Population Pharmacokinetics of Amikacin in Korean Clinical Population
Seong Bok Jang(1), Min Soo Park(2), Young Goo Song(3), Kyungsoo Park(1)
(1)Department of Pharmacology, College of Medicine, Yonsei University, Seoul, Korea; (2)Paediatrics, College of Medicine, Yonsei University, Seoul, Korea; (3)Internal Medicine, College of Medicine, Yonsei University, Seoul, Korea
Objectives: This study aimed to develop a pharmacokinetic model of amikacin and to assess the influence of demographic and clinical covariates in Korean patients.
Methods: Amikacin pharmacokinetics was studied in 308 Korean adult patients who received 125-1000mg once- or twice-daily dosing of amikacin. Peak and trough plasma samples at steady state were drawn in every patient, with the peak sample drawn 30-60 minutes after the completion of 30 minute infusion of amikacin and the trough sample drawn within 60 minutes before the next infusion. The 308 patients were randomly allocated into an index dataset (n=200) and a validation dataset (n=108).Covariate selection was made using a step-wise approach within NONMEM 7, using forward addition and backward deletion followed by model refinement . The predictive performance of the developed model was evaluated by the percent prediction error defined as the typical predicted concentration minus the measured concentration divided by the typical predicted concentration .
Results: Amikacin population pharmacokinetics was best described by a one-compartment model with a proportional inter-individual error model and a combined intra-individual error model, and the FOCE interaction method was used. For covariate selection, the effects of creatine clearance and ward (intensive care unit versus general ward) were found significant for clearance, and the effects of body weight and cholecystitis were found significant for volume of distribution, with creatine clearance most significant (p<<0.0001), and body weight next (p<0.0001) although somewhat different from the finding in the literature. The estimates of pharmacokinetic parameters for a typical individual were 2.5 L/hr for clearance, and 15.4 L for volume of distribution. Interindividual variabilities (CV%) were 32% and 9% for clearance and volume of distribution, respectively. The mean (sd) of percent prediction errors was 0.56(26.9)% at peak concentrations and -108(473)% at trough concentrations, which were not significantly different from zero (p=0.527 and 0.095, respectively).
Conclusions: Our results show that the developed population pharmacokinetic model may be used as a basis to find an optimal amikacin dose in Korean patient population without a significant bias. Further studies will be needed to validate the proposed results.
 Wählby U, Jonsson EN., Karlsson MO. Comparison of Stepwise Covariate Model Building Strategies in Population Pharmacokinetic-Pharmacodynamic Analysis. AAPS PharmSci. 2002;4(4):1-12
 Rosario MC, Thomson AH, Jodrell DI, Sharp CA, Elliott HL. Population pharmacokinetics of gentamicin in patients with cancer. Br J Clin Pharmacol. 1998;46(3):229-36.
 Lugo G, Castañeda-Hernández G. Amikacin Bayesian forecasting in critically ill patients with sepsis and cirrhosis. Ther Drug Monit. 1997;19(3):271-6.