Maturation of GFR in Preterm and Term Neonates Reflected by Clearance of Different Antibiotics

Document Type

Abstract

Publication Date

2011

Abstract

Objectives: Throughout infancy, renal function matures resulting in differences in glomerular filtration rate (GFR) at different stages of development. These developmental changes in GFR were previously quantified in (pre)term neonates aged up to 1 month on the basis of the clearance of amikacin. In this developmental renal excretion model (1), the maturation of GFR was predicted by birth weight (BWb) and postnatal age (PNA). The aim of this study is to assess model performance when this developmental renal excretion model (1) is used to describe maturation in clearance of other renally excreted antibiotics in (pre)term neonates. Using this approach a distinction is being made between system specific and drug specific information in paediatric pharmacokinetic models.

Methods: For the netilmicin dataset, 386 netilmicin concentrations were available from 97 (pre)term neonates (BWb 470-3000 g, PNA 1-30 days)(2). The vancomycin dataset contained 752 vancomycin concentrations from 273 preterm neonates (BWb 385-2550 g, PNA 1-30 days)(3). A pharmacokinetic model was developed for both netilmicin or vancomycin using the developmental renal excretion model for amikacin clearance in neonates (1):

CLi=CLp*{((BWb/BWBmedian)^1.34)*(1+0.213*(PNA/PNAmedian))}

Using this approach, CLp is considered a drug specific property and was therefore estimated for each of the drugs separately. The remaining information in this equation is considered system specific information which can be applied for all renally excreted drugs. The descriptive and predictive performance of models developed using the developmental renal excretion model (1) were compared with comprehensive covariate models (4) for netilmicin or vancomycin respectively, by evaluation of the objective function (OFV), basic goodness-of-fit plots, NPDE and the individual and population parameter estimates versus most predictive covariate (4).

Results: The descriptive and predictive properties of the models developed using the developmental renal excretion model, were similar compared to the comprehensive covariate models for basic goodness-of-fit plots and NPDE. In agreement the models that were developed using the developmental renal excretion model, in the comprehensive covariate models BWb and PNA were identified as most predictive covariates for clearance. The comprehensive covariate models had only a slightly lower objective function (netilmicin p<0.05, vancomycin p<0.001) compared to the models using the developmental renal excretion model.

Conclusions: Use of the developmental renal excretion model quantifying maturation in GFR mediated amikacin clearance for the analysis of netilmicin and vancomycin clearance in neonates, results in adequate descriptive and predictive performance. We conclude that the application of system specific information may lead to optimization of sparse data analysis in children.


Share

COinS