재단사업

1998 ㅣ -

Aminoglycosides (AG) and vancomycin are frequently prescribed for the treatment of serious bacterial infections. Since both agents have similar pharmacokinetics, it would be desirable to estimate vancomycin's pharmacokinetic parameters by using the patient's measured aminoglycoside pharmacokinetic parameters or vice versa. The objectives of this study were to determine if pharmacokinetic relationships exist between AG and vancomycin with respect to pharmacokinetic parameters in Korean patients in phase I and to evaluate the clinical utility of these ralationships by applying the derived regression equations to a different patient group in Phase II. 81 patients in phase I and 39 in phaseII receiving concurrent intravenous vancomycin and AG were enrolled in this retrospective study. Steady-state serum concentrations of AG and vancomcycin were obtained and pharmacokinetic parameters (Ke, t_1/2, Vd, Cl) were calculated using first-order pharmacokinetic equations. In phase I, there was significant linear correlation between AG and vancomycin pharmacokinetic parameters. The pharmacokinetic parameters of vancomycin derived from AG were superior to those derived from creatinine clearance (Clcr). Correlation coefficient between Ke, t_1/2, Vd, and Cl values of AG and vancomycin were 0.862, 0.874, 0.438, and 0.787, respectively. Correlation coefficient between Ke, t_1/2, Vd, and Cl values of vancomycin and those derived from Clcr were 0.728, 0.622, 0.408, and 0.675, respectively. In phase II, the correlation between pharmacokinetic parameters of AG and vancomycin may be beneficial for predicting pharmacokinetics of vancomycin, especially in Ke when AG concentrations have already been obtained. Regression equations derived from correlation between AG and vancomycin were less biased in Ke and more precise in Vd. But, vancomycin's Vd and Cl from current regression equations were slightly overpredicted. Thus, revised regression equations about prediction of Vd and Cl, and its prospective evaluation will be completed. In conclusion, regression equations derived from AG may be safely used in predicting vancomycin pharmacokinetic parameters.