POPULATION PHARMACOKINETIC MODEL FOR GENTAMYCIN AND ITS PREDICTIVE VALUE

AIM To construct a population pharmacokinetic model to describe gentamycin concentrations in serum in newborn infants and to validate the predictive ability of this model. METHODS Data used in this study were obtained from 30 neonates with 80 serum samples. A one-compartment open model was used to describe the kinetics of gentamycin after intravenous infusion. Following Sheiner's idea of population pharmacokinetics, a programs to estimate population parameter and individual parameter of gentamycin was made. The target function minimality was obtained from Monte Carlo algorithm. The predictive ability of the developed model was evaluated by computing precision and accuracy of serum concentration predicted using the parameter estimates. RESULTS Fitted population pharmacokinetic parameters (mean±standard deviation) were as follows:k_e: 0.220±0.022 h^-1, V_d: 0.51±0.06L·kg^-1, Cl: 112±10 mL·h^-1·kg^-1. For the population analysis sample and the predictive sample, predicted and observed concentrations were all close with correlation coefficient 0.920 and 0.946, respectively. Mean prediction error (ME) and root mean squared error (RMSE) were 0.001 mg·L^-1 and 0.84 mg·L^-1 for the predictive sample, respectively. CONCLUSION Observed gentamycin serum concentrations were explained very well by this model. We propose the use of this population pharmacokinetic model to optimize gentamycin clinical therapies in our institution and others with similar patient population characteristics.