Dopamine is not an independent risk factor for reduced amikacin clearance in extremely low-birth-weight infants

Introduction. Important inter-individual variability in amikacin clearance was observed in preterm infants, only in part explained by gestational age (GA), birth weight, or coadministration of nonselective cyclo-oxygenalse (COX) inhibitor. We therefore evaluated whether dopamine had an additional effect on amikacin clearance. Methods: Clinical characteristics (GA, weight COX inhibitor, dopamine, prenatal betamethasone) and amikacin pharmacokinetics were retrospectively collected in a cohort of preterm infants (GA of <31 wks, early neonatal life on respiratory support between January 1, 1999 and January 6, 2005). Pharmacokinetics were calculated by assuming a one-compartment model with instantaneous input and first-order output based on paired samples collected for therapeutic drug monitoring before and following second administration. Monovariate analysis (Spearman, Mann-Whitney U test) was used to study the impact of clinical characteristics on amikacin clearance, and logistic regression was used to assess their potential independent effect. Results. Paired amikacin samples were available for 240 neonates (mean GA, 28 wks; birth weight, 1042 g). Amikacin clearance was 0.46 (range, 0.09-2.33) mL/kg/min and distribution volume was 0.54 (range, 0.17-2.31) L/kg. GA, birth weight, COX inhibitor, and dopamine had a significant effect on amikacin clearance. In a logistic regression model, dopamine was no longer a significant variable when GA, birth weight, or cotreatment of a nonselective COX inhibitor was entered as second variable. Conclusions: Dopamine is an indicator but not an independent marker of reduced amikacin clearance in early neonatal life in extremely low-birth-weight infants. Therefore, neither dose nor interval should be adapted when dopamine is prescribed, if GA and coadministration of nonselective COX inhibitors already have been taken into account.