BACKGROUND Dopamine is an agonist of alpha, beta, and dopaminergic receptors with varying hemodynamic effects depending on the dose of drug being administered. The purpose of this study was to measure plasma concentrations of dopamine in a homogeneous group of healthy male subjects to develop a pharmacokinetic model for the drug. Our hypothesis was that dopamine concentrations can be predicted from the infusion dose using a population-based pharmacokinetic model. METHODS Nine healthy male volunteers aged 23 to 45 yr were studied in a clinical research facility within our academic medical center. After placement of venous and arterial catheters, dopamine was infused at 10 microg x kg(-1) x min(-1) for 10 min, followed by a 30-min washout period. Subsequently, dopamine was infused at 3 microg x kg(-1) x min(-1) for 90 min, followed by another 30-min washout period. Timed arterial blood samples were centrifuged, and the plasma was analyzed by high-performance liquid chromatography. Mixed-effects pharmacokinetic models using NONMEM software (NONMEM Project Group, University of California, San Francisco, CA) were used to determine the optimal compartmental pharmacokinetic model for dopamine. RESULTS Plasma concentrations of dopamine varied from 12,300 to 201,500 ng/l after 10 min of dopamine infusion at 10 microg x kg(-1) x min(-1). Similarly, steady-state dopamine concentrations varied from 1,880 to 18,300 ng/l in these same subjects receiving 3-microg x kg(-1) x min(-1) infusions for 90 min. A two-compartment model adjusted for body weight was the best model based on the Schwartz-Bayesian criterion. CONCLUSIONS Despite a homogeneous population of healthy male subjects and weight-based dosing, there was 10- to 75-fold intersubject variability in plasma dopamine concentrations, making standard pharmacokinetic modeling of less utility than for other drugs. The data suggest marked intraindividual and interindividual variability in dopamine distribution and/or metabolism. Thus, plasma dopamine concentrations in patients receiving dopamine infusion at identical rates may vary profoundly. Our data suggest that dosing dopamine based on body weight does not yield predictable blood concentrations.