To evaluate, using Monte Carlo simulation, the pharmacodynamics (PD) of empirical antibiotic monotherapies for serious infections consistent with Canadian intensive care unit (ICU) surveillance data. Meropenem, piperacillin/tazobactam and cefepime, along with ceftobiprole, a broad-spectrum cephalosporin active against methicillin-resistant Staphylococcus aureus (MRSA), were tested at standard and highest recommended doses with and without prolonged infusion times (t'). Population pharmacokinetic models were used to simulate antibiotic serum concentrations (n = 5000). Cumulative target attainment (CTA) at >50%, >75% and 100% fT( > MIC) (percentage of time free concentrations exceed the MIC) targets were determined based on ICU surveillance data including 4798 pathogens, most commonly methicillin-susceptible S. aureus (20.1%), Escherichia coli (15.2%) and Pseudomonas aeruginosa (12.3%). With standard doses, ceftobiprole (500 mg every 8 h, t' 2 h) had 0.90 CTA at the >50% fT( > MIC) target while meropenem (1 g every 8 h, t' 0.5 h), piperacillin/tazobactam (3.375 g every 6 h, t' 0.5 h) and cefepime (2 g every 12 h, t' 0.5 h) reached >50% fT( > MIC) in 0.79-0.82 of the population (0.84-0.88 when MRSA was excluded). Piperacillin/tazobactam had the largest reduction in CTA at the >75% and 100% fT( > MIC) targets requiring prolonged infusions to maintain comparable PD. For all agents, prolonged infusions and/or high doses were required to achieve >0.9 CTA at the lowest target, to reach higher targets or to cover less susceptible pathogens such as P. aeruginosa. This study provides important comparative data on empirical antibiotic monotherapies in an ICU setting including preliminary data on ceftobiprole. Ceftobiprole was most active overall, but similar to meropenem, piperacillin/tazobactam (lowest target only) and cefepime when MRSA was excluded. Prolonged infusions in particular and high doses were effective at improving antibiotic PD.