Premium
Population pharmacokinetic parameters of vancomycin in critically ill patients
Author(s) -
LlopisSalvia P.,
JiménezTorres N. V.
Publication year - 2006
Publication title -
journal of clinical pharmacy and therapeutics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.622
H-Index - 73
eISSN - 1365-2710
pISSN - 0269-4727
DOI - 10.1111/j.1365-2710.2006.00762.x
Subject(s) - nonmem , pharmacokinetics , volume of distribution , medicine , population , intensive care unit , therapeutic drug monitoring , critically ill , renal function , dosing , vancomycin , urology , anesthesia , biology , environmental health , bacteria , genetics , staphylococcus aureus
Summary Background: Intensive care unit patients are a highly heterogeneous population. Accurate dosing for this population requires characterization of the appropriate pharmacokinetic parameters. Objective: To estimate population pharmacokinetic parameters of vancomycin (VAN) in adult critically ill patients and to establish the predictive performance of the resulting model. Patients and method: Fifty critically ill patients with suspected or documented infection with VAN‐sensitive micro‐organisms were included. Thirty patients and 234 serum concentration–time sets obtained during clinical routine monitoring were used to estimate the pharmacokinetic parameters (group A). An open bicompartimental model with intermittent intravenous administration was used to adjust the data. Data were evaluated using a nonlinear mixed effects model ( nonmem software). Forty plasma concentration–time data sets from 20 patients were used for validation using the Bayesian method (group B). Results: There was a linear relationship between creatinine clearance ( Cl cr ) and VAN clearance ( Cl VAN ). The inclusion of the non‐renal clearance ( Cl nr ) (intercept of Cl VAN vs. Cl cr relationship) improved the model significantly ( Cl nr 17 mL/min). The volume of distribution seems to be larger than previously reported: volume of the central compartment ( V c ) was 0·41 L/kg and volume of the peripheral compartment was ( V p ), 1·32 L/kg. The mean error (bias) and mean absolute error (precision) for predicting subsequent peak concentrations were −2·16 and 9·28 mg/L and for trough concentrations, −0·22 and 3·87 mg/L respectively. Conclusion: The use of population‐specific pharmacokinetic parameters and Bayesian forecasting improves dosage–regimen design.