developed using the same data as the maturation model. The predictive performance of the models was tested in 88 children of different age groups. The maturation and allometric models predicted morphine clearance in children with the same degree of accuracy or error. Out of 88 subjects, the prediction error of 50% or less was observed in slightly >60% of the subjects. Almost 40% of the subjects showed a prediction error >50%, 20% of which showed an error ≥ 100%. Although both the maturation and allometric models provided a good prediction of morphine clearance in many children, they were less accurate for many others. High intersubject variability in morphine clearance probably led to less than adequate performance of the models. However, there could be many drugs for which intersubject variability in clearance might not be as high as morphine clearance and in those situations these models could perform reasonably well.