Population Pharmacokinetics Of Tobramycin In Neonates
D.Santos Buelga; M.E.Mendez*; M.J.Garcma; M.Pardo*; A.C.Falcao**; A.Dominguez-Gil.
Department of Pharmacy and Pharmaceutical Technology. Faculty of Pharmacy. University of Salamanca. Salamanca. Spain, Hospital del Rio Hortega. Valladolid. Spain, *** Lab. Pharmacology. Faculty of Pharmacy. University of Coimbra. Coimbra. Portugal.
Tobramycin is commomly used in newborn infants with suspected or proved bacterial infection. The serum concentrations of tobramycin are routinely monitored in these patients to individualize therapy. The aim of the present work was the explanation and interpretation from a clinical point a view of the dynamic pharmacokinetics parameters obtained through a mixed effect modeling using the NONMEM software package.
The model-building process was based upon retrospective information from 101 hospitalized patients with a total of 283 measured serum tobramycin concentrations. There were 58 female and 43 male neonates with a birth weight ranged from 0.8 to 4.25 kg, gestational ages from 30 to 42 weeks and postnatal ages until 2 weeks. The pharmacokinetic model used was a one-compartment open model with first-order absorption (i.m. administration) and first-order elimination. The effect of several covariates on clearance (CL) and volume of distribution (V) was investigated (sex, birth weight, birth height, actual weight, gestational age, postnatal age, serum creatinine concentration and concomitant therapy with theophylline).
In the final model the actual weight (WT), in a linear fashion, was the single best estimator of tobramycin clearance and volume of distribution:
CL (L/h) =THETA(1) * WT THETA(1) = 0.0486 (L/h/kg) ETA(CL) = 28.7%
EPSYLON = 37%
V (L) = THETA(2) * WT THETA(2) = 0.658 (L/kg) ETA(V) = 29%
The effect of covariates like postnatal age, sex and theophylline on clearance will be discussed because their inclusion on the final model could be supported by some (but not all) of the statistical criteria used during the model-building process. The predictive capacity of these models (including the final model) will be shown using a validation study in order to obtain the clinical impact of their application in clinical practice. The identification of influential patient factors and the quantification of their influence on tobramycin disposition allows for îa priori" estimates of tobramycin pharmacokinetic parameters in neonates. In addition, this findings can be generally applicable for the dosage recomendations and prospectively to forecast tobramycin concentrations with a Bayesian kinetic parameter estimation programe.