Population pharmacokinetics and relationship between demographic and clinical variables and pharmacokinetics of gentamicin in neonates

Population pharmacokinetic parameter estimates were calculated from 725 routine plasma gentamicin concentrations obtained in 177 neonates of 24 to 42 weeks' gestational age in their first week of life. Kel increases and V/W decreases with increasing gestational age. Almost identical results were obtained with iterative two-stage Bayesian fitting (MW\PHARM 3.30) as with a non-parametric maximization algorithm (NPEM2). The effect of various covariates on drug disposition was investigated retrospectively using multiple regression analysis. Predictive power for Kel increases with rising gestational age. For neonates </=28.5 weeks and neonates >28.5 weeks and </=30.9 weeks, the predictive power of the regression equation for Kel was relatively low (r2 respectively 0.270 and 0.364). Better predictivity was found for Kel at gestational ages >30.9 weeks (r2 = 0.482), with gestational age, postnatal age, and Apgar score at 5 minutes being predictors. A very strong correlation existed between volume of distribution and weight (r2 = 0.83). Volume as a function of weight could be described with low predictivity by gestational age and to a lesser degree by Apgar score at 5 minutes (r2 = 0.298). The developed models need appropriate prospective clinical validation.