INFLUENCE OF CLINICAL-DIAGNOSIS IN THE POPULATION PHARMACOKINETICS OFAMIKACIN IN INTENSIVE-CARE UNIT PATIENTS

The aim of the present study was to analyse the pharmacokinetic behavi our of amikacin in intensive care unit (ICU) patients using a mixed-ef fect model and sparse data collected during routine clinical care. The patient population comprised 158 medical ICU patients divided into tw o groups: one for computing the population model (n = 120) and the oth er for validation (n = 38). A 1-compartment model was used and the fol lowing covariates were tested for their influence on clearance (CL) an d volume of distribution (Vd): age, gender, weight, parenteral nutriti on, creatinine clearance, duration of therapy and clinical diagnosis. The nonlinear mixed-effect model (NONMEM) was used to assess the popul ation pharmacokinetic model of amikacin in this patient population. In this study, the final population model accounting for amikacin pharma cokinetics in ICU patients was: CL = 0.93 CLCR(1 + 0.22 Trauma), Vd = 0.39 TBW (1 + 0.24 Sepsis), where CLCR and TBW corresponded to the pat ients' creatinine clearance and total bodyweight, respectively. The 'T rauma' and 'Sepsis' variables referred to the clinical diagnosis of th e patients. This model was subsequently used to predict amikacin serum levels obtained in the validation population by a priori and Bayesian methods. The predictive performance was adequate for clinical purpose s, pointing to the feasibility of our population model to provide refe rence values for a priori prediction as well as the Bayesian approach for individualisation of amikacin therapy in ICU patients.