To determine the likelihood that antibiotic regimens achieve bactericidal pharmacodynamic exposures against common nosocomial pathogens. Pharmacodynamic Monte Carlo simulation model. Microbiologic data generated from isolates from the 14 centers in the United States in the 2004 Meropenem Yearly Susceptibility Test Information Collection (MYSTIC) surveillance study. Five thousand simulated patients with infection. Pharmacokinetic profiles of the patients were simulated to determine the bactericidal cumulative fraction of response (CFR) for commonly used intravenous regimens of cefepime, ceftazidime, ceftriaxone, ciprofloxacin, imipenem, levofloxacin, meropenem, and piperacillin-tazobactam against Pseudomonas aeruginosa, Acinetobacter baumannii, Escherichia coli, and Klebsiella species. Ciprofloxacin and levofloxacin had CFRs among the lowest of all drugs against all pathogens, especially P. aeruginosa (40.4-65.5%) and A. baumannii (43.6-48.2%). The low CFR of about 78% against E. coli with these two agents was of particular concern. Among the beta-lactams, only high-dose cefepime and ceftazidime regimens achieved CFRs of greater than 90% against P. aeruginosa, followed by cefepime 2 g every 12 hours and the carbapenems (86.3-89.7%). No regimen achieved an optimum CFR for A. baumannii. All beta-lactam regimens achieved a greater-than-90% likelihood of having bactericidal CFRs against Enterobacteriaceae. Because of the continual evolution of resistance among gram-negative bacteria in the United States, reevaluation of optimum dosing strategies for beta-lactam and fluoroquinolone antibiotics is necessary.