Standard doses of piperacillin/tazobactam (9-13.5 g over 24 h) administered by continuous infusion (CI) routinely provide serum concentrations in excess of the susceptibility breakpoint (< or =16/4 micro g/ml) for most Enterobacteriaceae. Since the breakpoint of this agent for Pseudomonas aeruginosa is considerably higher (< or=64/4 micro g/ml), the likelihood of obtaining adequate drug exposures with these doses against this bacterium is currently unknown. Monte Carlo simulation was utilized to determine the probability of obtaining adequate piperacillin concentrations above its MICs for P. aeruginosa in patients receiving CI. MICs of 557 P. aeruginosa isolates were determined by E-test and a distribution was constructed for the 496 susceptible isolates. Using a previously validated population pharmacokinetic equation, steady-state serum concentrations were estimated for 210 patients who received piperacillin/tazobactam via CI. A Monte Carlo simulation was performed to predict the probability of obtaining concentrations at the MIC, 2 x MIC, 4 x MIC, 5 x MIC, and 6 x MIC for patients infected with susceptible P. aeruginosa isolates. MICs ranged from 0.09 to 64 micro g/ml with modal and median values of 3 and 4 micro g/ml, respectively. Steady-state concentrations of 51.14 +/- 17.52 micro g/ml were estimated in our patient population. The simulation resulted in a median level of exposure 12.62 times the MIC. The level of certainty of obtaining concentrations at the MIC, 2 x MIC, 4 x MIC, 5 x MIC, and 6 x MIC for piperacillin administered by CI was 97, 93, 85, 81, and 77%, respectively. Despite concern for the place of CI piperacillin/tazobactam in the management of P. aeruginosa infections due to the higher established breakpoint, these data suggest a high probability of achieving adequate drug exposure against susceptible isolates with this dosing regimen.